GSA Fleet: Information on the Effect of Donating Cars to	 
YouthBuild USA and Potential Benefits to Rural Youthbuild	 
Participants (08-DEC-06, GAO-07-153).				 
                                                                 
To assist youth who live in high poverty rural areas obtain and  
retain jobs, YouthBuild USA, a national nonprofit organization,  
has proposed providing donated used cars to selected low-income  
youth in rural communities. YouthBuild USA's proposed program	 
hinges on receiving donations of used cars from the federal	 
government's General Services Administration (GSA). This report  
discusses (1) the effect of donating 1 to 5 percent of selected  
GSA used cars on GSA's fleet vehicle sales operations, (2) what  
studies have shown with respect to the benefits that car	 
ownership or access may hold for low-income individuals, and (3) 
what studies of selected low income car ownership programs and	 
experiences of these programs have shown with respect to the	 
benefits of participant car ownership. In conducting this study, 
GAO examined auction data from GSA, reviewed academic studies on 
the benefits of car access in gaining employment, and interviewed
officials of six existing low income car ownership programs.	 
-------------------------Indexing Terms------------------------- 
REPORTNUM:   GAO-07-153 					        
    ACCNO:   A63966						        
  TITLE:     GSA Fleet: Information on the Effect of Donating Cars to 
YouthBuild USA and Potential Benefits to Rural Youthbuild	 
Participants							 
     DATE:   12/08/2006 
  SUBJECT:   Disadvantaged persons				 
	     Employment assistance programs			 
	     Federal aid to localities				 
	     Federal property					 
	     Gifts or gratuities				 
	     Nonprofit organizations				 
	     Program evaluation 				 
	     Rural economic development 			 
	     Transportation					 
	     Youth						 
	     Youth employment programs				 
	     Program goals or objectives			 
	     GSA Interagency Fleet Management System		 
	     Low Income Car Ownership Programs			 
	     HUD Youthbuild Program				 

******************************************************************
** This file contains an ASCII representation of the text of a  **
** GAO Product.                                                 **
**                                                              **
** No attempt has been made to display graphic images, although **
** figure captions are reproduced.  Tables are included, but    **
** may not resemble those in the printed version.               **
**                                                              **
** Please see the PDF (Portable Document Format) file, when     **
** available, for a complete electronic file of the printed     **
** document's contents.                                         **
**                                                              **
******************************************************************
GAO-07-153

   

     * [1]Report to the Ranking Minority Member, Subcommittee on Financial
       Institutions and Consumer Credit, Committee on Financial Services,
       House of Representatives

          * [2]December 2006

     * [3]GSA FLEET

          * [4]Information on the Effect of Donating Cars to YouthBuild USA
            and Potential Benefits to Rural Youthbuild Participants

     * [5]Contents

          * [6]Results in Brief
          * [7]Background
          * [8]GSA Fleet Would Face Reduced Revenues and Need Legislative
            Authority to Donate Cars

               * [9]GSA Faces Reductions in Revenues from Donating Cars It
                 Would Normally Auction
               * [10]GSA Would Need Statutory Authority to Donate Vehicles
                 Directly to YouthBuild USA

          * [11]Available Research Suggests That Car Access Leads to Jobs and
            Other Benefits

               * [12]Studies Reported Employment Benefits from Car Access
               * [13]Studies Could Not Be Used to Estimate How Much
                 Youthbuild Participants Would Benefit from Having a Car

          * [14]LICO Programs Report Positive Results, but Limitations
            Restrict Projecting Results to YouthBuild USA's Program

               * [15]Studies and Experiences of Six LICO Programs Report
                 Positive Results for Participants
               * [16]Limitations in Studies Do Not Allow for Outcomes to Be
                 Projected to YouthBuild USA's Proposed LICO Program
               * [17]Differences in Design and Participants the Programs
                 Serve Also Limit Application of Study Results to YouthBuild
                 USA's Proposed Program

          * [18]Agency Comments and Our Evaluation

     * [19]Objectives, Scope, and Methodology
     * [20]Summary of Studies on the Effect of Car Access on Employment
     * [21]Comments from the General Services Administration
     * [22]GAO Contact and Staff Acknowledgments

Report to the Ranking Minority Member, Subcommittee on Financial
Institutions and Consumer Credit, Committee on Financial Services, House
of Representatives

December 2006

GSA FLEET

Information on the Effect of Donating Cars to YouthBuild USA and Potential
Benefits to Rural Youthbuild Participants

Contents

Figures

December 8, 2006Letter

The Honorable Bernard Sanders Ranking Minority Member Subcommittee on
Financial Institutions     and Consumer Credit Committee on Financial
Services House of Representatives

Dear Mr. Sanders:

Youth between the ages of 16 and 24 who live in high poverty areas can
face significant obstacles to finding employment and receiving job
training, including low levels of academic achievement, limited work
experience, and a scarcity of jobs in their communities. In rural
communities, these youth may face the additional challenge of a lack of
transportation to get to available job opportunities because of scarce
public transportation and, according to job training and other social
service providers, having a car is often a necessity for obtaining and
keeping a good job. To address this challenge, YouthBuild USA, a national
nonprofit organization working to increase the number of youth
transitioning out of poverty, has proposed providing donated vehicles to
selected low-income youth in rural communities to travel to job training
sites and to work. YouthBuild USA's proposed Rural Initiative Low Income
Car Ownership (LICO) program hinges on receiving donations of used cars
from the federal government's General Services Administration (GSA).

