Michael P. Walsh, Deputy Assistant Administrator for
Mobile Source Air Pollution Control

Important Notice
This Regulatory Analysis was completed prior to the EPA
decision to delay the model year of compliance from 1983 to 1984.
Thus, the analyses presented here were done under the assumption
that LDTs would meet the standards beginning in the 1983 model
year. While shifting the first model year of compliance to 1984
will have an effect on the analyses and their conclusions, they
still remain valid. (The following paragraphs illustrate the kinds
of effects which occur.) For this reason we have chosen to go
ahead and publish the document as it is.
The delay in the regulations has an effect on the environ-
mental and economic analyses of Chapter IV and V because of the
dependence of the analyses on various growth projections. Given
another year, the objective criteria surrounding the arrival of new
LDT control systems will have changed slightly. For example, the
constantly changing mix of emission contributions to urban air
quality will be a little different in 1984 than 1983. Likewise,
the increased production of LDTs which we expect to happen during
1984-88 relative to 1983-87 will mean that the aggregate cost of
the package over 5 years will be slightly greater; the cost per
vehicle, on the other hand, will be slightly less, since more
vehicles will share the aggregate cost. These kinds of effects on
the environmental and economic analyses are not great, and hence,
we do not feel a need to adjust them to reflect the 1-year shift in
the regulations.
As a final note, the cost-effectiveness numbers computed in
Chapter VII should not be affected by the delay in the regulations.
This is simply because the changes in the economic cost and envi-
ronmental benefit (which make up the calculation) are so small that
their ratio is not noticeably affected.

This document has been prepared in satisfaction of the Regu-
latory Analysis required by Executive Order 12044 and the Economic
Impact Assessment required by Section 317 of the amended Clean Air
Act. This document also contains an Environmental Impact Statement
for the Final Rulemaking Action.

Table of Contents
I.	Summary 			1
A.	Industry Description 		2
B.	Impact on the Environment		2
C.	Costs		3
D.	Alternatives . 			3
E.	Cost Effectiveness		4
II.	Introduction 		5
A.	Light-Duty Truck Emission Regulation
History and Background 		5
B.	Description of Final Rulemaking 		10
C.	Organization of the Regulatory Analysis 		13
III.	Description of the Product and the Industry	16
A.	Description of Light-Duty Trucks 		16
B.	The Light-Duty Truck Industry 		20
C.	Light-Duty Truck Sales 		25
D.	Other Considerations . . . 		33
IV.	Environmental Impact		38
A.	Background		38
B.	Primary Impact		42
C.	Potential Secondary Environmental
Impacts		57
D.	Irreversible and Irretrievable Committment
of Resources ............. 		S3
E.	Relationship of Short-Term Uses of the
Environment to Maintenance and Enhancement of
Long-Term Productivity 		58
V.	Economic Impact		61
A.	Cost to Truck Manufacturers		64
B.	Costs to Users of Light-Duty Trucks 		75
C.	Aggregate Costs		79
D.	Sensitivity of the First Price Increase
to Changes in Key Analysis Parameters		80
E.	Socio-Economic Impact 		87
VI.	Alternative Actions 	 ......	96
A.	Introduction					96
B.	Alternative Standards 		96
C.	Alternatives to Specific Elements
of the Rulemaking		97
D.	Alternative Timing for Implementation		98

Table of Contents (cont'd)
VII. Cost Effectiveness	100
A.	Methodology	100
B.	Background	100
C.	Summary	101
D.	Overall Rulemaking 		102
E.	Redefinition of Useful Life	102
F.	Allowable Maintenance Restrictions 		Ill
G.	Selective Enforcement Auditing (SEA) 		Ill
H.	Inspection and Maintenance (I/M) 		117
I.	Idle Test	119

As the total amount of urban emissions from light-duty vehi-
cles is reduced, the portions which light-duty trucks (LDTs) and
heavy-duty vehicles (HDVs) contribute becomes increasingly signifi-
cant. For example, it is expected that the LDT and HDV fractions
of total mobile source urban hydrocarbon (HC) emissions will climb
from 14% and 12%, respectively, in 1976 to 22% and 21% in 1995.
The carbon monoxide (CO) fractions will increase from 13% and 15%,
for LDTs and HDVs respectively to 28% and 18%. Nitrogen oxides
fractions will increase from 9% and 23% to 10% and 39%. It is in
light of these expectations that Congress has mandated stricter
controls on the gaseous emissions from heavy-duty engines used in
HDVs and from light-duty trucks in the 6,000 to 8,500 lbs. GVWR
range (hereafter called "heavy" light-duty trucks). It is also in
light of these expectations that EPA is considering stricter
controls on emissions from light-duty trucks in the under-6,000
lbs. GVWR range (hereafter called "light" light-duty trucks).
This rulemaking follows in part from the Congressional re-
quirement that EPA prescribe by regulation standards for heavy
LDTs which by 1983 will require 90% reductions in HC and CO emis-
sions, relative to a baseline of uncontrolled (1969) heavy LDTs.
The remaining part of the rulemaking follows from the general
Congressional directive that EPA establish standards for emissions
from new motor vehicles which cause or contribute to air pollution
which endangers public health or welfare.
The purpose of this regulatory analysis is to present the
results of EPA analyses of the environmental and economic impacts
and the cost effectiveness of the proposed regulations. The reader
will find chapters devoted as well to the make-up of the light-duty
truck industry and to alternative actions considered by the Agency.
The regulations include the statutory 1983 HC and CO 90
percent reduction standards for heavy light-duty trucks. These
standards are 0.8 g/mi HC and 10 g/mi CO.
The regulations also include several other standards.
These are (1) HC and CO standards for light LDTs, equal to the
standards for heavier LDTs; (2) idle CO standards applicable to
gasoline-fueled LDTs; and (3) a zero-emissions standard for
crankcase emissions from all diesel LDTs.
Although some of the standards for heavy LDTs are being
proposed under one statutory authority while the remaining stan-
dards are being proposed under another statutory authority, the
levels of all standards will be the same for all LDTs. The class
will remain a single class, with a single test procedure and a
single set of certification test vehicles.

Selective Enforcement Auditing (SEA) procedures for LDTs are
being revised in this package to conform to the improved sampling
system recently enacted for heavy-duty engines and vehicles.
These revisions include a 10% acceptable quality level (AQL) to
replace the current 40% AQL applicable to LDTs.
Additional changes appear in the rule. A new definition of
"useful life" is introduced. "Useful life" will be the average
period of use up to vehicle retirement or engine rebuild or re-
placement, to be determined by the manufacturer and stated on the
tune-up label. Revisions to the restrictions on the maintenance
performed on vehicles during durability testing and recommended to
purchasers are also promulgated.
A.	Industry Description
The light-duty truck industry uses both gasoline and diesel
engines and consists at present of five domestic manufacturers
which together account for 94 percent of the U.S. market and five
foreign manufacturers. All of the manufacturers with the exception
of International Harvester also produce passenger cars. The LDT
market is dominated by GM, Ford, and Chrysler, who together account
for about 88 percent of sales. Imports are a small but rapidly
increasing fraction of the market.
Sales of LDTs in 1978 were about 3.37 million units. LDT
sales have been increasing faster than sales of passenger cars over
a period from 1974 to 1978. However the growth rate is not ex-
pected to be the same in later years. In the past two years energy
availability and price has effected the sale of LDTs, whose fuel
economy is lower than LDVs. Extrapolation of sales growth over the
past decade predicts that 1984 sales will be about 3.5 million
units. By 1987, the last year covered by the cost analysis of this
document, sales will be about 4.0 million units. Light-duty diesel
truck sales are estimated at over .3 million; the 1987 sales
estimate will double to over .6 million.
Vehicles in the light-duty truck class are mainly pick-up
trucks and vans used primarily for personal transportation.
Those produced by the manufacturers which also produce passenger
cars share many components with those cars. Emission control
technology used on passenger cars has generally been easily adapted
for use on light-duty trucks. This is expected to be the case in
the future as well.
B.	Impact on the Environment
The projected improvement in light-duty truck emissions and
the ensuing decrease in total urban emissions will have a signifi-
cant positive effect on the environment. Light-duty trucks will
exhibit by 1999 reductions of 55% in HC emissions, and 62% in CO
emissions, relative to their performance if the 1979 standards and
other certification requirements were not changed.

On the basis of these reductions and the air quality models
currently approved for use by the states, EPA estimates that as a
result of the final rulemaking, by 1999 urban ambient levels of
oxidant will be reduced by 1 percent to 2 percent, levels of carbon
monoxide by 4 percent.
Secondary emission effects; water, noise and energy consump-
tion effects; and commitments of scarce resources as a result of
promulgation of the 1984 regulations are all expected to be negli-
gible. Effects on urban areas are expected to be limited to the
improvement in urban air quality stated in the previous paragraph.
C.	Costs
The increased costs which manufacturers of light-duty trucks
will have to bear, before passing them on to their customers, as a
result of the 1983 regulations consist primarily of the cost of
installation of new emission control systems. Development programs
and certification testing are the other components of the costs to
manufacturers, but are much smaller.
Selling prices for LDTs are estimated to increase by an
average of about $95 as a result of the proposed action.
Aggregate cost for the first five year's of compliance will be
about $1.29 billion (present worth on January 1, 1983, assuming a
10% discount rate). The aggregate cost per light-duty truck sold
in the first five years will be about $95 (present worth on
January 1 of the year of production).
The 1983 regulations will increase the average retail selling
price of a vehicle by about 1 percent to 2 percent. The Ford
Econometric Model which accounts for elasiticity of demand, in-
creased first costs and greater ownership costs, gives an average
long term sales decrease of about 0.2 percent due to the proposed
regulations. The impact on both price and sales volume will be far
less than normal annual changes. There should be no noticeable
effect on industry employment or productivity.
The increase in the selling price of light-duty trucks is
estimated to contribute about 0.0075 percentage points of rise in
the Consumer Price Index in 1983. As the public will receive air
quality benefits in exchange for higher LDT prices, this rise in
the Consumer Price Index cannot properly be termed inflation.
D.	Alternatives
The alternatives evaluated are in three areas: 1) Alternative
Standards, 2) Alternatives to Specific Elements of the Rulemaking,
3) Alternative Timing for Implementation.
1. Alternative Standards - Two options were available. The
first option concerned the stringency of the standards; the second

involved dividing the light-duty truck class into subcategories and
establishing separate standards for each subcategory. EPA does not
wish to promulgate standards more stringent than those of the NPRM,
nor has it identified any basis for a less stringent level.
Concerning the option of class subdivisions, it would be outside
the range of the rulemaking to change the class limits at this
2.	Alternatives to Specific Elements of the Rulemaking
Alternatives including the redefinition of useful life, in-use
durability testing, allowable maintenance regulations, AQL, and
diesel crankcase control are not discussed in this document but are
referred to the appropriate sections of the Summary and Analysis of
3.	Alternatives Timing for Implementation - EPA proposed
the light-duty truck regulation for 1983. A detailed analysis
concerning the provisions of the 1977 Clean Air Act and their
relations to any minimum leadtime was referred to the Summary
and Analysis of Comments. The conclusion reached is that the
1983 model year is a readily attainable compliance deadline
and that there are no legal barriers to EPA's promulgation.
E. Cost Effectiveness
An analysis of the cost effectiveness of each major element of
the regulation package and the overall package has been done. The
analysis developed benefits expressed as tons of pollutant removed
(HC or CO) over the average lifetime of an individual vehicle along
with total costs for the same lifetime.
The overall incremental lifetime cost effectiveness has been
established at $164/ton for HC and $12/ton for CO. Incremental
lifetime benefits are set at about .3 tons/vehicle for HC and about
4 tons/vehicle for CO.

A. Light-Duty Truck Emission Regulation History and Background
1. History to Date
The first Federal regulations for the control of motor vehicle
emissions placed trucks and similar vehicles having gross vehicle
weight ratings (GVWR) of 6,000 pounds or less (typically "half ton"
pick-ups and vans) in the light-duty vehicle (LDV) class. The LDV
class also included passenger cars, and so, beginning in the 1968
model year, trucks of 6,000 pounds GVWR or less were subject to the
first Federal passenger car emission standards. Under this clas-
sification, these vehicles would have been required to meet the
same statutory emission reductions for hydrocarbons (HC), carbon
monoxide (CO) and oxides of nitrogen (N0x) as passenger cars, on
the same schedule. However, lawsuits were filed by the manufac-
turers disputing EPA1s classification. These challenges were
successful, and as a result, EPA created the light-duty truck class
(to 6,000 pounds GVWR) effective for the 1975 model year. The
numerical standards for the new class were slightly greater than
for the light-duty vehicle class (see Tables II-A and B for summar-
ies of Federal and California standards for light-duty vehicles
and light-duty trucks).
While the 1975 LDT standards were numerically greater, and
therefore in some cases less stringent than standards for passenger
cars of the same year, they were more stringent than the heavy-
duty engine standards, which applied to trucks on the upper side of
the 6,000 pound GVWR dividing line. Manufacturers elected to
"shift" part of their pick-up and van production to the other side
of this dividing line by building larger numbers of trucks and vans
over 6,000 pounds GVWR (heavy "half ton" and three quarter ton
vehicles). The motivation was a less costly emission control
system and the elimination of the need for catalysts (and their
resulting need for unleaded gasoline). In response to this GVWR
"migration", EPA expanded the light-duty truck class to include
most vehicles up to 8,500 pounds GVWR. This change, effective for
the 1979 model year, brought the vast majority of "personal use"
light trucks into one homogeneous class for the first time. The
LDT standards for HC, CO and NOx were revised for the same year to
reflect advances in emission control technology as demonstrated on
light-duty vehicles. EPA considered the larger weights and road
loads (aerodynamic and tire drag) typical of light-duty trucks when
it set the 1979 standards. The standards also reflected certain
test procedure revisions.
The fuel evaporative emission standards for the LDT class have
followed a somewhat different schedule. The first evaporative
emission standard was set at 2 grams per test for the 1975 model

Table II-A
Emission Standards for Light-Duty Vehicles
Federal	California

Particulate 9/

NOx Evap.l/

,	_
















TJ	Gasoline vehicles only (g/test).
2/	Federal standard instituted as part of the 1975 waiver,
3/	SHED test.
4/	Compliance with 0.39 g/mile non-methane is optional.
1980 and later
5/	Manufacturer must elect one option (A or B) for both
-	1981 and 1982.
6/	Optional 100,000 mile durability NOx standard.
7/	Possible waiver to 7.0 g/mile.
8/	Possible diesel waiver to 1.5 g/mile.
9/ Diesel vehicles only.

Table II-B
Emission Standards for Light-Duty Trucks
1968-74 7/
1979	3/
1982	2/
1983	b/
1985 6/
Evap.1/	Particulate 8/
6.0 2/
2.0 2/




-8,500 GVWR





as HD





































1/ Gasoline vehicles only (g/test).
T/ SHED test.
3/ Federal LDT class was expanded for 1979 MY to include most
vehicles up to 8,500 lbs. GVWR.
4/ Compliance with 0.39 g/mi non-methane is optional.
*5/ Optional 100,000 mile durability NOx standard.
6/ Clean Air Act requires 90% HC and CO reduction in 1983 and 75%
NOx reduction in 1985 for vehicles over 6,000 lbs. GVWR.
JJ Prior to 1975 LDTs up to 6,000 lbs. GVWR were classified as
~ LDVs.
8/ Diesel vehicles only.

year, measured using a carbon trap procedure. For 1978 a better
measurement procedure (the SHED test) was adopted and the standard
was set at 6 grams per test which, with the more accurate test
procedure, was more stringent than the old 2-gram standard. In
1979, trucks in the 6,000 to 8,500 pound GVWR range came under this
6-gram standard. A 2-gram, SHED-based standard has recently been
set for the 1981 model year.
Finally, standards for particulate emissions from diesel
light-duty trucks have also recently been promulgated. These
standards are required by the Clean Air Act as amended, but the Act
left EPA to determine their level. EPA has set a standard of
0.60 grams per mile for 1982, and a tighter standard of 0.26 grams
per mile for 1985.
It should be pointed out that the state of California, under
waivers granted by EPA, has instituted stricter standards with
earlier implementation dates than the nationwide levels. Cali-
fornia also has divided the under-8,500 GVWR group into several
narrower subclasses, each with its own set of emission standards.
While EPA puts a curb weight ceiling of 6,000 pounds on the light-
duty truck class, California does not. However EPA allows vehicles
above this ceiling, which could be certified to the less stringent
heavy-duty engine standards, to be certified to the LDT standards
and test procedures. This mitigates the effect of the curb weight
disparity on manufacturers.
2. Clean Air Act Provisions
From the rulemaking which established the separate light-duty
truck class up until this rulemaking, light trucks have been
regulated under the general authority in the Clean Air Act.
Congress had set no specified reductions for these vehicles to
meet. This general statutory authority is as follows:
The Administrator shall by regulation prescribe (and from
time to time revise) . . . standards applicable to the emis-
sion of any air pollutant from any class or classes of new
motor vehicles . . . which in his judgement cause, or contri-
bute to, air pollution which may reasonably be anticipated to
endanger public health or welfare. . .
Any regulation . . . shall take effect after such period
as the Administrator finds necessary to permit the development
and application of the requisite technology, giving appro-
priate consideration to the cost of compliance within such
(Section 202(a) of the Clean Air Act)
With the Clean Air Act Amendments of 19 77, Congress created a
statutory heavy-duty class. All vehicles over 6,000 pounds GVWR
are considered to be "heavy-duty" under the Clean Air Act. Fur-

ther, these vehicles are required to meet a minimum 90% reduction
in HC and CO by the 1983 model year. A minimum 75% decrease in NOx
is also required by 1985. These reductions are measured from
gasoline-fueled "heavy-duty" vehicles produced in "baseline" model
years: 1969 for HC and CO, and 1973 for NOx. These reductions
closely parallel those required for light-duty vehicles. However,
EPA is allowed to give manufacturers more time to apply the appro-
priate technology to "heavy-duty" vehicles than Congress itself
gave for manufacturers of light-duty vehicles. As a result of this
legislation, the top portion of EPA's light-duty truck class, the
6,000-8,500 pound GVWR portion, is Congressionally defined as
"heavy-duty". As such, it is required to meet the statutory
reductions set forth above. (EPA is meeting the statutory require-
ments for the remainder of the Congressionally defined "heavy-duty"
class in a separate rulemaking.)
The Clean Air Act provides a mechanism for dealing with the
situation which would arise if EPA later found that compliance with
standards based on the minimum reductions in HC, CO and NOx set by
the Act could not be achieved on schedule. The Act allows EPA to
revise the 90%-reduction HC and CO, and 75%-reduction NOx standards
towards less stringency, but only if the EPA Administrator finds,
. . . that compliance with the emission standards other-
wise applicable for such model year cannot be achieved by
technology, processes, operating methods, or other alterna-
tives reasonably expected to be available for production for
such model year without increasing cost or decreasing fuel
economy to an excessive and unreasonable degree.
(Section 202(a) of the Clean Air Act)
Revised standards may apply only for a period of three years, after
which either more stringent revised standards or the original
statutory standards must be established.
In addition to this mechanism for revising the statutory
standards should they prove infeasible, the Clean Air Act provides
for a system of nonconformance penalties. Such penalties would be
paid by a manufacturer who, by necessity or by choice, produces
vehicles which do not comply with the statutory standards. In
exchange, the manufacturer would be permitted to certify and sell
such nonconforming vehicles. EPA must make this system operational
only if the Administrator does not determine the statutory stan-
dards to be practicable.
The law is completely mute on either a definition or standards
for vehicles which, because their GVWR is less than 6,000 pounds
are not "heavy-duty" vehicles, and, because of the court decision
which required EPA to remove them from the light-duty vehicle class
are not light-duty vehicles. These vehicles, the bottom portion of
the light-duty truck class as EPA's regulation define it, are
still included in the general statutory authority described above.

