United States
Environmental Protection
Agency
Air
Office of Air and Radiation
(ANR-443)
Washington, DC 20460
EPA
Employee Commute Options
Guidance
December 1992
-------
1.0 Introduction
The Clean Air Act Amendments of 1990 (CAAA) require severe and
extreme ozone nonattainment areas and serious carbon monoxide
nonattainment areas to establish programs aimed at reducing
commute trips to the worksites of large employers. The concerns
that lead to the inclusion of this Employee Commute Options (ECO)
provision in the Act are that more people are driving than ever
before and they are driving longer distances. Traffic
congestion has increased significantly as a result. Moreover,
the increase in congestion exacerbates the emissions impact of
the increased number of vehicle miles traveled.
The increase in drivers and the increase in the number of miles
traveled currently offset a large part of the emissions
reductions achieved through the production and sale of vehicles
that operate more cleanly. It is widely accepted that shortly
after the year 2000, the increased emissions caused by more
vehicles being driven more miles under more congested conditions
will outweigh the fact that each new vehicle pollutes less,
resulting in an increase in emissions from mobile sources. The
legislative history of the employer trip reduction provision
includes the following statement:
"If we are ever to reduce or even simply avoid increases in
vehicle pollution, it is not enough to control only the
pollution which each car emits. The use of the car must be
examined as well because the growth in vehicle miles
-------
traveled threatens to overwhelm what can be achieved through
even the toughest tailpipe standards. ... [ The provision
is] intended to reduce future growth of vehicles miles
traveled, but at the same time provide enhanced mobility to
serve ever increasing travel demands.”
The first trip reduction programs in the country were designed to
reduce congestion. Since 1982, approximately 55 Trip Reduction
Ordinances (TRO) have been enacted in the United States. One
such program was established in the City of Pleasanton,
California. Established in October 1984, this TRO was designed
to reduce peak hour traffic by 45%. As air pollution became more
of an issue in some areas, TRO’s were implemented to improve air
quality not only traffic congestion. In July of 1988, the South
Coast Air Quality Management District in California passed
Regulation XV designed to reduce emissions from vehicles
operating between 6:00 am and 10:00 am on weekdays. Three
Average Vehicle Ridership (AVR) targets, based on location within
the District were designed to result in approximately a 25%
increase in average vehicle ridership in the Los Angeles, San
Bernardino, Orange and Riverside Counties. The term Average
Vehicle Ridership is comparable to the term average vehicle
occupancy (AVO), as used below in this guidance.
2.0 The Purpose of This Guidance
The purpose of this guidance is to inform the affected State arid
local jurisdictions of the Clean Air Act requirement, to provide
guidance on preparing an approvable State Implementation Plan
2
-------
(SIP) revision, and to discuss various approaches which may help
areas achieve Clean Air Act targets through implementation
strategies that are the least burdensome and costly to both
affected employers and employees.
This guidance is intended to assist States in developing
approvable SIP revisions, but does not establish or affect legal
rights or obligations. It does not establish a binding norm and
it is not finally determinative of the issues addressed. EPA
approval of any particular SIP revision will be made by applying
the applicable law to the specific provisions in the SIP.
This guidance was initially to be included in the General
Preamble to Title I of the Clean Air Act 157 FR 13498, April 16,
1992]. As a part of the preparation of that document, this
guidance went through an informal comment period which included a
public meeting held June 25-26, 1991 in Reston, Virginia. An
announcement of this public meeting was published in the Federal
Register. In addition, several drafts of the guidance were sent
to a wide spectrum of interested parties for comment including a
review panel for The National Association of Regional Councils.
Subsequently, it was decided that this guidance would be
published separately from the General Preamble.
3
-------
3.0 The Clean Air Act Requirement
Section 182(d) (1)(B) requires that States with severe and extreme
ozone nonattàinmênt areas shall submit a SIP revision requiring
employers with 100 or more employees in such areas to implement
programs to reduce work related vehicle trips and miles traveled
by employees. At a minimum, the SIP revision shall require that
each employer increase its average passenger occupancy per
vehicle (APO) in commuting trips between home and the workplace
during peak travel periods “by” (or in EPA’s interpretatIon of
the intended meaning, “1:0 a level”) not less than 25% above the
average vehicle occupancy (AVO) for all such trips in the
nonattainment area at the time the SIP revision is submitted.
(The AVO refers to the baseline for the nonattainznent area or for
a zone within the nonattainment area if it is divided into zones.
The APO applies to employers.) The SIP revision shall be
submitted no later than November 15, 1992. The revision shall
require that within 2 years after the date the SIP revision is
submitted, employers must submit compliance plans that
convincingly demonstrate that compliance will be achieved no
later than four years after the SIP revision is submitted.
Section 187(b) (2) requires that States with serious CO
nonattairiment areas also adopt such programs.
EPA interprets Section 182(d) (1) (B) of the Act to mean that each
State with a severe or extreme ozone nonattainment area or a
serious carbon monoxide nonattaininent area will establish a
4
-------
process of compliance plan submission, approval, periodic
reporting on target achievement, and periodic compliance plan
revision that aims at the required target.