GSA purchases new vehicles and then leases them to federal agencies
through its Fleet program. Each year, GSA sells at auction approximately
40,000 of these vehicles, most of which are 3 to 4 years old. The proceeds
from these auctions help fund purchases of new vehicles for the Fleet
program. Under its proposed LICO program, YouthBuild USA is seeking
donations of 1 to 5 percent of the vehicles GSA auctions off annually,
which it would then provide to participants in selected rural Youthbuild
programs. Because GSA's Fleet program is currently self-sustaining and
YouthBuild USA does not yet have experience operating a LICO program, you
asked us to report on the implications of YouthBuild USA's proposal.
Specifically, this report (1) assesses the effect of donating 1 to 5
percent of selected GSA used cars on GSA's fleet vehicle sales operations,
(2) describes what studies have shown with respect to the benefits that
car ownership or access may hold for low-income individuals, and (3)
describes what studies of selected LICO programs and experiences of these
programs have shown with respect to the benefits of participant car
ownership.

To assess the effect on GSA's fleet vehicle sales operations from donating
vehicles, we focused our analysis on GSA compact sedans to determine the
average opportunity cost--the reduction in revenue for each compact sedan
GSA would donate--and the total reduction in sales revenue GSA would face
by donating 1 to 5 percent of its used compact sedans (112 to 559 cars) to
YouthBuild USA's proposed Rural Initiative LICO program. To describe the
results studies have shown with respect to benefits of car access (that
is, owning a car or having access to one), we identified and reviewed
academic studies that had been subject to a peer review and spoke with
experts. To describe what some LICO programs have shown with respect to
the benefits of participant car ownership, we identified six LICO programs
that had been the subjects of external reviews of their programs'
outcomes. We also met with officials from these LICO programs to learn
about how their programs operated and how they reported outcomes. Appendix
I provides additional details on our objectives, scope, and methodology.
We conducted our work from May 2006 to November 2006 in San Francisco,
California, and Washington, D.C., in accordance with generally accepted
government auditing standards.

Results in Brief

GSA's annual sales revenue would be reduced by an estimated $600,000 to $3
million by donating 1 to 5 percent of the compact sedans from its Fleet
program (112 to 559 cars) to YouthBuild USA each year. GSA receives no
direct appropriations to operate the Fleet and sustains the program
through the fees it charges to federal agencies for leasing vehicles and
the proceeds from selling its used vehicles. GSA currently does not donate
vehicles from its Fleet program. For GSA Fleet to donate cars directly to
YouthBuild USA, GSA would need new statutory authority because such direct
donations would deviate from the existing process for disposing of surplus
federal property. To recover the reduction in revenues it would face from
donating vehicles to YouthBuild USA, GSA officials indicated that the
agency would seek appropriations and consider increasing the leasing rates
it charges the federal agencies that lease vehicles from it. However, GSA
would need additional statutory authority to increase its leasing rates to
recover the costs of a donation program because presently its rates may
only reflect the costs of operating and replacing its fleet. GSA officials
also indicated they would consider keeping vehicles longer than the 3 to 5
years they currently do, which would result in an older fleet with higher
maintenance costs.

Taken as a whole, available studies consistently reported that car access
increases the likelihood that individuals with low incomes (such as rural
Youthbuild participants) obtained jobs. The research lists several reasons
as to why having access to a car leads to better chances of finding a job,
such as the possibility that a car allows an individual to search for a
job over a wider geographic area. Differences between the individuals who
were part of these studies and rural Youthbuild participants did not allow
us to use the studies' results to identify the degree to which
participants in YouthBuild USA's proposed LICO program would benefit from
having a car. For example, the individuals in four of the studies analyzed
were mainly urban welfare recipients who tended to be older, more
educated, and more likely to be employed than the average participant in
YouthBuild USA's proposed LICO program.

Similar to the studies on car access generally, six studies of LICO
programs and the experiences of officials of these programs indicate that
participants reaped benefits from owning a car, such as getting and
retaining jobs, earning higher wages, and spending more time with their
families. However, the studies themselves also had methodological
constraints (such as low response rates in surveys of participants) that
make it difficult to project from their results. For example, a study of
one program reported that 75 percent of respondents said that they got a
job that paid higher wages, and 55 percent reported obtaining better
quality day care for their children as a result of securing a car through
the program. The officials operating this LICO program noted that
participants found jobs (and kept them longer) and improved their quality
of life. However, researchers obtained responses from 38 percent of
participants in the program, which is too low of a response rate to apply
the results to all of the program's participants. LICO program officials
also noted that their programs' designs were different than the YouthBuild
USA proposal, which could also limit the applicability of the studies'
outcomes. For example, most LICO programs we reviewed require participants
to obtain a loan to purchase the car, while YouthBuild USA proposes to
give cars to participants for a one-time fee of $450. LICO program
officials believe that the loan aspect of their programs requires
participants to devote more resources and effort toward obtaining a car
and, as a consequence, participants become more invested in achieving the
goals of the programs. LICO program officials also noted that their
programs provided participants additional support, such as financial
literacy training and arrangements for covering car repair costs, that is
not available under YouthBuild's  proposed LICO program.

We make no recommendations in this report. We provided a draft of this
report to GSA for its review and comment. GSA found the report to be
accurate as it pertained to the description of the GSA Fleet program but
expressed concerns with the potential use of its Fleet for a car donation
program.

Background

GSA purchases about 35,000 to 40,000 vehicles annually for its Fleet
program and manages an inventory of almost 200,000 vehicles, including
sedans, passenger vans, trucks (light, medium, and heavy), buses,
ambulances, alternative fuel vehicles, and limited special purpose
vehicles. GSA then leases these vehicles to 75 participating federal
agencies^1 in the United States, Europe, and Puerto Rico.^2 As part of its
leasing arrangement with these agencies, GSA provides maintenance,
repairs, fuel, and management of accident claims and gets reimbursed for
these costs by the participating agencies.^3 As part of a regular
replacement schedule, GSA sells older vehicles in its fleet. The agency
uses a nationwide network of commercial auction firms to dispose of and
sell about 35,000 to 40,000 of its used vehicles annually. Federal
agencies may dispose of property, such as GSA's vehicles, only in the
manner authorized by statute. Specifically, GSA auctions vehicles from the
Fleet program under the "exchange/sale" authority contained in the Federal
Property and Administrative Services Act (Federal Property Act).^4 Under
this authority, an executive agency may acquire personal property by
exchanging or selling similar items and applying the exchange allowance or
proceeds of sale, in whole or in part

payment, for the property acquired.^5 GSA uses the sales proceeds from
these auctions to help purchase new vehicles.^6 Figure 1 illustrates GSA's
process for leasing and subsequently auctioning vehicles.