The general statutory authority also includes the "heavy-duty"
class during the period until 1983 for HC and CO and until 1985 for
NOx. For model years prior to 1983, the Clean Air Act gives
guidance as to how the general authority is to be applied to the
"heavy-duty" class.
The Administrator shall prescribe regulations . . . applicable
to emissions of carbon monoxide, hydrocarbons, and oxides of
nitrogen from classes or categories of heavy-duty vehicles . .
. manufactured during and after model year 1979. Such regu-
lations applicable to such pollutants from such classes or
categories of vehicles or engines manufactured during model
years 1979 through 1982 shall contain standards which reflect
the greatest degree of emission reduction achievable through
the application of technology which the Administrator deter-
mines will be available for the model year to which such
standards apply, giving appropriate consideration to the
cost of applying such technology within the period of time
available to manufacturers and to noise, energy, and safety
factors associated with the application of such technology.
(Section 202(a) of the Clean Air Act)
For model years 1983 and 1984, only the very general guidance
applicable to all emission standards applies (in addition to the
minimum 90% HC and CO reductions).
All motor vehicle emission standards established under the Clean
Air Act must apply to vehicles or engines for their "useful life."
The Act directs the EPA Administrator to prescribe regulations
under which the useful life will be determined. The Act constrains
the useful life of vehicles in the light-duty truck class (both
above and below 6,000 pounds GVWR) to be 5 years or 50,000 miles
(or the equivalent), whichever first occurs, unless the Admini-
strator determines that a longer period is appropriate.
B. Description of Final Rulemaking
1. Revised and New Emission Standards
EPA has established HC and CO exhaust emission standards for
light-duty trucks for the 1983 and later model years and has chosen
to implement the statutory 90 percent reduction in HC and CO
emissions. A "baseline" emissions testing program was designed and
conducted as a means of finding the level of emissions from
earlier, uncontrolled LDT's. Using this baseline, the resulting 90
percent reduction standards are 0.8 g/mi for HC and 10 g/mi for
After the completion of testing and the proposal of these
standards, we discovered that the contractor which conducted the
testing had adjusted a number of carburetors improperly. After ten
of the trucks were readjusted and tested again, a recalculation of
the average baseline emission results revealed that the properly

adjusted vehicles produced a somewhat "cleaner" baseline than the
original sample. The finalized standards, .8 g/mi for HC and 10
g/mi for CO, represent reductions of 88 percent for HC and 87
percent for CO from the final baseline. Because of the delays that
would be involved in reproposing the standard at a lower level,
combined with the relatively small environmental benefit which
might be realized, the standards are being finalized as proposed.
EPA also sets new standards for emissions of CO under idle
operation. EPA set the level of this standard to be equal to 10%
of a sales-weighted average of idle emissions of 1969 heavy-duty
engines. EPA has tested a sample of 1969 heavy-duty engines._3/
Since there is no reason to suppose that 1969 light-duty trucks
have significantly different idle emissions, it would be redundant
to test the latter when the heavy-duty engine baseline will be
available. Based on testing the heavy-duty testing the idle
standard will be approximately as follows:
Pollutant	Level
CO	0.47 percent at
curb idle
EPA also promulgates a zero-emissions standard for crankcase
emissions (blowby) from diesel LDTs.
The revised HC and CO standards and idle standards, as they
apply to the 6,000 to 8,500 lbs. GVWR portion of the LDT class are
mandated by statute. The HC and CO standards and idle standards,
as they apply to the under-6,000 lbs. GVWR portion and the diesel
crankcase emission standard, are promulgated under the general
authority of the Clean Air Act.
EPA does not revise the fuel evaporative emission standard of
2.0 grams per test now scheduled to take effect in 1981, the 1982
exhaust particulate standard (diesel vehicles only) of 0.6 grams
per mile, nor the prohibition against crankcase emissions from
gasoline-fueled vehicles. The practical effect of these standards
on the design of light-duty trucks may be affected by other aspects
of this action, notably the revisions to the AQL and the existing
definition of "useful life".
2. Idle Test Procedures
Accompanying the idle standard are testing procedures for
measuring vehicle idle emissions. Identical to the idle test
procedures published in the heavy-duty regulations (45 FR 4136),
the procedures are less restrictive than the standard Federal Test
The recently adopted "short test" for idle emissions from
light-duty trucks (45 FR 34802) is compatible with this test
procedure. Differences between the two test procedures allow

the certification procedure to be performed using the CVS/bag
sampling equipment employed for the standard Federal Test Proced-
ure. The certification procedure also uses somewhat more stringent
instrument specifications than the short test to allow for more
precise repeatability of certification data than is needed for a
field test procedure. Vehicles passing the certification test
procedure should also pass the short test.
3. Kevised Definition of "Useful Life"
EPA is amending the current definition of "useful life1' for
light-duty trucks. The amendment would bring the period of use
specified in the definition into closer agreement with the periods
of use actually seen by light-duty trucks before they are scrapped
or their engines are replaced or rebuilt. The proposed definition
states that the useful life of a light-duty truck is the average
period of use of the truck before vehicle retirement or engine
replacement or rebuild. The manufacturer will determine what this
period of use is in terms of years, miles, or hours of operation.
However, the minimum useful lifa will be 5 years or 50,000 miles
(the minimum required by the Clean Air Act) or the period of the
baaic mechanical warranty covering the engine assembly, whichever
is greater.
The amended definition will apply to the warranty, recall,
and certification provisions of the Clean Air Act. That is,
manufacturers will be required to furnish owners with Section
207(a) and 207(b) (once 207(b) is implemented) warranties covering
the period of use specified in the amended definition. A manufac-
turer would also be liable for recall of a category of its light-
duty trucks if the EPA Administrator determines that a substantial
number of the category does not conform to the emission standards
during that period. And the longer useful life definition will be
incorporated into the certification procedures via deterioration
factors, as described in the next subsection..
A. Revised Certification Requirements Regarding Durability
EPA had proposed a substantially revised durability test
procedure with this rulemaking. However, EPA is delaying the
finalization of this proposed procedure to optimize all com-
ponents of the program. A revised durability test procedure is
expected to be implemented in conjunction with the statutory
Light-Duty Truck NOx emission standard.
Beginning in 1983, and until finalization of a revised dura-
bility test procedure, the burden of durability testing will be on
the manufacturers. The manufacturers will determine their deter-
ioration factors in programs which they design and submit these
deterioration factors to SPA as part of the certification process.

5.	Revised Requirements Regarding Maintenance
EPA is revising the existing requirements regarding main-
tenance that may be performed on test vehicles and recommended
to purchasers of LDTs. The revision will deregulate a number of
maintenance items for which EPA currently prescribes minimum
service intervals. Manufacturers will be free to recommend any
service intervals for these items, but will still be required to
follow their own recommendations when maintaining test vehicles.
Not all maintenance items are to be deregulated, however. Emis-
sion-related maintenance items, the neglect of which has an
immediate effect on emission levels or can have irreversible
effects on other parts of the emission control system, will con-
tinue to be regulated. The manufacturer will be required to show
that it's service intervals for these maintenance times are techno-
logically necessary before it may perform the items on its test
vehicles or recommend them to its customers.
6.	Selective Enforcement Audit Program (SEA), Production
Compliance Auditing (PCA) and Nonconformance Penalties
(NCP )
The SEA program is an assembly-line emissions testing program
used to aid in ensuring that light-duty trucks produced meet
the emissions level to which they are certified. The process of
conducting an SEA begins with a test order issued by EPA to a
manufacturer. Each manufacturer will be assigned a limit on the
number of test orders which EPA may issue during a model year,
based on its projected annual sales. The goal of SEAs is to
ascertain whether or not the production LDTs tested meet a 10
percent Acceptable Quality Level. An AQL of 10 percent allows for
emissions measurement error and quality control aberrations which
can not be totally eliminated at the assembly-line. Therefore, a
10 percent AQL in effect requires every LDT to meet applicable
emission standards.
Failure of an SEA audit may result in suspension or revocation
of the certificate of conformity for that configuration. To have
the certificate reinstated subsequent to a suspension, or reissued
subsequent to a revocation, the manufacturer must demonstrate, by
passing a follow-up SEA audit, that improvements, modifications, or
replacement have brought the original LDT configuration, the
modified LDT configuration, or its replacement into compliance.
The regulations include hearing provisions vfriich allow the manu-
facturer to challenge EPA's suspension or revocation decision based
on application of the sampling plans or the manner in which the
tests were conducted.
In the NPRM EPA had proposed a mechanism for production
compliance audits (PCA) and nonconformance penalties (NCP) appli-
cable to LDTs above 6000 pounds GVWR. Since the proposed emission
standards were considered feasible for all manufacturers to meet,

NCPs were not to be made available in the system as proposed. In
finalizing the rulemaking package, EPA continues to find the
standards feasible for all manufacturers. The role of NCPs in this
situation, however, is still under review by the Agency. There-
fore, the PCA/NCP portions of the proposal are not being finalized
in the present rulemaking.
C. Organization of the Regulatory Analysis
This Regulatory Analysis presents an assessment of the envi-
ronmental and economic impacts of the proposed light-duty truck
regulations. It provides a description of the information and
analyses used to review all reasonable alternative actions.
The remainder of this document is divided into five major
sections. Chapter III presents a general description of light-duty
trucks, a description of the manufacturers of this equipment, and
the market in which they compete.
An assessment of the primary and secondary environmental
impacts attributed to the proposed LDT regulations is given in
Chapter IV. The degree of control reflected by the promulgated
standards is described and a projection of air pollutant emission
factors for the national LDT population, with the promulgated
standards in place through 1999, is presented. The impacts of
these regulations on urban emissions and the expected air quality
benefits are considered. Secondary effects on other air pollutant
emissions, water pollution and noise are also discussed in this
section, as are commitments of scarce resources and impacts on
urban areas.
An examination of the cost of complying with the regulations
is presented in Chapter V. These costs include those incurred to
develop and install emission control equipment on light-duty
trucks, the costs to certify, and any increased vehicle operating
costs which are expected to occur. Analysis is made to determine
aggregate cost for the 1983-87 time frame. Finally, the impact
that this regulation will have on industry and consumers will be
reviewed, including the impact on the general level of prices in
the economy.
Chapter VI will identify and discuss the alternatives to the
proposed action, their expected impacts, and the reasons none has
been adopted instead of the actual final regulations.
Chapter VII will present a cost effectiveness analysis of the
final action and compare the results of this analysis with those
done on other mobile source control strategies.

EPA Contract No. EPA 68-03-2683, "Baseline Characterization of
Emissions from Medium-Duty Gasoline Vehicles Tested on a
Chassis Dynamometer," with EG&G Automotive Research, Inc.
A more complete discussion of the baseline program is con-
tained in Issue K - "Numerical Standards and Standard Deriva-
tion," found in the Summary and Analysis of Comments document.
T. Cox, G. Passavant, and L. Ragsdale, "1969 Heavy-Duty Engine
Baseline Program and 1983 Emission Standards Development," EPA
report, May 1979.

A. Description of Light-Duty Trucks
1. Definition of Light-Duty Trucks
The light-duty truck (LDT) class includes all motor vehicles
which have a gross vehicle weight rating (GVWR) of 8,500 pounds
(3,546 kg) or less, have a vehicle curb weight of 6,000 pounds
(2,722 kg) or less, have a basic vehicle frontal area of 45 square
feet (4.3 square meters) or less, and which are: (1) designed
primarily for purposes of transporting property or are derived from
such vehicles, (2) designed primarily for transporting persons and
have a seating capacity of more than 12, or (3) are available with
special features enabling off-street or off-highway operation and
use. Heavier light-duty trucks are those with GVWRs greater than
6,000 but less than or equal to 8,500 pounds.
Initially, Federal regulations classified all vehicles with
GVWRs of 6,000 pounds or less as light-duty vehicles (LDVs);
vehicles with GVWRs greater than 6,000 pounds were subject to
heavy-duty engine (HDE) requirements. Beginning with the 1975
model year, non-passenger car vehicles with GVWRs of 6,000 pounds
or less were reclassified as light-duty trucks. This ruling
resulted in an increase in the number of vehicles certified with
GVWRs greater than 6,000 pounds, since these vehicles would then be
subject to standards less stringent than those applicable to the
LDT class. As a result of this migration, EPA expanded the LDT
class GVWR ceiling to 8,500 pounds, effective for the 1979 model
year. The above definition reflects this change.
The proposed emission regulations for 1983 and later model
year light-duty trucks contain a definition of what is meant by
the phrase: "special features enabling off-street or off-highway
operation and use." This is meant to include those vehicles
which: (1) have four-wheel drive, and (2) have at least four of
the following characteristics (calculated when the vehicle is at
curb weight, on a level surface, with the front wheels parallel to
the vehicle's longitudinal centerline, and with the tires inflated
to the manufacturer's recommended pressure): (a) an approach angle
of not less than 28 degrees, (b) a breakover angle of not less than
14 degrees, (c) a departure angle of not less than 20 degrees, (d)
a running clearance of not less than 8 inches, and (e) front and
rear axle clearances of not less than 7 inches each (see Figure
The automotive and truck industries have traditionally used
GVWR categories for vehicle classification purposes. Historically,
classes I and II have included vehicles with GVWRs between 0 and
6,000 pounds, and between 6,001 and 10,000 pounds, respectively.

Figure III-A
FRONT	8 ~~ breakover ANGLE

Since all vehicle production and sales data are still reported on
the basis of these categories, an estimate must be made concerning
the percentage of those trucks with GVWRs between 0 and 10,000
pounds which also have GVWRs greater than 8,500 pounds. Based upon
production data for recent years, EPA has estimated this percentage
to be approximately 5.5 percent. It is anticipated that this
figure will increase to roughly 13 percent in the next few years as
manufacturers re-rate some of their vehicles to GVWRs greater than
8,500 pounds in order to make them subject to less stringent
emission and fuel economy standards.2/
2. Use of Light-Duty Trucks
Light-duty trucks are produced in a wide variety of body-types
encompassing a wide variety of possible functions. Virtually all
light-duty trucks have two axles and four wheels, and most are
equipped with gasoline-powered engines and two-wheel drive. A
small number make use of diesel engines, and an increasing per-
centage are equipped with four-wheel drive. The three largest
categories of light-duty trucks are, in order: pickups, vans, and
utility vehicles. Together, these categories comprise approxi-
mately 95 percent of all U.S. light-duty truck sales (see Table
III-A). The remaining sales include station wagons built on truck
chassis, passenger carriers, and multi-stop vehicles.
Pickups have an enclosed cab, with varying amounts of seating
and storage space. Behind the cab is an open, flat load-bed, with
a hinged rear gate. Pickups can be equipped with caps which
enclose the load-bed or with camper units of varying size. The
pickup category includes three major types: conventional, compact,
and car-type pickups. The compact pickups are, so far, nearly all
either foreign or captive imports. Car-type pickups have a cab and
front end similar to those of a passenger vehicle.
Vans have an enclosed load area which is typically connected
with the driver's compartment, and can be used for transporting
cargo and/or passengers, and for personal and recreational pur-
poses. Most vans have very short hood lengths, allowing improved
visibility and maneuverability.
The third major category of light-duty trucks is composed of
general utility vehicles such as the Cherokee, Scout II, Blazer,
Bronco, etc. These vehicles, in general, are capable of trans-
porting both passengers and cargo, and of pulling fairly large
trailers. Approximately 95 percent of the utility vehicles sold in
1978 were equipped with four-wheel drive.3/ Table III-A shows a
breakdown of U.S. new truck retail deliveries by body type for the
years 1976 to 1978.
Since they are available in a variety of styles, and because
of their versatility, light-duty trucks are capable of being put to

Table III-A
U.S. New Truck Retail Deliveries by Body Type, 1976-1978
	(GVWR < 8,500 pounds)	
Body Type
Conventional Pickup
Compact Pickup
Car-Type Pickup
Van and Cut-Away
Station Wagon
(Truck Chassis)
Passenger Carrier
Source: Estimated from Ward's Automotive Yearbook.

a wide range of uses, both private and commercial. Most can be
used for transporting either heavy and/or bulky loads or moderate
numbers of passengers. Some, with various possible configurations,
can perform both functions (such as a passenger van with removable
seats). Because of their heavy construction, LDTs are often
well-suited for trailer-towing. Also, many LDTs are capable of
rough, off-road use and operation under adverse driving conditions,
particularly when equipped with four-wheel drive. In general, LDTs
are better able to perform these functions than are passenger cars,
which have limited load and passenger-carrying capacity, are
lighter in construction, have less-powerful engines, and are
usually confined to on-road use. Clearly, these characteristics
make LDTs much more attractive and capable than passenger cars for
performing certain functions.
Private uses can include such activities as personal transpor-
tation, moving and hauling, travel, sport and recreation, etc.
Commercially, LDTs are used for such purposes as delivery of goods
and services, public and personal transport, moving and hauling,
trailer-towing, off-road service, etc., and find such uses in
a variety of businesses and industries. Table III-B shows a
percentage breakdown of light trucks by major use in 1972.
An important consideration with respect to the use of LDTs is
the use and non-use of their capabilities. Clearly, many of the
needs currently met by LDTs could not be reasonably met by other
means or modes of transportation, and it can be expected that this
will continue to be the case. In many other instances, however,
consumers may buy a LDT because of anticipated requirements it can
fill, but may then fail to make efficient use of its capabilities.
For instance, a buyer, attracted by its utility, versatility, and
durability, might purchase a 1-ton pickup when he may have only
occasional light hauling to do. In this case, a smaller pickup may
be sufficient to meet the user's requirements, with greater use of
the vehicle's capability. As economic and energy constraints
become more pronounced, it is likely that consumers will tend to
purchase LDTs more suited to the jobs they will be required to
perform. This trend would result in higher sales of smaller,
lighter, more fue1-efficient LDTs, with (consequently) lower
B. The Light-Duty Truck Industry
1. Structure
Light-duty trucks can be divided into two primary categories:
domestic and imported. There are five manufacturers of LDTs in the
United States; these are: General Motors (GM), Ford, Chrysler,
American Motors (AMC), and International Harvester (IHC). With the
exception of International Harvester, these manufacturers also
produce light-duty vehicles (passenger cars), resulting in much
common technology between the two types. General Motors and
Chrysler each operate two LDT-producing divisions: for GM,

Table III-B
Percentage Breakdown of Light Truck Class by Major Use, 19
	(GVWR j 10,000 pounds)	
Use	%
Personal Transportation
Wholesale and Retail
For Hire
Forestry and Lumbering
All Other
Source: Census of Transportation, 1972, Truck
Inventory and Use Survey: U.S. Summary; U.S.
Bureau of the Census.

Chevrolet and GMC; and for Chrysler, Dodge and Plymouth. American
Motors operates the Jeep division. International Harvester, which
also produces a large number of heavy-duty vehicles, markets its
LDTs under the name of Scout. The domestic manufacturers' relative
market shares (see Table III-C) have been rather stable over the
past few years, although the penetration by foreign manufacturers
has been increasing significantly.
The two foreign manufacturers with the largest sales of LDTs
imported for sale under their own names are Toyota and Nissan
(Datsun). Another company, Toyo Kogyo, manufactures and sells a
moderate number of trucks under the Mazda name. Toyo Kogyo, Isuzu,
and Mitsubishi manufacture large numbers of trucks for sale by
domestic companies (captive imports). Other foreign companies
which produce LDTs for sale in the United States are Suzuki and
Volkswagen; these companies have, so far, had small sales or have
only recently entered the market. Nearly all of the trucks pro-
duced by these companies, as either foreign or captive imports, are
compact pickups. A large number of LDTs are manufactured in
Canadian plants of domestic companies and imported into the United
States; these trucks will be considered as part of the domestic
2.	Sales and Revenues
Financial sales data show the largest of the domestic LDT
manufacturers to be General Motors, followed by Ford, Chrysler,
International Harvester, and American Motors. Table III-D shows
for each of these companies their total sales, net income, and
average total number of employees (1978 data). It must be recog-
nized that these figures are company-wide totals, and not just
those pertaining to LDT production.
The only company showing a net loss for 1978 was Chrysler,
with a loss of 204.6 million dollars. Because of its weak finan-
cial condition, Chrysler has sought, with some success, government-
sponsored financial aid.
As was stated previously, International Harvester, although a
manufacturer of light- and heavy-duty trucks, is not a manufacturer
of automobiles. There has traditionally been a large carryover of
technology from the light-duty vehicle industry to the light-duty
truck industry. Because of this, IHC lacks the broad technological
base which is available to the other LDT manufacturers, all of whom
also market automobiles. This situation was a primary factor in
NHTSA's decision to subject IHC LDTs to fuel economy standards less
stringent than those applicable to the rest of the industry for
3.	Employment
In addition to financial data, Table III-D shows corporate
average employment levels in 1978 for the five domestic LDT

Table III-C
U.S. LDT Sales by Manufacturer, 1978
	(GVWR i 8,500 pounds)	
Manufacturer	Sales
GM	1,434,011
Ford	1,152,610
Chrysler	382,046
AMC	154,553
1HC	34,081
Total Domestic	3,157,301
Toyota	94,882
Nissan (Datsun)	93,336
Toyo Kogyo (Mazda)	4,708
Other	17,340
Total Imported	210,266
TOTAL	3,367,567
Source: Estimated from Ward's Automotive Yearbook, 1979
Automotive News, 1979 Market Data Book Issue.

Table III-D
1978 U.S. Vehicle Manufacturer Information
Company	Total Sales ($)	Net Income ($)	Employees
AMC	2,585,428,000	36,690,000	27,517
Chrysler	13,618,300,000	-204,600,000	157,958
Ford	42,784,100,000	1,588,900,000	507,000
GM	63,221,100,000	3,508,000,000	839,000
IHC	6,664,350,000	186,680,000	95,450
Source: Moody's Industrial Manual, 1979.

manufacturers. Because of the high degree of integration between
the LDV and LDT industries (except in the case of IHC), only an
estimate can be made regarding the number of employees involved in
the production of LDTs. This estimate was made as follows: the
corporate employment figures of GM, Ford, Chrysler, and AMC were
totaled (1,531,000), and this figure was divided by the total
number of vehicles produced by these four companies (18,640,700); a
ratio of 0.082 employees per vehicle was obtained. This is a rough
approximation of the number of employees required for the produc-
tion of each vehicle. This same ratio of employees to vehicles was
assumed to be applicable to LDT production. Multiplying this
figure by the total number of LDTs produced in the United States
during 1978 (3,083,647) yields an industry-wide employment estimate
of approximately 253,000. A similar analysis indicated an approx-
imate yearly payroll of 5.5 billion dollars for the U.S. LDT
industry in 1978. (Base data for this analysis was obtained from
Ward's Automotive Yearbook, 1979.)
C. Light-Duty Truck Sales
1. Historical Sales
The past few years have seen a tremendous increase in LDT
sales. This growth has been greater than that for light-duty
vehicles. From Table III-E, it can be seen that from 1974 to 1978,
LDT sales increased by nearly 58 percent, while passenger car sales
increased by about 27 percent. The growth in total LDT sales can
be attributed to an increase in the sale of heavier LDTs (those
with GVWRs greater than 6,000 pounds). Since 1974, sales of LDTs
with GVWRs less than or equal to 6,000 pounds have actually de-
creased, as Table III-F shows, while sales of trucks with GVWRs
between 6,001 and 10,000 pounds have increased by 37 percent.
A closer look will now be taken at some of the constituents of
these sales figures. First, California LDT sales will be con-
sidered. Table III-G shows California and 49-state sales for
1973-1978. These figures are for heavy as well as light-duty
trucks, but it is assumed that the percentages would be nearly the
same for only light-duty sales. As can be seen, California sales
as a percentage of total U.S. sales decreased from 1973 to 1977,
but showed a slight increase in 1978. Sales projections for 1980
indicate an increase to 9.4 percent (see Table I1I-J).
Sales broken down on the basis of domestic, captive, and
imported LDTs are shown in Table III-H for the years 1975-1978.
These figures show that sales in each of the three categories
increased significantly during this time period.
Another way in which to consider LDT sales is on the basis of
manufacturer. Table III-I shows LDT sales by manufacturer for the
period 1974-1978. With the exception of IHC, all LDT producers
increased their sales during these years.

Table III-E
U.S. New Car and Truck Sales, 1974-1978
Est imated.
Source: Ward's Automotive Yearbook, 1979; Automotive News, 1979
Market Data Book Issue.