Areas that have trip reduction programs in effect, either
mandatory or voluntary, that are anticipated to produce
significant increases in the area AVO in the short term have
expressed concern that establishing 1992 as the base year will
penalize them and will discourage employers from cooperating with
current programs. Where such programs are fully implemented and
meet all requirements for an approvable SIP under Sections 110
and 182 of the CA A, the areas may, if they have not already,
make a complete SIP submittal and receive full EPA approval. By
electing to use an earlier SIP submittal, such areas will be able
to recognize the AVO increases resulting from the ongoing efforts
of their program.
States with marginal, moderate, or serious ozone nonattainment
areas or with moderate Co nonattainment areas are not required to
implement an employer trip reduction program as outlined in
Section (d) (1) (B). They may, however, elect to implement such a
program and to follow this guidance. Reductions in emissions
from the ECO program in such areas may receive SIP credit toward
required emission reduction demonstrations, provided that certain
criteria relating to quantification, permanence, and
5
-------
enforceability of credits are satisfied. Emission reduction
estimation techniques will be addressed in a separate guidance.
4.0 Definitions
Definitions of several terms are not provided in the Act itself
but are needed to establish the parameters of employee commute
options programs. States developing ECO SIP submissions may use
the following interpretive definitions or may document that
alternative definitions are more appropriate.
uPeak travel periods” refers to the those hours between which the
morning commute occurs Monday through Friday. EPA believes that
the intent of the Act is to significantly reduce single occupancy
vehicle commute trips to and from work and has defined the peak
travel periods to include either the hours between 6:00 a.m. and
10:00 a.m. or any other period which captures 85% of commute
trips between 5:00 a.m. and 11:00 a.xn. as determined by the
State.
“Work—related trips” and “commuting trips between home and the
workplace” include trips with stops en route to work during the
peak travel periods.
“Vehicle” refers to a highway vehicle powered by a gasoline or
diesel internal combustion engine with fewer than nine seating
positions for adults. States may proposer in their SIP
6
-------
submissions, factors to be applied to the vehicle count which
would reflect the lower emission levels from alternatively fueled
vehicles if they are certified by a government authority as being
substantially lower emitting in actual use than vehicles
generally purchased in the area. EPA will work directly with
interested States on the development of such factors.
Although the statutory language does not explicitly allow
alternatively fueled vehicles to receive less than the full count
of a regular gasoline or diesel fueled vehicle, it is clearly the
intent of the Act as a whole to promote technologies that reduce
emissions. Therefore, EPA feels that certain alternatively
fueled vehicles known to be lower emitting in actual use than
other vehicles arriving at the worJcsIte, may be exempted from
counting as a full vehicle.
“Employee” means any person employed by a firm, person(s),
business, educational institution, non-profit agency or
corporation, government department or agency or other entity, in
a full-time or part—time position who either reports to work or
is assigned primarily to a worksite 80 or more hours per 28—day
period in either a permanent or temporary capacity, on either a
contract or employed basis, excluding volunteers. States may set
up a de minimis level for “temporary capacity,” for example, a
period of one month or less or a different period that is equally
7
-------
appropriate as applied to the area’s employment practices and
ambient air quality conditions.
“Employer” means any person(s) , firm, business, educational
institution, government department or agency, non-profit agency
or corporation, or other entity which employs 100 or more persons
at a single worksite. Only such worksites are subject to the
trip reduction requirement; smaller worksites of the same
employer are not subject to the trip reduction requirement.
Several subsidiaries or units that occupy the same worksite and
report to one common governing board or governing entity, are
considered to be one employer.
EPA interprets that it was Congress’ intent to target employers
who have enough employees arriving in the peak period to
establish a viable ECO program. Congress selected a threshold of
100 total employees, presumably with an awareness that for the
many employers near this threshold there would be fewer than 100
employees arriving during the peak period itself since some
employees will work other schedules. However, there is no
indication that Congress meant the 100 employee criterion to be
applied rigidly in situations in which the majority of an
employer’s workforce follows a non—standard schedule, and the
number of employees arriving in the peak period is both small
enough to be considered de minimis and small enough to make
ridesharing and special employer-provided services difficult. An
8
-------
employer of 100 workers split evenly between three shifts would
have about 33 employees arriving during the peak period. It is
EPA’S judgment that fewer than 33 employees who report to work
during the peak travel period do not constitute enough employees
commuting at that time for an employer to implement a viable trip
reduction program and that such a situation is de minimis.
Therefore, a de minimis exemption may be made at the State’s
option whereby employers with worksites at which fewer than 33
employees report to work during the peak travel period are not
subject to the requirements.
Except provided for in the immediately preceding paragraph, in
determining the number of its employees, an employer includes all
employees from all shifts, seven days a week, not only those who
commute during the peak travel period for that worksite. The
number of employees an employer has is determined as the number
of employees on the payroll (excluding temporary employees whose
term of employment was below the de minimis threshold as averaged
over a year-long period).