Figure 1: GSA Leasing and Auction Process

Welfare reform experts contend that transportation is an important element
in assisting former welfare recipients with finding employment.^7 However,
they also contend that public transportation is not always convenient or
accessible and, for some families receiving assistance, driving is the
best option. To address this issue, several communities started LICO
programs as highly individualized initiatives designed to meet local
transportation needs. In 2002, the National Economic Development and Law
Center documented at least 60 LICO programs across the country serving
welfare recipients and the working poor by helping with the high costs
associated with car ownership, including maintenance, repairs, and
insurance.^8 Typically, these programs rely on older cars received through
donations from individuals. They employ a number of strategies that
include making affordable and reliable used vehicles directly available to
customers or providing low-cost loans to enable individuals to buy
vehicles. Today, there are over 160 documented programs across the country
serving the car ownership needs of low-income individuals.

YouthBuild USA, which to date has not operated its own LICO program, has
proposed a Rural Initiative LICO program that would rely on 3- to
4-year-old vehicles donated by GSA to provide affordable and reliable
transportation for rural youth. YouthBuild USA is a national nonprofit
organization that provides staff training and technical assistance to the
nationwide network of almost 200 local Youthbuild programs.^9 The local
programs serve youth between the ages of 16 and 24 and focus primarily on
providing training in the building trades. YouthBuild USA is proposing to
obtain donated vehicles directly from GSA Fleet and provide these vehicles
to eligible youth so that they can continue in the job training program or
have reliable transportation to work sites or college after they have
graduated from the program. Under the proposal, YouthBuild USA will
identify eligible rural Youthbuild programs and youth at these sites who
would benefit from a donated vehicle. The program would have several
requirements for participants, including possession of a valid driver's
license, eligibility for insurance, good attendance in the Youthbuild
program, and successful completion of a 6-week car ownership course. 
According to program officials, YouthBuild USA would hold the title of the
car for 3 years, during which the participant would have to demonstrate a
good track record for preventative maintenance in order to fully own the
vehicle. Rural Youthbuild sites participating in the program would have to
demonstrate financial stability and the capacity to administer and provide
project oversight on the local level. Figure 2 presents a diagram of the
proposed YouthBuild USA Rural Initiative LICO program.

Figure 2: YouthBuild USA's Proposed Rural Initiative LICO Program

GSA Fleet Would Face Reduced Revenues and Need Legislative Authority to
Donate Cars

If GSA were required to donate 1 to 5 percent of its compact sedans to
YouthBuild USA (112 to 559 cars), its annual sales revenue would be
reduced by an estimated $600,000 to $3 million. GSA Fleet, which manages
the agency's program, receives no direct appropriations and depends on the
sale of these vehicles to sustain its operations. GSA does not currently
donate Fleet vehicles. Furthermore, it would need new statutory authority
to be able to donate them directly to YouthBuild USA because this would
deviate from the existing process for disposing of excess federal
property. If GSA were required to donate cars directly to YouthBuild USA,
it would seek an appropriation to recover the reduction in revenues this
would cause and consider increasing its leasing rates to federal agencies.
However, GSA would need additional new statutory authority to allow it to
increase its leasing rates for the purpose of recovering costs associated
with donating cars.

GSA Faces Reductions in Revenues from Donating Cars It Would Normally
Auction

GSA would face reduced sales revenues of an estimated $600,000 to $3
million per year if it donated 1 to 5 percent of its used compact sedans
from its Fleet program (112 to 559 cars) to YouthBuild USA's proposed
Rural Initiative LICO program rather than sell these cars through selected
auction houses around the country. From fiscal year 2002 through fiscal
year 2006 (as of August 2006), GSA, on average, auctioned 11,171 compact
sedans each year, with a mean sales price of $5,511. GSA officials
indicated that they base their decision to sell their used cars on a
combination of factors intended to maximize their revenues. For example,
they look at the expected sales proceeds of the vehicle based on its age
and mileage to determine at what point they will get the maximum value for
selling their used cars. They currently use the following age and mileage
guidelines for selling their used cars:

o3 years old and 36,000 or more miles, or

o4 years old and any miles, or

oany age and 60,000 or more miles.

In addition, GSA tracks the resale market to determine the high and low
points of the market to help decide when to sell and what types of
vehicles to sell. Finally, they look at events that could affect the used
car market. For example, Hurricane Katrina increased the demand for used
vehicles in parts of the country that were not affected by the hurricane.

If GSA donated to YouthBuild USA 1 percent of the compact sedans it
normally sells at auctions, this would be about 112 cars with a total
reduction in estimated sales revenues of about $600,000. If GSA donated 5
percent of the compact sedans it normally auctions, this would be about
559 cars with a total reduction in estimated sales revenues of about $3
million. Figure 3 shows the range of reduction in sales revenue from GSA
donating 1 to 5 percent of its compact sedans to YouthBuild USA (112 to
559 cars).

