IDENTIFICATION
OF DANGEROUS LEVELS
OF LEAD IN PAINT,
DUST, AND SOIL
TLE IV OF TSCA
SECTION 403
AN ECONOMIC ANALYSIS
DRAFT: January 10, 1994
-------
Associates Inc.
IDENTIFICATION
OF DANGEROUS LEVELS
OF LEAD IN PAINT,
DUST, AND SOIL
TITLE IV OF TSCA
SECTION 403
AN ECONOMIC ANALYSIS
DRAFT: January 10, 1994
Prepared by:
Abt Associates Inc.
4800 Montgomery Lane
Bethesda, MD 20814
Contract Number 68-D2-0175
Work Assignment 2-08
Preparedfor:
Nishkam Agarwal
Regulatory Impacts Branch
Economics, Exposure and Technology Division
Office of Pollution Prevention and Toxics
U.S. Environmental Protection Agency
Authors:
Greg Michaels
Frank Letkiewicz
Alice Tome
Brad Firlie
Amy Benson
Leland Deck
Hampdcn Square • Suite 600 • 4800 Montgomery Lane • Bethesda. Maryland • 20814-5341 • (301)913-0500 • Fax. (301) 652-3618/3839
-------
ACKNOWLEDGEMENTS
We would like to acknowledge the contributions of Dr. Nishkam Agarwal who directed
this project for the Regulatory Impacts Branch, Office of Pollution Prevention and Toxics,
U.S. Environmental Protection Agency. His guidance set the agenda for the entire analysis
and sharpened the focus of its results. We also appreciate the contributions of Dr. Gary
Cole, also of the Regulatory Impacts Branch, to the early development of the model which
was employed in this study.
Finally, we would like to recognize the efforts ofToddAagaard, Karl Kuellmer,
Michael Mailer, and Cassandra De Young ofAbt Associates. Their work to distill large
amounts of data and other information was critical to the conduct of our analyses and to the
presentation of our results in this report.
-------
TABLE OF CONTENTS
EXECUTIVE SUMMARY i
ES.l BACKGROUND ' ' .' ' i
ES.2 AGENCY APPROACH TO SECTION 403 REQUIREMENTS ..... i
ES.3 PURPOSE OF THE ANALYSIS u
ES.4 SYNOPSIS OF THE BENEFIT-COST APPROACH ii
ES.4.1 General Overview ii
ES.4.2 Abatement Decision Rules iii
ES.4.3 Costs '.'.'. iv
ES.4.4 Benefits viii
ES.S BENEFIT-COST ANALYSIS OF ALTERNATIVE HAZARD
LEVELS xii
ES.5.1 Results for the Five Abatement Decision Rules xii
ES.7 SENSITIVITY ANALYSIS xvi
ES.8 IMPACTS OF THE PROPOSED RULE xviii
ES.9 CONCLUSION '. . xx
1. INTRODUCTION 1_!
1.1 PROVISIONS OF RULE 1-1
1.2 STATUTORY AUTHORITY 1-1
1.3 PURPOSE AND CONTENTS OF REPORT 1-4
2. REGULATORY BACKGROUND 2-1
2.1 INTRODUCTION 'm'm\'m \2-\
2.1.1 Lead as a Public Health Problem 2-1
2.2 REGULATION OF LEAD PRODUCTS, ENVIRONMENTAL AND
WORKPLACE RELEASES OF LEAD, AND LEAD IN DRINKING
WATER 2-2
2.2.1 Lead in Paint .2-2
2.2.2 Lead in Gasoline 2-2
2.2.3 Other Lead-based Products 2-2
2.2.4 Environmental and Workplace Releases of Lead 2-3
2.2.5 Lead in Drinking Water 2-3
2.2.6 Resulting Reduction in Blood Lead Levels 2-4
-------
2.3 EFFORTS TO REDUCE LEAD-BASED PAINT, DUST AND SOIL
IN RESIDENTIAL AREAS 2-4
2.3.1 Current Estimates of Exposure 2-4
2.3.2 Federal Regulatory Activities to Decrease Exposure to
Lead-Based Paint in Erieting Housing 2-5
2.3.3 Federal Guidelines and Other Activities Related to Lead
in Soils and Dust 2-8
23.4 State and Local Programs to Reduce Exposure to Lead-
based Paint, Dust and Soil 2-9
2.3.5 Benefits of Defining a Lead Standard for Paint, Dust,
and Soil 2-13
2.4 REFERENCES 2-14
3. PROBLEM DEFINITION AND REGULATORY OPTIONS 3-1
3.1 RISK SUMMARY 3-1
3.1.1 Characterization of Exposure 3-3
3.1.2 Determining Blood Lead Distributions 3-11
3.1.3 Estimated Incidence of Adverse Health Effects 3-17
3.1.4 Extrapolation of Fust Model Year Results to Full
Modelling Time Frame 3-22
3.1.5 Discussion of Results for Baseline Risk Assessment . . . 3-35
3.2 MARKET FAILURE 3-40
3.3 NEED FOR FEDERAL REGULATION 3-43
3.4 REGULATORY OPTIONS 3.44
3.4.1 Information Provision 3.44
3.4.2 Other Regulatory Options 3-46
****•••••*••••••••••••••••••••••» J™^fO
4. COSTS 4_!
4.1 METHOD FOR COST ANALYSIS 4-1
4.2 DATA '. '.4_!
4.2.1 Testing and Abatement Costs of Lead in Dust 4-2
4.2.2 Testing and Abatement Costs for Lead-based Paint .... 4-3
4.2.3 Testing and Abatement Costs of Lead in Soil 4-11
4.2.4 Combined Abatement Scenario Costs 4-15
4.2.5 Enforcement Costs 4-16
4.3 SAMPLE COST CALCULATION 4-20
4.4 RESULTS . . 4-23
4.4.1 Total Abatement Costs 4-23
4.4.2 Quantity of Hazardous Waste Generated by
Abatement 4-27
4.5 REFERENCES . . .
•••••••••••••••••••••••
-------
5. BENEFITS 5_1
5.1 GENERAL ASSUMPTIONS . . . . .5-1
5.1.1 Decision Rule Assumptions 5-2
5.1.2 Abatement Choice Assumptions 5-5
5.1.3 Post-Abatement Exposure Condition Assumptions 5-5
5.2 BENEFITS MODELING PROCESS 5-8
EFFECTS 5.9
5.4 VALUATION OF BENEFITS 5-17
5.4.1 Valuing Lost IQ Points 5-17
5.4.2 Valuing Increased Educational Resources 5-20
5.4.3 Valuing Neonatal Mortality 5-21
5.5 COMPUTING BENEFITS FOR FULL MODEL PERIOD . . 5-21
5.6 RESULTS OF MONETIZED BENEFITS 5-23
6. BENEFIT-COST ANALYSIS 6-1
6.1 BASIS FOR EVALUATION .6-1
6.2 ALTERNATIVE DECISION RULES ; 6-1
6.2.1 Voluntary Optimum Decision Rule 6-2
6.2.2 Decision Rules Based Upon Induced Abatements 6-8
6.2.3 Other Decision Rules to Consider 6-25
6.3 SUMMARY AND CONCLUSION 6-31
7. SENSITIVITY ANALYSES 7-1
7.1 POTENTIAL SENSITIVITY ANALYSES .7-1
7.2 ALTERNATIVE COST ASSUMPTIONS 7-3
7.3 ALTERNATIVE DISCOUNTING PROCEDURE 7-10
7.4 SUPPLEMENTAL BENEFITS FOR ADULTS AND EXISTING
CHILDREN 7_24
7.5 REFERENCES 7-29
7.A APPENDIX 7-30
8. IMPACTS OF THE PROPOSED RULE 8-1
8.1 REGULATORY FLEXIBILITY ANALYSIS .... . . . . . .' . . . . . .' . 8-1
8.1.1 Reason and Legal Basis for Agency Action 8-1
8.1.2 Definition of Small Entity and Affected Populations .... 8-2
8.1.3 Data Availability 8-5
8.1.4 Regulatory Options 8-5
8.2 PAPERWORK REDUCTION ACT ANALYSIS . 8-5
8.3 TRADE IMPACTS ANALYSIS 8-6
8.4 ANALYSIS OF IMPACTS ON TECHNOLOGICAL INNOVATION . . 8-6
-------
8.5 EQUITY IMPACTS ANALYSIS 8-6
8.5.1 Age of Housing Stock 8-7
8.5.2 Regional Distribution 8-8
8.53 Cost of Housing 8-9
8.5.4 Income 8-10
8.5.5 Affordability 8-11
8.5.6 Race 8-22
8.5.7 Other Sociofeconomic Variables 8-22
8.5.8 Data Limitations 8-24
8.5.9 Environmental Equity Conclusions 8-24
8.6 REFERENCES 8-25
-------
EXECUTIVE SUMMARY
ES.l BACKGROUND
On October 28, 1992, Congress enacted the Housing and Community Development Act
of 1992. Title X of that Act - the Residential Lead-Based Faint Hazaid Reduction Act of 1992
~ included among other things an amendment adding "Title IV - Lead Exposure Reduction" to
the Toxic Substances Control Act (TSCA). Among the various requirements of TSCA Title IV
are those set forth under Section 403 - Identification of Dangerous Levels of Lead. Section 403
reads as follows:
"Within 18 months after the enactment of this tide, the Administrator shall
promulgate regulations which shall identify, for purposes of this title and the
Residential Lead-Based Paint Hazard Reduction Act of 1992, lead-based paint
hazards, lead-contaminated dust, and lead-contaminated soil."
ES.2 AGENCY APPROACH TO SECTION 403 REQUIREMENTS
The Agency's approach to the TSCA Section 403 requirements is to establish quantitative
standards for lead in residential soil and dust, and a qualitative standard for lead in paint. That
is, for soil and dust, EPA will identify specific, measurable levels of lead in these media as the
hazard levels. For lead paint, the Agency has determined that the condition and location
(accessibility) of the surfaces bearing the lead paint are more indicative of the potential for a
hazard to be present than are measurements of the amount of lead present on those painted
surfaces. Consequently, the lead paint hazard will emphasize those characteristics rather than
a quantifiable level of lead on painted surfaces.
The Agency recognizes that, from a public health perspective, human exposure to lead
at any level is undesirable. However, the Agency also recognizes that the costs to society and
individuals to reduce lead exposure from paint, soil and dust can be substantial. Therefore, from
an overall public welfare perspective,, it is important to balance properly the public health
benefits achieved from reducing exposure to lead with the costs that are incurred by society to
achieve those benefits.
The Agency also recognizes that while the incidence and severity of adverse effects of
lead generally operate along a continuum of exposure levels, there are different degrees of
potential human health hazards that suggest different forms of remedies to reduce or prevent
exposure. The analysis in the current report illustrates the significant contribution to social
welfare that could be made by fitting a remedial response to the particular hazards of a home.
The Agency will also suggest the types of responses believed to be appropriate for hazards of
different magnitudes and for different sources (namely, soil, dust, and/or paint). It is the
Agency's intent that by helping to set priorities in addressing exposure sources, it will aid in
maximizing lead exposure reductions within constraints imposed by practical resource
limitations.
Abt Associates, Inc. i Dnifti January JQ, 1994
-------
With respect to paint, the Agency recognizes that a dominant pathway of exposure is
through the contamination of dust and soil. Consequently, efforts undertaken to reduce exposure
to lead in dust and soil can effectively reduce a substantial portion of exposure due to lead in
paint, even without specific lead paint abatement efforts. However, in those cases where the
lead paint surfaces are in a deteriorating condition, there is an increased potential for both direct
ingestion from accessible surfaces, and an increased potential for continual recontamination of
dust and soil even with dust and/or soil abatement. As a result, the Agency has elected to focus
its paint hazard standard on the condition and location of the areas having lead paint.
ES.3 PURPOSE OF THE ANALYSIS
The goal of the economic analysis presented here is to help inform the decisions
regarding the specific choices of hazard levels to be identified under Section 403. This report
uses benefit-cost analysis to gain insights into possible hazard levels based upon the amount of
lead in paint, soil, and dust and upon the condition and location of lead-based paint. The results
of these benefit-cost analyses identify optimal lead hazard levels that maximize the net benefits
of undertaking lead exposure reduction actions.
ES.4 SYNOPSIS OF THE BENEFIT-COST APPROACH
ES.4.1 General Overview
Benefit-cost analyses provide an orderly framework both for evaluating the impacts of
a specific hazard level and abatement strategy, and for comparing the relative merits of
alternative levels and strategies. A complicating aspect of evaluating the impacts of the Section
403 regulations, both in terms of their costs and benefits, is that these standards do not
specifically require that any exposure reduction activities be undertaken as a result of their
promulgation. That is, these standards do not themselves compel any public or private entity
to comply with the levels set. However, the Agency recognizes that these hazard levels will be
used widely by federal, state, local and private entities to guide on-going and future efforts to
manage the hazards of lead in paint, soil and dust. Therefore, the Agency considers it
appropriate to attribute changes in the nature or extent of abatement to this rule. By following
this approach, the Agency can arrive at a set of standards which, when acted upon by public and
private entities, will maximize net social welfare.
Because the Section 403 rules do not mandate specific actions, it is not possible to
analyze fully developed regulatory alternatives per se. Instead, an approach has been taken that
considers what the benefits and costs would be if some broadly defined abatement efforts are
undertaken for reducing lead in these media. These abatement alternatives attempt to capture
the range of some reasonable extremes of full-scale lead abatement actions and exposure
reduction efforts conducted on a more limited scale.
In carrying out these analyses, a baseline estimate was first made of the magnitude and
value of the health damages that would be incurred by several generations of children assuming
Abt Associates, Inc. Jj Draft, January 10. 1994
-------
that future exposure levels to lead in paint, soil, and dust are not changed from current
conditions. Against this baseline, analyses were performed of the residual damages that would
be incurred if lead levels in paint, soil and dust were reduced to lower levels as a result of the
broadly defined abatement activities noted above, -reflecting assumptions made about their
efficacy. The benefits of these activities were then determined from the difference between the
estimated baseline damages and the residual, post-abatement damages. These estimated benefits
were then compared with the estimated costs of carrying out the various abatement activities,
identifying the hazard levels generating health benefits greater than costs and, in particular, the
hazard levels maximizing the net benefits to society.
ES.4.2 Abatement Decision Rules
This analysis considers five different decision rules under which homeowners initiate
abatement. It is assumed that this abatement is prompted by and takes place just prior to the
birth of a child, in order to avoid subsequent exposures for the newborn. These decision rules
vary in the degree to which EPA's guidance, primarily in the form of hazard levels, induces the
abatements. At one extreme is a decision rule in which EPA provides no hazard levels.
Instead, households are assumed to obtain perfect information about the benefits and costs of
varioius abatement options for their respective homes and to undertake the one, if any, with the
highest net benefits. The decision rule is called voluntary optimum. It has the highest net
benefits of all decision rules considered.
This decision rule also serves as a benchmark against which the other four decision rules
can be judged, both in terms of net benefits achievable and the nature of the information
households have. Of course, a primary role of promulgating hazard levels is to provide
households with information that will induce abatements where appropriate. This objective is
complicated by the fact that there is a tradeoff between conciseness in the way hazard levels are
promulgated and the appropriateness of any set of hazard levels for a particular home.
Conciseness may be an objective in the design of hazard levels to be promulgated because it is
likely that the more concise the hazard levels are, the more accessible the hazard information
is for homeowners. If, for example, EPA promulgates a hazard level of 2,300 ppm for soil, a
homeowner with a soil reading above this level could take the hazard level as unambiguous
guidance to abate soil. However, it may not be appropriate for all homeowners with soil
readings greater than 2,300 ppm to initiate soil abatement.
This tradeoff is considered in the other four decision rules evaluated in the benefit-cost
analysis. These decision rules vary in the extent to which recommendations are made for the
three different media - soil, dust, and paint - and therefore in the extent to which homeowners
may be induced to take action. Under some decision rules, a recommendation is made for only
one medium, such as dust, and under other decision rules, recommendations are made for all
three media. Yet, the hazard levels considered are all very simple. The recommendations
considered are ones which can be expressed as one set of numbers, such as one hazard level for
soil for all houses in the country. While concise, such recommendations cannot avoid
introducing some errors in abatement choices since the recommendations are never exactly
tailored to any particular house's circumstances.
Abt Associates, Inc. ft Drafti January I0i 1994
-------
Under all four decision rules, a recommendation is made to homeowners to conduct
abatement if lead-based paint is in bad condition. Under the first decision rule, this is the only
explicit recommendation made. Under the second decision rule, a recommendation is also made
to abate if lead exceeds a threshold in one particular medium. Each medium is considered in
turn, so that there are three examples under this decision rule - what happens if a hazard level
for soil alone is promulgated, what happens if a hazard level for dust alone is promulgated, and
what happens if a hazard level for paint alone (based upon the amount of lead present in addition
to condition) is promulgated. The third decision rule examines the net benefits from
promulgating hazard levels for two media in tandem rather one medium alone. All combinations
(soil/dust, soil/paint, and dust/paint) are evaluated. The fourth decision rule evaluates the net
benefits of promulgating all three hazard levels (soil, dust, and paint) together as well as the
paint condition recommendation.
For any given medium under a decision rule, a wide range of levels were evaluated in
the benefit-cost analysis. When more than one medium is combined as a set of candidate hazard
levels, such as soil and dust hazard levels under the two-media decision rule, the number of
potential combinations is enormous. All soil levels from 100 ppm to 3,000 ppm are combined
as candidate hazard levels with all dust levels from 100 ppm to 2,000 ppm. However, this
analysis focuses primarily on the set of hazard levels that have the highest net benefits. The best
set for each decision rule is compared with the best sets for all other decision rules. In this way,
it is possible to identify the best combination of media for which hazard levels could be
promulgated and the best values those hazard levels should take.
ES.4.3 Costs
The cost analysis considered both testing and abatement costs for all three media. Unit
cost estimates were developed first and then applied to the relevant population in each decision
rule to arrive at total costs. Each activity is discussed separately below.
Unit Costs
Testing
The unit testing costs for each media depend on the number of samples taken, the
sampling method and the cost of analysis. Using the Agency draft testing standards as a guide
for the number of samples, the estimated unit testing costs are $230 for interior paint, $1 IS for
exterior paint, $138 for soil and $230 for dust.
Abatement
Abatement unit costs depend primarily on the activities involved. Ten specific
abatements choices were considered. These include two dust abatements (recurrent and
nonrecurrent); two soil abatements (high-end and low-end soil abatements); two paint abatements
(high-end and low-end paint abatements) and four combined paint and soil abatements (high-end
paint with high-end soil; high-end paint with low-end soil; low-end paint with high-end soil; and
low-end paint with low-end soil). The components of each abatement scenario are shown in
Exhibit ES-1.
Abt Associates, Inc. iv Draft. January 10, 1994
-------
EXHIBIT ES-1
Summary of Abatement Scenarios
Abatement Scenario
Nonrecurrent Dust
Recurrent Dust
High-end Paint
Low-end Paint
High-end Soil
Low-end Soil
High-end Soil and High-end Paint
High-end Soil and Low-end Paint
Low-end Soil and High-end Paint
Low-end Soil and Low-end Paint
Activities
Families moved off-site, hard surfaces (floors,
woodwork, window wells and some furniture)
vacuumed with a high-efficiency particle
accumulator (HEPA) vacuum. Hard surfaces also
wiped with a wet cloth (an oil treated rag was used
on furniture) following vacuuming.
Every ten years a thorough cleaning as listed under
nonrecurrent dust. Every month an additional
standard house cleaning consisting of general
dusting, vacuuming, cleaning bathrooms and wiping
window sills.
Full abatement of windows, doors, woodwork and
walls by removal or replacement and a high-end
dust abatement.
Replacement of windows and high-end dust
abatement.
Removal of six inches of top soil, installation of a
barrier and replacement of soil with new soil tested
under ISO ppm lead, resodding and high-end dust
abatement
Resodding including grading but no removal of
existing grass plus high-end dust abatement.
Combination of high-end soil abatement and high-
end paint abatement.
Combination of high-end soil and low-end paint
abatement.
Combination of low-end soil and high-end paint
abatement.
Combination of low-end soil and low-end paint
abatement.
Abt Associates, Inc.
Draft, January 10, 1994
-------
Two types of dust abatement were modeled. Nonrecurrent dust abatement assumes a
one-time high-end cleaning during which the occupants are moved off-site; hard surfaces are
vacuumed with a high efficiency particle accumulator vacuum and wiped with a wet rag (or an
oil treated rag in the case of furniture). Based upon available information, the cost was
estimated to be $750. In addition to being considered as a separate abatement, the high-end dust
abatement was included as a component of both the paint and soil abatements as noted below.
The second form of dust abatement involved an initial high-end dust abatement, as described
above, followed by monthly routine house cleaning with additional high-end abatements every
ten years. This recurrent dust abatement scenario reduces the dust level to 100 parts per million
(ppm). The cost of the routine house cleaning was estimated to be $38 per month, repeated
monthly. The present value of recurrent dust abatement, calculated over the fifty vear lifetime
of the model, is $7,676.
For paint abatement, the high-end level of abatement assumed full abatement of windows,
doors, woodwork and walls, as well as high-end level dust abatement (see above). The present
value cost of the high-end paint abatement was estimated to be $10,500 per home. It is assumed
to be permanent and to render the home free of lead-based paint. The surrogate used to model
low-end paint abatement was replacing only windows, and consequently was applicable only to
those homes having windows with lead-based paint present. As with the high-end abatement,
the cost of one-time high-end dust abatement was also included. The unit cost (i.e., cost per
home) was estimated to be $2,750, and the abatement was assumed to be permanent for the
windows. The effectiveness was assumed to be equivalent to reducing the dust lead level by
8.6%, the portion windows contribute to total lead-based paint area in the 1989-90 Housing and
Urban Development Department (HUD) survey results.
The high-end soil abatement was assumed to involve removing the top six inches of soil,
filling the yard with new soil having a lead level below 150 ppm, and resodding. High-end dust
abatement in these homes was also included. Soil with a lead level above 2000 ppm was
assumed to be treated as hazardous waste; thus, hazardous waste transportation and disposal
costs were added. In addition, if the house had exterior lead-based paint, the cost of its removal
was included because this abatement was needed to ensure the effectiveness postulated. The cost
of the high-end soil abatement was estimated to be $7,998 if the soil was not considered a
hazardous waste and if there was no lead in the exterior paint. Waste disposal added $8,414 and
exterior paint abatement added $5,000 to the abatement cost. The high-end soil abatement was
assumed to be permanent. The low-end soil abatement scenario assumed resodding with some
preparatory work on the ground and was estimated to cost $2,860. It was assumed that the low-
end soil abatement results in a post-abatement soil level equivalent to 500 ppm. The low end
soil abatement was assumed to require repeated resodding every five years for a net present
value unit cost of $7,493.
Total Costs
The total costs are shown in Exhibit ES-2 for the five decision rules evaluated. Each
total cost has a testing and an abatement component. The total testing costs depend on the
decision rule and the number of media tested. Testing takes place at the birth of the first child
and total discounted testing costs over the fifty year span of the model range from $14.9 billion
for two media to $24.3 billion for all three. The abatement costs range from $13.9 billion for
Abt Associates, Inc. vi DmA Jmuary ]Q 1994
-------
EXHIBIT ES-2
Total Costs for Five Alternative Decision Rules
'•
2.
3.
4.
5.
Decision Ruin*
Voluntary
Optimum1*
Paint Condition
Onlyc
Single
Medium
Pbu
Condition11
2-Media
Phn
Condition*
3«.
3b.
3e.
4a.
4b.
4e.
3-Medh Pliu
Condition'
Soil
(pom)
-
-
2.300
-
-
2.300
2.300
-
2.300
Dull
(ppm)
-
-
-
1.200
-
1.200
-
1.200
1.200
hint
(XRF.
rag/cm1)
-
-
-
-
20
-
20
20
20
Nonintact Paint
Abatement
Recommended
No
Yea
Yea
Yea
Yea
Yea
Yea
Yea
Yea
Abatement
Costs
(S million)
13.196
24.668
21.344
29.646
25.013
31.903
28.689
29.991
32.248
Tealing Costs
(S million)
24.222
14.982
14.982
24.346
14.982
24.346
14.982
24.346
24.346
Total Coata
($ million)
38.118
39.650
43.326
53.992
39.995
56.249
43.671
54.337
56.594
Coata by Type of Abatement Chosen* ($ million)
HP
20.134
20,134
20.260
20.429
20.260
20,429
20.555
20,555
LP
26
3.517
3.442
3.462
3,567
3.387
3.493
3.551
3.438
HS
272
272
272
272
IS
1.992
3,475
1.420
3.475
3,475
1.420
3.475
RD
971
978
978
978
HP/HS
HP/LS
943
943
943
943
943
943
LP/HS
87
87
87
87
LP/LS
74
350
74
74
350
390
74
330
NRD
=^=
11,878
2,150
2.150
2,150
2.150
=^^™—
'Candidate hazard levels examined ranged up to 3000 ppm for soil, 2000 ppm for dust and 20 rag/cm1 in paint.
bEach home selects abatement (or no abatement) that has highest net benefits.
cAbatement is recommended for homes with more than five square feet of lead-based paint in nonintact condition, regardless of XRF level or net benefits. Home
owners choose the paint abatement method that generates the highest net benefits. Results are reported only for homes that exceed recommended levels.
dWithin the full range of individual soil, dust and paint hazard levels that could be set as a threshold for action, with no constraints placed on the other two
media, the levels specified in the table maximize the net benefits. Results are reported only for homes that exceed recommended levels.
eWithin the fall range of individual soil, dust and paint hazard level combinations that could be set as a threshold for action, with no restriction on the other
medium, the levels specified in the table maximize the net benefits. Results are reported only for homes that exceed recommended levels.
fWithin the full range of individual soil, dust and paint hazard level combination that could be set as a threshold for action, with no restriction on the other
medium, the levels specified in the table maximize the net benefits. Results are reported only for homes that exceed recommended levels.
SAbalement Codes: High Paint(HP); Low Paint(LP); High Soil(HS); Low Soil(LS); Recurrent Dust (RD); High Paint and High Soil(HP/HS); High Paint and
Low SoiI(HP/LS); Low Paint and High Soil (LP/HS); Low Paint and Low Soil (LP/LS); Nonrecurrent Dust (NRD). The abatement activities were
described in Exhibits 4.1-4.6.
-------
the voluntary optimum to $32.2 billion for the three media plus paint condition constrained
optimum. The higher value of the latter is a result of households being required to choose more
expensive abatements to meet the constraining criteria than they would have chosen voluntarily.
The majority of abatement costs of the constrained decision rules result from abating all
nonintact paint. The total costs, including both testing and abatement, range from $38 billion
to $56 billion.
ES.4.4 Benefits
The benefits associated with various combinations of hazard levels and abatement choices
have been examined using the following three primary measures:
• Changes in the characteristics of the population blood-lead distributions for children;
Estimates of avoided incidence of adverse health effects, specifically, avoided loss
of IQ points, avoided incidence of IQ < 70, avoided incidence of PbB > 25 /tg/dl, and
avoided incidence of neonatal mortality; and
• Monetary value of the avoided adverse health effects noted above.
Exhibit ES-3 summarizes the benefits for ten decision rule options. Included there are
the nine options that are described above in which the hazard levels for paint, soil and dust were
derived primarily from net benefits considerations. For the tenth decision rule, hazard levels
were obtained that would minimize an individual's risk of experiencing high blood-lead levels.
The rule was designed to result in less than 10% risk of blood-lead exceeding 10 /tg/dl, less than
5% risk of blood-lead exceeding 15 j*g/dl, and less than 1 % risk of blood-lead exceeding 20
pg/dl. Setting hazard levels of 500 ppm for soil and 400 ppm for dust (with paint in bad
condition also inducing abatement) accomplishes the risk-based goal.
It is important to note that the benefits estimated here include only those that were
quantifiable given currently available data. These types of benefits are comparable to those that
have been estimated for children in conjunction with other recent regulations aimed at reducing
environmental exposure to lead. However, other potential benefits exist for reducing childhood
lead exposure, such as avoided impairment of certain metabolic processes and possible avoidance
of cancer. In addition, benefits realized to the adult population from reduced exposure to lead
in residential settings have not been included in the base analysis because available risk
information is considered weak in the risk assessment community. These benefits are estimated,
however, in the sensitivity analysis.
The largest benefits overall are estimated to derive from the 500/4007- decision rule, with
approximately $67 billion in benefits over the full 50-year modeling time frame. This outcome
was expected since the environmental exposure to lead is reduced more than under all the other
decision rules resulting in the lowest post-abatement geometric mean blood-lead values (2.45
pg/dl), the lowest 90th percentile value (6.43 jig/dl) and the largest values for avoided incidence
of IQ point loss, IQ < 70, PbB > 25 /ig/dl and incidence of neonatal mortality.
Abt Associates, Inc. viii Draft, January 10, 1994
-------
Exhibit ES-3
Summary of Estimated Benefits
Baseline
Decision Rule:
Voluntary Optimum
Paint Condition Only
Single Media: 2300/-/-
Single Media: -/I200/-
Single Media: -1-120
2-Media: 2300/12007-
2-Media: 2300/-/20
2-Media: -/1200/20
3-Media: 2300/1200/20
2-Media Option to Minimize High
Blood Leads: SOO/400/-
Post-Abatement Population Blood Lead
Distribution Characteristics fog/Hi)
OM
4.06
2.52
3.82
3.78
3.32
3.82
3.30
3.78
3.32
3.30
2.45
OSD
2.45
2.70
2.47
2.43
2.39
2.47
2.35
2.43
2.37
2.35
2.2
Med.
3.91
2.40
3.72
3.71
3.36
3.72
3.36
3.71
3.36
3.36
2.56
90th ft-tile
13.30
9.47
12.55
11.99
9.81
12.55
9.67
11.99
9.81
9.67
6.43
Afoided IQ Point Loss
Total
0
1,912,011
319,818
490,626
1,316,456
321.631
1,402,677
492,439
1,318,269
1,404,490
2,751,452
Avg./Child.
0
1.34
0.86
1.16
1.94
0.86
1.98
1.15
1.93
1.97
1.58
Avoided Incidence of:
IQ <70
0
5,434
1,081
1,909
4,874
1,084
5.305
1.912
4,877
5,308
8,823
PbB>25
0
63,422
14,744
30,736
68,513
14,745
77,479
30.376
68.513
77.479
94.656
Neonatal
Mortality
0
0
48
48
48
48
48
48
49
49
75
Vahie of Benefits orer
FuD Model
Tbneframe ($B)
=^=^=;^=
$0.0
$48.2
$7.3
$11.2
$32.9
$7.4
$34.9
$11.2
$33.0
$35.0
$66.7
=^=^=^=
Note: Except for the monetary value of the benefits shown in the last column, benefits presented hen are for first model year cohort.
-------
The Voluntary Optimum rule elicits the largest benefits among those decision mles
derived from net benefits considerations. The value of the benefits for the Voluntary Optimum
rule over the full modeling time frame is $48.2 billion. Although the post-abatement geometric
mean for the Voluntary Optimum at 2.52 jtg/dl approaches the value estimated for the 500/400/-
nile noted above, its geometric standard deviation (GSD) of 2.7 is significantly larger than the
GSD of 2.2 for the 500/4007- rule. This difference in the spread of values translates into a
higher 90th percentile value (9.47 pg/dl) for the Voluntary Optimum rule and a lower incidence
of avoided blood lead values above 25 /xg/dl. It is also noteworthy that the Voluntary Optimum
rule results in no avoided cases of neonatal mortality, since this rule does not induce any high-
end paint abatement (a necessary condition to obtain this benefit).
As noted in the preceding section, the nine decision rules arrived at using net benefits
consideration include the constraint that lead-based paint in bad condition will induce paint
abatement regardless of the XRF value. Listed as Paint Condition Only, Exhibit ES-3 shows
that this component alone elicits $7.3 billion in benefits over the full modeling time frame.
Among the decision rules setting specific hazard levels arrived at using net benefits
considerations, the largest benefits are seen for those which include dust at 1200 ppm among the
hazard levels (i.e., -/1200/-, 2300/1200/-, -/1200/20, and 2300/1200/20). The total benefits for
these four rules range from $33.0 to $34.9 billion, with comparable impacts on blood lead
distributions, avoided loss of IQ points, avoided incidence of IQ < 70, avoided incidence of
PbB > 25 /ig/dl, and avoided neonatal mortality. Among these four decision rules, slightly
higher benefits accrue for the two that also include soil at 2300 ppm.
Of the three rules setting hazard levels that do not involve a dust component, the largest
benefits are seen for the two involving soil at 2300 ppm (2300/-/- and 2300/-/20). These rules
have total benefits estimated at $11.2 billion, with virtually identical results for the health effects
avoided.
The lowest benefits among the rules setting specific hazard levels come from the single
media rule setting paint at an XRF of 20 (-/-/20). The total benefits for this rule are estimated
to be $7.4 billion. Note, however, that the paint condition component alone is estimated to
provide $7.3 billion, indicating that setting the additional hazard level at an XRF of 20 for all
paint regardless of condition provides little additional benefit above that resulting from abatement
of paint in bad condition.
Exhibit ES-4 displays the relative contribution of the monetized value of each of the
categories of benefits based on the first model year abatements for each decision rule. The
relative contributions of the benefits categories are comparable for the entire modeling period.
By far, the major contribution to the value of the benefits derives from the avoided loss
of IQ points. For all of the decision rules, this component of the benefits contributes between
75 % and 90% of the value of the benefits. The contributions of the avoided incidence of IQ <
70 and of blood-lead levels > 25 pg/dl are comparable for each decision rule, generally
contributing between 5% and 7% of the total benefits each.
Abt Associates, Inc. x 0^ january JQ, 1994
-------
Exhibit ES-4. Distribution of Monetized Benefits by Category for
First Model Year Abatements.
1.1
II
ii
c •g
» 2
m ;r
$3
$2
$1
$0
Total Benefits
Avoided IQ Loss
Avoided Neonatal Ktortaity
Avoided PbB iCS
Avoided KLOQ
Awoided IQ<70
Avoided PbB>fi5
Avoided Neonatal
Mortality
Avoided IQ Loss
Total Benefits
Abt Associates, Inc.
XI
Draft, January 10, 1994
-------
The most variable contributor to the value of the benefits is avoided neonatal mortality.
Except for the Voluntary Optimum and the 500/400/- decision rules, the monetized value of
these benefits is comparable, approximately $140 million for the first year. For the Voluntary
Optimum, neonatal mortality avoidance makes no contribution, while for the 500/400/- rule the
value is about $220 million. Excluding the voluntary optimum, the avoided neonatal mortality
benefits as a percentage of the total are highest for those decision rules with the lowest total
benefits, and highest for those with the lowest benefits. For example, in the Paint Condition
Only rule, the first year benefits are estimated to be about $890 million, and the neonatal
mortality benefit at $138 million comprises about 16% of the total. By contrast, the
2300/1200/20 rule has total first year benefits of $3.5 billion, of which neonatal mortality at
$141 million is only 4% of that total. For the 500/4007- rule, where the total first year benefits
are highest at $6.6 billion, the neonatal mortality is also highest among all rules at $220 million.
However, as a percentage of the total, these benefits constitute only about 3.3% for the
500/400/- rule.
ES.5 BENEFIT-COST ANALYSIS OF ALTERNATIVE HAZARD LEVELS
ES.5.1 Results for the Five Abatement Decision Rules
Exhibit ES-5 presents the benefit-cost results for each of the five decision rules. The first
comparisons made in this section focus on the net benefits of each decision rule exclusive of
testing costs. This orientation makes it possible to distinguish decision rules in terms of the net
benefits of abatement. The final comparisons in this section integrate information on the testing
costs necessary to implement each of the decision rules.
Net Benefits Excluding Testing Costs
The voluntary optimum clearly sets the standard for judging the performance of the other
four decision rules. It generates net benefits of $34.3 billion before testing costs are considered.
The net benefits of the voluntary optimum surpass those of the next best alternative by nearly
$30 billion. More than 45 million homes undertake abatement, a substantial number especially
given the assumption under this decision rule that only homes obtaining positive net benefits
initiate abatement. Exhibit ES-6 shows that all but one of the abatements constructed for this
analysis would be chosen. Nonrecurrent dust abatement is the leading choice, by far, and low-
end soil abatement is a distant second. All other decision rules entail at least two million more
paint abatements, primarily because of the recommendation to abate lead-based paint in bad
condition.
This conclusion is borne out by the results for the second decision rule, which is based
upon the paint condition recommendation. Net benefits are negative (-$17.3 billion) even before
testing costs of nearly $15 billion are considered. Virtually all of the 7.1 million abatements
conducted under this decision rule are motivated by the assumption that all homeowners having
paint in bad condition will undertake the best form of abatement available to them. This
decision rule provides a telling example of the tradeoff between conciseness in the form that
Abt Associates, Inc. xii Draft. January 10, 1994
-------
Exhibit ES- 6
Benefit-Cost Results for Five Alternative Decision Rides
2.
Condition
Condition
3b.
3c.
4b.
4c.
o meaia riuo uonaiuon
Ippm)
2,300
2.300
2,300
2,300
(ppm)
1,200
1,200
1,200
1.200
mg/cm*
20
20
20
20
Abatement
Recommended
Yes
Yes
Yes
($ million)
7,319
11,186
32,945
7,383
34,920
11,249
34,984
Abatement
Costs
(9 million)
24,668
28,345
29.646
25,014
31.903
28.689
29.992
32,248
(Exclusive of
Testing Costs)
(9 million)
(17,349)
(17.159)
3.299
(17.631)
3.017
(17,440)
3,017
2.736
Testing
Costs
($ million)
24,222
14,987
24.222
24.227
14,906
24,227
24,222
24,227
24^3T
Net Benefits
(Including
Testing Costs)
10,072
(32,336)
(41.381)
(20,928)
(32,537)
(21.210)
(41,662)
(21,210)
(21,495)
Total Number
of Abatements
(1000s)
45.166
7.063
8,069
15,602
7.164
16,197
8.170
15.702
16.297 |
Number of
Abatements with
Negative Net
0
6,686
7,164
7.114
6.786
7.683
7.266
7.216
7.684 |
Candidate hazard levels examined ranged up to 3000 ppm for soil, 2000 ppm for dust, and 22 mg/cm' for paint.
Voluntary Optimum:
Paint Condition Only:
Single Medium:
2-Media:
3-Media:
Each home selects abatement (or no abetement) that has the highest net benefits
Abetement is recommended for homes with more then five square feet of lead-based paint in nonintaet condition, regardless of XRF level or net
benefits. Homeowners choose the paint abatement method that generates the highest net benefits.
Within the full range of individual soil, dust, and paint hazard levels that could be set as a threshold for action, with no constraints placed on the other
two media, the levels specified in the table maximize net benefits.
Within the full range of individual soil, dust, and paint hazard level combinations that could be set as a threshold for action, with no restriction on the
other medium, the levels specified in the table maximize net benefits.
Within the full range of individual soil, dust, and paint hazard level combinations that could be set as a threshold for action, the levels specified In the
table maximizes net benefits.
-------
Exhibit ES - 6
Distribution of Abatement Choices for Five Alternative Decision Rules
4.
Condition
2-Media Plus
Condition
3a.
3b.
3c.
4a.
4b.
4c.
(ppm)
•
2.300
•
•
2.300
2.300
2,300
(ppm)
•
-
1.200
•
1.200
•
1.200
1.200
(mg/cm*)
-
-
-
20
-
20
20
20
Abatement
Recommended
Yes
Yes
Yes
Number of Homes Abated by Abatement Type (1000s)
HP
4,160
4,160
4,186
4,221
4,186
4,221
4,247
4.247
LP
2,774
2.716
2,731
2.814
2,672
2,756
2,770
2,712
HS
0
0
74
0
74
0
74
74
LS
0
1.006
411
0
1,006
1.006
411
1,006
RD
0
0
• o
276
0
276
0
276
276
HP/HS
0
0
0
0
0
0
0
0
0
HP/LS
0
114
114
114
114
114
114
114
114
LP/HS
0
0
0
18
0
18
0
18
18
LP/LS
O
16
74
16
16
74
74
16
74
NRD
44,567
0
0
7.777
0
7.777
0
7.777
7,7i1
Total
45.166
7,063
8,069
15,602
7,164
16.197
8.170
16.702
16.297
Candidate hazard levels examined ranged up to 3000 ppm for soil, 2000 ppm for dust, and 22 mg/cm* for paint.
Abatement Codes
HP = High Paint Abatement
LP = Low Paint Abatement
HS = High Soil Abatement
LS = Low Soil Abatement
RD = Recurrent Dust Abatement
Voluntary Optimum:
Paint Condition Only:
Single Medium:
2-Media:
3-Media:
HP/HS = High Paint, High Soil Abatements
HP/LS = High Paint, Low Soil Abatements
LP/HS = Low Paint, High Soil Abatements
LP/LS = Low Paint, Low Soil Abatements
NR D = Nonrecurrent Dust Abatement
Each home selects abatement (or no abatement) that has the highest net benefits
Abatement is recommended for homes with more than five square feet of lead-based paint in nonintact condition, regardless of XRF level or net
benefits. Homeowners choose the paint abatement method that generates the highest net benefits.
Within the full range of individual soil, dust, and paint hazard levels that could be eet as a threshold for action, with no constraints placed on the other
two media, the levels specified in the table maximize net benefits.
Within the full range of individual soil, dust, and paint hazard level combinations that could be set as a threshold for action, with no restriction on the
other medium, the levels specified in the table maximize net benefits.
Within the full range of individual soil, dust, and paint hazard level combinations that could be eet as a threshold for action, the levels specified In the
table maximizes net benefits.
-------
EPA's guidance could take and the resulting inaccuracies in decisions made about abatement.
The last column of Exhibit ES-5 shows the number of abatements induced by this
recommendation that result in negative net benefits. Of the 7.1 million abatements conducted,
95%, or 6.7 million homes, would be expected to undertake abatement, because it has been
recommended, even though it results in negative net benefits.
This outcome is possible not because homeowners are making irrational decisions, i.e
taking steps known to generate negative net benefits, but because they are relying on limited
information relative to what homeowners are assumed to have under the voluntary optimum
Under this and all subsequent decision rules, it is assumed that the decision to abate is induced
by exceeding one or more hazard levels and not by calculating the net benefits directly. If
homeowners had made the calculation, as was assumed in the voluntary optimum, they would
not have undertaken the abatement. Clearly the results from the second decision rule show that
there is room for finetuning the recommendations made to homeowners in the form of hazard
levels. The third decision rule adds one more dimension to EPA's potential recommendations -
a hazard level for one of the three media (soil, dust, and paint) while keeping the paint
condition criterion of the second decision rule. Two of the cases under this single-medium rule
single hazard levels for soil and paint, also result in negative net benefits and have only slightly
lower error rates than the previous decision rule. For soil and paint, 89% and 95% of their
respective sets of abatements have negative net benefits.
The prospects for using a single hazard level for dust are more compelling The net
benefits of a dust hazard level of 1,200 ppm are $3.3 billion. The number of homes generating
negative net benefits is high, 7.1 million, but the number of abatements induced is also high
15.6 million, resulting in an error rate of 46%. As Exhibit ES-6 shows, 52% of these
abatements (8 out of 15 million) entail some form of dust abatement only. This proportion is
lower than that exhibited in the voluntary optimum, where approximately 99% of homes choose
some form of dust abatement, because of the high number of homes assumed to undertake paint
abatement to comply with the paint condition recommendation.
The fourth decision rule adds yet another dimension to the candidate hazard levels
combining them for two media rather than just one. To satisfy recommendations based upon
these hazard levels, such as the case where the soil hazard level is 2,300 ppm and the paint
hazard level is 20 mg/cm2, it is assumed that any homeowner whose home exceeds either one
of these thresholds would undertake the best soil and/or paint abatement that makes it possible
to go below any binding threshold (as well as to meet the paint condition criterion). Exhibit ES-
6 shows the distribution of abatements that would be induced under each set of two-media hazard
levels. Under this decision rule, the versions including a dust hazard level are the only ones
having positive net benefits before testing costs are considered. In each case, the optimal hazard
level for dust is 1,200 ppm, the same as the optimal level for the single-medium rule. However,
adding one more dimension to the decision rule constrains household decisions enough to reduce
benefits slightly. The net benefits of either two-media rule involving dust are $3.0 billion. The
optimal hazard level to combine with a dust hazard level is 2,300 ppm for soil or 20 me/cm2 for
paint.
The fifth and final decision rule presents the best set of three-media hazard levels The
resulting optimum brings together the hazard levels identified as optimal in the single-medium
Abr AssociaKs, Inc. Jamary
-------
and two-media decision rules - 2,300 ppm for soil, 1,200 ppm for dust, and 20 mg/cm2 for paint
plus the paint condition criterion - but at the cost of lower net benefits. Adding a third hazard
level to the decision restricts the leeway in household choices yet again. This lowers the net
benefits by approximately 10% and 20% compared respectively to the best single-medium and
two-media cases.
In closing this discussion of the relative net benefits of different decision rules, it is also
useful to point out that besides their clear differences in net benefits, the voluntary optimum
differs substantially from the other four rules in the number of homes abated. The four decision
rules based upon qualitative or quantitative hazard levels induce no more than 16.3 million
abatements, of which 7 to 8 million have negative net benefits. The voluntary optimum leads
to nearly three times as many abatements - more than 45 million abatements. None entails
negative net benefits. This finding raises the possibility that better decision rules could be
created that are both implementable, which is the advantage of the decision rules based upon
hazard levels, and that lead to positive and substantial net benefits, which the voluntary optimum
does. Since the information bases assumed for the first group and for the voluntary optimum
differ substantially, it appears that creating a better means for conveying useful information to
guide homeowners' abatement decisions could be a productive route for improving upon the
decision rules investigated here.
Net Benefits Including Testing Costs
The inclusion of testing costs does not alter the ranking of nine decision rules considered
here although it does lower the net benefits. The voluntary optimum still has the highest net
benefits ($10 billion). All subsequent decision rules have negative net benefits. The top four
among these are based upon a single medium (dust=1200), two media (soil=2300/dust= 1200,
dust=1200/paint=20), and three media (soa=2300/dust=1200/paint=20). The overall net
benefit of any of these decision rules is approximately minus $21 billion, once testing costs are
considered. The three lowest-ranking decision rules all have in common that they are based in
some way upon a paint hazard level. The qualitative hazard level based upon paint condition
criterion ranks seventh, the two-media rule based upon soil and paint ranks eighth, and the
single-medium hazard level based upon paint alone ranks ninth. These outcomes and the finding
of significant negative net benefits associated with paint abatement highlight the potentially
significant influence of the assumptions made regarding the effectiveness and cost of paint
abatement.
ES.7 SENSITIVITY ANALYSIS
The number of parameters considered in this report is limited to three items that had a
significant likelihood of influencing the policy-relevant outcomes.
First, the report considers the impacts that using low and high unit cost estimates have
on the benefits and costs of different decision rules and their optimal hazard levels. These
Abt Associates, Inc. XVI D^fi January 10, 1994
-------
assumptions provide a basis for comparison with the "medium" cost estimates used in the main
analysis. This basis establishes possible boundaries for the benefit-cost results and the estimated
optimal hazard levels. While the low and high unit cost estimates used in the sensitivity analysis
are not the ultimate minimum and maximum values, they were constructed to be representative
of the lower and upper ranges of values observed in practice. As such, these unit cost estimates
are probably appropriate for testing the boundaries of the benefit-cost analysis.
The bounding exercise for costs indicates that the findings regarding optimal dust and
paint hazard levels may not be affected by a better representation of the distribution of abatement
costs. Dramatic upward and downward revisions applied simultaneously to all abatements did
not change the optimal hazard levels for dust and paint. This does not however categorically
rule out revisions in cost estimates which could affect the optimal dust and paint hazard levels.
The sensitivity analysis does show however that the optimal soil hazard level could be
susceptible to changes in assumptions regarding the costs of abatement. While the optimal soil
hazard level held constant at 2,300 ppm for an upward revision in all abatement costs, it fell to
1,500 ppm when all costs were lowered. Consequently, it appears that the evidence for setting
a single hazard level for soil is not clearcut. Instead, a range from 1,500 to 2,300 ppm is
supported by the model when bounding cost assumptions are applied.
The second focus of the sensitivity analyses in this report is the discount rate used to
express the monetary value of future benefits and costs in contemporary terms. The selection
of a discount rate is one of the most commonly debated features of benefit-cost analyses of
environmental policies. A rate of 7% was used in the main analysis of this report. An alternate
approach, which has been used by the Agency in some other regulatory analyses, involves a two-
stage discounting procedure that employs both 3 % and 7%. The impact of this approach on the
findings of the main analysis was evaluated in the sensitivity analysis.
The sensitivity analysis revealed that the two-stage discounting procedure raises the
possibility of a wider range of optimal hazard levels for dust and soil than was implied in the
main analysis. The range for the optimal dust hazard levels is from 300 ppm to 1,200 ppm
The range for the optimal soil hazard level is 1,000 ppm to 2,300 ppm. The current analysis
cannot categorically support the selection of one hazard level for dust and soil from each of these
ranges. However, several things remain constant between the findings of the main analysis and
those of the sensitivity analysis. The voluntary optimum is far away the decision rule generating
the highest net benefits and the paint condition rule typically the least (and virtually always
negative). The types of decision rules that involve hazard levels for dust and soil that have the
highest net benefits are generally the same under the two-stage procedure as they are in the main
analysis. They are the single-medium dust rule, the two-media rales based upon soil and dust
and upon dust and paint, and the three-media rale.
The third focus of the sensitivity analyses in this report considers the possible benefits
to adults and children already in the home at the time that abatement takes place. The main
analysis presented in this report was based upon a model developed to consider the benefits to
children from the time of birth until the age of seven years from the abatement of lead
contamination. Not only was the risk assessment focused on this population alone but the
behavioral assumptions regarding lead abatement were integrally linked to the impending births
Abi Associates, Inc. xvii Dmft Januafy I
-------
of children. This model structure reflects the fact that this population has been considered a
primary target for measures to prevent residential exposures to lead-contaminated paint, dust,
and soil. However, other populations may also benefit from lead abatement. This sensitivity
analysis focuses on existing children in the household who are under the age of seven and on
adults in the household.
There are significant barriers to conducting a refined risk assessment for these two
populations. Little is known about the size of the IQ benefits to these existing children who only
partially avoid lead exposure during the critical seven-year period of intellectual development
since abatement was initiated sometime after they were bom. For adults, risk assessment is
complicated by the lack of information on the relationships between lead levels in paint, soil and
dust and blood lead levels for populations over the age of seven. For both populations the
current model has shortcomings because abatements are triggered by impending births rather
than the circumstances of either existing children or adults. For these reasons, the overall
estimates in this sensitivity analysis should be viewed as illustrative only, in light of the
unrefined and somewhat arbitrary assumptions needed to generate the supplemental benefit
estimates.
This analysis found an optimal hazard level for soil of 1,400 ppm. The optimal hazard
level for dust was 400 ppm. The experience with paint hazard levels under this sensitivity
analysis reproduced that of the two-stage procedure, where the optimal hazard level was
variously 4 or 20 mg/cm2, depending upon the decision rule, but the highest net benefits were
based upon 20 mg/cm2.
Taken together, the three sensitivity analyses presented in this chapter raise the possibility
of a wider range of potentially optimal hazard levels than the findings in the main analysis
imply. The optimal dust hazard level may be as low as 300 ppm or as high as 1,200 ppm. The
main analysis found a dust hazard level of 1,200 ppm. The optimal soil hazard level may be
as low as 1,000 ppm or as high as 2,300 ppm. The main analysis found a soil hazard level of
2,300 ppm. In the main analysis and in the sensitivity analyses, the highest net benefits for paint
were associated with a hazard level of 20 mg/cm2. Finally, a qualitative hazard level based upon
paint condition typically entails negative net benefits with the exception of two cases: this
particular sensitivity analysis which linked supplemental benefits to paint abatement specifically
and, the shortest-term amortization case (10 years) under the two-stage discounting procedure.
These two cases still seem rare enough to raise doubts about the desirability of a paint condition
criterion given the paint abatement options constructed for this analysis.
ES.8 IMPACTS OF THE PROPOSED RULE
Two types of impacts received primary attention in this report. The first set of impacts
relate to the regulatory flexibility of a rulemaking for Section 403. Although no formal
regulatory impact analysis has been conducted, since Section 403 does not require specific action
to abate residences, this report identifies small entities likely to be induced to conduct abatement
activities because of Section 403 and evaluates the available data to quantify any effects on these
Abt Associates. Inc. XVlii Draft, January 10, 1994
-------
small entities. Small landlords are the likeliest small entity to be affected. Data on small
landlords are not organized in an accessible form for a national analysis. Consequently, further
analysis of this impact could require resource-intensive data collection. The second set of
impacts relate to concerns about disproportionate burdens being placed on particular categories
of households and individuals by actions under Section 403. These impacts were examined using
socioeconomic information on a sample of homes and residents from the Department of Housing
and Urban Development (HUD) survey of residential lead contamination that was a basis for the
benefit-cost analysis in this report, as well as income information from the U.S. 1990 Census.
The results of this equity analysis are discussed below.
Existing lead-based paint hazards are a risk to all segments of our population living in
pre-1980 housing. However, the HUD survey does indicate that some segments of our society
are at relatively greater risk than others. In particular, the residents of older, low cost housing
are exposed to a disproportionately greater share of lead potential hazard than other housing
units. The housing stock in the North-East (and to some extent the Mid-West) includes a larger
share of such units than other regions, creating a regional inequity in the prevalence of the
problem. Because poorer people usually occupy low-cost housing, the hazards
disproportionately fall on lower income sub-populations (especially households living in poverty,
with annual incomes below $10,000), creating an income inequity. Finally, the relatively larger
share of African-Americans in the lower income groups results in racial inequity.
Although the baseline risks from lead-based paint disproportionately fall on poorer sub-
populations, abatement may well be more likely to occur in housing units occupied by wealthier
households. Most of the abatements under the Lead-Based Paint Hazard Reduction Act will be
voluntary, and wealthier households are more likely to have the means to abate an existing
problem in their home, or avoid moving into a housing unit with a known lead-based paint
hazard. An analysis conducted for this report shows that income constraints could have a
significant impact on the number of abatements conducted as the result of promulgating hazard
levels under Section 403. When abatement decisions are constrained by income limits, the
number of abatements undertaken falls substantially, raising the possible dilemma that some
necessary abatements will not be undertaken because of income constraints.
However, determining the ultimate implications for equity is more complicated. The
abatements that are not affordable tend to be the ones that have negative net benefits. Although
unaffordable, such abatements are already questionable from the perspective of social welfare.
Net benefits actually rise, and in certain cases become positive, when the number of abatements
are constrained by limits on income. This finding underscores how any policy intended to
address income obstacles to abatement should be paired with an effort to fit each house with a
suitable abatement choice. This analysis also shows that if subsidies were used to enable
abatements with positive net benefits that would otherwise be prohibited by income constraints,
the funding needs could be significant. Approximately $45 to $319 million would have to be
spent annually over SO years.
Abt Associates, Inc. xix Dmft January ]0, 1994
-------
ES.9 CONCLUSION
The highest net benefits identified in this analysis derive from determining the best
abatement for each individual home. Once testing is taken into account, net benefits of $10
billion are possible. This approach, known as the voluntary optimum, takes the concept of
hazard levels to the extreme since it is essentially equivalent to setting hazard levels for each and
every combination of paint condition and soil, dust, and paint levels, among other factors.
Among the class of more feasible sets of hazard levels, the primary candidate for
consideration is a single dust hazard level of 1,200 ppm, given the assumption that an additional
recommendation will be made to homeowners that non-intact lead-based paint be abated. This
hazard level generates the highest net benefits among all of the alternative sets of hazard levels.
Although the highest, the overall net benefits of this hazard level are negative (-$21 billion) once
testing costs of $24 billion are taken into account. Slightly lower net benefits can be achieved
by combining this dust hazard level with a soil hazard level of 2,300 ppm and a paint hazard
level of 20 mg/cm2. For any of these sets of candidate hazard levels it appears that the net
benefits are substantially lower because of the abatement aimed at non-intact lead-based paint.
The support for these specific hazard levels is very consistent across different decision
frameworks within the main analysis but the range of possible hazard levels may be broadened
once the results of sensitivity analyses or other decision factors are taken into account. The
sensitivity analyses indicate that the optimal dust hazard level could be as low as 300 ppm and
that the optimal soil hazard level could be as low as 1,000 ppm. The validity of these lower
estimates hinges on the weight given to the alternate assumptions made in the sensitivity
analyses. Other alternative hazard levels besides those cited above may also be contenders if
other criteria besides economic efficiency will be considered in EPA's decisionmaking. One
such possibility is to choose hazard levels to keep the risk of exceeding a blood lead
concentration of IS ug/dl below 5%. This analysis indicates that taking this approach may be
costly. The net benefits of one such set of hazard levels (soil =500/dust=400) are negative
(-$19 billion) before testing costs are included.
Abt Associates, Inc. XX Dmftt January JQ, 1994
-------
1. INTRODUCTION
1.1 PROVISIONS OF RULE
Section 403 in Title IV of the Toxic Substances Control Act (TSCA) directs EPA to
promulgate regulations that identify lead-based paint hazards, lead-contaminated dust, and lead-
contaminated soil. Section 403 is one portion of the Residential Lead-Based Paint Hazard
Reduction Act of 1992 (the Act), which requires that the United States Environmental Protection
Agency (EPA), the Department of Housing and Urban Development (HUD), and other Federal
Agencies develop a national strategy to build the infrastructure necessary to eliminate lead-based
paint hazards in all housing. Clearly identifying what constitutes a lead-based paint hazard is
an important step in encouraging effective action to evaluate and reduce the lead-paint hazards
in the Nation's housing stock.
The Act establishes Federal grants and other programs that create a partnership among
all levels of government and the private sector in order to best mobilize national resources to
reduce lead-based paint hazards. Many of the activities involving the identification of lead-based
paint hazards in the Act (including inspections and risk assessments by appropriately trained and
certified personnel, as well as abatement of any hazards identified) are voluntary. However, in
certain circumstances the activities are required. Section 403 is consistently used throughout the
Act to identify those paint, soil and dust conditions that are affected by the Federal and state
programs included as a part of the national strategy to eliminate the hazards of lead-based paint
in residences. Exhibit 1.1 summarizes the provisions of the Act that directly rely on the §403
identification of a hazard.
1.2 STATUTORY AUTHORITY
On October 28, 1992 Congress enacted the Housing and Community Development Act
of 1992, which includes 16 separate Tides. Title X of that Act is entitled the Residential Lead-
Based Paint Hazard Reduction Act of 1992. Title X is composed of five subparts, including
Subtitle B which amends TSCA by adding a new Title IV-Lead Exposure Reduction. TSCA
Title IV includes twelve sections, from §401 through §412. Section 403 - Identification of
Dangerous Levels of Lead is very brief; the complete text reads as follows:
"Within 18 months after the enactment of this title, the Administrator shall promulgate
regulations which shall identify, for purposes of this title and the Residential Lead-Based
Paint Hazard Reduction Act of 1992, lead-based paint hazards, lead-contaminated dust,
and lead-contaminated soil."
The Section 403 identification of lead-based paint hazards is used not only in Title IV of
TSCA, but throughout the Act as the basis for determining the appropriate response to the
Abt Associates. Inc. 1-1
January ]Q
-------
Exhibit 1-1
Relationship of §403 Identification with Other Provisions of the Lead-Based Paint Hazard Reduction Act
Section
§101 l(a)
Affected Housing Stock or Entity
Affordable non-public housing that is not
federally owned or assisted housing
Relationship
§403 Identification used to establish eligibility for
receiving HUD grants for interim controls or abatement
of lead-based paint hazards.
§1012
Various housing receiving assistance under the
Cranston-Gonzalez National Affordable
Housing Act
1) §403 Identification used to require reduction of
hazards in course of rehabilitation projects receiving less
than $25,000 per unit in federal funds, and abatement of
hazards in rehabilitation projects receiving more than
$25,000 per unit.
2) §403 Identification used to establish eligibility for
receiving federal funds for interim controls or abatement
of lead-based paint hazards.
3) §403 Identification used to establish eligibility for
including inspection and abatement costs in determining
maximum monthly rents in federally assisted rental
property.
§1013
Federally owned housing being sold
1) Housing built prior to 1960: Inspection and
REQUIRED abatement of lead-based paint hazard (as
identified by §403).
2) Housing built between 1960 and 1978: Inspection and
written notification to buyer of all lead-based paint
hazards (as identified by §403)
§1014
Low-income housing units under jurisdiction of
Cranston-Gonzalez National Affordable
Housing Act.
§403 Identification used to estimate number of housing
units in a jurisdiction occupied by low-income families
that have a lead-based paint hazard. Information shall
be used in preparing a housing strategy .
§1015
Private housing.
§403 Identification used by Inter-Agency Task Force to
recommend programs and procedures for financing
inspections and abatements.
-------
Section
§1017
§1018
§1021
TSCA Title IV,
§402
§1021
TSCA Title IV,
§405
§1021
TSCA Title IV,
§406
§1021
TSCA Title IV,
§408
Affected Housing Stock or Entity
Federally supported inspections, risk
assessments, interim controls and abatements
Sale or lease of all housing stock constructed
before 1978.
Persons offering to eliminate lead-based paint
hazards.
Information on identifying and eliminating
lead-based paint hazards.
Lead Hazard Information Pamphlet.
All executive, legislative and judicial branches
of the federal government having jurisdiction
over property, or engaged in activities that may
result in a lead-based paint hazard.
Relationship
§403 Identification used in Guidelines for conducting
federally supported lead-based paint hazard reduction.
Requires notification to buyer of any known lead-based
paint hazards (as identified by §403). Buyer has right to
perform inspection before being obligated by contract for
sale or lease.
Training and certification requirements for all persons
involved with identifying and eliminating lead-based
paint hazards (as identified by §403).
1) Clearinghouse and hotline to provide information on
identifying, reducing and eliminating lead-based paint
hazards (as identified by §403).
2) Establish protocols and performance characteristics
for products sold to reduce or eliminate lead-based paint
hazards (as identified by §403).
Required pamphlet to explain lead-based paint hazards
(as identified by §403) and approved methods to
eliminate those hazards.
All requirements in Lead-Based Paint Hazard Reduction
Act of 1992 shall apply to all federal departments,
agencies and instrumentalities.
Abt Associates, Inc.
1-3
Draft, January 10, 1994
-------
existence of a lead-based paint hazard. Section 1004(15) defines a lead-based paint hazaid as:
"The term "lead-based paint hazard" means any condition that causes exposure to lead
from lead-contaminated dust, lead-contaminated soil, lead-contaminatedpaint that is
deteriorated or present in accessible surfaces, friction surfaces, or impact surfaces that
would result in adverse human health effects as established by the appropriate Federal
agency."
This definition is repeated in TSCA Section 401(10), except the responsibility to establish what
constitutes a hazard is clearly given to EPA;
"...adverse human health effects as established by the Administrator under this Title."
The proposed Section 403 identification of lead-based paint hazards is used throughout the Act
to establish requirements and eligibility for programs dealing with the national strategy for
reducing the risks of lead-based paint.
Section 1004 (16 & 17) goes on to define lead-contaminated dust as:
"surface dust in residential dwellings that contains an area or mass concentration of lead
in excess of levels determined by the appropriate Federal agency to pose a threat of
adverse health effects in pregnant women or young children"
and lead-contaminated soil as:
"bare soil on residential real property that contains lead at or in excess of the levels
determined to be hazardous to human health by the appropriate Federal agency."
1.3 PURPOSE AND CONTENTS OF REPORT
The purpose of this regulatory impact analysis is to evaluate the effects of defining
various lead hazard levels in paint, soil and dust. The primary impacts are the costs of lead
abatements conducted in response to the regulation and the health benefits that accrue to children
from a reduced exposure to lead. The report follows the standard outline for a regulatory impact
analysis. Chapter 2 describes past regulatory actions to reduce risks from lead. Chapter 3
details the method used to evaluate the risks to children from lead exposure and explains the
method for quantifying the benefits of reduced exposure. This chapter also defines the market
failure that indicates a need for federal regulation, presents regulatory options that might be
considered, and gives an overview of the analytic approach. The costs of lead testing and
abatement are shown in Chapter 4 along with the total cost results for the regulatory options
under consideration. Chapter 5 presents the quantified benefits of the regulatory options. The
benefit-cost analysis in Chapter 6 explains how the regulatory options were identified based on
the value of the net benefits and presents results for each of the possible options. Chapter 7
presents sensitivity analyses to characterize the model uncertainties. The final chapter (Chapter
8) indicates the data available for evaluating the impact of this regulation on small entities
(businesses and governments), discusses the regulatory impacts on trade, technological
innovation and equity, and presents equity analyses for adults and children.
Abt Associates, Inc. 1-4 Draft, January 10, 1994
-------
2. REGULATORY BACKGROUND
2.1 INTRODUCTION
2.1.1 Lead as a Public Health Problem
Exposure to lead is one of the most serious public health problems currently facing the
United States (ATSDR, 1988). Lead's advantages, including its malleability, resistance to
corrosion, good insulation, and low cost, have made lead attractive for many applications; lead
has been used in gasoline, ceramics, paint, and several other products. These uses have resulted
in lead's release to and distribution in all environmental media, which has complicated efforts
at reduction (ATSDR, 1988). Much of lead in the environment is accessible to humans through
a variety of exposure pathways, and since it does not degrade, continued use of lead results in
accumulation in the environment. Human exposure to lead is of concern because it interferes
with the normal functioning of cells causing a range of toxic effects in the nervous, red blood
cell, and kidney systems (ATSDR, 1988). Fetuses and young children exposed to lead are
especially at risk from damages to the developing brain and nervous system (CDC, 1991).