“Worksite” means a building, or grouping of buildings located
within the same nonattainment area, or in the same AVO zone of
the nonattainment area if it is divided into AVO zones, which are
in actual physical contact or separated only by a private or
public roadway or other private or public right-of-way, and which
9
-------
are owned or operated by the same employer or by employers under
common control as described under the employer definition.
5.0 SIP Submittal
5.1 Introduction
The Act requires that States submit an ECO SIP revision not later
than November 15, 1992. The ECO SIP needs to include the
following:
• the AVO for the nonattainment area or for each zone if
the area is divided into zones,
• the target APO which must be no less than 25% above the
AVO(s),
• a process for compliance demonstration, and
• enforcement procedures to ensure submission and
implementation of compliance plans by subject
employers.
5.2 AVO Calculation
The baseline AVO may be estimated by simply dividing the number
of employees who report to worksites or other related activity
centers between the peak period inclusive Monday through Friday
by the number of vehicles in which these employees report over
that five-day period. A telephone survey, employer administered
10
-------
survey, and/or available census data may be used to determine the
AVO. Statistical sampling is acceptable.
If a survey is used to estimate the AVO, it should cover a
typical five-consecutive-weekday period, excluding any holidays,
and should occur during a time without holidays bordering the
weekend on either side of the selected week. For example, the
week before Labor Day weekend should not be considered a typical
week. If the ECO program specifies a season for determining
employer APO (see below) the AVO should be determined so as to
include the same season.
The Act states that the AVO applies to all commuting trips
between home and the workplace during the peak travel period.
Therefore, all commuters including those who work for employers
with less than 100 employees and who commute during the peak
travel periods must be included in this estimation.
The Act states that the AVO is for “such trips in the area at the
time the tSIP] revision is submitted.” The AVO may be estimated
prior to the SIP submittal so long as the estimation includes a
projection of the estimated AVO for the time the SIP revision is
submitted. EPA interprets the time of SIP submittal to encompass
a period up to a year prior to the date of SIP submittal.
11
-------
Each State or local jurisdiction calculating the AVO for its
nonattaimnent areas should work in close cooperation with local
governments as outlined in Section 174 of the Clean Air Act.
Each State or other lead agency will also decide whether
nonattainment areas will be subdivided into AVO zones and how
such determinations will be made. The AVO and target APO for
each area and zone will be included in the SIP revision due
November 15, 1992.
5.3 PO Calculation
The target APO for employers is not less than 25% above the AVO.
The APO for each employer with 100 or more employees may be
calculated as the number of employees reporting to the worksite
during the peak travel periods inclusive Monday through Friday
divided by the sum of the number of vehicles in which employees
report during those peak travel period plus or minus any APO
credits.
APO = # employees reporting to worksite during peak period
# vehicles in which employees report ± APO credits
This guidance allows States to provide for vehicles that are
lower emitting than those generally purchased in the area to
count as a percentage of a vehicle, as a measure to help make
this program more cost effective without compromising the
statutory requirement.
12
-------
Average vehicle occupancy and employer APO compliance may be
determined during the ozone season that occurs during the summer
months in cold climate regions, or on a year-round basis at the
State’s option. States may choose, for example, to have all
employer compliance surveys conducted during the summer for areas
where summer is the only ozone season, or may allow such surveys
to be distributed throughout the year. In the latter case,
although one employer may submit its plan at a different time of
year than another employer, each employer would be on a 12 month
or 24 month plan submittal cycle. This would allow employers to
have their compliance reports and next-cycle plans due at the
same time of year each year their plans are due.
5.4 Compliance Demonstration Process
State or local law must establish ECO requirements for employers
with 100 or more employees at a worksite within severe and
extreme ozone nonattainment areas and serious carbon monoxide
areas. Automatic coverage of those employers should be included
in the law. In addition, States should develop procedures for
notifying employers regarding the ECO requirements.
Finally, States and/or local laws must require that initial
compliance plans “convincingly demonstrate” prospective
compliance. Approval of the SIP component addressing the ECO
provision will depend on the ability of the State/local
regulations to ensure that the CAAA requirement that initial
13
-------
compliance plans “convincingly demonstrate” compliance will be
met. This demonstration can take on any of four forms or any
combination of these.
One option is for the State to include in the SIP evidence that
agency resources are available for the effective plan-by-plan
review of employer—selected measures to ensure the high quality
of compliance plans, and that plans that are not convincing will
be rejected.
A second option is for the regulations in the SIP to contain a
convincing minimum set of measures that all employers must
implement. These measures will be subject to review and approval
by EPA as adequate when the SIP is processed.
A third option is for the regulations in the SIP to provide that
failure by the employer to meet the target APO will result in
implementation of a regulation—specified, multi—measure
contingency plan. This plan will be reviewed by EPA as for
adequacy when the SIP is processed.