Table III-F
U.S. Truck Sales, 1974-1978
Year	0-6,000 lb GVWR	6,001-10,000 lb GVWR
1978	1,143,064	2,408,269
1977	1,218,094	1,903,103
1976	1,284,876	1,439,103
1975	1,204,259	895,758
1974	1,616,309	639,689
Source: Automotive News, Market Data Book Issues.

Table III-G
California and 49-State Truck Sales, 1973-1978
California Sales
as Percent of
Source: Ward's Automotive Yearbook, 1979, p. 178.

Table III-H
U.S. LDT Sales, 1975-1978
(rounded to nearest hundred)
Year	Total, 0-8500	lb* Domestic	Captive	Import
1978	3,356,000	3,005,000	140,700	210,300
1977	2,949,600	2,630,400	133,300	185,900
1976	2,574,600	2,334,900	100,300	139,400
1975	1,984,500	1,744,000	102,800	137,700
* Estimated.
Source: Ward's Automotive Yearbook; Automotive News, Market Data
Book Issue.

Table III-I
U.S. LDT Sales by Manufacturer, 1974-1978
(estimated for GVWR S 8,500 pounds)

* Includes non-captive imports.
Source: Automotive News, Market Data Book Issue.

Several important factors become apparent from these data. A
slump in vehicle sales {both car and truck) is clearly noticeable
for the years 1973 to 1975. This is most likely due to the com-
bined effects of an economic slowdown and gasoline supply and price
problems. These inter-related factors can have considerable
influence on the automotive industry. Also of interest is the
remarkably rapid growth of LDT sales in the late 1970's. A large
factor in this surge must be the sudden "popular" image of LDTs,
particularly of vans and sport/utiLity vehicles. An increased
number oE LDTs were being used for sport and recreation purposes.
Another trend worthy of consideration is the increased sales of
heavier LOTs. While lighter IDT sales dropped from 1974 to 1978,
heavier LDT sales rose significantly. An obvious spur to this was
the re-rating of some vehicles to higher GVWs in order to circum-
vent emission and fuel economy regulations.
2. Projected Sales
It can be expected that factors which influenced LDT sales in
the past will continue to do so in the future. Primary among these
will be energy availability and price. In the past two years,
gasoline prices have practically doubled, and supply has at times
been a problem.. This undoubtedly has an effect on the sale of
LDTs, whose fuel economy is lower than that of LDVs. By all
indications, 1979 LDT sales will be down considerably from 1976
levels* Also of primary importance are economic influenced.
Historically, economic slow-downs and recessions have had a pro-
nounced impact on the automotive industry. This will obviously
continue to be the case.
Other factors, possibly of secondary importance, include fuel
economy and emission standards, re-rating, and social influences.
Manufacturers may find it necessary to shift production toward
smaller, lighter LDTs j if not to meet government regulations, then
perhaps to meet increased public demand Cor more efficient vehicles
and to compete with foreign manufacturers. There will, of course,
always be some demand for larger LDTs by those whose needs can not
be met by smaller trucks. In addition, manufacturers are likely to
re-rate some of their trucks in order to make them subject to less
stringent regulations. Social factors will also continue to be
felt by the LDT industry. IE a negative image of truck ownership
and use develops, for whatever reason, a significant decline in
sales could result. Conversely, a positive image would be likely
to have a boosting effect on sales.
Givea the dependence of LDT sales on a variety of complex
variables, the difficulties in forecasting such sales can be
recognized. A number of such efforts have been undertaken, with
wide variations in results. EPA's projections, based on informa-
tion from Data Resources Inc., NHTSA, previous EPA studies, and
historical data, are presented in Table III-J. Although total LDT
sales are expected to increase through 1987, domestic production is
anticipated to level off around 1985, while import, captive, and

Table III-J
LPT Sales Projections (millions)
Total U.S. Sales

0-10,000 lb
Dome stic
Import/Capt ive

GVW 2/
Gas LPT 5/
LPT 4/
49 States (non-California)
0-8500	Domestic	Import/Captive
GVW 6/	Gas LPT	LPT 7/	LDPT 8/
1983	2.84	2.29	.29	.26
1984	3.13	2.50	.32	.31
1985	3.35	2.62	.34	.39
1986	3.49	2.64	.36	.49
1987	3.61	2.63	.37	.61
1988	3.79	2.61	.39	.79
T7Pata Resources Inc. Long Term Review, Fall 1979.
2/ 87 percent of the 0-10,000 lb. GVW sales projections. This 87
percent figure was taken from the Rulemaking Support Paper, Light-
Duty Truck Fuel Economy Standards Model Years 1982-1985, Pecember
1979,	NHTSA, POT.
3/ Figures used are based on percentages from the Regulatory
Analysis of the Light-Puty Piesel Particulate Regulations February
1980,	OMSAPC, EPA.
4/ Using 1980 projected sales, imports and captives represented
12.5	percent of total sales.
5/ Total 0-8500 - (Imports/Captives + LPPT).
6/ Using 1980 sales projections data, non-California sales were
90.6	percent of all LPT sales.
7/ Using 1980 sales projections, non-California sales of captives
and imports was 74.2 percent of the total.
8/ Using 1980 sales projections, non-California sales of standard
size light trucks was 92.9 percent of the total.

diese1-powered truck sales continue to increase. An increase
in import and captive truck market shares would seem to suggest a
trend toward LDTs with lower GVWRs.
Unexpected events and trends could easily shift sales from the
projected values, but these figures represent EPA's best estimate
based on available data and the most likely scenario of develop-
ments for the next several years.
D. Other Considerations
1.	Diesel Engine Penetration
A large unknown with respect to LDTs is their future use of
diesel engines. Diesels are attractive in our current fuel-
conscious era because of their significant gas-mileage improvements
over gasoline-fueled engines, as well as for their high durability
and their relative ease of maintenance. Diesels are also rela-
tively low in HC and CO emissions. They do, however, have moderate
levels of NOx emission and high levels of particulate emission.
Concerns about these latter two factors are preventing current
large-scale dieselization of the LDT fleet. Possible adverse
health effects of particulate emissions are currently under in-
vestigation. However, as Table III-K indicates, manufacturers
have already increased the use of diesel engines in LDTs (along
with LDVs and HDVs). The largest increase is seen to be in the
lighter LDT category. If the problems of NOx and particulate
emissions can be satisfactorily resolved, coming years are likely
to bring considerable growth in the use of diesel engines for LDTs.
Table III-J shows EPA's estimate of this growth for 1983-1987.5/6/
2.	Fuel Economy Standards
Another consideration relevant to LDTs is their fuel economy.
Gas mileage has become an item of primary concern for buyers of
LDVs, and it can be reasonably expected that this will be reflected
in LDT sales as well. NHTSA is applying fuel economy standards of
increasing stringency to LDTs. Table III-L shows current and
proposed standards (the figures for 1983 to 1985 are proposed
ranges within which the finalized standards are likely to fall).
The limited production line category includes only International
Harvester, which was deemed unable to satisfy the standards set for
the rest of the industry in 1980 and 1981. It is EPA's belief
that decreased vehicle emissions and increased vehicle fuel
efficiency are not incompatible goals.
3.	Emission Standards
Presented in Table II-B are past, present, and proposed
Federal and California emissions standards for LDTs. California
has adopted standards based on vehicle inertia weight which are
stricter than the comparable Federal standards. EPA's final 1983
LDT standards are within the capability of manufacturers applying

Table III-K
U.S. Factory Sales of Diesel Trucks by GVW
Year	0-6,000 lb	6,001-10,000 lb
1978	35,019	990
1977	2,392	1,128
1976	-	1,596
1975	-	1
Source: MVMA Motor Vehicle Facts and Figures, 1979, p. 14.

Table III-L
Fleet Average Fuel Economy Standards
Model Year
1983	(proposed)
1984	(proposed)
1985	(proposed)
Imports Other
16.0	16.0
16.7	16.7
Captive	Limited
Imports Other Production
14.0	14.0	14.0
15.0	15.0	14.5
Source: Rulemaking Support Paper, Light Truck Fuel Economy Stan-
dards, Model Years 1982-1985; NHTSA, DOT; December, 1979.

state-of-the-art technology. It is expected that manufacturers
will make use of much of the technology developed initially for the
LDV class. This includes such methods and devices as: exhaust gas
recirculation, turbocharging, catalytic converters, air injection,
and electronic controls. New devices such as trap-oxidizers for
particulates, and high-technology engines are currently being
researched and developed.7/
4. High-Altitude Standards
High-altitude areas are defined to be those elevations greater
than 4,000 feet (1,219 meters) above sea level. Outside of Cali-
fornia, there are 112 U.S. counties located substantially above
4,000 feet in elevation. In 1977, LDT sales in high-altitude areas
comprised approximately 5.5 percent of national LDT sales.
Vehicles which, after being designed and adjusted for use at
low altitudes, are operated under high-altitude conditions experi-
ence a degradation in emissions performance. In particular,
emissions of hydrocarbons and carbon monoxide are significantly
higher for maladjusted vehicles at high altitudes than at low
altitudes. For this reason, and since air pollution in a number of
high-altitude areas (primarily urban) is a serious problem, EPA has
considered and proposed measures affecting vehicles destined for
sale and/or use at high altitudes.
The Clean Air Act Amendments of 1977 require that, beginning
in 1984, all vehicles be able to comply with emission standards at
all altitudes. EPA has proposed that for the 1982 and 1983 model
year, vehicles be able to meet the standards applicable to the
altitude at which they are sold and be modifiable to meet the
standards at either high- or low-altitude. The proposed high-
altitude standards which 1982 and 1983 model year LDTs would be
required to meet when tested at a reference altitude of 5,400 feet
(1,650 meters) are: 2.0 g/mile HC and 26 g/mile CO in 1982, and 1.0
g/mile HC and 14 g/mile CO in 1983. The NOx standards would be
equivalent to those applicable at low-altitude. In addition,
EPA has proposed that manufacturers be required to provide adjust-
ment instructions for improving high-altitude emissions performance
for 1968 and later model year light-duty vehicles and trucks.8/9/

1/ Federal Register, Gaseous Emission Regulations for 1983 and
Later Model Year Light-Duty Trucks; EPA; Thursday, July 12,
2/ Rulemaking Support Paper: Light Truck Fuel Economy Standards,
Model Years 1982-1985; Office of Automotive Fuel Economy
Standards, NHTSA, DOT; December, 1979.
3/ Ward's Automotive Yearbook, 1979, p. 135.
4/ Preliminary Regulatory Analysis: Light Truck Fuel Economy
Standards, Model Years 1982-1985, NHTSA, DOT, December, 1979.
5/ Neil M. Szigethy, "Will Diesels Dominate?" Fleet Specialist;
May/June, 1979, pp. 31-39.
6/ Regulatory Analysis: Light-Duty Diesel Particulate Regula-
7/ Draft Regulatory Analysis: Proposed Emission Regulations for
1983 and Later Model Year Light-Duty Trucks; ECTD, OMSAPC,
EPA; June 28, 1979.
8/ Draft Regulatory Analysis: Environmental and Economic Impact
Statement for the Proposed 1982 and 1983 Model Year High-
Altitude Motor Vehicle Emission Standards; SDSB, ECTD, OMSPAC,
9J Federal Register, Control of Air Pollution from New Motor
Vehicles and New Motor Vehicle Engines, Submission of Altitude
Performance Adjustments; EPA; Thursday, January 24, 1980.

A. Background
The Clean Air Act as amended in 1970 contained many provi-
sions aimed at removing harmful pollutants from the air we breathe.
Among other things, the 1970 Act called for the establishment of
National Ambient Air Quality Standards. These levels were to be set
such that there would be no danger to public health and welfare.
To date, ambient air quality standards have been set for seven
pollutants: particulate matter, sulfur dioxide (SO2), carbon
monoxide (CO), nitrogen dioxide (NO2) , ozone (of which hydro-
carbons (HC) is the main precursor), hydrocarbon (HC), and lead
(Pb). Mobile sources are major contributors to the emissions of
all of these pollutants except SO2. This regulation package
concerns the establishment of standards for HC and CO from light-
duty trucks.
Both HC (in its role as an ozone precursor) and CO emissions
have been related to adverse health effects. Detailed information
on the health affects of HC and CO will not be discussed in depth
in this Regulatory Analysis since such information is well docu-
mented elsewhere.^/ Briefly, HC emissions react with sunlight to
form ozone and other photochemical oxidants. Ozone is a pulmonary
irritant that affects the respiratory mucous membranes, other lung
tissues, and respiratory functions. CO when inhaled replaces
oxygen in the blood. The presence of CO adversely affects the
carrying and delivering capacity of oxygen by the blood.
Although significant improvements have been made in air
quality since 1970, a review of air quality monitoring data makes
it clear that additional reductions in HC and CO emissions will be
necessary if ambient air quality goals set by Congress in the Clean
Air Act are to be achieved. On March 3, 1978, EPA published in the
Federal Register a listing on a State-by-State, pollutant-by-
pollutant basis, of the attainment status of every area of the
Nation (43 FR 8962). This information, compiled by the respective
States and reviewed by EPA, was the most accurate picture available
of the nation's air quality status as of the adoption of the Clean
Air Act Amendments. These data indicated that of 3215 counties or
county equivalents covered by those designations, 607 (19 percent)
were classified as nonattainment for photochemical oxidant, and 190
(6 percent) were classified as nonattainment for carbon monoxide.
Nonattainment status indicates that the given area fails to meet
the primary national ambient air quality standard (NAAQS) for the
pollutant under consideration based upon either direct air quality
monitoring or indirect estimates for areas lacking monitoring
data. Current non-attainment data is available to indicate the
changes which have occurred since 1977. As of July, 1979, the
non-attainment designations include 586 (18 percent) counties for
ozone and 164 (5 percent) for carbon monoxide.

Since the U.S. population is not uniformly distributed, but
rather is concentrated in urbanized areas, the above geographically
based figures are not representative of the proportions of popu-
lation actually exposed to excessive ambient pollutant concentra-
tions. Indeed, it is the very fact of urbanization which has led
to many of our air pollution problems. For example, the nonattain-
ment areas for ozone include 103 out of a total of 105 urban
areas in the U.S. with populations greater than 200,000 (the
exceptions being Honolulu, Hawaii, and Spokane, Washington). The
103 areas represent an exposure of over 100 million people.
Clearly, there is a great need to reduce pollutant (or pollu-
tant precursor in the case of ozone) emissions in the urban areas
of the U.S. So long as large numbers of people continue to be
exposed to concentrations in excess of the NAAQS, further emission
reductions must be sought.
Mobile sources have been recognized for some time as major
sources of hydrocarbons (ozone precursors) and carbon monoxide.
Light-duty vehicles in particular have been the focus of consider-
able control work since the late 1960's. However, as light-duty
vehicle emissions grow smaller, other source categories such as
light-duty trucks grow in proportional significance. The wisdom
of controlling light duty truck emissions is evident when these
emissions are placed in the context of other sources of these same
In order to properly assess mobile source emissions and their
control, it is best to look at urban areas where historically the
NAAQS contraventions have occurred. In this way a truer perspec-
tive of the air quality impact of mobile sources can be obtained.
It is in these urban areas that improvements are most needed.
The selection of the areas to analyze will be discussed in detail
below in Section 2. The HC analysis will be done on an Air
Quality Control Region (AQCR) basis. CO, on the other hand, will
be analyzed on a county basis. This is due to the more localized
nature of CO problems. Fifty seven AQCRs have been selected for
HC, and 50 counties for CO. Hydrocarbons analyzed include only
non-methane hydrocarbons since the methane fraction is non-reac-
t ive.
Figures IV-A and IV-B present breakdowns of non-methane
hydrocarbon (NMHC) and CO emissions into various source categories
for the selected regions. These figures give the 1976 emission
levels along with projected levels out to 1999. The data presented
in these figures represent what is considered the base case. That
is, it projects future light-duty truck emissions as if no new
regulations beyond those already in existence were promulgated.
For other source categories, known future control programs are
included. For example, heavy-duty trucks are projected based
upon the 1984 implementation of the regulations finalized in the
January 21, 1980 Federal Register (45FR 4136).2/ The base case
also assumes the successful implementation of i7m programs (since
the analysis is of non-complying regions).

14,000 .
12,000 .
(1000 tons)
10,000 .
4,000 .
Figure IV - A
Annual Non-Methane Hydorcarbon
Emissions for 57 Urban Regions

W , J. V





18,000 .
16,000 „
14,000 .
(1000 tons)
10,000 -
6,000 .
4,000 .
Figure IV-B
Annual Carbon Monoxide
Emissions for 50 Urban Counties
1990	1995	1999

For non-methane hydrocarbons, mobile sources currently repre-
sent approximately 33 percent of the urban emissions (Fig. IV-A).
With the regulations already in effect this percentage is expected
to decline to 16 percent by 1995.
Mobile source carbon monoxide emissions currently repre-
sent over 80 percent of the urban emissions (Figure IV-B). This
amount is expected to decline to 53 percent by 1999. No signi-
ficant change in stationary source emissions is expected for
CO. However, since CO problems are often attributed to high
localized concentrations during periods of high traffic density,
stationary sources have minimal impact on CO air quality prob-
lems .
Light-duty vehicles (passenger cars) contribute the major
portion of mobile source NMHC and CO emissions. The 1976 emis-
sion levels from light-duty vehicle and other mobile sources,
and projections of the future urban emissions are given in Figures
IV-C and IV-D. Again, these projections are for the base case
of no new light-duty truck regulations. The figures give a
general overview of the contribution to air pollution that each
class of vehicles is expected to make through 1995, and of the
distribution of the burden of control of emissions from all mobile
sources. From these figures it can be seen that emissions from
light-duty trucks will grow in proportion to emissions from heavy-
duty trucks and light-duty vehicles. This apparent inequitable
distribution of the burden for reducing mobile source emissions
can be in part accounted for by the past need to concentrate
control efforts on light-duty vehicles where potential gains were
the highest.
It is evident from the figures that for both NMHC and CO,
light-duty trucks represent a growing proportion of emissions.
For hydrocarbons, light-duty trucks go from 14 percent of the
total in 1976 to 22 percent in 1999. For carbon monoxide the
figures are 13 percent in 1976 and 29 percent in 1999. Thus,
control of light-duty trucks is extremely important in any overall
strategy for reducing emissions sufficient to meet ambient air
quality standards. The remainder of this chapter will address the
environmental impact which would result from imposition of the
light-duty truck emission control strategies considered as part of
this rulemaking.
B. Primary Impact
1. Emission Factors
Considerable work has been done within EPA in an attempt to
determine accurate emission factors for mobile sources. This work
depends heavily on in-use vehicle testing under EPA's Emission
Factor Program. To answer the question of how well vehicles
perform in actual use, EPA has administered a series of exhaust
emission surveillance programs. Test fleets of consumer-owned

4,000 .
3,000 .
(1000 tons)
2,000 .
HD Diesel
HD Gas
LD Truck
LD Vehic
Figure IV-C
Annual Mobile Source
Non-Methane Hydrocarbon
Emissions for 57 Urban Regions
1980	1985




(1000 tons)
HD Diese:^HJ 1%
HD Gas
LD Truck
LD Vehicle
Figure IV-D
Annual Mobile Source
Carbon Monoxide Emissions
For 50 Urban Counties

vehicles within various major cities are selected by model year,
make, engine size, transmission, and carburetor in such proportion
as to be representative of both the normal production of each model
year and the contribution of that model year to total vehicle miles
traveled. These programs have focused principally on light-duty
vehicles and light-duty trucks.
The data collected in these programs are analyzed to provide
mean emissions by model-year vehicle in each calendar year,
change in emissions with the accumulation of mileage, change
in emissions with the accumulation of age, percentage of vehicles
complying with standards, and effect on emissions of vehicle
parameters (engine displacment, vehicle weight, etc.). These
surveillance data, along with prototype vehicle test data, assembly
line test data, and technical judgment, form the basis for the
existing and projected mobile source emission factors. 3/
For this regulatory analysis, changes have been made to the
emission factors for heavy-duty and light-duty trucks. The emis-
sion factors found in the mobile source emission factors document
for heavy-duty vehicles are based upon steady-state data gathered
on the 9-mode and 13-mode test procedures. In the course of
developing the recently finalized heavy-duty engine regulations,
EPA has accumulated substantial data on the transient emissions of
heavy-duty engines. Both the CAPE-21 data gathering program and
resultant transient test procedure were designed to accurately
characterize in-use operation and therefore in-use emission.
Therefore, the available transient test data has been used to
revise the heavy-duty truck emission factors which are currently
being used. The emission factors for future heavy-duty engines
have also been revised to reflect accurately the final standards
and the implementation of Selective Enforcement Auditing with a 10
percent acceptable quality level. Refer to Appendix A of the
Regulatory Analysis accompanying the heavy-duty engine rulemaking
for details of the methodology and the calculations.4/
The light-duty truck emission factors need revision from those
presented in the emission factors document to reflect the imple-
mentation of parameter adjustment regulations in 1981 and 1982 and
to accurately reflect the impact of the final light-duty truck
regulations of this rulemaking. Revised emission factors for
pre-1983 light-duty trucks are those presented in "MOBILE 1 Modifi-
cations for the LDT Regulation Analysis." The factors contain
corrections for 1981 and 1982 light-duty trucks to account for the
implementation of parameter adjustment regulations in those model
year vehicles. Related adjustment of the computation methodology
for I/M credits for light-duty trucks to prevent double counting
was also done. This was accomplished by the relatively simple
mechanism of limiting the overall I/M plus parameter adjustment
credit to values characteristic of currently estimated benefits of
an I/M program applied to current (non-parameter adjustment)
vehicles. The limits used are 25 percent for HC and 35 percent for
CO. For 1983 and later light-duty trucks, emission factors have

been derived in Chapter VII using the methodology developed for the
heavy-duty rulemaking.
Light-duty trucks currently are powered almost exclusively by
gasoline-fueled engines. Only three diesel engine families are
currently certified for light-duty trucks. However, this is
expected to increase substantially in future years. For example,
EPA has projected dieselization of light-duty vehicles to reach 20
percent by 1991.5/ Based upon the similarity of usage between
light-duty trucks and light-duty vehicles, the same estimated
diesel fractions will be used for light-duty trucks. In order to
correctly assess the emissions impact of light-duty trucks, this
diesel fraction must be accounted for, since diesel engines have
different emission rates than gasoline-fueled engines. In the
analysis of air quality impacts, the gasoline-fueled engine emis-
sion rates and diesel engine emission rates will be combined
according to their respective sales fractions to give a single
emission rate. The reason for such combination is that the base-
line emission inventory used for the EKMA and Modified Linear
Rollback models (see discussion below in section 2 and 3) does not
contain separate categories for gasoline-fueled and diesel light-
duty trucks.
Estimates of diesel light-duty truck emission rates which will
be used in this analysis are based upon rather limited data. Table
G-2 of Issue G - Technological Feasibility, of the Summary and
Analysis of Comments, contains 1980 diesel light-duty truck certi-
fication data. Also presented in that table are results for the
light-duty vehicle version of the GM light-duty truck engine.
Because of 1980 light-duty vehicle standards, the light-duty
vehicle version of the engine incorporates redesigned injectors and
EGR which result in lower emission rates. The average results for
the families presented in that table are shown below:
Engine Family	Average Emissions
Manufacturer (cubic inch displacement)	HC CO NOx
GM LDT 09J9Z	.76 2.0 2.0
GM LDV	03J9ZG	.27 1.15 1.6
(Oldsmobile)	(350)
IHC	SD-33T	.42 1.9 1.5
VW	DP	.32 .90 1.1
In computing light-duty truck emission rates, the GM light-duty
truck emission rates will be replaced by those of the light-duty
vehicle for 1983 and beyond. Such improvements will be necessary
to meet the light-duty truck standards.