Knowledge of some of lead's negative health effects dates back about 2000 years.
Reproductive and developmental effects of lead were recognized in the 18th and 19th centuries
in the United States in female lead workers and wives of lead workers. These women
demonstrated problems including sterUity, spontaneous abortion, stillbirth, and premature
delivery, and their offspring exhibited high mortality, low birth weight, convulsions and other
effects. This recognition resulted in better industrial hygiene which in turn, reduced
reproductive problems (ATSDR, 1988).
The prevalence of direct lead poisoning in children was first examined in Australia in the
1890s and traced to exterior lead-based paint (ATSDR, 1988). In the U.S., physicians
eventually defined lead poisoning in children as a clinical entity in the early 20th century after
a study reported that lead caused acute encephalopathy in a number of children. In the 1930s
and 1940s, epidemiologic data on childhood lead poisoning began to expand and accelerated
through the 1960s. Rudimentary screening of children in the 1950s and 1960s clearly showed
that they were being exposed to excessive amounts of lead. Prevalence of lead poisoning was
especially high among inner city youth. More massive screenings in the 1970s resulted in the
recognition of lead poisoning as a widespread public health problem (ATSDR, 1988).
Lead exposure's prominence as a public health concern has been due to increased blood
lead levels. Although the average blood lead levels in the U.S. population are estimated to have
dropped in the last two decades (U.S. EPA 1989b, 1991a), levels are about 15 to 30 times
higher in some U.S. populations than the pre-industrial average of about 0.5 jig/dl (ATSDR,
17OO).
The recognition of lead's adverse effects has resulted in lowering the blood lead level that
triggers medical intervention. In 1970, the U.S. Public Health Service published guidelines that
set the level at 60 Mg/dl (CDC, 1991). Shortly thereafter, the CDC set the guidelines at 40
/ig/dl, then revised the recommendations to 20 jig/dl, and finally to set them at the current level
of 10-14 jig/dl in 1991. Levels higher than this range trigger various intermediate actions; a
Abt Associates, Inc. 2-1 Draft. January 10, 1994
-------
child with a blood lead level between 15-19 pg/dl should have nutritional and education
interventions; and a blood lead level greater than 20 pg/dl should prompt medical evaluations
and environmental investigations (CDC, 1991).
The following sections focus on regulations designed to decrease exposure to lead.
Section 2.2 discusses laws and regulations designed to decrease exposure to lead in a variety of
media, including products, releases, and past use of lead in plumbing. Section 2.3 focuses
solely on efforts at the federal, state and local levels to decrease exposure to lead remaining in
residential areas (including lead in paint, dust, and soil).
2.2 REGULATION OF LEAD PRODUCTS, ENVIRONMENTAL AND WORKPLACE
RELEASES OF LEAD, AND LEAD IN DRINKING WATER
Lead content in some products has been prohibited or restricted. Also, environmental
releases to air and water and in waste have been controlled. OSHA has set limits on workplace
concentrations. In addition, efforts have been made to remediate exposure to lead already in the
environment from its use in drinking water systems.
2.2.1 Lead in Paint
In the 1950s, the paint industry voluntarily restricted sale of paint with lead content
greater than one percent (Mushak and Crocetti, 1990). Subsequently, the Lead-Based Paint
Poisoning Prevention Act, enacted in 1971, prohibited the use of paint with greater than one
percent lead (by weight of nonvolatile solids) in certain federally-owned or federally-assisted
housing (HUD, 1990). As a result of 1976 amendments to this Act, lead paint was redefined
as paint containing more than 0.06% lead by weight (HUD, 1990). In 1978, the U.S. Consumer
Product Safety Commission banned both the sale of such lead-based paint to consumers and its
use in residences or on other consumer-accessible surfaces (16 CFR 1303).
2.2.2 Lead in Gasoline
The Clean Air Act of 1970 (CAA) first controlled the use of lead in gasoline because it
rendered catalytic converters inoperative. In response, the use of lead in gasoline declined
significantly during the 1970s. In 1986, the U.S. acted to phase out the use of lead in gasoline
entirely (51 FR 24606). Currently, the U.S. restricts the amount of lead allowed per liter of
leaded gasoline to 0.026 grams. Effective in 1988, the United States also required all new light
duty vehicles and trucks, motorcycles and heavy duty gasoline engines to operate on unleaded
gasoline. (Unleaded gasoline is defined as gasoline containing no more than 0.01 g/1 of lead.)
In addition, as of 1992, motor vehicle engines and non-road engines that require leaded gasoline
are prohibited. As of December 31, 1995, a total ban on leaded gasoline and lead gasoline
additives for highway use will be in place.
2.2.3 Other Lead-based Products
The U.S. canning industry has voluntarily phased out the use of lead solder in food cans
since alternative, affordable processes for sealing the seams of tin containers are available (FDA,
1992a). U.S. industry and the U.S. Food and Drug Administration have also undertaken efforts
Abt Associates. Inc. 2-2 Draft. January 10. 1994
-------
to control lead exposure from ceramic ware (FDA, 1991), foil on wine bottles, and crystalware
(FDA, 19925). Eight U.S. states have adopted legislation to limit the levels of lead in packaging
materials (EPA, 19915).
2.2.4 Environmental and Workplace Releases of Lead
Under authority of the Clean Air Act, EPA has established standards of performance
designed to limit emissions of air pollutants from lead smelting and processing facilities. In
addition, lead emissions from these and other industries are controlled via facility-specific
permits written by states. These permits are designed to reduce emissions to the extent needed
to meet EPA's national ambient air quality standard for lead of 1.5 /tg/m3 (quarterly average)
established in 1978 (43 FR 46246). '
Under the Clean Water Act, federal effluent guidelines and pretreatment limits for lead-
containing effluents have been established for over 20 industries. These limits help achieve
state-promulgated surface water quality standards (which may be based on water quality criteria
published by EPA). The effluent limits are implemented by states through facility-specific
permits and, depending on state water quality standards, may be more stringent than federal
effluent requirements.
Releases of lead as solid waste are regulated under the Resource Conservation and
Recovery Act (RCRA). A waste is defined as hazardous if, when tested, the leachate from the
waste contains more than 5 ppm (40 CFR 261.24). In addition, certain lead-containing wastes
are separately listed as hazardous wastes. All of these wastes must be properly managed and
disposed (40 CFR 260-270).
EPA has also initiated a voluntary program to reduce lead emissions, based on the Toxics
Release Inventory reporting. This project, called the "33/50 Program", encourages industry to
curtail emissions of 17 toxic pollutants, including lead (U.S. EPA, 1992). The specific aim of
the project is to obtain commitments from companies to voluntarily reduce emissions, effluents
and off site transfers of the subset of the 17 pollutants that are applicable to their operations in
two phases ~ 33 percent by 1992 and 50 percent by 1995 - using 1988 as the baseline year.
As of 1992, 850 companies had agreed to participate in this program.
Efforts to reduce exposure to lead releases in the workplace have included setting
permissible workplace air concentrations of lead and permissible blood lead levels in workers
(Niemeier, 1991). The current Permissible Exposure Limit (PEL) is 50 jig/m3 for most
industries except the construction industry (OECD, 1993). Under the Residential Lead-based
Paint Hazard Reduction Act, passed in October 1992, OSHA is required to issue interim
regulations lowering the limit for the construction industry (ACELP, 1993).
2.2.5 Lead in Drinking Water
Exposure to lead in drinking water has continued because of past use of lead in plumbing.
EPA has acted to reduce these exposures through recent comprehensive measures (Mushak and
Crocetti, 1990). In rules promulgated in 1991, the U.S. EPA outlined new treatment
requirements for drinking water systems (56 FR 26460). The regulation requires tap water
Abt Associates. Inc. 2-3 Draft, January 10. 1994
-------
sampling from high risk homes (e.g., lead service lines or lead soldering installed since 1982).
If at least 10 percent of home tap samples exceed IS /xg/1 (the "action level"), corrosion control
treatment and public education is required. Replacement of lead service lines is required if
corrosion control fails to bring water lead levels below the "action level." EPA has also issued
a maximum contaminant level goal of zero for lead in drinking water.
2.2.6 Resulting Reduction in Blood Lead Levels
These regulations and other efforts have clearly reduced exposure and blood lead levels
significantly. Although no recent national study has measured human blood lead, it is estimated
that average concentrations have dropped over the last two decades from about 15-20 pg/dl to
approximately 5 pg/dl (EPA, 1991a). In particular, reductions of lead in gasoline have
contributed dramatically to reductions hi blood lead levels. Several studies have specifically
examined the relationship between blood lead and the lead content of gasoline, and have found
a strong positive correlation (Schwartz and Pitcher, 1989, Rabinowitz and Needleman, 1983).
Annest et al. (1983) noted a 37 percent drop in blood lead levels from 1976 to 1980 correlated
with a reduction hi gasoline lead, while Schwartz and Pitcher (1989) estimated that as much as
SO percent of blood lead hi the U.S. hi the late 1970's may have been attributable to lead in
gasoline.
Reductions in dietary lead have also contributed to declining exposures. Dietary lead
intake for a two-year-old child has dropped from about S3 /tg/day hi 1978 to an estimated 13.1
/xg/day in 198S; comparable declines have been seen hi adults (U.S. EPA, 19895). These trends
are attributable to the reduction in gasoline lead emissions (and resulting reductions hi deposition
of lead from air to soil) and the voluntary phaseout of lead-soldered cans by U.S manufacturers
since the 1970s. It can be calculated that these changes hi lead exposure from food have led to
reductions of 1-2.S /ig/dl in average blood lead levels (U.S.EPA, 1989b).
2.3 EFFORTS TO REDUCE LEAD-BASED PAINT, DUST AND SOIL IN
RESIDENTIAL AREAS
One of the largest remaining lead exposure sources for children is existing reservoirs of
lead-based paint, dust and soil present in many residential areas (ATSDR, 1988). In an effort
to reduce exposure to residential lead hazards, regulatory efforts have been increasing for
several years to address these hazards.
2.3.1 Current Estimates of Exposure
Although new paint containing lead was banned for use in residences in 1978, exposure
to existing lead-based paint has continued due to prior uses in residential and other buildings.
In addition, leaded house paint can contribute to lead hi ulterior dust and soil. There also
remains a significant soil burden of lead from leaded gasoline and lead smelter emissions.
Several studies have demonstrated positive correlations between blood lead levels and
lead in paint, soil and dust (Gilbert et al., 1979 as cited in CDC, 1991; Charney et al., 1980,
Chamey et al., 1983, and Bellinger et al., 1986 as cited in HUD, 1990; Clark et al., 1991).
Exposures are especially high in children. In a 1988 Report to Congress on the extent of lead
Abt Associates, Inc. 2-4 Draft. January 10, 1994
-------
poisoning in children, ATSDR stated that the existing leaded paint in U.S. housing and public
buildings is "an untouched and enonnously serious problem" (ATSDR, 1988). The Centers for
Disease Control conveys the seriousness of home lead exposure as a contributor to elevated
childhood blood lead by stating that lead poisoning exists in our society primarily because of
exposure in the home (CDC, 1991). However, since only about 5 percent of children are
screened, most children with lead poisoning probably are not identified. Infants and toddlers
are especially susceptible to lead in the home because they may ingest lead paint chips, dust and
soil and because of the way they metabolize lead. Older children, up to at least 8 years old, are
also at increased risk (ATSDR, 1988).
Exposure continues mainly from paint in older homes, since houses built after 1978 are
presumed to be free of lead paint. A 1987 HUD survey of 284 homes built before 1980
indicated that, for privately owned houses, an estimated 57.4 million (74%) of all pre-1980
private homes have some lead-based paint (HUD, 1990). Of these units, an estimated 9.9
million units have families with children younger than seven years old. These units (with both
lead-based paint and young children) represent 71 percent of all pre-1980 housing units occupied
by families with children under seven. Houses built before 1940 have the highest prevalence
of lead-based paint in either the interior or exterior (90% of all pre-1940 houses), whereas 62 %
of the houses built between 1960 and 1979 contain lead-based paint. Of the homes with lead
paint, 3.8 million homes in which young children live have peeling paint or excessive lead-
containing dust.
A significant number of public housing units also contain lead paint. A small survey of
public housing conducted in 34 cities in 1984 and 1985 show that in housing built before 1950,
81 percent of the units sampled contained lead-based paint, whereas a smaller proportion (48%)
of the sampled units built between 1960 and 1977 had leaded paint (HUD, 1990). Preliminary
results from the public housing portion of the HUD survey conducted in 1987 indicate a greater
percent of public housing with lead paint (HUD, 1993); about 91 % of the sample of 97 public
housing units investigated had lead-based paint somewhere in the interior or exterior of the unit,
although many with lead-based paint had fairly low levels (HUD, 1993). Families of all
socioeconomic classes may live in older housing, and thus be exposed to lead paint (ATSDR,
1988). However, families with the lowest incomes are disproportionately found in older housing
(ATSDR, 1988). *
2.3.2 Federal Regulatory Activities to Decrease Exposure to Lead-Based Paint in Existing
Housing
Federal regulatory efforts and guidelines to limit exposure to lead-based paint in the
existing housing stock have evolved over the past twenty years. The following two sections
chronicle these activities in detail.
The Lead-based Paint Poisoning Prevention Act and Amendments
The Lead-based Paint Poisoning Prevention Act of 1971 (LBPPPA) and subsequent
amendments (1973, 1976, 1987, and 1988) have resulted in a number of federal regulatory
activities to reduce exposures and risks from lead paint in housing. In addition to setting limits
on the use of lead paint as described above, the Act established grants for lead-poisoning
Abt Associates, Inc. 2-5 Draft, January 10, 1994
-------
screening and treatment, and required a report to Congress on methods of abatement (HUD
1990).
Abatement of Lead-based Paint Hazards in Federally-associated, Public and Indian
Housing. The 1973 amendments required HUD to eliminate, as much as was practical, hazards
of lead-based paint poisoning in pre-1950 housing covered by housing subsidies and applications
for mortgage insurance and in all pre-1950 federally owned housing prior to sale. HUD acted
by issuing regulations to warn tenants and purchasers of HUD-associated housing of the
"immediate hazard" of lead-based paint (defined as conditions associated with deteriorating lead
paint surfaces). A1983 court action resulted in broadening the definition of "immediate hazard"
to include intact paint; this definition was subsequently signed into law in 1987. In regulations
issued by HUD in 1986 and 1987, the construction cutoff date was changed from 1950 to 1973
in most cases. HUD again changed the cutoff date in response to 1987 amendments to the
LBPPPA; the new date became 1978 for all programs (HUD, 1990). The 1988 Amendments
to the LBPPPA specified the level which defines a lead paint surface as 0.5% by weight or 1.0
mg/cra2 (AECLP, 1993). HUD has also promulgated rules to eliminate lead paint hazards in
public and Indian housing (Mushak and Crocetti, 1990). In these types of units that have
children younger than 7 years old, inspections for defective paint surfaces are required; if a child
has an elevated blood lead level, then the house must be inspected for chewable and defective
surfaces, and abatement is required in dwellings, common areas, or public child care facilities
within the public housing.
Grants. The LBPPPA authorized funding for States and cities to conduct extensive
screening programs to identify lead-poisoned children, refer them for medical treatment,
investigate their houses for lead, and require abatement (HUD, 1990).
Research and Reports to Congress. The 1971 LBPPPA required a report to Congress
on the "nature and extent of the problem of lead-based paint poisoning" and methods of removal.
Then, the 1987 amendments required an extensive research and demonstration project on lead-
paint testing and abatement technologies in HUD-owned housing, as well as additional reports
to Congress (HUD, 1990). In response to another mandate of the 1987 amendments, HUD
conducted a survey of the distribution of lead-based paint in the nation's housing stock and
submitted a report on the results for privately-owned housing to Congress in a comprehensive
plan for abating paint in private housing. Additional amendments in 1988 required a
demonstration of abatement techniques in public housing as well as a comprehensive plan to
address abatement in public housing (HUD, 1990).
Interim Guidelines. In response to a need for better guidance on testing, abatement,
remediation, disposal, and worker protection, HUD published interim guidelines related to these
activities and issues in 1990; these guidelines were specifically related to the concerns of public
housing agencies. The guidelines have been used subsequently in the abatement demonstration
in public housing (HUD, 1990; EPA, 1993).
The Residential Lead-based Paint Ha^rd Reduction Act
The most recent statutory activity related to the reduction of lead paint hazards is the
enactment of the Residential Lead-based Paint Hazard Reduction Act in October of 1992. Also
Abt Associates, Inc. 2-6 Draft, January 10. 1994
-------
known as Title X of the Housing and Community Development Act of 1992, this Act amends
sections of the LBPPPA and adds a new section (Title IV) to the Toxic Substances Control Act
(TSCA), in addition to other important new provisions. Described as "the most comprehensive
and significant lead poisoning prevention legislation in more than two decades" (AECLP, 1993),
the Act aims to provide attainable goals for reducing lead hazards in residential settings by
targeting specific housing in the greatest need of abatement (AECLP, 1993).
Federalty-owned and assisted housing. Title X allows for more targeted lead hazard
evaluation and reduction activities in federally-owned and assisted housing (AECLP, 1993).
Whereas provisions under the 1987 amendments to LBPPPA indicated that any and all houses
built before 1978 that contain lead-based paint constitute hazards that may be acted upon, Title
X provides a more strategic approach to reducing the hazards from lead-based paint. This
approach involves requirements for risk assessments, inspections, and interim controls for pre-
1978 housing (targeted housing) and also requires deadlines for action. Title X also extends
federal lead-based paint requirements to all housing that receives more than $5,000 in project-
based assistance under any federal housing or community development program (in addition to
the federally assisted and insured houses covered under previous Acts) (Section 1012); inclusion
of these houses significantly expands the universe of federally-insured and assisted housing
subject to lead-based paint related requirements (AECLP, 1993).
Additional provisions apply to federally-owned housing being sold (AECLP, 1993).
Properties built prior to 1978 must be inspected and their condition disclosed to the prospective
buyer. Units built before 1960 that have lead-based paint (defined as priority housing) must be
abated (Section 1013).
Private housing. Private housing has received greater focus under Title X than under
LBPPPA. Although states, local governments or common law still determine whether landlords
provide safe housing, Title X includes several features to encourage evaluation and reduction of
lead-based paint hazards in private housing. First, Title X formalized into law a grant program
run by the Department of Housing and Urban Development for reducing lead-based paint
hazards in low-income privately owned housing. Grants awarded to state and local governments
for this purpose include $47.7 million for 1992 and $100 million for 1993 (Section 1011). Title
X authorizes an increase of $250 million in grants for 1994, to be determined by subsequent
Appropriations Acts (AECLP, 1993).
Other Title X provisions also affect targeted private housing (AECLP, 1993). These
mandates include integration of lead hazard evaluation and reduction into local housing
programs, and certain disclosure and warning requirements to be met at the time of sale or rental
of any pre-1978 housing unit (Sections 1014 and 1018). The Act also requires establishment of
a national task force on lead-based paint hazard reduction and financing; this group is to be made
up of an array of groups involved in housing, real estate, insurance, lending, abatement, and
other groups (Section 1015).
Safety of residents and workers. This law requires promulgation of a number of
regulations addressing the safety of workers undertaking interventions and safety of families who
live or will live in treated housing. Section 1021 amends TSCA by adding a new Title IV,
which primarily addresses EPA requirements for contractor training and certification. This
Abt Associates, Inc. 2-7 Draft, January JO, 1994
-------
regulatory analysis supports the development of the regulation that responds to TSCA § 403 (in
§ 1021); this regulation requires EPA to define a "lead-based paint hazard" and dangerous levels
of lead in dust and soil. EPA must also set standards of minimum performance for lead-based
paint activities and ensure that individuals engaged in activities are trained, that training
programs are accredited, and that contractors are certified (TSCA § 402). HUD and EPA are
to assist in funding state certification and training programs and to issue standards for a model
state program (TSCA § 404). In addition, EPA must assure that a program is in place to certify
environmental sampling laboratories and must provide for development of products and devices
for testing and abatement (TSCA § 405). To further protect abatement workers (and other
construction workers), OSHA is required to issue interim final regulations on the maximum
permissible limit of lead in air at construction sites (§ 1031 and 1032).
Education regarding lead paint hazards. Title X also mandates a variety of public
educational efforts. A hotline designed to inform the public about lead hazards was set up soon
after passage of the 1992 Act. The National Clearinghouse on Childhood Lead Poisoning was
then established in April, 1993. The Consumer Product Safety Commission, in coordination
with EPA, is also developing educational materials such as information to be displayed by
hardware stores that sell paint removal products (AECLP, 1993).
Research and development. A variety of research is also required under Title X. EPA
is required to conduct a study on the hazard potential of renovation and remodeling and must
publish results by April, 1995 (Section 1021: TSCA 402). The new TSCA Section 405 requires
a study on the methods to reduce occupational lead exposures and a study of the sources of lead
exposure in children to be issued as a report to Congress. Section 1051-1053 of Title X requires
research and evaluation on various lead-based paint testing and abatement topics; five million
dollars is appropriated for this research in each of the years 1993 and 1994.
2.3.3 Federal Guidelines and Other Activities Related to Lead in Soils and Dust
Guidelines for levels of lead in soil and dust
As mentioned above, EPA is required to determine dangerous levels of lead in dust and
soil under Title X. In the meantime, however, interim guidelines for abatement of lead-based
paint in housing set by the Department of Housing and Urban Development recommend
clearance levels for lead in household dust of 200 fig/ft2 for floors, 500 pg/ft2 for window sills
and 800 /tg/ft2 for window wells (HUD, 1990). No guidelines currently exist for residential
soils, but EPA has adopted interim guidance for levels to be attained at remediated hazardous
waste sites. The interim guidance recommends soil concentrations between 500 and 1000 mg/kg
(EPA, 1989a).
Other activities
Under authority of Title m of the 1986 Superfund amendments, EPA has funded projects
in Boston, Baltimore, and Cincinnati to test the health effects of abating soil with high lead
content in residential areas. This research is being considered in developing the final guidance
on soil clearance levels.
Abt Associates, Inc. 2-8 Draft, January 10, 1994
-------
2.3.4 State and Local Programs to Reduce Exposure to Lead-based Paint, Dust and Soil
Activity to address lead-based paint hazards has recently increased at die state and local
levels, although certain areas (e.g., Baltimore) have had programs in place for many years. The
following sections describe typical features in lead poisoning prevention projects and discuss
notable state and local systems.
State activities
In 1991, CDC issued a policy statement on lead-poisoning prevention that included
several recommendations. A survey was then administered to state officials in 1992 to determine
their lead poisoning prevention and lead abatement activities and whether they had adopted the
CDC recommendations. States did not respond to every question in the survey. For one
question, thirty-seven of 46 responding states indicated that they were coordinating prevention
activities between housing and environmental agencies (Fischer and Boyer, 1993). Nineteen of
47 states had a program at the state level for monitoring health and environmental follow-up of
children with high blood lead levels. Twenty-four of 28 responding states reported the ability
to assure medical and environmental follow-up for more than 50% of children with blood lead
levels of 20 pg/dl and greater (Fischer and Boyer, 1993).
Several states have requirements and standards specifically related to lead abatement.
Seventeen states have authority to require abatement or remediation of lead hazards and eighteen
have adopted abatement standards (Farquhar and South, 1993). At least twelve states have
specific standards for soil (ranging from 100 ppm to 1000 ppm) and twelve states can require
abatement of lead in soils (Farquhar and South, 1993; Mn Dept. of fflth, n.d.). Table 2.1 lists
standards for paint, dust and soil adopted by several states; these standards represent levels that
trigger abatement or remediation and/or levels that must be achieved during abatement or
remediation. In cases where abatement has been required, states have consistently ordered
removal of paint up to five feet from the floor in order to protect children (HUD, 1990). The
principal sources of funding for abatement (as indicated by 23 states) have been local funds and
money spent by owners of property with lead hazards (Fischer and Boyer, 1993). However,
HUD grants for 1992 and 1993 are specifically slated for lead abatement in California,
Massachusetts, Minnesota, New Jersey, Rhode Island, and Wisconsin (NCLSH, 1993; AECLP).
Traditionally, state activities addressing lead-based paint and lead poisoning in children
have not been extensive. Blood screening has been the primary program activity of states,
usually provided by walk-in clinics or special screening campaigns. The extent of screening,
however, varies widely because of budget constraints (HUD, 1990). Environmental intervention
has often only occurred after identification of a poisoned child. In states that have authority to
require abatement, the authority usually occurs through negotiation so as not to cause undue
financial hardship. In addition, states have not usually provided funds for abatement activities
recommended as a result of medical and environmental intervention (HUD, 1990).
Abt Associates, Inc. 2-9 Draft, January 10, 1994
-------
Exhibit 2-1
Standards for Paint, Dust, and Soil Adopted by Selected States
Massachusetts
0.5% by weight
or
1.2 mg/cm2
200 (floor)
500 (window sill)
800 (window well)
1000
by
weight
105 CMR 460.020;
460.170
Maryland
0.5% by weight
or
0.7 mg/cm2
200 (floor)
500 (window sill)
800 (window well)
none
COMAR 26.02.07
Minnesota
0.5% by weight
or
1.0 mg/cm2
80 (floor)
300 (window sill)
500 (window well)
100 by
weight
Mn Dept of fflth,
1993
Rhode Island
0.05%
or 500 ppm
on non-intact
surfaces
200 (floor)
500 (window sill)
800 (window well)
1000
by
weight
R 23-24.6-PB
While most states generally have had limited programs, Maryland and Massachusetts
have traditionally had more extensive systems (HUD, 1990). In 1972, Massachusetts established
the first statewide program. Subsequently, Massachusetts has had some of the highest screening
penetration rates in the country (Prenney, 1987). The features that distinguish both the
Maryland and the Massachusetts programs include the following (HUD, 1990):
• Both states have a high level of interagency involvement which provides an
effective mechanism for policy development and implementation. Before formal
legislation was passed, each state formed a policy task force representing a cross-
section of agencies. This multidisciplinary approach was then written into
legislation in each state.
• Both states have methods of enforcing abatement requirements. In Massachusetts,
property owners who fail to comply with abatement orders are liable for actual
and punitive damages. Under Maryland's real property code, tenants may deposit
their rents in an escrow account held by the district court when landlords fail to
remove lead-based paint that is accessible to children.
• Both states provide some level of quality control over lead testers, abatement
contractors, and abatement inspectors. Massachusetts requires training and
licensing of abatement contractors and inspectors, and testing laboratories must
be certified. Maryland requires that workers be trained in safe and appropriate
abatement procedures (MDE, 1992) and has established a training program
employing private and public training organizations that are certified by the state.
Abt Associates, Inc.
2-10
Draft, January 10, 1994
-------
• Both states have loan or grant programs to provide abatement funds for property
owners with limited resources. Massachusetts has also established a $1,000 tax
credit for private property owners doing lead-based paint abatement. (A new bill
that has passed the Massachusetts house would increase the tax credit to $2,500
(Carroll, 1993).)
• Both have attempted to provide relocation resources for families during
abatement. However, the availability of suitable interim housing is a problem.
The State of Maryland has given the City of Baltimore a grant for "lead-safe"
houses to be used for transitional housing during abatement.
• Both states require that all cases of lead poisoning be reported to the state health
department. Private physicians must screen all preschool children for lead
poisoning and report cases of children with high blood lead levels to appropriate
authorities for followup. In 1991, a year after Massachusetts passed its regulation
requiring screening of all pre-school children, the screening rate of children aged
6 months to 6 years was 74% compared with 50% in 1989 (MDH, n.d.).
• Both states have legislation that calls for investigation, testing, and approval of
new abatement or containment technologies. Maryland has been involved in
ongoing research on the effects of lead dust on blood lead in abatement workers
and the development of testing protocols for encapsulation products. Maryland
has pioneered the development of standards and procedures for worker protection
during abatement, dust containment, and post-abatement cleanup, inspections, and
clearances. These methods have provided much of the basis for the National
Institute of Building Science's guidelines for testing and abatement of lead-based
paint in housing, which in turn, became the basis for the HUD interim guidelines
(HUD, 1990).
Rhode Island was able to draw from features of the existing Massachusetts lead poisoning
prevention and abatement program in addition to adding aspects of its own (Vanderslice, 1993).
Rhode Island's Lead Poisoning Prevention Act was passed in 1991. Under the law, the
Department of Health is authorized to expand blood lead screening to all children under six and
set up a public education program. Houses where children with blood lead levels of 25 pg/dl
or greater have been identified must be inspected, and nonintact lead hazards in these homes
must be abated (RTDH, 1993). The Rhode Island program differs from the Massachusetts
system by requiring comprehensive inspection of soil and exterior dust, whereas Massachusetts
requirements include only inspection of interior dust and paint (RTDH, 1993). Rhode Island also
generally requires less extensive clean-up (i.e., lead does not need to be removed if it is kept
intact within the home) (R 23-24.6-PB; 105 CMR 460.020; 460.170). However, a new bill
passed by the Massachusetts House would allow less expensive options for decreasing exposure
to lead paint (Carroll, 1993).
Several other states have notable programs for reducing lead-based paint hazards.
California initiated its activities in 1986 which included: screening of children in three
geographical areas, establishing a program to reduce exposure, reporting high blood lead levels
to the Department of Health Services, and submitting a policy report to the legislature on future
Abt Associates, Inc. 2-11 Draft, January 10, 1994
-------
lead poisoning prevention programs (Florini et al., 1990). Minnesota was one of three states
(including Massachusetts and Maryland) to upgrade its lead paint abatement practices at a time
when the standard practices had not been changed for 40 years (Farfel and Chisolm, 1990).
Other states, including Missouri, Louisiana, Vermont and New Hampshire, are in the process
of setting up childhood poisoning prevention programs (Farquhar, 1993).
Local Programs
Local lead-poisoning programs are similar to state programs in several organizational and
programmatic features (HUD, 1990). For instance, programs are usually located in the health
department. In addition, resources for carrying out local activities have been limited.
Differences between typical state schemes and selected city programs lie more in the
extent than in the substance of the activities (HUD, 1990). In general, city programs are more
focused and seem to receive higher priority, which may be due to the urgency of the lead-paint
problem in larger cities.
In the Comprehensive and Workable Plan for the Abatement of Lead-Based Paint in
Privately Owned Housing (HUD, 1990), the Department of Housing and Urban Development
outlined several distinguishing features of local programs as determined by investigation of ten
selected cities:
• A city that is governed both by local ordinances and state regulations for lead-
poisoning prevention and detection activities usually has local laws that are more
stringent than state laws and may supersede the state requirements.
• In addition to providing intervention after cases of lead poisoning have been
detected, local programs may require intervention as a result of targeted
inspection or tenant complaints. Several cities, including Baltimore, Chicago,
Louisville, New York, and Philadelphia, are authorized to take such preventive
measures.
• In general, the city programs show more cooperation and coordination between
agencies.
• City programs usually screen for high blood lead levels more systematically and
target high-risk areas for screening.
Under Title X, several cities and one county have recently received increased funding
for lead abatement. These localities include Boston, Baltimore, Cleveland, and Alameda County
in California (NCLSH, 1993). In addition, other recent funding, authorized by the 1986
amendments to Superfund, has been given to Boston, Cincinnati, and Baltimore to evaluate the
impact of residential lead contaminated soil abatement on children's blood lead levels (Weitzman
et al., 1991; Cook, 1993).
Of the childhood poisoning prevention programs on the local level, Baltimore has one
of the most extensive schemes. As early as 1951, the city banned the use of lead paint in the
Abt Associates, Inc. 2-12 Draft, January 10, 1994
-------
interior of residences (Mushak and Crocetti, 1990). When a lead-poisoned child is identified,
the health department must be notified and the housing unit is inspected (BCHD, 1993). When
the city inspects a building and finds a lead-based paint hazard, a violation notice is issued to
the property owner, who must abate the lead hazard (BCHD, 1993). Baltimore is currently
running a pilot program allowing alternative abatement strategies (instead of complete removal
of lead hazards) to make lead homes safe. These strategies involve stabilization of identified
lead-based paint hazards, with complete replacement required only for windows and certain
surfaces in poor condition. Dust samples are taken for two years after stabilization.
2.3.5 Benefits of Defining a Lead Standard for Faint, Dust, and Soil
Although several states and localities have taken action on lead-based paint, many have
no standards for paint, dust, and soil abatement (nor standards for any one of the three media).
In addition, of the states and local areas that do have standards, the level of paint, dust, and soil
considered unacceptable differs among states. By providing definitions at die federal level for
lead paint hazards and dangerous levels of lead in dust and soil, those states that do not have
standards may be prompted to adopt standards more quickly. In addition, the federal guidelines
will provide consistency between the states.
Abi Associates, Inc. 2-13 Draft, January 10, 1994
-------
2.4 REFERENCES
Alliance to End Childhood Lead Poisoning (AECLP). 1993. Understanding Title X: A
Practical Guide to the Residential Lead-based Paint Hazard Reduction Act of 1992.
AECLP, Washington, DC. January.
Annest, J., Piikel, J., Makuc, D., Neese, J. Bayes, D., and M. Kovar. 1983. Chronological
trend in blood lead levels between 1976 and 1980. New England Journal of Medicine
308:1373-1377.
Baltimore City Health Department (BCHD). 1993. Personal communication between Abt
Associates and Michael Wojtowycz, November 2.
Carroll, M. 1993. Easing law on lead paint approved. The Boston Globe, p. 35. July.
Centers for Disease Control (CDC). 1991. Preventing Lead Poisoning in Young Children.
U.S. Department of Health and Human Services, Public Health Service. Atlanta, GA.
October.
Clark, S., R. Bomschein, P. Succop, S. Roda, and B. Peace. 1991. Urban lead exposures of
children in Cincinnati, Ohio. Chemical Speciation and Bioavailability, 3(3): 163-171.
Farfel, M.R. and J.J. Chisolm, Jr. 1990. Health and environmental outcomes of traditional and
modified practices for abatement of residential lead-based paint. American Journal of
Public Health, 80 (10): 1240-1245.
Farquhar, D. 1993. Personal Communication with Doug Farquhar, National Conference of
State Legislatures. September 30.
Farquhar, D. and L. South. 1993. Lead Poisoning Prevention: Directory of State Contacts
1993. National Council of State Legislatures. Denver, CO. May.
Fischer, D.B. and A. Boyer. 1993. State activities for prevention of lead poisoning among
children-United States, 1992. Journal of American Medical Association, 269(13)-1614-
1616.
Florini, K.L., G.D. Krumbhaar, and E.K. Silbergeld. 1990. Legacy of Lead: America's
Continuing Epidemic of Childhood Lead Poisoning. Environmental Defense Fund,
Washington, DC. March.
Maryland Department of the Environment (MDE), Division of Lead Poisoning Prevention.
1992. Lead Paint Hazard Fact Sheet #1: The abatement of lead paint hazards. June.
Maryland Department of the Environment (MDE). 1988. COMAR 26.02.07: Procedures for
Abating Lead Containing Substances From Buildings. August 8.
Abt Associates, Inc. 2-14 Draft, January 10, 1994
-------
Massachusetts Department of Health (MDH), Childhood Lead Poisoning Prevention Program.
n.d. Statement describing the Massachusetts response to the 1991 CDC Statement.
Massachusetts Department of Public Health. 1992. 105 Code of Massachusetts 460.000: Lead
Poisoning Prevention and Control. April.
Minnesota Department of Health (Mn Dept. fflth). n.d. Summary of Rules Governing Lead
Abatement Methods and Standards for Lead in Paint, Dust, Drinking Water and Bare
Soil.
Mushak, P., and A. Crocetti. 1990. Methods for reducing lead exposure in young children and
other risk groups: an integrated summary of a report to the U.S. Congress on childhood
lead poisoning. Environmental Health Perspectives, 89:125-135.
Organisation for Economic Cooperation and Development (OECD). 1993. Risk Reduction
Monograph No. 1: Lead. Background and National Experience with Reducing Risk.
Paris: Environment Directorate, OECD. Document No. OCDE/GD(93)67.
Prenney, B. 1987. The Massachusetts lead program: moving toward phase 2. Prevention
Update. Developed by the Maternal and Child Health Consortium Project of AAUAP
and the National Coalition on Prevention of Mental Retardation. April.
Rabinowitz, M. and H. Needleman. 1983. Petrol lead sales and umbilical cord blood lead
levels in Boston, MA. Lancet, 8314/5 (1):63.
Rhode Island Department of Health (RIDH). 1993. Personal communication between Abt
Associates and Bob Vanderslice, Chief of the Office of Environmental Health Risk
Assessment, October 28.
Rhode Island Department of Health. 1993. R 23-24.6-PB: Rules and Regulations for Lead
Poisoning Prevention.
Schwartz, J. and H. Pitcher. 1989. The relationship between gasoline lead and blood lead in
the United States. Journal of Official Statistics, 5:421-431.
U.S. Department of Health and Human Services, Public Health Service, Agency for Toxic
Substances and Disease Registry (ATSDR). 1988. The Nature and Extent of Lead
Poisoning in the United States: A Report to Congress. July.
U.S. Department of Health and Human Services, Public Health Service, Centers for Disease
Control (CDC). 1991. Strategic Plan for the Elimination of Childhood Lead Poisoning.
February.
U.S. Department of Health and Human Services, Public Health Service, Food and Drug
Administration (FDA). 1991. Fact Sheet: Lead in China Dishes Lawsuit in California;
Lead in Solder Used in Cans, Crystal, and Food Wrappers. November 18.
Abt Associates, Inc. 2-15 Draft, January 10, 1994
-------
U.S. Department of Health and Human Services, Public Health Service, Food and Drug
Administration (FDA). 1992a. Personal communication with Dr. Michael Bolger,
February, 1992.
U.S. Department of Health and Human Services, Public Health Service, Food and Drug
Administration (FDA). 1992b. Statement by Michael R. Taylor, Deputy Commissioner
for Policy, Food and Drug Administration, Public Health Service, Department of Health
and Human Services Before the Ad Hoc Subcommittee on Consumer and Environmental
Affairs, Committee on Government Affairs, U.S. Senate. March 27.
U.S. Department of Housing and Urban Development (HUD). 1990. Comprehensive and
Workable Plan for the Abatement of Lead-Based Paint in Privately Owned Housing:
Report to Congress. Washington, DC. December.
U.S. Department of Housing and Urban Development (HUD). 1993. Personal Communication
with Steven Weitz. September 29.
U.S. Environmental Protection Agency (EPA). 1989a. Interim Guidance on Establishing Soil
Lead Cleanup Levels at Superfund Sites. Office of Solid Waste and Emergency
Response. OSWER Directive Number 9355.4-02.
U.S. Environmental Protection Agency (EPA). 1989b. Review of the National Ambient Air
Quality Standards for lead: Exposure Analysis Methodology and Validation. Office of
Air Quality Planning and Standards. Research Triangle Park, N.C. June.
U.S. Environmental Protection Agency (EPA). 1991a. U.S. Environmental Protection Agency
Strategy for Reducing Lead Exposures. February 21.
U.S. Environmental Protection Agency (EPA). 1991b. Summary of Public Comments:
Comprehensive Review of Lead in the Environment under TSCA. November 22.
Prepared for the Chemical Control Division, Office of Pesticides and Toxic Substances.
Prepared by AMS, Inc. Contract Number TV-82228V, Subcontract Agreement No. 6,
Task 2.
U.S. Environmental Protection Agency (EPA). 1992. EPA's 33/50 Program Second Progress
Report: Reducing Risks Through Voluntary Action. Office of Pollution Prevention and
Toxics. TS-792A. February.
U.S. Environmental Protection Agency (EPA). 1993. Personal Communication with Brian
Cook. June 18.
Weitzman, M.,etal. 1991. Boston Lead-in-soil/Lead Free Kids Demonstration Project. Draft.
Abt Associates, Inc. 2-16 Draft, January 10, 1994
-------
3. PROBLEM DEFINITION AND REGULATORY OPTIONS
This chapter characterizes the lead contamination problem to be addressed under Section 403
and presents a rationale for government intervention. A risk summary, provided in the first
part of this chapter, presents quantitative estimates of exposures, blood lead distributions, and
incidences of adverse health effects. The second part of this chapter presents an evaluation of
the market failure associated with these residential lead risks, the need for federal regulation,
and a discussion of possible regulatory option.
3.1 RISK SUMMARY
This section describes the risk assessment modeling procedures, information sources,
and assumptions used to estimate the incidence of adverse health effects to children resulting
from exposure to lead present in paint, soil and dust in residential settings. This risk
assessment model is used to support the Section 403 impact analysis by determining both the
baseline incidence of health damages expected in the absence of actions induced by Section
403, and the benefits that will result from various exposure reduction actions that may result
from the implementation of Section 403 rules.
The risk assessment model has three major components:
• Characterization of lead exposure from residential paint, soil and dust;
• Calculation of blood lead distributions resulting from these exposures; and
• Prediction of the incidence of adverse health effects from the blood lead
distributions.
Each of these components of the risk assessment model is discussed later in this
chapter. First, however, it is important to discuss some of the key underlying assumptions and
premises for the risk assessment model.
In this analysis, it is recognized that the presence of lead in paint, soil and dust is a
long-term environmental problem. Even though lead paint has not been used for residential
purposes since 1979, and major historical sources of lead deposition to soil such as automotive
emissions from leaded gasoline have been eliminated or severely curtailed, the existing stock
of lead in paint, soil and dust from these past sources will remain a major source of exposure
to children for many generations. Consequently, the risk assessment model has been
constructed to address not only the exposure and health risks to those children currently living
in lead-contaminated homes, but also the risks that children who are born into these homes
over the next several decades will face if abatement actions are not taken.
Abi Associates. Inc. 3-1 Draft. January 10.1994
-------
To incoiporate this consideration, the risk assessment model is built around the concept
of annual cohorts of children being born into homes over a SO year period, beginning in 1994.
Based on Census Bureau population projections and other assumptions, the model incorporates
estimates of the number of homes and births expected for each year of this time frame. It is
convenient to view the modeling conceptually as involving an iterative, stepwise process where
separate calculations are made of the incidence of adverse effects for each of these SO annual
cohorts of children, which are then summed to obtain the total for the full modeling period.
Computationally, however, the modeling process involves instead a determination of the
incidence of these adverse effects for the cohort of children born in first year the modeling
time frame. This result is then "multiplied" using factors reflecting birth rates over the SO
year period and the changes in the housing stock characteristics to obtain the incidence for all
children born over the full modeling time frame. Results for individual years are not,
however, explicitly isolated.
Because of this modeling procedure, most of the Risk Summary discussion focuses on
the first year of the model, which as noted above has been set at 1994. In Section 3.1.4, the
derivation of the factors used to multiply the first year results to the full 50 year time frame
are presented.
The terms "baseline" and "first model year" are used throughout this section. These
are not synonymous terms. The term "baseline" refers to the analyses of exposure and
incidence of adverse effects assuming there are no Section 403-induced changes. Subsequent
analyses are performed in which it is assumed that different types of exposure reduction
actions are induced by Section 403, and the results of these are then compared with this
baseline. "First model year" simply refers to the results of either the baseline analysis or the
alternative exposure assumption analyses for the first model year cohort. In all cases, these
first model year impacts are computed, and those results are then extrapolated to the full 50
year modeling time frame through the use of multipliers, as noted previously.
It is important to note that the baseline analysis reflects an assumption that no specific
abatement actions will be performed to reduce current and future exposure to lead from paint,
soil, and dust in the absence of promulgated Section 403 regulations. As discussed later, the
risk assessment model does include assumptions regarding the disappearance of older homes
over time, which has the effect of reducing the probability that a child will be born into a
home with lead paint in future years. However, in the baseline hazard assessment analysis, it
is assumed that no abatements of lead paint, soil or dust will be performed in the absence of
Section 403 regulations. This is obviously not the case, since many states and municipalities
are currently implementing lead abatement programs. However, it is difficult to estimate how
many such abatements will be done over the next several decades.
The "no abatement" assumption for the baseline clearly results in an overestimate of the
damages expected in the absence of the Section 403 rule as well as the benefits of having the
rule. Nevertheless, using the "no abatement" assumption in the baseline provides a common
basis for comparing the effectiveness of alternative hazard levels, and it is not expected to
Abt Associates, Inc. 3-2 Draft, January 10.1994
-------
affect the outcome of the analysis in terms of identifying the range of hazard levels that
maximize net benefits.
3.1.1 Characterization of Exposure
The purpose of the exposure characterization component of the model is to define the
distribution of lead levels in paint, soil, and dust in privately-owned housing stock in the US.
The exposure assessment also addresses other characteristics of these homes that affect
children's exposure, particularly the condition of the lead paint.
The distribution of current lead levels in paint, soil and dust in the US housing stock is
derived from the results of the survey sponsored by the US Department of Housing and Urban
Development. That survey was conducted in 1989-1990 to provide better estimates of the
extent of lead paint hazards in the Nation's private housing stock. The results of that survey
have been detailed by HUD in its December 1990 Report to Congress entitled Comprehensive
and Workable Plan for the Abatement of Lead-Based Paint in Privately Owned Housing
(HUD, 1991).
The HUD survey focused on privately-owned, occupied homes built prior to 1980.
HUD estimated that at the time its survey was conducted, there were approximately 77 million
pre-1980 homes in the US. The focus of the HUD survey on pre-1980 homes reflects the ban
on the use of lead-based paint for residential purposes in 1978 by the Consumer Product Safety
Commission acting under the authority of the Consumer Product Safety Act.
There were a total of 284 homes sampled in the HUD survey. The survey sample
design involved a stratification of the pre-1980 housing stock into six groups reflecting three
construction-period categories (pre-1940, 1940-1959, and 1960-1979) and two dwelling types
(single family and multifamily). To adjust for disproportionate sampling within these six
strata, as well as to correct for recognized disproportionate sampling with respect to census
region and presence/absence of children under age 7, the 284 HUD samples were given
"weights" by HUD so that the results from the 284 samples could be extrapolated to the
national total of 77 million pre-1980 homes. These HUD-specified weights were used in the
risk assessment modeling performed here, with additional adjustments made to them as
described later to accommodate the post-1980 housing stock.
Lead measurements of interior and exterior paint, exterior soil, and interior dust were
taken at each of the 284 HUD sample homes. Generally, measurements of lead in these media
were made at several locations and surfaces in each sample home. Other information relevant
to assessing exposure to lead in these homes was also obtained, such as the existence and
extent of damaged surface area of paint. The following briefly describes how these HUD
measurement data were used to characterize exposure potential in the model.
Lead in Paint. The most commonly used method to measure the level of lead on
painted surfaces in homes is the XRF (x-ray fluorescence) technique, which measures
lead in paint present on surfaces in units of mg/cm2. It should be noted that because of
Abt Associates. Inc. 3-3 Draft. January 10,1994
-------
limitations in this analytical method, low levels of lead paint reported by XRF
measurements (for example, in the range of approximately 1.0 mg/cm2 or less) are
considered much less reliable than are higher readings. For the purposes of this model,
XRF readings of 0.7 mg/cm2 were used as the cut-off to distinguish between homes
with and without lead based paint. That is, homes having reported XRF measurements
of 0.6 mg/cm2 or lower were considered to be free of lead paint.
Exposure to lead paint in the risk assessment model is associated primarily with interior
lead paint levels. In the HUD survey, interior XRF readings were taken at several
locations in each home, including one randomly selected wet room (i.e., rooms having
plumbing such as a kitchen or bathroom) and one randomly selected dry room.
Measurements were made on several substrates within those rooms, such as walls,
ceilings, windows, molding, door systems, and shelves. The value used to characterize
paint exposure potential in homes was the maximum interior XRF value, which is the
most frequently used measure to characterize lead paint levels in homes.
Data were also obtained on the XRF value for exterior paint. This information was
used in the model mainly in the abatement cost analysis to identify those homes
undertaking soil abatement that would also require exterior paint abatement to be fully
effective. Exterior lead paint information was also used in conjunction with the interior
paint reading to identify the number of lead-free homes in the HUD sample. As
discussed later, this information was needed to adjust the weighting factors to simulate
changes hi future characteristics of the housing stock.
The HUD survey also provided information on the condition of the lead paint in these
homes. For the purposes of this model, housing units reported to have more than 5 ft2
of damaged interior lead paint were classified as "bad condition" homes, as discussed
further below.
Lead in Soil. In the HUD survey, residential soil readings were taken near the
entrance to the home, at the drip line, and at a remote location. Soil lead
measurements were reported in parts per million (ppm). To be most representative of
the overall levels to which children are exposed, the arithmetic average of the
individual soil lead levels measurements was used to characterize each home.
The definition of lead-contaminated soil under Section 401 of TSCA Tide IV refers
specifically to "bare soil on residential real property that contains lead at or in excess of
the levels determined to be hazardous" by EPA under the section 403 regulations.
It has been noted that exposure of children to lead from soil is enhanced when they play
on nongrassy surfaces rather than on grass-covered areas (Madhavan et al., 1989;
Lewis and Clark County Health Department et al., 1986). It has been suggested that
contact with bare soil areas may result in increased ingestion of soil particles by
children. Also, bare soils may contribute more to household dust than covered soils.
While the enhancement of lead exposure from bare soils is often noted in the technical
Abt Associates, Inc. 3-4 Draft, January 10.1994
-------
literature, no studies were found that specifically addressed a quantitative difference in
children's blood lead levels as a function of the degree of soil cover. Most studies
concerning the relationships between soil lead and blood lead levels do not note the
condition of the soil with respect to grass or other form of cover.
In the risk analysis and benefit-cost modeling that has been performed for this impact
assessment, it has not been possible to specifically isolate and focus on lead hazards
associated with bare residential soils separately from other residential soils that are
partially or completely covered with grass or some other form of ground cover.
One major impediment is that the data in the HUD survey (which provides the basis for
the national estimates of the distribution of lead levels in residential soils used in the
modeling performed here) does not include any indication as to the condition of the
soils in the sample homes. Consequently, we cannot stratify the HUD-measured levels
of lead in the soils of the US privately-owned housing stock in terms of the condition of
the soil as bare or covered. No information from other sources concerning the national
incidence of lead levels as a function bare or covered soil in residential settings is
known to be available. Therefore, we are unable to estimate either the number of
homes that have bare soil, or the distribution of lead levels in bare soils and covered
soils.
A second impediment is that in the IEUBK model there is no differentiation made
between bare and covered soil in estimating intakes. That is, lead intake via soil is
addressed in the IEUBK model in terms of a daily ingestion of soil. (This is assumed
to be 45% of the combined daily soil and dust intake which ranges from 85 to 135
mg/day depending upon age.) The model does not suggest different default values
either for the soil fraction value or for the total ingestion amount of soil and dust as a
function of prevailing soil conditions. However, it seems reasonable to assume that for
a given lead concentration in soil, lead intake (and therefore blood lead levels) would
be greater for children regularly exposed to bare soils than for children exposed to soils
that have some form of cover. Though not stated explicitly, it appears that the assumed
intake values in the IEUBK model reflect an averaging of a range of intakes that may
include contact with both bare and covered soils.
In terms of the aggregate, baseline risk assessment, the inability to specifically address
bare and covered soils may result in an "averaging out" of the overall health damages
by overestimating damages for children exposed to a given lead level in covered soils
while underestimating damages for other children exposed to similar lead levels in bare
soils. The potential effects of this averaging out on the benefit and benefit-cost
implications of setting a soil hazard level is discussed further in Chapters 5 and 7.
Lead in Dust. In the HUD survey, floor dust lead concentrations (in ppm) were
obtained for a wet room, dry room, and at the entry way. An arithmetic mean of these
measurements was used to characterize the floor dust concentration for each HUD
Abt Associates. Inc. 3-5 Draft. January 10,1994
-------
home. Dust lead measurements were also taken for window wells and window sills,
but were not used in the averaging. Dust loading measurements, repotted in units of
ug/ft2- were also taken in the HUD study. However, the model for predicting blood
lead levels from exposure to dust (as described in the next section) requires dust
concentrations, and cannot use dust loading values directly.
For the purposes of the risk assessment model, each of the 284 HUD sample homes
represents a group or category of homes. The lead paint, soil, and dust characteristics for
every home in each of these groups is given by the measured values in the corresponding HUD
sample home.
The number of homes in each category is given by the weighting factor applied to the
HUD sample which, as described previously, accounts for several sampling strata
characteristics. The sum of the HUD weights is approximately 77 million homes, which is the
estimated current size of the privately-owned housing stock built prior to 1980. As described
in Section 3.1.4, below, it is estimated that the total privately-owned housing stock in 1994
(the year that is used as the starting point for the risk assessment modeling) is approximately
96 million. This implies that approximately 19 million homes were built from 1980 through
1994 hi addition to the 77 million built prior to 1980. Unfortunately, there is no comparable
survey providing useful lead paint, soil, and dust measurement data for the homes built from
1980 to the present. To incorporate post-1980 homes into the model, the two assumptions
were used.
First, it was assumed that these homes will be free of both ulterior and exterior lead-
based paint. Second, it was assumed that soil and dust lead levels in post-1980 homes will
follow a pattern similar to soil and dust lead levels in pre-1980 sample homes that are also free
of lead paint. Accordingly, the weights of the subset of pre-1980 homes in the HUD survey
that were found to be free of both interior and exterior lead-based paint were adjusted upward
(proportionately) to account for the additional post-1980 homes such that the sum of all of the
weights totaled 96 million.
Exhibits 3-1, 3-2, and 3-3 summarize the resulting distribution of 1994 homes by paint,
soil and dust lead levels, respectively. These distributions reflect the data obtained for the 284
HUD homes and the adjusted HUD weighting factors to extrapolate from those 284 samples to
the 96 million occupied, privately-owned homes estimated for 1994.
The data shown in Exhibit 3-1 indicate the pervasiveness of lead paint in homes despite
the ban on its use in 1979. Over 40% of homes, some 40 million, are still expected to have
interior lead paint present in 1994. While the majority of these have maximum interior XRF
values in the low end of the range (1 to 6 mg/cm2), there is still a substantial number of homes
with lead paint present at very high XRF levels. For example, there are about 3.8 million
homes estimated to have lead based-paint present with a maximum interior XRF value at or
above 10 mg/cm2.
Abt Associates, hie. 3-6 Draft, January 10,1994
-------
The distribution of soil lead levels shown in Exhibit 3-2 indicates that just over half of
all homes have average soil lead levels below 100 ppm. About 13% of homes have soil lead
levels at or above 500 ppm, and only 1.6% are estimated to have lead levels above 3,000 ppm.
The maximum average value observed from the HUD survey was 8,800 ppm, affecting just
over 200,000 homes.
Relative to soil lead levels, there is a much higher frequency of dust levels above 100
ppm (over 96%) as well as a higher incidence in the middle range of values, with some 36%
exceeding 500 and about 15% exceeding 1,000 ppm. The maximum dust lead concentration
found was 5,900 ppm.
Exhibit 3-1
Summary of Distribution of Maximum Interior
XRF Values for 1994 Homes
Maximum Interior
XRF Measurement
22
20
13
12
11
10
9
8
7
6
5
4
3
2
1
0
Estimated Number and
Percent of 1994 Homes
174,136
1,308,115
116,914
233,828
334,584
1,596,469
1,453,400
679,926
717,116
1,735,126
646,947
634,211
2,881,551
4,902,934
21,242,577
57,675,167
0.18%
1.36%
0.12%
0.24%
0.35%
1.66%
1.51%
0.71%
0.74%
1.80%
0.67%
0.66%
2.99%
5.09%
22.05%
59.87%
Cumulative Number and
Percent of 1994 Homes
174,136
1,482,250
1,599,164
1,832,992
2,167,576
3,764,045
5,217,445
5,897,371
6,614,486
8,349,612
8,996,559
9,630,771
12,512,322
17,415,256
38,657,833
96,333,000
0.18%
1.54%
1.66%
1.90%
2.25%
3.90%
5.41%
6.12%
6.86%
8.66%
9.34%
9.99%
12.99%
18.08%
40.13%
100.00%
Abt Associates. Inc.
3-7
Draft, January 10,1994
-------
Exhibit 3-2
Summary of Distribution of Average
Soil Concentrations for 1994 Homes
Average Soil
Concentration (ppm)
8800
5800
3100
3000
2300
2200
1700
1500
1400
1300
1200
1100
1000
800
700
600
500
400
300
200
100
0
Estimated Number
of 1994 He
217.940
120,342
811,603
420,357
1,376,717
300,785
108,201
193,020
116,914
811,602
116,644
127,818
2,516,509
885,948
373,906
941,777
2,960,312
1,594,968
3,887,101
6,292,020
22,015,661
50,142,856
and Percent
>mes
0.23%
0.12%
0.84%
0.44%
1.43%
0.31%
0.11%
0.20%
0.12%
0.84%
0.12%
0.13%
2.61%
0.92%
0.39%
0.98%
3.07%
1.66%
4.04%
6.53%
22.85%
52.05%
Cumulative Number and
Percent of 1994 Homes
217,940
338,282
1,149,884
1,570,241
2,946,959
3,247,744
3,355,945
3,548,965
3,665,879
4,477,480
4,594,124
4,721,942
7,238,451
8,124,399
8,498,305
9,440,082
12,400,393
13,995,361
17,882,463
24,174,483
46,190,144
96,333,000
0.23%
0.35%
1.19%
1.63%
3.06%
3.37%
3.48%
3.68%
3.81%
4.65%
4.77%
4.90%
7.51%
8.43%
8.82%
9.80%
12.87%
14.53%
18.56%
25.09%
47.95%
100.00%
Abt Associates, Inc.
3-8
Draft, January 10.1994
-------
Exhibit 3-3
Summary of Distribution of Average Floor
Dust Concentrations for 1994 Homes
Average Dust
Concentration (ppm)
5900
5800
5300
4400
3600
3300
3200
2700
2500
2400
2100
1800
1700
1600
1500
1400
1300
1200
1100
1000
900
800
700
600
500
400
300
200
100
0
Estimated Number and
Percent of 1994 Homes
1,197,765
120,342
532,215
369,692
127,818
859,142
256,992
748,083
233,557
116,644
393,074
100,535
925,289
295,510
402,265
1,042,945
787,851
2,819,209
474,932
2,141,279
1,041,672
3,294,547
2,043,610
7,006,259
7,654,178
6,940,566
12,572,751
18,270,537
20,140,532
3,423,212
1.24%
0.12%
0.55%
0.38%
0.13%
0.89%
0.27%
0.78%
0.24%
0.12%
0.41%
0.10%
0.96%
0.31%
0.42%
1.08%
0.82%
2.93%
0.49%
2.22%
1.08%
3.42%
2.12%
7.27%
7.95%
7.20%
13.05%
18.97%
20.91 %
3.55%
Cumulative Number and
Percent of 1994 Homes
1,197,765
1,318,106
1,850,321
2,220,013
2,347,831
3,206,972
3,463,964
4,212,047
4,445,604
4,562,248
4,955,322
5,055,858
5,981,147
6,276,657
6,678,921
7,721,866
8,509,717
11,328,925
11,803,858
13,945,137
14,986,809
18,281,356
20,324,966
27,331,225
34,985,402
41,925,968
54,498,719
72,769,257
92,909,788
96,333,000
1.24%
1.37%
1.92%
2.30%
2.44%
3.33%
3.60%
4.37%
4.61%
4.74%
5.14%
5.25%
6.21%
6.52%
6.93%
8.02%
8.83%
11.76%
12.25%
14.48%
15.56%
18.98%
21.10%
28.37%
36.32%
43.52%
56.57%
75.54%
96.45%
100.00%
Abt Associates, Inc.
3-9
Draft, January 10,1994
-------
To cany out the risk assessment modeling, some additional stratification of the 284
categories of homes was also performed to reflect characteristics that affect exposure from
interior lead paint in these homes. Of the 284 HUD samples, 141 were found to have some
lead paint present (i.e., maximum interior XRF values of 0.7 mg/cm2 or more). Each of the
141 groups of homes represented by these samples were divided into eight subgroups. Note
that the lead paint, soil and dust levels in each of these subgroups remain the same as in the
original group from which each is derived. The eight subgroups were created using the
following three characteristics:
• Based on data presented in the HUD report two subcategories were created to
differentiate between homes having interior lead paint on windows (40%) and those
that do not (60%).
• Using HUD data, homes with lead paint were differentiated between those having lead
paint in bad condition (24%) and good condition (76%). The criteria used for bad
condition paint was that provided by HUD in its report as homes having more than 5
ft2 of damaged, painted surface area.
• Lastly, 25% of all homes were identified as ones in which a child would exhibit pica,
while children in the remaining 75% of homes would not have pica. (This and other
pica assumptions are discussed in Section 3.1.2.)
These percentages were applied to the weights for each HUD home having interior
paint to obtain new weights for each of the eight subgroups.1 For example, if one of the
original 141 HUD samples with lead paint had a weight of 10,000 (i.e., it represents 10,000 of
the 96 million homes in 1994), the eight subgroups created from it would be weighted as
follows:
1 An alternative approach was considered for incorporating the windows and paint condition characteristics
into the modeling. This was to simply consider these characteristics in an "all or none* manner for the homes
represented by each of the 284 HUD samples. For example, the national estimate of homes with paint in bad
condition could have been taken as the sum of the weights for each of the 284 HUD sample homes found to have
paint in bad condition. This would have implied that only those homes with those particular combinations of lead
paint, soil and dust levels have lead paint in bad condition, while homes with all other combinations of levels
have interior paint in good condition. Similarly, homes with lead paint on windows would have been restricted in
the model to only those homes with the particular paint, soil and dust lead combinations in the representative
HUD sample homes where lead paint on interior windows was observed. Given the relatively small size of the
HUD sample homes, it was judged that this "all or none" approach would be less representative of the prevalence
of these conditions across all combinations of lead levels in paint, soil and dust hi homes. Therefore, the
approach used here was to apply the frequency observed (weighted by the HUD sample weights) for these
characteristics across all homes with interior lead paint.
Abt Associates, Inc. 3-10 Draft, January 10,1994
-------
Subgroup Characterisics Weight Derivation
Lead paint on windows, good condition, no pica: 2,280 (= 0.4 * 0.76 * 0.75 * 10,000)
Lead paint on windows, bad condition, no pica: 720 (= 0.4 * 0.24 * 0.75 * 10,000)
Lead paint on windows, good condition, pica: 760 (= 0.4 * 0.76 * 0.25 * 10,000)
Lead paint on windows, bad condition, pica: 240 (= 0.4 * 0.24 * 0.25 * 10,000)
No lead paint on windows, good condition, no pica: 3,420 (= 0.6 * 0.76 * 0.75 * 10,000)
No lead paint on windows, bad condition, no pica: 1,080 (= 0.6 * 0.24 * 0.75 * 10,000)
No lead paint on windows, good condition, pica: 1,140 (= 0.6 * 0.76 * 0.25 * 10,000)
No lead paint on windows, bad condition, pica: 360 (= 0.6 * 0.24 * 0.25 * 10.000)
The presence or absence of lead paint on windows was used primarily in the benefits
analysis to differentiate between homes needing full paint abatement (those without lead on
windows) and those that could be abated by replacing the windows only.
The homes having the combination of both lead paint in bad condition and pica children
are particularly important for the risk analysis. For these children, the model used to predict
blood lead levels (described in Section 3.1.2) included special input assumptions for paint chip
ingestion, as well as for exposure through dust and soil ingestion.
The 143 groups of homes without interior lead paint were not further stratified in the
model. As a result of these assumptions, the US privately-owned housing stock was ultimately
stratified into a total of 1,271 subgroups. Of the original 284 categories based on the HUD
samples, 141 had interior lead paint, which were therefore stratified into 1,128 subgroups (8 x
141). Adding to these the 143 that did not have interior paint results in the total of 1,271.
Having the housing stock fully stratified and properly weighted to account for the 96
million homes in 1994, the model then applied the estimated probability of a child being bom
into any home of 0.03994 for 1994 to determine the number of homes in each strata expected
to have a child in the first model year (see Section 3.1.4 for additional discussion of birth
rates). Blood lead distributions, and the incidence of adverse health effects, were then
computed for the children born into each of the 1,271 categories of homes.
3.1.2 Determining Blood Lead Distributions
For each of the 1,271 categories of homes created in the model, an estimate was made
of the geometric mean blood lead for the children born into them, all of whom are assumed to
live in those homes from birth through age seven. The geometric mean blood lead estimates
were obtained using EPA's Integrated Exposure, Uptake and Biokinetic Model for lead,
hereafter referred to as the IEUBK model.
The IEUBK model has been developed by EPA to use as a tool for estimating the
geometric mean blood lead levels in populations of children exposed to various levels of lead
in environmental media. The ffiUBK model has been under development for several years,
and has been available in several interim versions. The version of the IEUBK model used for
this analysis became available in July 1993, and is currently undergoing validation studies.
Abt Associates. Inc. 3-11 Drafti January J0f 1994
-------
The IEUBK model is designed to use data on lead concentrations in air, water, soil,
household dust, diet, and paint chips to estimate the geometric mean (GM) blood leads for a
population of children exposed to those specified environmental concentrations. To account
for individual variability within that population of children exposed to similar environmental
levels, it is assumed that the overall distribution of blood lead levels for that population is
lognormal, with an assumption made for the geometric standard deviation (GSD) of that
distribution, as discussed further below.