A fourth option is for the regulations in the SIP to include for
employers who fail to meet the target APO financial penalties
and/or compliance incentives that are large enough to result in a
significant prospective incentive for the employer to design and
implement an effective initial compliance plan of its own.
14
-------
In addition, States and local jurisdictions need to include in
their ECO regulations penalties and/or compliance incentives for
an employer who fails to submit a compliance plan or an employer
who fails to implement an approved compliance plan according to
the compliance plan’s implementation schedule. Penalties should
be severe enough to provide an adequate incentive for employers
to comply and no less than the expected cost of compliance.
5.5 Implementation Approaches
It is very important that State and local jurisdictions design a
system that achieves local goals in the lowest cost manner. EPA
has based much of the previous discussion on the program in
Southern California, but it encourages States and local
jurisdictions to seek innovative methods which may produce lower
cost results. The averaging, trading and banking options,
discussed in section 5.7, have not been used in California during
the first years of the program, but are examples of features
which may lower the overall cost of compliance.
5.6 AVO Zones
Where there are important differences in terms of commute
patterns, land use, or AVO, the States may establish different
zones for the calculation of AVO.
15
-------
5.7 Averaging, Banking and Trading
EPA believes that States or local governments may wish to allow
employers in the same flonattainment area to aggregate APO credits
through averaging, banking and trading as discussed below.
Section 182(d) (1) (B) can be interpreted to allow aggregation of
APO credits among employers so long as each employer increases
average passenger occupancy per vehicle to a level not less than
25% above the AVO, considering any trading. The statutory
language merely requires each employer to increase APO, and does
not specify that the required increase in APO must come from a
given employer’s own employees. Consequently, the statutory
phrase “commuting trips between home and the workplace” can be
interpreted to refer to the trips by any employees in the area
rather than only the employees of a specific employer. Any State
or local jurisdiction that elects trading and banking options
must ensure that an effective tracking system is implemented and
maintained to ensure the integrity of such a credit system.
Programs that incorporate provisions to allow averaging or
trading of emissions generally lower the cost of achieving given
emission reduction targets by shifting the emissions reductions
to the sources that can reduce emissions the most efficiently.
Averaging: An employer with more than one worksite in a
nonattairuuent area may average its APO across those worksites in
the nonattainment area. If the nonattainment area is divided
16
-------
into zones, the employer may only average those worksites within
each zone (unless as described below average commute miles values
for each worksite are applied in banking or a trade). The
average P20 is calculated by adding all of the employees at all
of the worksites in the numerator and dividing by all of the
vehicles in which these employees report to the worksites. If
averaging across worksites is allowed, each compliance plan
should still be worksite specific.
P20 Credit: In order to implement trading or banking in an ECO
program, the State will need to create a unit of P20 credit that
can be banked or traded and can be used to meet the P20 targets.
Employers can readily calculate the maximum number of vehicles
that may arrive at their worksite by dividing the number of their
empj.qyees arriving at the worksite during the peak period by the
to.L get P20. In their simplest form, P20 credits available for
averaging, banking or trading would be based on the number of
vehicles below the maximum allowed in order to meet the target
P20. If credits are defined so simply, the ECO regulations must
have some provisions, for example case by case review of proposed
trades or objective criteria for determining which employers may
trade with each other, to ensure that trades result in
substantially the same reduction in vehicle use and emissions as
if the trade had not occurred. Alternatively, States may include
in their definition of their P20 credits a factor for the average
commute miles of an employer’s employees. An employer could
17
-------
establish standard distances from each worksite based on zip
codes, and from these establish the roundtrip commute distance
for each employee.
When an employer exceeds the target PO, the employer has managed
to have fewer vehicles arrive at the workaite than the maximum
allowed to meet the target APO. Multiplying the number of
vehicles that could have arrived at the worksite but did not by
the average commute miles of the employees results in vehicle
mile credits that an employer may bank or trade. Vehicle mile
credits are APO credits that take into account the distance
commuted to work by an employer’s employees.
For example, assume Employer I had 20 fewer vehicles arriving at
the worksite than were allowed in order to meet the target PO.
The average commute miles for employer A’s employees is 10 miles.
Employer A would have 200 vehicle mile credits available to bank
or trade. Employer B has exceeded the number of vehicles that
can arrive at the work place by 5. Employer B’s average commute
is 20 miles. Employer B can then purchase 100 vehicle miles
credits from employer A in order to meet the target APO. It
should be noted that an employer who does not bank or trade
credits to demonstrate compliance is not required to consider the
distances traveled by its own employees.
18
-------
The State or local government law may allow employers to bank APO
credits to be applied to future compliance demonstrations. It is
EPA’s position that in terms of public health benefits, early
reductions achieved through banking of APO credits offsets later
application of banked credits because as the fleet turns over and
cleaner fuels are employed each vehicle trip generates less
emissions. EPA believes that the use of the banked APO credits
will not materially affect attainment by the required date. EPA
recommends that State and local jurisdictions encourage
banking of APO credits because of the immediate reductions in
emissions that are realized through such a program.