Th e engine families do not represent equal sales fractions for
1980. Therefore, the question of sales weighting arises. Cur-
rently the GM diesel represents some 76 percent of the diesel
sales. However, as more manufacturers introduce more diesel model
light-duty trucks, that share will surely drop. Since the VW and
IHC engines probably represent typical small and medium sized
engine emissions, and the GM a large engine, we will weight each
equally. Deterioration factor data for these engines is insuffi-
cient to establish whether the diesel deterioration rates will be
somewhat lower than those for gasoline-fueled engines as might be
expected. Therefore, the more conservative assumption that they
are not will be used. Deterioration factors of 1.4 for HC and 1.3
for CO over 100,000 miles have been derived for gasoline-fueled
light-duty trucks and will be applied to the diesel vehicles as
Using the starting emission rates found by averaging the GM,
IHC, and VW engines shown above (and substituting the Oldsmobile
data for 1983 and beyond), along with the deterioration factors as
just described, the light-duty truck diesel emission rates become:
Prior to 1983
HC - 0.5 + 0.02(M/10,000)
CO - 1.6 + 0.05(M/10,000)
1983 and Beyond
HC - 0.34 + 0.01(M/10,000)
CO - 1.3 + 0.04(M/10,000)
Here M corresponds to accumulated vehicle mileage. The HC and
CO deterioration factors have been expressed as equivalent additive
The general form of all the emission factors for mobile
sources is an equation with some starting new vehicle emission rate
plus a mileage dependent deterioration rate. This means that to
determine the emissions from a given vehicle one must know the
accumulated mileage. To determine the average emission rate for
the fleet made up by a given class of vehicles (for example,
light-duty trucks), it is necessary to account for the fact that
the on-the-road fleet consists of a mix of vehicles of varying ages
and model years. The appropriate emission rate is applied to each
fraction of the fleet and the fractions are summed into a com-
When vehicles meeting a new emission standard are introduced
into the on-the-road fleet, they at first represent only a small
fraction of the whole fleet. As time passes, the newer technology
vehicles come to represent a larger and larger share of the entire
fleet. This means that the composite emission rate for the entire
fleet will show a gradual change in response to new standards,
rather than a sudden change.

2.	Lifetime Emissions
One way to examine the effect of the rulemaking action is to
compare the emissions of vehicles' built to meet the requirements of
the rulemaking with the emissions of earlier vehicles. Using the
emission factor equations, the total lifetime emissions of a given
model year vehicle may be estimated. This will be done for 1969
(the "baseline" model year for deriviation of the standard), 1982
(representing vehicles built the year prior to the new standards),
and 1983 (year of implementation for this rulemaking) model year
vehicles. The calculations will use average vehicle lifetimes of
120,000 miles for light-duty trucks.6/ Lifetime per-vehicle
average emissions are given in Table IV-A.
The impact of the new standards on vehicles produced for 1983
(or later) is clearly evident in this data. Compared to emissions
from 1982 vehicles, 1983 vehicles are reduced 76 percent for HC
and 82 percent for CO in the case of gasoline-fueled vehicles. For
diesel engines HC is reduced 37 percent and CO is reduced 20
percent between 1982 and 1983 vehicles.
With reference to the 1969 baseline year, gasoline-fueled
light-duty truck HC emissions are reduced 89 percent and CO
emissions are reduced 91 percent in 1983 vehicles.
3.	Reduction in Urban Emissions From Light-Duty Trucks
We have seen that as new light-duty trucks are put into use
and older ones retired, the emissions of the average light-duty
truck on the road will decrease. The resulting composite emis-
sion factors can be used to project changes in annual emissions
from the entire fleet. The same can be done for other mobile
source categories as well. To make the projections, the changes in
composite emission rates are used along with estimated growth rates
in total vehicle miles traveled to modify the baseline emission
inventory for future years. Projections are also made of changes
in stationary source emission rates depending on present and
anticipated stationary source control programs.7/
For hydrocarbons, the exhaust emissions themselves are an
indirect rather than a direct problem. That is, the principal
harmful effect of HC emissions stems from the photo-chemical
reactions leading to ozone formation. The reaction process can
take several hours, by which time the pollutants involved are
transported and dispersed over broad areas. Therefore, the hydro-
carbon emissions have been analyzed on an Air Quality Control
Region (AQCR) basis. The AQCR's selected were those non-Califor-
nia, non-high-aLtitude regions violating the ozone standard (or
estimated to be violating where actual sampling data is missing) in
a 1975-1977 base period. California regions were excluded since
California has its own emission standards. High altitude regions
were excluded because the emissions data used in the analysis is

Table IV-A
Lifetime Emissions for Light-Duty Trucks (Tons)
Model Year
Class + Pollutant	1969 1982 1983
Gasoline fueled
HC	0.86 0.38 0.09
CO	9.9	5.0	0.9
HC	N/A	0.08 0.05
CO	N/A	0.25 0.20

not considered representative of high altitude conditions. A
separate detailed analysis would have to be done to assess the
impact of these regulations on high-altitude areas. This selection
process led to a set consisting of 57 AQCRs to be analyzed for
hydrocarbons. In addition, because methane emissions are non-
reactive and do not contribute to ozone formation, the emission
inventories compiled for analysis will be based upon non-methane
hydrocarbons (NMHC).
Carbon monoxide emissions, in contrast to hydrocarbons,
frequently create localized problems of high concentrations. These
are often associated with urban core areas experiencing high
traffic densities. It is desirable, therefore, to analyze CO on
a more localized basis than AQCRs. This has been done by using a
county based inventory. As for HC, only non-California non-high-
altitude areas were selected. In addition, regions in Alaska have
been excluded. A significant part of th^ CO problem in Alaskan
regions appears to be related to high CO emissions from vehicle
operated at low temperatures. The emission factors used in this
analysis are not representative for consistent low temperature
operation. The result is 50 counties exceeding the CO standard for
a 1975-1977 base period.
Following the selection of areas to be analyzed, an emission
inventory for each region was compiled. The most recent year for
which complete information could be obtained was 1976. This data
then forms the basis for future projections. Compilation of the
baseline and projection for future years is an involved process
entailing a number of assumptions. These are discussed in detail
in supporting documents.7/8/ Two assumptions are important to
highlight here. The first is the assumption that light-duty
vehicle and light-duty truck i/M programs will be implemented in
all the areas analyzed by 1982. Since all the areas chosen are
areas exceeding the HC and CO standards, such programs are expec-
The second assumption concerns projected growth rates for
various source categories in future years. For non-methane hydro-
carbons, rollback projections were made for a range of growth
rates. The high and low end of these ranges differ by one or two
percent. For this analysis we will use the growth rates of the low
growth option. For mobile sources these rates appear most consis-
tent with what appears likely because of energy costs and related
matters. The high growth assumptions would increase the absolute
levels of emissions and decrease the absolute levels of air quality
benefits projected by the models somewhat. They would, however,
make little difference in the relative change from the base case to
the control case. The maximum air quality benefits would peak in
1995 rather than in 1999 if the high growth case were chosen. For
heavy-duty vehicles, other specific adjustments in growth rates are
also required. Annual vehicle miles traveled (VMT) are expected to
decline for gasoline-fueled engines by about 2 percent per year,
while diesel VMTs will increase by about 5 percent per year. These

rates reflect increased use of diesel engines in the heavy-duty
industry, largely because of energy considerations.4/
Projections for both emission data and air quality data are
made on an AQCR by AQCR basis (or county by county for CO).
However, the underlying assumptions on emission factors are not
region specific. Rather, they represent typical nationwide values.
Because of this, only average results for all regions will be used
for analysis.
Figure IV-E and IV-F provide a comparison of the projected
mobile source emissions for the base case of no new light-duty
truck regulations with the projected emissions for the final
regulations. They cover the years from 1990-1999. By 1990,
light-duty trucks would account for 22 percent of mobile source
NMHC and 26 percent of mobile source CO emissions. The substantial
reductions in light-duty truck emissions expected are clearly
indicated. For HC, in 1999 the reduction reaches 55 percent.
For CO, in 1999 the reduction is 62 percent. These percentages are
measured in comparison to the. base-case emissions for the same
year, 1999.
Expressed as a percentage of all mobile source emissions, the
impact of the final rulemaking is as follows. Hydrocarbons are
reduced 12 percent in 1995 and 13 percent in 1999. Carbon monoxide
is reduced 17 percent in 1995 and 21 percent in 1999.
4. Ambient Air Quality Impact of Regulation
Using the emission rates previously discussed, an anal-
ysis was done of the air quality impact in each of the selected
regions.8/ The Modified Rollback method was used for oxidant and
CO to project future air quality improvements for each region. In
addition, the Empirical Kinetic Modeling Approach (EKMA) was also
used for oxidant. The EKMA procedure has been developed by EPA in
an attempt to provide an improved analysis of the relationship
between oxidant and precursor emissions while avoiding the com-
plexity of photochemical dispersion models.j)/ There is uncertainty
over the applicability of EKMA, so that both EKMA and rollback were
used to provide a range of possible air quality impacts.
In preparing the air quality projections, baseline emission
rates for various source categories were taken from the National
Emissions Data System (NEDS), and projections for future control
strategies plus growth rates were made. In combination with the
mobile source projections, this data allowed an evalutation of air
quality improvements to be expected. With both Modified Rollback
and the EKMA approach, the relative changes from strategy to
strategy are more reliable than predictions of absolute levels of
air quality. Therefore, the results will be expressed as percen-
tage gains over baseline between various strategies. In addition,
although the individual regions used in the analysis can be identi-
fied, the results are not considered accurate enough to be used for

(1000 tons)
HD Diesel
HD Gas
LD Truck
LD Vehicle
0 1
(a) (b)
Figure IV-E
Annual Non-Methane Hydrocarbon
Emissions For Baseline Case
and Control Case
(57 Urban Regions)
Note: Bar (a) = Baseline
Case of no new light-duty
truck regulations.
Bar (b) = Control case with
new light-duty truck regulations.
(a) (b)

3,500 -
(1000 tons)
1,500 .
HD Diesel
HD Gas
LD Truck
LD Vehicle
(a)	(b)
Figure IV-F
Annual Carbon Monoxide
Emissions for Baseline Case
and Control Case
(50 Urban Counties)
(a)	(b)
(a)	(b)
Note: Bar (a) = Baseline
Case of no new light-duty
truck regulations.
Bar (b) = Control case with
new light-duty truck regulations,

a region by region review of the regulations. Rather, averages
over all areas analyzed will be used. The average air quality
improvements are given in Table IV-B.
The modified linear rollback and EKMA models differ by a
factor of nearly 2 to 1 for ozone reductions. However, they each
indicate similar percentage gains from implementing the new stan-
dards. For the 1990-1999 period, improvements of 1 percent to 2
percent in ozone are projected to result from implementing the
light-duty truck regulations.
Table IV-B indicates that carbon monoxide will be improved 3
percent in both 1990 and 1995, and 4 percent in 1999.
The significance of a percentage gain in air quality in terms
of progress toward attainment of standards depends upon the origi-
nal levels. For example, a 2 percent improvement in air quality
may be sufficient to bring a region that is already close to the
standard into compliance, whereas in a region experiencing very
high levels (relative to the standard) that 2 percent would re-
present a totally inadequate reduction. In a region already
meeting the standards, such a further gain would increase the
margin for compliance. The question could then be posed: "How many
areas originally exceeding air quality standards are brought into
compliance by implementing the new emission standards?" In Table
IV-C the air quality improvements are analyzed in this fashion.
Considering the ozone results first, the difference in
absolute reductions predicted by modified rollback versus EKMA
noted in Table IV-B are again apparent. While modified rollback
indicates that 96-98 percent of the regions originally violating
the ozone standard will come into compliance in the 1990's, EKMA
puts that percentage at 72 percent. Therefore, as noted earlier,
caution must be used in interpreting results from either model in
absolute terms. For example, the indication from modified rollback
that nearly all violating regions will meet the ambient ozone
standard by 1999 should not be considered reliable. Rather, the
relative change attributable to implementation of the new regu-
lations is the item of maximum accuracy. The table indicates that
implementation of the light-duty truck regulations will result
in approximately a 0 percent (EKMA) to 2 percent (rollback) reduc-
tion in the number of violating regions.
The cautions noted for ozone are equally important in inter-
preting the CO results in Table IV-C. Only rollback applies to
this case, and that model indicates that with either strategy, all
regions analyzed will attain the CO standard by 1990. However, it
has already been noted that it is not within the ability of this
model to accurately predict absolute air quality levels. There-
fore, the indication of all regions meeting the standard is incon-
clusive. As an illustration of the accuracy required to accept the
absolute projections, in the final rollback projections for 1999
only 94 percent of the regions are in compliance with the standard

Table IV-B
Average Air Quality Percent Reductions
From 1976 Base Year
(Modified Linear Rollback/EKMA)
Base Case
Implement LDT


Base Case
Implement LDT

Table IV-C
Percentage of Regions Originally Violating
Air Quality Standards Brought Into Compliance
(Modified Linear Rollback/EKMA)
Base Case
Implement LDT


Base Case
Implement LDT

by a margin of greater than 20 percent for the base case. For the
control case, that result changes by 4 percent to a value of 98
percent. Inaccuracies on the order of 20 percent or greater are
more than possible in the present air quality analysis, and would
markedly change the absolute levels of predictions. However, such
inaccuracy would probably be relatively constant from strategy to
strategy and lead to consistent relative effects. Unfortunately,
since changes in air quality produced by the new regulations do not
become significant prior to 1990, no clear conclusions can be drawn
about the effect these regulations will have on attainment status.
However, as noted, based upon the number of regions within 20
percent of the standard, implementing the light-duty truck regula-
tions produces a 4 percent improvement.
C. Potential Secondary Environmental Impacts
1. Sulfuric Acid Emissions
A recent EPA reportlO/ provides an in-depth review of the
current status of sulfate emissions from mobile sources. On a
nationwide basis, mobile sources represent less than 2 percent of
the total man-made sulfur oxides. However, with the introduction
of the catalyst/air pump technology to control HC and CO emissions
from mobile sources, there exists the potential for a significant
source of mobile related sulfate emissions in the form of sulfuric
acid aerosol. While of negligible magnitude on a regional basis,
mobile source sulfuric acid emissions could produce a significant
localized urban sulfate concentration in urban street canyons, or
congested urban freeway situations. Moreover, mobile source
sulfates differ from stationary source sulfates in that they are
emitted in the form of a fine sulfuric acid mist and the particles
tend to remain near ground level.
The increase in sulfate emissions due to the use of oxidation
catalyst/air pump control systems on passenger cars and light-duty
trucks has been of considerable concern to EPA. In pre-model year
1975 non-catalyst systems, most of the fuel sulfur leaves the
vehicle after combustion as SO2. In oxidation catalyst/air
pump systems used on recent model year automobiles and light-duty
trucks, a small amount (less than 10 percent 10/) of the sulfur is
converted by the catalyst to SO3. The SO3 combines with water
in the exhaust to form sulfuric acid aerosol.
Extensive efforts have been made within government and indus-
try to improve the information about mobile source sulfate emission
factors, sulfate air quality modeling techniques and sulfate health
effects as a function of exposure level. In addition, technology
assessment work is proceeding to identify how sulfates are formed
in catalyst/air pump systems, and to develop other low sulfate
producing catalytic control systems such as the three-way catalyst.
According to current data, the extent of sulfate emissions is much
less than early concerns had anticipated. Major adverse health and
welfare effects from mobile source sulfates are unlikely.10/ Table

IV-D indicates sulfuric acid emission rates for several mobile
source categories.
Implementing the new light-duty truck standards is not ex-
pected to increase present mobile source sulfate emission or to
present a future problem. Catalyst systems are already in use on
light-duty trucks. Insofar as the 1983 standards might lead to
some increased use of three-way system there could be a decrease in
sulfate emissions.
2. Water Pollution, Noise Control, Energy Consumption
Complying with the light-duty truck regulations is expected to
have negligible impact on water pollution, or on the ability of the
light-duty truck manufacturers to meet present and future noise
emission regulations. Implementing these regulations can be done
with no fuel economy penalty. In fact, the analysis of fuel
economy impact done in the Summary and Analysis of Comments
indicates that there should be a net gain in fuel economy between
1982. and 1983. For futher discussion of fuel economy, the reader
is referred to Issue L of the Summary and Analysis of Comments.
D.	Irreversible and Irretrievable Committment of Resources
A small additional committment of platinum and palladium
will be required over and above that needed for current light-duty
trucks which already employ catalysts. This increase will result
from the need to improve catalyst durability and meet lower emis-
sion standards. The incremental demand in 1985 would be approxi-
mately 38,600 troy ounces of plantinum and 16,200 troy ounces of
palladium. These figures are based upon vehicle sales, catalyst
loadings and catalyst sizes developed in Chapter V and the Summary
and Analysis of Comments (Issue F - Economic Impact). In the event
that recycling of catalyst noble metals becomes economical in
future years, this incremental demand could be offset or elimin-
ated .
E.	Relationship of Short-Term Uses of the Environment to Mainte-
nance and Enhancement of Long-Term Productivity
More stringent control of light-duty truck emissions than
that currently imposed will result in substantial decreases in
hydrocarbon and carbon monoxide emissions from this source. This
reduction will be beneficial and aid in the long-term attainment
and maintenance of acceptable air quality.

Table IV-D
Approximate Mobile Source
Sulfuric Acid Emission Rates 10/
H2S04 Conversion	H2S04
Source Category	 	Rate (%)	(mg/mile)
Non-catalyst car	1	1
Oxidation catalyst car	10	10-15
3-way catalyst car	5	4
Light-duty diesel car	2	9
Heavy-duty diesel truck	2	50
Aircraft gas turbine	0.03	N/A

_1/ For a current review of this data, as well as citations to
other reports on health effects of HC and CO, see "Health
Effects of Exposure to Low Levels of Regulated Air Pollutants
- A Critical Review," Benjamin A. Ferris, Jr., M.D., Journal
of the Air Pollution Control Association, Vol. 28, No. 5, May
2/ For details on assumed future strategies for other source
categories see "Data Assumptions and Methodology for Assessing
the Air Quality Impact of Proposed Emission Standards for
Heavy-Duty Vehicles," EPA Air Management Technology Branch,
Office of Air Quality Planning and Standards, November 1979.
3/ A complete presentation of mobile source emission factors,
including future use projections, can be found in EPA-400/9-
78-005, "Mobile Source Emission Factors - Final Document,"
March 1978.
4/ "Regulatory Analysis and Environmental Impact of Final Emis-
sion Regulations for 1984 and Later Model Year Heavy-Duty
Engines," EPA Office of Mobile Source Air Pollution Control,
December 1979.
5/ "Summary and Analysis of Comments on the Notice of Proposed
Rulemaking for the Control of Light-Duty Diesel Particulate
Emissions from 1981 and Later Model Year Vehicles," EPA Ofice
of Mobile Source Air Pollution Control, October 1979. Table
6/ "Average Lifetime Periods for Light-Duty Trucks and Heavy-Duty
Vehicles," EPA Report SDSB 79-24, G. Passavant, November 1979.
TJ "Data Assumptions and Methodology for Assessing the Air
Quality Impact of Proposed Emission Standards for Heavy-Duty
Vehicles," EPA Air Management Technology Branch, Office of
Air Quality Planning and Standards, November 1979.
8/ "Air Quality Impact of Final LDT Emission Standards - SummaTy
of Results," EPA, April 1980.
9/ "Uses, Limitations and Technical Basis of Procedures for
Quantifying Relationships Between Photochemical Oxidants and
Precursors," EPA-450/2-77-021a, US EPA, Research Triangle
Park, NC, November 1977.
10/ "Emissions of Sulfur-Bearing Compounds From Motor Vehicles and
Aircraft Engines," Report to the United States Congress,
EPA-600/9-78-028, August 1978.