Exhibit 3-4 summarizes the IEUBK input assumptions used for this analysis. Shown
there are assumptions regarding levels of lead in each medium, daily intake of those media,
and absorption of lead from each source. The levels of lead in air and water, and the dietary
intake values are kept constant for all children, using the values shown in the Exhibit 3-4. The
input values for soil and dust are those in each housing group obtained from the HUD data as
described above. The intake of lead for the subgroup of children assumed to be ingesting lead
paint chips was estimated as follows.
An estimate was provided by Elias (1993) that children ingesting paint chips ingest an
average of 2.5 chips per week, each 1 cm2. It was also assumed that these chips are 0.1 mm
thick, resulting in an overall size of 0.01 cm3. It was also assumed that the density of a lead
paint chip is 2 g/cm3, a typical value for solid materials.
A draft IEUBK guidance manual (EPA, 1991) provided a relationship for estimating
the lead concentration in a paint chip from the XRF measurement, which as noted previously
has the units of mg/cm2. This relationship is:
10'100°M8pb P« XRF (mg/cm2) Equation 3. 1
Therefore, a paint chip from a lead-painted surface having an XRF measurement of 10
would be estimated to have 100,000 ng of lead (or, 0. 1 g of lead) per gram of paint chip.
Combining the above assumptions provides an estimate of the daily intake of lead from
paint chip ingestion as a function of the XRF value:
?days
Equation 3.2
sipb per XRF
day
Therefore, a child ingesting paint chips in a home where the interior XRF value is 10
mg/cm2 is assumed to have an intake of 700 ug/day from this source.
Abl Associates. Inc. 3-12 Draft. January 10.1994
-------
Exhibit 3-4
Summary of Parameter Values Used in IEUBK Model
Air Parameters:
All air parameters use default values
Vary air concentration by year?
Outdoor air lead concentration (ug/m3):
Indoor air concentration (% of outdoor value):
No
0.10
30%
Diet Intake Parameters:
Age:
Diet Intake
(ug/day):
Diet parameters were reduced to 50% of default values.
0-1
2.75
1-2
2.89
2-3
3.25
3-4
3.12
4-5
3.01
5-6
3.17
6-7
3.5
Water Intake Parameters
Lead concentratio
Age:
Drinking water
consumption (L/day):
: All water parameters use default values
D in water: 4 ug/L
0-1
0.02
1-2
0.05
2-3
0.52
3-4
0.53
4-5
0.55
5-6
0.58
6-7
0.59
Soil and Dust Intake Pan
Soil/dust ingestioi
Age:
Total soil + dust
intake (g/day):
uneters: Soil and dust levels are input. All other parameters use default values.
i weighting factor: 45% soil : 55% dust
0-1
0.085
1-2
0.135
2-3
0.135
3-4
0.135
4-5
0.1
5-6
0.09
6-7
0.085
Absorption method values: (* Indicates change from default)
Soil
Dust
Water
Diet
Alternate
Total
Absorption
(percent)
0.3
0.3
0.5
0.5
0.1
Fraction of
Total Assumed
Passive Absorption
0.05 *
0.05 *
0.05 *
0.05 *
0.05 *
Abt Associates, Inc.
3-13
Draft, January 10,1994
-------
As indicated in the previous section, homes with lead paint were stratified to isolate the
subset having children ingesting lead paint chips. It was estimated from the HUD data on
paint condition that 24% of homes with interior lead paint have non-intact paint. It was also
assumed that 25 % of children exhibit pica. This estimate was based on Baltrop (1966) as cited
in HUD (1991) that the frequency of pica among children in inner cities is 20-30%. A lower
estimate of pica incidence has been provided by Mahaffey (1993) based on an analysis of
NHANES n data indicating that the incidence of pica among children 0.5-3 years old is 11 %.
This lower estimate became available late in the modeling process. Incorporating it will, of
course, reduce the computed effect of paint chip ingestion on blood lead levels. Since there is
a concern that this may already be underestimated in the model, as discussed in Section 3.1.5
below, it was decided not to include this lower pica frequency estimate at this stage.
However, it may be included in a subsequent sensitivity analysis.
The combined conditions of non-intact paint and pica children therefore implies an
overall estimate that 6% (i.e., 25% of 24%) of children in homes with interior lead paint will
ingest lead paint chips.
It is important to note that for children in the remaining 94% of homes, the blood lead
geometric means estimated from the ffiUBK model are not affected either by lead paint XRF
value or by the condition of the paint in those children's homes. That is, pica and non-pica
children bom into homes having paint in good condition have no difference in their calculated
blood lead levels, all other exposure conditions being equal. Similarly, non-pica children in
homes with lead paint in bad condition have the same calculated blood lead as those with lead
paint in good condition, all other exposure conditions being the same. Exhibit 3-5 shows the
predicted blood lead levels from the EEUBK model for these various combinations of pica and
condition for a given set of paint, soil and dust lead levels.
Exhibit 3-5
Comparison of Predicted Blood Lead Levels for Pica/Non-Pica Children
in Homes with Lead Paint in Bad or Good Condition
Lead Paint in Bad Condition
Lead Paint in Good Condition
Pica Children
13.89 ug/dl
6.93 ug/dl
Non-pica Children
6.93 ug/dl
6.93 ug/dl
Assumed exposure conditions: Interior XRF = 5 rag/cm2
Soil = 500 ppm
Dust = 500 ppm
Exhibit 3-6 provides a summary of predicted geometric mean blood leads for soil and
dust lead concentration combinations ranging between 100 and 1,000 ppm each. These
estimates, which exclude any contribution from paint chip ingestion, indicate that the
geometric means increase by about 0.5 ug/dl (ranging from about 0.35 to 0.70 ug/dl) for each
Abt Associates, hie. 3-14 Draft, January 10,1994
-------
increase of 100 ppm in either soil or dust. For example, at a soil level of 500 ppm and dust
level of 100 ppm, the estimated geometric mean is 4.55 ug/dl. At this same soil level, with
dust increased to 200 ppm, the geometric mean blood lead increases to 5.18 ug/dl, a change of
0.63 ug/dl. The magnitude of the blood lead changes per 100 ppm of either soil or dust
becomes lower as the soil and/or dust level becomes higher.
Exhibit 3-7 shows the impact of paint chip ingestion on predicted geometric means.
For a given soil and dust concentration, the geometric mean of children ingesting paint chips is
estimated to generally increase about 1 to 2 ug/dl per unit change in the XRF measurement,
with the amount of increase less at high soil/dust/XRF combinations than at lower values.
Abt Associates, Inc. 3-15 Draft, January 10,1994
-------
Exhibit 3-6
Estimated Blood Lead Geometric Means (ug/dl) for Various Soil-Dust
Combinations (excluding paint chip ingestion)
SQL
Dust Cone, (pom)
100
200
300
400
500
600
700
800
900
1000
100
2.33
3.04
3.72
4.37
5.01
5.62
6.21
6.78
7.33
7.87
200
2.91
3.60
4.26
4.89
5.51
6.10
6.68
7.23
7.77
8.29
300
3.47
4.14
4.78
5.40
6.00
6.58
7.13
7.67
8.20
8.71
400
4.02
4.67
5.29
5.89
6.47
7.03
7.58
8.10
8.61
9.11
Concentration (D
500 | 600
4.55
5.18
5.78
6.37
6.93
7.48
8.01
8.52
9.02
9.50
5.06
5.67
6.26
6.83
7.38
7.91
8.43
8.93
9.42
9.89
pm)
700
5.56
6.16
6.73
7.28
7.82
8.34
8.84
9.33
9.80
10.26
800
6.05
6.63
7.18
7.72
8.25
8.75
9.24
9.72
10.18
10.63
900
6.52
7.08
7.63
8.15
8.66
9.15
9.63
10.10
10.55
10.98
1000
6.98
7.53
8.06
8.57
9.06
9.55
10.01
10.47
10.91
11.33
Abt Associates, Inc.
3-16
Draft, January 10, 1994
-------
Exhibit 3-7
Estimated Geometric Means (ug/dl) for Children Ingesting
Paint Chips with Indicated XRF/Soil/Dust Levels
Soil (ppm): 100
Dust (ppm): 100
XRF: 0
1
2
3
4
5
6
7
8
9
10
500
500
2.33
4.47
6.38
8.09
9.64
11.06
12.35
13.54
14.64
15.65
16.60
6.93
8.60
10.10
11.48
12.73
13.89
14.96
15.96
16.89
17.75
18.56
1000
1000
11.33
12.60
13.77
14.85
15.86
16.79
17.66
18.48
19.28
19.97
20.65
The EBUBK model provides age-specific estimates of the geometric mean blood lead
for given exposure conditions at ages ranging from birth through seven years. For this
analysis, the blood lead geometric mean predicted for age three was selected to use for
estimating health damages. This age was selected because blood lead levels tend to peak at
this age. It is also consistent with assumptions that cognitive effects are expected to occur only
after having elevated blood lead levels for a period of 3 to 4 years.
\
As noted previously, the ffiUBK model produces an estimate of the geometric mean for
an assumed lognormal distribution of blood lead levels for the population of children exposed
to similar environmental levels of lead. In this risk analysis, geometric mean estimates are
made for 1,271 separate populations of children based on the stratification of homes as
described in Section 3.1.1.
The variability of blood lead levels within in each of these 1,271 populations is
estimated by using a standard assumption that the geometric standard deviation (GSD) is 1.6,
regardless of the predicted geometric mean or the specific characteristics of the house where
those children are born. The assumption that the GSD is 1.6 for all subpopulations, provided
by Schwartz (1993a), is the best preliminary estimate of the overall population blood lead
distribution GSD obtained from the recent NHANES m study. As discussed further below,
this assumption that the overall population variance also applies uniformly to each subgroup
Abt Associates. Inc.
3-17
Draft. January 10.1994
-------
within that population results in an inconsistent outcome. That is, by assuming that the GSD is
1.6 for each of the 1,271 subgroups, and then aggregating those 1,271 blood lead distributions
to arrive at an overall population blood lead distribution, the resulting GSD for this overall
population distribution is necessarily larger than the 1.6 value used for each subgroup. This
effect will be discussed more fully in Section 3.1.5.
3.1.3 Estimated Incidence of Adverse Health Effects
The estimates of the incidence of adverse health effects resulting from exposure to lead
in residential paint, soil and dust were derived primarily from the blood lead distributions
obtained for each of the 1,271 categories of homes. The methodology used to obtain these
estimates is essentially identical to the methodology that has been used previously for
estimating the baseline health effects and benefits for regulating lead levels in gasoline and
drinking water.
There are several categories of adverse health effects that have been associated with
environmental exposure to lead. These include:
Adults:
Hypertension
Non-fatal heart attack and non-fatal stroke
Premature death from all causes
Possible cancer
Reproductive effects (women only)
Infants and Children:
Neonatal morality
Cognitive effects, including reduced intelligence
Interference with growth
Interference with nervous system development
Metabolic effects, impaired heme synthesis, anemia
Possible cancer
In this analysis, only effects on children have been considered. Inadequate data are
available to fully quantify the relationships between levels of lead in residential paint, soil and
dust and adult blood lead levels.
Abt Associates, Inc. 3-18 Draft, January 10.1994
-------
The adverse health effects included for children in this risk analysis are the effects on
intelligence and neonatal mortality. Specifically, the effects on intelligence included in the
analysis are:
• IQ point decrements
• Incidence of IQ < 70
• Low level cognitive damage, estimated from the incidence of blood lead levels >
25 ug/dl.
The incidence of each of these adverse effects was estimated separately for the annual
cohort of children in each of the 1,271 housing groups, with the total for all children in that
year's cohort obtained by summing across all subgroups. Again, the blood lead distribution
for each of these 1,271 subgroups were defined by the GMs obtained from the ffiUBK model
and the assumed GSD of 1.6.
As discussed further below, the estimates of the incidence of neonatal mortality do not
directly involve the use of the blood lead distributions obtained from the EEUBK model.
IP Point Decrements. The estimate of IQ point losses was obtained using a dose-
response relationship of 0.25 points lost per ug/dl of blood lead, as provided by Schwartz
(1993b). To obtain the total number of IQ points lost in a population of children, the 0.25
points lost per ug/dl change in blood lead is multiplied by the average blood lead level for that
population. Note that the value obtained from the IEUBK model is the geometric mean for
that distribution, not the arithmetic mean. To adjust for this, the relationship between the
expected value and the geometric mean of a lognormal distribution was used:
E(X) = exPrin(GM)+(ln(G2SD))21 Equation 3.3
where E(X) is the expected value (mean) of the distribution, GM is the geometric mean, and
GSD is the geometric standard deviation. Taking the log of both sides gives:
ln(E(X)) = ln(GM) + (ln(G2SD))2 Equation 3.4
Rearranging and exponentiating gives the ratio between the mean and the GM of:
in(E(X)) - in(GM) = ('n(G^D))2 Equation 3.5
(In(GSD))2
. [
T
GM 2 Equation 3.6
Ma Associates, Inc. 3-19 Draft, January 10,1994
-------
E(X)_ r(ln(GSD))21
GM '"T. 2J
Equations.?
For a GSD of 1.6, the resulting ratio between E(X) and GM is 1.117:
E(X) _J(ln(1.6))21
:cxpf " •• I
1 2 J
Equation 3.8
Therefore, the total lost IQ points for each group was estimated as:
GM«1.117*0.25*(Pop)k Equation 3.9
where (Pop)k is the number of children in the kth group of homes. Thus if a group of homes
has 10,000 children and an estimated GM from the IEUBK model of 4 ug/dl, the estimated IQ
points lost among these children due to lead is 1 1 , 170.
Incidence of IP < 70. The estimated incidence of IQ values below 70 was derived
using the blood lead distributions for each housing strata in a manner similar to that above. In
this case, however, the dose-response function is not constant across all blood lead values as in
the case of IQ point losses. Rather, a piece wise linear function is used to relate the
probability of IQ <70 to blood lead as shown in Exhibit 3-8.
Using the data shown in Exhibit 3-8, standardized estimates are first made of the
expected incidence of IQ < 70 per unit population for blood lead distributions having GMs
ranging from 0.5 up to SO ug/dl (in 0.5 ug/dl increments), each with a constant GSD of 1.6.
For a given housing group with a particular GM predicted from the IEUBK, the incidence of
IQ < 70 is calculated simply as the unit value of IQ <70 obtained from the standardized
estimates for a distribution with that GM, multiplied by the number of children associated with
that housing group.
Abt Associates, Inc. 3-20 Draft. January 10.1994
-------
Exhibit 3.8
Elements of Piecewise Linear Function for Estimating
Probability of IQ < 70 as a Function of Blood Lead (PbB) Range
PbB Range (ug/dl)
0-5.0
5.1-7.5
7.6 - 10.0
10.1 - 12.5
12.6 - 15.0
15.1 - 17.5
17.6 - 22.5
22.6 - 25.0
> 25.0
Slope
2.04 x 10-4
4.88 x lO"4
1.068 x 10-3
1.044 x 10'3
9.76 x lO"4
1.26x10-3
1.328x10-3
1.532 x 10-3
1.464x10-3
Intercept
3.60 x 10-3
2.18x10-3
-2.17x10-3
-1.93 x 10-3
-L08 x 10-3
-5.34 x ID'3
-6.53 x 10-3
-1.112xlO-2
-9.42 x ID'3
Abt Associates, hie.
3-21
Draft, January 10.1994
-------
Blood Lead Levels > 25 ug/dl. The estimate of children having blood lead levels
above 25 ug/dl, which is used as a surrogate indicator of the need for compensatory education
due to low-level cognitive damage, is derived directly from the blood lead distributions for
each strata using the normal distribution function with the estimated geometric mean obtained
from the DEUBK model and the assumed geometric standard deviation of 1.6. The probability
of exceeding 25 ug/dl obtained from the normal distribution function for a given subgroup of
homes is then applied to the total number of children in that subgroup.
Neonatal Mortality. The estimation of the incidence of neonatal mortality was arrived
at differently from the cognitive damages estimates described above. The neonatal mortality
estimate is based on an approach provided in the CDC (1991) report. The CDC report used
the following assumptions, which were also adopted here:
• The risk of infant mortality is 0.0001 per ug/dl of maternal blood lead. This is
based on relationships between maternal blood lead and gestational age, and
between gestational age and infant mortality as provided by Dietrich et al. (1987).
• The presence of lead paint in a home corresponds with a 2.13 ug/dl increase in
maternal blood lead. This value is used in the CDC report, attributed to Bornschein
(personal communication).
The calculation of the incidence of neonatal mortality is therefore limited to those
homes that have interior lead paint. No distinctions were made between paint in good
condition or bad condition, or between XRF levels. The birth rate for the first year cohort
(0.03994 births/house, see Section 3.1.4) was used to estimate the number of pregnant women
in each housing group for that year. Therefore, in the first model year, a housing group with
lead paint having a weighting factor of 100,000 units would have an estimated 0.85 additional
cases of neonatal mortality relative to a comparable size group with homes not having lead
paint. This results from:
100,000 houses • 0.03994 births/house • 2.13 ug/dl • 0.0001 additional deaths per ug/dl
= 0.85 additional deaths Equation 3.10
3.1.4 Extrapolation of First Model Year Results to Full Modeling Time frame
As discussed previously, the risk assessment modeling was premised on the assumption
that presence of lead in paint, soil and dust in homes in the private housing stock will continue
to affect children born into those homes over many decades. In the preceding sections, the
description of the modeling methodology used to estimate the incidence of adverse health
effects resulting from lead in paint, soil and dust focused on a single year's cohort, specifically
that of the first year model year, i.e., 1994.
Abt Associates, Inc. 3-22 Draft, January 10,1994
-------
To cany out the modeling over the 50 year time frame of 1994 to 2043, it was first
necessary to estimate both the size of the housing stock for each model year and the number of
new children entering into that housing stock. Obtaining estimates of new children for this
time frame was relatively straightforward. The Bureau of Census has published a document
entitled Population Projections of the United States by Age, Sex, Race and Hispanic Origin:
1992 to 2050 (Bureau of Census, 1992a). The estimated number of children under 1 year of
age was used as the estimate of new children for each year.
Obtaining estimates of the total housing stock for each year was less straightforward
since there was no published projection of housing levels found for the modeling time frame
comparable to the population projections data noted above. Data were available from the 1992
Statistical Abstract of the United States (Bureau of Census, 1992a) and the Forecast of
Housing Activity (NAHB, 1992) that provided estimates through the year 2000. To estimate
the number of homes from 2000 through 2043, a series of analyses were performed on
historical and projected data comparing number of persons in various age groups per
household from 1960 to 2000 to find the best indicator to use for the 2000 to 2043 projections.
Of the several approaches compared (specifically, using adults per household of ages 18+,
21+, 25+ and 18-64), the best predictor was found to be a ratio of 1.85 adults 21 + years old
per household for the period 1980-2000.
Using that ratio with the population projections through 2050 for adults 21 and over
from the sources noted above, the estimate of the total occupied housing stock for each year
through 2050 was obtained. An adjustment was also made to eliminate public housing, to
limit the analysis to privately-owned, occupied housing. Exhibit 3-9 summarizes the
projections for housing stock and children < 1 year old for the years 1990 through 2050.
Also shown there is the probability of a new child being born into a home in each year, which
is simply the number of children < 1 year old divided by the number of total occupied homes
for that year.
In addition to the overall change in the size of the housing stock, the model
incorporated assumptions regarding the distribution of homes between those with and without
lead paint. As noted in Section 3.1.1, it was assumed that all homes built between 1980 and
1994 are free of lead paint and have lead soil and dust characteristics like those in the pre-1980
housing stock without lead paint. Similarly, it was assumed that all new housing added to the
stock for the remaining 49 years of the model time frame would be added proportionately to
the groups of homes without lead paint
It was also assumed that existing homes disappear from the housing stock at a rate of
0.5% per year, based on data derived from the Annual Housing Survey Components of
Inventory Change: 1973 to 1983 (Bureau of Census, 1991). While this factor is assumed to
be the same for all types of homes regardless of its age or the presence/absence of lead paint,
the effect in this model is to reduce the proportion of lead paint homes relative to the total
number of homes over the modeling period. In 1994, it is estimated from the HUD data that
approximately 51 million of the 96 million privately-owned occupied homes have either
interior or exterior lead paint, comprising 53 % of the total.
Abt Associates, Inc. 3-23
Draft, January 10,1994
-------
Exhibit 3-9. Estimafc
and Child
Kama
JWU
2026
2030
2031
2033
2035
2036
2038
2043
2044
204S
2048
20SO
93*345
121.571
123.S29
131.559
135.206
136.031
137.682
139.332
140.157
141.807
145.620
146.341
147,061
149.224
150.666
ren < 1 Year Old for 1990 - 2050
Occupied PrivttelrOwncd OuUraK 1 Y«r Pratabilfeof.
Hmm OU Chiy < i Y nu
91.177
92.991
95.219
98.562
100.790
101.904
105.905
109.755
110.717
111.680
115.931
118.124
121.413
124.484
126.243
127.122
128.001
128.881
130.639
131.518
132.397
133.190
134.774
136.359
137.152
138.736
139,529
141.009
142.385
143.073
143,761
145.826
146.514
147.202
4.008
3.912
3.817
3.793
3,890
3.904
3.957
3,984
4.038
4.093
4.149
4.177
4.204
4.232
4.260
4.287
4.315
4.327
4.338
4.350
4.361
4.373
4.419
4.431
4.458
4.485
4.513
4.540
4.567
4.621
4,676
4.748
4.771
4.793
4,816
4.861
4.906
4 928
0.0335
0.0328
0.0328
0.0328
0.0328
0.0328
0.0328
0.0328
0.0327
0.0327 |
0 0327 11
Abt Associates, Inc.
3-24
Draft, January 10,1994
-------
Over the SO year period, with an annual loss rate of 0.5%, these 51 million homes
would decrease to about 39.7 million (51 * 0.99550). At the same time, the total occupied
privately-owned housing stock in 2043 is estimated to be 142.4 million. Therefore, in 2043,
the final year of the modeling time frame, houses with either interior or exterior lead paint are
estimated to comprise only 28% of the total.
The modeling of these temporal changes in the size of the housing stock, proportion
of lead paint housing, and birth rates could be done by "brute force," iterating through each
of the 50 years of the model time frame to calculate the incidence separately in each year, and
then sum across all years to arrive at the total. A computational shortcut was used, however,
taking advantage of the fact that children born into any one of the 1,271 housing categories
will have the same predicted blood lead distribution regardless of the year in which those
children are bom. The number of children to whom that distribution applies will vary from
year to year reflecting changes in the size of the housing stock and the birth rate. However,
the blood lead distribution for children in that category of homes will remain the same.
The computed incidence of adverse health effects is a function of blood lead
distributions and the number of children in the population characterized by those distributions.
Since the predicted blood lead distributions are constant in each category of homes over time,
the total incidence of adverse effects to all children bom into a particular category of homes
over time is related to that total number of children. Therefore, the ratio of total children
expected to be bom into those homes over the full modeling time frame to the number of
children born in the first year can be used as a multiplier to apply to the incidence of adverse
effects calculated for the children bora in the first year to obtain the incidence of adverse
effects for children bom over the entire modeling time frame.
Sections 3.1.4.1 and 3.1.4.2 discuss the derivation of these multipliers for homes with
lead paint and homes without lead paint, respectively. There are different multipliers for these
two types of homes because of the difference in housing dynamics for each (i.e., the number
of lead paint homes is decreasing over time while the number of non-lead paint homes is
increasing). Note, however, that among all HUD-based categories of homes with lead paint,
the multiplier is the same. Similarly, there is just one multiplier for all of the HUD-based
categories of homes without lead paint.
The notation used for the variables in these derivations tends to be intricate, and so a
summary of the key variables with various notations is provided in Exhibit 3-10.
Abt Associates, Inc. 3-25 Draft, January 10,1994
-------
Exhibit 3-10
Summary of Key Variables Used in Derivation
of Model Multipliers
B,
rT|LP
^J30
CFBHJ>
CFB|LP
pFBILP
*TSO
QTILP
H.LP
pTjNLP
'"ESO
p.FB|NLP(j)
*T50
TT*NLP(J)
*i
Probability of having a child bom into a home in Model Year
/..
Total number of children bom into lead paint homes over the
full modeling time frame.
First bom children in lead paint homes in Model Year 1 .
First born children in lead paint homes in Model Year i.
Total first bom children in lead paint homes over the full
modeling time frame.
Total number of children born into lead paint homes over the
full modeling time frame in which the first child is bom in
Model Year i..
Number of lead paint homes hi Model Year i that have not yet
had a child born into them. For /= 1 this is the number of lead
paint homes in 1994 ( = 49, 130 million).
Total number of children bom into non-lead paint homes over
the full modeling time frame.
Number of first bom children in non-lead paint homes added to
the housing stock in Model Year / .
Number of non-lead paint homes added to the housing stock in
Model Yeary that have not yet had a first child born into them
in Model Year / . For i = 1 , j = 1 this is the number of non-lead
paint homes in the housing stock in 1994 (= 47,203 million).
Each multiplier involves two components. The first component is to account for the
total number of "first bom" children in homes over the full modeling time frame. That is, for
1994 (the first model year) we estimate the number of children bom based on the number of
homes present and the per-home probability of a child being born. In the second year, another
portion of homes will have a first bom child based on the number of homes in the second year
that haven't yet had a child and the per-home birth probability for the second year. Similarly,
over each of the remaining 50 years, there will be a particular number of "first bom" children.
The first multiplier component addresses the total of these first bom children during the SO
year period of 1994 to 2043.
The second component of the multiplier accounts for the expected number of additional
children bom subsequently into those homes that have already had a first child born into them.
For this second component, we have estimated the expected number of additional children that
Abt Associates, Inc.
3-26
Draft. January 10.1994
-------
would be born during additional 50 year time period beyond the birth of the first child. That
is, for homes having a first child born in the first model year of 1994, the second part of the
multiplier reflects an estimate for additional children born in these homes through 2043. For
homes not having a first child born until the last model year of 2044, the multiplier reflects the
expected number of additional children born in those homes through 2093.
Multiplier for homes with lead paint
The overall multiplier for homes with lead paint can be expressed as:
C71""
MU>=£5§F Equation 3-11
where: C£jf is the total number of children bom into homes with lead paint, and
CfB|LP is the number of children bom into homes with lead paint in Model Year 1
(1994).
As noted above, the first component of the multiplier addresses the total number of first
born children over the 50 year modeling time frame. The number of first births in lead paint
homes in each model year can be computed as:
CH,IP=(H*U>)(B) Equation 3-12
where: Cf81"" is the number of first births in lead paint homes in year i;
H""* is the number of lead paint homes in that category in Year i that have not yet had
a child born into them; and
B, is the per home probability of a birth for year i.
The total number of first births in lead paint homes is, then, the sum of the numbers for
each individual year:
CzT" = BH^XB,) Equation 3.13
1=1
From the data presented in Exhibit 3-9, we know the value for each B, (where / = 1
for 1994). We also know from the HUD data the value for H*"", which is simply the total
number of lead paint homes in 1994, estimated to be 49 million. (Note: This number is higher
than the 38.7 million shown in Exhibit 3-1 as having interior XRF values greater than 0, since
the definition of lead paint homes used here includes both those with interior and exterior lead
paint.
Abt Associates, Inc. 3-27
Draft. January 10.1994
-------
We do not know, directly, the values for H^ for Model Years 2 through SO; however,
these values can be derived as follows. For Model Year 2, the number of lead paint homes
that have not yet had a child born into them is:
-(H7P.B1)).0.995 Equation 3.14
That is, the number of homes at the start of Model Year 2 not yet having a first birth is equal
to the number at the start of in Model Year 1 not having a first birth, less the number that do
have a child bom in Model Year 1, reduced by 0.5% to account for loss in the housing stock
that year.
Similarly, the number for Model Year 3 can be expressed as:
H?* = (H;1* -(H? .B2)).0.995 Equation 3.15
Note that the values for Model Year 3 are expressed in terms of Model Year 2 values, which
in turn are expressed in terms of Model Year 1 values. Therefore, the expression for Model
Year 3 can be written in terms of Model Year 1 values as:
Hi"" = (H2nj>-(H^J>.B2)).0.995
= (H;LP).(1-B2).0.995 Equation 3.16
= (H|")*(1-B1)«0.995*(1-B2)«0.995
Similarly, the value for each subsequent year can also be expressed in terms of the birth
rates for each year and the (constant) value H*"".
This can be generalized as:
H;u> = H;Lp.(n;(l-B,)).0.9951-1 Equation 3.17
Substituting this for H*Lpinto Equation 3-13 gives:
50
Equation 3.18
We can now compute the number of first bom children in the lead paint homes from
the birth rate information in Exhibit 3-9, and the number of lead paint homes. We can also
express an interim multiplier as the ratio of all first bom children in lead paint homes over the
50 year time frame to the first bom children in lead paint homes in Model Year 1 as:
Abt Associates, Inc. 3-28 Draft, January 10,1994
-------
Equation 3.19
It is useful to note that for this multiplier, the value for H*""drops out. Therefore, the
first birth component of the multiplier for lead paint homes is a function only of the birth rates
over the SO year period and the assumed loss rate of 0.5% per year. From the data in Exhibit
3-11, it can be shown that the value for Mra|U>is:
19.08 Equation 3.20
The summation values for each model year that result in 0.7623 for the numerator are
shown in Exhibit 3-11. The denominator value of 0.03994 is the per home birth probability
for 1994.
As noted above, this interim multiplier addresses only the number of total first births in
lead paint homes. That is, the total number of first births in lead paint homes from 1994 to
2043 is 19.08 times the number of first births in 1994.
To obtain the second component of the multiplier that accounts for additional children
expected to be bom into the homes subsequent to the first child, we begin with the 1994 birth
rate multiplied by SO to obtain the expected number of children over an ensuing SO year
period. This value is:
(0.03994) (50) = 1.998 Equation 3.21
However, since these homes are being lost from the stock at a rate of 0.5% per year, it
is necessary to adjust this number downward to account for homes that will be lost before this
additional SO years is complete. The probability that a home survives through 50 years with
an annual loss rate of 0.5 % is given by:
(1 - 0.005)50 = (0.995)50 = 0.778 Equation 3.22
Abt Associates, Inc. 3-29 Draft, January 10, 1994
-------
Exhibit 3-11. Elements of Summation Component for
i
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
IS
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
B;
0
0.0399
0.0392
0.0386
0.0379
0.0372
0.0366
0.0359
0.0359
0.0358
0.0357
0.0356
0.0355
0.0355
0.0354
0.0353
0.0352
0.0352
0.0351
0.0350
0.0349
0.0348
0.0347
0.0346
0.0345
0.0344
0.0343
0.0342
0.0341
0.0339
0.0338
0.0336
0.0335
0.0333
0.0332
0.0330
0.0329
0.0328
0.0328
0.0328
0.0328
0.0328
0.0328
0.0328
0.0328
0.0328
0.0328
0.0328
0.0328
0.0328
0.0328
n-B,-)
1.0000
0.9601
0.9608
0.9614
0.9621
0.9628
0.9634
0.9641
0.9641
0.9642
0.9643
0.9644
0.9645
0.9645
0.9646
0.9647
0.9648
0.9648
0.9649
0.9650
0.9651
0.9652
0.9653
0.9654
0.9655
0.9656
0.9657
0.9658
0.9659
0.9661
0.9662
0.9664
0.9665
0.9667
0.9668
0.9670
0.9671
0.9672
0.9672
0.9672
0.9672
0.9672
0.9672
0.9672
0.9672
0.9672
0.9672
0.9672
0.9672
0.9672
0.9672
n',»-'>-i>
..
1.0000
0.9601
0.9224
0.8868
0.8532
0.8215
0.7914
0.7630
0.7356
0.7093
0.6840
0.6596
0.6362
0.6136
0.5919
0.5710
0.5508
0.5315
0.5128
0.4949
0.4776
0.4610
0.4450
0.4297
0.4148
0.4006
0.3868
0.3736
0.3609
0.3486
0.3368
0.3255
0.3146
0.3041
0.2940
0.2843
0.2750
0.2660
0.2572
0.2488
0.2407
0.2328
0.2251
0.2178
0.2106
0.2037
0.1970
0.1906
0.1843
0.1783
(0.995)*'1
__
1.0000
0.9950
0.9900
0.9851
0.9801
0.9752
0.9704
0.9655
0.9607
0.9559
0.9511
0.9464
0.9416
0.9369
0.9322
0.9276
0.9229
0.9183
0.9137
0.9092
0.9046
0.9001
0.8956
0.8911
0.8867
0.8822
0.8778
0.8734
0.8691
0.8647
0.8604
0.8561
0.8518
0.8475
0.8433
0.8391
0.8349
0.8307
0.8266
0.8224
0.8183
0.8142
0.8102
0.8061
0.8021
0.7981
0.7941
0.7901
0.7862
0.7822
50
I(fi;i1-Bj)).{Bj)-0.995M
i th Year Quantity
of SunviiBtion
_ ,
0.0399
0.0375
0.0352
0.0331
0.0311
0.0293
0.0276
0.0264
0.0253
0.0242
0.0232
0.0222
0.0212
0.0203
0.0195
0.0187
0.0179
0.0171
0.0164
0.0157
0.0150
0.0144
0.0138
0.0132
0.0127
0.0121
0.0116
0.0111
0.0106
0.0102
0.0097
0.0093
0.0089
0.0086
0.0082
0.0079
0.0075
0.0072
0.0070
0.0067
0.0065
0.0062
0.0060
0.0058
0.0055
0.0053
0.0051
0.0049
0.0047
0.0046
0.7623
Abt Associates, Inc.
3-30
Draft. January 10,1994
-------
Combining these, the expected number of children born into homes having had a first child is:
(1.998) (0.778) = 1.55 Equation 3.23
Therefore, for each first born child in a home in each year, there is an expected
number of 1.55 additional children over the ensuing 50 years.
The total number of children over the full modeling time frame expressed in terms of
the number of first born children each year is:
r = Equation 3.26
1=1
and, as can be seen from the numerators in Equations 3-19 and 3-20,
H;LP. 0.7623 Equation 3.27
the overall multiplier for lead paint homes can be expressed as:
MLP = Cf 2.55. H;"*. 0.7623
2.55.0.7623
= 0.03994 Equation 3.28
= 48.65
Therefore, to estimate the total incidence of adverse effects for children born in lead
paint homes, the incidence of adverse effects obtained for the first year cohort is multiplied by
48.65.
Abt Associates. Inc. 3-31 Draft Jmuary ]Q ]994
-------
Multiplier for homes with no lead paint
Obtaining the multiplier for the homes without lead paint is conceptually similar to the
derivation for homes with lead paint, but the computation is complicated by the nature of the
change in the size of the stock of these homes over the SO year modeling period. For lead
paint homes, the stock was assumed to change with a constant loss rate over the SO year period
of 0.5% per year. As a result, a constant term could be used in the Equations (see, for
example ???) to reflect this rate. In the case of homes without lead paint, however, the annual
rate of change is not constant. As shown in Exhibit 3-9, the total number of privately-owned
homes increases each year from 1994 through 2043. It is assumed that the new homes added
each year are free of lead paint.
It is important to note that the number new homes added to the housing stock in a given
year is not just the difference between the total for that year and the total for the previous year.
Since older homes (both with and without lead paint) are disappearing at a rate of 0.5% per
year, the number of new, no lead paint homes each year is the sum of the difference between
the totals for those years, plus 0.5 % of the preceding year's total stock.
To derive the multiplier for homes without lead paint, it is useful to consider these
homes in terms of SO separate groups. The first group is the non-lead paint homes that are
present in Model Year 1. The remaining 49 groups are the new, non-lead paint homes added
to the stock each remaining year. It is assumed that each of these SO groups has the same loss
rate characteristic as noted in the preceding section for homes with lead paint. That is, it is
assumed that they disappear from the stock with the same annual loss rate of 0.5 %. By using
this approach, we can develop equations similar to those for homes with lead paint (i.e.,
having the constant loss rate of 0.5% per year) to determine the total number of first births in
these homes from the year they are added to the stock through the end of Model Year 50. The
first birth obtained for each of these 50 groups can then be summed, and a multiplier obtained
expressed as the ratio of this sum to the number of first birth children in non-lead paint homes
in Model Year 1.
The number of first births over the full SO years just in the non-lead paint homes that
are present in Model Year 1 is expressed as:
so
Equation 3.29
so
which is similar in form to Equation 3-18.
For the new lead paint homes added in Model Year 2, the number of first births
through Model Year 50 is given by:
Abt Associates, Inc. 3-32 Draft. January 10.1994
-------
so
jo Equation 3.30
1=2
Note that here the summation begins with Model Year 2.
Similarly, for homes that enter the stock in model year 3, the number of first births
over the duration of model year 3 to model year 50 is:
Equation 3.31
1=3
Again, the summation in this case begins with Model Year 3.
The total number of first births in homes without lead paint is, then, the sum of these
first birth calculations for each of the 50 groups of homes:
so
CFB/NLP _ V1 r-FB'NIJ'G) f *• t -»o
iso -2-c«o Equation 3.32
Exhibit 3-12 shows the first birth values calculated for each of 50 groups of non-lead
paint homes over the 50 year time frame, as well as the sum of these, shown there to be
73,648 M children. To convert this to an interim multiplier addressing only first births, both
sides of Equation 3.32 are divided by the number of first births in non-lead paint homes in
Model Year 1:
so so
-,FB/NLP
C -
M /NLP = cFB/NLPd) = JgFB/NLp(i) = ^»NLp(i)>B Equation 3.33
Using the values shown in Exhibit 3-12 of 73,648 for the numerator and 47,203 for the
number of non-lead paint homes in Model Year 1 in the denominator, as well as the value of
0.03994 in the denominator for the per home probability of birth described previously, an
interim "first birth" multiplier is obtained as:
E,«aaon3.34
Abt Associates, Inc. 3-33 Draft. January 10.1994
-------
Exhibit 3-12
Summary of New Non-Lead-Faint Homes and First Births
in These Homes over the 50 Year Model Period
Ywfrt
1
2
3
4
5
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
New, No-Lead-Paint Homes in this
Model Year
47.203
1.596
1.601
1.607
1.612
1.618
1.624
1.478
1.482
1.487
1.492
1.497
1.502
1.506
1.511
1.516
1.521
1.660
1.665
1.670
1.676
1.681
1.687
1.692
1.698
1.703
1.709
1.497
1.502
1.506
1.510
1.515
1.519
1.524
1.532
1.537
1.454
1.462
.470
1.474
.478
1.482
.490
.390
1.393
1.397 1
Total No-Lead-Paint Homes with Tint Births'
Through 2043
Total "First Births' in This Year's New
35.982
1.207
1.201
1,195
1.189
1.182
1.176
1,060
1.053
1,046
1.038
1,030
1,021
1,011
1,001 ^___^
991
1,053
1,039
1,025
1,010
994
960
942
922
902
769
749
729
707
685
661
637
584
556
, 497
438
407
374
339
303
265
185
132
90
46
Abt Associates, Inc.
3-34
Draft, January 10, 1994
-------
Therefore, the total number of first births in homes with no lead paint over the SO year
model period is computed as 39.06 times the number of first births in non-lead paint homes in
model year 1. To obtain the overall multiplier to account for both first birth and subsequent
births in these homes, the additional factor of 2.55 is applied, as described previously for
homes with lead paint. Thus, the overall multiplier is:
MUNLP = (39 06)(2.55) = 89.60 Equation 3.35
Therefore, to obtain the total incidence of adverse effects for children bom into homes
without lead paint, the incidence computed for children bom into those homes in the first
model year is multiplied by 89.60.
3.1.5 Discussion of Results for Baseline Risk Assessment
The discussion of the results of the baseline risk assessment focuses on the blood lead
distribution obtained for the first year of the model period. Because most of the other
measures of adverse health effects (e.g., incidence of IQ point losses) are difficult to interpret
until combined with information on the value of those damages and the societal benefits
attained by avoided those damages, these latter aspects are addressed in subsequent chapters of
this document addressing benefits and benefit-cost comparisons.
As discussed in the preceding sections, the blood lead distribution for the first model
year cohort of children is obtained by separately estimating blood lead distribution geometric
means for 1,271 groups of children associated with homes that are differentiated by their paint,
soil and dust levels (plus paint condition and pica potential). The geometric means are based
on the value obtained from the ffiUBK for children exposed to those conditions continually
from birth through age 7, with the blood lead for age three used as the representative value.
Along with these estimated geometric means, the blood lead distributions for these 1,271
groups of children are all assumed to have a geometric standard deviation of 1.6.
Exhibit 3-13 depicts the distribution of geometric means resulting from this modeling
process. As shown by this histogram, the distribution of the GMs is right skewed, with most
children expected to be in homes where GMs are expected to be in the range of 1 to 6 ug/dl,
with generally decreasing frequencies of GMs at higher values. The maximum GM predicted
for any housing category was 30.1 ug/dl.
Again, these values are the predicted geometric means for 1,271 groups of children.
The overall population distribution of blood lead levels was obtained by first estimating the
number of children in each 1 ug/dl range from 1 to 50 ug/dl (plus a total for those exceeding
50 ug/dl) for each of these 1,271 distributions using the standard normal distribution function.
The geometric standard deviation used in all of these computations was assumed to be a
constant 1.6 for all group, as discussed before.
Abt Associates, Inc. 3-35
Draft, January 10,1994
-------
Exhibit 3-13
Distribution of Predicted Baseline Geometric Means for First Year Cohort
25% T
§
| 20% -
S
1
I 15% 4
= 10% 4
5% -
0%
«— CN en *• in CD r*. CD
Gwmetric Mean PbB (ugfdQ
Exhibit 3-14
Distribution of Predicted Baseline Population Blood Lead Levels for First Model Year Cohort
(1994)
18.00% -
I 18.00% -
^ 14.00%
^ 12.00% -
E 10.00% •
2
iZ 8.00% -
5 6.00% -
3
g 4.00% -
4 2.00% -
000% -
1
Illlllllll...... ....
1 2 3 4 5 6 7 B 9 1011 1Z 13 14 IS 1617 18)92021 22 23 2425 X 27 28 29 30 31 32 3334 36 3637 38 39 40 41 42 4344 45 46 47 48 49 60
Blood lend Laval (ugMD
Abt Associates, Inc.
3-36
Draft, January 10, 1994
-------
Exhibit 3-15
Characteristics of Baseline Population Blood Lead Distribution and a
Distribution with GM = 5, GSD = 1.6
Mean
Geometric Mean
Geometric Standard Dev.
Median
90th Percentile
95th Percentile
% > lOug/dl
% > 15 ug/dl
% > 20 ug/dl
% > 25 ug/dl
From Modeled Baseline From a Distribution with
Distribution GM = 5.0 and GSD = 1.6
6.09 ug/dl
4.06 ug/dl
2.45 ug/dl
3.91 ug/dl
13.32 ug/dl
18.85 ug/dl
15.99%
8.03%
4.37%
2.47%
5.58 ug/dl
5.00 ug/dl
1.60 ug/dl
5.00 ug/dl
9. 13 ug/dl
10.83 ug/dl
7.01%
0.97%
0.16%
0.03%
To arrive at the overall population blood lead distribution for the first year cohort, the
predicted number of children in each 1 ug/dl range for all 1,271 groups were added together.
Exhibit 3-14 provides a histogram of the overall distribution arrived at by this process.
Exhibit 3-15 summarizes the characteristics of this distribution.
The mean shown in Exhibit 3-15 was obtained by assuming a mid-point value in each
of the 50 concentrations ranges (e.g., a value of 5.5 ug/dl was used for the children predicted
to fall between 5 and 6 ug/dl). The geometric mean was obtained in a similar manner by
taking the natural logs of these values, computing the mean of the log values and
exponentiating. The geometric standard deviation was obtained from the normal formula for a
sample standard deviation, using the log transformed values and weighting the differences
between the mean of the logs and the log of the mid-points of each range by the number of
children in that range.
For comparison, values are also included in Exhibit 3-15 for a lognormal distribution
having the parameters of GM=5 and GSD=1.6, which has been suggested by Schwartz
(1993a) as the current population blood lead distribution parameters for children under 7.
Exhibit 3-16 superimposes a histogram based on the GM=5, GSD=1.6 parameters onto the
baseline population blood lead distribution histogram shown previously in Exhibit 3-10.
The central tendency values for the modeled baseline distribution and the GM=5,
GSD=1.6 distribution are slightly lower than those for the GM=5, GSD=1.6 distribution,
reflected by the lower values for the geometric mean and median. The GM=5, GSD=1.6
distribution indicates higher values than the model in the range of abut 5 to 11 ug/dl, after
which point the model values exceed the GM=5, GSD=1.6 distribution values.
Abt Associates, Inc.
3-37
Draft, January 10, 1994
-------
Cumulatively, the modeled distribution is far more right skewed than the GM=5, GSD=1.6
distribution, however. For example, only 7% of the population would be expected to exceed
10 ug/dl from the GM=5, GSD=1.6 distribution, whereas the model predicts over twice that
frequency (16%).
Exhibit 3-16
Comparison of Baseline Blood Lead Distribution with GM=6, GSD=1.6 Distribution
20.00%
18.00%
= 16.00%
§ 14.00%
u
m 12.00%
• 10.00% •
f 8.00% •
M
E 6,00% -
o
* 4.00% -
2.00%
0.00%
• Btjeline Distribution
D GM-5,6SD-1.6Distibution
Blood Lead Level (ug/dD
The GM=5, GSD=1.6 distribution also predicts only about 0.03% of the population
exceeding 25 ug/dl, whereas as the model predicts almost 2.5%. Schwartz (1993a) has
suggested that this value is currently about 0.5% to 1 % based on a preliminary assessment of
the NHANES m data.
It is useful to examine the source of the population above 25 ug/dl predicted by the
model. Exhibit 3-17 provides a summary of the distribution of the blood lead levels above 25
ug/dl predicted by the model among different exposure conditions. As shown there, children
with pica in homes having lead paint in bad condition have the highest individual risk of blood
leads exceeding 25 ug/dl, at 9.2%. By comparison, children in homes with either interior of
exterior lead paint, but without ingesting paint chips have a predicted incidence of blood lead
above 25 ug/dl of about 3 %. Children in homes without lead paint are predicted to have blood
leads above 25 ug/dl only 1.6% of the time.
Notwithstanding the higher predicted rate of high blood leads for children ingesting
lead paint chips, the overall number of such children predicted is only about 9% of the total
having these high blood leads. Other sources have suggested that paint chip ingestion is a
much larger contributor to these high blood leads than these results show. For example,
Shannon and Graef (1992) found that paint chip ingestion was the origin of elevated blood
leads for 80-90% of toddlers with elevated blood leads entering the Lead/Toxicology Program
at Boston's Children Hospital. For infants, by contrast, paint chip ingestion appeared to
account for only about 20% of the cases.
Abt Associates, Inc.
3-38
Draft, January 10, 1994
-------
Exhibit 3-17
Blood Leads > 25 Related to Exposure Conditions
II Total Children
I Exposure Conditions in First Year # > 25 % > 25
[interior lead paint in bad condition,
|[paint chip ingestion
[interior lead paint (may also have
exterior lead paint), any condition,
no paint chip ingestion
llExterior lead paint, no interior lead
llpaint, no paint chip ingestion
|No interior or exterior lead paint, no
llpaint chip ingestion
94,271
1,476,903
406,957
1,937,141
8,688
43,971
12,416
31,548
9.22%
2.98%
3.05%
1.63%
Totals: 3,915,272 96,623 2.47%
Percent of PbB >25 related to paint chip ingestion
Odds ratio of PbB > 25 for Lead paint present : No lead paint
present
8.99%
2.02
Also, Schwartz and Levin (1991) found that the odds ratio of having blood leads
exceeding 30 ug/dl ranged from 5.70 to 12.81 given paint lead exposure. The population
studied was children i n Chicago for 1976 through 1980, when leaded gasoline was still a
major contributor of elevated blood leads. One would expect, therefore, a higher odds ratio
currently than observed in those data. However, the model results indicate an odds ratio of
only about 2.
Except for 8,688 children predicted to have blood leads above 25 ug/dl due to paint
chip ingestion, all other cases of high blood leads in the model are occurring from ingestion of
soil and/or dust having somewhat elevated lead levels. The lead levels in these soils and dusts
for most of the approximately 88,000 non-pica children with high blood leads tend to be above
1,000 ppm. Approximately 76,000 of these cases (86%) occur in cases when dust levels
exceed 1,000 ppm, and 32,500 (37%) occur when soil levels exceed 1,000 ppm.
Since there is some overlap between these, a weighted average of the soil and dust lead
levels was computed, using 55% dust and 45% soil (the assumption made for ingestion in the
ffiUBK model). Over 81,000 children with blood leads above 25 ug/dl (92% of the non-pica
group) are exposed to weighted average soil/dust concentrations of 1,000 ppm or more, and
about 54,000 (61 %) are at or above weighted average soil/dust concentrations of 2,000 ppm.
Abt Associates, Inc.
3-39
Draft, January 10,1994
-------
3.2 MARKET FAILURE
From an economic perspective, one necessary condition for regulatory intervention is the
existence of an inefficiency in the allocation of resources. This inefficiency is commonly labeled
a market failure since the market is the mechanism assumed to make efficient resource
allocations possible. The cause of a market failure can come from one or more of several
sources. These include poorly defined property rights (such as negative externalities, common
property resources, and public goods); imperfect markets for trading property rights (because
of a lack of perfect information or of contingent markets; monopoly power; distortionary taxes
and subsidies and other inappropriate government interventions); and the divergence of private
and social discount rates.2
The occurrence of any of these conditions justifies further inquiry into the need for
government intervention to reduce inefficiencies in the allocation of society's resources. This
section considers whether any of these conditions are linked to excess exposures from lead
contamination in residential soil, dust, and paint. If so, a better understanding of the nature of
the inefficiencies involved may facilitate the design of effective interventions. The specific
intervention considered here is the promulgation of hazard levels as mandated by Section 403.
The strongest case for the existence of a market failure can be built on the apparent lack
of perfect information. The right information is an important prerequisite to the demand for
abatement. The homeowner making the abatement decision has to know the levels of lead in
soil, dust, or paint; know what risks are implied by these levels; know the significances of these
risks; and know what can be accomplished through various forms of abatement. Clearly,
without knowing there is a lead problem, the homeowner will have too low a demand for
abatement. Misinformation on the other attributes of the abatement decision can also distort the
demand for abatement. Research into public views of risk indicate how common misperceptions
are in the assessment of latent risks like those associated with lead contamination. These
misperceptions can be biased upward or downward, resulting respectively in excess and
insufficient demand for abatement. Finally, reliable information on the relative and absolute
effectiveness of different abatement alternatives could be a significant obstacle.
The market itself has not provided a means for correcting the situation. Although
businesses offering testing or abatement services should find it in their vested interest to offer
the kinds of information cited above, this possibility has not closed the information gaps for the
public. One impediment may be public uncertainty about the reliability of the information that
such businesses would provide. Their information may be unreliable because they are not fully
competent to assess the lead contamination and what needs to be done, because the businesses
are subject to moral hazard (which occurs, for example, when a firm tells a homeowner that
there is a lead problem that warrants a certain abatement it can perform when the abatement is
not necessary or suitable), or both. Since many homeowners may lack easy access to
independent sources of information to motivate their abatement decisions, doing nothing may be
the likely response.
This taxonomy was developed from (Axelrad, 1993) and (Boadway, 1979).
Abt Associates, Inc. 3-40 Draft, January 10, 1994
-------
While lamentable, this lack of action is understandable given limits on the time and
money that a homeowner can actually spend on obtaining information needed for many different
decisions. Even though homeowners, as parents, may be deeply concerned about the welfare
of their children - a key target of exposure from lead contamination - there are a host of other
issues besides lead which affect their children's welfare and for which parents need information
to make important decisions. These other information needs compete with the information needs
of the lead abatement decision for scarce household resources. Given how little abatement has
been initiated by homeowners relative to the prevalence of the problem, the likelihood that there
is insufficient demand for abatement and that information gaps contribute to this circumstance
appears to be high. In conclusion, it appears that at least one condition associated with market
failures holds and, consequently, that inefficiencies may characterize the market for lead testing
and abatement.
Before a final determination can be made about the inefficiencies associated with the lack
of information, the costs of spanning the information gaps must be considered. One of the more
important unknown variables in setting hazard levels under Section 403 is what approaches to
making information available will actually get homeowners to test, to consider the abatement
alternatives, and to undertake abatement where appropriate. This analysis represents one step
in shedding light on approaches that reduce possible inefficiencies.
A central question is whether government intervention can make the right information
available to increase the demand for abatement. In attempting to answer this question, it is
helpftil to consider the public good aspect of promulgating hazard levels. To the extent the
public finds them credible and takes steps to measure lead contamination, the hazard levels
provide an independent benchmark for action, lessening at least part of the information needed
to make an abatement decision. As such, hazard levels can qualify as a public intermediate good
since they can be used simultaneously by many households in making their abatement
decisions.3 Whether the hazard levels are a public good or not depends finally on whether the
abatements induced by the hazard levels result in benefits exceeding the costs. If so, the hazard
levels are a public good. If not, they are a public bad. The analysis in this report addresses this
issue directly by attempting to discriminate between the forms and magnitudes of hazard levels
that lead to positive net benefits and those that do not.
Other potential causes of market failure may characterize the persistence of lead
contamination in residences. By undertaking abatement, the owner of a home creates positive
externalities for any occupants outside of his or her immediate family, such as renters, if the
owner is a landlord, and subsequent owners who are occupants of the home. If these renters
and subsequent owners are fully informed about the implications of lead contamination, the
market may adequately compensate the original owner for undertaking lead abatement and no
externality impedes the abatement decision. If they are not fully informed, then the original
owner will not be sufficiently compensated for services provided to the renters and subsequent
owners. Under these circumstances, too few abatements will be undertaken. It is difficult to
measure how large this problem is since it requires information on the stock of knowledge about
lead problems held by tenants and purchasers today and in the future.
The term "public intermediate good" and its definition are adapted from Boadway, 1979.
Abt Associates. Inc. 3-41 ^ Januafy ]Q
-------
Compounding the problem of undercompensation is a divergence between social and
private discount rates, which matters since this analysis anticipates that occupants as much as
fifty years in the future can potentially benefit from the abatement of a given house. Even if
each renter or subsequent owner is willing to pay the full market value of the externality
provided by the original owner's undertaking lead abatement, it is likely that the private estimate
of the present value of these future payments to the original owner will be smaller than the
present value based on the social rate of discount. Consequently, by relying on private
decisions, fewer abatements will be undertaken. The size of this effect could be estimated using
the framework applied in this analysis but has not been conducted to date.
This review suggests there is one or more market failures affecting decisions regarding
the abatement of residential lead contamination. The lack of perfect information is a primary
culprit. However, the evidence is not conclusive. The ultimate determination of a market
failure depends on whether gains in efficiency can be accomplished by some form of
intervention. An allocation of resources is deemed inefficient if someone can be made better
off without making someone else worse off. That is a core question of this analysis. The
discussion of risks from lead contamination does indicate a substantial potential for making
individuals' better off by reducing residential exposures from soil, dust, and paint.
Consequently, one part of the definition of an inefficient allocation has been met. However, to
determine whether the other part of the definition can be satisfied depends on the outcome of
the benefit-cost analysis itself. If the benefits of reducing lead exposures exceed all costs, it is
possible to accomplish this without making others worse off. The costs that have to be
considered include the costs of getting the right people to decide to abate, to choose the right
abatement, and to perform and maintain the abatement in the expected manner as well as the
direct costs of testing and abatement.
Abt Associates, Inc. 3-42 Draft, January 10, 1994
-------
3.3 NEED FOR FEDERAL REGULATION
In the Residential Lead-Based Paint Hazard Reduction Act of 1992 ("the Act"), the
United States Congress identified the elimination of lead-based paint hazards as a national goal.
Congress found that the Federal Government must take a leadership role in building the required
infrastructure, including an informed public, State and local delivery systems, certified
inspectors, contractors, and laboratories, trained workers (§1002(8)). By identifying what
constitutes a lead-based paint hazard (defined as paint, dust or soil conditions that would result
in adverse human health effects), Section 403 creates a crucial portion of the integrated federal
regulatory approach necessary to adequately inform the public of the dangers of lead-based paint,
and to implement other portions of the Act that require mandatory action if a lead-based paint
hazard exists.
The proposed federal identification of lead-based paint hazards will provide guidance to states,
localities and individuals in protecting citizen's rights. Such guidance will promote partnerships
in developing the most cost-effective ways to address lead-based paint hazards. The Act
encourages the individual States to adopt the federal §403 regulations, as well as federal
regulations from other sections of the Act, to the specific conditions that exist in the States by
utilizing existing State and local programs. Further, States have the option of imposing
requirements which are more stringent than the federal procedures. Thus the States may respond
to regional diversity and local social choice by building upon the §403 identification while
guaranteeing the minimum rights of all citizens.
The Act authorizes certain federal expenditures to partially achieve the national goal of
eliminating lead-based paint hazards. Authorized federal expenditures include federal grants for
evaluating and reducing lead-based paint hazards in non-publicly owned or assisted housing, risk
assessments and interim controls in federally assisted housing, and inspections and abatement
of lead-based paint hazards in all federally owned housing constructed prior to 1960. The
Section 403 identification of lead-based paint hazards is necessary to implement these federal
expenditures in a manner that develops the most promising cost effective methods.
Abt Associates, Inc. 3-43 Draft, January 10, 1994
-------
3.4 REGULATORY OPTIONS
Four general classes of instalments are options for government intervention. These are
(1) information provision and labelling, (2) performance or technical standards, (3) bans or
restrictions on use, and (4) economic incentives. The first of these is most closely linked to the
primary condition contributing to a market failure, as described in Section 3.2. Consequently,
directly addressing the lack of adequate information will be the focus of the discussion in this
section, and in the analysis of this report. Examples of how the other three classes of
instruments could be applied are presented but only to illustrate their potential. Further analysis
will be required to determine how viable they are.
3.4.1 Information Provision
A draft regulator's guide on economic incentives under TSCA identifies three
circumstances that are particularly favorable to making the provision of information an
appropriate instrument for intervention (Eyraud, 1993). The first circumstance - that there is
a significant lack of information generating exposure problems - has already been identified as
a strong likelihood. To rectify this circumstance, a corollary condition has to be met. The new
information has to be able to induce exposure-reducing behavior. Information programs are
appealing as a means of intervention in part because they do not impose direct burdens on the
economy. One of the dangers, though, is that the absence of a direct burden will come at the
expense of being ineffective. This does not have to be the case. Collectively, environmental
and other public health programs have amassed substantial experience in learning about what
works and what does not in risk communication. This expertise will have to be tapped to render
any information approach effective.
The second circumstance favorable to effective information provision is one where the
exposure is not created by an externality beyond the exposed individuals' control. In other
words, the affected population has to be able to put the information to good use. While
externalities between current abatement and future beneficiaries were identified as a possible
cause of market failure, these do not prevent information from being effective. For example,
if a house is not abated over the next 30 years, at least the occupants at that future time can
decide to undertake abatement if they have the right information to motivate their decision. This
circumstance appears to apply to the exposures from lead in residential soil, dust, and paint.
There is however at least one major exception. Financial constraints can prevent even the best
informed household from taking effective steps to reduce exposure. As homeowners, households
may not be able to afford abatement. As renters and buyers, they not be able to afford housing
that has been abated. It is important to note, however, that this hindrance is not unique to an
information approach. It is likely to affect other interventions to the same or greater degree.
Abt Associates, Inc. 3-44 Draft, January 10, 1994
-------
The third circumstance favorable to the use of an information approach is one where
other interventions would lead to greater and significant economic impacts. Although its effects
are indirect (working through changes in behavior rather than by direct enforcement), an
information approach does create economic impacts. Whether these economic impacts' are
greater or less than those of other interventions is unknown at this time because other
interventions have not been studied in as much depth.
Scope of Analysis
This analysis focuses on the influence that a particular type of information - hazard levels
- can have on the abatement decisions of homeowners. Promulgating such hazard levels is one
means of implementing Section 403 of TSCA, which calls for EPA to identify lead-based paint
hazards, lead-contaminated dust, and lead-contaminated soil. One objective in promulgating
such hazard levels is to fill part of the information gap that has been linked to sluggish rates of
abatement of lead-contaminated homes. Specifically, these hazard levels are intended to indicate
thresholds at which EPA recommends that certain forms of abatement take place. As such, they
can lower the information costs for homeowners' making a decision about whether to abate and
as a result increase the demand for abatement. It is important to note that by providing such
hazard levels, EPA will not be eliminating the information costs altogether. For example, the
costs of testing for levels of lead in soil, dust, and paint are still substantial. These are
considered in this analysis. Also, any public information campaign to motivate households to
be concerned about and test for lead contamination (analogous possibly to the public campaign
currently being waged for radon) will impose costs. These have not been explicitly considered
here.
The potential form of the hazard levels covers a wide range. At one extreme, the levels
of action would be set uniquely for each individual home, taking its circumstances into account
to determine whether soil, dust, and/or paint should be abated and, if so, how. At the other
extreme, one hazard level would be set for each medium (soil, dust, and paint) for the entire
nation. This analysis constructs different decision rules that reflect each of these extremes. A
brief synopsis is provided here and greater detail in Chapter 6.
The first decision rule considered in this analysis has the highest information requirement.
The public is assumed to have access to the same information that was used in this analysis
(e.g., blood lead levels; reductions achievable from different abatements; the monetary values
of these blood lead reductions; abatement costs; and the levels of lead in soil, dust, and paint
and the condition of paint in the home). Having the highest information requirement has its
rewards. Called the "voluntary optimum," this rule generates the highest net benefits of all
decision rules evaluated since the information can be used to target abatement decisions so well.
By providing guidance to households in the form of hazard levels at which abatement
should be undertaken, EPA can lower the actual information costs for the household abatement
decision. In the other decision rules, households are assumed to have less information overall
and therefore less leeway to choose on their own, instead being induced by EPA's hazard levels
to initiate abatements of certain types. As a result, some households also make mistakes, such
as abating when a benefit-cost analysis does not show that it is warranted or abating one
medium, such as soil, when abating another, such as dust, would be more productive. It follows
Abt Associates, Inc. 3-45
Draft, January 10, 1994
-------
that the net benefits of these decision rales are not as high as those that can be achieved under
the voluntary optimum, where more information is available. What these rules have in their
favor though is their simplicity. The simpler the action levels, the less complicated is the
decision for each household. If one or more action levels are exceeded, then a particular
medium should be abated. At its simplest, one set of levels of action can be promulgated for
the entire nation.
These decision rules are of four types, reflecting varying degrees. The first type of
decision rule addressed paint condition. It assumes that all homes with non-intact paint will
undertake paint abatement. The second type of decision rule included this criterion as well as
a hazard level for each of the three media that are addressed by this rule - soil, dust, and paint.
These are labeled "single-medium constrained" rules. The third type of decision rale
encompasses hazard levels for two media, as well as condition. The fourth type of decision rale
addresses all three media and condition. This analysis compares the net benefits of each of these
decision rales for the abatements that are actually induced by the respective hazard levels.
3.4.2 Other Regulatory Options
Other regulatory instruments may also be effective for addressing the market failures that
have led to inadequate abatement of lead. These alternative instruments have not been examined
to the same extent as the primary instrument considered -information provision. Suggestions for
alternatives that might be investigated further are provided in Exhibit 3-18. The list is meant
to illustrative rather than exhaustive, particularly where economic incentives are concerned. The
feasibility and advisability of these alternatives could vary widely.
Abt Associates, Inc. 3-46 Draft, January 10, 1994
-------
Exhibit 3-18
Other Regulatory Alternatives
Type of Instrument
Possible Application
Labelling
Section 1018 of the Residential Lead-Based Paint
Hazard Reduction Act of 1992 requires the disclosure of
lead-based paint or any known paint hazards in the sale
of target housing (any housing constructed prior to
1978). This provision could be extended to provide
information on soil and dust hazards, not just paint, at
the time of sale of any housing, not just target housing.
Technical and
Performance Standards
Hazard levels could be enforced through performance or
technical requirements. Owners of homes where
children are present would, for example, have to keep
paint in good condition, and reduce and keep soil and
dust contamination below the hazard levels. Technical
standards could specify exactly what abatements are
necessary if hazard levels are exceeded.
Bans and Restrictions of
Use
Restrictions could be placed on the access of young
children to homes where lead contamination is of
concern. These restrictions could include exclusion
from occupying such homes or from spending extensive
amounts of time in them, or prohibitions from accessing
particular areas, such as rooms with paint in
deteriorated condition or bare play areas outdoor where
soil contamination is high.
Economic Incentives
A quota could be established for the numbers of homes
allowed to have excess lead contamination. The quota
could be allocated by a system of marketable
allowances. Homes without allowances would have to
undertake abatement or accept restrictions on their
accessibility to young children.
Abt Associates, Inc.
3-47
Draft, January 10, 1994
-------
3.5 REFERENCES
Axelrad, D. 1993. Guidance on the Preparation of Economic Analyses and Regulatory Impact
Analyses in OPFT. Regulatory Impacts Branch, Office of Pollution Prevention and
Toxics, U.S. Environmental Protection Agency, Washington, D.C., January, pp. 26-27.
Baltrop, D. 1966. The Prevalence of Pica. American Journal of Disabled Children, 112:116,
as cited in HUD, 1990.
Roadway, R.W. 1979. Public Sector Economics. Little, Brown and Company (Inc.), Boston
MA, pp. 29-43.
Bureau of Census, U.S. Department of Commerce, 1991. Annual Housing Survey Components
of Inventory Change: 1973 to 1983.