No portion of a nonattainment area can be exempt from employer
requirements as stated in Section 182 (d) (1) (B). However, within
a nonattainment area, certain employers may have a measured APO
below the target APO and acquire APO credits to meet the target.
Full documentation of the traded APO credits should be supplied.
Unused APO credits may be traded more than once.
6.0 Costs and Benefits
EPA suggests that States, local jurisdictions and employers
carefully consider the strategies outlined in this guidance to
minimize the costs of complying with ECO targets. The strategies
that can minimize the social costs of ECO include, for example,
the trading of AVO reduction credits and the use of AVO zones
within the nonattainment area. The two main factors to consider
19
-------
in determining the net costs of the ECO program are the social
costs of ECO, which is related to the expenditures which
employees and/or employers must make, and the benefits of the
program. (A detaIled explanation of ECO program cost estimation
is included in an attached appendix.)
In estimating the costs of ECO it is important to distinguish
between expenditures and social costs. Employers’ and/or
employees’ expenditures do not equal the total social costs,
because part of these expenditures are “transfers” of resources
from one party to another. The social costs include the value of
the additional time, effort and inconvenience that commuters
experience when they reduce motor vehicle use with a resulting
increase in their employer’s APO.
A rough estimate of the total direct social cost of the ECO
program for the 11.7 to 13 million employees who work for covered
employers in the ten major nonattainment areas is $1.2 to $1.4
billion per year.’ This estimate is based on evidence that
compensation of $6.17 per day is representative of the cash
incentive employees who forego the use of their single occupancy
1 This range is a function of two variables —— the number of employees
affected, and the current percentage of solo drivers assumed. The $1.2 billion
dollar estimate assumes 11.7 million affected employees based on figures provided
by the nonattairunent areas subject to ECO and a 73% drive alone rate based on 1990
Census statistics. The $1.4 billion dollar estimate assumes 13 million affected
employees based on 1986 Census data for the number of employees in the nonattainment
areas subject to ECO multiplied times 45% (the percent of all workers who work for
employers with more than 100 employees in the Chicago metropolitan area.) The $1.4
billion dollar estimate also assumes a national drive alone rate of 75% based on a
1987 report by the ENO Foundation for Transportation, Inc.
20
-------
vehicle would have to receive in order for employers to achieve a
20% reduction in automobile use. This compensation estimate is
derived from studies of commuter choices in downtown Los Angeles.
It is uncertain how closely these represent conditions in the
other affected nonattainment areas. (See complete discussion
and explanation of methodology in Appendix.)
EPA believes that the establishment of ECO programs may yield
important benefits, beyond the obvious one of emission reduction.
ECO programs will reduce the use of single—occupant automobiles
for commuting by employees working at covered employers,
potentially reducing congestion, fuel use, and commuting time for
all motorists still using the highways during the commute period,
including noncommuters. (The values of changes in congestion,
fuel use, and time for employees who make a switch are reflected
in the estimate of $6.17 per day.) Also, less need for parking
space may release valuable real estate for alternate uses. In
addition, states may use the reductions in emissions in achieving
compliance with their nonattaininent areas’ reasonable further
progress and VMT offset requirements, thereby avoiding the cost
of other measures which would have been needed.
Extrapolating from recent work by Shoup and Wilson 2 , the value
of commuting time reductions may be estimated to range from
2 Donald C. Shoup and Richard W. Wilson “Commuting, Congestion, and Pollution:
The Employer—Paid Parking Connection” prepared for and presented at the Congestion
Pricing Symposium, May 1992.
2].
-------
approximately $447 - $511 million dollars annually 3 . This
social cost savings combined with the uncalculated pollution
benefits, at least partially, offsets the $1.2 to $1.4 billion
cost. However, this estimate of time savings is only an
approximation, as the true savings depends on the subjective
value commuters place on time savings, and on the specific
relationships between reduced road volume and time savings on the
many roadways that may be affected. Moreover, additional
uncertainty in estimating time savings (and emission reductions)
arises from the possibility that reduced congestion and shorter
commuting times may lead some employees of non—covered employers
to switch away from transit or car pools back to single-occupant
vehicles. As a result, VMT and commute times stabilize at higher
values than if there had been no such “latent demand” effect (but
still no higher than without the ECO program under any
circumstance).
7.0 Best Practices
Because various trip reduction programs including transportation
demand management (TDM) programs have been in place for almost
ten years, there is some experience to draw on for employers who
are subject to the ECO provision.
Shoup and Wilison estimate that each vehicle represents $262 dollars in
congestion—related social costs. The $262 cost per vehicle for each additional car
on the road reflects an increase in: 1) accidents 2) time/cost of congestion 3)
fuel costs. $447 million 11.7 million (affected workers) x .73 (percent of drive
alone based on 1990 Census) z .20 (percent that need to switch) x $262. $511
million = 13 mi1li n (number of employees in nonattainment areas subject to ECO z
.45) x .75 (percent of drive alone based on 1987 ENO Foundation report) x .20
(percent that need to switch) x $262.