This chapter examines the costs associated with the 1983 LDT
emission standards and related control strategy being finalized in
this rulemaking. Costs of this rulemaking lie in three major
areas: development and testing, certification, and emission
control hardware. A few LDTs may experience a change in
operating costs as a result of these regulations.
The vehicle manufacturers must bear the initial costs of
development, certification, and^ emission control hardware. These
costs will be added to the truck selling price, marked up, and then
passed on to the ultimate purchaser.
Due to the complex nature of the LDT emission standards
program and the current dynamic nature of the LDT market a few
words of explanation are necessary.
Vehicles sold in California must certify to a different set of
emission standards than those sold in the remaining 49 states.
(See Chapter III). For this reason, this analysis will exclude
California sales from any analysis or costs. California sales of
LDTs are 8-10 percent of the total LDT market ._1/ EPA has used a
figure of 9.4 percent of all imports and captives.^/ Using these
figures and some data received from NHTSA in their recent LDT fuel
economy standards proposal EPA has developed sales projections for
the years 1983-1987. Table V-A gives both 50 states and Federal
only sales projection for the analysis period.
The light-duty truck market of the 1980's is very difficult to
project. Rising fuel prices and fuel economy pressures in general
have halted the sales boom of the late 1970's and led the market
into a very dynamic state. Increased dieselization and the effects
of fuel economy and exhaust emission standards will lead to sub-
stantial changes in the characteristics of the market over what is
present in 1980.
At least four areas should be addressed to explain the forces
acting on the market: dieselization, imports, new engine lines,
and fleet wide engine downsizing.
Although light-duty diesel trucks (LDDT) are only about 1.5 2/
percent of the current light-duty truck sales, both EPA and the
manufacturers expect this percentage to increase substantially in
the 1980's. In the light-duty diesel particulate rulemaking
action, EPA addressed this changing market share and arrived at the
following figures for LDDTs: 1983(8.9 percent), 1984(9.5 percent),
1985(11.4 percent), 1986(13.8 percent), 1987(16.5 percent). This
analysis used these percentages in the sales projections discussed

Table V-A
LPT Sales Projections (millions)



Domest ic


States (Exclude
s California)




See Chapter III for more detail.

The market share of imports and captive imports has increased
greatly in the last year. However, beginning in 1980 captive
imports are no longer used to determine compliance with fuel
economy standards, which may lead the corporate importers to slowly
replace their captive imports with their own smaller light trucks.
As a result of this possible action, the foreign manufacturers of
these captive imports may choose to open their own retail outlets
and sell under their own names instead of as captives. This
situation remains so uncertain that EPA will assume that whether
the light-duty trucks from Isuzu, Mitsubishi, and Toyo Kogyo are
sold as captive imports or pure imports, their market shares remain
relatively stable.
EPA expects that most, if not all, of the domestic manufac-
turers will be introducing new smaller light-duty trucks in the
1980's. NHTSA's rulemaking support paper outlines some of these
plans.3/ These new truck families will be used as replacements for
some fraction of the sales of the larger, less fuel efficient
engines. The anticipated introduction of these new truck families
further indicates the dynamic nature of the light-duty truck
Finally, in addition to the introduction of new truck lines
with smaller engines, EPA expects that domestic manufacturers will
adjust their sales mixes such that they sell fewer of their large
CID engines and more of the smaller more fuel efficient engines.
The 1980 sales projection data submitted by the manufacturers
breaks neatly into three specific cylinder/CID groups (see Table
below). However, EPA expects that in the mid-eighties a large
shift will occur in these percentages. The current and future
market splits are shown below:
Sales Splits
EPA Market
Number of	Engine	1980 Market	Projection
Cylinders	CID Range	Percentage	(1983-1987)
4	0-200	11%	15%
6	200-300	19	45
8	300-400	70	40
It is clear that the light-duty truck fleet sold in the
mid-eighties will have considerably different sales mix character-
istics than that certified for 1980. However, neglecting the
changes in sales mix, the emission control hardware changes which
would be necessary to the 1980 LDT fleet are representative of
those incremental changes which would be necessary on the new sales
mix of four, six, and eight cylinder engines sold in the mid-
eighties. This relevancy of the 1980 fleet will be used to later
predict the emission control hardware cost.

A. Cost to Truck Manufacturers
1. Development and Testing
EPA expects that manufacturers will incur development costs
related to redesigning their emission control related hardware for
the full useful life. It is difficult to estimate the actual
development and testing costs each manufacturer or vendor will
However, it is quite likely that the bulk of this redesign and
development cost will be borne in improving catalyst, EGR, air
injection, and electronic engine controls (EEC). As a means of
estimating this hardware redesign and development cost EPA will
conservatively use the same amount of research and development
investment that went into the initial components, but inflated to
1980 dollars. For catalysts ($5.11), EGR ($1.37), air injection
($1.23) and electronic engine controls (EEC) ($1.23).4/ These
figures sum to $8.94, but EPA will conservatively assume $10 to
cover any minor changes in material which might increase component
durability or optimization costs which might be desirable to
improve efficiency or decrease cost.
To allocate this development cost, EPA has relied on manufac-
turer's comments to the 1979 LDT standards. Based on the manufac-
turers comments, the development and testing costs will be alloca-
ted over a five year period with the bulk of the cost being in-
curred in the first three years. Using this methodology and a
development cost of $10 per truck sold, the costs are apportioned
according to the following schedule:
EPA expects that the 1983 standards will require more develop-
ment work than the 1979 standards primarily because of the in-
creased useful life period and the 10 percent AQL. For the 1979
standards an R&D cost of $90 million was estimated. EPA believes
the $165 million in development costs in this analysis is ample to
cover any redesign, development testing, or emission control system
optimization efforts undertaken.
2. Emission Control System Costs
a. Gasoline-Powered LDTs
EPA expects the manufacturers to continue the use of oxidation
catalyst/air injection/EGR systems to achieve compliance with the
1983 emission standards. In addition, EPA expects that manufac-
turers will use a form of electronic engine controls (EEC).

Specifically, EPA expects manufacturers will use EEC to
control spark timing and to modulate EGR when necessary. Con-
trolling spark timing during cold start will aid in the reduction
of HC and CO emissions by minimizing catalyst light-off time and
then permitting optional spark timing during other driving modes.
Some families may choose to replace their current EGR system with a
modulated EGR system to achieve any small NOx reductions which
might be required. The emission levels of the current LDT fleet
indicate that not all LDTs will require EEC and/or modulated EGR to
reduce emissions, however, this analysis has conservatively assumed
that all manufacturers use EZC.5_/
To determine the hardware related costs of complying with the
1983 emission standards and control strategy, EPA studied the
emission levels and emission control strategies used on all engine
families in the 1980 LDT fleet .5^/ Using the results of this
analysis EPA has estimated the costs of the hardware necessary for
each engine family to achieve 4,000-mile emission levels below the
expected average target levels.
These hardware costs generally fall into three major areas:
catalytic converter upgrades, air injection system upgrades, and
electronic engine controls (EEC).
The primary changes in catalytic converters are in the areas
of converter volume and noble metal loading. These changes will
increase catalyst efficiency and durability.
Air injection systems on about eight engine families will
have to be upgraded as a result of the more stringent emission
standards. Of the eight families affected, four will have to
replace pulse air systems with mechanical air pumps, three will
have to add a mechanical air pump where none is presently used, and
one will have to add a pulse air system. The improved air injec-
tion systems in these 8 engine families will be necessary to insure
increased oxidation of HC and CO during all driving modes.
EPA has assumed that all LDTs use EEC in 1983 to control spark
timing and modulate EGR. The EEC system will entail an electronic
control unit plus simple sensors for spark control and EGR modula-
tion. This is a very conservative approach since not all LDTs will
require EEC with spark control and modulated EGR.5/ Fuel economy
pressures may force the manufacturers to use EEC however.
The results of EPA's engine family by engine family analysis
are shown in Tables V-B, V-C and V-D for eight, six, and four
cylinder engines respectively. These tables show the expected
hardware cost increase for each family, but the family names have
been deleted to protect the confidentiality of the manufacturers
sales projections which have been used in the sales-weighting
process. The use of the current manufacturers sales projections
allows for the maintenance of the same relative market shares
throughout the period. The summary and analysis of comments

Table V-B
Emission Control Hardware Cost: 8 Cylinder
	1980 Dollars	
Percent of 8	Air Pump	Converter	Electronic Engine
inder Sales
Upgrade Costs
Upgrade Costs
Control Systems
Sales-Weight Cost per LDT: $77.41

Table V-C
Emission Control Hardware Cost: 6 Cylinder
1980 Dollars
Percent of 6	Air Pump
Family Cylinder Sales Upgrade Costs
1	12.64	$0
2	7.41	0
3	13.45	0
4	41.24	0
5	23.86	23
6	1.38	0
Converter Electronic Engine
Upgrade Costs Control Systems Total
$13	$60	$73
13	60	73
25	60	85
2	60	62
16	60	99
46	60	106
Sales-Weight Cost per LDT:

Table V-D
Emission Control Hardware Cost: 4 Cylinder
1980 Dollars
Percent of 4
Cylinder Sales
Air Pump
Upgrade Costs
Upgrade Costs
Electronic Engine
Control Systems
Sales-Weight Cost per LDT: $112,50
* Includes $24 in stainless steel exhaust and unleaded fuel restrictor costs
necessary due to the first time use of catalytic converter technology.

document supporting this rulemaking outlines the actual compliance
steps anticipated in each case.
The costs for these hardware changes was estimated using the
data and methodology in a cost estimation report prepared under
contract for EPA.4/ This report was used to estimate costs for
catalytic converter upgrades, air injection systems, and EEC (ECU
plus sensors). For the few families which require modulated EGR,
EPA has estimated that it would have approximately the same cost as
the current EGR system which it is replacing so no significant
increase in first cost would occur. This methodology was altered
by allowing for the effects of inflation and using more realistic
profit and overhead margins at the corporate and dealer level. The
inflation rate used was 8 percent per annum, which is slightly
greater than the new car CPI values for 1978 and 1979 (6.2 percent
and 7.4 percent respectively).^/ The overhead and profit margin
used is the same as in the recent heavy-duty engine rulemaking
action (11.A percent overhead, 17.6 percent profit).^/ Also, 1980
noble metal prices for platinum and palladum were used.
b. Light-Duty Diesel Trucks (LDDT)
In 1980 three manufacturers certified light-duty diesel
trucks: General Motors, International Harvester and Volkswagen.
Of the three truck families, only the General Motors family did not
meet all of the 4,000-mile target emission levels.
In its light-duty diesel particulate rulemaking action, EPA
estimated a cost of $30 per engine for this engine family to
reduce its gaseous emissions. This would involve the addition
of EGR, injector redesign and possibly some other minor engine
modifications. By 1983, all of the necessary changes will have
been made on GM's light-duty diesel passenger car fleet. There-
fore, the only substantial change in GM's LDDT fleet will be to
incorporate the changes into this engine family.
Light-duty diesel trucks will also have to comply with the new
diesel crankcase control requirements. Presently, only one family,
that from International Harvester, does not have a closed crank-
case. EPA has determined that the cost to close the crankcase is
about $6 per engine when a simple cyclonic separator is employed.8/
As stated previously, EPA expects that sales of LDDT will
increase substantially during the mid-eighties, The 1980 projected
sales of about 1.5 percent of the market is expected to rise to an
average of 12 percent of the market during the period 1983-1987.
EPA has reason to believe that other manufacturers will be intro-
ducing LDDT lines to gain a portion of this increasing market.
It is reasonable that the new light-duty truck engines intro-
duced in the mid-eighties will have emission characteristics
similar to the three families now produced. Some of them will be
inherently very clean and have no emission problem, but some

others will have problems with HC and NOx. As a means of estimat-
ing this market, EPA will use the same basic split as is present
today, thus meaning that one-third of the engines will require
gaseous emission reductions and one-third will require closing of
the crankcase. Under this scenario, a per engine cost of twelve
dollars is anticipated (,33($30 + $6)).
c. Fleetwide Emission Control Hardware Costs
Having now computed the hardware cost for gasoline and diesel
light-duty trucks to meet the emission target levels, it only
remains for these costs to be spread over the entire LDT fleet.
These costs will be allocated according to the scenario
developed in the first few pages of this chapter. That is, for
gasoline-powered LDTs EPA expects 15 percent 4 cylinder (less than
200 CID), 40 percent 6 cylinder (200-300 CID) and 45 percent 8
cylinder (greater than 300 CID). EPA is anticipating that light-
duty diesel trucks will average twelve percent of the market over
the period 1983-1987. Using this scenario, the per truck emission
control hardware cost can be determined as shown below.
Hardware Cost =.12($12)+.88(.45($77)+.40($77)+.15($113)) = $74
If the fleet wide sales mix were to remain at current levels (11%
4 cylinder, 19% 6 cylinder, 70% 8 cylinder) this cost would be
3. Certification Costs
Certification is the process in which EPA determines whether a
manufacturer's light-duty trucks conform to applicable regula-
tions. The manufacturers must prove to EPA its trucks are designed
and will be built such that they are capable of complying with the
emission standards over their full useful life. The certification
process begins by a manufacturer submitting a certification appli-
cation to EPA. Subsequently, two steps occur.
The first step involves the determination of preliminary
deterioration factors for the regulated pollutants. The light-duty
truck manufacturer may determine these preliminary deterioration
factors in any manner it deems necessary to insure that the fac-
tors it submits to EPA for certification purposes are accurate and
representative of the deterioration expected over the full useful
life. Manufacturers must state that their procedures follow sound
engineering practices and specifically account for the deteriora-
tion of EGR, air injection, and catalyst systems as well as other
critical deterioration processes which the manufacturer may identi-
fy. In addition, when applicable, the manufacturer must state that
the allowable maintenance intervals were followed in determining
the preliminary deterioration factors. The manufacturers would
submit preliminary deterioration factors, based on the revised
useful life definition, in each case where current certification

procedures require testing of a durability vehicle. Beyond these
requirements EPA would not approve or disapprove the durability
test procedures used by the manufacturers.
Step two involves emission data vehicles. Although the number
of emission data vehicles per family is not fixed, a reasonable
range is 2-8 with the 1980 average being approximately 4 per engine
family. These vehicles would be operated for 4,000 miles before
the emission test. The preliminary deterioration factor would be
multiplied times the 4,000-mile emission test results to predict
whether the emission data vehicles would meet the standards for
their full useful life. If the emission data vehicles are predict-
ed to pass the standards over the full useful life, then the engine
family is granted certification.
As can be seen in the discussion above, each engine family
will incur costs in two distinct areas: preliminary deterioration
factor assessment and emission data vehicles.
To determine their preliminary deterioration factors EPA
expects the manufacturers will use a procedure similar to that used
now. However, due to the new useful life definition EPA expects
that manufacturers will run their durability vehicles for 100,000
miles as opposed to the current 50,000 miles. As a general rule
manufacturers run from one to two durability vehicles per family.
The calculation of the industry's cost for finding these
preliminary deterioration factors under this assumption is shown in
Table V-E. The unit costs are based on EPA estimates of what
manufacturers have spent on testing durability data vehicles in the
past.9_/ The estimates were made in 1975, but have been adjusted
for inflation and the impact of the longer testing period. The
twenty-eight emission tests allow for testing at 5,000-mile inter-
vals during the expected 100,000 miles of test operation and eight
tests associated with maintenance. Based on the current number of
engine family control system combinations EPA expects there will be
about 50 combinations certified by all manufacturers in 1983.
Total industry costs of determining preliminary deterioration
factors are estimated at $31 million.
The testing of emission-data vehicles will not be affected by
regulations, except that carry-over of emission test results from
previous model years will be disrupted. EPA's method for estimat-
ing the cost impact of this disruption is to assume that no emis-
sion-data carry over is possible in the first effected year. This
overestimates the incremental impact in the first year, since 100
percent carry over would not have been possible in any case. But
the disruption also has an incremental impact in the second and
following years. EPA reasonably assumes that the various errors
cancel each other.
The calculation of the costs of testing emission-data vehicles
is shown in Table V-F. Again, unit costs are derived from EPA

Table V-E
Certification Costs Associated with Determining
Preliminary Deterioration Factors
1980 Dollars
Cost per Engine Family - Emission Control	System Combination
Prototype Vehicle	$ 35K
Mileage Accumulation to 100,000 miles	$263K
Maintenance and Overhead
Twenty-Eight Emission Tests	$ 12K
at $400 per test
Total Cost Industry Wide
Assuming fifty engine-system combinations will be tested
and two vehicles per engine family.
100 x $310K - $31.0M Total Cost

Table V-F
Certification Costs Associated with
Emission Data Vehicle Testing
Cost per Emission Data Vehicle
Prototype Vehicle	$13.8K
Mileage Accumulation to 4,000 miles,	$ 8.8K
Maintenance, and Overhead
Two Emission Tests at $400 per test	$ 0.8K.
Total Cost Industry Wide
Assuming 200 emission data vehicles from 50 families:
200 x $23.4K - $4.68M Total Cost

estimates made in 1975. EPA also estimates that 200 emission-data
vehicles will require testing, an average of four per engine
family-control system combination. (The ratio of emission-data
vehicles to combinations is not fixed by regulation, so there is
some variability in it from combination to combination and from
year to year.) Emission-data costs total about $4.68 million.
The total certification cost ($35.68M) may be overestimated by
as much as 20 percent if California does not adopt the revised
durability testing procedure. To be conservative EPA has included
the cost for recertification of the California families even though
actual California sales have been excluded from the other costs and
4. Selective Enforcement Auditing Costs (SEA)
In addition to the revised emission standards for 1983, EPA is
implementing changes in the LDT sampling system and acceptable
quality level. The costs associated with the SEA program can be
divided into two main categories: sampling system changes and 10
percent acceptable quality level costs (AQL).
a.	Sampling Plan Changes
The incremental costs of changing from a batch sampling plan
to a sequential sampling plan are negligible. There may be a small
cost decrease due to a slight decrease in the average number of
engines in an audit (at a 40 percent AQL).
b.	10 Percent AQL Costs
Incremental costs associated with going from a 40 to a 10
percent AQL lie in three areas: formal SEA testing costs, self
audit testing costs, and costs associated with meeting the lower
target levels.
Under the sequential sampling system the average number of
vehicles tested at a 40 percent AQL is sixteen (assumes 40 percent
non-compliance). However at a 10 percent AQL the average sample
number is thirteen (assuming 10 percent non-compliance).10/
So there would be a formal SEA testing cost decrease of at the very
least $1,200 per audit (assuming no vehicle break-in period). This
cost is small compared to others discussed in this chapter, so it
will conservatively be neglected.
Although EPA solicited manufacturers comment on any incremen-
tal increases in self auditing which may be required, only Chrysler
responded in the affirmative. They estimated a one time cost of
$1.7 million for equipment and $300,000 a year for testing. Since
Chrysler did respond in the affirmative their costs will be in-
cluded. Perhaps the reason for the lack of response from the
manufacturers is because they are already very close to achieving
the compliance levels necessary for the 10 percent AQL.11/ Cali-

fornia audit data shows non-compliance levels of 5.1 percent for
HC, 6.2 percent for CO, and 9.4 percent for NOx.
With the change to the emission standards and the useful life
definition it is difficult to determine precisely the incremental
hardware cost of going from a 40 percent AQL to a 10 percent
On a fleetwide basis, EPA expects the same hardware to be used
regardless of the AQL. Gasoline-powered LDTs will still use air
injection, EGR, oxidation catalysts, and electronic engine con-
trols. As outlined in the summary and analysis of comments docu-
ment, EPA expects that one engine family will have to add mechan-
ical air pumps instead of a pulse air system at a fleetwide per
vehicle cost of about $1 per LDT. In addition, the more stringent
HC and CO targets will probably force an incremental increase in
noble metal loading of not more than 0.1 grams of platinum on
average. On a fleetwide basis these costs are only about $2 per
truck. For diesel LDTs, the implementation of the 10 percent AQL
will force GM to add EGR to their 350 CID diesel at a cost estima-
ted at $15 per truck. Using the fleet projections discussed
earlier and taking engines of this type to be one-third of the
light-duty diesel truck sales in the mid-eighties, the per vehicle
cost on a fleetwide basis is about $.60 per engine. All of the
hardware costs discussed in this paragraph have already been
included in the emission control hardware costs in section A.
Incremental hardware costs of the 10 percent AQL are estimated at
$4 per LDT. These costs are relatively small primarily because the
change in emission targets is relatively small and the degree of
conformity during production is already at the levels required for
a 10 percent AQL.
5. Total Costs to Manufacturers
The four main costs to manufacturers (development and testing,
certification, emission control hardware, and SEA related expendi-
tures) are summarized in Table V-G. All costs are in 1980 dollars.
The total cost shown in Table V-G, $1,418 billion dollars (undis-
counted) provides sufficient funds for the manufacturers to deal
with all aspects of this regulatory strategy.
B. Costs to Users of Light-Duty Trucks
1. Increases in First Costs
The added cost to manufacturers for development and testing,
certification, SEA, and emission control system hardware will be
passed on to purchasers of light-duty trucks. The amount a manu-
facturer must increase the price to recover its expenses depends on
the timing of the costs and of the revenues from sales and on the
cost of capital to the manufacturer. Table V-G showed the manner
in which the manufacturers costs are distributed over the period
1981-1987. The cost of capital used is 10 percent, and EPA has

Table V-G
Total Costs to Manufacturers for Trucks
Produced During 1983-1987
1980 Dollars
Cost 1/
Cost 1/
Costs 1/
Emission Control
Hardware 2/
$ o
$ o
$ 55M
26 7M
Totals	$165M	$35.7M	$3.2M	$1215M	$1418.9M
17Fixed cost.
1} Each year's costs are approximately 1/3 fixed cost (tooling,
overhead, etc.) and 2/3 variable cost (material, labor, profit,

allowed the recovery of all fixed investment within five model
years. The expected average, sales-weighted first price increase
for 1983-1987 vehicles is $95 and is comprised of $17 for R&D, $4
for certification and SEA testing and $74 for emission control
hardware. The range of this first price increase varies from $61
to $203 for gasoline-powered LDTs and reflects the differences in
emission control hardware costs. For LDDTs the first price in-
crease ranges from $21 to $51.
2. Maintenance Costs
For the vast majority of the engine families and trucks
produced in the mid-eighties EPA is expecting no change in operat-
ing costs. However, two light-duty truck families will be using
catalyst technology for the first time in 1983 and thus will have
decreased operating costs associated with fewer exhaust system
replacements and fewer spark plug replacements. The use of unlead-
ed gasoline combined with material improvements in the exhaust
system will reduce maintenance costs associated with exhaust system
replacement. EPA estimates that approximately one entire exhaust
system replacement will be saved over the vehicle lifetime. Spark
plug life will be increased substantially over the current inter-
vals as a result of the use of unleaded fuel.
To estimate these cost savings, exhaust system and spark plug
replacement costs, the current and future spark plug replacement
costs, the current and future spark plug replacement intervals, and
a mileage accumulation rate for light-duty trucks must be known.
EPA ascertained exhaust system and spark plug replacement
costs using parts plus labor replacement cost estimates received
from several retail dealers. For a set of four spark plugs esti-
mates ranged from $6.80 to $7.68. EPA used the lowest point in
these estimates, approximately $6.80 per set. Exhaust system
replacement costs ranged from $150 to $210 for parts and labor.
Since at least one complete exhaust system will not have to be
replaced, EPA will conservatively use the lowest cost of this range
or $150 per replacement.
The current spark plug intervals for these two families are
both 15,000 miles.2/ The new maintenance interval is 30,000 miles.
For exhaust systems EPA has estimated that with the use of unleaded
gasoline only one replacement late in the sixth year would be
required and a second replacement could be eliminated. (see Table
V-H) .
Although the financial savings will not be computed or ac-
counted for in this analysis, one other potential savings does
exist. With the change from leaded to unleaded fuel accompanying
the use of catalyst technology, blowby emissions of lead into the
crankcase will also be eliminated. This in turn, should allow an
increase in the oil change interval, and thus savings to the owner
over the vehicle life.