Bureau of Census, U.S. Department of Commerce, 1992a. Population Projections of the United
States by Age, Sex, Race and Hispanic Origin: 1992 to 2050. Current Population
Reports P2S-1092.
Bureau of Census, U.S. Department of Commerce, 19925. Statistical Abstract of the United
States. 112th Edition.
Centers for Disease Control (CDC) U.S. Department of Health and Human Services, 1991.
Strategic Plan for the Elimination of Childhood Lead Poisoning. February.
Dietrich, K.N., Krafft, K.M., Shukla, R., Bomschein, R.L., Succop, P.A. 1987. The
neurobehavioral effects of prenatal and early postnatal lead exposure. In: Schroeder,
S.R., ed. Toxic Substances and Mental Retardation: Neurobehavioral Toxicology and
Teratology. Washington DC: American Association of Mental Deficiency, 1987: 71-
95 (Monograph No. 8).
Elias, R. 1993. US EPA, Office of Research and Development. Personal communication
through Karen Hogan, US EPA Office of Pollution Prevention and Toxics, July 7.
Eyraud, J. 1993. Economic Incentives Under TSCA: A Regulator's Guide. Volume I.
Review Draft. For the U.S. Environmental Protection Agency, Office of Pollution
Prevention and Toxics, Regulatory Impacts Branch. Washington, D.C., February, pp
3-12.
Lewis and Clark County Health Department, Montana Department of Health and Environmental
Sciences, Center for Disease Control (Public Health Service, U.S. Department of Health
and Human Services), and U.S. Environmental Protection Agency. 1986. East Helena,
Montana Child Lead Study, Summer 1983. Final Report, 36.
Madhavan, S., Rosenman, K.D., and Shehata, T. 1989. Lead in Soil: Recommended
Maximum Permissible Levels. Environmental Research, 49: 136-142.
Abt Associates, Inc. 3-48 Draft, January 10, 1994
-------
Mahaffey, K. 1993. US EPA, Office of Research and Development. Personal communication
through Karen Hogan, US EPA Office of Pollution Prevention and Toxics, July 7.
National Association of Home Builders of the United States (NAHB), 1992. Forecast of
Housing Activity. November.
Schwartz, J. 1993a. US EPA, Office of Policy Planning and Evaluation. Personal
Communication, April 29.
Schwartz, J. 1993b. Beyond LOEL's, p Values and Vote Counting: Methods for Looking at
the Shapes and Strengths of Associations. NeuroToxicology, 14(2-3): 237-246.
Schwartz, J., Levin, R. 1991. The Risk of Lead Toxicity in Homes with Lead Paint Hazard.
Environmental Research, 54: 1-7.
Shannon, MW. Graef, J.W. 1992. Lead Intoxication in Infancy. Pediatrics, 89(1): 87-90.
U.S. Department of Housing and Urban Development (HUD). 1990. Comprehensive and
Workable Plan for the Abatement of Lead-Based Paint in Privately Owned Housing:
Report to Congress. Washington DC. December.
U.S. Environmental Protection Agency (EPA). 1991. Guidance Manual for Site-Specific Use
of the U.S. Environmental Protection Agency Lead Model (Draft). December.3.5
Abl Associates, Inc. 3-49 Draft, January 10, 1994
-------
4. COSTS
4.1 METHOD FOR COST ANALYSIS
The general method used for the cost analysis was to develop a unit cost for the testing
and abatement of lead in each medium (dust, paint and soil) for a single house and then apply
this unit cost to the housing population to be abated. Abatements vary in effectiveness, duration
and cost. To explore these variations, ten abatement strategy scenarios were considered along
with the option not to abate. Abatements occurred just prior to the birth of the first child in a
home. All cost results are presented in present value terms. Nominal costs from several studies
have been used. Because of the narrow time range of the data around the base year (1987-1993
for a 1990 base year) and because the variation caused by changes in money value over time was
within the uncertainty of the estimates, presenting the costs in constant dollars would have little
effect. The next section describes the data available on the costs of testing and abatement for
lead in dust, paint and soil. Section 4.3 shows a sample cost calculation and Section 4.4
presents results.
4.2 DATA
To carry out the analysis, unit costs for testing and high-end and low-end1 levels of
abatement of lead in dust, paint and soil were estimated from the available data. Primary
sources of the data were the "Comprehensive and Workable Plan for the Abatement of Lead-
Based Paint in Privately Owned Housing, Report to Congress" (HUD, 1990), the results from
the Urban Soil Lead Abatement Demonstration Projects, and selected interviews with lead testing
and abatement firms as well as landscapes, commercial cleaning services and hazardous waste
disposal firms. The data were generally point estimates and in some cases represented costs for
somewhat different services. For example, some estimates included relocation of the house
occupants during abatement while others did not. Frequently the exact services included in a
single estimate were not listed. This was particularly true of paint abatements where the
information on the quantity of paint removed and the post-abatement testing were not clearly
described. The analysis presents reasonable high and low cost estimates for each abatement
scenario by combining the estimates obtained from these sources as described below. A medium
value is also presented. In general this value is the mean of the high and low values.2 The
relative merits of each abatement depend not only on the costs of the abatement itself but also
on its effectiveness and duration. Again information from the sources listed above were
combined with the analyst's best judgment to create an estimate. These aspects of abatement
(cost, effectiveness and duration) are discussed separately below for each medium. The costs
The terms high-end and low-end refer to the level of abatement activity taking place. For each level high,
medium and low costs have been estimated that reflects the range of costs for a particular activity level.
The only exception occurs for the medium exterior paint abatement cost. The $5,000 value used was based
on abating a duplex and was considered more accurate than taking the mean of the high ($10,000 for a single family
home) and low ($3,000 for a single unit of a multifamily dwelling) estimates.
Abt Associates, Inc. 4-1 Draft, January 10, 1994
-------
of abating different media are then combined for each of the scenarios to provide a total cost for
each abatement strategy.
4.2.1 Testing and Abatement Costs of Lead in Dust
Dust Testing
The few dust testing costs that have been reported in the literature are highly variable
This is probably due to differences in protocol such as the number of samples and the method
of sampling. The Agency is in the process of developing its own dust testing standards The
current draft standards call for, at a minimum, collection of two dust samples in each of four
rooms and additional samples if the paint is in poor condition. Further samples are required in
common areas. Since there is no standard testing protocol, we combined the draft protocol with
information obtained from lead abatement firms on the cost of testing. One reputable firm in
New England reported a cost of inspection of $128 (for paint,soil and dust) plus $15 per sample
analyzed (Ulluci, 1993). Combining one-third of the $128 inspection cost with the testing costs
of nine samples results in a $178 dust testing estimate. The calculation is shown below.
0.33 x $128 + 9 x $15 = $178 Equation 4.1
A further survey of three lead inspection firms in the Mid-west gave an average labor estimate
of six hours per complete inspection of lead in paint, soil and dust at about $40 per hour The
calculation for dust assumes that one third of that labor was needed to inspect for dust A
second estimate of the total cost of dust testing was calculated using this value for inspection and
the mid-point of the range of sampling analysis costs, $12-27, reported by nine Minnesota lead
analysis firms. The resulting total testing cost, with ten samples required, is $280 as shown
below.
2 x $40 + 10 x $20 = $280 Equation 4.2
Averaging the two calculated values ($178 -I- $280)/2 yields $230 for an estimate of total dust
testing costs.
The calculated testing estimates are very uncertain. The method used for sampling the
dust was not recorded for any of the firms and could be an important cost factor. In addition
the values calculated were based on a limited review of the costs of testing. It is also likely that
a range of quality coexists with the range of price and some quality control monitoring would
be necessary in any dust testing scenario.
Dust Abatement
Cost. The cost of lead dust abatement depends on the thoroughness of the cleaning and
whether rugs, furniture and duct work are replaced. Two sets of estimates are available The
first is from the Urban Soil Lead Abatement Demonstration Project cities which reported costs
of $134-458 in Boston (Weitzman et al., 1992) and $1,216 in Cincinnati (Clark and Bornschein
Abt Associates, Inc. 4_2
Draft, January 10, 1994
-------
1993). The second is from the HUD Report to Congress which estimated a range of dust
abatement costs from $505-730 per cleanup (HUD, 1990). For the high-end abatement option,
the Cincinnati estimate was used as the high cost estimate while the mid-point of the Boston
range, $300, was used as the lower value. The mean of the two was the medium cost value as
shown in Exhibit 4-1. The abatement included moving families off-site while the floors,
woodwork, window wells and some furniture were vacuumed with a high-efficiency particle
accumulator vacuum. Hard surfaces were also wiped with a wet cloth following vacuuming.
The low-end dust abatement strategy is defined as an additional house cleaning, including
general dusting, vacuuming, cleaning bathrooms and wiping window sills. The low cost estimate
of $38 was based on an informal survey of cleaning services-an estimate of about $25 per hour
for cleaning a 3-room apartment in an hour and a half (Rohmer, 1993). The high cost estimate
was calculated assuming a six room house at the same hourly rate for a cost of $75 per cleaning.
The medium cost estimate, $57, was the mean of the high and low cost values.
Effectiveness. The effectiveness of dust abatement depends on the circumstances in
which it is used. A single cleaning, the nonrecurrent dust abatement scenario, will reduce the
dust level to the lower of the dust level reported in the survey or the dust level calculated from
the soil and paint lead levels as described in Chapter 3. Alternatively, recurrent dust abatement
is assumed to reduce the dust lead level to 100 parts per million (ppm).
Duration. The duration of the dust abatement also depends on the circumstance in which
it is used. If there are no external sources of lead the high-end dust abatement is assumed to
be permanent. If there are external sources of lead the high-end dust abatement is assumed to
be done every ten years and the low-end dust abatement is assumed to be done every month to
maintain the abated level.
4.2.2 Testing and Abatement Costs for Lead-based Faint
Interior Paint Testing
The cost of testing for lead-based paint depends both on the type of testing done and the
number of samples taken. The Agency is currently developing a testing protocol to standardize
the number and type of tests required, and once the standards are finalized, variability based on
these factors will be reduced. A typical testing plan requires a visual inspection of paint
condition and determination of the lead content of painted surfaces by either in situ analysis
using a portable x-ray fluorescence (XRF) spectrum analyzer or by off-site laboratory analysis
of paint chip samples. X-ray fluorescence analysis has the advantage of being both lower in cost
and non-destructive when compared with laboratory paint chip analysis. However, XRF
readings are not as accurate as laboratory analysis (HUD, 1990).
Few paint testing costs are reported in the literature. The lead-based paint testing
protocol for the Housing and Urban Development Department's 1989 lead in housing survey
reported a $375 cost per unit (HUD, 1990). The cost included five interior readings and five
exterior readings (using a portable XRF spectrum analyzer), a visual inspection, detailed
estimation of the amount of paint surface and an interview with the owner. The $375 value
Abt Associates, Inc. 4-3 Draft, January 10, 1994
-------
EXHIBIT 4-1
Summary of Abatement Costs for Lead in Dust*
Dust Abatement
Activity
Level
High
Low
Activities
Families moved off-site, hard
surfaces (floors, woodwork, window
wells and some furniture) vacuumed
with a high-efficiency particle
accumulator (HEPA) vacuum. Hard
surfaces were also wiped with a wet
cloth (an oil treated rag used on
furniture) following vacuuming
One additional standard house
cleaning per month including general
dusting, vacuuming, cleaning
bathrooms and wiping window sills
Cost
Estuuflte
Level
High
Medium
Low
High
Medium
Low
Costs
$ Value
$1,216
Recommend using $1200''
$750
$134-458
Recommend using $300°
$75 Based on $25/hr for a 6 room house
$57
$38
1 .5 hours for a 3 room apartment at $25/hour
Source
Clark et al., 1993
Avenge of high
and low
Weitzman ct ftl.
1992
Rohmer, 1993
Average of high
and low
Rohmer, 1993
Hfotfanen
based on scenario. For a single
cleaning, the dust level is the
lower of the HUD reported dust
and that calculated for the soil
and XRF levels. For recurrent
cleanings, the dust level is
reduced to 100 ppm.
(Only used as maintenance of the
high-end dust abatement)
Dun**,
PornuMiil
if no lead
source ia
present
IMMMtk
•nDnui
'Sources for cost estimates are listed in the fifth column of the table.
^Using only two significant figures for the estimate.
'Midpoint of the range rounded to two significant figures.
Abt Associates, Inc.
4-4
Draft, January 10, 1994
-------
reported may thus overestimate the tnie value of interior lead-based paint testing alone because of the
inclusion of detailed inspection and cataloging of the amount of paint present and the collection of
exterior samples. A reputable contractor in New England quoted a price of $55 for 25 XRF readings
and $128 for sample collection and travel as part of a complete lead inspection (Ullucci, 1993) If one-
third of the collection and travel is attributed to the lead paint inspection (assuming soil and dust account
for the remaining two-thirds) the total inspection and testing costs will be about $100 as calculated
below if 25 XRF readings are taken.
0.33 x $128 + $55 « $100 Equation 4.3
As mentioned in Section 4.2.1, a survey of three lead inspection firms in the Mid-west gave an average
labor estimate of a six hours per complete inspection of lead in paint, soil and dust at about $40 per
hour. This information was combined with sample analysis costs from a list of Minnesota de-leading
contractors (primarily from the Minneapolis/ St. Paul area) that showed a range of lead paint sample
analysis costs of $18-27 per sample. An estimated lead testing cost of $360 per house was calculated
by assuming that sampling costs are $23, the mid-point of the range, and that twelve samples were taken
and analyzed (based on a protocol of two per room for four rooms as is likely in the EPA protocol and
four in the entryways or common areas). In addition, it was assumed that lead paint sampling labor is
one-third of the total inspection labor. The value is calculated below.
2 x $40 + 12 x $23 « $360 Equation 4.4
Our final estimated cost of $230 is the average of the estimated costs from the New England and
Minnesota data, ($100 + $360)/2 = $230. The HUD cost value was not included because of the
additional cataloging and interviewing it involved. Lead abatement inspections are still in their infancy
and with more competition, the price could be lowered; however, as standards are instituted, the price
may rise because of training costs and performance requirements. In addition, if the lead paint testing
is not earned out in conjunction with inspections of other media, the cost of travel and collection could
nse. The variation in costs is likely to be large given the range of house sizes and the possible variation
in the number of samples taken.
Exterior Paint Testing
The cost of testing for exterior lead-based paint was estimated at $115, half the value of the
interior lead-based paint testing. Because exterior lead-based paint covered surfaces tend to be larger
and less varied than interior surfaces, fewer samples would probably be needed to establish the extent
of lead-based paint. In this analysis it is assumed that half as many samples are taken outside as are
taken inside. Six samples are taken, one for each of the four sides of a house and two for added
features such as porches and doors, so the sampling time and testing costs should be approximately half
the value of interior lead-based paint testing which requires twelve samples. Four environmental lead
analysis firms were contacted to confirm the two to one ratio of interior to exterior paint testing Only
two of the firms conduct this type of testing on private homes and both use a ratio of approximately
two-to one. Gene Sparrow of Advanced R&D, St. Paul, Minnesota takes about 12 interior samples and
between 4 and 8 exterior samples. Scott Askew of Nova Environmental Services, Incorporated takes
15-20 samples inside and 7-8 samples outside.
Abt Associates, Inc. 4-5
Draft, January 10, 1994
-------
Interior Faint Abatement
Cost. Similar to the cost of testing, the estimated cost of interior lead paint abatement can vary
widely. Some of the variation is caused by differences in the extent of abatement. For example,
replacing the windows is less expensive than removing all the molding, doors and wall paint. Some
of the variation is due to the size of the house and some is due to regional cost variation. The average
cost reported by local abatement programs for lead abatement is $2,100 in 1989 dollars, accoiding to
a Center for Disease Control (CDC) report based on local projects that included abatement of common
areas and exterior when necessary, costs of materials, labor insurance, overhead, woricer protection and
cleanup (CDC, 1991). The estimated costs in the HUD Report to Congress are much higher, except
in the "Units with Interior Lead-Based Paint (LBP) Only" category, which estimates $1,808 per unit
of removal. HUD also found that over 50 percent of all housing units with lead-based paint could be
abated by removal for under $2,500 per unit. The Boston Globe reported the average cost of a de-
leading job for a single unit in Massachusetts as $3,450 in 1990 where valid methods of abatement
include stripping, removal or enclosure of the lead paint (Carroll, 1993). Based on the values for
interior paint abatement estimated in a Michigan grant proposal to the CDC, the complete abatement
cost estimates range from $16,000 per unit for a ten unit apartment building to $23,000 per unit for a
single family house. The proposal also reports an average cost of $15,000 for units abated by Maryland
and Minnesota under their state plans. No distinction is made on this last value between interior and
exterior lead-based paint abatement. Telephone calls to a small sample of de-leading contractors in the
United States produced a range of estimates from $4,000 to $12,000 for a complete interior lead paint
abatement. As part of the Boston study of the effects of soil abatement, 46 homes were de-leaded with
a reported average cost of $7,500 for interior paint de-leading. If the costs of moving, storage,
alternative housing, inspections, monitoring and clearance samples are added, the costs could rise to
$10,000-$10,500 (Weitzman, 1992). A typical interior lead-based paint abatement in Maryland was
reported to be about $12,000 although there is considerable variation (Morris, 1993). Exhibit 4-2
summarizes the various cost estimates for interior paint abatement.
The high-end paint abatement option for our model is complete removal of lead-based paint
including full abatement of windows, doors, woodwork and walls plus a high-end dust abatement. The
range of abatement costs presented above reflects differences in house size, quantity of lead paint
present and abatement technique used. For this analysis we developed high, medium and low costs for
the high-end abatement activities. Typical low cost for high-end abatement was $7,500 while $12,000
was considered representative of recent high cost experience in complete interior lead paint removal
including disposal of debris as a non-hazardous waste (excluding dust abatement). The medium cost
($9,750) reflects the mean of high and low values for high-end paint abatement. If dust abatement
(Section 4.2.1) is included, the total low, medium, and high cost high-end paint abatements are $8,250,
$10,500, and $12,750 respectively. High-end dust abatement is included in this calculation to ensure
that the full abatement effectiveness is achieved. These values are shown in Exhibit 4-3.
While complete removal of all lead paint represents a high-end abatement, replacement of
windows reflects a typical partial abatement. Paint on windows is preferentially prone to peeling and
flaking due to movement of the sash and the increased exposure to sunlight and weathering. The cost
of replacing a window ranges from $131 for an aluminum window to $262 for a wooden sliding window
(R.S. Means, 1987). An informal interview with Vance Morris of Maryland's Department of Housing
and Community Development confirmed $200 as an average value for a window replacement (Morris,
1993). Information on the average number of windows per house is difficult to obtain since this
Abt Associates, Inc. 4-6 Draft, January 10. 1994
-------
EXHIBIT 4-2
Summary of Interior Lead-based Faint Abatement Costs
Cost
$1,808
$2,100
$2,500
$3,450
$4000-$12,000
$7,500
$10,000-$10,500
$12,000
$15,000
$16,000 per unit
for a ten unit
apartment
$23,000 for a
single family
house
Activities Covered
Removal of lead-based paint in units with
only interior lead-based paint
Average lead abatement cost reported by
local abatement programs
Value at which 50% of all housing units with
lead-based paint could be abated by removal
Average cost of a de-leading job for a single
unit in Massachusetts in 1990 by stripping,
removal or enclosure
Complete lead abatement, method
unspecified
Average cost of de-leading 46 homes in
Boston
Average cost of de-leading 46 homes in
Boston if moving, storage, alternative
housing, inspections, monitoring and
clearance samples are added
Typical interior lead-based paint abatement
in Maryland, method unspecified
Stated as average cost for unit abated in
Maryland and Minnesota, may include
exterior paint abatement as well
Interior lead-based paint abatement cost
estimates
Source
HUD, 1990
CDC, 1991
HUD, 1990
Carroll, 1993
Sample of U.S. de-
leading contractors
Weitzman, 1992
Weitzman, 1992
Morris, 1993
Michigan, 1991
Michigan, 1991
Abt Associates, Inc.
4-7
Draft, January 10, 1994
-------
EXHIBIT 4-3
Summary of Abatement Costs for Lead in Paint*
Activity
Level
High
Low
Activities
Full abatement of windows, doors,
woodwork and walls, includes mean
high-level dust abatement necessary to
ensure abatement to effectiveness
level postulated values do not include
disposal of abatement debris as a
hazardous waste
Replace 15 windows @ $200 each
plus mean high-level dust abatement
to ensure abatement to effectiveness
postulated
Replace 10 windows ® $200 each
plus mean high-level dust abatement
to ensure abatement to effectiveness
postulated
Replace 5 windows @ $200 each,
plus mean high level dust abatement
to ensure abatement to effectiveness
postulated
Cost
Estimate
Level
High
Medium
Low
High
Medium
Low
Costs
$ Value
Elements: $12,000b + $750°
Total: $12,750
Elements: $9.750* + $750°
Total: $10,500
Elements: $7,500* + $750°
Total: $8,250
Elements: $3,000* + $750°
Total: $3,750
Elements: $2,000* + $750°
Total: $2,750
Elements: $1000h + $750°
Total: $1,750
Source
Morris, 1993
Average of high
and low estimates
Weitzman, 1992
Morris, 1993
irf!fn«i*tV«Mnn»
ttllCCDVCBCSS
Abated to a paint
lead level
removal of all
paint plus
removing all the
paint ingestion for
pica children
Abated to a dust
lead level that is
91. 4% of the
original dust level
plus removing ell
the paint ingestion
for pice children
(See Section 4.2.2)
Duration
•J^______»
rGllllUWIH
Permanent (for
windows)
snown in me mm co limn 01 me tame.
0 High-cost estimate for High-end interior lead-based paint abatement.
c Average-cost estimate for High-end lead dust abatement.
d Mediu
stesbi
! for High-end interior
lead-hiued
paint abatement.
e Low-cost estimate for High-end interior lead-based paint abatement.
f High-cost estimate for Low-end interior lead-based paint abatement.
> Medium-cost estimate for Low-end interior lead-based paint abatement.
h Low-cost estimate for Low-end interior lead-based paint abatement.
Abt Associates, Inc.
4-8
Draft, January 10, 1994
-------
information is included in neither the American Housing Survey nor the decennial census. The low
medium and high estimates were calculated using 5, 10, and IS windows3, respectively; the resulting
overall cost estimates of low-end paint abatement are $1000, $2,000 and $3,000. Including high-end
dust abatement brings the totals to $1,750, $2,750, and $3,750. The equation is shown below.
LPAC = Wx $200 + $750 Equation 4.5
where:
LPAC = Low-end Paint Abatement Cost, and
W = Number of windows.
The Michigan proposal estimates a value of between $3000 and $6500 for replacement of windows in
one unit of a ten unit apartment and a single family dwelling, respectively. The lower costs are used
for the modelling effort with the understanding that the actual value could vary greatly depending on
the type and number of windows replaced. These values do not explicitly assume debris disposal costs
but if the waste is considered non-hazardous its disposal cost is within the uncertainty of the cost
estimate. The Agency has not yet made a decision on whether paint abatement waste will be exempted
from the Resource Conservation and Recovery Act hazardous waste definitions. Under current
regulation, only those portions of the waste that fail the Toxicity Characteristic Leaching Procedure
(TCLP) for lead are considered hazardous waste.4 Disposal costs depend on the quantity being
discarded. If each abatement produced an average of 217 pounds of hazardous waste, as was generated
per housing unit in the Housing and Urban Development abatement demonstration project, and cost the
reported $255 to discard, the paint abatement costs would increase between two and nine percent
depending on whether the low-end or high-end abatement scenario was chosen (US EPA, 1992a).
Effectiveness. The removal of all the lead-based paint in a home (the high-end paint abatement
option) was assumed to have an effectiveness equivalent to eliminating the paint contribution to lead dust
as well as removing the possibility of high blood lead levels resulting from pica (the consumption of
non-food items). However, little data are available regarding the effectiveness of abatement techniques.
The HUD demonstration study did conduct follow-up on lead paint abatement effectiveness but the
Personal communication with Gopaul Ahluwalia at the National Association of Home Builders, 1993. Mr.
Ahluwalia estimated the average number of windows in a new single family home (17) and the average number per
unit in a new multifamily apartment building (9) based on a recent construction-material-usage data base. There
are two trends in home building that need to be considered before using these estimates as the number of windows
present in homes built prior to 1980 (our population of interest for lead abatement). The first is that new homes
are larger now than in the past and second, homes are currently built with more windows to increase light in the
house. No quantitative information was available about the latter trend but the former trend was compensated for
by multiplying the 1993 average number of windows by the ratio of the avenge square feet per single family home
in 1980 compared to that estimated for 1993 (1600 sq. ft./2100 sq. ft.) resulting in an estimate of 13 windows per
average single family home and 7 windows per unit for a multifamily dwelling built in 1980. The range investigated
in the model 5-15 brackets these estimates. Obviously, very large single family homes can have many more
windows than the reasonable high value of 15 used in this analysis.
Personal communication with Rajani Joglekar, Office of Solid Waste and Emergency Response, US EPA, 1993.
Abt Associates, Inc. 4-9 Draft. January 10. 1994
-------
resulting values are reported as dust lead loadings which are not directly translatable into the dust lead
concentration values used in our model.
Window replacement was estimated to contribute 8.6 percent of the benefits that would accrue
from abatement by removal. The dust level is directly tied to the benefits thus the post window
abatement dust level is 91.4 percent (100 - 8.6 percent) of the original dust level. This value was based
on a calculation from the HUD survey data that showed about 8.6 percent of the lead-based paint
accessible to children was on windows. This value may underestimate the effectiveness of the
abatement in homes where only the windows have lead-based paint or contribute to the lead dust levels.
However, the value may overestimate the effectiveness in houses with a large amount of accessible lead
paint in places other than windows. In addition, the threat of elevated blood lead levels due to pica was
removed from homes that had window replacement.
Duration. Removal of lead-based paint from a house results in permanent abatement as long
as all the paint was removed as postulated in the high-end option. There should be no further lead
contribution from interior paint under this option. In the low-end option where the windows were
replaced, there is also permanent removal of a portion of lead-based paint. Thus, the overall
assumption is permanent duration of paint abatement.
Exterior Faint Abatement
Cost. The cost of exterior lead-based paint abatement depends on many factors including the
size of the house, whether or not the dwelling is a multi-unit abode and the method of abatement. The
estimates for low, medium and high costs were obtained from the Michigan report. The report listed
costs of re-siding for a unit of an apartment complex, a duplex, and a single family home, as $3,000,
$5,000 and $10,000 respectively; these values were taken as low, medium and high exterior paint
abatement cost estimates (Michigan, 1991). The HUD Report to Congress shows exterior abatement
costs for homes that have only exterior lead-based paint as $2,841 for encapsulation and $4,791 for
removal. However, if the costs quoted in HUD for abating interior lead-based paint are subtracted from
the costs of abating units with both interior and exterior lead-based paint, the cost of abating exterior
lead-based paint is estimated to be $6,600. While the latter value may overestimate the cost because
some of the difference may be due to increased interior lead-based paint abatement costs, it reflects
some of the variation among homes. The Boston experience with de-leading the interiors of 46 homes,
as reported in Weitzman, is $5,700 per home (Weitzman, 1992). These values confirm the range of
values chosen as our estimates.
Effectiveness. Abatement of exterior lead-based paint in this analysis is used only to ensure that
the soil abatement effectiveness projected below (in Section 4.2.3) is achieved. As a consequence, the
effectiveness is not independent of the soil abatement value. However the contribution of exterior lead-
based paint to soil has been assumed to be eliminated by the abatement as described in Exhibit 4.4.
Duration. Exterior lead-based paint abatement is assumed to be permanent because all paint has
been removed.
Abt Associates, Inc. 4-10 Draft, January 10, 1994
-------
4.2.3 Testing and Abatement Costs of Lead in Soil
Soil Testing
As with paint, soil testing costs can vary considerably based on the number of samples taken,
the method of analysis used and the region of the country. The cost of soil sampling reported by
Baltimore in the Urban Soil Lead Abatement Demonstration Project is $825 for the 30 samples that
were taken and analyzed for each house (Farrell, 1993). The HUD survey took only nine samples per
house but did not report the cost of sampling (HUD, 1990). The Agency is in the process of developing
a standardized testing protocol. For the purposes of this study a four sample protocol was adopted,
based on two composite dripline samples and two play area samples. If there is more than one play
area, then the sampling cost would rise. Estimated costs are based on two data sources. The first is
a reliable lead contracting firm in New England that provided information indicating that each soil
sample analysis costs $16 and overall inspection costs $128 for paint, lead and soil (as discussed in
Section 4.2.1). By assuming that one-third of the inspection time is devoted to soil sampling and
assuming four soil samples are taken, the soil testing costs are about $106 as shown below.
0.33 x $128 + 4 x $16 = $106 Equation 4.6
The second data source is a list of laboratories and consultants in Minnesota that provide environmental
lead analysis. This list shows a range of $12-32 for sample analysis costs. Assuming two hours of
labor (at $40 per hour) for inspection and sampling of soil, and analysis of four samples at $22 each,
(the mid-point of the range), an estimate of $170 for soil testing results is calculated below.
2 x $40 + 4 x $22 * $170 Equation 4.7
We used the average of the two calculated values, ($106+170)72 = $138, as the soil testing cost. The
sampling for Baltimore was not included in the estimate because we lacked information on how the
analysis was conducted. However, the wide range of the estimates suggests that further investigation
may be warranted.
Soil Abatement
Cost. The costs of soil abatement are tied to the area treated, the method used and whether or
not the waste is considered hazardous under the Resource Conservation and Recovery Act (RCRA).
Residential soil abatement is a relatively new industry and no standards have been established on what
constitutes a complete abatement. The three participants in the Urban Soil Lead Abatement
Demonstration Project, (Boston, Baltimore, and Cincinnati) provided costs for a high-end abatement
process. The procedure involved removal of six inches of top soil, installation of a barrier (Boston
only), disposal of the contaminated soil as non-hazardous waste and replacement with new soil with less
than 150 ppm lead (or less than 50 ppm in Baltimore and Cincinnati) (Elias, 1993). The costs ranged
from $2,400 per property in Cincinnati, to $4,896 per property in Baltimore, to $6,600 and $9,600 per
property under two separate contracts in Boston (Bias, 1992). The higher of the two Boston studies
was chosen as the high representative of the costs of high-end residential soil abatement because of the
use of the barrier. Because simultaneous dust abatement is necessary to achieve full effectiveness for
Abt Associates, Inc. 4-11 Draft, January 10. 1994
-------
soil abatement, this value was combined with the high-end dust abatement cost, described in Section
4.2.1, for a total of $10,350. The lower cost estimate used the Baltimore value of $4,896, which,
including high-end dust abatement sums to $5,646. The Baltimore study was preferred in this instance
to the Cincinnati study because the latter study primarily abated playgrounds rather than residential
lawns. Exhibit 4-4 summarizes the soil abatement cost information.
In certain cases, three other costs were added to the high-end soil abatement. The first is the
cost of exterior paint abatement if the house has an exterior XRF reading greater than 1 mg/cm2. The
exterior paint abatement costs have been described above. Because exterior lead-based paint is a
primary source of lead soil contamination, it is necessary to abate the exterior paint to maintain the low
soil lead levels achieved by soil abatement. Two additional costs are incurred if the soil must be
transported and disposed as a hazardous waste under RCRA. Soil is considered hazardous if it fails the
Toxicity Characteristic Leaching Procedure (TCLP) for lead. Many factors affect the leaching
characteristics of lead in soil including the soil type and pH. As a conservative estimate we assumed
soil with lead levels greater than 2000 ppm will fail the test. Thus houses with soils that fail the TCLP
will incur the additional soil disposal and transportation costs. Based on an interview with Chemical
Waste Management the cost of stabilization and landfilling of lead contaminated soil is between $230
and $275 per ton (Donegan, 1993).
The high cost option uses the quantity of soil removed from lawns under the 1989 Boston
contract (41 cubic yards) (Weitzman, 1992). The soil abatements from the Demonstration Project
already include the cost of disposing of the soil as a non-hazardous waste. To avoid double counting
the disposal the cost of $250 per ton for hazardous waste was reduced by $35 per ton for landfilling
non-hazardous waste, which results in a cost of $215 per ton. The $215 per ton is multiplied by 41
cubic yards and by 1.3 tons per cubic yard for a total disposal cost of $11,460. For the lower cost
estimate, the Baltimore experience of an average of 14.2 cubic yards of soil removed per lawn was
used; this resulted in disposal costs of $3,968 (Farrell, 1992). A sample calculation is shown below.
ton
yd3 x 1.3 21 x $215 per ton = $3,968 Equation 4.8
yd
A typical transportation cost for hazardous waste is $425 per 22 ton dump truck for under 100
miles (Donegan, 1993). Assuming the truck is full and that the disposal site is within 100 miles the cost
is $425/22 tons equalling $19.32 per ton. This is $19.32 times 41 cubic yards times 1.3 tons per cubic
yard, equalling $1,040 for the high end cost. A sample calculation is shown below.
$19.31 per ton x 14.2 yd3 x 1.3 — * $360 Equation 4 9
yd3 ^
The low-end abatement is a resodding of the lawn. The critical determinant of this cost is the
area resodded. The high-end estimate was calculated using the average size of lawn abated in the
Boston experience of the Urban Soil Lead Abatement Demonstration Project, equal to 2,141 square feet.
The average cost of resodding, including preparing the ground and applying the sod but not removing
Abt Associates, Inc. 4-12 Draft, January 10, 1994
-------
EXHIBIT 4-4
Summary of Abatement Costs for Lead in Sou*
Soil Abutment
Activity
Level
High
Activities
Removal of 6 inches of top (oil,
barrier installed and new soil
(tested at under 150 ppm) used as a
replacement, resodding plus high-
end dust abatement, plus hazardous
waste disposal of soil > 2000 ppm
plus the transportation costs for
disposing of hazardous soil.
Average of High and Low Costs
As above without the barrier and
using the Baltimore cost of
abatement plus dust abatement plus
10.9 cubic meters (14.2 cu.yd) soil
removed
Cost
Estimate
Level
High
Medium
Low
Costs
$ Value
>2000 ppm
Elements6: $9,600 + $750 +
$ll,460d + $1.040+ $10,000
(Only if exterior paint is
present)
Total: $22,850 + $10,000
(Only if exterior paint is
present)
>2000 ppm
Elements6: $7,248 + $750 +
Vl.l\4 + $700 + $5,000
(Only if exterior paint is
present)
Total: $16,412 + $5,000 (Only
if exterior paint is present)
> 2000 ppm
Elements*: $4,896 + $750 +
$3,968d + $360 + $3,000
(Only if exterior paint is
present)
Total: $9,974 + $3,000 (Only
if exterior paint is present)
£2000 ppm
Elements6: $9.600 + $750 +
$10,000 (Only if exterior paint is
present)
Total: $10,350 + $10,000 (Only
if exterior paint is present)
£2000 ppm
Elements': $7,248 + $750 +
$5.000 (Only if exterior paint is
present)
Total: $7,998 + $5,000 (Only if
exterior paint is present)
£2000 ppm
Elements11: $4,896 + $750 +
$3,000 (Only if exterior paint is
present)
Total: $5,646 + $3.000 (Only if
exterior paint is present)
Source
Weitzman, 1992 for soil
abatement protocol, Elias,
1993 for soil abatement
cost and Donegan, 1993
for hazardous waste
costs and Michigan, 1991
for exterior lead-paint
abatement
Avenge of high and low
values for soil abatement
and Michigan, 1991 for
exterior paint abatement
costs
EPA, 1993 for amount of
soil and Elias, 1993 for
cost of soil abatement and
Donegan, 1993 for cost of
disposal and transport of
hazardous waste and
Michigan, 1991 for
exterior paint abatement
costs
Effectiveness
Soil is abated
to 100 ppm
Duration
Pornuiiwnt
Abt Associates, Inc.
4-13
Draft. January 10, 1994
-------
EXHIBIT 4-4
Summary of Abatement Costs for Lead in Soil*
SoQAbal
Activity
I JMMj
Level
Low
count
Activities
Resodding 2.141 iquara feet ®
$1 .4S/square foot plus mean high-
end dud abatement
Activities include resodding a little
grading but no removal of existing
grass
Avenge of high and low
Resodding 770 square feet 9
$1 .45/squire foot plus mean high-
end dust abatement (10.9 cu.m. per
property x 3S.3 cu ft/cu.m at 6")
Cost
Estimate
Level
High
Medium
Low
Costs
$ Value
Elements': $3,104 + $750
Total: $3,854
Elements': $2,1 10+ $750
Total: $2,860
Elements*: $1,116 + $750
Total: $1,866
Source
Weitzman, 1992 for size of
lawn and Degen, 1993 for
cost of resodding
Average of high and low
EPA, 1993 Baltimore
experience for size of lawn
and Degen, 1993 for cost
of resodding
Effectiveness
Soil is abated
toSOOppm
Duration
5 years
* $9.600 = High-cost/High-end soil; $750 = Avenge-cosl/High-end dust; $11.460 = High-cost soil disposal; $1,040 = High-cost transport; $10,000 = High-cost exterior lead paint abatement
« $9.600 = High-cost/High-end soil; $750 = Average-cost/High-enddust; $10,000 = High-cost exterior lead paint abatement.
« Assumes removal of 6 inches and is the difference of hazardous waste dispossl and regular landfilling since the Urban Soil Lead Abatement Demonstration Project values include soil disposal
$7,248 = Medium-cost/High-endsoil; $750 = Average-cost/High-enddust; $7,714 = Medium-cost soil disposal; $700 = Medium-cost transport;
$5,000 = Medium-cost exterior lead paint abatement.
f $7,248 = Medium-cost/High-endsoil; $750 = Average-cost/High-enddust; $5,000 = Medium-cost exterior lead paint abatement.
* $4,896 = Low-cost/High-end soil; $750 = Average-cost/High-end dust; $3,968 = Low-cost soil disposal; $360 = Low-cost transport; $3,000 = Low-cost exterior lead paint abatement
? $4,896 = Low-cost/High-end soil; $750 = Average-cost/High-end dust; $3.000 = Low-cost exterior lead paint abatement.
I $3.104 = High-cost/Low-endsoil lead abatement; $750 = Average-cost/High-enddust abatement.
J $2.110 = Medium-cost/Low-end soil lead abatement; $750 Average-cost/High-enddust abatement.
k $1,116 = Low-cost/Low-end soil lead abatement; $750 = Average-cost/High-enddust abatement.
Abt Associates, Inc.
4-14
Draft, January 10, 1994
-------
existing sod, is about $1.45 per square foot based on informal interviews with landscapes. The
total high cost is thus $3,104. The low estimate of $1,116 is calculated using the same cost per
square foot applied to the average size of the lawn abated in Baltimore (770 square feet); this
area was calculated based on the amount of soil removed. The medium estimate is the mean of
the high and low estimates. Including dust abatement, the totals are $3,854, $2,860, and $1,866
as shown in Exhibit 4-4. The high cost estimate is shown below as a sample calculation.
$1.45 x 2,141 ft1 + $750 = $3,854 Equation 4.10
These estimates may be low compared to resodding an entire suburban lawn of 2,500 to 7,000
square feet (a typical size range) but may represent a realistic value if only a portion of the lawn
is resodded. In addition, the cost of resodding can vary based on the physical layout of the
property. For example, if trees are present, additional labor will be required to sod around such
obstacles raising the $1.45 per square foot estimate.
Effectiveness. The high end abatement is assumed to reduce the soil lead level to 100
ppm, the average of the lead levels in the replacement soil in the Urban Soil Lead Abatement
Demonstration Project (US EPA, 1993). The low-end estimate for the effectiveness of resodding
is 500 ppm since in the past, the Agency has used this value as an action level. No
measurements of the effectiveness of resodding were found.
Duration. The high-end soil abatement is permanent as long as there are no external sources
of lead. The low-end abatement (resodding) was assumed to last five years based on
expectations of sod durability obtained from landscaping firms in an informal survey.
Effect of Bare Soil on Soil Abatement Cost Scenarios. The cost of the high-end soil
abatement scenario may be slightly higher for covered soil than for bare soil if extra sod must
be removed from the yard; the low-end soil abatement may not be needed at all if the sod
covering the soil is intact. The effectiveness of the two abatement scenarios as measured by the
post-abatement soil lead level is unaffected by the initial condition although resodding would not
achieve significant reduction in exposure if the soil were already covered. Initial soil coverage
conditions would have no affect on the abatement duration. As mentioned in Chapter 3, data
limitations make us unable to specifically address bare and covered soils. Qualitative discussions
of the implications of this limiation for the benefit and benefit-cost results are contained in
Chapters 5 and 7.
4.2.4. Combined Abatement Scenario Costs
Exhibits 4-1 through 4-4 showed the costs of abatements in each medium. These abatements
were combined into ten plausible scenarios that are summarized in Exhibits 4-5 and 4-7 below.
In practice there will be considerable variation in the costs. A formal survey of current lead
abatement practice could provide better estimates of the true costs; in addition, a controlled study
could provide better data on the effectiveness and duration of the abatement techniques. In
further work, a sensitivity analysis is planned to address the uncertainty in these estimates. The
exhibits show the ten abatement strategies analyzed in this report (the eleventh "no abatement"
Abt Associates. Inc. 4-15 Draft. January 10. 1994
-------
has no cost) and the lifetime cost of the strategy in the year of abatement. A seven percent
discount factor was used to calculate the net present values. The values are presented in
Exhibits 4-5 and 4-6. The calculation of abatement costs that occur over the homes' lifetime,
low-end soil and recurrent dust, were further discounted by 0.5% per year to account for the
removal of homes from the housing stock.
Because damages to children occur only through age seven in the main model analysis,
abatement should only occur when a child under the age of seven is resident in the home. The
values used here include all years, once an abatement is undertaken. This provides an accurate
estimate for those abatements that are permanent but it overestimates the cost of recurrent
abatements because the abatement activity, such as resodding, should only occur when a child
under seven is resident.
4.2.5 Enforcement Costs
There are no enforcement costs associated with Section 403. Section 403 requires that the
Agency set hazard levels for lead in paint, soil and dust that will be used in other sections of
Title X to trigger abatements. The enforcement costs of these actions, however, are not
attributable to Section 403 but to the section of the rule requiring the abatement. All abatement
activity under Section 403 is voluntary and thus incurs no enforcement cost.
4.2.6 Implementation Costs
The implementation costs associated with Section 403 are of two types. The first is the cost
of setting and promulgating the Section 403 hazard levels themselves; a negligible cost compared
to the funding appropriated in Tide X for abatement ($250 million in 1994). The second
implementation costs would be those of states or localities that voluntarily use the hazard levels
set by the Agency as action levels in their own lead management programs. The size of these
costs depend on the current level of activity at the state and local level, whether the hazard levels
the Agency sets are above or below those of the programs in place, and the number of programs
that implement the hazard levels. If the Agency levels are more stringent than current practice,
implementation costs could be significant; however if the Agency levels are higher than those
in practice implementation costs will be negligible. No quantitative evaluation of the
implementation costs was attempted because reliable information on current and expected future
programs at the state and local level was not available. If implementation costs are proportional
to the number of homes affected, which could be the case if state or local authorities decided
to track homes to assure abatement, then the inclusion of implementation costs in the benefit-cost
analysis would favor higher hazard levels over lower ones, all other things equal, since the
number of homes to be tracked would be lower under the latter than the former.
Abt Associates, Inc. 4-16 Draft, January 10, 1994
-------
EXHIBIT 4-5
Summary of Unit Abatement Costs by Scenario
Single Media Abatement
Scenarios
Activities
Unit Cost
High-end Paint Abatement
Paint abatement and dust
abatement
High Estimate - $12,750
Medium Estimate- $10,500
Low Estimate- $8,250
Low-end Paint Abatement
Window abatement and
dust abatement
High Estimate - $3,750
Medium Estimate- $2,750
Low Estimate - $1,750
Non-Recurrent Dust
One high-end dust
abatement
High Estimate - $1200
Medium Estimate - $750
Low Estimate - $300
Recurrent Dust
High-end dust every 10
years, Low-end dust
monthly
Ten year Monthly Total Cost Estimate3
High
High
Low
Low
Medium
Medium
High
Low
High
Low
High
Low
$14,621
$8,524
$12,925
$6,828
$13,773
$7,676
High-end Soil Abatement
Complete removal and
replacement of soil
(includes disposal and
transportation of
hazardous waste for
concentrations above 2000
ppm and also exterior
paint abatement with
additional cost for those
houses with exterior paint)
High cost
>2000 ppm and exterior paint - $32,850
£2000 ppm and exterior paint - $20,350
>2000 ppm and no exterior paint - $22,850
£2000 ppm and no exterior paint - $10,350
Medium cost
>2000 ppm and exterior paint - $21,412
£2000 ppm and exterior paint - $12,998
>2000 ppm and no exterior paint - $16,412
£2000 ppm and no exterior paint - $7,998
Low cost
>2000 ppm and exterior paint - $12,974
£2000 ppm and exterior paint - $8,646
>2000 ppm and no exterior paint - $9,974
£2000 ppm and no exterior paint - $5,646
Low-end Soil Abatement
Resod every 5 years,
High-end dust abatement
at first resodding
High3 - $10,669
Medium3 - $7,493
Low1 - $4,316
rate.
Includes discounting at 7% to the year of abatement and removal of housing stock at a 0.5%
(Total cost would be further discounted to 1994.)
-------
EXHIBIT 4-6
Summary of Combined Abatement Strategies
Net Present Value Costs for Combined Scenarios
Scenario
High-end Paint and High-end Soil
High-end Paint and Low-end Soil
Cost Elements
>2000ppm
High*
$12.000 + $750 + $9,600 + $750 +
$11.460 + $1,040 + $10,000 (Only if
exterior paint is present)
Medium1*
$9,750 + $750 + $7,248 + $750 +
$700 + $7,714 + $5,000 (Only if
exterior paint is present)
Lowc
$7.500 + $750 + $4.896 + $750 +
$3,968 + $360 + $3.000 (Only if
exterior paint is present)
£2000 ppm
High
$12,000 + $750 + $9,600 +
$750+ $10,000 (Only if
exterior paint is present)
Medium
$9,750 + $750 + $7,248 +
$750 + $5,000 (Only if
exterior paint is present)
Low
$7,500 + $750 + $4,896 +
$750 + $3,000 (Only if
exterior paint is present)
Highd
$12,000 + $750 + $10,669
Medium6
$9,750 + $750 + $7,493
Low'
$7,500 + $750 + $4,316
Unit Costs - Discounted at 7 percent to the year of abatement
> 2000 ppm
High
$35,600 +$10,000 (Only if
exterior paint is present)
Medium
$26,912 +$5,000 (Only if
exterior paint is present)
Low
$18,224 + $3,000 (Only if
exterior paint is present)
High
$23,419
High
$23,100 +$10,000 (Only if
exterior paint is present)
Medium
$18,498 + $5,000 (Only if
exterior paint is present)
Low
$13,896 +$3,000 (Only if
exterior paint is present)
Medium
$17,993
Low
$12,566
• $12,000 = High-cost/High-end interior lead paint abatement; $750 = Average-cost/High-end dust abatement; $9,600 = High
-------
EXHIBIT 4-6
Summary of Combined Abatement Strategies
Net Present Value Costs for Combined Scenarios
Low-end Paint and High-end Soil
Low-end Paint and Low-end Soil
>2000ppm
High*
$3.000 + $750 + $9.600 + $750 +
$11.460 + $1,040 + $10.000 (Only if
exterior paint is present)
Medium11
$2.000 + $750 + $7.248 + $750 +
$7.714 + $700 +$5,000 (Only if
exterior paint is present)
Low'
$1.000 + $750 + $4.896 + $750 +
$3,968 + $360 + $3,000 (Only if
exterior paint is present)
£2000 ppm
High
$3.000 + $750 + $9.600 +
$750+ $10,000 (Only if
exterior paint is present)
Medium
$2,000 + $750 + $7.248 +
$750+ $5,000 (Only if
exterior paint is present)
Low
$1,000 + 750 + $4.896 +
$750 + $3,000 (Only if
exterior paint is present)
High!
$3.000 + $750 + $10.669
Medium1
$2.000 + $750 + $7,493
Low1
$1,000 + $750 + $4.316
>2000 ppm
High
$26,600 + $10.000 (Only if
exterior paint is present)
Medium
$19. 162 +$5.000 (Only if
exterior paint ii present)
Low
$11, 724 + $3,000 (Only if
exterior paint is present)
£2000 ppm
High
$14, 100 +$10.000 (Only if
exterior paint is present)
Medium
$10.748 + $5,000 (Only if
exterior paint is present)
Low
$7,396 + $3,000 (Only if exterior
paint is present)
$14,419
$10,243
$6,066
8 $3,000 = High-cost/Low-end interior lead paint abatement; $750 = Average-cost/High-end dust abatement; $9,600 = High-cost/High-end soil lead abatement; $750 = Average-cost/High-end dust abatement;
$11,460 = High-cost hazardous soil disposal; $1.040 = High-cost soil transport; $10,000 = High-cost exterior lead paint abatement.
$2,000 = Medium-cost/Low-end interior lead paint abatement; $750 = Average-cost/High-end dust abatement; $7,248 = Medium-cost/High-end soil lead abatement; $750 = Average-cost/High-end dust abatement;
$7,714 = Medium-cost hazardous soil disposal; $700 = Medium-cost soil transport; $5,000 = High-cost exterior lead paint abatement.
1 $1,000 = Low-cost/Low-end interior lead paint abatement; $750 = Average-cost/High-end dust abatement; $4,896 = Low-cost/High-end soil lead abatement; $750 = Average-cost/High-end dust abatement;
$3,968 = Low-cost hazardous soil disposal; $360 = Low-cost soil transport; $3,000 = High-cost exterior lead paint abatement.
1 $3,000 = High-cost/Low-end interior lead paint abatement; $750 = Average-cost/High-end dust abatement; $10,669 = High-cost/Low-end soil with resod every 5 yean and High-end dust abatement at first resodding
(see Exhibit 4-5).
k $2,000 = Medium-cost/Low-end interior lead paint abatement; $750 = Average-cost/High-end dust abatement; $7,493 = Medium-cost/Low-end soil with resod every 5 years and High-end dust abatement at first resoddina
(see Exhibit 4-5).
$1,000 = Low-cost/Low-end interior lead paint abatement; $750 = Average-cost/High-end dust abatement; $4,316 = Low-cost/Low-end soil with resod every 5 years and High-end dust abatement at first resodding
(see Exhibit 4-5).
-------
4.3 SAMPLE COST CALCULATION
A sample calculation is presented below for the total cost of each abatement scenario
being analyzed. These equations build on those developed in Chapter 3 Section 1.4.
The optimum net benefit model developed for Section 403 calculates total benefits and costs for
a given set of assumptions regarding anticipated lead abatement choices for a 50 year time
period.6 Arriving at a present value estimate of costs or benefits for a 50 year time frame
requires the application of a multiplier to the base year's cost and benefits.
**(*fybl99JF Equation 4. 1 1
where: C = present value of total cost for 50 year period
u = unit cost of selected abatement for sample point y
n = number of sample points
Ny = number of homes (sample weight) for sample point y
bim = birth rate for base year (0.03994)
F = multiplying factor used to inflate cost to 50 years.
If, in Equation 4-1 1, F were equal to 1 then C would equal the total cost for the base year. F
is a function of birth rate, housing removal and growth rate, and discount rate. Because homes
that contain lead paint are removed and replaced with homes that do not, the lead paint homes
and non-lead paint homes have different growth rates and thus different inflation factors.
The multipliers used here are just the multipliers for the first year, first births derived
in Chapter 3 discounted to 1994 using a seven percent rate. For lead-paint-based homes this
value is 8.79; the derivation is shown in Exhibit 4-7. For lead-paint-free homes the multiplier
is larger (12.48) because the population grows each year. (See Exhibit 4-8 for the calculation.)
Note that Equation 4.1 1 is also used to calculate the costs of lead testing. In this case,
the unit cost u is the cost of testing and depends on the number of media tested. For example',
if interior paint and dust are both tested, then the unit testing cost is $460. If all three media
(interior paint, dust and soil) are tested, the unit cost is $598. Exterior paint testing is done only
when high-end soil abatement is chosen and adds another $115 to the total unit testing costs.
Fifty years was chosen as the modelling period because the net present value of the benefits accruing to
children born in the fiftieth year are less than one percent of the dollar value in year fifty using a seven percent
discount rate. Note benefits do not begin to accrue until age 18.
Abt Associates. Inc. 4-20 Draft. January 10. 1994
-------
EXHIBIT 4-7
Calculation of Lead Paint House Cost Mnltinfier
Year©
1995(1)
1999(5)
2024(30)
2025(31)
2027(33)
2028(34)
2030(36)
2032(38)
2033(39)
2034(40)
2035(41)
2038(44)
2040(46)
2042(48)
Amnd Cffmpomats of Abatement
Dedsraus Made -
0.9385
0.8835
0.8298
0.7797
0.7349
0.6910
0.6629
0.6341
0.6066
0.5804
0.5554
0.5330
0.5101
0.4882
0.4673
0.4486
0.4294
0.4111
0.3936
0.3768
0.3456
0.3310
0.3171
0.3038
0.2910
0.2788
0.2664
0.2553
0.2440
0.2340
0.2236
0.2145
0.1967
0.1887
0.1816
0.1748
0.1619
0.1558
0.1388
0.1336
0.1286
0.1237
0.1191
(Interest Rate)1
1.0700
1.1449
1,2250
1.3108
1.4026
1.6058
1.7182
1.8385
1.9672
2.1049
2.2522
2.4098
2.5785
2.7590
2.9522
3.1588
3.3799
3.6165
3.8697
4.1406
4.4304
4.7405
5.0724
5.4274
5.8074
6.2139
6.6488
7.1143
7.6123
8.1451
8.7153
9.3253
9.9781
10.6766
11.4239
12.2236
13.0793
13.9948
14.9745
16.0227
17.1443
18.3444
19.6285
21.0025
22.4726
25.7289
27.5299
Abatement Cost Lead Homes Multiplier
>M !
-------
EXHIBIT 4J8
Calculation of Non-Lead
Hones Multiplier
1 ' .
1
7
9
10
11
13
14
IS
17
18
19
20
22
23
25
26
27
29
30
31
33
34
35
38
39
40
42
43.
45
46
47
49
Year
1994
1995
1998
2000
2001
2002
2003
2004
2005
2006
2007
2008
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2025
2026
2027
2028
2029
2030
2031
2032
2033
2034
2035
2036
2037
2038
2039
2040
2041
2042
2043
Births
1.885
1.832
1.782
1,693
1.653
1.616
1499
1,584
1.S69
1,555
1,542
1.530
1,519
1,508
1.498
1.488
1.483
1.478
1,473
1,469
1.465
1.462
1.459
1,456
1.454
1.452
1.440
1,429
1,418
1.408
1.382
1.374
1,367
1.361
1.357
1.354
1.351
1.348
1.346
1.343
1,341
1,340
1.338
1.337
1.331
1.327
1,322
(Interest Rate)1
1.00
1.07
1.14
1.23
131
1.40
1.50
1.61
1.72
1.84
1.97
2.10
2.25
2.41
2.58
2.76
2.95
3.16
3.38
3.62
3.87
4.14
4.43
4.74
5.07
5.43
5.81
6.21
6.65
7.11
7.61
8.15
8.72
9.33
9.98
10.68
11.42
12.22
13.08
13.99
14.97
16.02
17.14
18.34
19.63
21.00
22.47
24.05
25.73
27.53
Total Discounted Births
Multiplier (Tout Discounted Births/First Year Births)
Discounted Births II
1885.40
1712.04
1556.64
1417.12
1291.64
1178.61
1076.64
995.93
921.73
853.47
790.63
732.75
679.40
630.20
584.81
542.90
504.19
469.45
437.24
407.36
379.63
353.89
329.98
307.76
287.10
267.90
250.03
231.74
214.89
199.34
184.99
171.75
159.51
148.20
137.74
128.05
119.09
111.02
103.51
96.53
90.03
83.99
78.36
73.12
68.25
63.71
59.47
55.37
51.57
48.03
23522.69
12.48
Abt Associates, Inc.
4-22
Draft, January 10, 1994
-------
4.4 RESULTS
A typical regulatory impact analysis will evaluate the costs of various regulatory
alternatives. The objective of this Section 403 analysis is to choose alternative hazard levels for
lead in paint, soil and dust. As a method of investigating the impacts of various levels of lead
on abatement, the net benefits are used as a tool to identify possible regulatory hazard levels.
The consequence of using net benefits as the identifying criterion is that the costs presented in
this section are intimately tied to the benefits analysis discussed in Chapter 5 and even more so
to the benefit-cost results examined in Chapter 6. A brief discussion of the approaches used in
the model is contained in Chapter 3, and a thorough discussion in Chapter 6.
4.4.1 Total Abatement Costs
To serve as an appropriate guide for public policymaking, the cost estimates used in this
analysis should be consistent with the concept of social costs. Because the data sources generally
provided estimates of private costs, it is reasonable to determine whether there are any
divergences between these estimated costs and social costs. This evaluation must be conducted
for each of the economic resources used in response to the promulgation of hazard levels under
Section 403. For the current discussion, these resources are classified generally as labor and
capital and discussed in turn below.7
The most common source of divergence between private and social costs of labor is
associated with unemployment. When unemployment exists, the costs of employing additional
labor may be lower than the prevailing wage implies. Consequently, if the regulation being
considered here causes unemployed labor to be used, the price of labor is not accurately
represented by the prevailing wage. Gramlich proposes conditions to be met if unemployed
labor is to be valued at less than market rates in benefit-cost analyses (Gramlich, 1981). Given
the fifty-year timeframe of this analysis, it is unclear whether two of these conditions can be
met, except intermittently, in the case of the hazard levels to be set under Section 403. The first
is that the reduction in unemployment must be sustained, meaning that reduced unemployment
today does not lead to inflation that generates more unemployment later. The second condition
is that the abatements and other actions induced by the promulgation hazard levels do indeed
lead to unemployment reductions that existing policies would not have addressed. Abatements
induced by this regulation will create additional demand for labor but it is unclear how much
unemployed labor will be provided to meet this additional demand. In part, making this
determination is difficult since the size and composition of the pool of unemployed labor will
fluctuate as the economy changes over the next 50 years.
It is hard to summarize here the possible divergences between the social and private costs of a third class
of resources - the environment - since that in many ways is the focus of this analysis, as the discussion of market
failures in Chapter 3 indicated. Instead, any divergences between social and private costs of using the environment
are discussed as they arise throughout this report. One example of such a divergence comes in the form of the
positive externality to future residents from the abatement of a given home, which is modelled in the benefit
estimation.
Abr Associates. Inc. 4-23 Draft. January 10. 1994
-------
Without an indication that the social costs of labor will be less than the private costs
during the time period of this analysis, it was assumed that private labor costs adequately reflect
social costs. This position is bolstered by the likelihood that the estimates of private costs used
in this analysis may be too low during periods of high demand for the type of labor used in
abatement. In short, there are downward and upward tendencies to the labor estimates used here
from the social costs of labor they are meant to represent. These tendencies imply that any point
estimate of labor costs may have a great deal of variance associated with it. However, the
direction of any net biases are unknown.
With regard to capital, one of the most common forms of divergence between social and
private costs hinges on whether investment is displaced by the lead abatements and other actions
induced by this rule. The opportunity cost of displaced investment is the present value of the
consumption stream that could be generated by that investment. This concept is also known as
the shadow price of capital. At one extreme, if investment is completely displaced, capital costs
should be multiplied by the shadow price of capital, which is estimated to be approximately 2.5,
in order to express the capital costs in terms of consumption, and then discounted back to the
present using the social rate of time preference (Scheraga, 1989). At the other extreme, if no
investment is to be displaced, abatement costs must be funded entirely from current consumption
(implying a shadow price of capital equal to one). An assumption of no displacement of
investment was applied in the main analysis. An alternative approach which allows for
displacing investment over varying timeframes is considered in sensitivity analyses in Chapter
7. These analyses are based upon a two-stage discounting procedure.
The total cost of any of the decision rules is the sum of the testing and abatement costs.
These, in turn, are a function of the number of homes undergoing testing and abatement and the
type of testing or abatement being done. Exhibit 4-9 shows the costs by abatement type for each
of the five decision rules used in this analysis. (See Chapter 3 Section 3.4 for a discussion of
the decision rules.) Exhibit 4-10 shows the number of homes abated by type of abatement for
each rule. The results discussed here were calculated using the medium value of the abatement
costs discounted at seven percent as shown in Exhibits 4-5 and 4-6. Results for high and low
abatement costs as well as costs calculated using the two-stage discounting procedure will be
discussed as part of the sensitivity analyses in Chapter 7.
The total testing costs for each decision rule are shown in Exhibit 4-9. For the four
decision rules that did not have an explicit dust condition or did not permit non-recurrent dust
abatement as an option, only interior paint (in pre-1980 homes) and soil were tested. The
resulting total testing costs were $14,982 million. For the remainder of the decision rules, all
three media were tested. In the voluntary optimum case, no exterior paint testing was required
because the high-end soil abatement was never chosen. The total testing costs were $24,222
million. In the remaining four decision rules, exterior paint testing costs were added for a total
of $24,346 million. The testing costs and abatement costs are summed to yield the total costs
discussed below.
Abl Associates. Inc. 4-24 Draft, January 10. 1994
-------
EXHIBIT 4-9
Total Costs for Five Alternative Decision Rules
'•
2.
3.
4.
5.
Decision Rule.8
Voluntary
Optimum''
Paint Condition
Only6
Single
Medium
Plus
Condition*1
2-Media
Plui
Condition6
3a.
3b.
3c.
4a.
4b.
4c.
3-Media Pliu
Conditionf
Soil
(ppm)
-
-
2.300
•
-
2.300
2.300
-
2.300
Dint
(ppm)
-
-
-
1.200
-
1.200
-
1.200
1.200
Paint
(XRF.
mi/em')
-
•
-
-
20
-
20
20
20
Noninlact Paint
Abtttcmcnt
Recommended
No
Ye.
Ye.
Ye.
Ye.
Ye.
Ye.
Ye.
Ye.
Abatement
Coiti
(S million)
13.896
24.66S
28.344
29.646
25.013
31.903
28.689
29.991
32,248
Testing Costs
(S million)
24.222
14.982
14.982
24.346
14.982
24.346
14.982
24.346
24.346
Total Costs
($ million)
38.118
39.630
43.326
53.992
39.995
56.249
43.671
54.337
56.594
Costs by Type of Abatement Chosen1 ($ million)
HP
20.134
20.134
20.260
20.429
20.260
20.429
20.555
20.555
LP
26
3.517
3.442
3.462
3.567
3.387
3.493
3.551
3.438
HS
272
272
272
272
IS
1.992
3.475
1.420
3.475
3.475
1.420
3.475
RD
978
978
978
978
HP/HS
HP/LS
943
943
943
943
943
943
943
943
LP/HS
87
87
87
87
LP/LS
74
350
74
74
350
350
74
350
NRD
11.878
2.150
2.150
2.150
2.150
choose the paint abatement method that generates (he
•Candidate hazard levels examined ranged up to 3000 ppm for soil. 2000 ppm for dust and 20 mg/cm'
•Each home selects abatement (or no abatement) that has highest net benefit.
'Abatement a recommended for homes with more than five square feet of lead -based paint in nonintact condition, regardless of XRF level or net benefits. Home
highest net benefits. Results are reported only for homes that exceed recommended levels.
dWithin the Aill range of individual soil, dust and paint hazard levels that could be set as a threshold for action, with no constraints placed onl eh other two media, the levels specified in the table maximize the net benefits Results
are reported only for homes that exceed recommended levels.
•Within the full range of individual soil, dust and paint hazard level combination that could be set as a threshold for action, with no restriction on the other medium, the levels specified in the table maximize the net benefits
Results are reported only for homes that exceed recommended levels.
'Within the fall range of individual soil, dust and paint hazard level combination that could be set as a threshold for action, with no restriction on the other medium, the levels specified in the table maximize the net benefits
Results are reported only for homes that exceed recommended leveb.
tAbalement Codes: High PabiKHP): Low PainKLP): High Soil(HS); Low Soil(LS); Recurrent Dust (RD): High Paint and High Soil(HP/HS); High Paint and Low Soil(HP/LS); Low Paint and High Soil (LP/HS)- Low Paint
and Low Soil (LP/LS): Nonrecurrent Dust (NRD). The abatement activities were described in Exhibits 4.1-4.6.
Abt Associates, Inc.
4-25
Draft, January 10, 1994
-------
EXHIBIT 4-10
Distribution of Abatement Chokw for Five Alternative Decision Rube
1.
2.
3.
4.
5.
Deciiion Rules*
Voluntary
Optimum''
Punt Condition
Onlyc
Single
Medium
Phis
Condition"1
2-Media
Plus
Condition*
3*.
3b.
3c.
4a.
4b.
4c.
3-Media Plus
Condition'
Soil
(ppm)
-
-
2.300
-
-
2.300
2.300
-
2.300
Dust
(ppm)
-
-
-
1.200
-
1.200
-
1.200
1.200
Paint
(XRF.
mg/cirf)
-
-
-
-
20
-
20
20
20
Noninlacl
Paint
Abatement
Recommended
No
Yea
Yes
Yea
Yes
Yes
Yes
Yes
Yes
Total
Number of
Abatements
(Thousands)
45.165
7,064
S.070
15.603
7,164
16.197
8.169
15.702
16.297
HP
4.160
4.160
4.186
4.220
4.186
4.220
4.246
4.246
LP
21
2.774
2.716
2.731
2.814
2.672
2.755
2.770
2.712
HS
74
74
74
74
Number of Abatements by Type Chosen1
(Thousand.)