Figure 3: Estimated Average Annual Reduction in Sales Revenue to GSA Fleet
from Donating Compact Sedans Normally Sold at Auction

GSA Fleet, which manages the agency's program, does not receive direct
appropriations from Congress; therefore, GSA officials indicated that GSA
Fleet would need to replace the reduction in sales revenues from donating
cars in order to continue to sustain its operations. Currently, to support
its Fleet operations, GSA relies on the proceeds of its auction of used
vehicles and the income it receives from the rates it charges agencies
that lease vehicles from it. According to GSA officials, in setting its
leasing rates, it is allowed to include an increment to these rates to
cover inflation on its current inventory of vehicles, as well as to cover
the estimated replacement cost of these vehicles to meet the demand of
agencies that lease from the Fleet program. This increment is known as
replacement cost pricing. GSA officials indicated that a revolving fund
sustains the Fleet program, with the revenues it receives from auctioning
and from leasing vehicles, offsetting the expenditures for operating the
entire fleet of vehicles.

GSA Would Need Statutory Authority to Donate Vehicles Directly to
YouthBuild USA

In order for GSA to donate cars from its Fleet program directly to
YouthBuild USA, it would need new statutory authority to deviate from the
existing process for disposing of surplus federal property.^10 GSA Fleet
does not participate in this process. Specifically, under the process for
disposing of surplus federal property, federal agencies must determine if
any property under their control is excess, or no longer needed, within
the agency. If this is the case, they must then report this to GSA, which
first determines if any other federal agency needs the property. If no
other agency needs it, GSA declares the property to be "surplus" and can
dispose of it in a number of ways, such as by selling it or destroying it.
GSA also has authority to donate this property but may not do so directly
to specific private organizations, such as YouthBuild USA. Instead, it
must donate the property through state agencies, which, in turn, donate it
to public agencies or certain nonprofit educational or public health
institutions or organizations.

According to GSA officials, if they were given the statutory authority
explicitly allowing them to donate vehicles from their Fleet program to
YouthBuild USA, they would likely first seek appropriations to the General
Supply Fund to make up for the reductions in revenues associated with
donating the vehicles.^11 In addition, they would consider increasing the
rates they charge federal agencies to lease vehicles, in order to recover
this reduction in revenues. However, GSA currently lacks authority to pass
on the costs of donations to its client agencies. Specifically, the
Federal Property Act specifies how GSA is to set prices to recover the
costs of operating the fleet. The pricing formula (through which GSA sets
its leasing rates) specifies that prices should cover the costs of
operating the fleet and may include an increment for the cost of replacing
fleet vehicles and related equipment.^12 Because the costs associated with
donations of fleet vehicles would not be costs of operating or replacing
the fleet and related equipment, statutory authority would be needed in
order to increase the rates GSA charges its client agencies to recover the
costs of donating cars to YouthBuild USA.

GSA officials indicated there are different ways they might consider
raising leasing rates if given authority to do so but also expressed
concerns about the effect such an increase might ultimately have on the
viability of the Fleet program. For example, GSA could raise its rates
only for those agencies that lease compact sedans; this would result in a
1 percent increase in order to recover the reduction of $600,000 in sales
revenue. To recover the reductions from donating 5 percent of its compact
sedans ($3 million in sales revenue), GSA might raise its leasing rates by
about 5 percent. Another option GSA could pursue would be to raise its
rates across all agencies that lease vehicles from it, regardless of car
type. In this case, all agencies that lease vehicles from GSA Fleet (not
just those leasing compact sedans) would subsidize the vehicle donations.

GSA officials indicated that if the agency raised its rates too high, this
could affect the economic viability of the Fleet program. Nothing requires
federal agencies to lease vehicles from GSA; they can lease from private
companies if they choose to do so. GSA officials told us that the agency
is currently less expensive than the private sector in terms of leasing
rates. GSA officials indicated that, while raising the rate GSA charges to
lease vehicles in order to recover the reduced sales revenue from donating
1 to 5 percent of its compact sedans to YouthBuild USA would probably not
cause the federal agencies to stop leasing from GSA, the officials are
more concerned that donating vehicles would inspire other nonprofit
organizations to seek donations from GSA Fleet. According to GSA
officials, they have received such requests in the past. If GSA were
required to donate a greater percentage of its Fleet vehicles to nonprofit
organizations, then GSA's leasing rates could eventually become higher
than its competitors in the private sector and eventually drive GSA out of
the leasing business.

Another option GSA officials cited does not require a legislative change.
Specifically, GSA officials indicated that it could decrease the amount of
funds it uses to buy new cars each year. Because GSA would have less sales
revenue due to donating some of the cars it normally would sell, GSA would
have less funds to buy new cars. According to GSA officials, to compensate
for the purchase of fewer new cars, GSA would keep its cars in the Fleet
program longer than the current 3- to 5-year time period. Furthermore,
because these cars would be older than GSA's current fleet, they would
require more maintenance, which would result in higher maintenance costs.
Due to the prospective nature of a donation program, GSA officials
indicated that they did not have sufficient information to give a precise
estimate of the increased maintenance costs GSA would incur. Finally, GSA
officials indicated that keeping cars longer might mean more downtime for
some of these cars, which might also result in a reduction of revenues
from leasing.

Available Research Suggests That Car Access Leads to Jobs and Other
Benefits

Taken as a whole, the seven studies we reviewed consistently indicated
that owning a car or having access to one increases the likelihood that
low-income individuals (such as rural Youthbuild participants) get a job
(see app. II for more information on each of the studies we reviewed).^13
According to recent research, this is because a car likely allows a person
to search for a job over a wider geographic area and to work during hours
when public transit is not available. While each of the studies had some
methodological limitations, they all produced fairly consistent results on
the effects of car access on employment and hours worked. Differences
between the populations in these studies and rural Youthbuild participants
did not allow us to use the studies' results to identify the degree to
which participants in YouthBuild USA's LICO program would benefit from
having a car.