Table V-H
Exhaust System and Spark Plug Savings
Exhaust System Replacement
Number of Spark Plug	With and Without

Average Annual
Replacements Based
on Intervals
Year 1/
Mileage 2/
Mileage 2/
15,000 miles - 30
,000 miles


TJ Year of light-duty truck usage.
21 Average Lifetime Periods for Light-Duty Trucks and Heavy-Duty Vechiles, US EPA, OMSAPC, ECTD, SDSB
79-24 Glenn W. Passavant, November 1979.

Ffnally^ t0 compute the discounted values of these savings,
the average mileage accumulation rate for light-duty trucks (<6000
Lbs. GVWR) must be known. This was taken from an EPA technical
report and is shown in Table V-H together with the exhaust system
and spark plug computations.
Using the data in Table V-H and the standard 10 percent
discount rate, the average spark plug and muffler savings is
estimated at $115 (discounted) per vehicle effected.
Since sales from these two engine families are only about four
percent of the total,2/ the average savings per LDT is only $4.61
3. Fuel Economy and Fuel Costs
One means of meeting the target emission levels is what
EPA would call the "quick fix technology" approach. This would
involve the use of the same basic emission control hardware
as is currently used; and achieving emissions reductions through
engine or emission control system calibrations and start cata-
lysts. This is the type of technology which some manufacturers
used to meet the 1980 California emission standards. If this
"quick fix technology" is used, then a 3 to 4 percent fuel economy
penalty is poss ible.12/
However, EPA has no reason to believe tliat this fuel economy
penalty need occur. If the manufacturers use the electronic engine
controls and upgraded catalyBt/air injection systems described in
the summary and analysis of comments and costed in this chapter,
the necessary emission reductions should be achievable with no fuel
economy penalty. If a manufacturer has trouble meeting the target
emission levels without a fuel economy penalty, then the option of
using a 3-way catalyst system is available. This 3-way system
would cost about $96 more per LDT than the emission control system
outlined previously for an 8 cylinder#engine.V EPA's assessment
Df the fuel economy impact of these regulations is reinforced by
NHTSA's analysis for their proposed light-duty truck fuel economy
standards for model years 1982-1985, which shows a fleetwide fuel
economy gain from MY 1982 to MY 1983.13/
The two LDT families which will be using catalyst technology
for the first time will also require more expensive unleaded fuel
for the first time. A long term unleaded-leaded price differential
of three cents per gallon is anticipated.14/
Using a sales-weighted average fuel economy of 21.6 miles per
gallon for these two engine families,^/ a lifetime period of
122,000 milealS/a and an unleaded fuel differential of three cents
per gallon, lifetime fuel coats will be increased by $114 (dis-
counted) for each of these light-duty trucks, on a per LDT basis
this cost is only $4.56 (discounted).

4. Total Costs to Users
To summarize, users of light-duty trucks can as a result of
these regulations expect to pay an average of $95 more for 1983
model year trucks than for comparable models purchased in 1982, in
1980 dollars. Operating costs will not increase for the average
vehicle. The purchasers of light-duty truck models which will
require unleaded fuel for the first time can expect their increased
costs for unleaded fuel to be offset by saving in maintenance
expenses for spark plugs and the exhaust system.
C. Aggregate Costs
The aggregate costs to the nation of complying with the 1983
Federal LDT emission regulations consist of the sum of increased
costs for development, new or upgraded emission control hardware,
certification costs and selective enforcement auditing at a 10
percent AQL. Two truck models will also require unleaded gasoline
for the first time, but this expenditure should be offset by
savings in maintenance costs related to spark plugs and the exhaust
system. All of these costs will be calculated for a five-year
period of compliance.
The five-year costs of compliance are dependent on the
number of light-duty trucks sold during the period. The accuracy
and validity of projecting vehicle sales as far into the future
as 1987 is problematic, so cost estimates based on such projec-
tions are subject to some qualification. However, because the
largest portion of the costs in this analysis are variable costs
(hardware) and not fixed costs (certification, R&D) the accuracy
of the sales projections is not as important as might be the
case in some other rulemaking actions. Future sales of LDT for
this rulemaking action were discussed previously, and are shown in
Table V-A.
The various costs associated with this rulemaking action will
occur in different periods. In oVder to make all costs comparable,
the present value at the start of 1983 of the aggregate costs has
been calculated, based on a discount rate of 10 percent. Use of a
discount rate emphasizes that because of the time value of money, a
cost incurred now is worth more to the nation than a cost incurred
in the future.
The calculation of the present value in 1983 of the aggregate
costs, with the assumptions required for the calculation is shown
in Table V-I. The aggregate cost of complying with the new regula-
tions for the five-year period is equivalent to a lump sum invest-
ment of about $1.29 billion (1983 dollars) made at the start of
1983. Expressed in other terms, the aggregate cost of compliance
is equivalent to an investment of $95 per LDT made at the start of
the year the LDT is produced.
It is estimated that LDTs over 6,000 lbs. GVWR comprise 45
percent of the LDT group.16/ In accordance with the Clean Air Act,

Table V-I
Present Value in 1983 of the Aggregate Cost
of Compliance for the 1983-1987 Model Years
Year U
Cost 2/
$ 70.950K
Present Value in 1983 2/
$ 85.850K
Undiscounted Aggregate Cost: $1,478,825K
1983 Present Value of Aggregate Costs: $1,285,638K
1/ Costs are assumed to occur at the start of each year.
Redesign and development
Emission control system
Estimated at $165 million and
allotted over the five year
period (1981-1985) according
to the formula: Cost (mil-
lions) ¦ 21846 - ll(year).
1982 - Preliminary deteriora-
tion factor assessment and
emission data vehicles.
Same as model year.
1982 (hardware), 1983-1987
2/ 1980 dollars, includes profit and overhead.
3/ 10 percent discount rate.

Table V-J
Aggregate Costs of Compliance for
Vehicles Produced During 1983-1987
(Discounted at 10% to January 1, 1983)
Redesign and Development	$ 228,382K
Certification	50,630K
SEA Testing (self audit)	4,026K
Manufacturing of Hardware	$1,002,600k
Total Discounted Cost	$1,285,638K

Table V-K
Undiscounted Costs of Compliance Per Vehicle
for Vehicles Produced During 1983-1987
Redesign and Development
Cert ificat ion
SEA Testing (self audit)
Manufacturing of Hardware

Undiscounted Cost per Vehicle

Table V-L
Discounted Costs of Compliance Per Vehicle
for Vehicles Produced During 1983-1987
Redesign and Development
SEA Testing (self audit)
Manufacturing of Hardware

Discounted Cost per Vehicle
and First Price Increase

any vehicle over 6,000 lbs. GVWR is a heavy-duty vehicle and must
meet the heavy-duty engine emission standards. Using the appropri-
ate portions of the fixed, and variable costs the expenditure for
meeting the reductions required by statutes is about $580 million
dollars (discounted).
For ease of reference, the components of the cost of compli-
ance and the different ways of expressing it are summarized in
Tables V-J, V-K, and V-L.
D. Sensitivity of the First Price Increase to Changes in Key
Analysis Parameters
In any analysis of this type large changes in any key projec-
tions or assumptions may lead to significant changes in the econom-
ic impact of the regulations to the ultimate consumer. This
economic impact may take the form of either changes in the first
price incrase or operating cost changes.
EPA has identified six major areas to use in this sensitivity
analysis: hardware cost, discount rate, sales projections,
fleet sales mix projections, fixed cost expenditure rate, and
changes in operating costs.
1. Hardware CqbC
EPA's estimated hardware costs could be in error if, for
example, noble metals increase or decrease drastically or 3-way
catalyst systems are required on a percentage of the eight cylinder
Four cases will be analyzed:
1} 25 percent increase in noble metal prices
2)	25 percent decrease in noble metal prices
3)	20 percent of eight cylinder engines require 3-way
4)	40 percent of eight cylinder engines require 3-way
Using the noble metal costs and hardware cost methodology used
previously, the table below gives the impact of these changes on
the hardware cost and the anticipated first price increase of
Incremental Change in
First Price Increase
25% increase in noble metal prices
25% decrease in noble metal prices
20% of 8 cylinder use 3-way
40% of 8 cylinder use 3-way

2. Discount Rate
In the current times of sluggish motor vehicle sales and
unstable economic conditions some manufacturers may choose to
finance their capital requirements through the capital markets.
Current costs of capital (prime rate) have varied between ten and
twenty percent in the past two to three years. To gauge the
impact of changes in the prime rate on the first price increase,
two additional discount rates will be tested: 12.5 percent and 15
percent. The results are shown below:
3. Sales Projections
As it is with any other product, the cost per item is depen-
dent on the number of units sold and thus, the first price increase
is dependent on the number of light-duty trucks sold.
To determine the sensitivity of the first price increase to
total sales, fleets of fifteen percent larger and smaller than that
used in this analysis will be tested. With all other inputs the
same as in the base case, the sensitivity of the first price
increase to total sales is shown below.
4. Fleet Sales Mix Projections
Due to market pressures, fuel economy pressures, and emission
standards, EPA is expecting a shift in the market from its present
mix of LDDT and 4, 6, and 8 cylinder gasoline-powered trucks.
At least two different areas of this sales mix should be
investigated: dieselization and engine sizing mix changes.
An average dieselization percentage of 12 percent is expected
through the period 1983-1987. To test the sensitivity to dieseli-
zation, values of 6 percent and 18 percent will be used. To gauge
the sensitivity of the analysis to the engine size mix, three
additional scenarios will be tested: 8 cylinder dominant, 6
cylinder dominant, and 4 cylinder dominant. The results of this
analysis are shown below:
Discount Rate	Incremental Change in
First Price Increase
Discount Rate	Incremental Change in
First Price Increase

Fleet Change
Incremental Change in
First Price Increase
6% Increase in Dieselization
6% Decrease in Dieselization
8 cylinder dominant
(70% 8 cylinder, 20%
6 cylinder, 10% 4 cylinder)
6 cylinder dominant
(30% 8 cylinder, 55%
6 cylinder, 15% 4 cylinder)
+ $0
4 cylinder dominant
(35% 8 cylinder, 35%
6 cylinder, 30% 4 cylinder)
5. Fixed Cost Expenditures
Manufacturers may choose to expend the expected fixed costs at
a different rate than was used in this analysis (Table V-G). The
year in which the fixed costs are incurred is important in deter-
mining the effect of fixed costs on the first price increase due
primarily to the compound effects of discounting and overhead/
This analysis will examine two cases: 1) all fixed costs are
incurred before production begins and 2) all fixed costs are
incurred before the anticipated 1985 NOx regulations. For this
analysis, certification and SEA expenditures will be incurred as
shown in Table V-G, but the R&D expenses will be spread evenly over
the term allowed. For case one above this is a two year period and
for case two a four year period. The results of this analysis are
shown below.
6. Changes in Operating Costs
The emission control technology described in this chapter and
in the technological feasibility discussion is expected to provide
compliance with emission standards with no fuel economy penalty and
possibly a slight fuel economy improvement.
If the EEC/catalyst/air pump system were to yield a fuel
economy loss or gain this would have a substantial impact on the
total economic burden of this regulation.
Expenditure Rate
2 years (evenly 1981 and 1982)
4 years (evenly 1981 through 1984)
Incremental Change in
First Price Increase

To estimate the potential impact of a slight change in fuel
economy, EPA has analyzed the impact of a one percent gain or loss
in fuel economy from the 1980 CAFE levels of 16 mpg for 2 wheel
drive and 14 mpg for 4 wheel drive trucks. The results of this
analysis are shown below:24/
Incremental Cost
Fuel Economy Change	Change per Vehicle
+1%	-$83
-1%	+$83
In addition, the possibility remains that some manufacturers
may choose to meet the emission standards using larger, more
heavily loaded catalysts, increased air injection and engine
modifications, instead of EEC systems. EPA's fuel economy analysis
in the summary and analysis of comments concluded that a 3 to 4
percent fuel economy loss would be possible if the pure catalyst
system was used. However, the pure catalyst system is about $30
cheaper than the EEC system. Using the same methodology as refer-
enced previously and the $30 decrease in first price yields the
cost changes shown below:
Pure Catalyst System/	Incremental Cost
Fuel Economy Change	Change Per Vehicle
-3%	+$219
-4%	+$302
Figure V-A summarizes the cost and cost effectiveness for both
the EEC and pure catalyst control system for a range of fuel
economy changes.
As can be seen in Table V-M which summarizes this analysis,
the greatest sensitivity is related to changes in operating costs.
EPA fully expects that fuel economy pressures will force the
manufacturers to use the EEC based system rather than a pure
catalyst system. Thus, any fuel economy penalty will be averted.
The greatest sensitivity of the remaining parameters is related to
hardware costs. This is as expected since the hardware related
costs account for 78 percent of the expected first price increase
(see Table V-L). Even the most dramatic departure from the
scenario analyzed would yield only a 16 percent increase in the
first price increase. From the benefits shown later in Chapter
VII, the average first price increase would have to be well in
excess of $145 before cost effectiveness would become a concern.
E. Socio-Economic Impact
1. Impacts on Manufacturers
a. Capital Expenditures
The promulgation of the 1983 LDT emission regulations will

Figure V-A

Table V-M
Economic Impact Sensitivity
Absolute Change
in the Average First Price
Case Tested	Increase Expected ($95)
25% increase in noble :
metal prices
25% decrease in noble i
metal prices
20% of 8 cylinders use
40% of 8 cylinders use
12.5% discount rate

15% discount rate

Sales 15% greater

Sales 15% less

18% dieselization

6% dieselization

8 cylinder dominant

6 cylinder dominant

+ $
4 cylinder dominant

Fixed costs (2 years)

Fixed costs (4 years)

+1% fuel economy (EEC)

-1% fuel economy (EEC)

-3% fuel economy (pure
-4% fuel economy (pure
Note: All values rounded to the nearest dollar.
* Operating cost change only.
** Operating cost as offset by $30 saving on first price in

cause the manufacturers of theae vehicles to spend about $200
million for development and certification and an additional average
of $243 million a year Cover the first five years) for production
of emission control systems over and above those required to meet
current standards. These costs will be paid ultimately by in-
dividuals who buy light"duty trucks, but the manufacturers will be
required to bear the initial cost burden for this work. This
regulation, therefore, will require manufacturers to generate
additional capital between promulgation of the final rule and 1983,
either internally or on the capital markets, sufficient to meet
each year's costs.
Table V-G showed the manufacturers costs by category and
year. Coats are first incurred in 1981 as redesign and development
begins, but the first opportunity to recover costs via price
increases will be in 1983,
Manufacturers should have little trouble financing the re-
quired investment. If the total pre-1984 development, and certifi-
cation costs plus the 1983 emission control hardware costs are
distributed among the manufacturers in proportion to their shares
of the LOT market,2/ the required investment by each manufacturer
is as follows '.17/
Required Investment
General Motors
. 2M
Toyo Kogyo
These investments are small when compared to the total capital
requirements of the manufacturers in this period. For example,
Ford would have to invest about 8.3 percent of its 1978 corporate
profits and General Motors would have to inveBt only 3.7 percent of
its 1978 corporate profits. Chrysler's investment will have to be
funded from Limited loan guarantees. However, in all cases, mast
if not all of the investment will be recovered through. Che ficet
price increase.
b. Effects on the Demand for LDTs
Changing the prices of light-duty trucks may, of course,
impact sales. An average first cost increase of $95 means a
selling price rise of about 1-2 percent. This increase is less
than the usual annual increase of 5 to 7 percent.6/

EPA knows of no specific estimates of the price, elasticity
of demand for LDTs. The short term elasticity of demand for LDVs
and LDTs has been estimated at -0.70.19/ Lacking any other esti-
mates, this short term elasticity of demand has been assumed as
valid for LDTs. In a report by EPA's Office of Noise Abatement
Control, a long-term elasticity of demand of -0.32 was cited for
all trucks and buses.20/ Considering that LDTs comprise the vast
majority of vehicles in this group, and the long-term elasticity of
demand for heavy-duty vehicles is near -0.7, using the -0.32 as the
long-term elasticity demand for light-duty trucks, is a reasonable
estimation. One method of estimating the sales decrease is the
Ford Econometric Model.21/ This model accounts for first cost
increases, changes in fuel economy, and increases costs of opera-
tion and maintenance. Using some of the data presented in this
chapter and making some simple assumptions as to LDT average retail
price and fuel economy, the model predicts that LDT sales should
decrease about 0.5 percent in the short term and about 0.2 percent
in the long term as a result of these regulations. Sales by some
smaller manufacturers of LDTs may decline more than those of larger
manufacturers due to their smaller sales volume over which the
development, certification and tooling costs can be amortized. The
small decrease in total industry sales, due to these regulations,
will be more than overcome by normal sales growth and thus can be
expected to have no noticeable effect on any single manufacturer's
sales .
It is unlikely that the sales mix between LDTs, HDVs, and LDVs
will be significantly affected. Only commercial concerns would
consider switching from an LDT to an HDV for delivery or other
purposes, but the greater selling price and operating costs of
heavy-duty vehicles would greatly deter this switching. Some
vehicle owner's who drive LDTs for pleasure may choose to switch to
LDVs, or lower GVWR LDTs, as a result of this action.
Since commercial LDTs are used primarily for intracity deliv-
ery, no switch to shipping freight by rail or air is feasible.
It is not expected that the promulgation of the regulations
will have any long term impact on employment or productivity in the
light-duty truck industry, since industry wide sales will be
affected little.
2.	Impact on Users
Users of light-duty trucks will be affected by the higher
vehicle costs. The expected average, sales-weighted first cost
increase of $95 should not substantially impact the owner's ability
to pay for new LDTs. These regulations will increase costs only
.08 cents per mile over the useful life, an insignificant fraction
of current costs.
3.	Effects of Energy Use
EPA expects compliance with these regulations to be based upon

continued use of catalyst plus air pump system and the addition of
the newly developed electronic engine controls. These regulations
will have no negative fuel impact and will not inhibit manufac-
turers' ability to comply with fuel economy standards for LDTs.
(See fuel economy issue in the Summary and Analysis of Comments).
4.	Inflationary Effects - Consumer Price Index
The consumer price index (CPI) is one of the primary indica-
tions for changes in the general price level. It is estimated that
light-duty trucks contribute about 0.5 percent to the CPI deter-
mination.22/ Combining this percent contribution with the average
estimated price increase of about 1.5 percent will give only a
.0075 percent increase in the CPI. Needless to say this increase
is negligible compared to other elements of the CPI. Therefore,
EPA concludes that these emission regulations will have no signi-
ficant price level impact. Further, since the public will receive
air quality and related health improvements in exchange for the
higher LDT prices, the rise in the CPI that will occur cannot
properly be termed inflation.
5.	Balance of Trade Effects
The increase in the precious metal loadings of catalytic
converters may cause an increase in the imports of platinum and
palladium. Before one can quantify this impact, an engine family
by engine family analysis of the increases noble metal loadings is
necessary. This is found in the Economic Impact issue of the
Summary and Analysis of Comments document supporting the rule-
making. This data must then be sales weighted, considering only
domestically produced engines.
Accomplishing this analysis yields a per LDT increase of .338
grams of platinum and .091 grams of palladium. The current value
of these increased metal imports is about $3.90 per domestic LDT.
Using the sales projections in Table V-A this comes to an increase
in imports of about $9.9 million dollars per year. Assuming
increases in domestic imports of platinum and palladium neglects
the real possibility that recycled precious metals from used
catalysts may be commerically available by the mid-eighties.
Another major balance of trade impact is related to the first
price increase in imported LDTs. Based on the hardware costs in
Table V-D (most imports are 4 cylinder) an average first price
increase of $134 can be expected for imported LDTs. The expected
first price increase for imported LDTs is larger than average due
to the first time use of catalytic converter technology on a high
percentage of imported LDTs. If one assumes a constant market
share for imported LDTs this yields a loss in the balance of trade
by an average of about $45 million dollars per year.
6.	Local and Regional Effects
The domestic light-duty truck manufacturers operate about 22

Table V-N
Light Truck Assembly Plants
in the United States
General Motors:
Fremont, California
Lakewood, Georgia
Baltimore, Maryland
Detroit, Michigan
Flint, Michigan
Pontiac, Michigan
St. Louis, Missouri
Lordstown, Ohio
Janesville, Wisconsin
San Jose, California
Louisville, Kentucky
Wayne, Michigan
Twin Cities, Minnesota
Kansas City, Missouri
Lorain, Ohio
Norfolk, Virginia
Warren, Michigan
St. Louis, Missouri
American Motors:
Toledo, Ohio
South Bend, Indiana
International Harvester:
Fort Wayne, Indiana
Westmoreland, Pennsylvan

plants in assembling their products. Their locations for each
manufacturer, are shown in Table V-N.23/ A total of twelve states
are included.
General Motors operates nine assembly plants to produce
light-duty trucks and vans. Some of these plants are also used to
assemble passenger cars. General Motors has plants located in
seven states with three in Michigan. Ford operates seven assembly
plants spread over seven states, Chrysler and AMC currently
operate two each, IHC and Volkswagen operate one each.
It is reasonable that any slight decrease in employment which
might be related to these regulations would be spread evenly across
the twenty-two plants affected. If production were to drop 0.2
percent as a result of these regulations, spreading that drop
evenly over the twenty-two plants would yield a drop of only 0.2
percent at each plant. This is a relatively incalculable impact
considering other factors affecting production, and thus only a
very insignificant drop in employment might result.
Offsetting this slight drop would be the jobs created or
sustained by the research and development effort anticpated. This
impact would be strongest at the large volume manufacturers (GM,
Ford, Chrysler) and at the vendors which produce emission related
In any event, the expected annual sales increases and the
effects of the dynamic condition of the LDT market will render the
employment impact of these regulations negligible.
As a result, EPA concludes that no locality or region will
suffer noticeable or disproportionate economic impact, positive
or negative, as a result of these regulations, and all areas will
benefit by the improvements in air quality these regulations will
br ing.