IS
577
1.006
411
1.006
1.006
411
1.006
RD
276
276
276
276
HP/HS
HP/LS
114
114
114
114
114
114
114
114
LP/HS
18
18
18
18
LP/LS
16
74
16
16
74
74
16
74
NRD
44.567
7.777
7.777
7.777
7.777
Dd thai generate* me
*Candidate hazard levels examined ranged up to 3000 ppm for soil. 2000 ppm for dust and 20 mg/cm*
bEach home selects abatement (or no abatement) that has highest net benefits
'Abatement is recommended for homes with more than five square feet of lead -based paint in nonintact condition, regardless of XRF level or net benefits. Home owners choose the punt abatement
highest net benefits. Results are reported only for homes that exceed recommended levels.
'Within the full range of individual soil, dust and paint hazard levels that could be set as a threshold for action, with no constraints placed onl eh other two media, the levels specified in the table maximize the net benefits. Results
are reported only for homes that exceed recommended levels.
eWhhin the full range of individual soil, dust and paint hazard level combination that could be set as a threshold for action, with no restriction on the other medium, the levels specified in the table maximize the net benefits.
Results are reported only for homes that exceed recommended levels.
'Within the full range of individual soil, dust and paint hazard level combination that could be set as a threshold for action, with no restriction on the other medium, the levels specified hi the table maximize the net benefits.
Results ire reported only for homes that exceed recommended levels.
(Abatement Codes: High Paint(HP); Low PanutLP); High Soil(HS); Low SoiKLS); Recurrent Dust (RD); High Paint and High Soil(HP/HS); High Paint and Low Soil(HP/LS); Low Paint and High Soil (LP/HS); Low Paint
and Low Soil (LP/LS); Nonrecurrent Dust (NRD). The abatement activities were described in Exhibits 4.1-4.6. discussed below.
Abt Associates, Inc.
4-26
Draft, January 10, 1994
-------
The voluntary optimum decision rule has the lowest total costs of any of the rules
considered ($38,118 million) and the largest number of abatements recommended, just over 45
million. The overwhelming majority of the abatements were non-recurrent dust, the lowest unit
cost abatement. Once constraining conditions are added to the decision rules, the total costs rise,
not because the number of abatements increases (in fact they decrease), but because the types
of abatements necessary to meet the decision rule requirements have higher unit costs.
The remaining eight decision rules all recommend nonintact paint abatement as either part
or all of the abatement. One of these decision rules recommends only nonintact paint be abated;
in this case, 7 million abatements are performed at a total testing and abatement cost of $40
billion. While fewer abatements are performed under this rule than under the voluntary
optimum, they are paint abatements (alone or in combination with soil abatements), that have
higher unit costs than non-recurrent dust abatement, the voluntary optimum preferred choice.
The lower testing costs under this decision rule result from testing only paint and soil. Because
dust abatement will not satisfy the constraining condition (that paint in bad condition be abated),
there is no need for dust testing.
The next seven constrained decision rules each have higher costs and a larger number of
abatements. The non-intact paint abatement condition, with a cost of $24.6 billion, represents
the majority of the abatement cost in all the remaining rules. The single media constrained cases
add to this base. A soil condition at 2,300 ppm adds about $3.7 billion to the paint condition
base cost, the 1,200 ppm dust condition adds both soil and dust abatements totaling $5 billion.
An intact paint XRF condition of 20 mg/cm2 adds less than half a billion dollars to the nonintact
paint abatement cost. The three two-media constrained decision rules also add costs to the paint
condition base. With a decision rule of abating soil to 2,300 ppm and dust to 1,200, the
additional cost is about $7 billion. Adding soil at 2,300 ppm and paint at 20 mg/cm2 adds only
$4 billion while adding dust (1,200 ppm) and paint (20 mg/cm2) conditions adds $5.3 billion.
Finally, the three media constrained case adds $7.6 billion to the paint condition base making
it the most costly option considered. Chapter 6 will compare the costs incurred to the monetized
benefits discussed in Chapter 5.
4.4.2 Quantity of Hazardous Waste Generated by Abatement
The amount of hazardous waste that will be generated under each decision rule, based
on the Resource Conservation and Recovery Act (RCRA) disposability requirements, is an
additional consideration when evaluating the hazard level choices. The applicability of the
RCRA standards and the potential quantity of waste generated is described below for each
medium. As was discussed in Section 4.2.2, the Agency has yet ruled on whether residential
lead abatement wastes will be considered hazardous waste. It is also possible that the hazardous
wastes generated will be treated, and disposed of as non-hazardous waste, in which case the
hazardous waste capacity would not be an issue.
Abt Associates, Inc. 4-27 Draft. January 10, 1994
-------
Paint
In the Housing and Uiban Development abatement demonstration project, an average of
217 pounds of hazardous waste was generated per housing unit in the three cities for which data
was available (US EPA, 1992a). This estimate does not include all the laige solid debris which
was not considered hazardous. The method of abatement used may affect the amount of
hazardous waste generated, however, chemical stripping and abrasive removal, methods more
likely to create hazardous waste, were not considered in our model, nor are they likely to be
widely used. Because the disposal cost is a small percentage of the overall paint abatement cost
no explicit consideration of hazardous waste disposal was included in the modelled abatement
cost. A second consideration however, is the total quantity of waste that would be generated
under each of the proposed decision rules. Exhibit 4-11 shows the total and annualized volume
of hazardous waste from paint abatement. These values are compared to the total quantity of
hazardous waste, 197,501,112 tons, generated in the United States in 1989 (US EPA, 1992b).
As Exhibit 4-11 shows, the hazardous waste generated by lead-based paint abatement is less than
a tenth of a percent of the total national annual hazardous waste generation.
Dust
Dust abatement is expected to generate a very small quantity (< 20 pounds/unit) of waste
that could be considered hazardous. This was assumed to be excluded from RCRA regulation
under the household waste exemption criterion (Fortuna, 1987).
SoU
The hazardous waste from soil was explicitly considered in the cost of soil abatement
when the soil was abated by removal. The medium cost estimate presented in this report
assumed removal of 35.9 tons of soil per unit. Using this value and the number of high-end soil
abatements as reported by decision rule in Exhibit 4-10, the total quantity of hazardous soil
generated was calculated for each decision rule. The results are shown in Exhibit 4-11. The
total annual contribution of soil abatement is less than one tenth of one percent of the total
hazardous waste generated annually. Based on these results the quantity of hazardous waste
generated is not a significant concern although the cost contribution of disposing of the waste
could be significant.
Abt Associates, Inc. 4-28 Draft, January 10, 1994
-------
EXHIBIT 4-11
Volume of Hazardous Waste Generated by Media for Five Alternative Decision Rules
1.
2.
3.
4.
S.
Dccuion Rules*
Voluntary
Optimum
Paint Condition
Only'
Single
Medium
Plus
Condition
2-Medh
Plus
Condition"
3a.
3b.
3c.
4a.
4b
4c.
3-Media Phu
Condition^
Soil
(ppm)
-
2.300
•
2.300
2.300
-
2.300
Dust
(ppm)
-
-
-
1.200
•
1.200
-
1.200
1.200
hint
(XRF.
mg/cm1)
-
•
•
-
20
20
20
20
Nonintict
Flint
Abatement
Recommend oo
No
Yes
Y«
Yes
Yes
Yea
Yea
Yea
Yea
Total Volume of
Hazaidoua Waste
Ocneraled Over Fifty
Yean
(Thousand Tom)
2
696
696
3.998
705
3.998
70S
4.008
4.008
Volume of Hazardous
Wute Generated Over
Fifty Yean by Media
and Decision Rule*
(Thousand Tons)
Paint
2
696
o9o
696
70S
696
70S
70S
70S
Soil
0
0
0
3.303
0
3.303
0
3.303
3.303
Volume of Hazardous
Waste Generated Annually
by Media and Decision
Rule
(Thousand Tons)
Paint
.04
14
14
14
14
14
14
14
14
Soil
0
0
0
66
0
66
0
66
66
Lead Abatement
Hazardous Waste aa
Percent of Total
Hazardous Waste
Generated Annually
•e.OOOl*
.007%
.007*
.04*
.007*
.04%
.007%
.04%
.04%
e owner* choose the paint abatement method
^Candidate hazard leveb examined ranged up to 3000 ppm Tor soil. 2000 ppm for dust and 20 mg/cm'
"•Each home select! abatement (or no abatement) that has highest net benefits
'Abatement b recommended for homes with more than five square feet of lead -based paint in nonintact condition, regardless of XRF level or net benefits. He
that generates the highest net benefits. Results are reported only for homes that exceed recommended leveb.
'Within the hill range of individual soil, dust and paint hazard leveb that could be set as a threshold for action, with no constraints placed ont eh other two media, die leveb specified bi die table maximize
the net benefits. Results an reported only for homes that exceed recommended leveb
'Within the full range of individual soil, dust and paint hazard level combination that could be set as a threshold for action, with no restriction on the other medium, the levels specified in the table maximize
the net benefits. Results are reported only for homes that exceed recommended leveb.
'Within the hill range of individual soil, dust and paint hazard level combination that could be set as a threshold for action, with no restriction on the other medium, the leveb specified in die table maximize
the net benefits. Results are reported only for homes that exceed recommended leveb.
Abt Associates, Inc.
4-29
Draft, January 10, 1994
-------
4.5 REFERENCES
Carroll, M. 1993. Lead-paint Battle May Get Less Costly. The Boston Globe, January 30.
Clark, S., R. Bomschein, P. Succop, S. Roda, and B. Peace. 1991. Urban Lead Exposures
of Children in Cincinnati, Ohio. Chemical Speciation and Bioavailability. 3(3): 163-
171.
Degen, J. 1993. Personal Communication with Joshua Degen, Earthscape, Inc., Brighton,
MA, June, 1993.
Donegan, T. 1993. Personal Communication with Thomas Donegan, Chemical Waste
Management, Oak Brook, IL. July.
Elias, R. 1993. Personal Communication with Rob Elias, U.S. Environmental Protection
Agency. March 16.
Farrell, K., et al. 1992. Baltimore Soil Lead Abatement Demonstration Project. Final
Report. July 31.
Fortuna, R. et al. 1987. Hazardous Waste Regulation the New Era. McGraw-Hill, Inc.
Gramlich, E. M. 1981. Benefit-Cost Analysis of Government Programs. Prentice-Hall, Inc.,
Englewood Cliffs, NJ, p. 67.
Michigan Department of Public Health (Michigan). 1991. Recommended Abatement
Procedures and Cost Estimate Data. December, 1991.
R. S. Means Company, Inc. 1988. Means Residential Cost Data 1988. Construction
Consultants and Publishers, Kingston, MA.
Morris, V. 1993. Personal Communication with Vance Morris, Maryland Department of
Housing and Community Development, Annapolis, MD.
Rohmer, R. 1993. Personal Communication with Rosemary Rohmer, Maid in the USA,
Boston, MA.
Scheraga, J. 1989. Supplemental Guidelines on Discounting in the Preparation of
Regulatory Impact Analyses. Economic Studies Branch, Office of Policy,
Planning and Evaluation, U.S. Environmental Protection Agency, Washington,
DC, March.
Ullucci, P. 1993. Personal Communication with Paul A. Ulluci, Technical Director, ESA
Laboratories, Bedford, MA.
U.S. Department of Health and Human Services Public Health Services Centers for Disease
Abt Associates, Inc. 4-30 Draft, January 10, 1994
-------
Control (CDC). 1991. Strategic Plan for the Elimination of Childhood Lead
Poisoning. February.
U.S. Department of Housing and Urban Development (HUD). 1990. Comprehensive and
Workable Plan for the Abatement of Lead-Based Paint in Privately Owned Housing:
Report to Congress. Washington, DC. December.
U.S. Environmental Protection Agency, 1992.(US EPA, 1992a) "Applicability of RCRA
Disposal Requirements to Lead-based Paint Abatement Wastes", Office of Pollution
Prevention and Toxics, June.
U.S. Environmental Protection Agency, 1992. (US EPA, 19925) "1989 National Biennial
Report of Hazardous Waste Generators and Treatment, Storage, and Disposal
Facilities Requested Under RCRA. Office of Solid Waste.
U.S. Environmental Protection Agency (EPA). 1993. Urban Soil Lead Abatement
Demonstration Project - Integrated Report. April IS. Draft.
Weitzman, M., et al. 1992. Boston lead-in-soil/lead free kids demonstration project. Final
Report.
Abt Associates, Inc. 4-31 Draft, January 10, 1994
-------
5. BENEFITS
5.1 GENERAL ASSUMPTIONS
Chapter 3 provided a description of the methodology used to produce the "baseline"
hazard assessment of children's exposure to lead from paint, soil and dust in US private
housing stock. A basic premise of the baseline hazard assessment is that in the absence of
establishing Section 403 hazard levels to induce property owners to perform abatements,
children born each year over the next several decades will experience exposure to lead from
paint, soil and dust in patterns that are, for the most part, similar to current exposure patterns.
As a result, they will also incur the health damages associated with those patterns of exposure.
As discussed in Chapter 3, the baseline hazard assessment does account for some
improvement in the profile of childhood lead exposure over the 50 year modeling time frame
to reflect the expected attrition of older housing stock where lead paint can still be found, with
a concurrent increase in the number and proportion of newer homes without lead paint.
However, it was assumed in the baseline hazard assessment that no property owners would
undertake specific abatements to reduce or eliminate lead currently present in paint, soil or
dust.1
To estimate the benefits expected to result from abatements assumed to be induced by
the Section 403 hazard levels, a series of "what-if analyses were performed. Three sets of
assumptions were necessary to conduct these "what if' analyses. These were:
1) A set of assumptions defining decision rules incorporating the Section 403
hazard levels;
2) A set of assumptions regarding the nature of the responses by property owners
to those hazard levels in terms of the specific abatement actions to be
undertaken; and
3) A set of assumptions regarding the change in children's exposure conditions as a
result of undertaking those abatements.
These three sets of assumptions are discussed further in the following sections.
1 As noted in Chapter 3, this 'no abatements" assumption incorporated in the baseline analysis is
recognized as being extreme, since there are abatements of paint, soil and dust currently being performed in the
absence of these regulations.
Abt Associates, Inc. 5-1 Draft. January 10, 1994
-------
5.1.1 Decision Rule Assumptions
The first set of assumptions, regarding the selection of hazard levels to be evaluated,
has been discussed in Section 3.4, and is dealt with in detail in Chapter 6. The specific
decision rules addressed are summarized in Exhibit 5-1.
As discussed in Chapter 3.4 and Chapter 6, the specific paint, soil and dust values
identified as the hazard levels in these decision rules were, with the exception of the 500/400/-
option, arrived at through the consideration of the estimated net benefits produced That is,
among all of the three media combinations, two media combinations, and single media options
of paint, soil and dust hazard levels considered, the decision rules with the hazard levels
presented in Exhibit 5-1 were the ones that maYimi™H net benefits, given those other
assumptions.
The Voluntary Optimum decision rule is a special case in which no specific hazard
levels are set for paint, soil or dust; rather it is assumed that individual property owners will
choose to perform abatements that provide maximum positive net benefits, or will perform no
abatement if none of the options produces positive net benefits.
The hazard levels in the 500/400/- decision rule that is included in this analysis ttalfnot
based on net benefits considerations. Rather, it was based on an individual risk perspective,
using the probability of exceeding certain critical blood lead levels as the "trigger" for
inducing abatements. Suppose for example that a household facing an abatement decision
(i.e., expecting a child in the ensuing year) has lead paint in good condition at an XRF of 19,
soil lead at 2,200 ppm, and dust lead at 1,100 ppm. Since these values are just below those
that would induce abatement in any of these media according to the net benefits analysis, it
would be assumed under those decision rules that no abatement would be done. However, the
blood lead geometric mean predicted by the ffiUBK for a population of children exposed to
these levels is 15.3 Mg/dl. At this level, and assuming a GSD of 1.6, the probability of
exceeding several frequently targeted blood lead levels of concern are:
81.7% chance of exceeding 10 /*g/dl
51.7% chance of exceeding 15/ig/dl
28.4% chance of exceeding 20 jtg/dl
14.8% chance of exceeding 25 /*g/dl
Because one of the primary objectives of setting hazard levels is to reduce or eliminate
children's risk of adverse health effects due to lead exposure from these sources, and because
CDC now considers blood lead levels down to 10 pg/dl to be of concern from a health effects
standpoint, it was felt necessary to also include consideration of a decision rule in the benefits
analysis that was aimed specifically at minimizing the incidence of these high blood lead
levels, notwithstanding the net benefits of that rule.
Abt Associates. Inc. 5-2 Draft, January JO. 1994
-------
Exhibit 5-1. Summary of Decision Rules and Hazard Levels
Decision Rule
Description
Basis
Voluntary Optimum
No specific hazard levels are set; individual properly
owners undertake those abatements (if any) producing
me maximum positive net benefits.
This decision rule produces the
maximum net benefits of all rules
considered.
Paint Condition Only
Interior lead paint in bad condition always induces
abatement, regardless of lead level; however, no hazard
levels set for paint in good condition, dust lead, or soil
lead.
Since all other decision rules (except the
Voluntary Optimum) include a hazard
level for paint based on bad condition,
mis decision rule provides a means oi
discerning the portion of the benefits
from those other rules contributed by this
component.
2300/1200/20
Interior lead paint in bad condition always induces
abatement, regardless of lead level; and hazard levels
set at:
Soil = 2,300 ppm
Dust = 1,200 ppm
XRF = /OTfor paint in good condition
This decision rule was found to have the
highest net benefits among all
combinations considered that included
specific hazard levels for all three media,
with paint condition included.
2300/-/20
Interior leacK-p&nt in bad condition always induces
abatement, regardless of lead level; and hazard levels
set at:
Soil =2,300 ppm
Dust = no level set
XRF = 20 for paint in good condition
This decision rule was found to have the
highest net benefits among all
combinations considered that included
specific hazard levels for soil and good
condition paint, with paint condition
included.
-71200/20
Interior lead paint in bad condition always induces
abatement, regardless of lead level; and hazard levels
set at:
Soil = no level set
Dust = 1200 ppm
XRF = 20 for paint in good condition
This decision rule was found to have the
highest net benefits among all
combinations considered that included
specific hazard levels for dust and good
condition paint, with paint condition
included.
2300/1200/-
Interior lead paint in bad condition always induces
abatement, regardless of lead level; and hazard levels
set at:
Soil = 2,300 ppm
Dust = 1,200 ppm
XRF = no level set
This decision rule was found to have the
highest net benefits among all
combinations considered that included
specific hazard levels for soil and dust,
with paint condition included.
2300/-/-
Interior lead paint in bad condition always induces
abatement, regardless of lead level; and hazard levels
set at:
Soil = 2,300 ppm
Dust = no level set
XRF = no level set
This decision rule was found to have the
highest net benefits among all
combinations considered that included
specific hazard levels for soil only, with
paint condition included.
-M200/-
Intenor lead paint in bad condition always induces
abatement, regardless of lead level; and hazard levels
set at:
Soil = no level set
Dust = 1,200 ppm
XRF <= no level set
This decision rule was found to have the
highest net benefits among all
combinations considered that included
specific hazard levels for dust only, with
paint condition included.
-/-/20
Interior lead paint in bad condition always induces
abatement, regardless of lead level; and hazard levels
set at:
Soil = no level set
Dust = no level set
XRF = 20 for paint in good condition
This decision rule was found to have the
highest net benefits among all
combinations considered that included
specific hazard levels for good condition
paint, with paint condition included.
SOO/400/-
Interior lead paint in bad condition always induces
abatement, regardless of lead level; and hazard levels
set at:
Soil = 500 ppm
Dust = 400 ppm
XRF = no level set
This decision rule was designed to
minimize individual risk of exceeding
specific target blood lead levels (see
discussion in text).
Abt Associates, Inc.
5-3
Draft, January 10, 1994
-------
Shown in Exhibit 5-2, below, are back-calculations of the value for the geometric
mean that will keep 90%, 95%, or 99% of the population below the specific blood lead target
of 10, 15, 20 and 25 jtg/dl (assuming in all cases a GSD of 1.6). So, for example, for an
individual child to have _>_ 90% chance that his or her blood lead will not exceed 10 /tg/dl (or,
conversely, < 10% chance that it will exceed 10 /tg/dl), that child's exposure conditions
should lead to an estimated geometric mean blood lead value of 5.48 /tg/dl or less. Similarly,
to have a 95% chance of having a blood lead below 15 /£g/dl, the predicted geometric mean
for those exposure conditions should be less than 6.92 /ig/dl.
Exhibit 5-2. Geometric Means (in /tg/dl) to Achieve
Indicated Blood Lead Targets
PbB target: 10 /tg/dl
15 /tg/dl
20 /xg/dl
25 MR/dl
GM to keep
90% of
population
below
indicated PbB
target
GM to keep
95% of
population
below
indicated PbB
target
GM to keep
99% of
population
below
indicated PbB
target
(Assumes a GSD of 1 .6)
5.48
8.21
10.95
13.69
4.62
6.92
9.23
11.54
3.35
5.03
6.70
8.38
Given the most recent guidance provided by CDC for blood lead levels of concern,
goals were assumed for a set of lead hazard levels that will limit individual risk of exceeding
key target blood lead levels to:
Approximately 90% chance of blood lead less than 10 /tg/dl, and
Approximately 95% chance of blood lead less than 15 /tg/dl, and
Approximately 99% chance of blood lead less that 20 /tg/dl.
Considering the geometric means in Exhibit 5-2, these goals would be met by the blood
lead geometric means in the bolded cells. For the purpose of this analysis, a value in the
middle of this range of 6.5 /tg/dl has been selected as the target geometric mean blood lead to
approximate the above stated risk targets.
Based on the IEUBK model runs, a series of soil and dust combinations were identified
that produce a geometric mean of approximately 6.5 /tg/dl. These combinations included
some having very high dust levels in combination with low soil levels (e.g., dust = 800 ppm;
soil = 11 ppm), and some having very high soil levels in combination with low dust levels
(soil = 958 ppm; dust = 25 ppm). The specific hazard levels selected for this decision rule
Abt Associates, Inc.
5-4
Draft, January 10, 1994
-------
were soil = 500 ppm and dust = 400 ppm, which avoided the extremes of combinations
producing the same risk levels. It should be noted that this decision rule also includes the
assumption that lead paint in bad condition will always induce abatement. No hazard level
was included for lead paint in good condition, since it was found that adding such a hazard
level would not substantially affect the predicted blood lead levels (see Section 5.3).
5.1.2 Abatement Choice Assumptions
The second set of assumptions are also discussed in part in Chapter 3.4 as well as in
Chapter 4. Briefly, it has been assumed that property owners facing an abatement decision
will choose an abatement type that will reduce their paint, soil and/or dust levels to be in
conformance with the Section 403 hazard levels, and will choose from among alternatives able
to accomplish this goal, the abatement alternative that maximizes the net benefits.
As described in Chapter 4, there are 10 specific abatement choices considered in this
analysis. These include two paint abatements (high-end and low-end paint abatements); two
soil abatements (high-end and low-end soil abatements); four combined paint and soil
abatements (high-end paint with high-end soil; high-end paint with low-end soil; low-end paint
with high-end soil; and low-end paint with low-end soil); and two dust abatements (recurrent
and non-recurrent).
5.1.3 Post-Abatement Exposure Condition Assumptions
The third set of assumptions, those regarding the change in exposure conditions as a
result of undertaking a specific type of abatement, were discussed in part in Chapter 4 with
respect to the effectiveness of the various types of abatement considered.
The assumed post-abatement conditions for each alternative are summarized in Exhibit
5-3. Post-abatement conditions for the combined paint and soil abatements simply reflect the
combination of post abatement conditions for each separately. The "calculated" dust values
referred to in Exhibit 5-3 are explained below.
It is generally recognized that the lead present in dust in homes originates primarily
from lead in paint at that home and in the soil in proximity to that home. As discussed in
Chapter 4, we have assumed that all paint and soil abatements will also include dust
abatement. Therefore, whenever homes perform abatements of paint or soil, there is an
expected concomitant reduction in the level of lead in the dust in that home. It was therefore
necessary to incorporate an algorithm in the benefits analysis to estimate what those post-
abatement dust levels would be.
Abt Associates, Inc. 5-5 Draft, January 10, 1994
-------
Exhibit 5-3.
Summary of Post-Abatement Conditions for
Various Abatement Alternatives
Abatement Alternative
Assumed Post-Abatement Conditions
High End Paint Abatement
• Paint:
•Soil:
No interior lead paint remains.
No change in soil lead concentration.
Uses lower of HUD value or calculated value.
Low End Paint Abatement
• Paint: No lead paint on windows; intact lead paint
remains on other surfaces.
• §eU: No change in soil lead concentration.
• Dust: Uses lower of HUD value or calculated value.
High End Soil Abatement
Paint: No change in interior lead paint levels oi
condition.
Soil: Lead concentration reduced to 100 ppm.
Dust: Uses lower of HUD value or calculated value
Low End Soil Abatement
Paint: No change in interior lead paint levels of
condition.
§pjl : Lead concentration reduced to 500 ppm
Dust: Uses lower of HUD value or calculated value.
Recurrent Dust Abatement
Paint: No change in interior lead paint levels of
condition.
SoU: No change in soil lead concentration.
Dust: Lead concentration reduced to 100 ppm.
Non-recurrent dust
• Paint: No change in interior lead paint levels of
condition.
Soil: No change in soil lead concentration.
Dust: Reduced to calculated value.
Abt Associates, Inc.
5-6
Draft, January 10, 1994
-------
The algorithms used were based on relationships provided in the December 1991 draft
Guidance Manual for the ffiUBK model. The basic relationships provided there were:
PAD = [366 + (83.5. (PbP -1.0))]+ [0.9• PbS ] (Equation 5-1)
where PbD is the dust lead concentration in ppm; PbP is the maximum interior paint XRF
measurement; and PbS is the soil lead concentration in ppm.
In the Guidance Manual, this relationship is modified slightly for low XRF values. If
the XRF value is greater than 0 but less than or equal to 1, the equation becomes:
PbD = 252 + [0.9 • PbS ] (Equation 5-2)
If there is no lead paint (i.e., XRF = 0), the relationship is:
HO = [0.9 • PbS] (Equation 5-3)
Two exceptions exist to using these calculated post-abatement dust levels. The first is
in the case of recurrent dust abatement. Because it is assumed that the recurrent dust
abatement option is undertaken for the express purpose of minimizing dust lead levels without
performing any paint or soil abatement, it was necessary to make an assumption as to the
effectiveness of this option. It is assumed in this case that the recurrent dust abatement will
yield an effective dust level is 100 ppm.
The second exception to using these calculated dust values for post-abatement
conditions is when the original HUD dust value is lower than the post-abatement value
calculated from the above algorithms. In these cases, the lower HUD value was used to avoid
having an outcome where the post-abatement dust level exceeded the baseline dust level.
As noted in Chapter 3, the risk modeling (and therefore the benefits modeling as well)
cannot at present differentiate between the health damages to children whose soil lead exposure
is predominately bare soil from the damages to children whose exposure is mainly from
covered soils. As discussed in Chapter 3, no information is available to differentiate between
bare and covered soils in the baseline risk assessment. It is assumed, however, that the post-
abatement soils are all grass-covered. Because the intake of lead from exposure to bare soils
is expected to be greater than from exposure to covered soils (all other factors being equal,
such as lead concentration and soil type), the benefits of soil abatement may be underestimated
for those cases where the starting condition is bare soil, and overestimated where the starting
condition is covered soils. As noted in Chapter 3, it is assumed that these factors largely
compensate for one another in the aggregate estimate of the baseline damages and the benefits
of soil abatement. It is not known, however, how reasonable this assumption is, since neither
the distribution of the incidence of bare vs. covered soils in residential settings, nor the
relationships between soil condition and blood lead levels are known. In Chapter 7, this issue
Abt Associates, Inc. 5-7 Draft. January 10, 1994
-------
is discussed further in the context of its potential effects on benefit-cost comparisons and
identifying hazard levels that maximise net benefits.
5.2 BENEFITS MODELING PROCESS
The modeling process for estimating the benefits resulting from various combinations
of decision rules and abatement choices parallels that described in detail in Chapter 3 for the
baseline hazard assessment. Consider, for example, the first year of the SO year modeling
time frame. In the baseline hazard assessment, the first year's cohort of children are assumed
to experience exposure to lead in paint, soil and dust as estimated from the HUD data and
other assumptions described previously. In essence, the abatement decision being made in the
baseline assessment in anticipation of these new children arriving is "no abatement."
Alternative analyses were therefore performed to calculate the effect of undertaking each of the
10 viable abatement options at each of these homes. To do this, the baseline levels of lead in
paint, soil, and/or dust were replaced with the assumed post-abatement conditions, as
summarized in Exhibit 5-3.
Using these post-abatement paint, soil and dust lead levels, the modeling process is
then identical to that used in the baseline hazard assessment. That is, these reduced exposure
levels are used in the ffiUBK model, maintaining all other ffiUBK assumptions as before, to
arrive at a new, lower estimate of the geometric mean for subpopulations of children in each
housing category. These lower geometric means, again with the assumption of a geometric
standard deviation of 1.6, are used to define the blood lead distribution for these children.
Using the first year cohort of children, an estimate is made of the incidence of IQ points lost,
cases of IQ < 70, incidence of blood leads > 25 jig/dl, and neonatal mortality. As
discussed in Chapter 3, the estimated incidence of these effects for the remaining 49 years of
the modeling time frame are obtained through the use of multipliers that reflect anticipated
changes in birth rates and housing stock levels over the full modeling period.
As indicated before, the benefits for undertaking a particular set of abatements in
response to a given decision rule is calculated as the difference between the baseline estimate
of the incidence of these effects and the estimates obtained with the assumed abatements
having been performed.
It is important to note that in modeling the benefits and costs of hazard level /
abatement choice combinations, we assume always that when abatements are done, they are
only done in conjunction with the anticipated arrival of a new child in the ensuing year. That
is, abatements are not assumed to be done on homes just because they exceed the lead paint,
soil and/or dust hazard levels. The imminent arrival of a new child is assumed to be the
trigger for making abatement decisions, using the Section 403 hazard levels to guide the
specific abatement choice. It is also important to note that the modeling of benefits has
assumed that property owners facing an abatement decision will always choose to perform
some abatement if they exceed specified hazard levels (i.e., there are no non-compliers).
Abt Associates. Inc. 5-8 Draft, January 10,1994
-------
This section presents the estimated benefits of the various Section 403 decision rules in
terms of impacts on population blood lead levels and avoided incidence of the specific adverse
health effects addressed, namely IQ point loss, IQ < 70, blood lead > 25 pg/dl, and neonatal
mortality. For the purpose of comparing the different decision rules, the blood lead
distribution changes and avoided incidence estimates presented in this section reflect only the
first model year. Since the multipliers used to determine the avoided incidence of these effects
across the entire modeling period are the same for the baseline and all post-abatement
scenarios, the relative order of benefits is the same in the first year as it would be over the full
modeling time frame.
Since most of the benefits computed are directly associated with the changes that
abatements have on the population blood lead distribution, it is useful to first compare how
different decision rules affect the blood lead distributions. Exhibit 5-4 provides a summary
comparison of the blood lead distributions for the baseline with those resulting from the
various decision rules for the first model year cohort. One general observation about the blood
lead changes (which applies to most other measures of benefits as well) is that the inclusion in
a decision rule of the XRF = 20 hazard level for paint in good condition has little or no effect
on the estimated benefits relative to a similar decision rule without the XRF constraint.
The largest downward shifts in the blood lead distributions were observed in the two
"special case" decision rules, i.e., the Voluntary Optimum and the 500/4007- rule. The latter
of these had the greatest impact, with a predicted downward shift in the geometric mean from
a baseline value of 4.06 fig/dl to 2.45 pg/dl. The Voluntary Optimum was close behind this
with a downward shift in the geometric mean to a value of 2.52 pg/dl. However, the
500/400/- decision rule has a much larger impact in terms of reducing the size of the upper tail
of the distribution, showing for example only about 2.4% of the population expected to be
above 15 /*g/dl, versus 9.0% above this level for the Voluntary Optimum.
Among the several three-media, two-media and single-media decision rules based on
maximum net benefits, the largest effects are seen for the options that include both soil at
2,300 ppm and dust at 1,200 ppm. Again, adding an XRF = 20 hazard level for paint in good
condition has no significant effect on the predicted blood lead distribution. The decision rule
options with dust of 1,200 without a soil hazard level results hi a greater downward shift in the
blood lead distribution than the options with a soil hazard level of 2,300 without a dust hazard
level. The least effective decision rules are those that place a hazard level only on paint in
good condition, with no constraints on either soil or dust. It should also be noted that
while the downward shift in the geometric mean is greater for the Voluntary Optimum than for
these rules based on maximum net benefits, all of these latter rules have lower predicted
GSDs, and consequently result in slightly lower portions of the population in the upper tail
than predicted for the Voluntary Optimum. This is primarily a result of the underlying
assumption that, in the rules based net-benefits, all homes with lead paint in bad condition will
undergo abatement when a child is expected, an assumption that is not included in the
voluntary optimum.
Abt Associates. Inc. 5-9 Draft. January 10, 1994
-------
Exhibit 5-4. Summary of Post-Abatement Blood Lead Distribution Characteristics by Decision Rule
Decision Rule
Baseline
Voluntary
Optimum
Paint Condition
Only
Soil = 2300
Dust = 1200
XRF = 20
Soil = 2300
Dust = 1200
No XRF value
Soil = 2300
No Dust Value
XRF = 20
No soil value
Dust = 1200
XRF = 20
Soil = 2300
No Dust value
No XRF value
No soil value
Dust = 1200
No XRF value
No soil value
No dust value
XRF = 20
Soil = 500
Dust = 400
No XRF value
Mean
6.08
4.11
5.75
4.64
4.64
5.58
4.73
5.58
4.73
5.75
3.24
Geometric
Mean
4.06
2.52
3.82
3.30
3.30
3.78
3.32
3.78
3.32
3.82
2.45
Geometric
Standard
Deviation
2.45
2.70
2.47
2.35
2.35
2.43
2.37
2.43
2.39
2.47
2.20
Median
3.91
2.45
3.72
3.36
3.36
3.71
3.36
3.71
3.36
3.72
2.56
90th
Percentile
13.30
9.47
12.55
9.67
9.67
11.99
9.81
11.99
9.81
12.55
6.43
95th
Percentile
18.83
13.39
17.73
12.78
12.78
16.65
13.19
16.65
13.19
17.73
8.12
% > 10
Mg/dl
15.96%
9.05%
14.63%
9.24%
9.24%
13.87%
9.57%
13.88%
9.57%
14.62%
2.39%
% > 15
Mg/dl
8.01%
3.81%
7.12%
3.13%
3.13%
6.33%
3.54%
6.33%
3.54%
7.12%
0.39%
% > 20
Mg/dl
4.35%
1.72%
3.77%
1.16%
1.16%
3.16%
1.50%
3.16%
1.50%
3.77%
0.08%
% >25
Mg/dl
2.46%
0.82%
2.08%
0.46%
0.46%
1.67%
0.69%
1.67%
0.69%
2.08%
0.02%
Mote: All decision rules except the Voluntary Optimum also include paint in bad condition as a hazard level regardless of XRF value;
XRF values shown in the above decision rules refer to good condition paint.
-------
Exhibit S-S summarizes the effect of these rules in terms of avoided IQ point loss and
avoided incidence of IQ < 70 for the first year cohort. In terms of IQ point losses avoided,
the order of benefits obtained by the decision rules is identical to that observed for the
downward shifts in the blood lead distributions. The 500/4007- decision rule and the
Voluntary Optimum show avoided IQ point losses in the first year of about 2.8 million and 1.9
million, respectively. It is noteworthy that the 2300/1200/20 and 2300/1200/- decision rules
have slightly lower total avoided IQ point losses of about 1.4 million, but that the average IQ
point loss per affected individual is almost 2 points, versus about 1.6 points per individual for
the 500/400 and voluntary optimum rules.
The benefits in terms of avoided incidence of IQ < 70 deviates slightly from the
avoided IQ point losses in that the Voluntary Optimum ranks slightly below the 2300/1200/20
and 2300/1200/- decision rules. This is in large part due to the effect of the larger residual tail
in the Voluntary Optimum discussed previously coupled with the piece wise linear regressions
used to estimate the incidence of IQ < 70 which has a higher probability at higher blood lead
levels.
Similarly, in the avoided incidence of blood lead levels above 25 ng/dl (Exhibit 5-6),
the general order is similar, but in this case the Voluntary Optimum falls below both the
2300/1200/20 and 2300/1200/- rules, as well as the -/1200/20 and -/1200/- rules. Again, this
is the effect of the Voluntary Optimum including no high end paint abatements noted
previously.
Exhibit 5-7 provides a summary of the resulting impact of the various decision rules on
limiting the individual risk of children to elevated blood lead levels. As indicated there, it is
estimated that in the baseline (no abatement) case, about 960,000 children in the first year
cohort are born into homes where the predicted geometric mean blood lead levels are above
6.5 ftg/dl. Assuming a GSD of 1.6 for these homes, the individual risk of exceeding 10 /tg/dl
is 18.0% or greater, of exceeding 15 /zg/dl is about 4.76% or greater, and of exceeding 20
/*g/dl is about 0.85% or greater. As shown in Exhibit 5-7, the 500/400/- decision rule
eliminates all cases of homes where the expected blood lead GM is .>. 6.5 /tg/dl, which was
the specific intent of this particular decision rule. The other decision rules result in
approximately 600,000 to 880,000 of the baseline 960,000 children in the first model year
being bom into homes where paint, soil and dust levels imply GMs at or above 6.5 /ig/dl.
Exhibit 5-8 shows the avoided incidence of neonatal mortality for the various decision
rules. Most of the decision rules result in essentially identical benefits of avoiding 48 or 49
cases of neonatal mortality in the first year cohort. These neonatal mortality cases avoided
result almost entirely from the required high end paint abatement in homes having bad
condition lead paint2 . The 500/400/- decision rule has a somewhat higher benefit of 75 cases
avoided, owing to the need to perform additional high-end paint abatements beyond those for
The current model assumptions for neonatal mortality associate this adverse effect with the presence of
any interior lead paint in the home. Since only high-end paint abatement eliminates all interior lead paint, only
high-end paint abatement will provide the benefit of reduced incidence of neonatal mortality.
Abt Associates. Inc. 5-11 Draft, January 10. 1994
-------
homes having bad condition in order to meet the more stringent dust level of 400 ppm. The
Voluntary Optimum is notable in that no cases of neonatal mortality are avoided, since this
option results in no high-end paint abatements being performed. (Note that in the baseline
estimate there are about 330 neonatal deaths in the first year as a result of maternal exposure to
lead paint.)
Abt Associates, Inc. 5-12 Draft, January 10, 1994
-------
Exhibit 5-5. Summary of Avoided IQ Point Loss and
Avoided Incidence of IQ < 70 by Decision Rule
Decision Rule
Baseline
Voluntary Optimum
Paint Condition
Only
Soil = 2300
Dust = 1200
XRF = 20
Soil = 2300
Dust = 1200
No XRF value
Soil = 2300
No Dust Value
XRF = 20
No soil value
Dust = 1200
XRF = 20
Soil = 2300
No Dust value
No XRF value
No soil value
Dust = 1200
No XRF value
No soil value
No dust value
XRF = 20
Soil = 500
Dust = 400
No XRF value
IQ Point Loss
Avoided (First
Model Year
Cohort)
0
1,912,011
319,818
1,404,490
1,402,677
492,439
1,318,269
490,626
1,316,456
321,631
2,751,452
Affected
Population of
Children (First
Model Year
Cohort)1
0
1,482,732
370,074
714,261
708,989
428,048
683,096
422,776
677,824
375,346
1,736,524
Average IQ
Point Loss
Avoided (First
Model Year
Cohort)
0
1.34
0.86
1.97
1.98
1.15
1.93
1.16
1.94
0.86
1.58
Avoided Cases
of IQ < 70
0
5,434
1,081
5,308
5,305
1,912
4,877
1,909
4,874
1,084
8,823
1 Total population in first model year cohort = 3,877,530
Abt Associates, Inc.
5-13
Draft, January 10, 1994
-------
Exhibit 5-6. Summary of Avoided Incidence of Blood Lead > 25 pg/dl
by Decision Rule
Decision Rule
Baseline
Voluntary Optimum
Paint Condition Only
Soil = 2300
Dust = 1200
XRF = 20
Soil = 2300
Dust = 1200
No XRF value
Soil = 2300
No Dust Value
XRF = 20
No soil value
Dust = 1200
XRF = 20
Soil = 2300
No Dust value
No XRF value
No soil value
Dust = 1200
No XRF value
No soil value
No dust value
XRF = 20
Soil = 500
Dust = 400
No XRF value
Avoided Incidence of FbB > 25 /ig/dl
0
63,422
14,744
77,479
77,479
30,736
68,513
30,736
68,513
14,745
94,656
Abt Associates, Inc.
5-14
Draft, January 10,1994
-------
Exhibit 5-7. Summary of Children hi First Year Cohort Remaining hi
Homes with Paint, Soil and Dust Lead Levels Implying
Geometric Mean Blood Lead Levels _>_ 6.5 /tg/dl
Decision Rule
Baseline
Voluntary Optimum
Paint Condition Only
Soil = 2300
Dust = 1200
XRF = 20
Soil = 2300
Dust = 1200
No XRF value
Soil = 2300
No Dust Value
XRF = 20
No soil value
Dust = 1200
XRF = 20
Soil = 2300
No Dust value
No XRF value
No soil value
Dust = 1200
No XRF value
No soil value
No dust value
XRF = 20
Soil = 500
Dust = 400
No XRF value
First Year Cohort of Children Remaining in
Homes with Predicted GM _> 6.5 pg/dl
< 0.85% chance of PbB > 20
< 3.76 % chance of PbB > 15
< 18.0% chance of PbB > 10
(assuming GSD =1.6)
960,066
607,312
883,301
636,768
636,768
861,764
636,768
861,764
636,768
883,301
0
Abt Associates, Inc.
5-15
Draft, January 10, 1994
-------
Exhibit 5-8. Summary of Avoided Incidence of Neonatal Mortality
by Decision Rule
Decision Rule
Baseline
Voluntary Optimum
Paint Condition Only
Soil = 2300
Dust = 1200
XRF = 20
Soil = 2300
Dust = 1200
No XRF value
Soil = 2300
No Dust Value
XRF = 20
No soil value
Dust = 1200
XRF = 20
Soil = 2300
No Dust value
No XRF value
No soil value
Dust = 1200
No XRF value
No soil value
No dust value
XRF = 20
Soil = 500
Dust = 400
No XRF value
Avoided Incidence of Neonatal Mortality
(in First year Cohort of Children)
0
0
48
49
48
48
49
48
48
48
75
Abt Associates, Inc.
5-16
Draft, January 10, 1994
-------
5.4 VALUATION OF BENEFITS
In Section 5.3, above, the discussion of health damages associated with childhood lead
exposure, and the benefits of reducing that exposure, focused on the incidence of adverse
health effects in terms of blood lead distributions, IQ point losses, avoidance of IQ < 70, and
neonatal mortality. To provide a basis for comparing the magnitude of the benefits resulting
from paint, soil and dust abatements with the estimated cost of conducting those abatements, it
is necessary to place a monetary value on the benefits. This section describes how these
benefits have been monetized.
5.4.1 Valuing Lost IQ Points
Available economic research provides little empirical data for society's willingness to
pay (WTP) to avoid a decrease in a child's IQ. As an alternative measure, it was assumed that
IQ deficits incurred through lead exposure will persist throughout the exposed child's lifetime.
Two consequences of this IQ decrement, representing a portion of society's full willingness to
pay, are then considered: the decreased present value of expected lifetime earnings for the
child, and the increased educational resources expended for a- child who becomes mentally
handicapped or is in need of compensatory education as a consequence of lead exposure. The
value of foregone earnings is addressed in this section.
The reduction in IQ has a direct and indirect effect on earnings. The direct effect is
straightforward - lower IQs decrease job attainment and performance. Reduced IQ also
results in reduced educational attainment, which, in turn, affects earnings and labor force
participation. Note that these effects on earnings are additive since the studies used for this
analysis have controlled for the direct and indirect effects separately.
Direct Effect of IQ on Wage Rate
Henry Aaron, Zvi Griliches, and Paul Taubman have reviewed the literature
examining the relationship between IQ and lifetime earnings (USEPA 1984). They find that
the direct effect, (schooling held constant) of IQ on wage rates ranged from 0.2 percent to
0.75 percent. Perhaps the best of these studies is Griliches (1977).3 He found the direct effect
of IQ on wage rates to be slightly more than 0.5 percent per IQ point. Because this value is
roughly the median estimate of the USEPA review of the literature, it is the value used in this
analysis.
Indirect Effect of IQ on Earnings
From Needleman et al. (1990) it is possible to estimate the change in years of schooling
attained per one IQ point change. Their regression coefficients for the effect of tooth lead on
Griliches used a structural equations model to estimate the impact of multiple variables on an outcome of
interest. This method has conceptual advantages over other empirical estimates used in the literature because it
successfully controls for the many confounding variables that can affect earnings.
Abt Associates, Inc. 5-17 Draft, January 10, 1994
-------
achieved grade provide an estimate of current grade achieved. However, many of these
children were in college at the time and are expected to achieve a higher grade level.
Following Schwartz (1990a), after adjusting the published results for the fact that a higher
percentage of children with low tooth lead were attending college, a 0.59 year difference in
expected maximum grade achieved between the high and low exposure groups was estimated.
It is assumed that educational attainment relates with blood lead levels in proportion to IQ.
The difference in IQ score between the high and low exposure group was 4.5 points. By
dividing 0.59/4.5 = 0.131, it suggests that the increase in lead exposure which reduces IQ by
one point may also reduce years of schooling by 0.131 years.4
Studies that estimate the relationship between educational attainment and wage rates
(while controlling for IQ and other factors) are less common. Chamberlain and Griliches
(1977) estimate that a one year increase in schooling would increase wages by 6.4 percent. In
a longitudinal study of 799 subjects over 8 years, Ashenfelter and Ham (1979) reported that an
extra year of education increased the average wage rate over the period, by 8.8 percent.
Conservatively, we use a lower bound by assuming one year of additional schooling increases
the wage rate by 6 percent. To arrive at the indirect effect of increased schooling, increased
wages per IQ point is calculated using: (6 percent wage increase/school year) x (0.131 school
years/IQ) = 0.786 percent increase in wages per IQ point.
There is one final indirect effect on earnings. Changes in IQ affect labor force
participation. Failure to graduate high school, for example, correlates with participation in the
labor force, principally through higher unemployment rates and earlier retirement ages. Lead
is also a strong correlate with attention span deficits, which likely reduce labor force
participation. The results of Needleman et al. (1990) relating lead to failure to graduate high
school can be used to estimate changes in earnings due to labor force participation. Using the
odds ratio from Needleman et al., it was estimated that a one IQ-point decrease would also
result in a 4.5 percent increase in the probability of failing to graduate. Krupnick and Cropper
(1989) provide estimates of labor force participation between high school graduates and non-
graduates, controlling for age, marital status, children, race, region, and other socio-economic
status factors. Based on their data, average participation in the labor force is reduced by 10.6
percent for persons failing to graduate from high school. Because labor force participation is
only one component of lifetime earnings (i.e., earnings = wage rate X years of work), this
indirect effect of schooling is additive to the effect on wage rates. Combining this estimate
with the Needleman result of 4.5 percent increase in the risk of failing to graduate high school
per IQ point, indicates that the mean impact of one IQ point loss is a (10.6% x 4.5%) = 0.477
% decrease in expected earnings from reduced labor force participation.
Combining the direct effect on wage rates of 0.5 percent with the two indirect effects
(0.786% for less schooling and 0.477% for reduced labor force participation) yields a total of
1.76 percent decrease in earnings for every loss of one IQ point.
4 Following Schwartz (1990a), this analysis uses the Needleman (1990) to quantify the change in grade
achievement from lead exposure.
Abt Associates. Inc. 5-18 Draft, January JO, 1994
-------
Value of Foregone Earnings
In the next step to monetize intelligence effects, the percent earnings loss estimate must
be combined with an estimate of the present value of expected lifetime earnings. Data on
expected lifetime earnings as a function of educational attainment and sex was reported for the
U.S. population in 1979 by the Bureau of the Census (USDOC 1983). Given the distribution
of the 1979 population with respect to age, educational attainment, and sex, Census used age
specific employment rates and average wage rates to estimate annual earnings as a function of
age, sex and education. Assuming various rates of real wage growth (productivity effect) and
discount factors, the annual earnings stream from age 18 to age 64 was collapsed to a series of
estimates of the present value of lifetime earnings using an assumption of 1 percent real wage
growth and a 7 percent discount rate. Men tend to earn more than women because of higher
wage rates and higher labor force participation. However, for both men and women, expected
lifetime earnings increase greatly with education.
The Census estimates were expressed in 1981 dollars and assumed that the
age/education specific employment and average wage rates would remain constant over time.
A number of issues must therefore be addressed in updating the Census estimates to 1990
dollars. First, educational attainment has changed since 1979, with a greater proportion of the
population attending college, especially a greater proportion of women. Second, wage rates
have increased both due to productivity effects (real wage growth) and inflation. Third, age-
specific employment rates may have changed. Women, in particular, are likely to have higher
rates of labor force participation than in 1979.
In revising the Census estimates, the first issue was addressed by using more recent
data on education. USDOC (1992) provides data on educational attainment for the 1991
population. For this analysis, data on the population over age 25 were used in order to remove
the influence of those individuals too young to have completed schooling. The population data
were used as weighting factors to derive a sex and education weighted average of expected
lifetime earnings. So constructed, the weighted average adjusts the estimate based on 1979
data to current levels of educational attainment. Lifetime earnings were thus calculated to be •
$177,000 for the average work force participant. The next step in adjusting the earnings
estimate is to apply an adjustment for wage growth to update from 1981 to 1990 dollars. The
Bureau of Labor Statistic's Employment Cost Index rose from a level of 67.2 in 1981 to 105.4
in 1990, an increase of about 57 percent. Updating the average lifetime earnings to 1990
dollars yields a revised estimate of $277,616.
While more recent age, sex, and education-specific employment rates could be used to
re-estimate labor force participation, a complete analysis would require steps to dampen the
effects of cyclical unemployment. Such an exercise would require considerable effort and is
beyond the scope of this analysis. To the extent that labor force participation has increased for
specific groups since 1979, the adjusted value presented here underestimates the true expected
lifetime earnings. For example, if the percentage of female children eventually joining the
permanent workforce is greater than the percentage of women over age 25 that worked in
Abt Associates, Inc. 5-19 Draft, January 10, 1994
-------
1979, the expected lifetime earnings of female children would be greater than estimated in this
analysis.
Note that use of earnings is an incomplete measure of an individual's value to society.
Those individuals who choose not to participate in the labor force for all of their working years
must be accounted for, since the lost value of their productive services may not be accurately
measured by wage rates. The largest group are those who remain at home doing housework
and child rearing. Also, volunteer work contributes significantly to social welfare and rates of
volunteerism tend to increase with educational attainment and income.5 If the opportunity
cost of non-wage compensated work is assumed to be the average wage earned by persons of
the same sex, age, and education, the average lifetime earnings estimates would be
significantly higher and could be approximated by recalculating the tables using full
employment rates for all age/sex/education groups. To be conservative, only the value of lost
wages is considered in this analysis.
The adjusted value of expected lifetime earnings obtained above is a present value for
an individual entering the labor force at age 18 and working until age 64. Because a lead-
induced IQ decrement occurs in infancy or childhood, the $277,616 figure must be further
discounted to the specific age at which the health effect is measured and adjusted for the
probability that the infant would survive to age 18. For an infant less than one year old, the
present value of lifetime earnings discounted at seven percent at age 18 and adjusted for
survival would be $80,587. Combining this value with the estimate of percent wage loss per
IQ point yields: $80,587 x 1.76 percent loss/IQ point = $1,414 per lost IQ point.
5.4.2 Valuing Increased Educational Resources
There are two categories of increased educational resources needed as a result of lead
exposure. First, lead exposure results in an increase in the number of children with IQs less
than 70 (note that IQ is not measured until age 7). As these children grow older, they will
need an education program tailored to the mentally handicapped. In addition, some children
whose blood lead is greater than 25 pg/dl will need additional instruction while attending
school later in life.
Children with IQs Less than 70
To value the reduction in the number of infants with IQs less than 70, the reduction in
education costs were measured - a clear underestimate of the total benefits.6 Kakalik et al.
(1981), using data from a study prepared for the Department of Education's Office of Special
Education Programs, estimated that part-time special education costs for children who
remained in regular classrooms cost $3,064 extra per child per year in 1978. Adjusting for
changes in the GNP price deflator yields an estimate of $6,318 per child in 1990 dollars. For
5 Statistical Abstract of the United States, 1986. Table No. 651, p. 383.
6 The largest part of this benefit is the parents' willingness to pay to avoid having their child become
mentally handicapped, above and beyond the increased educational costs.
Abt Associates, Inc. 5-20 Draft, January 10, 1994
-------
the calculations, this incremental estimate of the cost of part-time special education was used to
estimate the cost per year per child needing special education as a result of impacts of lead on
mental development. Costs would be incurred from grades 1 through 12. Discounting future
expenses at a rate of 7 percent yields an expected present value cost of approximately $33,346
per child (assuming compensatory education begins at age 7 and continues through age 18).
Note that this is an underestimate of the cost, since Kakalik et al. measured the increased cost
to educate children attending regular school - not a special education program.
Children with Blood Lead Levels > 25 pg/dl
When calculating the cost of compensatory education, three relatively conservative
assumptions were made. First, it is assumed that no children with blood lead levels below 25
jtg/dl would require compensatory education later in life. This is conservative since many
studies show cognitive effects at IS pg/dl. Second, it is assumed that only 20 percent of the
children above 25 pg/dl would be severely affected enough to require and receive some
compensatory education. Third, based on several follow-up studies that showed cognitive
damage persists for three years or more, even after blood lead levels are lowered, it is assumed
that each child who needed compensatory education would require it for three years (age 7
through 9).
For this analysis it is assumed 20 percent of the children with PbB > 25 pg/dl will
receive compensatory education for three years, but after that, will not.7 The Kakalik et al.
(1981) estimate of part-time special education costs for children who remained in regular
classrooms is also used to estimate the cost of compensatory education for children suffering
low-level cognitive damage. As indicated above, adjusting for changes in the GNP price
deflator yields an estimate of $6,318 per child in 1990 dollars. Discounting future costs at a
rate of 7% annually yields a present value estimate of $11,048 in 1990 dollars.
5.4.3 Valuing Neonatal Mortality
The value of avoiding a statistical death used in this analysis is $2 million. This value
is based on the lower estimate of a range of values provided in a review of studies quantifying
individual's willingness to pay to avoid risks to life by Fisher et al. (1989). The Fisher et al.
lower bound estimate of $1.8 million in 1986 dollars is adjusted for inflation and real income
growth to 1992 dollars using the Gross Domestic Product (GDP) implicit price index.
5.5 COMPUTING BENEFITS FOR FULL MODEL PERIOD
In Chapter 3, we provided a detailed description of the procedures used to inflate the
results obtained for the first year of the modeling time frame to arrive at the results for the
overall 50 year modeling period. As discussed there, the incidence of adverse effects (such as
7 See U.S. EPA (1986) for more detail on the data sources and the nature of the assumptions made to
quantify this benefit category.
Abt Associates, Inc. 5-21 Draft. January 10, 1994
-------
number of IQ points lost, cases of IQ < 70) obtained in the first model year in homes with
lead paint are multiplied by a factor of 48.65, and by a factor of 99.60 for homes without lead
paint to obtain the full 50 year incidence. The benefits of the various decision rules expressed
in terms of avoided incidence of these effects can also be determined for the full model period
by applying these factors to the avoided incidence calculated for the first model year in lead
paint and non-lead paint homes.
To estimate the total value of the benefits over the full 50 year time frame, a similar
pair of multipliers is applied to the value of the benefits obtained for the abatements performed
in the first model year. The expression for computing the total benefits is similar to that given
in Chapter 4 for computing total costs:
where:
Vt is the total value of the benefits
Ali is the avoided incidence of the specific effect of concern;
,* is the undiscounted unit value of the benefit in 1990 dollars, adjusted for additional
expected children in homes getting abated;
r is the discount rate, here 7%.
The adjustment for additional expected children noted above that is included in the
undiscounted unit value of the benefits is included in recognition of the potential benefits to
future children bom into those homes after the abatement is performed in response to the first
births in those homes. As noted in Chapter 3, for each "first born" child in a home, there is
an expected number of an additional 1.55 children in that same home over a subsequent 49
year period. Therefore, for each $1.00 in benefits for the first born child, an additional,
undiscounted $1.55 in benefits is expected for the abatement performed on that home (i.e., this
would be the value if those additional children were born in the same year as the first child).
To account for the expectation that those additional children will be born at some later time, it
is necessary to discount that amount. Discounting that additional $1.55 over the ensuing 50
years with a 7% discount rate, and taking the average of those discounted values results in an
estimated additional benefit of approximately $0.50 for each $1.00 in first year benefits.
Therefore, the undiscounted unit benefit cost provided in Section 5.4 are multiplied by 1.5 for
inclusion in Equation 5-4 , above.
Using procedures similar to those described in Chapter 3, the multipliers obtained are
8.80 for homes with lead paint, and 12.48 for homes without lead paint. That is, the total
value of the benefits is obtained by applying these multipliers to the undiscounted unit value of
the monetized benefit calculated for the abatements performed in the first model year.
Abt Associates. Inc. 5-22 Draft, January 10, 1994
-------
5.6 RESULTS OF MONETIZED BENEFITS
Exhibit 5-9 provides a summary of the total monetized benefits for the various decision
rules considered over the full modeling period. As indicated by the values shown there, the
order of the decision rules based on total monetized benefits is similar to the order obtained
from consideration of changes in the blood lead distribution and avoided IQ point losses.
The largest benefits, totaling $66.7 billion, axe achieved by the 500/400/- decision rule,
followed by the Voluntary Optimum at $48.2 billion. The benefits for the 2300/1200/20 and
2300/1200/- rules have nearly identical benefits of $35.0 and $34.9 billion, respectively.
Following these closely are the -/1200/20 and -/1200/- rules with benefits of $33.0 and $32.9
billion. Considerably lower total benefits are estimated for the 2300/-/20 and 2300/-/- rules,
each with approximately $11.2 billion. Lastly, the -/-/20 and the paint condition only rules
are estimated to produce total benefits of approximately $7.4 and $7.3 billion, respectively.
Exhibit 5-10 displays the relative contribution of the monetized value of each of the
categories of benefits based on the first model year abatements for each decision rule. The
relative contributions of the benefits categories are comparable for the entire modeling period.
By far, the major contribution to the value of the benefits derives from the avoided loss
of IQ points. For all of the decision rules, this component of the benefits contributes between
75 % and 90% of the value of the benefits. The contributions of the avoided incidence of IQ
< 70 and of blood lead levels > 25 /tg/dl are comparable for each decision rule, generally
contributing between 5% and 7% of the total benefits each.
The most variable contributor to the value of the benefits is avoided neonatal mortality.
Except for the Voluntary Optimum and the 500/400/- decision rules, the monetized value of
these benefits are comparable, approximately $140 million for the first year. For the
Voluntary Optimum, neonatal mortality avoidance makes no contribution, while for the
500/400/- rule the value is about $220 million. Excluding the voluntary optimum, the avoided
neonatal mortality benefits as a percentage of the total are lowest for those decision rules with
the lowest total benefits, and highest for those with the lowest benefits. For example, in the
Paint Condition only rule, the first year benefits are estimated to be about $890 million, and
the neonatal mortality benefit at $138 million comprises about 16% of the total. By contrast,
the 2300/1200/20 rule has total first year benefits of $3.5 billion, of which neonatal mortality
at $141 million is only 4% of that total. For the 500/400/- rule, where the total first year
benefits are highest at $6.6 billion, and the neonatal mortality is also highest among all rules at
$220 million. However, as a percent of the total, these benefits constitute only about 3.3% for
the 500/400/- rule.
Abt Associates, Inc. 5-23 Draft, January 10, 1994
-------
Exhibit 5-9. Summary of Monetized Value of Benefits for
Full Modeling Time Frame by Decision Rule
Decision Rule
Voluntary Optimum
Faint Condition Only
Soil = 2300
Dust = 1200
XRF = 20
Soil = 2300
Dust = 1200
No XRF value
Soil = 2300
No Dust Value
XRF = 20
No soil value
Dust = 1200
XRF = 20
Soil = 2300
No Dust value
No XRF value
No soil value
Dust = 1200
No XRF value
No soil value
No dust value
XRF = 20
Soil = 500
Dust = 400
No XRF value
Value of Benefits over
Full Modeling Time Frame
($ Million)
$48,190
$7,319
$34,984
$36,920
$12,249
$33,009
$11,186
$32,949
$7,383
$66,688
Abt Associates, Inc.
5-24
Draft, January 10, 1994
-------
Exhibit 5-10. Distribution of Monetized Benefits by Category for
First Model Year Abatements.
1.1
€> -°
3 ~
< eg
.- c
H
§ •
CO £
$2
$1
$0
Total Benefits
Avoided Q Loss
Avoided Neonatal Mortality
.voided PbB =26
Avoided Q<70
Avoided IQ<70
| Avoided PbB ^5
Avoided Neonatal
Mortality
Avoided IQ Loss
Total Benefits
Abt Associates, Inc.
5-25
Draft, January 10, 1994
-------
6. BENEFIT-COST ANALYSIS
The objective of this chapter is to evaluate alternative sets of criteria which are expected
to induce abatements of lead-contaminated soil, dust, and paint. These sets of criteria are called
decision rules. For paint, its condition is taken as the primary criterion. For soil and dust, the
criteria, called hazard levels, are expressed as concentrations of lead. This chapter also.
considers the possibility of making levels of lead contamination in paint a criterion for inducing
abatement.
6.1 BASIS FOR EVALUATION
The two previous chapters have presented the costs and benefits of different types of
abatements applied to homes with varying levels of lead in soil, dust, and paint and with paint
in varying conditions. While benefits and costs are of independent interest, considering benefits
or costs alone does not permit an evaluation of economic efficiency. On the contrary, it is not
unusual for the decision rule generating the greatest benefits to have the highest costs, possibly
even costs that exceed the benefits. It is also not unusual for the rule having the least costs to
generate very small benefits. To produce significant benefits it may be necessary to expend
significant resources.
To determine which circumstances are the ones that warrant a significant expenditure of
resources requires that the benefits and costs be considered in tandem. Considering them in the
form of net benefits (benefits minus costs) provides a basis for determining whether society will
be better off by implementing any particular decision rule. Comparing different decision rules
on the basis of their respective net benefits provides a means for identifying the best
opportunities to improve economic efficiency. For this benefit-cost analysis, net benefits were
calculated for thousands of different combinations of hazard levels for soil, dust, and paint.1
This chapter focuses its attention on the decision rules that generate the highest net benefits.
6.2 ALTERNATIVE DECISION RULES
The abatement choices made under different decision rules in this analysis can be
characterized in terms of differences in the information that the public is assumed to have. At
one extreme, the public is assumed to have access to the same information that was used in this
analysis: the blood lead levels expected among children if abatement does not take place; the
blood lead reductions achievable from different abatement scenarios; the monetary values of
these blood lead reductions; the cost of the alternative abatement approaches; and the levels of
lead in soil, dust, and paint and the condition of paint in the home. At the other extreme, all
households are assumed to know lead levels and paint condition but only the households that are
1 There are 14,972 different combinations of 30 soil concentrations, 20 dust concentrations, and 22 paint
concentrations. Each combination constituted a potential decision rule in the benefit-cost analysis. Two other
decision rules were also considered, a "voluntary optimum* and a qualitative hazard level based on paint condition,
bringing the total to 14,974 different decision rules considered.
Abt Associates. Inc. 6-1 Draft, January 10, 1994
-------
expected to abate (since their levels exceed candidate hazard levels and/or their paint is in bad
condition) have the set of information necessary to choose the abatement alternative that yields
the highest net benefits. Implicitly this assumes an ability to calculate benefits and costs for
abatement alternatives. However, such a calculation by households may not actually be
necessary. In practice, any household that chooses to abate will need to seek guidance regarding
the combination of soil, dust, and point abatements that is appropriate to its circumstances.
6.2.1 Voluntary Optimum Decision Rule
The first decision rule considered in this analysis generates the highest net benefits of all
decision rules evaluated. It serves as the benchmark against which all alternative decision rules
can be judged. This decision rule is called the voluntary optimum because it is based upon
households' determining which abatement alternative generates the highest net benefit and their
voluntarily choosing to undertake that optimal abatement. Decisions under the voluntary
optimum are assumed to draw upon sufficient information to determine whether it is optimal to
abate and if so, to choose the optimal abatement alternative. In subsequent decision rules,
households are assumed to have less information and therefore less leeway to choose on their
own, instead being induced by EPA's hazard levels to initiate abatements of certain types. The
evaluation of the voluntary optimum indicates that a substantial number of households would find
it in their vested interest to have their homes abated.2
Examples
To illustrate the implications of allowing households to choose their abatements this way,
examples for particular households are presented in Exhibit 6-1. Home A has a mean
concentration of lead in soil of 8,800 ppm, mean concentration of interior dust of 1,100 ppm,
and lead-based interior paint that is in damaged condition and has an XRF reading of 11.