Studies Reported Employment Benefits from Car Access

Taken as a whole, the seven studies we examined consistently indicated
that owning a car or having access to one increased the likelihood that
someone would get a job. For example, a 2005 study surveyed about 2,000
urban and rural welfare recipients in Tennessee starting in January
2001.^14 Researchers interviewed the same people every 6 months and asked
questions about car access and employment. The study indicated that
individuals with access to a car increased their chance of finding a job
and leaving welfare by about 59 percent. Another study obtained data from
a survey of a random sample of Alameda County, California, residents who
received welfare benefits in fiscal years 1992 to 1993.^15 Researchers
surveyed these same individuals again in fiscal years 1994 and 1995 and
reported that private automobiles were more effective than increased
public transit in moving participants from welfare to work. Another study
estimated that owning a car increased the likelihood of being employed by
about 20 percent.^16

Three of the seven studies reported that car access led to more work hours
(the other four studies did not address this issue). For example, one
study examined the effects of car access on weekly hours worked and
estimated that access to a car increased the number of hours worked by
nearly 9 hours per week.^17 Two other studies, focusing on car ownership,
estimated increases of 5 to 11 hours per week in the number of hours those
they studied worked.^18

The literature gives several explanations for these results. For example,
a car likely allows a person to search for a job over a wider geographic
area, expanding the number of employment opportunities and the chances he
or she will find a job. Also, a car likely allows a person to work hours
that are not traditionally supported by public transit, increasing not
only the hours he or she can work, but the job openings that may be
considered.^19

Studies Could Not Be Used to Estimate How Much Youthbuild Participants
Would Benefit from Having a Car

Because there are several major differences between the individuals that
the researchers studied and those in the rural Youthbuild population, we
could not extrapolate from this research to identify the degree to which
the rural YouthBuild USA population would benefit from having a car. For
example, the studies generally reviewed welfare recipients who tended to
be older and better educated than the rural Youthbuild population.
Research has shown that older, more educated individuals are more likely
to find employment (with or without a car) than younger and less educated
individuals.

In addition, four of the studies we reviewed were based on data from
mainly urban populations, while two others used national data without
controlling for urban/rural locations. Controlling for differences such as
urban and rural populations would be necessary to extrapolate from these
studies for the Youthbuild population because of various differences
between the two groups. For example, we have reported in the past that
rural residents generally have fewer public transportation options and
live in less densely populated areas than urban residents, making them
more dependent on car access or ownership.^20 Consequently, the effect of
car access in rural areas may be more difficult to discern and different
than in urban areas because rural residents have fewer transportation
options from which to choose. Also, rural unemployment rates are higher on
average than urban areas; as a result, the effect of increasing car access
on new employment may be substantially different than in urban areas.
Consequently, the research that did not distinguish rural and urban
populations did not allow us to gauge whether the rural YouthBuild USA
population would experience greater, equal, or lesser benefits than the
studies reported.

LICO Programs Report Positive Results, but Limitations Restrict Projecting
Results to YouthBuild USA's Program

Six studies of LICO programs, as well as the managers of these programs,
reported that participants were able to get a job or retain their current
jobs, received higher wages, and spent more time with their families as a
result of getting a car through the LICO programs. However, determining
the extent that Youthbuild participants would benefit from such a program
is not possible because of limitations in the methodologies of these
studies, differences between individuals served by these programs compared
with YouthBuild USA, and differences in the designs of the existing
programs and the YouthBuild USA proposal. For example, most of the studies
had low response rates from the participants, which prevented us from
projecting their results to all participants. In addition, in terms of
program design, the LICO programs studied are largely loan programs that
provide additional tools to aid participants, such as financial literacy
training, whereas YouthBuild USA proposes to donate cars and does not
envision additional support services. As a result, managers of the
existing LICO programs believed their participants' outcomes may be
different than YouthBuild USA's because the loan commitment, the financial
literacy training, and the low-risk profile of their participants
altogether increase the likelihood of success.

Studies and Experiences of Six LICO Programs Report Positive Results for
Participants

Six LICO programs that had been subject to an external formal evaluation
study and officials of these programs reported that participants found
jobs, retained their current ones, or increased their income as a result
of obtaining a car through the program. For example, one of the studies
reported that 60 percent of the participants found a job after getting the
car, and another study reported that 97 percent of the participants
surveyed attributed their ability to find a job or retain their current
one to the car they obtained through the program. With respect to
increasing participant income, all six of the studies reported gains in
earnings, and four of the six studies reported that from 72 to 80 percent
of the LICO program participants secured or reported receiving higher
wages. These studies reported that participants believed that their
increased wages were attributable to the vehicle they received from the
LICO program. Another study reported that the participants experienced
annual average income increases from nearly $5,500 to $7,900 after getting
a car. Additionally, officials of the LICO programs told us their
participants reported that having a dependable and affordable car helped
them find or retain their jobs, helped them receive higher wages and work
more hours, or find better jobs.

The LICO program studies and the LICO officials reported that the program
participants benefited from an improved quality of life as a result of
obtaining a car through the LICO programs. According to the studies,
participants reported that having a car enabled them to improve their
education and training, find better quality day care for their children,
and spend more time with their families. One official of a LICO program
added that participants stated that having a car was a life transforming
event, increasing their financial stability and allowing them to obtain
better housing or become homeowners. Other LICO program officials stated
that participants reported that the car had a positive effect on their
self-esteem or gave them hope for the future. LICO program studies and
officials stated that their participants reported that the car they got
through the program allowed them to shop at discount centers rather than
nearby more expensive convenience stores.