\J See Chapter III.
2/ Based on data gathered from EPA's Certification Division.
3/ Rulemaking Support Paper, Light Truck Fuel Economy Standards
~~ Model Years 1982-1985, U.S. DOT, NHTSA, December, 1979.
4/ Cost Estimations for Emission Control Related Components/
Systems and Cost Methodology Description, Leroy H. Lindgren,
Rath and Strong Inc, March 1978, EPA-460/3-78-002.
5/ See Issue F, Economic Impact, in the Summary and Analysis of
Comments supporting this rulemaking action.
6/ Based on Bureau of Labor Statistics data.
7/ Summary and Analysis of Comments to the NPRM: 1983 and Later
Model Year Heavy-Duty Engines, U.S. EPA, OMSAPC, December
8/ The six dollar cost is based on discussion with a Mercedes-
Benz retail dealer and represents an approximate replacement
cost for the closed crankcase system on their light-duty
diesel engines plus profit.
9/ EPA memo, Light-Duty Vehicle Certification Costs, D. Hardin,
Jr. to E. Brune, D. Kimball, and J. Marzen, March 13, 1975.
10/ Analytical Development of Sampling Plans for SEA, Sylvia
Leaver, EPA Office of Enforcment, MSED, December, 1978.
11/ Analysis of California Two Percent Audit Data, March 1980,
available in the Public Docket.
12/ See Issue L Fuel Economy in the Summary of Analysis of Com-
ments supporting this rulemaking action.
13/ Preliminary Regulatory Analysis of Light Truck Fuel Economy
Standards Model Years 1982-85, DOT, NHTSA, Office of Program
and Rulemaking Analysis December 1979, Table 111-12.
14/ Fuel Additive Issues and Petroleum Product Supply Effects,
February 1980, Sobotka and Company, Inc., pp. 20, 21, prepared
for EPA under contract 68-01-4939.

References (cont'd.)
15/ Average Lifetime Periods for Light-Duty Trucks and Heavy-Duty
Vehicles U.S. EPA, OMSPAC, SDSB 79-24, Glenn W. Passavant,
November 1979.
16/ Environmental Impact Statement - Emission Standards for
Light-Duty Trucks, U.S. EPA, OMSPAC, ECTD, November 1976.
17/ Required investment includes a portion of the fixed costs and
the 1983 hardware cost. The portion for each manufacturer is
based on it's portion of the total projected sales.
18/ Chrysler's cost includes its SEA hardware cost and 1983 SEA
testing costs.
19/ "Economic Analysis of Selective Enforcement Auditing Regula-
tions", U.S. EPA, Thomas J, Alexander, December 22, 1975.
20/ "Background Document for Medium and Heavy Truck Noise Emission
Regulations", Appendix C, EPA Office of Noise Abatement
Control, March 1976, EPA-550/9-76-008.
21/ Econometric model of new car sales presented by Ford Motor
Company in submission to the 1977 Suspension Hearing panel,
interoffice memo of January 27, 1976, J.V. Deaver to D.A,
22/ Preliminary Impact Assessment of the Non—Passenger Automobile
Fuel Economy Standards for Model Years 1980 and 1981, DOT,
NHTSA, Planning and Evaluation Office of Program Analysis,
November 29, 1977.
23/ Automotive News Market Databook Issue, April 1980 and Wards
Automotive Yearbook, 1979.
24/ This analysis used a fuel price of $1.30 per gallon and a
discount rate of 5 percent for the price of fuel over the
useful life periods and mileage accumulation rates shown in
reference 15 above for LLDT and HLDT. In addition, this
anlaysis used LLDT as 55 percent of the market and HLDT as 45
percent of the market. Based on historical sales data LLDTs
are 95 percent 2WD and 5 percent 4WD. HLDTs are 47 percent
2WD and 53 percent 4 WD.

A.	Introduction
As EPA has proceeded with the development of a final rule-
making based upon analysis of comments received in response to the
July 1979 proposal, alternatives and options in essentially all
aspects of the rulemaking have been evaluated. Most of the com-
ments received from manufacturers either explicitly or implicitly
involved alternatives to items which EPA had proposed. That is,
EPA was requested to evaluate eliminating, modifying or replacing
elements of the rulemaking proposal in a wide variety of ways based
upon what manufacturers perceived as defects in the proposal, or
more desirable alternatives. Some of the alternatives raised
during the comment period had already been analyzed by EPA, while
some had not.
In the Summary and Analysis of Comments detailed analysis of
all identified alternatives are developed. This document is
available in the public docket (OMSAPC-79-2) and the material it
contains will not be repeated in this chapter beyond the level of a
brief review of major alternatives considered. In addition to the
Summary and Analysis of Comments, Chapter VII (Cost Effectiveness)
of this Regulatory Analysis considers the emission benefits and
costs associated with each basic element of the rulemaking and
determines the resulting cost effectiveness.
The alternative evaluated by EPA will be considered in three
areas: 1) Alterantive standards, 2) Alternative to specific
elements of the rulemaking, 3) Alternative timing for implementa-
B.	Alternative Standards
There were two options available to EPA with regard to the
establishment of the standards. The first concerned the stringency
of the standards. The second involved dividing the light-duty
truck class into subcategories and establishing separate standards
for each subcategory.
Concerning the stringency level of the standards, the proposed
levels of 0.8g/mile HC and lOg/mile CO are being finalized. This
is being done in spite of the fact that revisions to the light-duty
truck baseline made since the proposal indicate that the statutory
90 percent reduction standards could be lower (0.6 HC, 8 CO).JL_/
Rather than adopt more stringent standards, EPA has retained the
original standards, as proposed. This has been done primarily for
the sake of promptly completing this rulemaking. The 0.8 and 10g/
mile standards do produce substantial benefits at an economical

cost. They also represent levels very close to the statutory 90
percent reduction levels (88 percent for HC and 87 percent for
CO). Therefore, they are being finalized for the 1983 model year.
Revised standards for HC and CO may however be considered as part
of future rulemaking.
EPA also has the option of adopting standards less stringent
than those proposed. This is not being done because of the sub-
stantial environmental benefits of the statutory reductions and the
ready feasibility of attaining those reductions.
In addition to the actual level of the standards, EPA de-
scribed at the time of the NPRM an option involving subdivision of
the current light-duty truck class into subcategories and estab-
lishing graduated standards. As described in the draft Regulatory
Analysis accompanying the NPRM, this alternate could have advan-
tages. By such an approach, the smaller and lighter light-duty
trucks could be controlled to more stringent levels and a greater
overall emission reduction would result. On the other hand, such
an approach could have a detrimental effect on fuel economy if it
were to discourage downsizing of vehicles. This could happen if a
manufacturer found himself facing more stringent emission standards
as a result of his desire to downsize.
As was the case with lower numerical levels for standards, it
would be outside the range of this rulemaking to change the class
limits at this time. At the time of the proposal, EPA indicated
its belief that retention of a single light-duty truck class was
the best option. However, EPA intends to continue its study of
this option. If at some future time the balance should shift
toward a subdivided light-duty truck class, then that approach
would appear in an EPA proposal.
C. Alternatives to Specific Elements of the Rulemaking
Since essentially all aspects of the proposal were questioned
during the comment period, EPA has analyzed all of these in the
course of developing the final rulemaking. Alternatives relating
to the rulemaking include redefinition of useful life, in-use
durability testing, allowable maintenance regulations, a 10 percent
acceptable quality level for assembly line testing, and diesel
crankcase control. For each, there is an appropriate portion of
the Summary and Analysis of Comments which can be consulted. In
addition, the cost effectiveness of each element is estimated in
Chapter VII of this Regulatory Analysis. It is important to
realize that Chapter VII is actually an analysis of alternative
rulemaking packages. Each cost versus benefit ratio is derived
from a comparison of the final rulemaking with a rulemaking package
not having the item being evaluated. Thus, each constitutes a
unique combination package, and each combination represents an
alternative approach to the rulemaking.

For the case of in-use durability testing, this review process
indicated that the proposal should not be promulgated at the
present time. For each of the remaining elements, the basic
approach originally proposed by EPA remains the best alternative.
However, modifications have been made in several of them to improve
their practicability or clarity.
D. Alternative Timing for Implementation
The last major area where alternatives were considered is that
of the first model year for which the final rule should be applied.
EPA had proposed the regulation for 1983. Many comments were
received indicating that in the context of such aspects of the
proposal as redefined useful life, allowable maintenance restric-
tions and a 10 percent AQL, manufacturers were doubtful that
compliance could be attained by 1983. In addition, legal issues
concerning provisions of the 1977 Clean Air Act Amendments and
their relation to any minimum mandated lead time were strongly
The timing for introduction of new regulations can have very
important consequences. From the manufacturers point of view it
affects the rate at which resources must be expended to attain
compliance, and possibly the very ability to comply. Environmen-
tally, timing defines the point at idiich desired emission reduc-
tions will begin to be realized. In considering these conse-
quences, an appropriate balance must be struck.
EPA has carefully considered the comments concerning lead
time. An analysis can be found in the Summary and Analysis of
Comments. In brief, the conclusion reached is that model year 1983
is a readily attainable compliance deadline, and that there are no
legal barriers to EPA's promulgation of that deadline. Therefore
EPA has decided to promulgate the final rules for the 1983 model

1/ See the Summary and Analysis of Comments, Issue K-Numerical
Standards/Standards Deviation.

A.	Methodology
Cost effectiveness is a measure of what might be termed the
economic efficiency of some action directed toward achieving some
goal. Expressed as cost per unit of benefit achieved, cost effec-
tiveness can be used to compare various alternative methods of
achieving the same goal. In the context of improving air quality,
the goal is to reduce emissions of harmful pollutants, and cost
effectiveness is expressed in terms of the dollar cost per ton of
pollutant controlled.
To evaluate cost effectiveness, two pieces of information on
the alternative being evaluated are needed. These are the cost of
the alternative and the benefits to be gained. Costs to be used in
this chapter will be total identified costs expressed on a per
engine basis, including both costs to the manufacturer and costs to
the operator (all discounted to January 1 of the model year in
which the vehicle is produced). These costs will be allocated
equally among the pollutants being controlled. The benefits will
be computed as total lifetime emission reductions per vehicle.
In this chapter, the rulemaking provisions will be subjected
to two distinct analyses. The first will be an incremental analy-
sis of each of the major components of the package. The second
will be an analysis of the package as an integrated strategy. The
purpose of these two approaches are different, and the reader is
cautioned against misinterpretations of the incremental analysis.
In the incremental approach, the effect on costs and benefits of
removing individual components will be examined. To varying
degrees, both costs and benefits of these components overlap and
several components of the package may act together to obtain a
given benefit. In such a case, loss of any one part of the package
can result in a disproportionate loss of benefits. There are so
many overlapping interrelationships that it would be impossible to
consider every possible combination of the various components of
the package. This analysis will instead look at the single set of
options produced by deleting each component one at a time. The
total loss of benefits produced by deleting a component will be
associated with the cost of that part of the package. Therefore, if
one were to simply sum incremental costs or incremental benefits as
an attempt at obtaining total costs or benefits, significant
amounts of double counting would occur. Such a procedure would be
invalid. The integrated cost effectiveness analysis must be used to
evaluate overall costs or benefits.
B.	Background
In the draft Regulatory Analysis which accompanied the pro-
posed regulations, a cost effectiveness analysis of the proposal

was carried out. That analysis considered the overall cost effec-
tiveness of the entire proposal as an integrated compliance strat-
egy. In the preamble to the NPRM it was noted that "(i)t is not
possible to present the individual cost-effectiveness values of
each element (e.g., change in durability testing, change in AQL,
etc.) of this proposal due to insufficient data. Moreover, the
individual elements are interrelated which makes it difficult to
isolate the benefits for each element. Removing one element might
seriously jeopardize the effectiveness of the other elements,"
(44FR 40793 July 12, 1979). For these two reasons, cost-effective-
ness values for individual elements were not presented.
During the course of the comment period on the proposed
regulations, EPA has endeavored to develop more data and establish
methods for estimating changes in emissions which could be asso-
ciated with changes in the various components of the package. This
effort has been sufficiently successful to allow estimated cost
effectiveness analyses for the main components of the rulemaking to
be undertaken.
The problem of interrelated benefits still exists, however.
It is important to bear in mind that the benefits and costs in
this analysis will overlap, and that summing them all would result
in double counting. For example, consider the case of extended
catalyst lifetimes required under the allowable maintenance provi-
sions and the revised useful life definition. The benefit of
increasing catalyst lifetimes is significant. However, if the
useful life remained at 50,000 miles, the intent of the allowable
maintenance provision for catalyst change intervals would be lost.
Therefore, incremental analysis of the allowable maintenance
interval and revised useful life will each separately be looking at
partly the same benefit in emission reductions.
Allowing the benefits to overlap in this fashion may appear to
give too much credit to individual elements of the package. This
is not true, since in each case the benefit considered will be the
best estimate of what the package would actually gain or lose if
that element were retained or removed. The purpose of an incremen-
tal analysis is to answer that question for each element. Although
it would be desirable, it is not the chief purpose of an incremen-
tal analysis to evaluate the benefits of the total package. The
benefit attributed to the overall integrated package will be
determined separately.
C. Summary
Using all data now available (both that generated by EPA and
that submitted to EPA during the public comment period on the
proposed regulations), an analysis of the cost effectiveness of
each major element of the regulation package and of the overall
package as a unit has been done. This analysis developed benefits
expressed as tons of pollutant removed (either HC or CO) over the
average lifetime of an individual vehicle along with total costs

for the same lifetime (discounted to year of sale).
Overall benefits and costs used as a starting reference the
existing regulations applicable to 1982 model year light-duty
trucks. That is, both overall benefits and overall costs were
developed as changes in relation to the case of the existing
regulations continuing in effect. Benefits and costs for the
individual elements of the package, on the other hand, were eval-
uated in terms of changes to the final package. The loss in
benefits that would occur if each element were removed from the
package was evaluated in comparison with the cost reduction that
would be produced by that same change. Figure VII-A summarizes the
benefits developed. Costs, benefits, and cost effectiveness are
tabulated in Table VII-A. Cost-effectiveness figures for other
mobile source control strategies are provided in Table VII-B for
comparison purposes.
D. Overall Rulemaking
Lifetime emissions for vehicles built to conform to this
rulemaking are given in Table IV-A. Also in that table are life-
time emission rates for model year 1982 light-duty trucks. The
difference between these two is the benefit to be realized by
implementing the rulemaking. That is,
(MY '82 lifetime emission) - (MY '83 lifetime emission) * net benefit
Using the data from Table IV-A, we have:
Benefit for HC ¦ 0.38 - 0.09 = 0.29 tons per vehicle
Benefit for CO ¦ 5.0 - 0.9 =4.1 tons per vehicle
The cost analysis of Chapter V determined costs attributable
to various aspects of the regulation package. These cost, given in
Table V-L, are reproduced below:
	Item		Discounted Cost per Engine
Redesign and Development	$16.87
Certification	3.74
SEA Testing (self auditing)	0.30
Hardware	74.00
Total cost per engine	$94.91
The total cost per engine, when split equally between HC and
CO, and combined with the per vehicle benefits will give the
cost-effectiveness values shown in Table VII-A.
E. Redefinition of Useful Life
In section D above, the lifetime emissions per vehicle

Figure VII-A
Incremental Lifetime Benefits
10 Percent AQL
Allowable Maintenance |
Useful Life £
Overall Package
0.1	0.2	0.3
Lifetime HC Benefit (tons/vehicle)
10 Percent AQL j
Allowable Maintenance
Useful Life
Overall Package
-i	1			r—	1	»	r
1	2	3	4
Lifetime CO Benefit (tons/vehicle)

Table VII-A
Incremental Lifetime Cost Effectiveness
Benefit Cost Effectiveness
Cost	(tons)		($/ton)	
Opt ion
Useful life
Allowable maintenance
10 percent AQL*
Overall package
*	Note: The ten percent AQL includes a NOx benefit of 0.04
tons. Cost is therefore dividied three ways. The cost effective-
ness for NOx is $31 per ton.

Table VII-B
Cost Effectiveness ($/Ton) Comparison
With Other Emission Control Strategies
Control Program
LDV Statutory
Standards b/
LDT Interim
Standards c/
After Control
Baseline Emission a/ Program Initiated a/
3.1	d/
5.2	e/
I/M for Existing
LDVs ff
Motorcycle Standards
1978/1979 £/
1980 +
1984 HDE Regs i/
HC - 1.
CO - 25
HC + NOx - 10
8-22.5 h/
Cost Effectiveness
~aj Emission Levels in grams/mile, except for HD vAiich are g/
b/ Report: Interagency Task Force on Motor Vehicle Goals Beyond
1980, March 1976.
cj "Environmental Impact Statement - Emission Standards for
Light-Duty Trucks," November 29, 1976.
d/ Trucks 0 - 6,000 lbs. GVWR.
e/ Trucks 6,001 - 8,500 lbs. GVWR.
f_J "Cost Effectiveness Estimated for Mobile Source Emission
Control," Vector Research, Inc. for EPA, January 1978.
gj "Environmental and Economic Impact Statement - Exhaust and
Crankcase Regulations for the 1978 and Later Model Year
h_/ Sliding Scale Based on Engine Displacement (cubic centimeters).
ij "Regulatory Analysis and Environmental Impact of Final Emis-
sion Regulations for 1984 and Later Model Years Heavy-Duty
Engines," EPA Office of Mobile Source Air Pollution Control,
December 1979.
7.7 2763
Neg. --
238	8 (gas)
253	- (diesel)

were calculated to be 0.29 tons of HC and 4.1 tons of CO. These
numbers, as has been noted, presumed that all other aspects of the
rulemaking were intact. The basic assumption made in that regard
was that the combined package would result in in-use emissions
which closely match the performance of certification vehicles. The
only exception was due to the failure of a small percentage (8.5
percent of total) of catalysts on a random basis near the end of
the vehicle useful life.
The evaluation of the new useful life definition will proceed
by estimating the loss of benefits and reduction in costs that
would occur if this element were removed from the package while all
other elements remained intact. This method will make it possible
to evaluate the impact of not implementing useful life on the
overall package while at the same time estimating the cost effec-
tiveness of this element.
1. Benefits
The extension of the useful life definition to the average
full lifetime rather than something approximating half of the full
life as is done in current practice has the effect of requiring
that vehicles will be able to meet emission standards throughout
their average life. This will require new vehicle emission rates
to be lower so as to not exceed the standards after accounting for
emissions deterioration over approximately twice the mileage
interval of current practice. Full life useful-life will also
require the use of control systems which are sufficiently durable
to last the vehicle's lifetime. This makes the useful life
change a key to the effectiveness of the allowable maintenance
provisions. In addition, since performance of emission related
maintenance will be a condition for maintaining warranty coverage,
it is likely that high mileage vehicles will be better cared for
than is currently the case.
The increased level of maintenance will have an associated
degree of emission benefits. However, a method for quantifying
those benefits is lacking. The remaining two aspects of full life
useful-life - lower initial emission rates and more durable com-
ponents - do provide a basis for estimating the benefits of this
element of the rulemaking. If full life useful life were dropped
in favor of the current 50,000-mile useful life then both of these
areas would suffer. Emission target levels would increase and
system durability would not have to be proven beyond 50,000 miles.
The latter fact would have its major emission impact in relation to
catalysts. If catalyst durability need only be proven to 50,000
miles then a "50,000 mile catalyst" will be used instead of a
"100,000 miles catalyst".
Following the procedure used in section G, these changes can
be quantified. The production target levels have been noted as
(0.81) x (max. legal level/df) for HC and (0.68) x (max. legal
level/df) for CO. The maximum legal level means the highest actual

emissicm rate which would round off to the emission standard. The
dfs used for the benefits of the overall package were 1.4 for HC
and 1.3 for CO over 100,000 miles. For a useful life reduced to
50,000 miles, these df's become 1.2 for HC and 1.15 for CO. The
resulting target levels are:
(0.81) x (0.85/1.2) = 0.57 g/mi HC
(0.68) x (10.5/1.15) - 6.2 g/mi CO
These emission levels form the starting point for vehicles
whose catalysts remain operational. Because of deterioration,
emissions will increase with time corresponding to our df's of
1.2/1.15 for 50,000 miles. Expressed in gram per mile the overall
emissions as a function of mileage (M) are:
HC « 0.57 + .023(M/10,000)	(VII-1)
CO - 6.2 + 0.19(M/10,000)	(VII-2)
In the absence of full useful life, we have noted that
catalysts designed for 50,000 miles would be used. However,
based upon probability, not all catalysts will have exactly the
same lifetime. Nor will all catalysts need to be changed at the
same point due to failure. We will treat each catalyst as having a
finite lifetime, beyond which emission performance will begin to
degrade at a rapid rate. This could result from occasional high-
temperature conditions or other operating conditions which will
affect system integrity, or randomly occurring factors during
catalyst system manufacturer which affect durability of the system
as extended mileage accumulates. A distribution generally found
appropriate for lifetime phenomena is the Weibull distribution.^/
This distribution has the form:
„ .	, ,M)b,	(VII-3)
F - 1 - exp l-(-Q J
To specify the function of equation VII-3, we will let the
nominal catalyst lifetime (50,000 miles in this case) correspond to
a failure rate of 5 percent. This gives the manufacturer a 95
percent confidence in catalysts performing properly for the desired
lifetime. We will further use a "Weibull slope" of b ¦ 3. Based
upon these two factors, the "characteristic value," 6, becomes
134,570 miles. A plot of this function is given in Figure VII-B.
If a catalyst were to fail on an in-use vehicle with extended
mileage, it is quite posible that it would not be replaced.
Therefore, average emission will increase somewhat near the end of
the useful life period. For these catalysts which fail, emission
rates characteristics of well maintained non-catalyst engines are
desired. Based upon a review of emission factors for light-duty
trucks,2/ starting emission rates of 2.56 g/mi HC and 31.5 g/mi CO