Relevant choices for this house are: high-end paint abatement, low-end paint abatement, high-end
soil abatement, low-end soil abatement, recurrent dust abatement, four combined abatements
(high-end paint abatement and high-end soil abatement, high-end paint abatement and low-end
soil abatement, low-end paint abatement and high-end soil abatement, low-end paint abatement
and low-end soil abatement). Nonrecurrent dust abatement is not a viable option because paint
and soil would remain sources of lead contamination.
The gross benefits considered by each household were assumed to include not only the private benefits
accruing to that household from current and future members but also the social benefits accruing to any household
living in the home in the 50 years after the birth of the child that is triggering the abatement decision. Based upon
a 7% discount rate, a substantial portion of the benefits from abatement, approximately 69%, are associated with
protecting the first child alone. (For households having additional children, the private proportion of the social
benefits will be higher.) In many cases, net benefits can be positive when only this portion of the benefits is
compared to costs. Where that is not the case, there is also the prospect that the homeowners can recoup all or part
of their abatement costs in an increased value of the home at the time of sale. This increased value may stem from
the possibility that the new owner places a premium on having a lead-abated home, from the spillover improvements
in the home that come from certain types of abatement (such as the aesthetic improvements that come from high-end
paint abatement), or from both.
Abt Associates. Inc. 6-2 Draft, January JO, 1994
-------
Under the voluntary optimum, home A chooses low-end soil abatement, which generates
net benefits of $3,607. High-end soil abatement would generate higher benefits but
disproportionately higher costs, so much so that the resulting net benefits are negative. The
damaged condition of paint does not justify paint abatement. Consequently, paint abatement
options generate negative net benefits, with the exception of combining low-end soil and low-end
paint abatement. This combination results in positive net benefits but lower than what can be
accomplished by low-end soil abatement alone since low-end paint abatement generates negative
net benefits.3
After the abatement, soil levels would be 500 ppm, predicted dust levels are still above
1000 ppm since paint contamination has not been abated, and some soil contamination still
exists. No further abatements appear to be warranted. As indicated above, undertaking paint
abatement in addition to soil abatement generates negative net benefits. Additional abatement
directed at the remaining dust, through recurrent dust abatement, also generates negative net
benefits.
Home B has approximately the same dust contamination as home A (1000 ppm) but no
lead-based paint and soil contamination of only 100 ppm. The only relevant abatement choices
for this home are recurrent and nonrecurrent dust. The former is not needed since there is no
known major source of lead. Negative net benefits for recurrent dust abatement reflect this
phenomenon. Instead, nonrecurrent dust abatement is chosen since it generates positive net
benefits of $2,614. After abatement, this home will have dust and soil levels of 100 ppm.
Home C provides an example where dust abatement is optimal even where lead
contamination is present in soil and damaged interior paint as well as in dust. This example
illustrates the substitutability among lead abatements for different media. For each of the three
media (soil, dust, and paint), some form of abatement could be selected that produces positive
net benefits. These are low-end paint abatement, high-end soil abatement, recurrent dust
abatement, and nonrecurrent dust abatement. Choosing the latter generates the highest net
benefits, $2,555, more than twice the net benefits from the next best alternative - low-end paint
abatement. Being able to exploit the substitutability by choosing the one with the highest net
benefits is one key to the high net benefits of the voluntary optimum. Inducing households to
meet medium-specific targets can reduce the achievable net benefits. For example, by setting
a soil hazard level of 1100, it is assumed that the owner of home C would be induced to
undertake high-end soil abatement since this is the only means for getting below this specific
hazard level.
The net benefits of combining low-end soil and low-end paint abatements are higher than the sum of each
of these abatements alone because the blood lead reductions implied by a given change in a lead source are not
constant. They are lower at higher levels.
Abt Associates. Inc. 6-3 Draft, January 10, 1994
-------
Exhibit 6-1
Examples of Abatement Chokes Under the Voluntary Optimum Decision Rule
Home
A
B
C
D
E
Soil
ppm
8800
100
1100
0
800
Dust
ppm
1100
1000
3300
200
1200
Paint
nig/
en1
11
0
5
20
1
Condition
Damaged
Intact
Damaged
Damaged
Intact
HP
(9336)
N/A
(4046)
(6227)
(8349)
LP
(2203)
N/A
1208
55
(2144)
HS
(9364)
N/A
157
N/A
(3055)
Net Benefits ($1990)
LS
3607
N/A
(1587)
N/A
(5181)
RD
(«6«3)
(4312)
92
(7307)
(4138)
HP/HS
(13963)
N/A
(«145)
N/A
(12097)
HP/LS
(2466)
N/A
(8062)
N/A
(14015)
LP/HS
(10520)
N/A
(1689)
N/A
(5782)
LP/LS
2356
N/A
(3240)
N/A
(7912)
NRD
N/A
2614
2555
N/A
(161)
Selected
AhafflamMf
Low-Cad
Sofl(LS)
Non-
Recurrent
Dot (NRD)
Non-
R0cwnnt
Dust (NRD)
Low-
End Paint
(LF)
None
Kev
HP = High-End Paint Abatement
LP = Low-End Paint Abatement
HS =• High-End Soil Abatement
LS = Low-End Soil Abatement
RD = Recurrent Dust Abatement
HP/HS = High-End Paint Abatement and High-End Soil Abatement
HP/LS = High-End Paint Abatement and Low-End Soil Abatement
LP/HS - Low-End Paint Abatement and High-End Soil Abatement
LP/LS = Low-End Paint Abatement and Low-End Soil Abatement
NRD = Nonrecurrent Dust Abatement
The designation of N/A for net benefits means that either the associated abatement is not applicable because no lead is
present from the source abated (i.e., soil, dust, or paint) or that the initial level of lead is at or below the assumed post-
abatement level. In either case, abatement of the source produces no benefits.
Abt Associates, Inc.
6-4
Draft, January 10, 1994
-------
In the voluntary optimum, paint abatement is the optimal choice in only a few homes.
Home D illustrates the special circumstances where low-end paint abatement was the optimal
choice. Essentially, paint abatement is the best choice because neither soil abatement nor dust
abatement is relevant. There is little or no contamination in these media; therefore no
substitutability among media to abate can be exploited. Paint abatement is also justified because
the XRF reading is so high. The latter results in expected damages from pica, which is assumed
to have a 25 % chance of being exhibited by the first child born into this home after abatement,
that are large enough to justify abatement. The net benefits from low-end paint abatement are
small but positive.
The final example, home E, shows that choosing not to abate can also be an optimal
choice, even when there is lead contamination in each of the three media - 800 ppm in soil, 1200
ppm in dust, and a maximum XRF of 1 for interior paint. It is optimal not to abate because
each abatement alternative has negative net benefits. This outcome further underscores how
other decision rules besides the voluntary optimum lead to lower net benefits. For example,
suppose that a decision rule were constructed based upon a soil level of 800 ppm and that the
owner of home E is induced to abate because this level has been set by EPA. At best, this
homeowner could expect to achieve only negative net benefits, -$3,055. In the aggregate, if
there are a large number of homeowners in similar circumstances, this particular hazard level
can lead to a substantial portion of the induced abatements having negative net benefits.
Consequently, how well a particular set of hazard levels performs relative to others in terms of
aggregate net benefits depends on how well it can avoid inducing abatements in homes where
they are not warranted.
Results for the Voluntary Optimum
Over the course of the 50 years of births modeled in this analysis, 45 million abatements
would be initiated under the voluntary optimum decision rule. As indicated in Exhibit 6-2,
approximately 99% are nonrecurrent dust abatements. Initial lead concentrations in the dust
range from 200 to 5,900 ppm. Lead concentrations in soil are less significant, ranging from 100
to 1,100 ppm. Low-end soil abatements account for a little more than 1 %. In these homes, soil
concentrations, which are much higher than for the homes receiving nonrecurrent dust
abatement, range from 3,000 to 8,800 ppm. Dust concentrations, which can be affected
indirectly by soil abatement, range from 1,100 to 5,800. The small remaining fraction of
abatements entail low-end paint abatement. As anticipated from the example cited above, the
interior XRF readings are high (20 to 22) and the dust and soil levels for the paint-abated homes
are low. Among the homes where the choice is made not to abate, the maximum contamination
levels can be as high as 3,100 ppm for soil, 2,500 ppm for dust, and an XRF of 22 for paint.
The present discounted value of the net benefits under the voluntary optimum is
approximately $34 billion. These results imply average discounted benefits of $1,067, average
discounted costs of $308, and average discounted net benefits of $759. The discounted net
benefits per home range from a low of $11 to a high of $13,126.
These figures do not reflect the testing and inspection costs that are a prerequisite to
determining what, if any, abatement is justified in a home. To determine lead levels in soil and
Abt Associates, Inc. 6-5 Draft, January 10, 1994
-------
dust in all homes where a birth is imminent and paint levels in homes where a birth is imminent,
the undiscounted cost of testing is $713.
Ideally this cost would be included in the benefit-cost calculation made for each house.
While the testing costs have to date not been included, subtracting the total testing costs from
the total net benefits provides a sufficient evaluation given certain assumptions. Excluding
testing costs from the individual abatement decision could have a significant effect on the gross
benefits but possibly not the number of abatements. Given estimated testing costs of $713 per
home, it is assumed that some homeowners would not choose to test if the expected pay-off from
the test (the likelihood of a positive test result times the net benefit from the best abatement
choice) is less than the cost of testing. Following this kind of logic in the modelling would
likely lower the testing costs, the number of abatements, and the benefits and costs. However,
for homeowners who choose to test without regard to the expected pay-off, the number of
abatements would not change. When the net benefits of abatement are positive, abatement will
be initiated. The testing costs are sunk costs that do not affect the decision to abate. In this
situation, the number of abatements and the gross benefits may not be too different from that
estimated currently for the voluntary optimum although the net benefits will.
By subtracting the total testing costs of $24 billion from the total net benefits, it is
possible to provide an estimate of net benefits under these circumstances. The resulting overall
net benefits of $10 billion from the voluntary optimum are still substantial.
Abt Associates, Inc. 6-6 Draft, January 10, 1994
-------
Exhibit 6-2
Voluntary Optimum
Abatement
Tolih
Statagr
Selected
Nona
LS
LP
NRD
Banalita
llnrillionl
2.789
26
45.376
48.190
Com
ItnrillionJ
1.802
26
11.878
13.888
•at Banatila
(•million)
0
787
0
33.487
34.284
Total Number
ol Abatemante
IIOOOi)
0
577
21
44.567
46.185
Proportion
ol
Abatemente
1.28%
0.05%
88.68%
Number of Abatemente
With Negative
0
0
0
0
0
• a* i j a
Individ ml
MB! BsiMfitti
0
670
47
11
Maximum
Individual
NetBenelita
0
8.843
55
13.126
Minlmuni
Soil Level
taunj
0
0
0
Maximum
Soil 1ml
fepm)
3.100
6.800
100
1.100
MI_iL«_B
ininiiniin
Duet level
fern)
0
1.100
ML
Hiaumun
Dual Laval
2.500
400
Minimum
Interior Paint
ft i ___n
0
0
20
0
Maximum
Interior Paint
22
11
22
13
Abate
ntCo
HP - High Paint Abatement
LP - Low Paint Abatement
HS - High Sol Abatement
LS - Low SoB Abatement
RD - Reeumnt Duet Abatement
HP/HS • High Paint, High Soil Abetomente
HP/IS - High Petal, Low Soil Abatement.
LP/HS - tow Paint. High SoB Abatemente
LP/LS » Low Paint. Low SoB Abatemante
NH D » Nonreemrent Duet Abatement
-------
6.2.2 Decision Rules Based Upon Induced Abatements
Since the current rate of abatement does not appear to be as great as that implied by the
voluntary optimum, it is reasonable to wonder what interventions are necessary to induce
homeowners to initiate abatement. Section 403 of TSCA proposes hazard levels that have the
potential to serve as a means of inducing abatement. The strong assumption made in this
analysis is that owners of homes where any given hazard level is exceeded and where a child
is about to be born will be induced to abate.
To evaluate a particular set of candidate hazard levels that EPA might consider setting,
this analysis characterizes these levels as restrictions to the behavior modelled under the
voluntary optimum. For example, suppose that the Agency were to set a soil hazard level of
1100 ppm and a dust hazard level of 1200 ppm. Using these criteria as constraints on each
household's decision of whether and how to abate, this analysis identifies the highest net benefits
achievable while adhering to EPA's hazard levels. Relative to the pure voluntary optimum, this
circumstance reflects a constrained optimization. Therefore, the net benefits will be lower since
households may be induced by EPA's guidance to make abatement choices which had lower or
even negative net benefits under the voluntary optimum. Referring to Exhibit 6-1 again, home
E would be induced by this set of hazard levels, particularly the 1200 ppm threshold for dust,
to undertake dust abatement even though the net benefits are negative.
Since EPA's promulgation of hazard levels does take the form of guidance for most
households and does not bind any household to adhere to them, the circumstances of the
individual household's decisionmaking are assumed to be different in this constrained
optimization that they were under the voluntary optimum. The difference can be described in
terms of differences in the information households are assumed to have about the benefits and
costs of lead abatement. Since information is costly to the household, the assumption of perfect
information being available to the household is a strong one. By providing guidance to
households in the form of hazard levels at which abatement should be undertaken, EPA can
lower the actual information costs for the household abatement decision. However, rather than
providing each household with perfect information, the hazard levels promulgated by EPA would
serve as general guidance and not the type of information that allows each household to
determine exactly what is best.
The result is that some households will make the same abatement choice as they would
if they had perfect information, some will be induced to choose a form of abatement that
provides lower net benefits, and some will be induced to abate when they would have chosen
not to under the voluntary optimum. Taking the candidate hazard levels mentioned (soil = 1100
ppm and dust = 1200 ppm), these outcomes can be illustrated using the homes highlighted in
Exhibit 6-1. Home A exceeds the soil hazard level. No change in behavior is necessary since
low-end soil abatement, which was chosen under the voluntary optimum, is sufficient to reduce
soil contamination below this threshold. Home C exceeds both the soil and dust thresholds. The
optimal choice under the voluntary optimum, nonrecurrent dust abatement, does not address soil
contamination. Consequently, the best choice that allows getting below both thresholds is high-
end soil abatement, which has positive net benefits of $157 but these are lower than those
achievable through nonrecurrent dust abatement ($2,555). Finally, the example of home E,
Abt Associates, Inc. 6-8 Draft, January 10, 1994
-------
where no abatement is optimal under the voluntary optimum, would be induced to choose
nonrecurrent dust abatement, with negative net benefits (-$161), to meet the dust hazard level.
Taken together, the overall net benefits for these three homes of selecting abatements are
positive. For the full set of homes considered in this analysis, the overall net benefits of
selecting abatements based upon a particular set of candidate could be positive or negative. Only
by setting hazard levels specific to each house is it possible to replicate the voluntary optimum.
Since it does not yet appear feasible for EPA to structure the hazard levels it promulgates this
way, a simpler and more general set of hazard levels were assumed to be more likely.
One important objective of this analysis was to identify sets of hazard levels of different
types which compromised least the net benefits achievable under the pure voluntary optimum.
The sets of hazard levels, referred to as decision rules, are of four types. See Exhibit 6-3. The
first type of decision rule addressed paint condition. It assumes that all homes with non-intact
paint will undertake paint abatement. The second type of decision rule included this criterion
as well as a hazard level for each of the three media that are addressed by this rule - soil, dust,
and paint. In Exhibit 6-3, these are labeled "single-medium constrained." The third type of
decision rule encompasses hazard levels for two media, as well as condition. The fourth type
of decision rule addresses all three media and condition. Each of these will be considered in
turn.
Exhibit 6-3
Taxonomy of Decision Rules
Decision Rules
Condition
Constrained
Single-Medium
Constrained
Two-Media
Constrained
Three-Media
Constrained
Note: A "Yes" inc
rule.
Soil
(ppm)
—
Yes
—
—
Yes
Yes
—
Yes
Dust
(ppm)
—
—
Yes
—
Yes
—
Yes
Yes
Paint
(XRF)
—
—
—
Yes
—
Yes
Yes
Yes
icates that a hazard level is fixed for that particular med
Paint
Condition
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
urn in the decision
Abt Associates, Inc.
6-9
Draft, January 10, 1994
-------
Qualitative Hazard Level for Faint Condition
The first type of decision rule targets the condition of paint only. Under this rule, only
houses with non-intact paint are induced to abate. Exhibit 6-4 presents the benefit-cost results
for this decision rule. The total number of abatements only reflect the abatement of homes with
lead-based paint in bad condition. The abatement of other homes, which would be expected
under the voluntary optimum, are not included, given the logic of this decision rule that only
the qualitative hazard level will induce abatement. Also, in contrast with the voluntary
optimum, certain homes are induced by the decision rule to undertake paint abatement even
though this creates negative net benefits. Of the 7.1 million abatements induced by this decision
rule, 6.7 million (95%) have negative net benefits. Overall, the net benefits for this decision
rule were negative (-$17.3 billion).
Exhibit 6-4 also shows the distribution of abatements by category. Since the decision
rule is specifically targeted at paint, only the paint abatement alternatives are chosen.4 High-end
paint is the optimal choice for 59% of the homes; low-end paint is the choice in 39% of the
homes. Owners of other homes chose combinations of paint abatement and soil abatement, since
these choices have higher net benefits than paint abatement alone. The numbers of high-end and
low-end paint abatements with negative net benefits are approximately proportional to the total
numbers of abatements associated respectively with each.
Exhibit 6-4 also presents the abatement-specific net benefits, both in aggregate and the
range of undiscounted values for individual homes. These results underscore the difficulty with
specifying an across-the-board rule. There are cases of positive net benefits for individual
homes undertaking low-end paint abatement (where the maximum net benefits are $7,142), high-
end paint abatement (where the maximum net benefit is $2,629) and combined low-end
paint/low-end soil abatements (where the maximum net benefits are $2,356). However, the
dominant outcome is that this decision rule entails substantial negative net benefits. Of the 7.1
abatements induced by this decision rule, 6.7 million (95%) generate negative net benefits.
Taken literally, a paint condition criterion requires abating pain in bad condition only. Neither paint
abatement alternative used in this analysis is tailored particularly to abating deteriorating paint alone. Instead, high-
end paint abatement entails complete and full abatement of interior lead-based paint and low-end paint abatement
involves full abatement of windows only. As such, the costs and effectiveness of these alternatives are taken as
proxies respectively for abatements addressing "extensive" and "limited" amounts of deteriorating paint.
Abt Associates, Inc. 6-10 Draft, January 10, 1994
-------
Exhibit 8 • 4
Pilnl Condition Criterion
hdueod
No
Fold:
suton
Bobctid
linn*
IHmB
IP
HPfl.8
Totol 60- Yoor
Bonollto
II million)
0
1.456
564
7.319
Total 60-Vur
Coolo
llmllllonol
0
20.134
1517
943
74
24.668
Totol 60-Voor
Not Bonollto
(Imllllanol
0
114.926)
12X1611
079)
17
117.349)
Totol Numbor
ol Abotomonto
(lOOO-ol
0
4.160
2.774
114
16
7X164
Proportion
ol
Abotomonto
56.88%
39.27%
1.61%
0.00%
Abotomonto
WlthNogotrm
Not Bonollto
IIOOO-o)
JD_
4JJ27
2.544
JJ4_
0
6.685
lodbbJnl
Not Bonollto
III
0
19.690)
12,657)
Morimin
IndMdal
Nit Binif Itt
in
0
2,829
7,142
U58
Mmknom
SollUvol
(ppml
0
0
0
3.100
8300
SoDlonl
(ppml
3.100
2.200
3.100
8.600
8.800
Minimum
OuotUnl
Ippml
0
0
1.100
•nxnm
DuolbMl
Ippml
6400
5400
1.100
laloriorPimt
1
1
11
Moxknm
mtortorPobri
Imalem')
22
11
AbMamani CodM
HP = High P.lnl Abnamenl
IP = Low Palm Abitamanl
HS " High Soil AbMamani
L8 = law Boll Abalomoni
RD » Reeurronl Dutt Abatanunl
HP/HS • High Palm. High Soil Abalamanla
HP/US . High Palnl. Low Soil AbalamonM
LP/HS = Low Paint. High Soil Abalomanta
IP/IS . Low Palm. Low Soil Abaiamonta
NR O • Nonracutram Dual Abaiamani
-------
Single Medium Hazard Levels
The effects of setting a hazard level for a single medium alone - for soil, dust, or paint -
in addition to a qualitative hazard level for paint condition are considered in this section. While
the intent of Section 403 is to set hazard levels for soil, dust, and paint, the results for these
single-medium decision rules illustrate that the individual media make very different
contributions to a three-media approach to addressing residential lead risks. As will be shown
below, decision rules based upon two of the three individual media (soil and paint) lead to
negative net benefits. The net benefits of the best dust hazard level are more than $20 billion
higher than the net benefits of either the best soil or paint hazard levels.
For each medium, a wide array of candidate decision rules are considered. Choosing the
best among these alternatives is based upon a comparison of net benefits. For soil, the benefit-
cost results for thirty alternative hazard levels ranging from 100 ppm to 3,000 ppm are given
in Exhibits 6-5 and 6-6. For paint, the results for twenty-two XRF readings from 0 to 21 are
presented in Exhibits 6-7 and 6-8. For dust, the results for twenty alternative hazard levels
ranging from 100 ppm to 2,000 are presented in Exhibits 6-9 and 6-10.
Given the framework of a single-medium decision rule combined with a paint condition
criterion, a hazard level of 2,300 ppm for soil would achieve the highest net benefits but these
are negative (-$17.2 billion). Separate analyses indicate that the paint condition criterion is
responsible for the bulk of the negative net benefits. This finding was anticipated from the earlier
results showing large negative net benefits from the previous decision rule involving paint
condition alone.
The optimal hazard level for paint is an XRF reading of 20 me/cm2 but again the net
benefits are negative (-$17.6 billion). While there are homes having XRF levels equal to or
greater than 20 whose owners choose paint abatement under the voluntary optimum, their
number was small (20,503) and they choose low-end paint abatement only. In contrast, the
owners of more than 4 million homes choose to conduct high-end paint abatement because this
decision rule combines the paint condition criterion with the single-medium paint hazard level.
Almost 95 % of the induced abatements have negative net benefits.
The best prospect among the single-medium decision rules is one based upon dust. A
single hazard level of 1200 ppm for dust combined with the paint condition criterion generates
net benefits of $3.3 billion. The net benefits are higher for the single dust hazard level because,
unlike the previous two cases, there is greater leeway for homeowners to choose abatements with
higher pay-off, especially but not exclusively, the two forms of dust abatement. Consequently,
the accuracy of this decision rule in targeting homes with positive net benefits is higher. Of the
15.6 million abatements induced by this decision rule, only 46% entail negative net benefits.
Abt Associates, Inc. 6-12 Draft, January 10, 1994
-------
Exhibit 6-5
Single Medium Hazard Levels: Soil
Decision Rules
Soil
(ppm)
2.900
2,800
2,700
2.600
2,500
2,400
2.300
2,200
2,100
2.000
,900
,800
,700
.600
,500
,400
1,300
1,200
1,100
1,000
900
800
700
600
500
400
300
200
100
Nonintact Paint
Abatement
Recommended
YES
YES
YES
YES
YES
YES
YES
YES
YES
YES
YES
YES
YES
YES
YES
YES
YES
YES
YES
YES
YES
YES
YES
YES
YES
YES
YES
YES
YES
YES
Benefits
($ millions)
9,983
,186
,186
,186
,186
,186
,186
11.186
14,645
15,240
15,240
15,240
15,240
15,240
15,571
15,571
15,750
15,906
16,864
17,091
17,441
18,708
18,708
20,539
20,628
21,211
31,968
34,207
37,582
40,049
Abatement
Costs
($ millions)
27,241
28,345
28,345
28,345
28,345
28,345
28,345
28,345
33,471
34,466
34,466
34,466
34,466
34,466
34,896
34,896
35,403
35,731
37,862
38,189
38,547
45,155
45,155
48,149
49,131
51,656
88,648
94,859
111,203
136,435
Net Benefits
($ millions)
(17,258!
(17,159!
(17,1591
(17,1591
(17,159]
(17,159!
(17,159]
(17,1591
(18,826]
(19,226]
(19,226)
(19,226)
(19,226)
(19,226)
(19,325)
(19,325)
(19,652)
(19,824)
(20,998)
(21,098)
(21,106)
(26,447)
(26.447)
(27,611)
(28,503)
(30,444)
(56,681)
160,651)
(73,620)
(96,386)
Total Number
of Abatements
(1000s)
7.750
8,069
8,069
8,069
8,069
8,069
8,069
8,069
10,139
10,486
10,486
10,486
10,486
10,486
10,649
10,649
10,761
10,828
11,445
11,513
11.586
13,189
13,189
14,104
14,320
14,883
16,922
18.777
23,259
29,157
Induced
Abatements
With Negative
7.154
7.154
7.154
7.154
7.154
7,154
7.154
7.154
9.224
9.571
9,571
9.571
9.571
9.571
9.734
9.734
9.845
9.913
10.530
10.597
10,606
12,209
12,209
13.124
13,340
13,912
16.722
18.577
23,059
28,958
-------
Exhibit 6-6
Single Medium Hazard Levels for Soil: Types of Abatement
Soil
(ppm)
2,900
2,800
2,700
2,600
2,500
2,400
2,300
2,200
2,100
2,000
,900
,800
,700
,600
,500
1,400
1,300
1,200
1,100
1,000
900
800
700
600
500
400
300
200
100
Nonintact Pain
Abatement
rlouulllllieiiuau
YES
YES
YES
YES
YES
YES
YES
YES
YES
YES
YES
YES
YES
YES
YES
YES
YES
YES
YES
YES
YES
YES
YES
YES
YES
YES
YES
YES
YES
YES
(1000s)
4,160
4,160
4,160
4,160
4,160
4,160
4,160
4,160
4,147
4,147
4,147
4,147
4,147
4,147
4,147
4,126
4,113
4.113.
4,100
4,086
3,898
3,898
3,848
3,807
3,714
3,426
3,394
3,263
2,918
(1000s)
2,716
2,716
2,716
2,716
2,716
2,716
2,716
2,716
2,716
2.708
2,708
2.708
2.708
2,708
2.708
2.708
2,694
2,685
2,685
2,677
2,668
2,545
2,545
2,512
2,485
2,425
2,237
2,216
2,131
1,905
HS
(1000s)
0
0
0
0
0
0
0
0
0
0
0
0
0
163
163
163
230
230
298
371
371
371
1,287
1,287
1,456
9,858
11,713
16,195
22,094
LS
(1000s)
686
,006
,006
,006
,006
,006
,006
,006
3,076
3,423
3,423
3,423
3,423
3,423
3,423
3,423
3,534
3,534
4,151
4,151
4,151
5.754
5,754
5,754
5,970
6,363
0
0
0
0
RD
(1000s)
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
HP/HS
(1000s)
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
13
13
26
40
40
40
90
90
122
847
879
1,010
1,355
HP/LS
(1000s)
114
114
114
114
114
114
114
114
114
126
126
126
126
126
126
126
147
147
147
147
147
336
336
336
377
438
0
0
0
0
LP/HS
(1000s)
0
0
0
0
0
0
0
0
0
0
0
Qj
0
0
0
0
0
8
8
17
26
26
26
58
58
80
553
574
659
885
^•^••^^•TB
LP/LS
(1000s)
74
74
74
74
74
74
74
74
74
82
82
82
82
82
82
82
96
96
96
96
96
219
219
219
246
286
0
0
0
0
NRD
(1000s)
0
o
0
o
o
0
o
o
0
0
0
o
0
0
0
0
o
o
0
0
0
o
0
0
0
0
o
0
0
0
-------
Exhibit 6-7
Single Medium Hazard Levels: Paint
Decision Rules
Paint
(mg/cm1)
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
0
Nonintact Paint
Abatement
Recommended
YES
YES
YES
YES
YES
YES
YES
YES
YES
YES
YES
YES
YES
YES
YES
YES
YES
YES
YES
YES
YES
YES
Benefits
($ millions)
7,383
7.383
7,870
7,870
7,870
7,870
7,870
7,870
7,870
8,069
8,118
9,011
10.502
11,719
12.170
12,383
16,558
17,460
17,574
18,491
19,479
25,516
Abatement
Costs
($ millions)
25,013
25.013
27.605
27,605
27.605
27.605
27,605
27.605
27.605
27.837
28.300
29,398
33,541
36,420
37,767
39,188
42,626
43,908
45,164
50,874
60,588
102,784
Net Benefits
($ millions)
(17,631)
(17.631)
(19.735)
(19,735)
(19,735)
(19,735)
(19,735)
(19,735)
(19,735)
(19,767)
(20,182)
(20,387)
(23,039)
(24,701)
(25,597)
126,805)
(26,068)
(26,448)
(27,591)
(32,382)
(41,109)
(77,268)
Total Number
of Abatements
(1000s)
7,164
7,164
7,920
7,920
7,920
7,920
7,920
7,920
7,920
7,987
8,123
8,316
9,238
10,078
10,471
10,885
11,888
12,262
12,628
14,293
17,126
29,431
Induced
Abatements
With Negative
Benefits (1000s)
6,786
6,786
7.542
7.542
7,542
7,542
7,542
7,542
7,542
7,582
7,718
7,861
8,784
9.623
10,016
10,430
10,712
10,982
11,348
13,013
15,846
27,934
-------
Exhibit 6-8
Single Medium Hazard Levels for Paint: Types of Abatement
Decision Rules
Paint
(mg/cm1)
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
0
Nonintact Paint
Abatement
Recommended
YES
YES
YES
YES
YES
YES
YES
YES
YES
YES
YES
YES
YES
YES
YES
YES
YES
YES
YES
YES
YES
HP
(1000s)
4,221
4,221
4,678
4,678
4,678
4,678
4,678
4,678
4,678
4,719
4,801
4,841
5,116
5,624
5,861
6,112
6,719
6,945
7,166
8,174
9,888
17,332
LP
(1000s)
2,814
2,814
3,113
3,113
3,113
3,113
3,113
3,113
3.113
3,139
3,193
3,219
3,584
3,915
4,071
4,234
4,630
4,778
4,923
5,580
6,699
11,560
HS
(1000s)
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
LS
(1000s)
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
RD
(1000s)
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
HP/HS
(1000s)
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
HP/LS
(1000s)
114
114
114
114
114
114
114
114
114
114
114
190
474
474
474
474
474
474
474
474
474
474
LP/HS
(1000s)
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
LP/LS
(1000s)
16
16
16
16
16
16
16
16
16
16
16
65
65
65
65
65
65
65
65
65
65
65
NRD
(1000s)
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
-------
Exhibit 6-9
Single Medium Hazard Levels: Dust
Decis
Diiot
(ppm)
2.000
.900
.800
.700
.600
,500
,400
.300
,200
,100
,000
900
800
700
600
500
400
300
200
100
on Rules
Mnnin4af»t Point
Abatement
Recommended
YES
YES
YES
YES
YES
YES
YES
YES
YES
YES
YES
YES
YES
YES
YES
YES
YES
YES
YES
YES
Benefits
Ifi millinnat
26,227
26,227
26,227
26,366
28,342
28,846
29,669
31,413
32,945
35,856
37,227
41,932
43,145
46,438
49,416
58,373
63,230
64,003
71,579
77,260
Abatement
UOS1S
($ millions)
26,656
26,656
26,656
26.682
27,218
27,292
27,557
29,128
29,646
35.245
37,354
42,245
44.872
50.736
53.947
64,534
70.785
83,966
121,989
155,953
Net Benefits
I ft millions!
\9 minions I
(429)
(429)
(429)
(317)
1,124
1,554
2,112
2,285
3,299
611
(127)
(313)
(1,727)
(4,298)
(4,531)
(6,161)
(7,555)
(19,963)
(50,411)
(78,693)
Total Number
. ...
of ADBtements
(1000s)
12.370
12,370
12,370
12,446
13,351
13,564
13,954
14,654
15,602
17,634
17,909
20,378
21.019
23,506
25,962
34,334
43,883
51,978
63,963
83,759
Induced
Abatements
With Negative
Benefits (1000s)
6,753
6.753
6,753
6,753
6,762
6.762
6,803
7.031
7.114
8,491
8.781
9,149
9,625
11,209
12,168
15,402
19.756
22.796
30,267
44.358
-------
Exhibit 6-10
Single Medium Hazard Level for Dust: Types of Abatement
Dust
(ppm)
2.000
,900
,800
,700
,600
,500
,400
1,300
1,200
1,100
1,000
900
800
700
600
500
400
300
200
100
Nonintact Paint
Abatement
Recommended
YES
YES
YES
YES
YES
YES
YES
YES
YES
YES
YES
YES
YES
YES
YES
YES
YES
YES
YES
YES
HP
(1000s)
4,160
4,160
4,160
4.160
4,160
4,160
4,160
4,177
4,186
4,289
4,302
4,397
4,345
4,449
4,454
4,469
4,476
4,125
4,413
4,167
LP
(1000s)
2,766
2,766
2,766
2,766
2,757
2,757
2,783
2,740
2,731
2,627
2,572
2,516
2,329
2,239
2,192
2,087
1,998
1,894
1,190
692
HS
(1000s)
0
0
0
0
74
74
74
74
74
74
74
222
222
148
804
967
967
2,092
5,751
8,983
LS
(1000s)
91
91
91
91
91
91
91
411
411
411
411
411
411
1,028
1,028
3,098
3,379
0
0
0
RD
(1000s)
67
67
67
67
67
67
108
202
276
1,581
1,824
2,811
3,352
4,281
4,521
5,359
6,150
10,772
17.372
22,126
HP/HS
(1000s)
0
0
0
0
0
0
0
0
0
0
21
35
35
45
73
73
73
1,026
1,460
2,205
HP/LS
(1000s)
114
114
114
114
114
114
114
114
114
172
314
314
439
439
452
541
623
0
0
0
LP/HS
(1000s)
8
8
8
8
18
18
18
18
18
9
9
28
42
32
19
19
19
19
0
0
LP/LS
(1000s)
16
16
16
16
16
16
16
16
16
16
0
0
0
0
0
0
0
0
0
0
NRD
(1000s)
5.148
5,148
5,148
5.224
6,055
6,268
6.591
6,903
7,777
8,455
8.381
9,642
9,845
10.845
12.420
17,721
26,198
32,050
33,776
45,587
-------
Exhibit 6-11 provides greater detail on the benefit-cost results for a hazard level of 1200
ppm. Of the 15.6 million abatements associated with this decision rule, 8.1 million (52%) entail
dust abatement alone (recurrent or nonrecurrent). It follows that the number of negative net
benefits induced by this decision rule is a smaller percentage than was the case for either of the
previous two decision rules, as indicated above.
Most of the successful targeting of homes for abatement is associated with the choice of
nonrecurrent dust abatement. Of the 7.7 million homes induced to undertake recurrent dust
abatement, 97% accrue positive net benefits. While these abatements are substantial, the two
other media are also addressed. Furthermore, all of the high-end and low-end soil abatements
generate positive net benefits, except when combined with paint abatement. The maximum net
benefits per home are as high as $533 and $8,843 respectively for high-end and low-end soil
abatement. All of the combined abatements involving high-end paint or high-end soil abatement
have negative net benefits. Finally, the vast majority of the high-end and low-end paint
abatements generate negative net benefits. Together, the net benefits of all high-end and low-end
paint equal minus $17 billion. Again, this result was anticipated from the evaluation of the
decision rule based on a paint condition criterion alone. Without the paint condition criterion,
the net benefits of this decision rule would have been substantially higher.5
It may be possible to improve the cost-effectiveness of paint abatement directed at non-intact paint and
consequently improve the overall net benefits. Currently, the benefit-cost analysis assumes that all paint in a home
will receive high-end paint abatement if there is non-intact paint that triggers abatement. More selective abatement
of non-intact paint only could lower the costs of this abatement option without compromising the benefits
significantly. Net benefits would improve as a result.
Abt Associates, Inc. 6-19 Draft. January 10, 1994
-------
EihRiita-lt
SingU Medium Hanrd Lml: Out - 1200
•witMntnt
YM
Told
Soloctod
HP
IP
HS
18
RO
HPAS
IPIHS
ipas
NRD
BrnnlHi
llnilllonl
5.301
1.385
290
1.892
ws"
564
65
91
22.472
32.945
Coilt
llmilltonl
20.260
3,461
272
1.420
878
943
87
74
2.150
29.646
Mm* OatBulIlM
Mil BIMllw
Itmlllbiu)
114,9601
12.076)
19
472
(92)
0791
1221
17
20.322
3.299
of AbilMiinli
liooo-ii
4,188
2.731
74
411
276
114
18
16
7.777
16.601
Proportion
el
Abitominti
2&83%
17.50%
047%
2.63%
1.77%
0.73*
0.11%
0.10%
4934%
Abitominti
WIlhNogitln
NitBonoflta
IIOOO'i)
0
4.053
2.527
0
0
202
114
18
0
200
7.114
Mlnknam
hdMdMl
MM BiMf IH
in
0
0.690)
12.5571
670
13.528)
(8,5141
0.8011
1356
1130)
Mixknum
MMdnl
HilBiMllte
l«)
0
2.629
7.142
533
8.843
1.694
12468
11.689)
USB
13.126
SoD Unl
(ppm)
0
0
0
1.100
100
3.100
1.100
0
MMtaam
SoD Unl
(pprnl
3,100
2.200
3.100
1.100
5,800
1.200
asm
1.200
QUO
600
DmtUml
Ippml
0
0
0
3,300
1400
1400
1.100
1.100
1JOO
HlMtlillMl
Ippml
1.200
3.300
sjm
3,800
1.200
3JOO
1.100
Mmtanm
iHlMffalP •>•!•!
ftngiml
0
1
1
5
0
B
10
B
11
0
MMknni
(mBlMl
22
22
22
B
0
13
11
11
11
10
AeMtflfiunt CoaM
HP = High Print Abatement
LP • Low Print Abatement
H8 - High Sell Abatement
L8 . Lew Sell Abatement
RO - Recurrent Dual Abatement
HP/H8 - High Paint. High Sell Abatement!
HP/18 - High Print. Lew Sell Abatementi
LP/HS n Lew Print, High Sell Abatementi
LP/LS n Lew Paint. Lew Sell Abatement.
NR D B Nonrecurrent Dual Abatement
-------
Two-Media Hazard Levels
In this section, decision rules based upon hazard levels for two-media plus the paint
condition criterion are evaluated. Under these circumstances, households have less leeway to
optimize their abatement decisions than they did under the single-medium decision rules and
much less than they did under the voluntary optimum. To satisfy recommendations based upon
a two-media decision rule, such as the case where the soil hazard level is 2,300 ppm and the
paint hazard level is an XRF of 20, it is assumed that any homeowner whose home exceeds
either one of these thresholds would undertake the best abatement that makes it possible to go
below both thresholds, as well as to meet the paint condition recommendation. In Exhibits 6-12
and 6-13, the results for the best cases of the three different combinations (soil/dust, soil/paint,
dust/paint) are presented. The best cases are defined as the combinations having the highest net
benefits.
As was the case under the single-medium decision rule, defining a hazard level for dust
appears to be critical. The two combinations based upon a dust hazard level of 1200 ppm (soil
= 2300 ppm and dust = 1200 ppm; dust = 1200 ppm and paint = 20 (XRF)) have the highest
net benefits. As shown in Exhibit 6-12, the net benefits are over $3 billion. The two dust-based
decision rules would induce abatement in 15.7 to 16.2 million homes and generate $33.0 to
$34.9 billion in benefits. In contrast, the decision rule combining soil and paint hazard levels
(soil = 2,300 ppm and paint = 20 mg/cmj) has much lower benefits ($11 billion) and,
accordingly, negative net benefits (-$17.4 billion).
The two rules based upon dust are approximately as accurate as the single-medium dust
hazard level was in inducing abatements leading to positive net benefits. The percentage of
abatements having negative net benefits is about 47% for the soil/dust combination or the
dust/paint combination. As shown in Exhibit 6-13, the effectiveness of these two combinations
stems from their being able to exploit nonrecurrent dust abatement, which is not a feasible
alternative when the hazard levels are based upon soil and paint.
For all three combinations, the paint condition criterion creates a tremendous burden.
This outcome is illustrated for the soil and dust combination, which has the highest net benefits
among all of the two-media decision rules. Exhibit 6-14 paints a picture similar to the one
already shown for dust in Exhibit 6-11. Certain abatements never have positive net benefits (the
combination of high-end paint and low-end soil abatements and of low-end paint and high-end
soil abatements), one form of abatement always generates positive net benefits (high-end soil),
and some abatements generate positive net benefits for some homes and negative net benefits for
others (high-end paint,low-end paint, low-end soil, recurrent dust, low-end paint and low-end
soil, and nonrecurrent dust). Most notable among the latter are the homes where high-end paint
abatement has been induced. The overwhelming majority of these (97%) have negative net
benefits.
The biggest difference between the single-medium rule based upon dust and the two-
media rule based upon soil and dust is that the two-media rule induces nearly half of a million
homes to undertake low-end soil abatement that results in negative net benefits. This difference
can be construed as the price to pay for specifying a two-media rather than a single-medium
decision rule.
Abt Associates, Inc. 6-21 Draft, January 10, 1994
-------
Exhibit 6-12
Banoflt-Coit Rnult* for Two Madia HIM Condition DeoMon Rule*
(ppml
Img/etn'l
Abatement
ReuoiHiiended
(Eidualve of
TeaOng Coote)
Coote
11000*1
•DI HOOOal
34.920
Condition
2.300
20
Yes
11.249
31,903
3.017
24,346
121.3291
29.699
117.4401
1.200
20
14.992
33.009
29.991
3.017
24.346
132,422)
121.3291
16.197
7.683
8.169
7.2BB
16,702
7.216
Candidate hazard lavah examined ranged up to 3000 ppm for toil. 2000 ppm for dint, and 22 mg/em* for paint.
-------
Exhibit 6 • 13
Distribution of Abatamant Choice, for Two Madia Plua ConoWon Dadalon Rulaa
Condition
Ippml
2,300
2.300
•
Dint
Ippml
1.200
-
1.200
Paint
lmg/om'|
-
20
20
NonlntaMt Point
Abatamant
Yoi
Yai
Yai
HP
4.180
4.220
4.246
LP
2.672
2.755
2.770
Numbar of Homaa Abated by Abatamant Typa |1000a|
H8
74
0
74
L8
1.006
1.006
411
RO
270
0
270
HP/H8
0
0
0
HP/18
114
114
114
LP/H8
18
0
18
LP/L8
74
74
10
NRD
7.777
0
7.777
Total
10.197
8.109
16.702
Candidate hazard lovah examined ranged up to 3000 ppm for aoil. 2000 ppm for dual, and 22 mg/cm> for paint.
Abatamant Code.
HP - High Paint Abatamant
LP = Low Paint Abatamant
HS = High Soil Abatamant
LS = Low Soil Abatamant
RD = Recurrent Du.t Abatamant
HP/HS - High Paint. High Soil Abatement.
HP/IS = High Paint. Low Soil Abatement.
LP/HS = Low Paint. High Soil Abatement.
UP/IS • Low Paint. Low Soil Abatement.
NR 0 - Nonrecurrent Dual Abatamant
-------
Exhfclte 14
Two Modta Haurd Unit: Soil - 2100. Oust - 1200
Abottnont
In
Told
StatogT
Sobctod
HP
IP
MS
IS
RO
HP/IS
LPIHS
LP/LS
NRO
BoiMlHo
II rallUonil
5.301
1,360
290
3.724
885
564
65
260
22.472
34.920
Com
llmlllbu)
20.260
3.387
272
3.475
979
943
87
350
2.150
31.903
Not Bono! Hi
flminioml
114.960)
12.0271
18
249
1921
0791
1221
1911
20.322
1.017
IIOOO'i)
4.186
2.672
74
1.006
276
114
IB
74
7.777
16.196
Proportion
ol
AbitmontB
2534%
16.50%
0.46%
6.21%
1.71%
0.70%
0.11%
046%
48.02%
Abotomonn
Whh Mogothn
Hot Bonolho
liomroi
0
4.053
1469
0
469
202
114
IB
59
200
7.583
Minimum
IndMdnl
•olBonolho
W
0
0.6901
B.557I
S33
02281
0.529
18.514)
0.8011
14.0011
11301
Mixknom
todMdtMl
Hit MMiM
i»i
0
2.629
7.142
533
8.843
1.694
12.466)
11.6891
2J56
13.126
SoOtmol
Ippml
0
0
0
1.100
3,000
100
3.100
1.100
3.100
0
BollUwl
Ippml
2.200
2,200
1.100
1.200
8.800
1.200
8.600
600
Mmknn
DatlMtl
Ippml
0
0
0
1.100
1.400
1.100
2.500
1.100
DHltml
Ippml
1.200
5.800
3.300
5,800
1.200
3,300
1.200
5,900
Mmknam
htntn Mill
(mpjiml
0
1
1
S
0
6
10
5
10
0
Mixknam
hfnfat Pitnf
tagfem1)
22
22
22
5
11
13
11
11
11
10
Abatement Coda
HP - High Pclnl Abatement
IP . Low Mm AbMement
HS = High Sail Abatement
L6 = Low Soil Abatement
RD a Recurrent Out! Abatement
HP/HS • High Paint. High Soil Abalomonti
HP/16 • High Paint. Low Soil Abatements
LP/HS o Low Paint. High Soil Abatemenn
IP/18 n Low Paint. Low Soil Abatements
MR 0 «• Nonrocurrant Dual Abatement
-------
Three-Media Hazard Levels
Exhibit 6-15 shows the benefit-cost results for the combination of three-media hazard
levels generating the highest net benefits (soil = 2300 ppm; dust = 1200 ppm; paint = 20
mg/cm2). Adding one more dimension to the decision rule does not change the optimal hazard
levels for soil, paint, or dust observed under the single-medium and two-media decision rules.
It does however change the net benefits, lowering them slightly since homeowners have one
more constraint on their decisions. The net benefits for the 2300/1200/20 rule are positive ($2.7
billion) but slightly smaller than those of the single-medium and two-media decision rules. Of
the 16.3 million abatements induced by these hazard levels, about 47% lead to negative net
benefits. The major difference between the abatements induced by this decision rule and those
induced by the single-medium and the two-media decision rules is that more high-end paint,
more low-end soil, and more low-end paint/low-end soil abatements are conducted.
6.2.3 Other Decision Rules to Consider
Among the set of decision rules that specify hazard levels, a rule based upon a dust
hazard level of 1200 ppm and a paint condition criterion generates the highest net benefits ($3.3
billion). It is important to note that, while this decision rule generates the highest net benefits,
other soil, dust, and paint hazard levels can generate net benefits of lower but similar magnitude.
Exhibit 6-16 shows other top combinations of different hazard levels. They each generate net
benefits within $2.2 billion of the highest net benefits achievable under a decision rule based
upon dust equal to 1200 ppm. This expanded set of alternatives does not offer hazard levels for
dust that are lower nor are there any two-media or three-media decision rules that are more
stringent, with one exception. A two-media decision rule based upon a soil hazard level of
2200 ppm and a dust hazard level of 1200 generates net benefits of $1.4 billion.
These and other alternatives may also be contenders for an optimal set of soil and dust
hazard levels if factors not yet encompassed in the current analysis are weighed in EPA's
decisionmaking. One such possibility is to choose the set of hazard levels that keeps the risk
at an individual home of exceeding a blood lead concentration of IS ug/dl below 5%. This can
be achieved with a hazard level of 500 ppm for soil and one of 400 ppm for dust, in addition
to the paint condition criterion. As shown in Exhibit 6-17, these hazard levels have a net benefit
of minus $18.8 billion.
All of the above decision rules considered so far, including the voluntary optimum, are
predicated upon testing homes. With this information, abatement decisions can be based upon
the circumstances of the homes, which offers the possibility of finding the best abatement for
each home and as a result, increasing the net benefits of abatement. An altogether different
approach would be to apply a given type of abatement to all homes, without conducting any
prior tests. Clearly, under this approach, some homes would be abated when abatement was not
appropriate, resulting in negative net benefits, but this is also the case for all of the decision
rules besides the voluntary optimum. The central issue is whether these kinds of errors would
be offset by savings in testing costs. The results for the voluntary optimum suggest one means
for implementing such an approach in a way that increases overall net benefits.
Abt Associates. Inc. 6-25 Draft, January 10, 1994
-------
Exhibit 8-IS
Thni Midli Hizard Lowlo: Soil - 2300. Dint - 1200. Interior Point - 20
Yn
retil
Stotogy
Sibctid
HP
IP
H8
18
RD
HPIL8
WHS
IP/IS
NRD
Bimllti
ItmlWonil
5.360
1.365
290
3.724
885
564
65
260
22.472
34.984
Coin
(IniDlonol
20.555
3.430
272
3.475
970
943
87
350
2.150
32.248
Hit BIIMI M
UmllNonol
115.1961
12.0731
18
249
1921
0791
122)
Oil
20.322
2.718
ol Abolonranto
IIOOO-o)
4.246
2.712
74
1,006
276
114
18
74
7.777
16.297
Proportion
ol
Abittmonti
26.06%
16.64%
045%
6.17%
1.70%
0.70%
0.11%
0.46%
47.72%
WllhNogilrn
Not BomlHo
IIOOO'o)
4.114
2.508
0
469
202
114
18
58
200
7.683
Minimum
MMdnl
Mot BomlHo
1*1
0
0.5571
12.2281
13.5261
B.514I
13.9011
14.0011
11301
MoKimra
InHluliliioil
UWIVIuUII
HotBonolHo
1*1
0
2.629
7.142
533
8.843
1.694
12.466)
11.689)
23SB
13.126
Minimum
Sod Ural
to*)
0
0
0
1.100
100
3,100
1.100
3,100
0
Mixknnm
Sod Ural
Ippml
2JOO
1200
2.200
1.100
8.800
1.200
8,800
1.200
600
Boat 1ml
Ippa)
0
0
0
3.300
1.100
1,400
1.100
2,500
1.100
1.300
Dot Loral
Ippml
1,200
1300
5.800
3,600
1.200
3,300
1.200
Mlnknn
hrtorior Point
IHIJtnvi
0
1
1
5
0
6
10
5
10
0
kMwbrPiuit
(mihni
20
22
22
S
11
13
H
11
11
10
Abatement Cooea
HP = High PMnt Abatement
LP - Low Print Abatement
HS = High Soil AbMariMnt
L8 = Lew Soil Abatement
RD = Reeuiient Du>t Abatement
HP/HS n High Palm. High Soil Abatement
HP/LS - High Palm, Lew Sell Abatamenio
LP/HS o Lew Palm. High Soil Abatement
LP/LS » Low Paint, Lew Soil Abalamenla
NH D • Nonteeurrant Dual Abatement
-------
Exhibit 6-16
Near Contenders for Highest Net Benefits
Soil
(ppm)
—
—
—
3.000
2.900
2.800
2,700
2,600
2,500
2,400
2,300
3,000
3,000
2,900
2,900
2,800
2,800
2,700
2,700
2,600
2,600
2,500
2.500
2.400
2.400
2.300
2.300
—
—
—
—
3.000
2.900
2.800
2,700
2,600
2,500
2.400
2,300
2.900
2.800
2.700
Dust
(ppm)
1.200
1.200
1,200
1,200
1.200
1,200
1.200
1,200
1.200
1,200
1,200
1,200
1,200
1,200
1,200
1.200
1,200
1,200
1.200
1.200
1,200
1,200
1,200
1,200
1.200
1,200
1,200
1.300
1.400
1.300
1.300
1,300
1,300
1,300
1,300
1,300
1.300
1,300
1.300
1,400
1,400
1,400
Paint
(mg/cmx
—
21
20
—
—
—
—
—
—
—
—
21
20
21
20
21
20
21
20
21
20
21
20
21
20
21
20
—
—
21
20
—
—
—
—
—
—
—
—
—
—
—
Total
Benefits
(* millions
32,945
33,009
33.009
34,920
34,920
34.920
34.920
34,920
34,920
34,920
34.920
34,984
34,984
34,984
34,984
34,984
34,984
34,984
34,984
34,984
34,984
34,984
34,984
34,984
34,984
34,984
34,984
31,413
29,669
31,477
31.477
33,389
33,389
33,389
33,389
33,389
33.389
33.389
33,389
32,847
32,847
32.847
Total
Costs
($ millions)
29,646
29,991
29.991
31,903
31,903
31,903
31,903
31.903
31,903
31,903
31,903
32,248
32,248
32.248
32,248
32,248
32.248
32,248
32,248
32.248
32.248
32,248
32,248
32,248
32,248
32,248
32,248
29,128
27,557
29,473
29.473
31.385
31.385
31,385
31,385
31.385
31.385
31.385
31.385
30,917
30,917
30,917
Net Benefits
(* millions)
3,299
3.017
3,017
3,017
3,017
3,017
3,017
3,017
3,017
3.017
3,017
2,736
2.736
2.736
2,736
2,736
2.736
2.736
2.736
2.736
2,736
2.736
2,736
2,736
2,736
2.736
2.736
2.285
2.112
2,004
2,004
2,004
2,004
2,004
2,004
2,004
2,004
2,004
2,004
1,929
1,929
1.929
Total Number
of Abatement!
(1000's)
15,602
15,702
15,702
16,197
16,197
16,197
16,197
16.197
16,197
16,197
16,197
16,297
16.297
16,297
16,297
16,297
16.297
16.297
16.297
16.297
16.297
16.297
16.297
16,297
16.297
16,297
16,297
14,654
13,954
14,754
14.754
15,249
15.249
15,249
15,249
15.249
1 5.249
15,249
15,249
14,869
14,869
14.869
Abatements
With Negative
Net Benefits
(1000's)
7.114
7.215
7.215
7,583
7,583
7,583
7,583
7,583
7,583
7,583
7,583
7,684
7,684
7,684
7,684
7,684
7,684
7,684
7,684
7,684
7,684
7,684
7,684
7,684
7.684
7,684
7,684
7,031
6,803
7,132
7.132
7.500
7.500
7.500
7,500
7,500
7,500
7,500
7.500
7.272
7.272
7,272
-------
Exhibit 6-16
Near Contenders for Highest Net Benefits
Soil
(ppm)
2,600
2,500
2,400
2,300
—
—
3,000
3,000
3,000
2,900
2,900
2,800
2,800
2.700
2,700
2,600
2,600
2,500
2,500
2,400
2.400
2,300
2,300
2,900
2,900
2,800
2,800
2,700
2,700
2,600
2,600
2,500
2,500
2,400
2,400
2,300
2,300
—
3.000
3.000
2,900
2.800
Dust
-------
Exhibit 6-16
Near Contenders for Highest Net Benefits
Soil
(ppm)
2.700
2.600
2.500
2.400
2,300
2.200
—
—
3.000
—
2.800
2.800
2.700
2,700
2.600
2,600
2,500
2,500
2,400
2,400
2,300
2,300
Dust
(ppm)
1,500
1,500
1.500
1.500
1,500
1.200
.500
.500
,500
,600
,500
,500
.500
.500
.500
,500
,500
,500
,500
,500
,500
,500
D^IM*
rBini
(mg/cm1
—
—
—
—
—
—
21
20
—
—
21
20
21
20
21
20
21
20
21
20
21
20
Total
Benefits
(ft millions)
32.024
32,024
32.024
32,024
32,024
38,380
28.910
28.910
30.821
28,342
32,088
32,088
32,088
32,088
32.088
32,088
32,088
32,088
32.088
32.088
32,088
32,088
Total
Costs
(ft millions)
30,652
30,652
30.652
30,652
30,652
37,029
27,637
27,637
29,548
27,218
30,997
30,997
30,997
30,997
30,997
30,997
30,997
30,997
30,997
30.997
30,997
30,997
Net Benefits
(ft millions)
.372
.372
.372
.372
.372
.350
.273
.273
1.273
1.124
.090
.090
.090
,090
,090
,090
.090
,090
.090
1.090
1,090
1,090
Total Number
of Abatements
(1000's)
14.479
14,479
14,479
14,479
14,479
18,267
13,665
13,665
14,159
13,351
14.579
14.579
14.579
14,579
14,579
14,579
14,579
14,579
14,579
14,579
14,579
14.579
Abatements
With Negative
Net Benefits
(1000's)
7.231
7.231
7.231
7.231
7.231
9,653
6.862
6,862
7,231
6,762
7.331
7.331
7.331
7.331
7,331
7.331
7.331
7,331
7,331
7,331
7,331
7,331
-------
Exhibit 6-17
Alternative Hazard Levels Based Upon Soil
500. Dust = 400
Soil
(ppm)
500
500
500
500
500
500
500
500
500
500
500
500
500
500
500
500
500
500
500
500
500
500
500
Dust
(ppm)
400
400
400
400
400
400
400
400
400
400
400
400
400
400
400
400
400
400
400
400
400
400
400
Paint
(XRF)
—
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
0
Total
Benefits
(* millions)
66,688
66,752
66,752
66,711
66.711
66.711
66,711
66,71 1
66,711
66.711
66,737
66,770
66,770
66.816
67,228
67,438
67.573
67,537
67,623
67.705
68,186
68,993
73,234
Total
Costs
(» millions)
85.445
85,790
85.790
86.910
86,910
86.910
86,910
86,910
86,910
86,910
86,998
87,229
87,229
87,481
88,478
89.138
90,248
91,457
91,703
92,577
96,479
105,526
146,212
Net Benefits
(« millions)
(18.757)
(19,038)
(19,038)
(20,199)
(20.199)
(20,199)
(20,199)
(20.199)
(20,199)
(20.199)
(20,261)
(20,460)
(20,460)
(20,665)
(21,250)
(21,699)
(22,675)
(23,920)
(24,079)
(24,871)
(28,293)
(36,533)
(72.978)
Total Number
of Abatements
(1000's)
45.563
45,664
45,664
45,728
45,728
45,728
45,728
45,728
45.728
45.728
45,728
45,728
45,728
45,728
45,728
45,797
46,049
46,049
46,049
46,304
47,099
49,537
59,039
Abatements
With Negative
Net Benefits
(1000's)
21,584
21.685
21,685
21.749
21,749
21.749
21,749
21,749
21,749
21,749
21,749
21,749
21,749
21,749
21,749
21,818
22,070
22,070
22,191
22,446
23,241
25,679
36,283
-------
Most of the abatements chosen under the voluntary optimum are nonrecurrent dust
abatements (99%) (Exhibit 6-2). Since the voluntary optimum decision rule presumes testing
prior to the abatement choice, this new approach must impose additional abatements costing no
more than the avoided testing costs ($24 billion) to improve on the overall net benefits of the
voluntary optimum rule. Given average discounted nonrecurrent dust abatement costs of $267
and the finding that 62.2 million homeowners chose not to abate under the voluntary optimum,
the incremental cost of imposing nonrecurrent dust abatement on all homes is $16.6 billion.
Some of these homes would also accrue benefits (even though net benefits are negative). There
would also be a small loss in benefits from those achieved under the voluntary optimum since
the homes choosing low-end paint and low-end soil abatements are not given the choice under
this rule. A preliminary estimate suggests a net increase in benefits of about $2 billion.
Adding this estimate to the net costs avoided (equal to $24.2 billion in testing costs
avoided minus $16.6 billion in additional abatement costs), there is a net gain of $9.6 billion.
In other words, with testing costs, the voluntary optimum decision rule generates net benefits
of $10 billion while this hybrid rule, based strictly upon nonrecurrent dust abatement, would
generate net benefits of approximately $20 billion. These higher net benefits can be achieved
through this uniform approach even though the percentage of homes with negative net benefits
(58%) is higher than that of the voluntary optimum (0%) or those of the best of the other
decision rules (about 47%).
It is unclear, however, how reliable this particular abatement approach based upon
nonrecurrent dust abatement is since it relies exclusively on one, substantial cleaning of every
home involved. Nonetheless, the improvements this approach offers underscores the important
role that testing costs may play in the abatement choices initiated by hazard levels set under
Section 403. It also suggests that there may be ways to construct guidance to the public which
takes testing costs into account.
In sum, further investigation of factors not fully considered in this analysis may reveal
additional opportunities for creating decision rules to address outstanding EPA concerns, and
possibly, to create higher net benefits.
6.3 SUMMARY AND CONCLUSION
Exhibits 6-18 and 6-19 summarize the benefit-cost information presented above for all
five sets of decision rules. The final comparison of these results integrates information on the
testing costs necessary to implement each of these decision rules. This information was
presented in the discussion of the overall net benefits of the voluntary optimum decision rule but
not for the other decision rules. Since there is variation in the tests that are required for
different decision rules (testing soil or dust for lead content or testing paint for condition or lead
content), the testing costs that are prerequisite to abatement decisions also vary.
The inclusion of testing costs does not alter the ranking of nine decision rules considered
here. The voluntary optimum still has the highest net benefits ($10 billion). All subsequent
decision rules have negative net benefits. The top four among these are based upon a single
medium (dust=1200), two media (soil=2300/dust=1200, dust=1200/paint=20), and three
media (soil=2300/dust=12007 paint=20). The overall net benefits of any of these decision
Abt Associates, Inc. 6-31 Draft, January 10, 1994
-------
rules is approximately minus $21 billion. The three lowest-ranking decision rules all have in
common that they are based in some way upon a paint hazard level. The qualitative hazard level
based upon paint condition criterion ranks seventh, the two-media rule based upon soil and paint
ranks eighth, and the single-medium hazard level based upon paint alone ranks ninth. These
outcomes and the finding of significant negative net benefits associated with paint abatement
highlight the potentially significant influence of the assumptions made regarding the effectiveness
and cost of paint abatement.
Another notable difference between the top-ranked decision rule and the next four is the
number of homes abated. The decision rules based upon qualitative or quantitative hazard levels
induce no more than 16.3 million abatements, of which 7 to 8 million have negative net benefits.
The voluntary optimum leads to nearly three times as many abatements - more than 45 million
abatements. None entails negative net benefits. This finding raises the possibility that better
decision rules could be created that are both implementable, which is the advantage of the
decision rules based upon hazard levels, and that lead to positive and substantial net benefits,
which the voluntary optimum does. Since the information bases assumed for the first group and
for the voluntary optimum differ substantially, it appears that creating a better means for
conveying useful information to guide homeowners' abatement decisions could be a productive
route for improving upon the decision rules investigated here. This may be an important point
of departure for subsequent investigations.
A more immediate point of departure for investigation is to put the current analysis into
perspective by considering the sensitivity of the results to various critical assumptions. Some
conclusions from this analysis have been strongly influenced by certain parameters used in the
current analysis. For example, the influence of assumptions regarding the cost and extent of
abatement of non-intact paint on net benefits has been highlighted. Alternative assumptions
about these unit costs and those of other abatements could affect the net benefits and the number
of abatements for each decision rule. The discount rate applied to benefits and costs also
appears to be particularly influential. The sensitivity of the benefits and costs of different
decision rules to these and other parameters is investigated in the next chapter.
Abt Associates. Inc. 6-32 Draft, January 10. 1994
-------
Exhibits- 18
Benefit-Cost Results (or Five Alternative Decision Rides
1.
2.
3.
4.
5.
Decision Rules
Voluntary Optimum
Paint Condition Only
Single Medium Plus
Condition
2-Media Plus
Condition
3a.
3b.
3c.
4a.
4b.
4c.
3-Media Plus Condition
Soil
(ppm)
-
-
2,300
•
•
2,300
2,300
•
2,300
Dust
(ppm)
-
-
-
1,200
•
1,200
-
1,200
1,200
Paint
mg/cm1
-
-
-
•
20
-
20
20
20
Nonintact Paint
Abatement
Recommended
No
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Benefits
($ million)
48,190
7,319
11,186
32,945
7,383
34,920
1 1 ,249
33,009
34,984
Abatement
Costs
($ million)
13,896
24,668
28,345
29,646
25,014
31.903
28,689
29,992
32,248
Net Benefits
(Exclusive of
Testing Costs)
($ million)
34,294
(17,349)
(17.159)
3,299
(17.631)
3.017
(17.440)
3,017
2,736
Testing
Costs
($ million)
24,222
14,987
24,222
24,227
14.906
24,227
24,222
24,227
24,231
Net Benefits
(Including
Testing Costs)
($ million)
10.072
(32.336)
(41,381)
(20,928)
(32.537)
(21.210)
(41.662)
(21.210)
(21,495)
Total Number
of Abatements
(1000s)
45.165
7.063
8.069
16.602
7,164
16.197
8,170
16,702
16,297
Number of
Abatemanta with
Negative Nat
Banafita (lOOOi)
0
6.686
7,164
7.114
6,786
7,683
7.266
7.216
7,684
Candidate hazard levels examined ranged up to 3000 ppm for soil, 2000 ppm for dust, and 22 mg/cm1 for paint.
Voluntary Optimum:
Paint Condition Only:
Single Medium:
2-Media:
3-Media:
Each home selects ebatement (or no abatement) that has the highest net benefits
Abatement is recommended for homes with more than five square feet of lead-based paint in nonintaet condition, regardless of XRF level or net
benefits. Homeowners choose the paint ebatement method thet generates the highest net benefits.
Within the full range of individual soil, dust, end paint hazard levels that could be set as a threshold for action, with no constraints placed on the other
two medie, the levels specified in the table maximize net benefits.
Within the full range of individual soil, dual, and paint hazard level combinations that could be set as a threshold for action, with no restriction on the
other medium, the levels specified in the table maximize net benefite.