Limitations in Studies Do Not Allow for Outcomes to Be Projected to
YouthBuild USA's Proposed LICO Program

Limitations in the methodologies in the six studies, such as low survey
response rates, restrict applying their employment outcomes to the entire
populations of the LICO programs studied or to participants in YouthBuild
USA's proposed LICO program. For example, three studies reported positive
outcomes for participants, but these outcomes were based on low response
rates from LICO program participants.^21 Specifically, these studies
reported the following:

oA study of one LICO program that helps its participants obtain
low-interest loans to purchase a used car  reported that, among the
participants surveyed, 75 percent said that they found a job that paid
higher wages, and 55 percent reported obtaining better quality day care
for their children as a result of obtaining a car through the program.
Officials of this LICO program noted that participants found jobs (and
kept them longer) and improved their quality of life as a result of
obtaining a car from the LICO program. However, while the study of this
program attempted to contact the 90 participants, it only received
responses from 34 participants (38 percent). This raises the possibility
that the 56 nonrespondents, nearly two-thirds of those in the program, may
have had different experiences from the 34 who responded.

oAnother study of a LICO program that relied on data on participants'
earnings and employment reported that 63 percent retained their jobs for
12 months after obtaining a car loan. However, data were not available for
all participants for the time periods of interest to the researchers, and
the study had to rely on samples of the 511 for whom data were available.
For example, there were only data on 131 program participants that went
back a full year before their loans were issued, which raises the
possibility that those for whom data was not available had different
outcomes from those for whom data was available.

oAnother study of a LICO loan program reported that 339 out of 353
participants surveyed experienced an improved quality of life in that they
were able to spend more time with their families. However, these 353
respondents were drawn from a pool of 2,200 that the study authors created
because they anticipated very high nonresponse rates, which raises the
possibility that the participants surveyed experienced very different
outcomes from those that were not surveyed.

Differences in Design and Participants the Programs Serve Also Limit
Application of Study Results to YouthBuild USA's Proposed Program

Officials of the LICO programs studied also told us that differences in
program design and population served to restrict the extent to which their
outcomes could be applied to YouthBuild USA's proposed LICO program. One
of the major differences between the design of the LICO programs studied
and YouthBuild USA's proposal is that most LICO programs we consulted
require participants to obtain a loan to purchase the car, while
YouthBuild USA proposes to give cars to participants for a one-time fee of
$450. Four of these six programs require participants to obtain a
low-interest loan to purchase a car, while YouthBuild USA proposes
donating cars to participants. The other two programs obtain used cars,
repair them (if necessary), and then sell them to the participants at the
cost of repairs or at a very low cost. One of these two programs may also
assist the participants in obtaining a loan,^22 and  the other requires
the participants to make payments for the purchase of the car. In order to
qualify for the loans, the programs require participants to have adequate
income to support car payments and meet creditworthiness standards. Once
they obtain the car, participants must make the loan payments for the term
of the loan. Because of these requirements, as well as participants'
greater reliance on their own resources, officials of these LICO programs
told us that they believe that their participants have greater incentives
toward ensuring that they realize the benefits of car ownership than do
individuals who receive a donated vehicle. Additionally, they believe that
their participants become more invested in getting jobs and achieving
other results that are consistent with the programs' goals than would
participants in a car donation program such as what YouthBuild USA
proposes.

These LICO programs provide additional support to their participants, such
as financial literacy training and car repair assistance, which might also
limit the application of the studies' results because YouthBuild USA's
proposal does not presently envision providing such services. Officials of
these LICO programs stated that, before allowing participants to obtain a
loan, it was important for them to understand how to budget their money
and deal with credit responsibly. They indicated that the financial
literacy training and subsequent counseling was an essential component of
their programs and tied to the success of participants fulfilling their
obligations on their loans, as well as getting or keeping their jobs.

Furthermore, most of the LICO programs implemented various arrangements
for costs of repairs as part of their programs. For example, two programs
developed plans to enable participants to cover the costs of minor repairs
by establishing accounts to which participants make small monthly
deposits, which they later may use for repair costs. Also, one of these
programs required participants to contribute a one-time fee of $250 to a
pool that could later be used to cover the costs of major repairs above
$700. Additionally, an official of one LICO program told us that it
addressed the issue of budgeting for repair costs by securing favorable
rates in advance at specific auto shops and requiring participants to go
to these specific vendors for repairs.

Finally, similar to the studies we reviewed related to car access in
general, the differences between the populations served by these LICO
programs and those YouthBuild USA proposes to serve also might limit the
extent to which the outcomes of these LICO programs could be applied to
YouthBuild USA's proposal. Our reviews of the participants' profiles of
the LICO programs studied found that more than half of the participants
were single women in their mid-20s to 30s, with children, and with some
education beyond high school. In contrast, participants in YouthBuild
USA's LICO program would primarily be men between the ages of 16 and 24
who would have not completed high school upon entering the Youthbuild
program. Officials of the LICO programs believed that the demographics of
a LICO program can influence the programs' outcomes. For example, these
officials indicated that participants who have responsibilities, such as
those with children, are more likely to have a lower risk profile and be
motivated toward getting a job and reaching other goals consistent with
those of the LICO programs.

Agency Comments and Our Evaluation

We provided GSA a draft of this report for its review and comment. In a
letter from the Assistant Commissioner for the Office of Vehicle
Acquisition and Leasing Services (see app. III), GSA described our report
as accurate as it pertains to our description of the GSA Fleet program.
GSA expressed concerns with the potential use of its Fleet for a car
donation program--namely, that the revenue lost would require direct
appropriations from the Congress or cost increases to its customer
agencies in the form of higher lease rates, potentially affecting the
viability of the GSA Fleet program. Additionally, GSA stated a concern
that creating a donation program to YouthBuild USA could lead other
organizations to seek donated vehicles as well.

GSA also stated that the report does not clearly address the question as
to why YouthBuild USA should receive donated GSA vehicles over other
worthy charitable organizations. In our review, we focused on the effect
of donating 1 to 5 percent of selected GSA cars on GSA's fleet vehicle
sales operations and the potential benefits to rural Youthbuild
participants from such donations. We did not, however, evaluate the merits
of providing vehicles to other charitable organizations.