22
-------
There are a number of factors that are commonly found at
companies with successful programs. These include:
• Support of both upper and middle management
• Appointment of a cpialified and committed employee
transportation coordinator (ETC)
• Effective market research on the commute patterns of
employees and transportation options available to
employees
• Selection of an appropriate and effective set of
measures including incentives and disincentives to
encourage employees to select non drive-alone modes.
There are many considerations to take into account when an
employer selects measures to put into a compliance plan. Key
among these include: the availability of parking, the cost of
parking, and the availability of mass transit. Of course each
worksite is unique and an effective program is designed to meet
the needs of employees at each particular worksite.
Some examples of measures employers may include in compliance
plans are listed below. This list is not all-inclusive and the
measures are not required per se.
1) Offer cash incentives
2) Cashing out parking 4
An employer offers to provide a cash allowance to an employee equivalent to
the parking subsidy that the employer would otherwise pay to provide the employee
with a parking space.
23
-------
3) Institute compressed work weeks
4) Charge those who drive alone for parking
5) Sponsor and/or subsidize carpools and vanpools
6) Subsidize use of public transit
7) Provide comprehensive rideshare matching service
8) Subsidize mid-day shuttles to local shopping areas
9) Provide company-owned vehicles for ridesharing
10) Offer preferential or subsidized parking for carpools
and vanpools
11) Provide a guaranteed ride home program
12) Improve facilities to promote bicycle use
13) Promote establishment of on-site amenities
14) Offer telecommuting and work-at-home options
In February 1990, the Federal aighway Administration (FHWA)
published the “ Evaluation of Travel Demand Management Measures to
Relieve Congestion.” The goal of the study was to determine
whether particular TDM efforts have had a measurable impact on
traffic volumes. In reviewing TDM programs in different regions
of the country, 11 employers were selected for case studies.
Determination of the percent of vehicle trips reduced was used as
the measure of success for each program. Based on this
criterion, several of the employers selected were very successful
in their TDM programs. The study concludes that TDM can reduce
low occupancy vehicle trips to a site and that the degree of
success is directly determined by the specific components of the
TDM program.
24
-------
The FHWA study also examined the factors which contributed to
successful programs. Employer size and the density of
development around an employer’s site were not factors which
predicted the success of a company’s efforts. One of the most
powerful indicators of program success was the offering of
transportation allowances or subsidies. Parking also plays a
critical role in influencing commuter behavior. Charging for
parking and/or restricting access to parking provide a
disincentive to solo-driving, but may also result in a cost
reduction or savings to the employer.
The top three employers in FHWA’s study all had restricted
parking and charged for parking. It appears that a very
effective approach is for an employer to implement an appropriate
combination of transportation allowances or subsidies, parking
charges, and parking restrictions. This approach may be the most
economic for employers as well if the parking charges exceed the
real cost of parking and are applied toward the subsidies given
to employees who use non drive-alone modes. One employer
supplied all employees with a monthly $40 transportation
allowance regardless of how they arrived at work. Parking was
restricted and employees driving alone were charged $40 for
parking although this was above the actual cost of parking to the
employer. The parking revenue was committed to finance direct
subsidies to carpoolers and transit users. Transit users were
given a $15 monthly pass discount and carpoolers were given a
25
-------
free parking space, both in addition to the $40 transportation
allowance. By restricting parking and charging for parking,
providing a transportation allowance, and subsidizing high
occupancy modes, this employer succeeded in shifting the drive
alone percentage of its employees from 89% in December 1986 to
54% in January 1988. This is a significant shift in just over a
year’s time.
Among the employers studied in the FHWA study, complete cost data
were available for only four of the eleven sites. The two most
effective programs were the only ones among the top five to have
cost information available. Interestingly, based on the net cost
of the TDM program, both employers came out ahead as a result of
implementing their TDM programs. Clearly, this does not indicate
that all successful ECO programs will result in a negative net
cost to the employer. However, it does demonstrate that. there
will likely be some employers among those who implement cost
effective measures to comply with the ECO provision of the Clean
Air Act who do so without enduring an undue economic burden.
Additional information about strategies that employers may use in
implementing the ECO requirements in the Clean Air Act may be
found in the U.S. EPA document, “Transportation Control Measure
Information Documents,” published by the Office of Mobile Sources
in May, 1992.
26
-------
APPENDIX A
METHODOLOGY FOR ROUGH ESTIMATION OF ECO SOCIAL COSTS
The discusSiOfl below outlines the derivation of a rough
estimate of the social costs for the ECO program.