Within the full range of individual soil, dust, end paint hazard level combinations that could be set as a threshold for action, the levels specified in the
table maximizes net benefits.
-------
Exhibit 6-19
Distribution of Abatement Choices for Five Alternative Decision Rules
'•
Condition
2-Media Plus
Condition
3a.
3b.
3c.
4a.
4b.
4c.
(ppm)
-
2.300
•
•
2.300
2.300
•
2,300
(ppm)
-
-
1.200
•
1.200
•
1.200
1.200
(mg/cm*
-
•
-
20
20
20
20
Abatement
Recommended
Yee
Yes
Yes
HP
4.160
4.160
4,186
4.221
4.186
4.221
4.247
4,247
LP
2,774
2,716
2,731
2,814
2,672
2,756
2,770
2,712
Number of Homes Abated by Abatement Type (1000s)
HS
0
0
74
0
74
0
74
74
LS
0
1,006
411
0
1.006
1.006
411
1,006
RD
0
0
0
276
0
276
0
276
276
HP/HS
0
0
0
0
0
0
0
0
0
HP/LS
0
114
114
114
114
114
114
114
114
LP/HS
0
0
0
18
0
18
0
18
18
LP/LS
0
16
74
16
16
74
74
16
74
NRD
44.567
0
0
7.777
0
7.777
0
7.777
7,777
Total
46.166
7,063
8.069
16,602
7,164
16,197
8.170
16.702
i 6,297
Candidate hazard levels examined ranged up to 3000 ppm for soil, 2000 ppm for dust, and 22 mg/cm1 for paint.
Abatement Codes
HP = High Paint Abatement
LP = Low Paint Abatement
HS = High Soil Abatement
LS = Low Soil Abatement
RD = Recurrent Dust Abatement
Voluntary Optimum:
Paint Condition Only:
Single Medium:
2-Media:
3-Media:
HP/HS = High Paint, High Soil Abatements
HP/LS = High Paint. Low Soil Abatements
LP/HS = Low Paint. High Soil Abatements
LP/LS = Low Paint, Low Soil Abatements
NR D = Nonrecurrent Dust Abatement
Each home selects abatement (or no abatement) that has the highest net benefits
Abatement is recommended for homes with more than five square feet of lead-based paint in nonintaet condition, regardless of XRF level or net
benefits. Homeowners choose the paint abatement method that generates the highest net benefits.
Within the full range of individual soil, dust, and paint hazard levels that could be set as a threshold for action, with no constraints placed on the other
two media, the levels specified in the table maximize net benefits.
Within the full range of individual soil, dust, and paint hazard level combinations that could be set as a threshold for action, with no restriction on the
other medium, the levels specified in the table maximize net benefits.
Within the full range of individual soil, dust, and paint hazard level combinations that could be set as a threshold for action, the levels specified In the
table maximizes net benefits.
-------
7. SENSITIVITY ANALYSES
Any modelling to support environmental policy analysis depends on making judgments
in how to depict human behavior, environmental circumstances, and their interactions. These
judgments draw variously on the analysts' expertise, existing knowledge about the problem at
hand, the advice of relevant specialists, and conclusions drawn from available data and studies.
One objective in combining insights gleaned from these sources is to fill the substantial gaps in
current understanding in a way that provides a "reasonable" characterization of the problem at
hand. Still, what constitutes a reasonable characterization may not be the definitive
characterization. It may apply to a large number but not all circumstances in the real world.
Or, it could be based on a set of assumptions about which even the best-informed analysts would
disagree. More than likely, it will be subject to both conditions.
This chapter considers the influence of alternative assumptions on the benefits and cost
of lead abatement in residential settings. The results are presented as sensitivity analyses
because they represent alternative specifications to the model used to calculate benefits and costs
rather than radical reconfigurations of the modelling framework. The aim is to determine
whether changing a relatively small number of parameters but in a credible way can lead to
substantial changes in the policy-relevant outcomes.
The number of parameters considered in this chapter is limited to a few selected items.
In part, this selection is due to model refinement which has already taken place. Yesterday's
"sensitivity analyses" became today's model enhancements, which were formally incorporated
into the framework. The selection also stems from the desire to focus on parameters that had
a significant likelihood of influencing the policy-relevant outcomes. By no means are the
parameters examined here the ones with the greatest potential for changing these outcomes.
Other modelling assumptions have greater potential but they often represent more controversial
assumptions. Finally, the selection derives from a limit on the number of sensitivity analyses
that can be generated. The binding constraint is not so much a limit on the number of modelling
iterations that can be run since with faster computing power, immense numbers of modelling
runs can be undertaken with changes in assumptions engineered from one iteration to the next.
The real limit is on the ability to organize the results from a large number of iterations in a way
that lends itself to reliable quality assurance, interpretation, and presentation.
In the next section, a brief overview of potential sensitivity analyses is provided. In the
three subsequent sections, sensitivity analyses based upon alternative cost and discounting
assumptions are presented.
7.1 POTENTIAL SENSITIVITY ANALYSES
Two sets of parameters will be discussed. The first set includes parameters tied primarily
to cost calculations and the second, ones that are relevant to benefit calculations. This
classification masks what may be the ultimate factor determining whether they should be
evaluated. It is the effect of changing these parameters on the optimal hazard levels of different
decision rules and the net benefits of these different decision rules which is one of the most
Abt Associates, Inc. 7-1 Draft. January 10, 1994
-------
policy-relevant outcomes that can come from any sensitivity analyses. However, presenting the
sensitivity analyses in this manner may give clearer indications on where these assumptions fit
into the model.
For the cost calculations, two general sets of assumptions stand out as candidates for
sensitivity analyses. The first set encompasses assumptions regarding the unit costs of abatement
of soil, dust, or paint. For soil abatement, possible variations include different assumptions
regarding the nature of abatement, the level at which soil is considered hazardous, and the costs
of transportation and disposal. For dust abatement, unit costs currently depend on the type and
frequency of different levels of abatement. For paint abatement, costs are highly dependent on
the condition and amount of lead paint present, among other factors. Some of these variations
may be captured explicitly or implicitly in the range of unit cost estimates. The results in
previous chapters were based upon a set of "medium" cost estimates. This chapter considers
the impacts that using low and high unit cost estimates have on the benefits and costs of different
decision rules and their optimal hazard levels. The second set of assumptions relates to
assumptions about testing costs. The numbers of soil, dust, and paint samples and the numbers
of households tested are critical components of the unit and aggregate costs, respectively.
Furthermore, the costs per sample could have considerable regional variation.
The benefit calculations depend on a wide array of assumptions that could be subjected
to sensitivity analyses. Because there are so many potentially influential assumptions, it is
important to narrow the set of potential candidates. Determining which assumptions are the best
candidates for sensitivity analysis and how the given assumptions should be altered complicates
the task of conducting the sensitivity analysis. For example, the effectiveness and durability of
various forms of abatement have not been thoroughly quantified by studies to date but for the
purposes of calculating benefits it is necessary to quantify them explicitly. It is possible that
there are alternate, credible assumptions besides those currently made in this analysis that would
alter the policy-relevant outcomes.
Some of the possible variations in assumptions are substantial enough to constitute major
model revisions rather than sensitivity analyses. For example, the main analysis did not include
benefits for children who already exist at the beginning of the time period covered by the model.
Although benefits for these children can be estimated, incorporating the influence of these
benefits on abatements decisions is more difficult and will require a significant adjustment to the
model. The potential benefits to adults from avoided high blood pressure effects from exposure
to lead have not been included in the model either. If exposures for adults to lead contamination
from residential soil, dust, or paint are significant, these benefits could have a substantial impact
on policy-relevant outcomes of the analysis, including the timing, number, and types of
abatements and most important, the optimal hazard levels for different decision rules. Like the
benefits for existing children, changing the model to make abatement decisions responsive to
adult exposures will require a significant adjustments since abatement will no longer be
exclusively induced by births. A further obstacle to incorporating such benefits is that the
evidence supporting their validity is very limited.
Questions have also been raised about the variation in benefits that derives from
differences in exposures from lead contamination in bare versus covered soil. Exposure to the
former entails higher ingestion of lead than exposure to the latter, under otherwise similar
Abt Associates, Inc. 7-2 Draft. January 10, 1994
-------
circumstances, which means that bare soil could have a lower optimal hazard level than covered
soil. The magnitude of these differences is unknown and little information is available to
suggest a conclusive resolution of this issue. Dealing with a related set of assumptions - soil and
dust ingestion rates employed in the ffiUBK model - offers another target for sensitivity analyses
but setting a new range for these values to address the bare/covered soil issue is almost an ad
hoc exercise.
Because of the breadth of the potential changes implicit in varying these assumptions, to
date a conservative approach has been taken toward what will be presented here. As mentioned
above, the first sensitivity analysis considers low and high unit cost estimates. The second
sensitivity analysis focuses on changing one particular parameter of the analysis that has a large
potential for affecting policy-relevant outcomes but is also relatively easy to implement. This
parameter is the discount rate used to express the monetary value of future benefits and costs
in contemporary terms. A rate of 7% was used for the analysis presented in previous chapters.
An alternate approach, which has been used by the Agency in some other regulatory analyses,
involves a two-stage discounting procedure that employs both 3% and 7%. Results using this
procedure will be presented below. The third sensitivity analysis illustrates the kinds of effects
that the potential benefits to adults and existing children could have on the benefit-cost analysis.
For this exercise, the benefits estimated for newborn children were supplemented by estimates
for adults and existing children but the basis for abatement decisions - the imminent birth of a
child - remains the same.
7.2 ALTERNATIVE COST ASSUMPTIONS
The information on unit costs of abatement provided in Chapter 4 provides the basis for
a wide array of different cost assumptions. For the abatement of one single medium, three
different estimates were developed - a high one, a low one, and a medium one. Once the
abatement of more than one medium is considered, the number of alternative cost combinations
that could apply to a given home grows exponentially. With the addition of one more medium
to abate, the number of cost estimates increases from three to nine. With the addition of a third
medium, the number of cost estimates rises to twenty-seven. If information on the likely
frequency of these different cost combinations were available, these different cost combinations
could be integrated explicitly in the abatement decisions of the homeowners, such as by having
homeowners consider the expected costs of high-end paint and low-end soil abatement, rather
than only the medium estimates. However, information on the distributions of abatement costs
was considered too limited to allow approximating such distributions reliably.
One alternative to trying to characterize the distributions of cost estimates in their
entirety is to look at the impact that the extreme values have on the policy-relevant outcomes.
On this basis, at least it would be feasible to establish boundaries on the possible benefit-cost
results and the estimated optimal hazard levels. While the low and high unit cost estimates
presented in Chapter 4 are not the ultimate minimum and maximum values, they were
constructed to be representative of the lower and upper ranges of values observed in practice.
As such, the low and high unit values of Chapter 4 are probably appropriate for testing the
boundaries of the benefit-cost analysis.
Abt Associates. Inc. 7-3 Draft, January 10, 1994
-------
The sensitivity analyses for the high and low cost scenarios focuses on comparisons of
the five alternative decision rules targeted in Chapter 6. To facilitate these comparisons, tables
were constructed for each scenario analogous to the ones presenting the benefit-cost results and
the distribution of abatement choices in Chapter 6 (Exhibits 6-18 and 6-19). Exhibits 7-1 and
7-2 present these findings for the high cost scenario. Exhibits 7-3 and 7-4 present those for the
low cost scenario.
For the high cost scenario, the most important rinding may be that the optimal hazard
levels are exactly the same for most versions of the decision rules. As shown in Exhibit 7-1,
for dust the optimal hazard level is still 1,200 ppm in all cases (single medium, two-media,
three-media). Furthermore, the decision rules with the highest net benefits are still those which
involve setting a dust hazard level. For paint the optimal level is still 20 mg/cm2. For soil, the
optimal hazard level is still 2,300 ppm in the decision rules having the highest net benefits (three
media, two media when combined with a dust hazard level) but is 3,000 ppm for decision rules
which have markedly lower net benefits. In all decision rules except the voluntary optimum,
the net benefits are negative. The net benefits for the voluntary optimum decline by 18% but
the number of abatements declines more dramatically, by 27%, as shown in Exhibit 7-2. The
numbers of abatements induced by the other decision rules stay basically the same. There are
minor shirts away from the more expensive abatements, such as high-end paint/low-end soil, but
mainly the distributions of abatements change little if at all.
For the low cost scenario, the optimal levels for dust and paint again remain at the levels
generated by the main analysis (1,200 ppm for dust and 20 mg/cm2 for paint) but the optimal
hazard level for soil is 35% lower. As shown in Exhibit 7-3, the optimal hazard level for soil
in all relevant decision rules is 1,500 ppm. This appears to be the most significant difference
in the results for the induced rules. Although the net benefits are substantially larger for all
decision rules and positive (exclusive of testing costs) for most, the net benefits are still negative
for the decision rules that had negative net benefits in the main analysis (the paint condition rule,
the single medium soil and paint decision rules and the two-media rule combining soil and
paint). Notably the number of abatements having negative net benefits does not decline
appreciably. In the case of the single-medium dust rule, for example, the decline is only 10%.
The other significant differences between the results for the main analysis and this
sensitivity analysis are associated with the voluntary optimum. As shown in Exhibit 7-4, the
number of abatements jumps from 45 million to almost 78 million. As before, an overwhelming
majority of these are nonrecurrent dust abatements (95 %). There is however a slight adjustment
to the distribution with the addition of high-end soil abatements, when there had been none
before, and with a six-fold increase in the number of low-end soil and low-end paint abatements.
In conclusion, this bounding exercise indicates that findings regarding optimal dust and
paint hazard levels may not be affected by a better representation of the distribution of abatement
costs. Dramatic upward and downward revisions applied simultaneously to all abatements did
not change the optimal hazard levels for dust and paint. This does not however categorically
rule out revisions in cost estimates which could affect the optimal dust and paint hazard levels.
It is important to remember that the abatement choices are the result of comparing all abatement
choices allowed by a given rule. If cost estimates for abatements to one medium in particular
are significantly revised without changes to the estimated abatement costs for other media, the
Abt Associates. Inc. 7-4 Draft, January 10, 1994
-------
optimal hazard level for that medium is likely to change. Whether this is a rare possibility or
not is unknown. Given current information it seems unlikely to occur with the estimated dust
abatement costs, given their significantly lower magnitudes, but it could affect the estimated
paint abatement costs, which already exhibit a very broad range.
This sensitivity analysis has shown that the optimal soil hazard level could be susceptible
to changes in assumptions regarding the costs of abatement. While the optimal soil hazard level
held constant at 2,300 ppm for an upward revision in all abatement costs, it fell to 1,500 ppm
when all costs were lowered. Although it has already been shown in Chapter 6 that certain
small changes in the hazard level for a particular medium do not change net benefits
significantly, in this case there is a significant change. In the main analysis, the best decision
rule, given that the soil hazard level is fixed at 1,500 ppm, has negative net benefits (-$7
billion).1 When the soil hazard level is instead allowed to fluctuate, the best decision rule
involving soil is one having a soil hazard level of 2,300 ppm. This rule has positive net benefits
($3 billion). Consequently, it appears that the evidence for setting a single hazard level for soil
is not clearcut. Instead, a range from 1,500 to 2,300 ppm is supported by the model when
bounding cost assumptions are applied.
The best decision rule when the soil hazard level is fixed at 1,500 ppm and the paint condition applies is
a two-media rule where the dust hazard level is 1,200 ppm.
Abt Associates, Inc. 7-5 Draft, January 10, 1994
-------
Exhibit 7-1
Benefit-Cost Results for Five Alternative Decision Rules. High Cost Scenario
I-
3.
4.
5.
Single Medium Plus
Condition
2-Media Plus
Condition
3a.
3b.
3c.
4a.
4b.
4c.
3-Medla Plus Condition
Sod
(ppml
-
-
3.000
-
-
2.300
3.000
-
2.300
Oust
Ippml
-
•
-
1.200
•
1.200
•
1.200
1.200
Paint
Img/cm'l
-
-
-
-
20
-
20
20
20
Nonbitaet Paint
Abatement
Recommended
No
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Benefits
I* mllDonl
43,122
6.991
9.983
32.604
7.055
34.907
10.047
32.668
34.970
Abatement
Costs
ItmODonl
15.008
30.135
34.239
37,166
30.662
40,820
34,666
37.592
41,246
Net Benefits
(Exclusive of
Tasting Costs)
28.114
(23.144)
(24,256)
(4.562)
(23,607)
(5,913)
(24,618)
(4,926)
(6,276)
Total Number
of Abatements
(1000s)
33.098
7,063
7.760
16,602
7,164
16,197
7,850
16,702
16,297
Number of
Abatements with
Negative Net
6,726
7,196
7,648
6.827
8,017
7,295
7,649
Candidate hazard levels examined ranged up to 3000 ppm for soil. 2000 ppm for dust, and 22 mg/cm* for paint.
Voluntary Optimum: Each home selects abatement (or no abatement) that has the highest net benefits
Paint Condition Only:
Single Medium:
2-Media:
3-Media:
Abatement Is recommended for homes with more than five square feet of lead-based paint In nonlntact condition, regardless of XRF level or net
benefits. Homeowners choose the paint abatement method that generates the highest net benefit*.
Within the full range of individual soil, dust, and paint hazard levels that could be set as e threshold for action, with no constraints placed on the other
two media, the levels specified in the table maximize net benefits.
Within the full range of individual soil, dust, and paint hazard level combinations that could be set as a threshold for action, with no restriction on the
other medium, the levels specified in the table maximize net benefits.
Within the full range of Individual soil, dust, and paint hazard level combinations that could be set as a threshold for action, the levels specified In the
table maximizes net benefits.
-------
Exhibit 7-2
Distribution ol Abatement Choices for Five Alternative Decision Rules. High Cost Scenario
2.
3.
4.
5.
Voluntary Optimum
Paint Condition Only
Single Medium Plus
Condition
2-Media Plus
Condition
38.
3b.
3c.
4a.
4b.
4c.
3-Media Plus Condition
(ppml
-
3.000
-
-
2.300
3.000
-
2.300
Ippml
-
-
1.2OO
-
1.200
-
1.200
1.200
Paint
(mB/cm'l
•
-
-
20
-
20
20
20
Nonlntact Paint
Abatement
Recommended
No
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Number of Homes Abated by Abatement Type (1000s)
HP
0
4.249
4,160
4.275
4,310
4.186
4,221
4,336
4.247
LP
0
2.774
2,716
2.731
2.814
2.672
2,756
2.770
2.712
H8
0
0
0
0
0
0
0
0
0
LS
257
0
688
411
0
1,006
688
411
1.006
RD
0
0
0
350
0
360
0
350
350
HP/HS
0
0
0
0
0
0
0
0
0
HP/L8
0
24
114
24
24
114
114
24
114
LP/H8
0
0
0
18
0
18
0
18
LP/LS
0
16
74
16
16
74
74
16
NRD
32,840
0
0
7.777
0
7.777
0
7.777
Total
33.098
7.063
7.750
16.602
7,164
16,197
7,850
15.702
Candidate hazard levels examined ranged up to 3000 ppm for soil, 2000 ppm for dust, and 22 mg/cm' for paint.
Abatement Codes
HP = High Paint Abatement
LP = Low Paint Abatement
HS = High Soil Abatement
LS = Low Soil Abatement
RD = Recurrent Oust Abatement
HP/HS = High Paint, High Soil Abatements
HP/LS = High Paint, Low Soil Abatements
LP/HS = Low Paint. High Soil Abatements
LP/LS = Low Paint, Low Soil Abatements
NR D = Nonrecurrent Dust Abatement
Voluntary Optimum:
Paint Condition Only:
Single Medium:
2-Media:
3-Media:
Each home selects abatement (or no abatement) that has the highest net benefits
Abatement Is recommended for homes with more than five square feet of lead-based paint in nonintact condition, regardless of XRF level or net
benefits. Homeowners choose the paint abatement method that generates the highest net benefits.
Within the full range of individual soil, dust, and paint hazard levels that could be set as a threshold for action, with no constraints placed on the other
two madia, the levels specified in the table maximize net benefits.
Within the full range of Individual soil, dust, and paint hazard level combinations that could be set as a threshold for action, with no restriction on the
other medium, the levels specified In the table maximize net benefits.
Within the full range of Individual soil, dust, and paint hazard level comblnationa that could be set as a threshold for action, the levels specified In the
table maximizes net benefits.
-------
Exhibit 7-3
Beneflt-Coit Results for Five Alternative Decision Rube. Lew Cost Scenario
'•
2.
3.
4.
5.
Voluntary Optimum
Paint Condition Only
Condition
2-Madia Plus
Condition
3a.
3b.
3c.
4a.
4b.
4c.
3-Media Plus Condition
Soil
Ippml
-
•
1.600
-
-
1.500
1.500
-
1.500
Dust
Ippm)
-
-
•
1,200
-
1,200
•
1.200
1,200
Paint
Img/em*)
-
-
-
-
20
-
20
20
20
Nonbitact Paint
Abatement
Recommended
No
Yea
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Benefits
II million)
58,703
7.566
15,647
33.193
7.630
39,382
15.711
33,256
39,446
Abatement
Costa
(»mflUon|
13.950
18.972
24.778
21.836
19,235
26,824
25,041
22,099
27.087
Nat Benefits
(Exclusive of
Testing Costs)
ItmlOionl
44.753
(11.405)
(9.130)
11.357
(11.605)
12,558
(9.330)
11.167
12.368
Total Number
of Abatements
77,738
10,649
15.602
7.164
18.776
10.750
15.702
18.877
Number of
Negative M*t
0
6.394
6.463
6.429
6.529
6,494
6.664
Candidate hazard levels examined ranged up to 3000 ppm for soil, 2000 ppm for dust, and 22 mg/cm' for paint.
Voluntary Optimum: Each home selects abatement (or no abatement) that has the highest net benefits
Paint Condition Only:
Single Medium:
2-Medla:
3-Media:
Abatement is recommended for homes with more than five square feet of lead-based paint In nonlntact condition, regardless of XRF level or net
benefits. Homeowners choose the paint abatement method that generates the highest net benefits.
Within the full range of individual soil, dust, and paint hazard levels that could be set as a threshold for action, with no constraints placed on the other
two media, the levels specified In the table maximize net benefits.
Within the full range of individual soil, dust, and paint hazard level combinations that could be set as a threshold for action, with no restriction on the
other medium, the levels specified In the table maximize net benefits.
Within the full range of individual soil, dust, and paint hazard level combinations that could be set as a threshold for action, the levels specified In the
table maximizes net benefits.
-------
Exhibit 7-4
Distribution of Abatement Choices lor Five Alternative Decision Rules. Low Cost Scenario
1.
2.
3.
4.
5.
Decision Rules
Voluntary Optimum
Paint Condition Only
Single Medium Plus
Condition
2-Media Rus
Condition
3a.
3b.
3c.
4a.
4b.
4c.
3-Media Plus Condition
Sod
(ppml
-
1,500
-
-
1,500
1,500
-
1,500
Dust
(ppml
-
-
1,200
-
1,200
-
1,200
1,200
Paint
(mg/em'l
-
-
-
20
-
20
20
20
Nonlntact Paint
Abatement
Recommended
No
Yea
Yea
Yea
Yea
Yea
Yea
Yes
Yea
Number of Homes Abated by Abatement Type (1000s)
HP
0
4,120
4,120
4,146
4,181
4,146
4,181
4,207
4,207
LP
131
2,716
2,708
2,672
2,756
2,664
2,747
2,712
2,704
HS
163
0
163
74
0
237
163
74
237
LS
3.545
0
3,423
411
0
3,423
3,423
411
3.423
RD
0
0
0
276
0
276
0
276
276
HP/HS
0
27
27
27
27
27
27
27
27
HP/LS
0
126
126
126
126
126
126
126
126
LP/H8
0
0
0
18
0
18
0
18
18
LP/LS
0
74
82
74
74
82
82
74
82
NRD
73.899
0
0
7.777
0
7.777
0
7.777
7.777
Total
77.738
7,063
10,649
16.602
7.164
18.776
10.760
16.702
18.877
Candidate hazard levels examined ranged up to 3000 ppm for soil, 2000 ppm for dust, and 22 mg/cm* for paint.
Abatement Codes
HP = High Paint Abatement
LP = Low Paint Abatement
HS = High Soil Abatement
LS = Low Soil Abatement
RD = Recurrent Dust Abatement
HP/HS = High Paint, High Soil Abatements
HP/LS = High Paint, Low Soil Abatements
LP/HS = Low Paint, High Soil Abatements
LP/LS = Low Paint, Low Soil Abatements
NR D = Nonrecurrent Dust Abatement
Voluntary Optimum:
Paint Condition Only:
Single Medium:
2-Media:
3-Media:
Each home selects abatement (or no ebatement) that has the highest net benefits
Abatement la recommended for homes with more than five square feet of lead-baaed paint in nonlntact condition, regardless of XRF level or net
benefits. Homeowners choose the paint abatement method that generates the highest net benefits.
Within the full range of Individual soil, duat, and paint hazard levels that could be set as a threshold for action, with no constraints placed on the other
two media, the levels specified in the table maximize net benefits.
Within the full range of individual soil, dust, and paint hazard level combinations that could be set as a threshold for action, with no restriction on the
other medium, the levels specified in the table maximize net benefits.
Within the full range of Individual soil, dust, and paint hazard level combinations that could be set as a threshold for action, the levels specified in the
table maximizes net benefits.
-------
7.3 ALTERNATIVE DISCOUNTING PROCEDURE
The selection of a discount rate is one of the most debated features of benefit-cost
analyses of environmental policies. Fait of the concern driving the debate stems from a common
feature of environmental policy, which is that costs tend to be incurred in the present or near-
term while benefits accrue mostly in the future and sometimes in the distant future. Because of
the exponential nature of discounting, an increase in the discount rate leads to a greater than
proportional reduction in future benefits (and costs to the extent there are any) and therefore,
so the debate goes, discounting biases benefit-cost analysis against environmental regulations.
The debate also has a theoretical dimension to it. As Pearce et al. point out, there are two
reasons for discounting, which do not necessarily lead to the same rate (Pearce et al., 1989).
One reason is to reflect the social rate of time preference. The other reason is to reflect the
social opportunity cost of capital. Either a choice must be made between using one or the other,
or, some method of combining the two must be employed.
In the main analysis, a single discount rate of 796 was applied to both benefits and costs,
as specified in recent guidance from the Office of Management and Budget. As an alternative,
a two-stage discounting procedure is considered in this sensitivity analysis. This procedure uses
the approach of combining the time and capital elements to discounting cited above and was
issued as supplemental guidelines for regulatory impact analysis by EPA in 1989 (Scheraga,
1989). The first step of the procedure calls for annualizing capitalize costs using the marginal
rate of return on private investment and adding any operation and maintenance costs to the
annualized figures. A rate of 7% and periods of 10, 20, 30, 40, and 50 years for amortizing
(annualizing) the capital cost were used in this application. The second step entails discounting
both the benefit and cost streams using the consumption rate of interest. As recommended in
the guidelines, a rate of 3% was used. The two-stage procedure is appropriate under certain
circumstances. The costs of abatement have to be passed on to consumers, since displaced
consumption is the means for expressing the opportunity cost of the abatement. That is the case
for owner-occupants undertaking lead abatement, since they are also the consumers in question,
and is a strong possibility for renters but is dependent on market conditions.
Using the two-stage discounting procedure produces a substantial shift in the magnitude
of the estimated benefits and costs of abatement, resulting in much larger net benefits.2 (See
Appendix 7-A for abatement scenario unit costs using the two-stage discounting procedure.) The
best single indicator of the greater net benefits is that many of the estimated optimal hazard
levels for dust and soil are much more stringent than those estimated in the main analysis.
Exhibit 7-5 summarizes the range of optimal hazard levels estimated using different amortization
periods in the two-stage procedure.
This outcome derives in part from the performance of the two-step procedure when the benefit and
cost streams are very different over time (Scheraga, 1989) and from the fact that the benefits in any given year are
about four times larger when the damages stemming from IQ losses are discounted at 3% rather than 7%.
Abi Associates, Inc. 7-10 Draft, January 10, 1994
-------
Exhibit 7-5
Optimal Hazard Levels:
Results for Different Amortization Assumptions
Decision Rules
1.
2.
3.
Single Medium Plus
Condition
2-Media Plus
Condition
la.
Ib.
Ic.
2a.
2b.
2c.
3-Media Plus Condition
Soil
(ppm)
Min
300
-
-
1,000
700
-
1,000
Max
1000
-
-
2,200
1,000
-
2,200
Dust
(ppm)
Min
-
300
-
300
-
300
300
Max
-
400
-
400
-
400
400
Paint
(rag/cm2)
Min
-
-
4
-
4
20
20
Max
-
-
4
-
4
20
20
The estimated optimal hazard level for dust experiences the clearest change from the main
analysis. The highest level under the two-stage procedure is 400 ppm, as opposed to an
optimum of 1,200 ppm in the main analysis. The clear demarcation between the hazard levels
found using the two-stage procedure and those found using the conventional 7% approach
underscores how critical the decisions regarding the appropriate discount rate are. In this case,
unlike in the sensitivity analysis for costs, policy-relevant outcomes for dust are significantly
different between those of a model based upon one set of assumptions and those of a model
based upon a credible set of alternative assumptions.
The estimated hazard level for soil is much more dependent on the amortization
assumptions made.3 At the high end of the assumed range, meaning that costs are amortized
over fifty years, an optimal hazard level of 2,200 ppm for soil is estimated. At the low end,
where costs are amortized over ten years, an optimal hazard level of 300 ppm is estimated, at
least for a single medium decision rule for soil. The range of values is narrower once the
comparison is put on more equal footing. When the comparison is restricted to the decision rule
having the highest net benefits under each amortization assumption, the optimal hazard levels
range from 1,000 ppm to 2,200 ppm. Exhibits 7-6 to 7-15 provide the benefit-cost results and
The amortization period is the time period over which the initial capita] investment is recovered.
It matters so much because the two-stage procedure shares reflects the opportunity cost of capital that cannot be put
to other uses. The longer the time period over which the capital is amortized, the longer the capital is displaced.
Five iterations of the two-stage discounting procedure were estimated, based on amortization periods of 10, 20,30,
40, and SO years respectively. The iteration based upon SO years results in the highest present value of the
aggregate abatement costs and the iteration based upon 10 years, the lowest. Consequently the 50-year version leads
to the least stringent hazard levels and the 10-year one to the most stringent levels.
Abt Associates, Inc.
7-11
Draft, January 10. 1994
-------
abatement distributions for each of the five amortization analyses: 50,40, 30,20, and 10 years.
These show that the turning point, from an optimal hazard level for soil of 2,200 ppm to one
of 1,000 ppm occurs between the analysis based upon a thirty-year period and that based upon
a twenty-year period. This finding raises the question of whether a choice between the higher
and lower hazard levels can be justified by definitively establishing what the appropriate
amortization period is.
The appropriate amortization period depends on what is assumed about the economic
depreciation of the lead abatement. Viewed as a form of home improvement, the lead
abatement can add to the capital value of the home. The capital value reflects the present
discounted value of the stream of services that come from the lead abatement in the future.
When those future services no longer have value, the capital is fully depreciated.
Determining when the lead abatement no longer provides services of value is difficult.
At one extreme, all valuable services disappear when the home, because it is to be replaced,
is destroyed. The frequency of such an occurence can be estimated from the predicted
disappearance rate for homes. In this analysis, homes are assumed to have half-lives greater
than the longest amortization period, seeming to indicate that an amortization period of fifty
years should be used. Still, this example offers insufficient guidance since the value of
having a lead-abated home could be zero even if the home is still standing. For example, if
lead-free homes are in much greater relative abundance in the future, the added value of a
lead-abated home should disappear.4
As a result, even though it is reasonable to expect that lead-abated homes could
command higher prices in the near future, it is unknown when the housing market will no
longer place an added value on a lead-abated home. It does not appear possible to make a
definitive choice between the more stringent optimal hazard levels based upon twenty or
fewer years for amortization and the less stringent levels based upon thirty or more years of
amortization.3
As for other rules which do not estimate optimal hazard levels, the amortization
assumptions vary in their significance. They are mostly irrelevant for the voluntary
optimum. The two-stage discounting procedure is significant only because of the magnitude
of the estimated net benefits and the numbers and types of abatements. The net benefits
Even today in urban housing markets where there are relatively few homes with lead contamination
the market value of a lead-abated activities would be small.
As a last resort, it might appear that a case could be made that the relevant time period is no more
than twenty years based upon the length of payback periods of home improvement loans and of loans made to real
estate developers and landlords. Each of these have been asserted to be less than twenty years typically. However,
these norms are not directly linked to the value of the services from the home improvement, which can last longer
than the life of the loan. Even if in fifteen years the investment is fully recovered for the homeowner who initiated
the home improvement, and capital is no longer displaced, the subsequent owner may still pay a premium for the
home because of the improvement. Once again, whether this is the case depends on the value that the housing
market places on lead-abated homes at that time. Consequently, it does not seem that the choice between amortizing
over twenty years or thirty years can be settled by appealing to current loan practices.
Abt Associates, Inc. 7-12 Draft. January 10. 1994
-------
range from $364 to $409 billion, or approximately eleven to twelve times larger. At these
levels, it is not surprising that more expensive abatements that had not been chosen before
are now chosen, at least to a small degree. While more than 86% of the choices are still
nonrecurrent dust abatements, all abatement alternatives are chosen under the 20-year, 30-
year, and 40-year amortization analyses and all but one hi the other two analyses. Finally, it
should be noted that the two-staged procedure does not change the relative ranking of the
voluntary optimum among all the decision rules considered. Furthermore, in absolute terms,
the gap between the net benefits of the voluntary optimum and those of the other rules
actually grows by a factor of three.
For the paint condition rule, the amortization assumptions in the two-stage discounting
procedure matter since they determine the difference between circumstances, where the net
benefits are even more negative than under the main analysis (-$24 billion versus -$17
billion), as is the case when amortization takes place over fifty years, and circumstances
where the net benefits are positive for the first time ($0.6 billion), as is the case when an
amortization of ten years is assumed. These findings do not change the general conclusion
from the main analysis that other decision rules have a far stronger economic justification
than the paint condition rule does.
In conclusion, the two-stage discounting procedure has the greatest implications for
choosing hazard levels under Section 403 because it raises the possibility of a wider range of
potentially optimal hazard levels for dust and soil. The gap for the dust hazard level ranges
from 300 ppm to 1,200 ppm.6 For the soil hazard level, the gap ranges from 1,000 ppm to
2,300 ppm. The current analysis cannot categorically support the selection of one hazard
level for dust and soil from each of these ranges.
Several important things remain constant between the main analysis and this
sensitivity analysis. The voluntary optimum is far away the rule generating the highest net
benefits and the paint condition rule typically the least (and virtually always negative). The
types of decision rules that involve hazard levels for dust and soil and that have the highest
net benefits are generally the same under the two-stage procedure as they are in the main
analysis. They are the single-medium dust rule, the two-media rules based upon soil and
dust and upon dust and paint, and the three-media rule.
Although the amortization discussion focussed on its influence on the optimal soil hazard level,
it also matters to the question of which dust hazard level is optimal. A level of 400 ppm is optimal if amortization
periods of forty or fifty years are assumed, and 300 ppm otherwise.
Abt Associates, Inc. 7-13 Draft, January 10, 1994
-------
Exhibit 7-6
Benefit-Coot Results for Five Alternative Decision Rules. Two Stage Dboounting: 60 Years
2.
3.
4.
5.
Decision Rules
Voluntary Optimum
Paint Condition Only
Single Medium Plua
Condition
2-Media Plua
Condition
3a.
3b.
3e.
4a.
4b.
4c.
3-Madia Plua Condition
Soil
Ippml
-
-
1.000
-
-
2.200
1.000
-
2.200
Oust
(ppml
-
-
-
400
-
400
•
400
400
Paint
(mgfem'l
-
•
-
•
4
•
4
20
20
NonurtMrt Paint
AD 8V0IVI0 HI
nSOOIItlffWfMiB a
No
Yea
Yea
Yea
Yea
Yea
Yea
Yea
Yea
Benefits
l» millionl
501.319
46.748
119.001
462.385
123.802
462.385
179.780
462.766
462.766
Abatement
Costa
It minion)
136.957
71.216
113.688
205.427
134.544
205.427
168.169
206.359
206.359
N01 BSftAffitV
(Eiohnlva of
Testing Costal
I* mBHonl
364.362
124.4671
5.313
256.959
(10.743)
256.959
11.611
256.408
256.408
Total Number
of Abatementa
. HOOOa)
79.268
7.063
11.586
43.883
12.262
43.883
15.869
43.983
43.983
Number of
Benaflta dOOOal
0
5.667
5.779
8.684
8.034
8.684
8.034
8.784
8.784
Candidate hazard levels examined ranged up to 3000 ppm lor aoil. 2000 ppm for duet, and 22 mg/om1 for paint.
Voluntary Optimum:
Paint Condition Only:
Single Medium:
2-Media:
3-Media:
Each home selects abatement (or no abatement) that has the highest net benefits
Abatement is recommended for homes with more then five square feet of lead-baaed paint in nonlntact condition, regardless of mg/cm' level or net
benefits. Homeowners choose the paint abatement method that generates the highest net benefits.
Within the full range of individual aoil. dust, and paint hazard levels that could be set as a threshold for action, with no constraints placed on the other
two media, the levels apecified in the table maximize net benefits.
Within the full range of individual aoil. dust, and paint hazard level combinations that could be sst as a threshold for action, with no restriction on the
other medium, the levels apaoifiad in the table maximize net benefits.
Within the full range of individual aoil. duat. end paint hazard level combinations that oould be eat aa a threshold for action, the levels specified hi the
table maximizes net benefits.
-------
Exhibit 7-7
Distribution of Abatement Choices for Five Alternative Decision Rule*. Two Stage Discounting: 60 Yeare
1.
2.
3.
4.
5.
Deoteiofi Rules
Voluntary Optimum
Paint Condition Only
Single Medium Plus
Condition
2-Media Plus
Condition
3a.
3b.
3o.
4a.
4b.
4e.
3-Modia Plus Condition
SoH
Ippml
•
1.000
-
•
2.200
1.000
-
2.200
Dust
(ppm)
-
-
400
-
400
-
400
400
Paint
(mg/om'l
-
-
-
4
•
4
20
20
Nonifft AOI rBint
Abatement
No
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Number of Homes Abated by Abatement Type 11000s)
HP
367
4.137
4.137
4.463
7.049
4.463
7.049
4.524
4.524
LP
187
2.521
2.507
1.990
3.806
1.990
3.748
2.030
2.030
HS
1.808
0
921
1.584
0
1.584
780
1.584
1.584
IS
3.616
0
3.534
2.762
0
2.762
2.828
2.762
2.762
RD
2.744
0
0
5.681
0
5.681
0
5.681
5.681
HP/HS
21
161
228
375
512
375
512
375
375
HP/LS
0
126
126
803
486
803
486
803
803
IP/HS
8
36
36
27
92
27
92
27
27
IPOS
16
82
86
0
317
0
375
0
0
NRD
70.500
0
0
26.198
0
26.198
0
26.198
26.198
Total
78.268
7.063
11.586
43.883
12.262
43.883
15.869
43.983
43.983
Candidate hazard levels examined ranged up to 3000 ppm for soil. 2000 ppm for dust, and 22 mg/cm' for paint.
Abatement Codes
HP = High Paint Abatement
LP = Low Paint Abatement
HS = High Soil Abatement
LS " Low Soil Abatement
RD = Recurrent Dust Abatement
Voluntary Optimum:
Paint Condition Only.
Singla Medium:
2-Media:
3-Media:
HP/HS - High Paint. High Soil Abatements
HP/LS = High Paint. Low Soil Abatements
LP/HS o Low Paint. High Soil Abatements
LP/LS o Low Paint. Low Soil Abatements
NR D " Nonrecurrent Duet Abatement
Each home select* abatement (or no abatement) that has tha highest net benefits
Abatement m recommended for homes with more than five square feet of lead-based paint in nonlntoot condition, regardless of mg/om' level or not
benefits. Homeowners choose the paint ebetement method that generates the highest net benefits.
Within the full range of individual soil. dust, and paint hazard levels that could be set as a threshold for action, with no constraints placed on tha other
two media, the levels specified in the table maximize net benefits.
Within the full range of individual eoil, dust, and paint hazard level combinations that could be set as a threshold for action, with no restriction on tha
other medium, the levels specified in the table maximize net benefits.
Within the full range of individual eoil. dust, and paint hazard level combinations that could be set as a threshold for action, tha levels specified In the
table maximizes net benefits.
-------
Exhibit 7-8
Benefit-Co* Result! for Five AHernetive Decision Rule.. Two Stege Discounting: 40 Years
I-
2.
3.
4.
5.
Voluntary Optimum
Paint Condition Only
Single Medium Plus
Condition
2-Modia Plue
Condition
3b.
3c.
4e.
4b.
4o.
3-Media Plua Condition
(ppml
•
-
1.000
-
-
2.200
1.000
-
2.200
Ippml
•
-
-
400
-
400
-
400
400
Paint
(mg/om'l
•
•
-
•
4
-
4
20
20
Nofiintaot Paint
Abatement
No
Yee
Yea
Yes
Yes
Yee
Yes
Yes
Yes
Benefits
ItmOBon)
502.912
47.301
119.554
462.385
126.107
462.385
182.085
462.766
462.766
Abatement
Costs
It rnOHonl
132.648
66.892
108.363
196.613
127.849
196.613
160.678
197.480
197.480
Prai Benefite
(Exclusive of
Testing Costs)
l« mitten)
370.264
(19.5911
11.191
265.772
(1.742)
265.772
21.406
265.286
265.286
Total Number
of Abatements
(1000s)
79.307
7.063
11.586
43.883
12.262
43.883
16.869
43.983
43.983
Number off
UmmmtluM- ju^t
ntjgnive ram
0
5.740
8.628
7.929
8.628
7.929
8.728
Candidate hazard levels examined ranged up to 3000 ppm for soil. 2000 ppm for dust, and 22 mg/om> for paint.
Each home eelects abatement (or no abatement) that has the highest net benefits
Voluntary Optimum:
Paint Condition Only:
Single Medium:
2-Media:
3-Media:
Abatement is recommended for homes with more than five square feet of lead-based paint in nonintaot condition, regardless of mg/cm1 level or net
benefits. Homeowners choose the paint abatement method that generates the highest net benefits.
Within the full range of individual soil. dust, and paint hazard levels that could be set as a threshold for action, with no constraints placed on the other
two medie. the levels specified in the table maximize net benefits.
Within the full range of individual soil. dust, and paint hazard level combinations that eould be eat as a threshold for action, with no restriction on the
other medium, the levels specified in the table maximize net benefits.
Within the full range of individual soil. dust, and paint hazard level combinations that could be eat as a threshold for action, the levels specified in the
table maximizes net benefits.
-------
Exhibit 7-9
Distribution of Abatement Choices for Five Alternative Deoblon Rule.. Two Stage Discounting: 40 Year.
2.
4.
Condition
2-Media Plua
Condition
3a.
3b.
3o.
4a.
4b.
4c.
3-Medie Plua Condition
Ippml
2.200
1.000
2.200
Ippml
-
•
400
-
400
400
(mg/om'l
•
•
4
-
4
20
20
Nonintaot Psint
Abatement
RttOORini 8IM Ad
Yea
Yea
Yea
Yea
Yea
Yea
. Number of Homee Abated by Abatement Type HOOOel
HP
4.188
4.188
4.463
7.258
4.463
7.2S8
4.524
4.524
LP
165
2.470
2.457
1.990
3.596
1.990
3.538
2.030
2.030
H8
921
1.584
0
1.584
780
1.584
1.584
18
3.557
3.534
2.762
0
2.762
2.828
2.762
2.762
RD
0
5.681
0
5.681
0
5.681
5.681
HP/MS
237
375
550
375
550
375
HP/IS
126
126
803
486
803
486
803
LP/H8
8
27
27
54
27
54
27
LP/LS
82
96
0
317
0
375
0
NRD
70.441
0
26.198
0
26.198
0
26.198
Total
43.883
12.262
43.883
16.869
43.983
Candidate hazard levels examined ranged up to 3000 ppm for eoil. 2000 ppm for duet, and 22 mg/cm' for paint.
Abatement Codes
HP - High Paint Abatement
LP - Low Paint Abatement
HS - High Soil Abatement
LS - Low Soil Abatement
RD " Recurrent Oust Abatement
Voluntary Optimum:
Point Condition Only:
Single Medium:
2-Media:
3-Media:
HP/HS - High Paint. High Soil Abatements
HP/LS - High Paint. Low Soil Abatements
LP/HS - Low Paint. High Soil Abatements
LP/LS = Low Paint. Low Soil Abatements
NR D = Nonrecurrent Duet Abatement
Each home aelaeta abatement lor no abatement) that has the highest net benefit*
Abatement is recommended for homes with more than five square feet of lead-baaed paint in nonfntaot condition, regardless of mg/cm« level or net
benefits. Homeowners choose the paint abatement method that generates the highest net benefits.
Within the full renge of individual aoil. duat. and paint hazard levels that could be eat as a threshold for action, with no constraint, placed on the other
two media, the levels specified in the table maximize net benefite.
Within the full range of individual aoil. dust, and paint hazard level combinations that could be eat aa a threahold for action, with no restriction on the
other medium, the levels specified in the table maximize net benefite.
Within the full range of individual aoil. duat. and paint hazard level combinations that could be aet aa a threahold for action, the levels apecified in the
table maximizes net benefits.
-------
Exhibit 7-10
Benefit-Coat Result* for Five Alternative Decision Rule*. Two Stage Diaoounting: 30 Yeara
'•
2.
3.
Single Medium Plus
Condition
Condition
3b.
3e.
4b.
4c.
Ippml
•
-
•
2.200
•
Ippml
•
-
300
-
•
300
Img/om'l
-
-
•
4
•
4
20
NofiintBot Pfllnt
Abatement
Hfloo n i Hie no ttd
No
Yea
Yee
Yea
Yea
Yea
Yea
Yes
Yea
Benefite
I* milHonl
512.547
48.663
121.102
539.335
131.779
535.036
189.535
535.210
535.210
Abatement
Coeta
1* minion)
134.511
62.418
102.885
234.110
122.576
253.424
155.221
253.630
253.630
Net Benefite
lEiohnhra of
Tearing Coatal
(»miIBon)
378.036
113.755)
18.217
305.225
281.613
33.314
281.680
281.580
Total Number
of Abatementa
MOOOe)
78.406
11.686
51.978
12.282
51.978
16.869
Number of
Negative Net
9.052
7.684
9.944
Candidate hazard levels examined ranged up to 3000 ppm for aoil. 2000 ppm for duet, and 22 mo/cm* for paint.
Each home eelecte abatement (or no abatement) that haa the highest net benefits
Voluntary Optimum:
Paint Condition Only:
Single Medium:
2-Media:
3-Media:
Abatement ia recommended for homes with more than five square feet of lead-baaed paint in nonintaet condition, regardleaa of rng/em* lava) or net
benefits. Homeowners chooae the pnnt abatement method that ganeratea the Mghaat net benefits.
Within the full range of individual aoil. duat. and paint hazard lavela that could be eat aa a threshold for action, with no constraints placed on the other
two media, the levels specified in the teble maximize net benefits.
Within the full renge of individual aoil. duat. and paint hazard level combinationa that could be eat aa a threshold for action, with no restriction on the
other medium, the levels apacified in the teble maximize net benefita.
Within the full renge of individual aoil. duat. and paint hazard level combinationa that could be eet aa a threshold for action, the level* spsoified In the
table maximizes net benefits.
-------
Exhibit 7-11
Distribution ol Abatement Choices for Five Alternative Decision Rules. Two Stage Discounting: 30 Veers
I-
2.
3.
4.
5.
Voluntary Optimum
Paint Condition Only
Single Medium Plue
Condition
2-Medie Plue
Condition
3a.
3b.
3c.
4a.
4b.
4c.
3-Media Plus Condition
Ippm)
•
1.000
•
2.200
700
-
2.200
(ppml
-
•
300
•
300
-
300
300
Img/om'l
-
•
•
4
-
4
20
20
Nonintaot Paint
Abatement
No
Yes
Yes
Yes
Ye*
Yes
Yes
Yes
Yes
Number of Homes Abated by Abatement Type 11000s)
HP
1.532
4.290
4.290
8.082
7.684
5.647
7.684
5.748
5.748
LP
213
2.368
2.354
1.885
3.171
1.885
3.113
•1.88S
1.885
HS
2.319
0
921
5.827
0
5.548
780
5.548
6.548
LS
3.444
0
3.534
0
0
0
2.828
0
0
RD
1.684
0
0
4.154
0
5.046
0
4.945
4.945
HP/HS
112
170
251
1.952
550
1.775
608
1.775
1.775
HP/LS
172
185
185
0
730
0
730
0
0
LP/HS
8
27
27
27
54
27
54
27
27
IP/LS
18
24
24
0
74
0
74
0
0
NRD
69.907
0
0
32.050
0
32.050
0
32.050
Total
79.408
7.063
11.588
61.978
12.262
51.978
16.869
51.978
Candidate hazard levels examined ranged up to 3000 ppm for soil. 2000 ppm for dust, and 22 mg/cm» for paint.
Abatement Codes
HP - High Paint Abatement
LP •> Low Paint Abatement
HS m High Soil Abatement
LS ° Low Soil Abatement
RD = Recurrent Dust Abatement
Voluntary Optimum:
Paint Condition Only:
Single Medium:
2-Media:
3-Medie:
HP/HS = High Paint. High Soil Abatements
HP/LS = High Paint. Low Soil Abatements
LP/HS = Low Paint. High Soil Abatements
LP/LS = Low Paint. Low Soil Abatements
NR D B Nonrecurrent Dust Abatement
Each home seleete abatement lor no abatement) that has the highest net benefit*
Abatement is recommended for homes with more than five square feet of lead-based paint in nonintaot condition, regardless of mg/em1 Isvel or not
benefits. Homeowners choose the paint abatement method that generates the higheat net benefit*.
Within the full range of individual soil. dust, and paint hazard level* that could be set as a threshold lor action, with no cor
two medie. the levels specified in the table maximize net benefits.
nt* placed on the other
Within the full range of individual eoH. dust, end paint hazard level combinations that could be eet as a threshold for action, with no restriction on the
other medium, the levels specified in the table meximize net benefita.
Within the full range of individual soil. duct, and paint hazard level combination* that could be set a* a threshold for action, the levels specified in the
table maximizes nst benefite.
-------
Exhibit 7-12
Benefit-Co*! Result* for Five AHernetive Deohien Rule*. Two Stage Discounting: 20 Year*
1.
2.
3.
4.
5.
Voluntary Optimum
Paint Condition Only
Single Medium Hue
Condition
2-Media Plua
Condition
3a.
3b.
3c.
4a.
4b.
4o.
3-Madia Plua Condition
Soil
Ippml
-
-
300
-
-
1.000
700
-
1.000
Dual
Ippml
-
-
-
300
-
300
-
300
300
Paint
Img/om'l
-
-
-
-
4
•
4
20
20
NOfnfllBOt r Blflt
Abatement
Hftooin mono AQ
No
Yea
Yea
Yea
Yea
Yea
Yea
Yea
Yea
Benefit*
1$ million)
533.242
49.546
260.929
539.335
135.152
540.104
236.203
639.335
540.104
Abatement
Coate
I* million)
142.387
56.524
228.815
234.110
112.951
234.617
183.622
234.110
234.617
N01 B0IWltta9
(Exotuslva of
Teetmg Coat*)
((rniUonl
390.854
16.978)
32.114
305.225
22.200
305.487
62.581
305.225
305.487
Total Number
of Abatement*
(lOOOal
79.689
7.063
18.777
51.978
12.262
52.045
18.171
51.978
52.045
Number of
Abatement* with
Negative Not
Benefita HOOOa)
0
5.150
9.684
9.052
7.274
9.052
8.790
9.052
9.052
Candidate hazard levels examined ranged up to 3000 ppm lor *oil. 2000 ppm for duat. and 22 mg/cm1 for point.
Voluntary Optimum:
Paint Condition Only:
Single Medium:
2-Media:
3-Media:
Each home select* abatement lor no abatement) that has the highest net benefit*
Abatement is recommended for homes with more than five square feet of lead-baaed paint in nonintect condition, regardlese of mg/om* level or net
benefit*. Homeowners ohooae the paint abatement method that generates the highest net benefits,
Within the full renge of individual aoil. duet, and paint hazard laveb that could be eat as a threshold for action, with no eonatralnta placed on the other
two madia, the level* apecified in the table maximize net benefits.
Within the full renge of individual aoil. duat. and paint hazard level combination* that could be eat e* a threshold for action, with no restriction on the
other medium, the level* epecified in the table maximize net benefit*.
Within the full range of individual aoU. duet, and paint hazard level combination* that could be set a* a threshold for action, the level* epeoified in the
table maximize* net benefit*.
-------
Exhibit 7-13
Distribution ol Abatement Choices for Five Alternative DeoMon Rides. Two Stage Discounting: 20 Yeara
1.
2.
3.
4.
5.
Deobion Rule*
Voluntary Optimum
Paint Condition Only
Single Medium Plus
Condition
2-Medie Plus
Condition
3a.
3b.
3c.
4a.
4b.
4c.
3-Media Plue Condition
Soil
(ppml
•
300
-
-
1.000
700
•
1.000
Dint
Ippml
-
•
300
-
300
•
300
300
Paint
Imgfom')
•
-
-
4
-
4
20
20
Noniiitoot Pflint
Abatement
No
Yes
Yea
Yes
Yes
Yes
Yes
Yes
Yes
HP
2.198
4.317
3.612
6.082
7.795
6.082
7.649
6.082
6.082
Number of Homes Abated by Abatement Type (1000s)
LP
190
2.319
1.998
1.885
2.966
1.885
2.779
1.885
1.885
H8
4.670
0
9.198
5.827
0
5.895
5.433
6.827
5.895
L8
1.092
0
0
0
0
0
477
0
0
RD
1.387
0
0
4.154
0
4.154
0
4.154
4.154
HP/H8
169
205
3.745
1.952
656
1.952
894
1.952
1.952
HP/IB
188
188
0
0
783
0
783
0
0
IP/HS
8
27
223
27
54
27
150
27
27
LP/L8
0
8
0
0
8
0
8
0
0
NRD
69.786
0
0
32.050
0
32.050
0
32.060
32.060
Total
79.688
7.063
18.777
61.978
12.262
52.046
18.171
51.978
52.046
Candidate hazard levels examined ranged up to 3000 ppm tor aoil. 2000 ppm for dust, and 22 mg/cm' for paint.
Abatement Codes
HP - High Paint Abatement
LP = Low Point Abatement
HS - High Soil Abatement
LS <• Low Soil Abatement
RO - Recurrent Dust Abatement
Voluntary Optimum:
Paint Condition Only:
Single Medium:
2-Medie:
3-Media:
HP/HS i High Paint. High Soil Abatements
HP/LS - High Paint. Low Soil Abatements
LP/HS - Low Point. High Soil Abatements
LP/LS - Low Paint. Low Soil Abatements
NR D - Nonrecurrent Dust Abatement
Each home selects abatement (or no abatement) that has tha highest net benefite
Abatement is recommended for homee with more than five equara feet of lead-based paint in nonintact condition, regardless of mg/cm' (oval or net
benefits. Homeowners choose the paint abatement method that generates the highest net benefite.
Within the full range of individual soil. duet, and paint hazard levels that could be set as e threshold for action, with no constraints placed on tha other
two media, the levels specified in the table maximize net benefits.
Within the full range of individual soil. dust, and paint hazard level combinations that could be sst ae a threshold for action, with no restriction on the
other medium, the levele specified in the table maximize net benefits.
Within the full range of individual eoil. duet, and point hazard level combinations that could be set'as a threshold for action, tha levele specified in the
table maximizee net benefits.
-------
Exhibit 7-14
Benefit-Cost Result, for Five Alternative Decision Rules. Two Stage Dboounting: 10 Year.
2.
4.
*•
Point Condition Only
Condition
2-Modia Plus
Condition
3b.
3c.
4a.
4b.
Ippml
-
-
-
1.000
700
-
Ippml
•
300
-
300
-
(mg/om'|
•
-
4
-
4
Abatement
Yea
Yea
Yea
Yea
Yes
Yea
Yea
Benefit.
<» million!
t
558.772
52.477
263.342
543.290
140.161
544.058
238.581
543.290
544.058
Abatement
Coata
I* million)
150.119
51.843
200.086
210.539
103.225
210.978
162.501
210.978
N0i BWMfil8
(Exohisivo of
Testing Coatel
408.653
63.257
332.751
36.936
333.080
76.081
Total Number
Of AOBf 0in0fn9
(1000.1
80.842
7.063
18.777
51.978
12.262
52.046
18.171
51.978
Number of
Negative Nat
0
8.386
8.352
6.512
8.352
Candidate hazard level, examined ranged up to 3000 ppm for soil. 2000 ppm for dust, and 22 mg/cm' for paint.
Each home selects abatement (or no abatement) that hee the higheat net benefrte
Voluntary Optimum:
Point Condition Only:
Single Medium:
2-Madia:
3-Media:
Abatement is recommended for homes with more than five square feat of lead-based paint in nonintoot condition, regardless of mg/orn' lava) or net
benefit.. Homeowner, chooaa the point abatement method that generate, the highest net benefits.
Within the full range of individual soil. dust, and paint hazard levels that could be sot as a threshold for action, with no constraints placed on the other
two madia, the levela apecified in the teble maximize net benefit..
Within the fuH range of individual eoil. duet, and point hazard level combinations that could be set as a threshold for action, with no restriction on the
other medium, the levels specified in the table maximize net benefits.
Within ths full range of individual aoH. duet, and point hazard level combinatione that could be eet aa a threshold for action, the levels specified In the
table maxhnizee net benefits.
-------
Exhibit 7-16
Distribution of Abatement Choice* lor Five Alternative Deonion Ride*. Two Stege Discounting: 10 Years
2.
3.
4.
5.
Decision Rules
Voluntary Optimum
Paint Condition Only
Single Medium Plus
Condition
2-Media Plus
Condition
3a.
3b.
3c.
4a.
4b.
4c.
3-Media Plus Condition
Soil
Ippml
-
300
-
-
1.000
700
•
1.000
Dust
Ippml
•
-
300
-
300
•
300
300
Paint
(mg/om'l
•
-
•
4
-
4
20
20
NOIWItBOf rBint
Abatement
RaTMMinniAlwlarf
No
Yea
Yea
Yes
Yea
Yea
Yes
Yea
Yea
Number of Homes Abated by Abatement Type HOOOel
HP
2.735
4.227
3.682
6.645
7.665
6.645
7.629
6.645
6.645
IP
178
2.202
1.828
1.871
2.764
1.871
2.682
1.871
1.871
HS
6.038
0
8.012
5.827
0
5.885
5.488
6.827
5.895
L8
537
0
0
0
0
0
411
0
0
RD
952
0
0
3.264
0
3.264
0
3.264
3.264
HP/HS
652
388
3.958
2.279
876
2.279
1.011
2.279
2.279
HP/IS
657
188
0
0
783
0
783
0
0
LP/HS
8
48
186
41
76
41
158
41
41
IP/18
0
0
0
0
0
0
0
0
0
NRD
68.085
0
0
32.050
0
32.050
0
32.050
32.050
Total
80.842
7.063
18.777
61.978
12.262
52.046
18.171
61.978
52.045
Candidate hazard laveb examined ranged up to 3000 ppm lor soil. 2000 ppm lor dust, end 22 mg/om' lor paint.
Abatement Codes
HP - High Paint Abatement
LP - Low Paint Abatement
HS - High Soil Abatement
LS » Low Soil Abatement
RD = Recurrent Duet Abatement
Voluntary Optimum:
Paint Condition Only:
Single Medium:
2-Media:
HP/HS - High Paint. High Soil Abatements
HP/IS - High Paint. Low Soil Abatements
LP/HS - Low Paint. High Soil Abatements
LP/LS - Low Paint. Low Soil Abatements
NR D - Nonrecurrent Dust Abatement
Each home selects abatement (or no abatement) that has the highest net benefits
Abatement is recommended lor homes with more then live square leet ol lead-based paint in nonintaot condition, regardless ol mg/om* level or net
benefits. Homeowners choose the paint abatement method that generates the highest net benefits.
Within the lull range ol individual aoil. dust, and paint hazard levels that could be eat aa a threshold lor action, with no constraints placed on the other
two media, the levels specified in the table maximize net benefite.
Within the lull range ol individual soil. dust, end paint hazard level combinations that could be eet as a threshold lor action, with no restriction on the
other medium, the levels specified in the table maximize net benefite.
Within the lull range of individual aoil. dust, and paint hazard level combinationa that eould be eet as a threshold lor action, the levels specified in the
table maximizes net benefits.
-------
7.4 SUPPLEMENTAL BENEFITS FOR ADULTS AND EXISTING CHILDREN
The main analysis presented in this report was based upon a model developed to
consider the benefits to children by reducing lead exposure from the time of birth until the
age of seven years through the abatement of lead contamination. Not only was the risk
assessment focused on this population alone but the behavioral assumptions regarding lead
abatement were integrally linked to the impending births of children. This model structure
reflects the fact that this population has been considered a primary target for measures to
prevent residential exposures to lead-contaminated paint, dust, and soil. There are however
other populations that may benefit from reducing lead exposures. '
One population that has not been included in the main analysis consists of children
under the age of seven who are already present in homes during the first seven years of the
modeling time frame. Because the model is currently structured to consider only lead
abatements in anticipation of a new child being bom into homes, these "existing" children do
not serve as triggers for making an abatement decision nor are the benefits from any
abatement calculated for them. These children need special attention only in the first seven
years of the fifty years of analysis since in subsequent years all children younger than seven
years old were bom during the time period being modelled.
Little is known about the size of the benefits to children who have been exposed to
high lead levels during, for example, the first three-and-half years of their lives but avoid
these exposures in the remaining three-and-half years during which IQ development, in
particular, could have been inhibited. For the purposes of this sensitivity analysis, the
original assumption regarding the motivation for abatement - the upcoming birth of a child -
was maintained but the benefits from abatement for the child about to be born were
supplemented by an estimate of the benefits to the expected number of children under seven
already in the household. The benefits for these existing children were a prorated portion of
the benefits expected for a newborn. In the first year of the model, for example, the
representative existing child was assumed to be three-and-a-half years old and to gain one-
half the benefit that a newborn would from abatement.
Other populations for whom potential benefits of lead abatements are not currently
considered in the model are older children and adults. Adverse health effects are associated
with elevated blood lead levels in individuals over the age of seven, and there is a
considerable body of information on the dose-response relationships between blood lead
levels in adults and blood pressure. There is, however, scant information on the
relationships between lead levels in paint, soil and dust and blood lead levels for populations
over the age of seven (the ffiUBK model only considers children up to seven years of age).
It is therefore difficult to quantify the changes in blood lead levels and the coincident changes
in health status of adults associated with residential lead abatements. As a "placeholder" for
estimating the potential impacts of lead abatements, a method has been developed that draws
upon information provided in recent report by the Centers for Disease Control (CDC). The
CDC cites findings of Bornschein that the blood lead levels of pregnant women in homes
with and without lead-based paint differed by 2.13 pg/dL (Centers for Disease Control,
1991). (As noted in Chapter 3, this information was also used in the main analysis for the
computation of the incidence of neonatal mortality.) In this analysis, this difference is
Abt Associates, Inc. 7-24 Draft. January JO, 1994
-------
applied to the case of adults living in homes where lead-based paint is abated in connection
with an impending birth to calculate the reduced probabilities of hypertension-related effects.
Reduced hypertension-related mortality is valued using $2 million per statistical life saved.
Certain reduced morbidity effects (non-fatal stroke and coronary heart disease) are valued
using 32% of the value of a statistical life saved. Avoided hypertension is valued using
estimates of medical costs avoided. In sum, the per-abatement adult benefit is estimated as
$5,814. More in-depth details can be found in (Abt Associates, Inc., 1993).
The overall estimates that derive from supplementing the newborn child benefits in
the main analysis with benefits to existing children and adults should be viewed as illustrative
only, in light of the unrefined and somewhat arbitrary assumptions needed to generate the
supplemental benefit estimates. The benefit-cost results are presented in Exhibit 7-16 and the
distribution of abatements selected in Exhibit 7-17.
For all decision rules, the net benefits are higher in this sensitivity analysis than they
were under the main analysis of this report (Exhibits 6-18 and 6-19). This is mostly
attributable to the inclusion of the adult benefits rather than the inclusion of the benefits to
existing children. Furthermore, the net benefits are positive for all decision rules, a rare
finding seen in only other version of the analysis (the two-stage discounting example based
on an amortization period of 10 years). This outcome underscores how radical a departure
the addition of adult benefits is from the main analysis. In the case of the voluntary
optimum, the net benefits are $7 billion higher. For the remaining decision rules, the
changes are even more dramatic. The net benefits are $20 to $27 billion higher than they
were for these rules in the main analysis.
In two cases (the paint condition only rule and the single-medium rule based upon soil
plus non-intact paint abatement), the number of abatements induced did not change. The net
benefits of these abatements changed dramatically though because of the supplemental adult
benefits associated with paint abatement. The number of abatements under the voluntary
optimum increased by 9 million homes. In the remainder of the cases the numbers of
abatements rose by 5 to 28 million homes.