GSA officials also provided a number of technical clarifications to our
report, which we have incorporated, as appropriate.

We are sending copies of this report to the Chairman of the Subcommittee
on Financial Institutions and Consumer Credit of the House Committee on
Financial Services and other interested congressional committees. We also
are sending copies to the Administrator of GSA. We also will make copies
available to others upon request. In addition, the report will be
available at no charge on the GAO Web site at [23]h  ttp://www.gao.gov.

me at (202) 512-8678 or [email protected] Contact points for our
Offices of Congressional Relations and Public Affairs may be found on the
last page of this report. GAO staff who made major contributions to the
report are listed in appendix IV.

Sincerely yours,

William B. Shear
Director, Financial Markets
     and Community Investment

Appendix I
Objectives, Scope, and Methodology

The objectives of this report were to (1) assess the effect of donating 1
to 5 percent of selected General Services Administration (GSA) used cars
on GSA's fleet vehicle sales operations, (2) describe what studies have
shown with respect to the benefits that car ownership or access may hold
for low-income individuals, and (3) describe what studies of selected Low
Income Car Ownership (LICO) programs and experiences of these programs
have shown with respect to the benefits of participant car ownership.

To assess the effect on GSA's fleet vehicle sales operations from donating
vehicles, we obtained data from GSA on the total number of compact sedans
its Fleet program sold and the average sales proceeds for these vehicles
for fiscal years 2001 through 2005 and for 2006 (up to August 2006). We
focused our analysis on GSA compact sedans sales because this is the type
of vehicle YouthBuild USA officials told us that they would primarily need
for their proposed Rural Initiative LICO program. We used GSA's data from
fiscal years 2002 through 2006 to calculate the average opportunity
cost--the reduction in revenue for each compact sedan GSA would
donate--and the total reduction in sales revenue GSA would face by
donating 1 to 5 percent of its compact sedans.^1 To determine how much GSA
might increase the rate it charges federal agencies that lease compact
sedans from it (in order to recover the reduction in sales revenues from
donating 1 to 5 percent of its compact sedans), we obtained data from GSA
on the number of compact sedans it currently leases and revenues it
receives from leasing these vehicles. This data allowed us to calculate
the amount (percentages) GSA would need to increase the rate it charges
federal agencies to make up the reduction in sales revenues from donating
1 to 5 percent of its compact sedans. We obtained written descriptions on
GSA's computer systems and procedures for ensuring that the agency has
verified Fleet transactions. In addition, the agency provided us with a
written description of how GSA Fleet handles any discrepancies found in
the data in its system. Based on this information, we determined that the
data that GSA provided to us were sufficiently reliable for the purposes
of this report. In addition, we obtained information on the options the
agency might pursue to make up for the reduction in revenues from
interviews with GSA officials. To determine the legal issues that would be
involved in a car donation program, we reviewed the legislation that
establishes GSA's authority to operate the Fleet program and for it to set
prices to recover its program costs. We also reviewed the legislation that
established the General Supply Fund and the purposes for which the fund
can be used and discussed these and related legal issues with officials
from GSA's General Counsel.

To describe what studies have shown with respect to the benefits that car
ownership or access may hold for low-income individuals, we conducted a
literature search of relevant studies, reviewed a list of studies from the
Web site of the National Economic Development and Law Center (NEDLC), and
interviewed two individuals who had conducted research on this issue.
Based on these efforts, we identified a large body of research on car
access and employment. We limited our review to seven studies that used
more sophisticated economic models that distinguished between car access
and several other factors that could affect the likelihood that an
individual found a job, such as education, age, and gender.^2 Our review
of these seven studies included an analysis of study methodologies, the
individuals that were studied, study results with respect to car access
and employment, and study limitations. In order to address the issue of
the extent to which the results of these studies could be applied to
YouthBuild USA's proposed LICO program, we compared the individuals
reviewed in these studies with rural Youthbuild participants. For a more
detailed summary of our review of these seven studies, see appendix II.

To determine what studies of selected LICO programs have shown with
respect to the benefits of participants' car ownership, we spoke with
experts in the field of LICO programs about the nature and extent of
existing studies of LICO programs. These experts identified several LICO
programs that had been subject to an independent evaluation of their
outcomes. Additionally, to identify other LICO programs that had been
subject to an independent evaluation and were located in rural areas, we
reviewed about 140 LICO programs listed on the NEDLC Internet Web site^3
and contacted those LICO programs on the NEDLC listing that appeared to be
located in rural and mixed rural-urban areas.^4 From our discussions with
the experts and contacts with the LICO programs, we identified and
selected the following six LICO programs that had been subject to an
independent evaluation:

oGood Wheels, Augusta, Maine

oGood News Garage, Burlington, Vermont

oWest CAP JumpStart, Glenwood City, Wisconsin

oNew Leaf Services, Inc., Decatur, Georgia^5

oWays to Work, Inc., Milwaukee, Wisconsin

oWorking Wheels, Seattle, Washington

Three of these programs were located in rural areas, two in urban areas,
and one in a mixed rural-urban area. Four of the six LICO programs were
loan programs, one was a car-donation and car-sales program, and the other
was a car sales program. We obtained and analyzed copies of the studies
for the six LICO programs in order to identify the reported outcomes and
potential limitations with these studies. Where possible, we reviewed
information from the LICO program Web sites to obtain information on how
these programs operate, what types of individuals they serve, and how many
cars they provide to their participants. We also interviewed LICO program
officials about their studies to gain a better understanding of the
reported outcomes and limitations with the studies. During these
interviews, we also obtained more background information about the LICO
programs' designs, populations served, and challenges the officials faced
in running these LICO programs. We also obtained the views of LICO program
officials on the extent to which they believe their programs' reported
outcomes could be applied to participants in YouthBuild USA's proposed
LICO program.