A key part of the costs of ECO program is the additional
time, inconvenience and expense that workers who drive to work
would incur if they had to commute by other means. A standard
economic approach to estimating such costs is to assume direct
cash compensation is offered to reduce employee car use for
commuting. Although many employers may achieve ECO goals by
other means, there is no reason to believe that these will on
average be more efficient, since none provides the same
flexibility to workers that they may have under cash
compensation.’ Estimation using direct compensation may use
empirical evidence which shows how commuter mode choice changes
in response to the various costs of commuting. However, it may
also be noted that the Clean Air Act, as interpreted by EPA, does
not make achieving the AVO target enforceable or make employers
begin with particular incentives unless the State chooses to do
so. It is possible, therefore, that not, all of the annual costs
estimated above will begin to occur right away. The approach
used here considers the social costs of ECO in two parts; (1) the
costs borne by workers who reduce car use in order to achieve ECO
goals; (ii) the costs incurred for program implementation and
administration.
Costs borne by Workers who Reduce Car Use :
The direct compensation that employers must pay employees to
reduce commuting by automobile can be used to measure the social
cost incurred by these employees as they adopt other, less-
preferred means of getting to work. To estimate the direct
compensation necessary to meet the requirement of §182(d) (1) (B)
of the CAP , empirical evidence can be used to estimate how
commuters choice of travel mode depends on cost. Such evidence
of commuter behavior provides a powerful tool to measure
commuter’s valuation of particular modes;
(empirical data) present real world results where
people have shifted their commuting modes in
response to changes in the monetary costs of
commuting. This method implicitly takes into
account all of the factors relating to the
‘Several studies conclude that cash incentives (and/or commute subsidies) are
one of the most powerful indicators of the success of employer trip reduction
programs. (See: Cambridge Systematic, Inc., “Effects of Demand Management and Land
Use on Traffic Congestion: Literature Review”, December 1991, page 93. Prepared
for USDOT, contract number DTFH61—91—C—00085. Also see: Comsis Corporation,
“Evaluation of Travel Demand Management Measures to Relieve Congestion”, February
1990, pages 25—27. Prepared for USDOT, report number FHWA—SA—90—005.)
1
-------
advantages and disadvantages of switching from
driving alone to carpooling or public
transportation. n2
Empirical results on the effect of direct compensation on
commuting mode choice are available from research on the demand
for parking by Shoup and Wilison (1992) . They used a
multinomial logit analysis of Commuting among office-workers in
downtown Los Angeles and found that workers who had to buy
parking at market prices used 17% fewer cars to commute to work
than did similar workers who received employee paid parking. In
their sample the market value of parking was $4.15 per car per
day in 1986 dollars, or $5.25 in 1992 dollars. 4 Assuming that
the market demand function for parking is linear, compensation of
$6.17 per car per day would be needed to achieve the §182(d) goal
of a 20% reduction in vehicle use. 5
This estimate of direct compensation costs must be applied
across the whole population of affected workers. 6 There are two
ways to estimate the number of affected employees. One approach
is based on figures provided by the nonattainment areas subject
to the ECO program resulting in an estimate of 11.7 million
affected employees. The second estimate is derived by taking the
total number of employees in the affected nonattainment areas
based on 1986 Census data and multiplying it by 45%, which is the
estimated percentage of employees at affected employers in the
Chicago metropolitan area. The second approach results in an
estimate of 13 million affected employees.
2 B. Galef, E. Chu, T. Bansal, ICF Inc. Memorandum to Tern Whale, E k. May
5, 1992.
D. Shoup and R. Wilison, “Employer—paid Parking; The Problem and Proposed
Solutions”, Transportation Quarterly, June 1992.
Using the CPI for 1986 of 110, and for 1992 of 140, the average parking
price in ‘92 dollars = $4.15 x (140/110) = $5.25.
The assumption of linearity is a first—order approximation of general
parking demand that is valid for small changes in price. We thus extrapolate to get
the price needed to induce a 20% reduction in automobile use: $5.25 x (20%117%)
$6.17.
The compensation estimate of $6.17 derived from Los Angeles commuters may
or may not be representative of other ECO cities. On the one hand, since Los
Angeles is one of the least compact ECO cities, we could expect that opportunities
for carpooling or mass transit would be more limited than in other ECO areas. In
this case, the $6.17 would be an overestimate of the compensation needed to reduce
commuting car use by 20% in other ECO cities. On the other hand, Shoup and
Wilison’s data are from office workers in the central business district of Los
Angeles. For these workers in this particular areas, the opportunities for
arranging carpools or using mass transit may be greater than for the rest of the Los
Angeles nonattainment area. Thus the estimate of $6.17 would be too low for the
whole Los Angeles nonattainment area. Given such offsetting effects and the lack
of other comparable data, we believe it is reasonable to rely on Shoup and Willson
to generalize across the other ECO cities.
2
-------
The next step is to determine the number of solo drivers
among the affected employees. Based on 1990 Census data, 73% of
the population drive to work alone. Based on a 1987 Eno
Foundation report, an estimate of the number of employees driving
alone would be 75% of the affected commuters. 7 Compensating 20%
of this group to switch to non-solo driving commuting would
require expenditures ranging from $2.7 to $3.1 billion. 8
The social cost of this direct compensation must be
calculated by netting out transfers. The amount of the transfers
depends on both marginal income tax rates and the characteristics
of the demand curve for automobile use for commuting. Assuming a
marginal tax rate of 25%, the additional amount of taxes
generated by the compensation payments ranges from $700 million
of the $2.7 billion to $780 million of the $3.1 billion. This is
a transfer to the government and not a social cost. Of the
remaining $2.0 - $2.3 billion, some is a pure gain to employees,
and some represents the cost to those employees who switch modes.