4 5 6 7 8 9 100,07) 2
3 4 5 6 7 8*

are appropriate. These can be combined with a df characteristic of
a well maintained non-catalyst vehicle of 1.1 to give the following
emission rates.
HC - 2.56 + 0.051(M/10,000)	(VII-4)
CO = 31.5 + 0.63(m/10,000)	(VII-5)
Combining equations (VII-l) to (VII-5), the overall average
emission rates will be:
HC = [0.57 + 0.023(M/l0,000)][1 - F] + [F][2.56 + 0.051(M/10,000)]
HC = 0.57 + 0.023(M/10,000) + F[2.0 + 0.028(M/10,000)]	(VII-6)
CO - [6.2 + 0.19(M/10,000)][1 - F] + [F][31.5 + 0.63(M/10,000)]
CO - 6.2 + 0.19(M/10,000) + F[25 + 0.44(M/10,000)]	(VI1-7)
To illustrate the effect of catalyst decay, equation (VII-6) is
plotted in Figure VII-C. Also shown is the result for a 100,000
mile catalyst lifetime.
Over the full useful life of 120,000 miles^/ equations (VII-6)
and (VII-7) yield total emission of 0.14 tons of HC and 1.5 tons of
CO. The net loss of benefits from eliminating the useful life
changes is the difference between these net emissions and those
previously determined for the full rulemaking. The result is:
0.14 - 0.09 ¦ .05 tons HC and 1.5 - 0.9 » 0.6 tons CO
Remembering that the benefit of increased maintenance by the
vehicle owner has not been quantified, these benefits should be
viewed as a lower limit of the potential available.
2. Costs
Apportioning the per-engine costs attributable to the change
in useful life is a difficult task. The methodology to be used for
estimating useful life cost has been developed earlier, in Chapter
V. That discussion notes that electronic engine controls and
upgraded air injection systems would be required for either case.
Costs for catalyst improvements, however, would be different.
Approximately 30 percent of catalyst costs (and the related R&D
for design) are attributed to the increased useful life. From the
data in Chapter V, this can be determined to be $4.27 for hardware
and $5.06 for R&D.
A second area of cost can be directly tied to the useful life
change. This cost is the cost of running extended durability
vehicles to determine a full-life deterioration factor. Of the
$3.74 certification costs quoted earlier, $1.44 is related to the
extended useful life. Total cost associated with the useful life

Figure VII-c
1983 Model Year
Fleet Average HC
Emission Rate vs. Mileage
Vehicle Mileage

change is therefore, $4.27 + $5.06 + $1.44 ¦ $10.77.
F.	Allowable Maintenance Restrictions
These regulations will affect a wide variety of emission
related components. The overall impact will include decreasing the
amount of emission maintenance required to maintain proper vehicle
emission rates. This will reduce the likelihood of excess in-use
emissions due to improper maintenance.
Benefits attributable to many of the maintenance items are
difficult to quantify. However, one of these, the catalyst change
interval, exerts what is perhaps the major emissions influence and
can be estimated. The allowable maintenance regulations will
result in a minimum catalyst change interval of 100,000 miles.
Depending upon the actual useful life to which various manufac-
turers will certify, catalyst lifetimes longer than 100,000 miles
may be used in some cases. However, we will use 100,000 miles for
the analysis. Without the allowable maintenance restrictions,
catalyst change intervals corresponding to current vehicle useful
lives (50,000 miles) are expected. Based upon the position that
catalysts on in-use vehicles are not likely to be changed, even
though catalyst changeover would be specified in the maintenance
instructions, then the in-use fleet would not perform as expected.
Emissions would increase after the 50,000 mile point and with a
shorter lifetime higher rates of catalyst failure would occur.
Emission rates for this situation would be essentially the
same as those developed for a 50,000 mile useful life in Section E
above. Lifetime loss of benefits from dropping the allowable
maintenance regulations would thus be the same as those developed
for useful life:
0.05	tons HC and 0.6 tons CO.
Costs would also be the same as those estimated for useful
life. Results are in Table VII-A.
G.	Selective Enforcement Auditing (SEA)
The question to be evaluated with regard to SEA concerns the
acceptable quality level (AQL) to be used in that program. This
level identifies the maximum failure rate that can occur in audits
of production LDTs before there is a significant probability of
failing an audit. The light-duty truck SEA programs will use a 10
percent AQL to allow for measurement error and quality control
1.	Benefits
The benefits will	be estimated by evaluating the change in
emissions which would	result if the AQL were relaxed from 10
percent to 40 percent.	In general, changing the AQL results in a

change in the mean production level target the manufacturer will
aim for. The degree of change can be calculated from statistical
considerations and there are various ways that these calculations
can be approached, all of which give similar results.4/
a. 10 Percent AQL
Manufacturers base their estimates of production line mean
values upon limited testing of pre-production vehicles (typically
3). In order to ensure that an SEA audit will be passed with some
desired confidence factor (we will use 90%), their target emission
levels will, of necessity, be some point below the required level
(because of production variability and the small sample size).
This point can be estimated by standard statistical techniques,
using the "t" statistic. The following relationships will be
LMT = low mileage target = max. legal level/df	(VII-8)
m = maximum desired production mean = LMT -1.28s
(at a 10 percent AQL)	(VII-9)
x = target new vehicle emission rate = m - s (t//TT)	(VII-10)
max. legal level " highest actual emission rate which when
rounded off will equal the standard
df = multiplicative deterioration factor
s = standard deviation of emission levels
t a "t" statistic for 90% confidence level and n-1 degrees
of freedom
n = sample size
To perform the calculations, estimates of deterioration
factors and emission variability (standard deviation) are required.
Both of these can be obtained from data available to the staff at
this time.
Deterioration factor estimates were given in the NPRM. These
were 1.5 for HC and 1.3 for CO over 50,000 miles and were described
as "typical of the upper end of the range of factors for current
catalyst systems," (44FR 40792 July 12, 1979). Having been used by
EPA, these values were frequently echoed back by commenters in
evaluating the impact of the proposed regulations. The values were
chosen by EPA to illustrate the fact that even with relatively high
deterioration factors, the standards were still feasible. It is
more appropriate in the present case (when we wish to make a best
estimate of the overall impact of the regulations) to choose

estimates of average expected deterioration factors rather than
upper limits. To do this, the staff has examined 1980 certifica-
tion data for light-duty trucks. The average deterioration factors
for 43 durability vehicles for 1980 are 1.17 HC/1.13 CO/1.01 NOx,
(Calculations for NOx will be included throughout this section.
The change in AQL affects the production mean levels for NOx even
though the standard is unchanged. These changes figure in the
costs and feasibility analyses.) To calculate our target values
we will use the following 50,000 mile and 100,000 mile df's:
For 50,000 miles : 1.2 HC/1.15 CO/1.02 NOx
For 100,000 miles: 1.4 HC/1.3 CO/1.04 NOx
100,000 miles is used because, as noted elsewhere, manufac-
turers are expected to either certify to a 100,000 mile lifetime,
or have a catalyst change at that point.
The extension to 100,000 miles relies on extending the deteri-
oration rates in a linear fashion from 50,000 miles to 100,000
miles. Data available to the staff indicates that this is appro-
priate. See for example the data submitted by American Motors on
high mileage vehicles, presented in Figures 1-4 of Issue G of the
Summary and Analysis of Comments (Technological Feasibility).^/
Data on variability was submitted by several commenters.
Variability in the form of s/x ratios was submitted by GM, IH, and
AM. Ford submitted variability measured as s/LMT. Although the
form of the Ford data is somewhat different, Ford indicated in its
submission that current engine emissions levels are such that x is
approximately equal to or somewhat less than LMT. Therefore, the
Ford s/LMT data can be used as an estimate of s/x. Variability
expressed as s/x is desired by the staff because of its judgement
that this measurement (known as the coefficient of variation) will
remain a relatively constant ratio as emission levels (x) go up or
down. (>/
The variability data submitted by commenters showed a wide
range of values. The Ford data was relatively low (.08 HC/.16
CO/.22 NOx) compared to the other submissions (generally in excess
of 0.30). Attempts to resolve the discrepancies led to the con-
clusion that the GM, IH, and AM data was not properly calculated
for our purposes. These firms had combined all sampling results
within a given engine family to compute means and standard devia-
tions. Doing this combines data on several different configura-
tions possibly existing within the same family. Since the emission
values of each configuration would tend to cluster at different
values, the variability calculated when the results were lumped
together could be substantially higher than the variability
existing within any individual configuration. SEA audits are done
on a configuration level, and it is the variability characteristic
of configurations which is needed. Therefore, the GM, IH, and AM
data cannot be used.

Fortunately, the Ford data was computed on a	configuration
specific basis, which probably accounts for the	lower values
supplied by Ford. Based on the Ford values, we will	use estimated
variability of 0.10 HC/0.20 CO/0.25 NOx.
The information on deterioration and variability can be used
with equations (VII-8), (VII-9), and (VII-10) to calculate emission
target values for a 10 percent AQL. The usual pre-production
sample size is 3 vehicles. Calculations will also be done for
sample sizes of 5 and 7. Estimates of the production mean/LMT
ratios will be made first, and then deterioration factors will be
incorporated to generate target production means.
We have:
x = m - s (t//n)
m = LMT - 1.28s
s/x = 0.10, 0.20 or 0.25
Combining these we get, depending on the s/x ratio used:
x = LMT-0.10x(l.28+t/Jn)
x(l.128+0.lOt/Tn) = LMT
x/LMT = l/(1.128+.10tA/Tr)
x = LMT- .20x( 1.28+t/Jn)
x(l.256+.20t/Vn) = LMT
x/LMT = 1/(1.256+.20t//n) I
x = LMT-.25x(l .28+t/JV)
x(1.32+.25tA/ff) = LMT
x/LMT = l/(1.32+.25t//n)
For sample size n * 3,5,7, the results are as follows:
n t x/LMT (s/x - 0.10) x/LMT (s/x = 0.20) x/LMT (s/x = 0.25)
3 1.886	.81	0.68	0.63
5 1.533	.84	0.72	0.67
7 1.440	.85	0.73	0.69
The above tabulation indicates some of the flexibility in-
herent in production target levels. For example, if a manufac-
turer's variability for CO were 0.25 instead of 0.20, increasing
the sample size from 3 to 5 vehicles would allow the same target
values to be maintained.
Tlie values of "t" used are those for a 90 percent confidence
level. Some manufacturers felt that a confidence level as high as
97 percent was needed for each individual parameter in order to
maintain an overall confidence of 90 percent for all three pol-
lutants. This argument was based upon the assumption that the
pollutant levels are independent of each other, which is not true.
HC and CO are strongly related. The analysis of feasibility
contained in the Summary and Analysis of Comments indicates that
most of the emission reductions required to meet the final regula-

tions will be needed for CO and NOx. Meeting the CO targets will
all but guarantee the HC levels. Some increased confidence level
might be desired to cover meeting both CO and NOx. However, this
increased confidence can be obtained at the same target levels by
increasing vehicle sample size. For example, going from 3 to 5
engines would increase the level of confidence to 95 percent or
Using equation (VII-8) along with the x/LMT ratios and deteri-
oration factors derived above, target emission rates can be com-
puted as follows:
(0.85/1.4)0.81 - 0.49 HC
(10.5/1.3)0.68 - 5.5 CO
(2.35/1.04)0.63 - 1.4 NOx
Allowing the emissions to increase according to the desired
df's will yield the following:
HC - 0.49 + 0.02(M/10,000)
CO - 5.5 + 0.16(M/10,000)
NOx - 1.4 + 0.0056(M/10,000)
Following the discussion of section E, the effect of catalyst
failures can be incorporated. NOx emissions are largely unaffected
by catalyst failure. Using equations (VII-4) and (VII-5) combined
with the above equations, the average emission rates for the 10
percent AQL case can be expressed as:
HC - 0.49 + 0.02(M/10,000) + F[2.07 + 0.031(M/10,000)]	(VII-11)
CO - 5.5 + 0.16(M/10,000) + F[26.0 + 0.47(M/10,000)]	(VII-12)
NOx - 1.4 + 0.0056(M/10,000)	(VII-13)
The fraction of failed catalyst will be as specified in
equation (VII-5). For a 100,000 mile catalyst, the "characteristic
value" 0 is 269,141. Lifetime emissions corresponding to equations
(VII-11), (VII-12) and (VII-13) can be calculated as 0.09 tons HC
0.9 tons CO and 0.19 tons NOx over 120,000 miles. These correspond
to emission rates for the complete rulemaking package.
b. 40 Percent AQL
In order to revise the target emission rates to reflect a 40
percent AQL, we need simply change equation (VII-9) to reflect a 40
percent rather than a 10 percent cutpoint.
m - LMT - 0.25s (40 percent AQL)

Carrying this result through	the calculations as done for the
10 percent AQL case will produce	the following, (again for our 3
different variabilities):
x/LMT ¦ 1/(1.025 + 0. 10t//"n)	(s/x = 0.10)
x/LMT = 1/(1.05 + 0.20t//n)	(s/x - 0.20)
x/LMT » 1/(1.06 + 0.25t//iT)	(s/x = 0.25)
For sample sizes of n = 3, 5,	7 the results are as follows:
x/LMT (s/x = 0.10)
x/LMT (s/x = 0.20)
x/LMT (s/x = 0.25)
Using these new ratios along with the applicable deterioration
rates, equation (VII-8) will yield the following target emission
(0.85/1.4)0.88 = 0.53 HC
(10.5/1.3)0.79 = 6.4 CO
(2.35/1.04)0.75 - 1.7 NOx
Continuing with the calculations, the 40 percent AQL equiva-
lents to equations (VII-11), (VII-12) and (VII-13) can be expressed
HC - 0.53 + 0.02(M/10,000) + F[2.03 + 0.031(M/10,000)]	(VII-15)
CO - 6.4 + 0.19(M/10,000) + F[25.1 + 0.44(M/10,000)]	(VII-16)
NOx = 1.7 + 0.0068(M/10,000)	(VII-17)
Lifetime emissions for equations (VII-15), (VII-16) and
(VII-17) corresponding to the 40 percent AQL case are 0.093 tons
HC, 1.1 tons CO and 0.23 from NOx. At a 10 percent AQL, the
emissions were 0.087 tons HC (expressed to 3 places) 0.9 tons CO
and 0.19 tons NOx. Loss of benefits associated with going from a
10 percent AQL to a 40 percent AQL is then 0.093 - 0.087 ¦ 0.006
tons HC, 1.1 - 0.9 ¦ 0.2 tons CO, and 0.23 - 0.19 = 0.04 tons
The amount of emission reduction attributable to the change of
AQL is quite small for light-duty trucks. This is particularly
true for the HC reduction of 0.006 tons. These small benefits are
simply reflecting the fact that the AQL has a relatively small
effect on the target emission rates. Thus, changing the AQL from
10 percent to 40 percent changes the projected targets for HC from

0.49 to 0.53. The reason why the targets show small sensitivity to
the AQL is the low amount of emission variability being seen in
light-duty trucks. Since vehicle-to-vehicle variations in emis-
sions are small, it only requires a small reduction in the average
emission rate to insure that 90 percent of the vehicles pass the
2. Costs
We have just noted that changing the AQL has a relatively
small impact on emissions. Consequently, changing the AQL has
little impact on the difficulty of meeting the standards or the
costs associated. Hardware related costs have been identified in
the Summary and Analysis of Comments as $0.92 for air pumps, $0.60
for diesel EGR, and $1.88 for catalyst loading changes. Added to
this would be a saving of $0.30 for SEA testing costs, for a total
cost of $3.70. Since the AQL change produces a NOx benefit as well
as an HC and CO benefit, cost effectiveness in Table VII-A is
calculated on the basis of dividing costs among three pollutants.
H. Inspection and Maintenance (I/M)
The analysis which has been done so far has contained no
specific reliance on I/M programs. In the overall rulemaking
neither specific benefits nor costs for I/M programs have been
included. However, there are ways in which I/M would enhance the
effectiveness of the rulemaking and help insure full realization
of possible benefits. Therefore, some discussion of I/M in rela-
tion to this rulemaking is appropriate even though it is not
required by the regulations being promulgated.
In the context of this rulemaking package, I/M can be viewed
as an "insurance policy" for many of the benefits. The presence of
an I/M program, which EPA expects would be implemented in those
areas requiring maximum benefits, will insure against neglect or
abuse of emission related systems by the vehicle owner. The two
principal areas when this might occur are misfueling with leaded
fuel or tampering with emission related hardware.
EPA has estimated that misfueling in light-duty vehicles
occurs in up to 8 percent of the vehicles .7j There are no cor-
responding estimates for heavy-duty vehicles since catalysts have
yet to be used. However, something similar seems possible. The
incentive for misfueling is largely an economic one, due to the
lower cost of leaded fuel compared to unleaded fuel. In an area
having an I/M program, the vehicle owner would be faced with a much
more powerful economic incentive against misfueling. This incen-
tive would be the cost of replacing the vehicle catalyst, which
would be over $300, should he fail the I/M test. Rather than
incur this expense, EPA believes the owner would avoid misfueling
his vehicle. Thus, I/M insures against the loss of benefits which
might result from misfueling without actually generating the costs
associated with catalyst replacement.

A similar situation would occur in relation to tampering.
Current engine systems are easy to adjust, and could be adjusted
differently for an I/M test than they are for normal operation.
For future engines this will not be the case. Engines complying
with the parameter adjustment regulations will be difficult to
adjust in such a way as to adversly affect emissions. The pot-
ential for costly repairs from failure of an I/M test (such as
replacing a damaged carburetor) would make the occurrence of such
maladjustment unlikely. Other forms of tampering, such as removal
of the catalyst or other components, would also be difficult enough
to be deterred by the need to pass an annual I/M inspection.
The above scenario allows a rough estimate to be made of the
benefits an I/M program might realize. Assumptions are as follows:
8 percent of the vehicles would be misfueled initially, without
I/M. Their catalysts would fail to non-catalyst emission levels
used earlier for failed catalyst emission rates (equations (VII-4)
and (VII-5)). An additional 8 percent of the catalysts will be
estimated to have failed by the end of the average useful life
period due to occasional misfueling. This is equivalent to 4
percent failed over the whole life in terms of emissions. Tam-
pering will be accounted for by including an additional 5 percent.
Catalyst failures would then total to 8 percent + 4 percent + 5
percent ¦ 17 percent. There will be a 17 percent shift in emis-
sions from rates for operating catalysts to rates for failed
catalysts. Referring to equations (VII-1), (VII-2) (VII-4), and
(VII-5) the change in emission can be expressed as:
HC Increase = 0.17[(2.56 + 0.051(M/10,000)) - (0.57 + 0.023(M/10,000))]
CO Increase = 0.17[(31.5 + 0.63(M/10,000)) - (6.2 + 0.19(M/10,000))]
Lifetime emission benefit of the I/M program using these
relations is .05 tons HC and 0.6 tons CO.
Cost for the I/M program consists of a $5 annual inspection
fee. On the belief that I/M will deter the problems of misfueling
and tampering, no other new costs will be incurred. The fee costs
over the vehicle Life (12 years), discounted to year of sale, are
$37.47. This is used to compute the cost effectiveness found in
Table VII-A.
It could be argued that once an I/M program is put in place to
deter tampering and misfueling, that some of the benefits derived
from other components of the overall regulation (useful life,
allowable maintenance) could be secured by I/M. However, it is the
intent of the regulatory strategy to force the design of durable
emission control systems that are not highly susceptible to mal-
maintenance. It is less costly for the consumer to pay for these
features as part of the new vehicle engine design than to have to
secure maintenance or replace parts later on. If the full useful
life and allowable maintenance provisions of the regulations were
dropped in favor of reliance of an I/M program to obtain the

related benefits, the cost of field maintenance and replacement
catalysts would then have to be charged to the I/M program.
Considering the catalyst situation alone makes this approach much
less efficient than the approach of retaining all parts of the
regulation package and backing it up with I/M. We have previously
estimated the incremental hardware cost of a full life versus a
half life system to be $22. The cost of a replacement catalyst
considering after-market parts markup is about $330.
I. Idle Test
The idle standard applies to CO emissions from gasoline fueled
engines. Based upon the idle emission data now available to
EPA, any emission reduction brought about by the need to certify to
an idle standard would be minimal. Costs associated with implemen-
tation of the idle test are only the actual cost of running the
additional certification test. No new test equipment is required.
There is also no impact on other costs (e.g. control hardware).
Expressed as a cost per vehicle, the costs are negligible. Because
this is so, a cost effectiveness computation would not be meaning-
ful and will not be attempted.

_1/ Discussed in many statistical texts. See, for example,
"Statistical Design and Analysis of Engineering Experiments,"
Lipson & Sheth, p. 36.
2/ "Mobile Source Emission Factors - Final Document," EPA-400/9-
78-005, March, 1978, TableII-1.
3/ "Average Lifetime Periods for Light-Duty Trucks and Heavy-Duty
Vehicles," EPA Report SDSB 79-24, G. Passavant, November
4/ One alternate procedure for calculating targets can be found
in the Regulatory Analysis of the final 1984 Heavy-Duty Engine
Gaseous Emission Regulations, Chapter VII, Section 5a.
5/ This data was originally submitted by AM to challenge the
extrapolation of df's derived at low mileage to high mileage
(what has been characterized as the "lever arm" effect). The
data does demonstrate that this procedure can result in major
errors for individual vehicles due to the large amount of
scatter in the data points. However, for the case at hand
here, this is not a problem because over the average of a
number of vehicles these errors (due to random scatter) will
cancel out. The importance of the data lies in its demonstra-
tion of basically linear emission deterioration over 100,000
6/ See for example the submission by Caterpillar of August 15,
1979 in response to the Heavy-Duty Engine Gaseous Emission
NPRM, pp. 10-13.
TJ Memorandum, "Fuel Switching," Benjamin Jackson, EPA Office of
Enforcement, August 2, 1979.