We conducted our work from May 2006 to November 2006 in San Francisco,
California, and Washington, D.C., in accordance with generally accepted
government auditing standards.

[This page is intentionally left blank]

Appendix II
Summary of Studies on the Effect of Car Access on Employment

^aLogit analysis is a regression technique used to address outcomes where
there are two possible categories, such as employed or not employed. The
multinomial logit is an extension of the logit method to cases where there
are more than two outcomes.

^b"Heckman selection" models are a regression technique used to address
concerns about obtaining biased estimates in cases where an outcome is not
observed unless a given event occurs. For example, hours and wages are not
observed unless someone is employed. Therefore, nonworkers given greater
access to jobs may not have the same outcome as someone who is already
employed.

^aThe two-stage least squares model is a regression technique used to
address concerns about simultaneity bias. For example, the concern that
car access may result from having a job rather than the job resulting from
car access.

^aAlthough "car access" is not specifically mentioned in this study, the
public transportation "barrier" would be removed by "car access."

^bLogit analysis is a regression technique used to address outcomes where
there are two possible categories, such as employed or not employed. The
multinomial logit is an extension of the logit method to cases where there
are more than two outcomes.

^aLogit analysis is a regression technique used to address outcomes where
there are two possible categories, such as employed or not employed. The
multinomial logit is an extension of the logit method to cases where there
are more than two outcomes.

^aLogit analysis is a regression technique used to address outcomes where
there are two possible categories, such as employed or not employed. The
multinomial logit is an extension of the logit method to cases where there
are more than two outcomes.

^aA linear probability model is a regression technique that is used to
estimate the probability that someone is in one of two categories--for
example, employed or not employed. This approach is less desirable for
categorical outcomes than logit analysis because the latter has more
desirable statistical characteristics that increase the precision of
regression estimates.

Appendix III
Comments from the General Services Administration

Appendix IV
GAO Contact and Staff Acknowledgments

William B. Shear, (202) 512-8678, or [email protected]

In addition to the individual named above, Bill MacBlane, Assistant
Director; Harold J. Brumm Jr.; Martin H. De Alteriis; Tami Gurley; Marc M.
Molino; Elizabeth A. Olivarez; Jose R. Pena; Diana Pietrowiak; David M.
Pittman; Paul G. Thompson; and James D. Vitarello made key contributions
to this report.

(250294)

www.gao.gov/cgi-bin/getrpt?GAO-07-153 .

To view the full product, including the scope 
and methodology, click on the link above.

For more information, contact William B. Shear at (202) 512-8678 or
[email protected]

Highlights of [27]GAO-07-153 , a report to the Ranking Minority Member,
Subcommittee on Financial Institutions and Consumer Credit, Committee on
Financial Services, House of Representatives

December 2006

GSA FLEET

Information on the Effect of Donating Cars to YouthBuild USA and Potential
Benefits to Rural Youthbuild Participants

To assist youth who live in high poverty rural areas obtain and retain
jobs, YouthBuild USA, a national nonprofit organization, has proposed
providing donated used cars to selected low-income youth in rural
communities. YouthBuild USA's proposed program hinges on receiving
donations of used cars from the federal government's General Services
Administration (GSA). This report discusses (1) the effect of donating 1
to 5 percent of selected GSA used cars on GSA's fleet vehicle sales
operations, (2) what studies have shown with respect to the benefits that
car ownership or access may hold for low-income individuals, and (3) what
studies of selected low income car ownership programs and experiences of
these programs have shown with respect to the benefits of participant car
ownership.

In conducting this study, GAO examined auction data from GSA, reviewed
academic studies on the benefits of car access in gaining employment, and
interviewed officials of six existing low income car ownership programs.

[28]What GAO Recommends

GAO is making no recommendations in this report.

GAO provided a draft of this report to GSA for its review and comment. GSA
stated that our report is accurate but expressed concerns with the use of
its Fleet vehicles for car donations.

If GSA annually donated 1 to 5 percent of the compact sedans available for
auction from its Fleet program (112 to 559 cars), its annual sales revenue
would be reduced by $600,000 to $3 million (see fig.). To donate cars
directly to YouthBuild USA, GSA would need new statutory authority to
deviate from the existing process for disposing of surplus federal
property. If it were given this authority, GSA would likely first seek
appropriations to recover the loss in sales revenue from the donations but
would also consider increasing its leasing rates for compact sedans.
However, GSA would also require new legislative authority to increase its
rates for this purpose because the current statute governing its Fleet
program does not allow it to pass on these costs to the agencies that
lease vehicles from it.

The seven studies GAO reviewed consistently found that owning a car or
having access to one increases the likelihood that low-income individuals
(such as rural Youthbuild participants) find a job. One reason for this is
that a car allows a person to search for a job over a wider geographic
area. Differences between the populations in these studies and rural
Youthbuild participants did not allow GAO to use this research to identify
the degree to which participants in YouthBuild USA's proposed program
would benefit from having a car.

Six studies of low income car ownership programs and the experiences of
those operating the programs indicated that participants got and retained
jobs, earned higher wages, and spent more time with their families as a
result of owning a car. However, it is difficult to project the results of
these studies to rural Youthbuild participants because of limitations in
the methodologies of the studies, differences between individuals served
by the programs and YouthBuild USA, and differences in the designs of the
existing programs and the YouthBuild USA proposal.

Estimated Average Annual Reduction in Sales Revenue to GSA Fleet from
Donating Compact Sedans Normally Sold at Auction

References

Visible links
27. http://www.gao.gov/cgi-bin/getrpt?GAO-07-153
*** End of document. ***