Although the change in benefits from the main analysis was only directly associated
with paint abatement, increases in the number of abatements were not restricted to paint
abatement. As shown in Exhibit 7-17, nonrecurrent dust abatement, abatements that included
high-end paint abatement, and low-end soil abatement showed the largest increases from the
results of the main analysis. Other forms of abatement in addition to paint abatement
increased in the optimum because formerly the houses needing paint abatement served as a
"roadblock" to going to lower soil or dust hazard levels in the optimum. These houses had
negative net benefits from abatement and lowered the aggregate net benefits in the main
analysis even though other houses, with the same soil and/or dust levels, had positive net
benefits from soil or dust abatement. In the aggregate, the latter were not large enough to
outweigh the negative net benefits of the former. Consequently, in the main analysis, higher
hazard levels had higher aggregate net benefits. Only once the benefits of paint abatement
were supplemented by the adult benefits was it possible for the aggregate net benefits of
more stringent soil and dust hazard levels to be the highest.
Abt Associates. Inc. 7-25 Draft. January 10. 1994
-------
This phenomenon helps to explain why the hazard levels for soil and dust could
decline when only the benefits for paint abatement were increased in this sensitivity analysis.
The dust hazard level exhibited the most consistent decline, from 1,200 ppm in the main
analysis to 400 ppm in all cases in this sensitivity analysis. This reduction in the optimal
dust level may have also been assisted by the possibility that lower dust levels can be
achieved through paint abatement. The fact that the optimal three-media decision rule
(1,400/400/20) and the optimal two-media decision rule involving dust and paint (--/400/20)
both induce large numbers of paint abatements while the paint hazard level remains high
seems to be consistent with this conclusion. These two decision rules and the other two that
involve a dust hazard level (the single-medium rule (-/400/~) and the two-media rule for
soil and dust (1,400/4007-)) have net benefits of approximately $29.5 billion, the highest
among the rules defining hazard levels.
In view of the uncertainties associated with the benefits for adults and existing
children in this particular sensitivity analysis, it is important to weigh its implications for
policymaking very carefully. Of the three sensitivity analyses presented in this chapter, this
one may rest on the weakest analytical footing. However, since the optimal hazard levels
identified in this analysis are consistent with the lower bounds of the optimal hazard levels
identified in the previous two sensitivity analyses, the burden of proof for the findings of this
particular sensitivity analysis is not high.
This analysis found an optimal hazard level for soil of 1,400 ppm. A level of 1,500
ppm was optimal in the low cost scenario while a level of 1,000 ppm was the lowest optimal
hazard level under the two-stage discounting procedure. The optimal hazard level for dust in
this sensitivity analysis was 400 ppm. The optimal level was higher in the high and low cost
scenarios (1,200 ppm) but the lower bound of the optimal hazard levels under the various
two-stage discounting scenarios was comparable (300 ppm). The experience with paint
hazard levels under this sensitivity analysis reproduced that of the two-stage procedure,
where the optimal hazard level was variously 4 or 20 mg/cm2, depending upon the decision
rule, but the highest net benefits were based upon 20 mg/cm2.
Taken together, the three sensitivity analyses presented in this chapter raise the
possibility of a wider range of potentially optimal hazard levels than the findings in the main
analysis imply. The optimal dust hazard level may be as low as 300 ppm or as high as
1,200 ppm. The main analysis found a dust hazard level of 1,200 ppm. The optimal soil
hazard level may be as low as 1,000 ppm or as high as 2,300 ppm. The main analysis found
a soil hazard level of 2,300 ppm. In the main analysis, and in these sensitivity analyses, the
highest net benefits for paint were associated with a hazard level of mg/cm2. Finally, a
qualitative hazard level based upon paint condition typically entails negative net benefits with
the exception of two cases: this particular sensitivity analysis which linked supplemental
benefits to paint abatement specifically and, the shortest-term amortization case (10 years)
under the two-stage discounting procedure. These two cases still seem rare enough to raise
doubts about the desirability of a paint condition criterion given the paint abatement options
constructed for this analysis.
Abt Associates. Inc. 7-26 Draft, January 10. 1994
-------
Exhibit 7-16
Beneflt-Coet Results for Flwa Alternative Decision Ride*. Adults and Existing Children
Decision Rule*
1.
2.
3.
4.
5.
Voluntary Optimum
Paint Condition Only
Singh Medium Phia
Condition
2-Modla Phit
Condition
3a.
3b.
3c.
4a.
4b.
4e.
3-Madia Phil Condition
Soil
(ppm)
-
-
2.300
.
.
1.400
2.300
.
1.400
Dual
(ppm)
-
-
-
400
.
400
.
400
400
Paint
mg/cm'
-
-
-
.
4
-
4
20
20
Nonintact Paint
AbfltBnwnt
noco rnrnonctod
No
Yaa
Yoa
YM
Yaa
Yaa
Yaa
Yaa
Yaa
Banafiti
(1 million)
100.116
37.015
44.081
123.754
70.910
123.754
72.853
1 24.247
124.247
Abatement
Costs
(I million)
58.720
34.838
41.192
94.262
62.053
94.262
63.473
94.749
94.749
Net Benaflta
(Exehialve of
Teatlng Coata)
It million)
41.396
2.177
2.890
29.492
8.858
29.492
9.380
29.498
29,498
Total Number
of AeMtwnonte
HOOOa)
54.558
7.063
8.069
43.883
12.262
43.883
1 2.673
43.983
43.983
Number of
AbfltwTMntc with
Negative Net
Benefit* (1000*1
0
4.241
4.710
11.168
6.733
11.168
5.733
11.168
11.168
Candidate hazard level* examined ranged up to 3000 ppm for *oil. 2000 ppm for du*t. and 22 mg/cm' for paint.
Voluntary Optimum:
Paint Condition Only:
Single Medium:
2-Madla:
3-Media:
Each home aalecta abatement (or no abatement) that ha* the highest net benefite
Abatement la recommended for home* with more then five equate feet of load-based paint in nonhrtaet condition, regardlm* of XRF level or net
benefit*. Homeowner* choose the paint abatement method that generate* the highest net benefit*.
Within the full range of individual toil. dust, and paint hazard level* that could be let a* • threshold for action, with no <
two media, the level* specified in the table maximize net benefit*.
•tralnt* placed on the ether
Within the full range of individual aoil. dint, and paint hazard level combination* that could be **t a* a threshold for action, with no restriction on the
other medium, the level* ipecified in the table maximize net benefit*.
Within the full range of Individual aoll. dint, and paint hazard level combinetione that could be eet a* a threshold for action, the level* specified In the
table maximizes net benefit*.
-------
Exhibit 7-17
Distribution of Abat«m«nt Cholera for Flw. Alternative Decision Riaee. Adults and Existing Children
1.
2.
3.
4.
b.
Decision Rulea
Voluntary Optimum
Paint Condition Only
Single Medium Plus
Condition
2-Media Plus
Condition
3a.
3b.
3c.
4e.
4b.
4e.
3-Media Plus Condition
Soil
Ippml
•
2.300
-
-
1.400
2.300
•
1.400
Duet
(ppm)
•
•
400
•
400
•
400
400
Paint
mg/cm'
•
•
•
4
-
4
20
20
Nonintact Paint
Abatement
R0CO IIIIIMIIUBU
No
Yes
Yea
Yea
Yea
Yea
Yea
Yea
Yea
HP
8.928
6.876
6.876
1 2.439
1 1 .479
1 2.439
1 1 .479
1 2.S40
1 2.540
Number of Homes Abated by Abatement Type (1000s)
LP
0
0
0
0
0
0
0
0
0
HS
0
0
0
967
0
967
0
967
967
LS
411
0
411
3.379
0
3.379
411
3.379
3.379
RD
0
0
0
879
0
879
0
879
879
HP/HS
0
0
0
233
0
233
0
233
233
HP/LS
314
188
783
1.269
783
1.269
783
1.289
1.269
LP/HS
0
0
0
0
0
0
0
0
0
LP/LS
0
0
0
0
0
0
0
0
0
NRD
44.906
0
0
24.716
0
24.716
0
24.716
24.716
Total
64.668
7.063
8.069
43.883
1 2.262
43.883
1 2.673
43.983
43.983
Candidate hazard levels examined ranged up to 3000 ppm for soil. 2000 ppm for dual, and 22 mg/cm' for paint.
Abatement Codee
HP - High Paint Abatement
LP - Low Paint Abatement
HS = High Soil Abatement
LS » Low Soli Abatement
RD = Recurrent Dual Abatement
Voluntary Optimum:
Paint Condition Only:
HP/HS - High Paint. High Soil Abatements
HP/LS - High Paint. Low Soil Abatements
LP/HS = Low Paint'. High Soil Abatements
LP/LS - Lew Paint. Low Soil Abatements
NR D • Nonrecurrent Dual Abatement
Each home selects abate
nt (or no abatement) that has the highest net benefits
Abatement is r<
mded for homes with more than five square feat of lead-baaed paint In nonlntaot condition, regardleaa of XRF (aval or net
benefits. Homeowners choose the paint abatement method that generates the highest net benefits.
2-Media:
3-Media:
i the full range of Individual soil. dust, and paint hazard levels that could be set as a threshold for action, with no constraints placed on the other
two madia, the levels specified In the table maximize net benefits.
Within the full range of individual aoil. dual, and paint hazard level combinations that could be set as a threshold for action, with no restriction on the
other medium, the levels specified in the table maximize net benefits.
Within the full range of individual aoil. duet, and paint hazard level combinations that could be set es a threshold for action, the levels specified In the
table maximizes net benefits.
-------
7.5
Abt Associates, Inc. 1993. Tide IV, Sections 402 and 404: Regulatory Impact Analysis.
Draft final report prepared for Nicolaas Bouwes, Regulatory Impacts Branch,
Economics, Exposure and Technology Division, Office of Pollution Prevention and
Toxics, U.S. Environmental Protection Agency, Washington, DC, October.
Centers for Disease Control, U.S. Department of Health and Human Services. 1991.
Strategic Plan for the Elimination of Childhood Lead Poisoning, February.
Pearce, D., A. Markandya, and E. B. Barbier. 1989. Blueprint for a Green Economy.
Earthscan Publications Ltd., London, p. 134.
Scheraga, J. 1989. Supplemental Guidelines on Discounting in the Preparation of
Regulatory Impact Analyses. Economic Studies Branch, Office of Policy,
Planning and Evaluation, U.S. Environmental Protection Agency, Washington,
DC, March.
Abt Associates, Inc. 7-29 Draft, January 10, 1994
-------
7.A APPENDIX
As discussed in Section 7.1, a two-stage discounting procedure was used as an alternative to
the straight seven percent discounting presented in Chapters 4-6. Exhibit 7.A-1 shows the
unit costs calculated for each of the ten abatement scenarios using amortization of twenty and
thirty years. Because the reflect the social costs of displacing capital, these unit costs are
higher than those calculated using a seven percent discount rate as shown in Exhibits 4-5 and
4-6.
Abt Associates. Inc. 7-30 Draft, January 10, 1994
-------
Exhibit 7.A-1
Summary of Abatement Strategy Unit Costs Using Two Stage Discounting
Abatement Scenario
High-end Paint Abatement
Low-end Paint Abatement
Non-Recurrent Dust
Recurrent Dust
Estimate Level
High
Medium
Low
High
Medium
Low
High
Medium
Low
Unit Cost by Amortization Period
10 Year
$15,485
$12,752
$10,020
$4,554
$3,340
$2,125
$1,457
$911
$364
20 Year
$17,905
$14,745
$11,585
$5,266
$3,862
$2,457
$1,685
$1,053
$421
30 Year
$20,139
$16,585
$13,031
$5,923
$4,344
$2,764
$1,895
$1,185
$474
40 Year
$22,106
$18,205
$14,304
$6,502
$4,768
$3,034
$2,081
$1,300
$520
50 Year
$23,771
$19,576
$15,381
$6,991
$5,127
$3,263
$2,237
$1,398
$559
Ten Year Monthly Total Coat P^timafo
High High $25,708
High Low $15,065
Low High $22,606
Low Low $11,964
Medium High $24,159
Medium Low $13,515
Abt Associates, Inc.
7-31
Draft, January 10, 1994
-------
Exhibit 7.A-1
Summary of Abatement Strategy Unit Costs Using Two Stage Discounting
Abatement Scenario
High-end Soil Abatement
Low-end Soil Abatement
Estimate Level
High cost
> 2000 ppm and exterior paint
£2000 ppm and exterior paint
>2000 ppm and no exterior paint
£2000 ppm and no exterior paint
Medium cost
> 2000 ppm and exterior paint
£2000 ppm and exterior paint
> 2000 ppm and no exterior paint
£2000 ppm and no exterior paint
Low cost
> 2000 ppm and exterior paint
£2000 ppm and exterior paint
>2000 ppm and no exterior paint
£2000 ppm and no exterior paint
Unit Cost by Amortization Period
High cost
$39,897
$24,715
$27,752
$12,570
Medium cost
$26,005
$15,786
$19,933
$9,714
Low cost
$15,757
$10,501
$12,114
$6,857
Hieh cost
$46,132
$28,578
$32,089
$14,535
Medium cost
$30,069
$18,253
$23,048
$11,232
Low cost
$18,220
$12,142
$14,007
$7,929
High cost
$51,888
$32,143
$36,092
$16,348
Medium cost
$33,821
$20,531
$25,923
$12,633
Low cost
$20,493
$13,657
$15,754
$8,918
Hithcost
$56,956
$35,283
$39,618
$17,945
Medium cost
$37,125
$22,536
$28.455
$13,867
Low cost
$22,495
$14,991
$17,293
$9,789
High cost
$61,245
$37,940
$42,601
$19,296
Medium cost
$39,920
$24,233
$30,598
$14,911
Low cost
$24,188
$16,119
$18,595
$10,526
High -$19, 178
Medium -$13, 3 10
Low - $7,442
Abt Associates, Inc.
7-32
Draft, January 10, 1994
-------
Exhibit 7.A-1
Summary of Abatement Strategy Unit Costs Using Two Stage Discounting
Abatement Scenario
High-end Paint and High-end Soil
High-end Paint and Low-end Soil
Estimate Level
High cost
>2000 ppm and exterior paint
£2000 ppm and exterior paint
> 2000 ppm and no exterior paint
£ 2000 ppm and no exterior paint
Medium cost
> 2000 ppm and exterior paint
£2000 ppm and exterior paint
> 2000 ppm and no exterior paint
£2000 ppm and no exterior paint
Low cost
> 2000 ppm and exterior paint
£2000 ppm and exterior paint
>2000 ppm and no exterior paint
£2000 ppm and no exterior paint
High
Medium
Low
Unit Cost by Amortization Period
Hieh cost
$55,382
$40,200
$43,237
$28,055
Medium cost
$38,757
$28,538
$32,685
$22,466
Low cost
$25,777
$20,521
$22,134
$16,877
$34,662
$26,062
$17,462
Hieh cost
$64,037
$46,483
$49,994
$32,440
Medium cost
$44,814
$32,998
$37,793
$25,977
Low cost
$29,806
$23,728
$25,593
$19,515
$37,083
$28,055
$19,027
High cost
$72,027
$52,282
$56,231
$36,487
Medium cost
$50,406
$37,116
$42,508
$29,218
Low cost
$33,524
$26,688
$28,785
$21,949
$39,317
$29,895
$20,473
Hieh cost
$79,062
$57,389
$61,724
$40,051
Medium cost
$55,330
$40,741
$46,660
$32,072
Low cost
$36,799
$29,295
$31,597
$24,093
$41,284
$31,515
$21,746
Hieh cost
$85,016
$61,711
$66,372
$43,067
Medium cost
$59,496
$43,809
$50,174
$34,487
Low cost
$39.569
$31,500
$33,967
$25,907
$42,949
$32,886
$22.823
Abt Associates, Inc.
7-33
Draft, January 10, 1994
-------
Exhibit 7.A-1
Summary of Abatement Strategy Unit Costs Using Two Stage Discounting
Abatement Scenario
Low-end Paint and High-end Soil
Low-end Paint and Low-end Soil
Estimate Level
High cost
>2000 ppm and exterior paint
£2000 ppm and exterior paint
> 2000 ppm and no exterior paint
2 2000 ppm and no exterior paint
Medium cost
> 2000 ppm and exterior paint
£2000 ppm and exterior paint
> 2000 ppm and no exterior paint
22000 ppm and no exterior paint
Low cost
> 2000 ppm and exterior paint
22000 ppm and exterior paint
>2000 ppm and no exterior paint
22000 ppm and no exterior paint
High
Medium
Low
Unit Cost by Amortization Period
High cost
$44,451
$29,269
$32,306
$17,124
Medium cost
$29,345
$19,126
$23,273
$13,054
Low cost
$17,882
$12,626
$14,239
$8,982
$23,732
$16,650
$9,567
Hidi cost
$51,398
$33,844
$37,355
$19,801
Medium cost
$33,931
$22,115
$26,910
$15,094
Low cost
$20,678
$14,600
$16,465
$10,387
$24,444
$17,172
$9,900
Hiiui cost
$57,810
$38,066
$42,015
$22,271
Medium cost
$38,165
$24,875
$30,267
$16,977
Low cost
$23,257
$16,421
$18,518
$11,682
$25,101
$17,654
$10,206
High cost
$63.458
$41,785
$46,120
$24.446
Medium cost
$41,893
$27.304
$33.223
$18.635
Low cost
$25,529
$18.025
$20,327
$12,823
$25,680
$18,078
$10,476
Hieh cost
$68,236
$44,931
$45,592
$26,287
Medium cost
$45.047
$29.360
$35,725
$20,038
Low cost
$27,451
$19.382
$21,858
$13,789
$26,169
$18,437
$10.705
Abt Associates, Inc.
7-34
Draft. January 10, 1994
-------
8. IMPACTS OF THE PROPOSED RULE
8.1 REGULATORY FLEXIBILITY ANALYSIS
8.1.1 Reason and Legal Basis for Agency Action
The Agency is identifying lead-based paint hazards, lead-contaminated soil and lead
contaminated dust under the Housing and Community Development Act which was enacted by
Congress on October 28, 1992 and contains 16 separate Titles. Title X of the Act is named the
Residential Lead-Based Paint Hazard Reduction Act of 1992 and is composed of five subtitles.
One of these, Subtitle B, amends the Toxic Substances Control Act (TSCA) by adding Title IV-
Lead Exposure Reduction. TSCA Title IV includes twelve sections, from §401 through §412.
Section 403, the subject of this analysis, reads as follows:
Sec. 403. Identification of Dangerous Levels of Lead
"Within 18 months after the enactment of this title, the Administrator shall
promulgate regulations which shall identify, for purposes of this title and the
Residential Lead-Based Paint Hazard Reduction Act of 1992, lead-based paint
hazards, lead-contaminated dust, and lead-contaminated soil."
Section 403 by itself requires only the identification of lead hazard levels and does not
require any specific action to abate residences whose contamination is above these levels. Thus
a formal regulatory flexibility analysis will not be done. Instead, this section will identify small
entities likely to be induced to abatement activity by the Agency's actions and discuss the
availability of data to quantify the effects. The approach taken is in keeping with the Regulatory
Flexibility Act charter which suggests:
"as a principle of regulatory issuance that agencies shall endeavor,
consistent with the objectives of the rule and of applicable statutes, to fit
regulatory and informational requirements to the scale of businesses,
organizations and governmental jurisdictions subject to regulation. To
achieve this principle, agencies are required to solicit and consider flexible
regulatory proposals and to explain the rationale for their actions to assure
that such proposals are given serious consideration."
The hazard levels the Agency sets will generate activity in two ways. First, voluntary abatement
is a likely response to the introduction of EPA these levels. This may impose a disproportionate
burden on lower income homeowners and renters and is addressed by comparing income to
abatement costs for individual homeowners as shown in Section 8.5. Second, the levels set in
Section 403 are used throughout Title X as the basis for determining appropriate responses to
the existence of lead-based hazards. The small business and government entities likely to be
affected by various provisions of Title X are discussed below.
Abt Associates, Inc. 8-1 Draft. January 10, 1994
-------
8.1.2 Definition of Small Entity and Affected Populations
The first step in the analysis is to define small entity and the affected populations in a
quantitative manner in order to determine the number affected. The definition should encompass
small firms, small non-profit organizations and small governments. The Section 403 regulation
pertains to home-owners and landlords; in these cases, the typical definition of small business
based on the number of employees is not relevant. However, the "EPA Guidelines for
Implementing the Regulatory Flexibility Act" allows a re-definition of small entity to suit the
regulatory circumstances if the Office of Advocacy of the Small Business Administration is
consulted and public comment on the proposed alternative definitions is obtained. For this
regulation we suggest the typical housing industry definition of a small landlord, one who owns
less than four units, as the appropriate small business definition. Definitions based on the
number of units are preferable to those based on revenue or number of employees because of
the variability of the latter two per housing unit. Small not-for-profit organizations could be
defined in the same manner as the small business. For small government jurisdictions, the
standard definition of a population less than 50,000 would apply.
Exhibit 8-1 summarizes the discussion in Chapter 1 of the Housing and Community
Development Act sections where the levels set in Section 403 are used. The final column
identifies the types of entities that may be affected. The sections of Title X that are most likely
to disproportionately burden small entities are Sections 1012 and 1018. The former requires
reduction of hazards in rehabilitation projects receiving less than $25,000 per unit in Federal
funds and abatement of hazards in those rehabilitation projects receiving more than $25,000 in
Federal funds per unit. These additional hazard reduction actions could prove a financial
burden. Section 1018 requires notification to the buyer of any property with a known lead-based
paint hazard and gives the buyer the right to perform an inspection before being obligated to a
contract for sale or lease. Small entities, housing owners, might be disproportionately affected
by a decrease in market value of a home if lead is found.
Abt Associates, Inc. 8-2 Draft, January 10, 1994
-------
EXHIBIT 8-1
Relationship of Identification of Lead Hazard Levels under § 403 with Other Provisions of the Lead-Based Paint Hazard Reduction Act
Section
Affected Housing Stock or Entity
Relationship
Small Entities Affected by (403
Hazard Levels
81011 (a)
Affordable non-public housing that ii not federally owned or assisted
housing
|403 Identification used to establish
eligibility for receiving HUD grants for
interim controls or abatement of lead-
based paint hazards.
No direct effects. Indirect effect is a
benefit for both large and small
landlords.
§1012
Various housing receiving assistance under the Cranston-Gonzalez
National Affordable Housing Act
I) |403 Identification used to require
reduction of hazards in course of
rehabilitation projects receiving less than
125,000 per unit in federal funds, and
abatement of hazards in rehabilitation
projects receiving more than $25,000 per
unit.
2) (403 Identification used to establish
eligibility for receiving federal funds for
interim controls or abatement of lead-
baaed paint hazards.
3) (403 Identification used to establish
eligibility for including inspection and
abatement costs in determining maximum
monthly rents in federally assisted rental
property.
Could add lead reduction or abatement
costs on top of rehabilitation coils. The
$25,000 ceiling is low enough that both
small and large landlords will be
affected. This might be a greater
financial burden to small landlords.
(1013
Federally owned housing being sold
1) Housing built prior to 1960:
Inspection and REQUIRED abatement of
lead-based paint hazard (as identified by
(403).
2) Housing built between i960 and 1978:
Inspection and written notification to
buyer of all lead-based paint hazards (as
identified by (403)
Federally owned housing primarily from
the Resolution Trust Corporation or
Veteran's Administration or Housing
and Urban Development. Federal
government is a large entity.
(1014
Low-income housing units under jurisdiction of Cranston-Gonzalez
National Affordable Housing Act.
(403 Identification used to estimate
number of housing units in a jurisdiction
occupied by low-income families that
have a lead-based paint hazard.
Information shall be used in preparing a
housing strategy .
Information dissemination only. No
significant direct or indirect effect on
small entities.
(1015
Private housing.
(403 Identification used by Inter-Agency
Task Force to recommend programs and
procedures for financing inspections and
abatements.
No direct effect on small entities.
Indirect effects could occur based on the
policies set.
Abt Associates, Inc.
8-3
Draft, January 10, 1994
-------
Section
51017
{1018
11021
TSCA Title IV. J402
§1021
TSCA Tille IV, (405
J1021
TSCA Title IV, 8406
11021
TSCA Title IV, §408
Affected Housing Slock or Entity
Federally supported inspections, risk assessments, interim controls and
abatements
Sale or lease of all housing stock constructed before 1978.
Persons offering to eliminate lead-based paint hazards.
Information on identifying and eliminating lead-based paint hazards.
Lead Hazard Information Pamphlet.
All executive, legislative and judicial branches of the federal
government having jurisdiction over property, or engaged in activities
that may result in a lead-based paint hazard
Relationship
§403 Identification used in Guidelines for
conducting federally supported lead-
based paint hazard reduction.
Requires notification to buyer of any
known lead-based paint hazards (as
identified by §403). Buyer has right to
perform inspection before being
obligated by contract for sale or lease.
Training and certification requirements
for all persons involved with identifying
and eliminating lead-baaed paint hazards
(as identified by §403).
1) Clearinghouse and hotline to provide
information on identifying, reducing and
eliminating lead-based paint hazards (as
identified by §403).
2) Establish protocols and performance
characteristics for products sold to
reduce or eliminate lead-bated paint
hazards (as identified by §403).
Required pamphlet to explain lead-based
paint hazards (at identified by §403) and
hazards.
All requirements in Lead-Based Paint
Hazard Reduction Act of 1992 shall
apply to all federal departments, agencies
and instrumentalities.
Small Entities Affected by §403
Hazard Levels
No significant direct or indirect effects
for small entities.
Small landlords could be burdened if the
market value of house decreased due to
the presence of lead; however if die
market reflects the higher value of a
lead free home once the market adjusts
there would be no differential burden
for small landlords.
These effects are covered under Section
402 Regulatory Flexibility Act.
No effect directly due to Section 403.
Information dissemination should have
no direct effect on small businesses.
i mo 11 covered under the Section 406
analysiB.
effect on small government!.
Abt Associates, Inc.
8-4
Draft. January 10, 1994
-------
8.1.3 Data Availability
Data on housing ownership and financial viability of landlords is available in proportion
to the amount of Federal assistance given. In general, the four levels of housing ordered by
decreasing amount of Federal assistance are: Federally owned, local public housing authority
owned, privately owned with public assistance, and privately owned. The information available
on each of these levels is discussed below. Information on Federally owned housing such as
insured properties acquired through default by the Department of Housing and Urban
Development (HUD), Resolution Trust Corporation, Veteran's Administration and Department
of Defense is available. However, since the Federal government is a large entity, the burden
of the Title X requirements will not be considered here. Public housing is owned by over 3,400
local public housing authorities ranging in size from less than 100 units to over 9,000 units. The
definition of small entity in this regard may be determined by the financial viability of the
authority. Data is available on the size of public housing authorities by units, revenue and
number of employees. Privately owned and Federally assisted housing consists of multifamily
rental properties under Sections 236 and 221(d)(3), project-based assistance for multifamily
rentals under the Section 8 Program and Fanners Home Administration owner-occupied and
rental programs under Sections 502, 504 and 515. The total number of units in HUD-based
multifamily programs is perhaps 1.4 million. HUD can identify inventory and age but has little
information on ownership except whether the owner is a non-profit or profit entity. Project
income expense data is maintained by HUD for some programs. Title X makes testing and
abatement an eligible expense for locally administered programs for owner and rental
rehabilitation; however, the number of rehabilitations is not known. The final housing
category, private market rentals and sales are covered by two provisions of Title X. Overall,
quantitative data on privately owned housing is unavailable. Large cities usually have property
directories that list owners; however, they are not in a standard format and accessing the data
could be difficult. In cities with rent control, such as Los Angeles and the District of Columbia,
the city directories could be searched by landlord to locate all small landlords. Unfortunately,
this information is not compiled and would be resource-intensive to collect.
8.1.4 Regulatory Options
Classic options available to protect small business are possible under Section 403.
Because Section 403 is a TSCA regulation, benefit-cost considerations are integral in setting the
regulatory levels. Exemption of certain classes of small landlords from the provisions or relaxing
the standards for small entities are two possibilities. Consideration of these options must include
their effect on the overall goal of Section 403, which is to set health-based hazard levels and to
the intent of the law, which is to protect the public health.
8.2 PAPERWORK REDUCTION ACT ANALYSIS
Setting of the lead hazard levels in paint, soil and dust does not, of itself, generate
additional reporting or record keeping requirements. Thus there is no increase in burden or
costs requiring analysis under the Paperwork Reduction Act. Regulatory requirements by other
Title IV sections (including sections 402 and 406) that rely on the hazard levels set under Section
403 rules are evaluated separately in other regulatory impact analyses.
Abl Associates. Inc. 8-5 Draft, January 10, 1994
-------
8.3 TRADE IMPACTS ANALYSIS
Section 403 of TSCA Title IV identifies lead hazard levels in paint, soil and dust for the
puiposes of Title IV and the Residential Lead-Based Paint Hazard Reduction Act of 1992.
The regulation requires certain abatement and notification activities to take place at these levels
• in subsets of domestic housing as was outlined in Section 1.2 Exhibit 1.1. All the entities
covered are domestic and there is no anticipated direct effect on international trade. Any
indirect effects are expected to be negligible. Abatements are expected in housing other than
those listed hi Exhibit 1.1 too. But these are all voluntary and more importantly domestic.
8.4 ANALYSIS OF IMPACTS ON TECHNOLOGICAL INNOVATION
No analysis of technological innovation was attempted at this time. However, on
examining the regulatory requirements, actions induced by the hazard levels that are set under
the regulation are likely to encourage innovation. Since the hazard levels may be used by states
and localities as required levels of abatement, a larger lead testing and abatement market is likely
to be created. This could then lead to greater competition and development of innovative testing
or abatement methods.
By setting numerical hazard levels, the type of innovation is limited to methods that allow
abatement to below the standard and testing methods that can quantitatively detect levels down
to the identified hazard level.
8.5 EQUITY IMPACTS ANALYSIS
The Agency is concerned about whether there are disproportionate burdens on particular
categories of households or individuals as a result of its actions. Although the Section 403
hazard levels in lead-based paint, soil and dust have not yet been established this analysis
investigates the equity impacts of four potential hazard levels. The hazard levels are:
• XRF ^ 6 mg/cm2 (maximum X-ray fluorescence reading in the housing unit).
• Both XRF ^ 6 mg/cm2 AND at least 10 per cent of the painted surface area is
damaged.
• Dust lead concentration £ 500 ppm.
• Soil lead concentration ^ 500 ppm.
Lead hazards as defined above can occur in virtually all segments of the American housing
stock. However, even though lead-based paint hazards are widespread throughout the United
States, and affect every socioeconomic group, the distribution of the hazards is not uniform with
respect to region of the country, age of housing stock, race or household income. Lead hazards
are more common hi older, low-cost housing units in the North-East and Mid-West than in other
units. Because these housing units tend to be occupied by households at or below the poverty
level, including a disproportionate share of African-Americans, these sub-populations are
potentially exposed to relatively more risks than other sub-populations. To the extent that
Abt Associates, Inc. 8-6 Draft, January 10, 1994
-------
Section 403 results in more abatement of hazards in housing (including paint, dust and soil
abatement), the segments of the U.S. population that are disproportionately exposed to the
hazards are likely to receive a larger share of the risk reductions. However, because most of
the abatements covered by Title IV are voluntary, relatively wealthier households are more likely
to proceed with the risk-reducing abatements. This section of the report describes the
distribution of lead-based paint hazards in the housing stock, and considers the environmental
equity implications of that distribution.
The distributions, hi the U.S. housing stock, of the four potential hazard conditions listed
above are estimated using data from the national survey of lead-based paint in housing sponsored
by the U.S. Department of Housing and Urban Development (HUD). The HUD survey was a
national stratified sample of 284 privately owned, occupied housing units built before 1980.
HUD developed weights for each observation, to create a weighted national sample representing
the 77.1 million privately owned and occupied housing units. The analysis in this section is
based on the HUD estimates of the national pre-1980 housing stock. These estimates are shown
in Exhibit 8-2.
EXHIBIT 8-2
National Prevalence of Lead-Based Paint Hazards
National Prevalence
in Pre-1980 Housing Stock
XRF £ 6
11%
XRF £ 6
and
> 10% Bad
Condition
4%
Dust
Concentration
S: 500 ppm
34%
Soil
Concentration
^ 500 ppm
11%
8.5.1 Age of Housing Stock
In general, lead-based paint hazards are more common in older housing stock. Even
though lead was not banned in household paint until 1978, the lead content of paint declined
after World War EL This is reflected in the distribution of the age of the housing stock that has
a lead-based paint hazard as shown in Exhibit 8-3. For example, 61 percent of the housing
stock with a maximum XRF reading of 6 mg/cm2 or more was built before 1930, even though
only 21 percent of the existing stock of pre-1980 housing units is that old. Shading indicates
a disproportionate prevalence of a potential hazard (i.e., the actual prevalence of a hazard in a
sub-group is at least 5 percent more than the sub-groups' share of the total national housing
stock). It is important to realize that even though the older units are more likely to have a paint
hazard, hazards do exist in some housing units of all ages.
Abt Associates, Inc.
8-7
Draft, January 10, 1994
-------
EXHIBIT 8-3
Distribution of Age of Housing Unit and Lead-Based Faint Conditions
Year Built
Pro 1930
1930 - '49
1950 - '65
1966 - '78
Total
Overall
Housing
Stock
Distribution
21%
13%
43%
22%
700%
XRF ;> 6
^•«*:;Si
14%
25%
-
700%
XRF Si 6
and
> 10% Bad
Condition
i^alB&S!
14%
47%
-
700%
Dust
Concentration
£: 500 ppm
$$$&»• .;!O%P'V'.' '.. ":
16%
23%
19%
700%
Soil
Concentration
^ 500 ppm
•&:£ 73i%1'" '"
7%
20%
-
700%
8.5.2 Regional Distribution
Because the North-East and Mid-West regions of the country tend to have relatively more
older housing units than the South and West, these regions would be expected to have more paint
hazards than regions with newer housing stock. However, the hazards are even greater than
would be expected if only housing age was considered. For example, of the national housing
stock with maximum XRF readings of 6 mg/cm2 or more, 57 percent occur in the North-East
region. Further, a total of 77 percent occur in the combined North-East and Mid-West regions;
whereas these two regions have only 47 percent of the total pre-1980 housing stock. In contrast,
the HUD survey found that in the West, which has 19 percent of the total housing stock, high
XRF readings occur in only 4% of all housing. Exhibit 8-4 shows the regional distribution of
the four lead-based paint hazards.
Abt Associates, Inc.
8-8
Draft. January 10, 1994
-------
EXHIBIT 8-4
Regional Distribution of Lead-Based Faint Conditions
Region
MidWest
Noith-East
South
West
Total
Overall
Housing
Stock
Distribution
25%
22%
34%
19%
700%
XRF * 6
20%
**js*!«c#
19%
4%
700%
XRF £ 6
and
> 10% Bad
Condition
"^'iiitat^^
, /"jjjP»'x.sSI
17%
-
700%
Dust
Concentration
^ 500 ppm
jiSB^vv:
^SiST''""-
27%
12%
700%
Soil
Concentration
^ 500 ppm
.',3&»«&"v
13%
17%
15%
700%
8.5.3 Cost of Housing
The equity effects of uneven regional and housing age distribution of the four hazards
are compounded by an uneven distribution of various demographic and socioeconomic sub-
populations. Although some of the nation's older housing stock is premium real estate, and
commands a high market price, in general older housing units are less expensive than newer
units. Thus, the higher prevalence of the four hazards in the older housing stock is related to
the fact that people living in lower cost housing are disproportionately exposed to the lead-based
paint hazards. The prevalence of the four hazards by monthly housing cost (measured as either
monthly rent or monthly mortgage payment to amortize a 10 percent mortgage in 30 years, not
including taxes or insurance) is shown on Exhibit 8-5. The least expensive housing (less than
$250 per month) has a disproportionately higher share of each of the four hazards than more
expensive housing.
Notice that the problem is not confined to low-cost housing. Even the most expensive
housing units have some prevalence of hazards, and comprise a disproportionately higher share
of all units with high XRF readings.
Abt Associates, Inc.
8-9
Draft, January 10, 1994
-------
EXHIBIT 8-5
Distribution of Monthly Housing Costs and Lead-Based Paint Hazards
Monthly
Housing
Cost
< $250
$250-$500
$500-$750
$750-1500
> $1500
Total
Overall
Housing
Stock
Distribution
17%
29%
18%
20%
16%
700%
XRF ^ 6
^.•'^t&^$
26%
13%
2%
^^ii^wsk
• --^rf *v»"~y™
100%
XRF Ss 6
and
> 10% Bad
Condition
8%
-
7%
700%
Dust
Concentration
£ 500 ppm
22%
15%
21%
18%
700%
Soil
Concentration
^ 500 ppm
^:32%t^
24%
23%
14%
7%
700%
8.5.4 Income
As would be expected, the relationship of lead-based paint hazards and income reflects
that poorer people tend to live in the lower-cost houses, and thus bear a disproportionate share
of the exposure to the four hazards. Exhibit 8-6 shows the distribution of household income and
the hazards and that while the poorest people are far more likely to live in housing with a paint
hazard, the hazards also occur for higher income households.
EXHIBIT 8-6
Distribution of Household Annual Income and Lead-Based Paint Conditions
Household
Annual
Income
< $10k
$10 - 20k
$20 - 30k
> $30k
NA
Total
Overall
Housing
Stock
Distribution
19%
16%
18%
40%
6%
700%
XRF 3» 6
42%
9%
16%
27%
5%
700%
XRF £ 6
and
> 10% Bad
Condition
52% ,
4%
..' -24%" ;:•••.-
20%
-
700%
Dust
Concentration
5: 500 ppm
" "
' . 28% Vr
12%
12%
'4S«L *
3%
700%
Soil
Concentration
^ 500 ppm
19%
9%
• »&#*•'
43%
5%
700%
Abt Associates, Inc.
8-10
Draft, January 10, 1994
-------
8.5.5 Affordability
The regulatory impact analysis has considered abatement decision rules where each
household chooses the type of abatement that maximizes its net benefits (both social and private
benefits e.g., damages avoided by future children born into the current householders' abated
house are included as well as the damages avoided by the householders' children). No
consideration has been made of whether the household could afford its optimal abatement choice.
The following analysis compares an affordability threshold based on household income to the
costs of the abatements chosen under the decision rules developed in this report. The issues
raised for public financing of abatements are also discussed.
Affordability Measure and Income
Affordability measures are typically based on either income or the value of the asset
being improved. For this analysis, we chose an income measure because of the short time
(seven years) over which the benefit is accrued and the availability of data. In the past, the
Agency has used a range of affordability measures. Typical measures for large scale capital
projects are 1.8-2.1 percent of median household income in perpetuity (U.S. EPA, 1990).
However, the case considered in this report is different because the benefits to a householders'
child (and thus the duration of payment) occur over seven years. Therefore, we assumed that
householders would be willing to spend as much as (5%) of their income annually if the duration
were limited to seven years. The long term cost of capital, seven percent, is used as the
discount rate.
The distribution of household incomes from the Housing and Urban Development
(HUD) study (Exhibit 8-6) served as the basis for this analysis. Two modifications were made
to the distribution; first the home owners who had not reported incomes were redistributed in
the same proportions as those who had reported, and second, the "greater than $30,000" income
category was divided into two ranges of equal frequency (those between $30,000 and $50,000,
and those greater than $50,000). Because $30,000 was the median income in 1990, the 75"1
percentile income, $50,000, was chosen as the range limit (U.S. Department of Commerce,
1992). The resulting household income distribution is shown in Exhibit 8-7.
EXHIBITS-?
Household Income Distribution and Affordability Threshold
Household Annual Income
< $10k
$10- <20k
$20- <30k
£30k - <50k
£50k
Overall Housing Stock
Distribution
21%
17%
19%
21.5%
21.5%
Affordability Threshold
$1,442
$4,325
$7,208
$8,650
$14,416
Abt Associates, Inc.
8-11
Draft, January 10, 1994
-------
Exhibit 8-7 also shows the affordable price (or threshold) associated with each income
level; households are assumed to pay no more than the affordability threshold for an abatement.
The affordable abatement cost was calculated from the midpoints of the income ranges except
in the final two categories where the lower limits of the category ($30,000 and $50,000) were
used. By using an affordability measure based on income, two further assumptions have been
made implicitly. First, we assume owners pay the cost of abatement and generate no increase
in property value as a result, and second, we assume that the cost of abatement is passed directly
to renters. If property values increase as a result of abatement, we have underestimated the
number of affordable abatements. Exhibit 8-8 compares the abatement costs, described in detail
in Chapter 4, to the affordability thresholds. Note that only nonrecurrent dust abatement is
affordable at all income levels and only 18 of the 95 income/abatement combinations are
affordable.
Abt Associates, Inc. 8-12 Draft, January 10, 1994
-------
EXHIBIT 8-8
Affordable Abatement Scenarios by Income Level for Five Percent of Income
Abatement Scenario
High-end Paint
Low-end Paint
High-end Soil
>2000 ppm and exterior paint
£2000 ppm and exterior painl
>2000 ppm and no exterior paint
£2000 ppm and no exterior paint
Low-end Soil
Recurrent Dust
Nonrecurrent Dust
High-end Paint and High-end Soil
>2000 ppm and exterior paint
£2000 ppm and exterior paint
>2000 ppm and no exterior paint
£2000 ppm and no exterior paint
High-end Paint and Low-end Soil
Low-end Paint and High-end Soil
>2000 ppm and exterior painl
£2000 ppm and exterior paint
>2000 ppm and no exterior paint
£2000 ppm and no exterior paint
f 4|W-fnd P"int And 1 4ia/^nH SAI|
Abatement Unit Cort
$10.500
$2.750
$21.412
$12,998
$16.412
$7.998
$7.493
$7.676
$750
$31.912
$23.498
$26.912
$18,498
$17.993
$24,162
$15.748
$19,162
$10.748
tin 74*
1 Income
Affordability
$20K
$4,325
X
X
$20->$30K
$7.208
X
$30->$50K
$8,650
>$50K
$14,416
Abt Associates, Inc.
8-13
Draft, January 10, 1994
-------
Results
Exhibit 8-9 shows the number of affordable abatements by scenario and decision rule
over the fifty year model lifetime. (The nine decision rules were explained in Chapter 6.) As
explained in Chapter 3, each abatement decision represents a portion of the housing stock. To
determine the number of unaffordable abatements we compared the five affordability thresholds
in proportion to their frequency in the housing stock distribution (as shown in Exhibit 8-7) to
the cost of the abatement chosen. Those abatements costing more than the affordability
threshold were considered unaffordable.
The percent of optimal abatements that are affordable range from over 99 percent in the
voluntary optimum decision rule to 43 percent for rules where doing only non-recurrent dust
abatement, the most affordable scenario, does not satisfy the decision rule constraints. The
affordability by abatement type (in Exhibit 8-9) shows where the majority of the unaffordable
abatements are for each decision rule. The constraint of abating all nonintact paint combined
with the relatively high cost of paint abatement causes these categories (high-end paint, low-end
paint, high-end paint and high-end soil, low-end paint and low-end soil, and low-end paint and
high-end soil) to account for the majority of the unaffordable abatements.
For every decision rule but the voluntary optimum, the total net benefits of the affordable
abatements are higher than the total net benefits of all abatements. This means that for all
decision rules other than the voluntary optimum, many of those who could not afford to abate
had individual negative net benefits. For example, the total net benefit for a decision rule based
on 1,200 ppm dust are $3.3 billion when affordability is ignored. Once affordability thresholds
are imposed, the net benefits rise to $15.6 billion. In the voluntary optimum, however,
individual benefits always exceed individual costs. Thus, eliminating any households because
they cannot afford to abate will reduce the net benefits.
From a social welfare perspective, only those homes with positive net benefits warrant
abatement. Those households who cannot afford to abate may be publicly funded. The
following discussion characterizes the number and value of the abatements that would qualify
for assistance under this perspective. First, Exhibit 8-10 shows the number of unaffordable
abatements which have positive net benefits, the majority of which are low soil abatements. The
total net benefits correspondingly range from $454 to $680 million or about $33 million to $49
million annually at a seven percent discount rate. The total cost, over the fifty year model
lifetime, of subsidizing these abatements ranges from $622 million for the paint condition only
to almost $2 billion for the three media constrained decision rule. This translates into
approximately $45 million to $145 million annually.1
If all the unaffordable abatements were funded, whether they result in positive net
benefits or not, the cost range would be $1.1 billion to $20.8 billion, which is equivalent to $80
The numbers presented represent an upper bound by subsidizing the total cost of the abatement. If each
household paid to its affordability threshold, the total cost over 50 years would range from $387 million to $1.3
billion. This translates into $28 million to $94 million annually.
Abt Associates, Inc. 8-14 Draft, January 10. 1994
-------
million to $1.5 billion annually.3 However, the range of net benefits would fall significantly.
The new range is minus $12.7 billion to positive $454 million, which, on an annual basis, is
minus $920 million to positive $33 million.
Note that this analysis only considers the affordability of the optimal abatement chosen
based on maximizing net benefits. A more in-depth analysis could introduce an affordability
constraint when the optimal abatement is being chosen, limiting the abatement choices to those
below an affordability threshold. This would decrease the total benefits but could also reduce
costs. Finally, the cost of testing for lead has not been included. These costs can be substantial,
$713 per unit, if all media are tested. Obviously, the number of unaffordable abatements would
increase if the testing costs were added.
Funding for lead abatements (including inspection) is formalized and authorized at $125
million for 1993 and $250 million for 1994 according to Title X, Section 1011. Additional
funds may be available through subsequent Appropriations Acts. This level of funding compares
favorably with the amount needed to subsidize unaffordable abatements that result in positive net
benefits. However, the funding may not be adequate to subsidize all unaffordable abatements
induced by certain decision rules. Further, the funding is only appropriated over two years and
the model project lifetime is fifty years. (Thus the same level of funding would have to be
appropriated each year for the next fifty years.) The case presented here does not consider all
the possible sources of funds for residential lead abatement. States and localities have a history
of funding such projects. It may be possible to extend the funds by providing concessionary
loans as Massachusetts has done. In this way, abatements can be performed on loans at
subsidized interest rates but at a lower cost to government than a grant program. The case
presented here represents an upper bound estimate of the annual cost to government of
subsidizing abatements.
These values represent the upper bound by subsidizing the total cost of the abatement. The total cost over
SO years would range from $780 million to $12.7 billion equivalent to $56 million to $920 million annually if each
household paid an amount equal to its affordability value.
Abt Associates, Inc. 8-15 Draft. January 10. 1994
-------
EXHIBIT 8-9
Affordability (at Five Percent of Income) of Abatement Choices for Five Alternative Decision Rules
Decision Rules
'•
2.
3.
4.
5.
Voluntary
Ootimum
Paint Condition
Only
Single
Medium
Plus
Condition
2-Media
Plus
Condition
3a.
3b.
3c.
4a.
4b.
4c.
3-Media Plus
Condition
Soil
(ppm)
-
•
2.300
~
*
2,300
2,300
~
2,300
Dust
(ppm)
-
•
-
1,200
~
1,200
™
1,200
1,200
Paint
(XRF,
mg/cm1)
-
•
-
*
20
•
20
20
20
Nonintacl Paint
Abatement
Recommended
No
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Total Net
Benefits
Exclusive ol
Testing
Costs
($ millions)
34,294
-17,349
-17.159
3,299
-17.631
3,017
-17,440
3,017
2,736
Total
Affordable
Abatements
(1000s)
(Percent of
Total)
44,832
99.26%
3,089
43.74%
3,488
43.23%
11,169
71.59%
3,134
43.75% "
11,391
70.33%
3,533
43.24%
11,213
71.41%
11,435
70 17%
Total Net
Benefits of
Affordable
Abatements
Exclusive
of Testing
Costa ($
millions)
33,840
•4,834
-4,711
15,635
-4,921
15,554
-4,798
15,548
15,468
Number and Percent of Total Homeowners with Unaffordable Abatements by
Type • (1000s)
HP
3,265
46%
3,265
40%
3,286
21%
3.313
46%
3.286
20%
3,313
41%
3,334
21%
3,334
20%
LP
4
0.01%
583
8%
570
7%
573
4%
591
8%
561
3%
579
7%
582
4%
570
3%
HS
42
0.27%
0
42
0.26%
42
0.27%
42
0.26%
LS
329
0.73%
573
7%
234
2%
573
4%
573
7%
234
1.49%
573
4%
RD
158
1%
158
1%
158
1%
158
1%
HP/HS
HP/LS
114
2%
114
1%
114
1%
114
2%
114
1%
114
1%
114
1%
114
LP/HS
14
0.09%
14
0.09%
14
0.09%
14
LP/LS
12
0.17%
58
1%
12
0.08%
12
0.17%
58
0.36%
58
0.71%
12
0.08%
58
NRD
0
Oft
0
0%
0
0%
0
0%
0
'Abatement Codes: High Paint(HP); Low Paint(LP); High Soil(HS); Low Soil(LS); Recurrent Dust (RD); High Paint and High Soil(HP/HS); High Paint and Low Soil(HP/LS); Low Paint and High
Soil (LP/HS); Low Paint and Low Soil (LP/LS); Nonrecurrent Dust (NRD). The abatement activities, wen described in Exhibits 4.1-4.6.
Abt Associates, Inc.
8-16
Draft. January 10, 1994
-------
EXHIBIT 8-10
AITordabilily (at Fire Percent of Income) of Abatement Chokes for Households with Positive Net Benefits for Five AltematiTe Decision Rules
Decision Rules
1.
2.
3.
4.
5.
Voluntary
Ootimum
Paint Condition
Only
Single
Medium
Plus
Condition
2-Media
Plus
Condition
3a.
3b.
3c.
4a.
4b.
4c.
3-Media Plus
Condition
Soil
(ppm)
-
•
2.300
-
-
2.300
2,300
~
2.300
Dust
(ppm)
-
•
•
1,200
•
1,200
-
1,200
1.200
Paint
(XRF.
ing/cm1)
-
•
•
-
20
•
20
20
20
Nonintact Paint
Abatement
Recommended
No
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Total Net
Benefits of
Unaffordable
Abatements
with Positive
Net Benefits
Exclusive of
Testing Costs
($ millions)
454
226
642
532
226
680
642
532
680
Total Costs of
Unaffordable
Abatements
with Positive
Net Benefits
Exclusive of
Testing Costs
($ millions)
1.141
622
1,679
1,728
622
1,976
1.679
1,728
1.976
Total Costs
of
Unaffordable
Abatements
1,141
17,545
19,726
19,223
17.787
20,595
19,968
19,464
20.837
Number and Percent of Total Homeowners with Unaffordable Abatements by
Type* (1000s)
HP
104
1.47%
104
1.29%
104
0.67%
104
1.45%
104
0.64%
104
1.27%
104
0.66%
104
064%
LP
4
0.01%
48
0.69%
48
0.60%
43
0.27%
48
0.68%
43
0.26%
48
0.59%
43
0.27%
43
0.26%
HS
42
0.27%
42
0.26%
42
0.27%
42
0.26%
LS
329
0.73%
306
4%
234
1.50%
306
1.89%
306
3.75%
234
1.49%
306
1 88%
RD
42
0.27%
42
0.26%
42
0.27%
42
0.26%
HP/HS
HP/LS
0
0%
0
0%
0
0%
0
0%
0
0%
0
0%
0
0%
0
0%
LP/H
S
0
0%
0
0%
0
0%
0
0%
LP/LS
12
0.17%
12
0.15%
12
0.08%
12
0.17%
12
0.08%
12
0.15%
12
0.08%
12
008%
NRD
0
0%
0
0%
0
0%
0
0%
0
0%
•Abatement Codes: High Paint(HP); Low Painl(LP); High Soil(HS); Low Soil(LS); Recurrent Dust (RD); High Paint and High Soil(HP/HS); High Paint and Low Soil(HP/LS); Low Paint and High
Soil (LP/HS); Low Paint and Low Soil (LP/LS); Nonrecurrent Dust (NRD). The abatement activities wen described in Exhibits 4.1-4.6.
Abt Associates, Inc.
8-17
Draft, January 10, 1994
-------
Alternative Threshold Analysis
Exhibit 8-11 shows which abatements are affordable if two percent of income over seven
years is used as the affordability threshold. The two percent level corresponds to past EPA
analysis although the period, seven years, is shorter as described above (U.S.EPA, 1990). Only
seven of the 95 income/abatement combinations are affordable. As expected, the percent of
affordable abatements decreases for each decision rule when compared to the five percent
affordability threshold. (See Exhibit 8-12.) Only 78 percent of the abatements are affordable
under the voluntary optimum; this falls to less than 25 percent for decision rules where
nonrecurrent dust abatement is not an option. Exhibit 8-13 shows that the total cost over fifty
years of subsidizing all households below the threshold which have positive individual net
benefits is $825 million to $4.4 billion or $60 to $319 million annually discounted at seven
percent. Corresponding net benefits are $334 million to $7.8 billion or $24 to $565 million
annually. This alternative analysis shows the importance of the affordability threshold chosen
to the quantity of abatements that are considered unaffordable.
Abi Associates, Inc. 8-18 Draft, January 10, 1994
-------
EXHIBIT 8-11
Affordable Abatement Scenarios by Income Level for Two Percent of Income
Abatement Scenario
High-end Paint
Low-end Paint
High-end Soil
>2000 ppm and exterior paint
£2000 ppm and exterior paint
>2000 ppm and no exterior paint
£2000 ppm and no exterior paint
Low-end Soil
Recurrent Oust
Nonrecurrent Dust
High-end Paint and High-end Soil
>2000 ppm and exterior paint
£2000 ppm and exterior paint
>2000 ppm and no exterior paint
£2000 ppm and no exterior paint
High-end Paint and Low-end Soil
Low-end Paint and High-end Soil
>2000 ppm and exterior paint
£2000 ppm and exterior paint
>2000 ppm and no exterior paint
£2000 ppm and no exterior paint
Isra/wtiiH Paint and lnw-»nrt Sail
Abatement Unit Cost
$10,500
$2,750
$21.412
$12,998
$16,412
$7.998
$7,493
$7,676
$750
$31,912
$23,498
$26,912
$18.498
$17,993
$24.162
$15,748
$19,162
$10,748
tin 741
Income
AfTordability
Threshold
<$10K
$577
$10->$20K
$1,730
X
$20->$30K
$2.883
X
X
$30->$50K
$3,460
>S50K
$5.767
Abt Associates, Inc.
8-19
Draft, January 10, 1994
-------
EXHIBIT 8-12
AfTordabilily (al Two Percent of Income) of Abatement Choices for Five Altematife Decision Rules
1.
2.
3.
4.
5.
Decision Rules
Voluntary
Ootimum
Paint Condition
Only
Single
Medium
Plus
Condition
2-Media
Plus
Condition
3a.
3h
3c.
4a.
4b.
4c.
3-Media Plus
Condition
Soil
(ppm)
-
"
2,300
•
"
2.300
2.300
"
2.300
Dust
(ppm)
-
•
•
1.200
~
1.200
"
1,200
1,200
Paint
(XRF.
ing/cm1)
-
•
~
•
20
'
20
20
20
Noninlacl Paint
Abatement
Recommended
No
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Total Net
Benefits
Exclusive of
Testing
Costs
($ millions)
34.294
-17.349
-17.159
3.299
-17.631
3.017
-17,440
3.017
2,736
Total
Affordable
Abatements
(1000s)
(Percent of
Total)
35,221
77.98%
1.720
24.35%
1,684
20.87%
7.837
50.23%
1.745
2435%
7.801
48.16%
1,708
20.91%
7,862
50.07%
7.825
48 02%
Total Net
Benefits of
Affordable
Abatements
Exclusive
of Testing
Costs ($
millions)
26,463
-1.278
-1,248
14.767
-1.306
14.797
-1.276
14,739
14,769
Number and Percent of Total Homeowners with Unaffordable Abatements by Type*
(1000s)
HP
4,160
59%
4,160
52%
4.186
27%
4.221
59%
4,186
26%
4,221
52%
4.247
27%
4,247
26%
LP
8
0.02%
1,054
15%
1,032
13%
1.038
7%
1,069
15%
1,015
6%
1,047
13%
1,053
7%
1,031
6%
HS
74
0.47%
0
74
0.46%
74
0.47%
74
0 45%
LS
577
1.28%
1,006
12%
411
3%
1,006
6%
1,006
12%
411
2.62%
1,006
6%
RD
276
2%
276
2%
276
2%
276
2%
HP/HS
HP/LS
114
2%
114
1%
114
1%
114
2%
114
1%
114
1%
114
1%
114
1 %
LP/HS
18
0.11%
18
0.11%
18
0.11%
18
LP/LS
16
0.22%
74
1%
16
0.10%
16
0.22%
74
0.46%
74
0.91%
16
0.10%
74
NRD
9,359
20.72%
1,633
10.47%
1,633
10.08%
1,633
10.40%
1,633
•Abatement Codes: High Paint(HP); Low Paint(LP); High Soil(HS); Low Soil(LS); Recurrent Dust (RD); High Paint and High Soil(HP/HS); High Paint and Low Soil(HP/LS); Low Paint and
High Soil (LP/HS); Low Paint and Low Soil (LP/LS); Nonrecurrent Dust (NRD). The abatement activities were described in Exhibits 4.1-4.6.
Abt Associates, Inc.
8-20
Draft. January 10, 1994
-------
EXHIBIT 8-13
Affordability (at Two Percent of Income) of Abatement Chokes for Households with PositiTe Net Benefits for Five Alternatta Decision Rules
Decision Rules
1.
2.
3.
4.
5.
Voluntary
Optimum
Paint Condition
Only
Single
Medium
Plui
Condition
2-Media
Plui
Condition
3a.
3b.
3c.
4a.
4b.
4c.
3-Media Plus
Condition
Soil
(ppm)
-
•
2,300
•
~
2.300
2,300
•
2,300
Dust
(ppm)
-
•
-
1,200
•
1.200
•
1.200
1.200
Paint
(XRF.
mg/cm*)
-
•
•
-
20
•
20
20
20
Nonintact Paint
Abatement
Recommended
No
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Tola! Net
Benefits of
Unaffbrdable
Abatements
with Positive
Net Benefits
Exclusive of
Testing Costs
($ millions)
7,831
334
1.065
5,141
334
5,401
1,065
5,141
5,401
Total Costs of
Unaflbrdable
Abatements
with Positive
Net Benefits
Exclusive of
Testing Costs
($ millions)
4,496
825
2,680
3,203
825
3,638
2,680
3.203
3.638
Total Costs
of
Unaffoidable
Abatements
Threshold ($
million)
4,496
22,487
26,210
25,801
22.801
28,104
26,524
26.115
28,418
Number and Percent of Total Homeowners with Unaffbrdable Abatements by
Type* (1000s)
HP
132
1.87%
132
1.64%
132
0.85%
132
0.00%
132
0.82%
132
1.62%
132
0.84%
132
0 81%
LP
8
0.02%
88
1.24%
88
1.09%
77
0.50%
88
0.00%
77
0.48%
88
1.07%
77
0.49%
77
0 47%
HS
74
0.47%
74
0.46%
74
0.47%
74
0 45*
LS
577
1.28%
537
6.65%
411
2.63%
537
3.32%
537
6.57%
411
2.62%
537
3 29%
RD
74
0.47%
74
0.46%
74
0.47%
74
0 45%
HP/HS
HP/LS
0
0%
0
0%
0
0%
0
0%
0
0%
0
0%
0
0%
0
0%
LP/HS
0
0%
0
0%
0
0%
0
0%
LP/LS
16
0.22%
16
0.19%
16
0.10%
16
0.22%
16
0.10%
16
0.19%
16
0.10%
16
0 10%
NRD
9,359
21%
1,591
10%
1.591
10%
1.591
10%
1,591
10%
•Abatement Codes: High Paint(HP); Low Paint(LP); High Soil(HS); Low Soil(LS); Recurrent Dust (RD); High Paint and High Soil(HP/HS); High Paint and Low Soil(HP/LS); Low Paint and High
Soil (LP/HS); Low Paint and Low Soil (LP/LS); Nonrecurrent Dust (NRD). The abatement activities were described in Exhibits 4.1-4.6.
Abt Associates, Inc.
8-21
Draft, January 10, 1994
-------
8.5.6 Race
Lead-based paint hazards are more likely to affect African-Americans than other racial
sub-populations. This is a result of both the larger African-American share of the population
in the North-East and Mid-West, and of the higher poverty rate for African-Americans. Exhibit
8-14 shows the distribution of the hazards by race. (Race is defined as the stated race of the
youngest person in the household.)
EXHIBIT 8-14
Distribution of Household Racial Composition and Lead-Based Paint Conditions
Race
African-
American
Hispanic
White
Other
Total
Overall
Housing
Stock
Distribution
9%
7%
78%
7%
700%
XRF > 6
.-'Oat-
~ff •• ' '. • > •
3%
66%
2%
700%
XRF2t 6
and
> 10% Bad
Condition
:*j$j|Njj. $
"• ' ^S'*^
4%
44%
-
700%
Dust
Concentration
^ 500 ppm
^'*xt$jjjj^^
^...,.-^-:^:.. -j
4%
71%
11%
700%
Soil
Concentration
^ 500 ppm
11%
3%
79%
7%
700%
8.5.7 Other Socioeconomic Variables
The prevalence of lead-based paint hazards based on other socioeconomic variables does
not show as dramatic a disproportionate prevalence as the region, income and race variables.
Exhibit 8-15 shows the prevalence for the following variables: ownership, presence of children,
presence of elderly. A summary of this dispropoitionality is as follows:
Ownership: rental units are somewhat more likely to have high XRF readings.
Presence of children (6 years old or less): there is no disproportionate prevalence
of lead-based paint hazards among units with young children.
Elderly (defined as at least one person over the age of 65 living in the unit): no
overall pattern of disproportionate prevalence, but units including elderly people
are somewhat less likely to have both high XRF readings and bad paint
conditions.
Abt Associates, Inc.
8-22
Draft. January 10, 1994
-------
EXHIBIT 8-15
Distribution of Other Household Demographic Characteristics
and Lead-Based Fault Conditions
Ownership
Rent
Own
Children
£ 6 Years
Present?
No
Yes
Adult
> 65 Years
Present?
No
Yes
Overall
Housing
Stock
Distribution
XRF ;> 6
XRF £ 6
and
> 10% Bad
Condition
Dust
Concentration
£ 500 ppm
Soil
Concentration
£ 500 ppm
35%
65%
- '4i*-Ki
55%
"wf^V-ijMS™":- %-^V •
59%
31%
69%
36%
64%
82%
18%
79%
21%
86%
14%
85%
15%
78%
22%
76%
24%
77%
23%
100%
-
78%
22%
80%
20%
Abt Associates, Inc.
8-23
Draft, January 10, 1994
-------
8.5.8 Data Limitations
The U.S. Housing and Urban Development survey provides the best nationwide data for correlating
socioeconomic information and residential housing lead levels. However, the data are limited. Because only
284 privately owned homes were sampled the actual number of homes in the various socioeconomic strata of
interest is small. The confidence intervals are thus correspondingly large. While the data may show trends,
the small sample size indicates caution should be used in interpreting the results.
8.5.9 Environmental Equity Conclusions
Existing lead-based paint hazards are a risk to all segments of our population living in pre-1980 housing,
and local, state and federal efforts to reduce the risks of lead-based paint must extend to all potentially affected
parties. However, the HUD survey does indicate that some segments of our society are at relatively greater
risk than others. In particular, the residents of older, low cost housing are exposed to a disproportionately
greater share of lead potential hazard than other housing units. The housing stock in the North-East (and to
some extent the Mid-West) includes a larger share of such units than other regions, creating a regional inequity
in the prevalence of the problem. Because poorer people usually occupy low-cost housing, the hazards
disproportionately fall on lower income sub-populations (especially households living in poverty, with annual
incomes below $10,000), creating an income inequity. Finally, the relatively larger share of African-Americans
in the lower income groups results in racial inequity.
Although the baseline risks from lead-based paint disproportionately fall on poorer sub-populations,
abatement may well be more likely to occur in housing units occupied by wealthier households. Most of the
abatements under the Lead-Based Paint Hazard Reduction Act will be voluntary, and wealthier households are
more likely to have the means to abate an existing problem in their home, or avoid moving into a housing unit
with a known lead-based paint hazard. Thus even though a national strategy of eliminating lead-based paint
risks targets a problem affecting a greater share of poor households and African-Americans, the impact of
income on the ability to undertake voluntary abatements may result in a more inequitable distribution of the
risks in the future.
As shown in Section 8.S.S the ability to afford abatements may have a serious impact on the number
of abatements undertaken voluntarily. In addition, when the affordable abatements are considered alone the
net benefits rise in all but the voluntary optimum decision rule. This implies that not all unaffordable
abatements may be worth subsidizing because some have negative net benefits. However, subsidizing only
those with positive net benefits raises a fairness issue, since many of the affordable abatements induced by the
decision rule had negative net benefits yet presumably were paid for privately by households. Should those
who cannot afford abatements be given the same treatment? In practice, can the homes with positive and
negative net benefits be identified in a cost effective manner? Finally, significant funds, sustained over time,
will be required to subsidize even those unaffordable abatements which have net benefits greater than zero.
Abt Associates, Inc. 8-24 Draft, January 10, 1994
-------
8.6
U.S. Environmental Protection Agency, 1990. (U.S. EPA, 1990) "National Characterization of Small
Communities: The Ability to Finance Wastewater Construction Projects", April.
U.S. Department of Commerce, 1992. (U.S. Department of Commerce, 1992) "Statistical Abstract of the
United States, 1992" U. S Government Printing Office.
Abt Associates. Inc. 8-25 Draft, January 10. 1994
------- |