Assuming as a first—order approximation that demand for parking
is linear in the relevant range, the social cost to the workers
who previously commuted by driving alone is $1.0 - $1.2
billion. 9 The assumption of linearity is generally seen as
appropriateF “the assumption of linear demand is a middle—ground
assumption rather than one that falls at either extreme. ”’°
Costs Due to Program Administration :
There are opportunity costs associated with the
administration and implementation of ECO. Every dollar spent on
commuting surveys, plan development, plan coordination, program
ENO Foundation for Transportation, Inc., “Commuting in america”, 1987, p. 53 .
They report that solo drivers account for 77.3% of metropolitan residents
nationwide.
8$27 billion/yr = $6.17/day x 260 working days/yr x 11.7 million workers x
73% (solo drivers) x 20% (switchers). $3.1 billion/yr $6.17/day x 260 working
days/yr x 13 million workers x 75% (solo drivers) x 20%(switchers).
It is important to point out that employers will have to compensate more than
just the 20% of solo drivers who agree to shift modes. Employees who, pre—ECO, did
not solo drive, will still demand the ECO compensation. These employees can
credibly threaten to put aside their bicycles, carpools or bus passes and resume
solo driving if they are not paid the same compensation as those who agree to stop
solo driving only to help achieve ECO goals. These compensations, however, are
transfers, and not net social costs, since these individuals do not have to change
their activities in anyway. The social costs associated with these payments are the
renegotiations of labor contracts that such compensation engenders. Such
renegotiations are potentially costly, but the absence of data prevents any
quantification of such costs.
$1.0 billion = $2.7 billion x 75% x 0.5. $1.2 billion = $3.1 billion 75%
x 0.5. The social costs to these workers would be higher if the demand curve was
concave to the origin, and lower if convex.
10 ICF Memo, May 5, 1992
3
-------
monitoring and employees’ time is one dollar of social cost. 11
Preliminary assessments of trip reduction programs suggest annual
administration costs of $3500 per year for a typical firm of 200
employees. This figure averages to about $17.50 per employee per
year, or $200 - $230 million across all covered workers.’ 2
(These administrative costs at covered firms will also have labor
market distortion affects, with resulting social costs. These
costs are likely to be small and so are not considered in
detail. 13)
Total Social Cost :
The total social cost of this ECO program are thus estimated
to fall in the range of $1.2 - $1.4 billion per year.’ 4 This
overview is meant to be illustrative, although parameter
estimates have been used which are consistent with the data.
This approach provides a reasonable estimate of the social costs
of the ECO, however, States and other responsible parties are
encouraged to improve upon this assessment by making further
refinements.
Actual employer ECO requirements will be determined on a State—by—State
basis. However, as is common with State Implementation Plan provision, EPA will
likely compel States to require firms to produce thorough and verifiable baseline
surveys, trip reduction plans, and monitor compliance regularly.
12 The $3500 estimate is derived from estimates of: survey costs of $300 per
year (5 days x 4 hr/day x $15/hr x 1 survey/yr); statistical analysis of survey data
of $400 per year (8 hr x $50/hr x 1 survey/yr); employee familiarization with ECO
of $1600 per year (1/2 hour per employee x 200 employees x $16/hr x 1 time/yE);
staff management of ECO of $1000 per year; and one—time ECO plan development of
$2000 per plan (about $200 per year). Data showing roughly similar or higher levels
of administrative costs comes from surveys of Regulation XV plans by the Chicago
Area Transportation Study (F. Gerald Rawlings letters to Andrew Plummer, May 20,
1992, and June 12, 1992). One plan they reviewed for a firm of 115 employees
reported annual survey costs of $1290, one—time plan preparation costs of $3000, and
annual ECO management costs of $4680 per year. Another firm, with 354 employees,
reported one—time plan preparation costs of $4350.50, and several thousand of
dollars in annual ECO management costs.
u Firms will endure an increase in the cost of employment for two reasons.
First, the administrative cost of implementing ECO has a component that increases
with employment, and second, the cost of compensating workers who would prefer to
drive to work raises costs above what they would otherwise be. Preliminary
investigations of the magnitudes of theses costs indicate that they are likely to
be low relative to the total costs of ECO.
I4 The total social cost equals the tax transfer and the pure gain subtracted
from the expenditures required to switch solo drivers to a non—SOV mode plus the
administrative costs. $1.2 billion = $2.7 billion — $700 million — $1.2 billion ÷
$200 million. $1.4 billion $3.1 billion — $780 million — $1.1 billion + $230
million.
4
------- |