Trade Adjustment Assistance: Labor Should Take Action to Ensure  
Performance Data Are Complete, Accurate and Accessible		 
(25-APR-06, GAO-06-496).					 
                                                                 
In the current tight budgetary environment, program performance  
is likely to be an increasingly significant factor used to help  
policymakers assess programs and determine funding levels. Given 
concerns over the quality of performance data for the Trade	 
Adjustment Assistance (TAA) program and the importance of having 
meaningful information to assess program performance, we examined
(1) whether the TAA performance data provide a credible picture  
of the program's performance, (2) what TAA performance data the  
Department of Labor (Labor) makes available to the public and	 
states and the usefulness of the data for managing the program,  
and (3) what Labor is doing to address issues with the quality of
TAA data submitted by states.					 
-------------------------Indexing Terms------------------------- 
REPORTNUM:   GAO-06-496 					        
    ACCNO:   A52440						        
  TITLE:     Trade Adjustment Assistance: Labor Should Take Action to 
Ensure Performance Data Are Complete, Accurate and Accessible	 
     DATE:   04/25/2006 
  SUBJECT:   Education or training costs			 
	     Government information dissemination		 
	     Locally administered programs			 
	     Occupational retraining				 
	     Performance management				 
	     Performance measures				 
	     Program evaluation 				 
	     Program management 				 
	     State-administered programs			 
	     Strategic planning 				 
	     Data integrity					 
	     Trade Adjustment Assistance 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-06-496

     

     * Results in Brief
     * Background
          * Services Available under the TAA Program
          * Certification Process and Eligibility Requirements
          * TAA Funding
          * TAA Performance Reporting System
          * Common Measures
          * Flow of Data for Collecting and Reporting TAA Data
     * TAA Data Do Not Provide a Complete, Credible Picture of the
          * TAA Performance Data Are Incomplete and May Be Inaccurate
          * Some States Are Not Using All Available Data Sources to Dete
          * Some States' TAA IT Systems Have Limited Capacity to Help En
          * Some States Told Us Limited Resources Contributed to TAA Per
     * Labor Makes Some TAA Performance Data Available, but Its Use
          * Publicly Available TAA Data Do Not Provide a Clear Picture o
          * Most States Find TAA Performance Data Useful, but Many Would
     * Labor Has Taken Steps to Improve the Quality of Performance
          * Data Validation Is Having Positive Effects but Does Not Addr
          * Labor Has Taken Several Other Steps to Help Improve TAA Data
          * Labor's Common Measures May Address Some Concerns, but State
     * Conclusions
     * Recommendations for Executive Action
     * Agency Comments
          * Web-Based Survey
          * Site Visits
     * GAO Contact
     * Acknowledgments
     * GAO's Mission
     * Obtaining Copies of GAO Reports and Testimony
          * Order by Mail or Phone
     * To Report Fraud, Waste, and Abuse in Federal Programs
     * Congressional Relations
     * Public Affairs

                 United States Government Accountability Office

Report to the Committee on Finance,

GAO

                                  U.S. Senate

April 2006

TRADE ADJUSTMENT ASSISTANCE

Labor Should Take Action to Ensure Performance Data Are Complete, Accurate, and
                                   Accessible

GAO-06-496

TRADE ADJUSTMENT ASSISTANCE

Labor Should Take Action to Ensure Performance Data Are Complete,
Accurate, and Accessible

  What GAO Found

The performance information that Labor makes available on the TAA program
does not provide a complete and credible picture of the program's
performance. Only half the states are including all participants, as
required, in the performance data they submit to Labor-states were more
likely to report participants who received training than those who
received other benefits and services but not training. In addition, many
states are not using all available data sources to determine participants'
employment outcomes. This may result in lower reported outcomes because
states may be inaccurately recording some workers as unemployed who
actually have jobs. To compile their performance data, some states are
using manual processes or automated systems that lack key capabilities to
help minimizes errors, but many states have plans to improve their
systems' capabilities.

Labor reports data on TAA activity levels, services provided to TAA
participants, and key performance measures. The performance data may be
useful for providing a long-term national picture of program outcomes, but
it represents participants who left the program up to 30 months earlier
and, thus, is not useful for gauging the TAA program's current
performance. Also, the performance information is not displayed using
categories that would be informative to policymakers, such as type of
service received and industry of dislocation. Most states find the
performance information they receive from Labor to be at least moderately
useful, but many want more information.

Labor has taken steps to improve the quality of TAA performance data, but
issues remain. In 2003, Labor began requiring states to validate their
data, and most states reported that this increased awareness of data
quality at the state and local level. However, the validation process does
not address the problem of participants being excluded from the
performance data. In fiscal year 2006, Labor instituted a set of common
measures, and many states reported they are experiencing delays in
implementing all required changes. States also expressed interest in
receiving more opportunities to share lessons learned on topics relevant
to TAA data quality.

Number of States Including All or Most Participants in TAA Performance
Data, by Type of Service Received

Types of TAA-funded services

Waivers, but no TAA-funded training

0 5 1015202530354045

Number of states

                 United States Government Accountability Office

Contents

Letter 1
Results in Brief 2
Background 4
TAA Data Do Not Provide a Complete, Credible Picture of the Program's
Performance 12
Labor Makes Some TAA Performance Data Available, but Its Usefulness Is
Limited 23
Labor Has Taken Steps to Improve the Quality of Performance Data, but
Issues Remain 31
Conclusions 39
Recommendations for Executive Action 41
Agency Comments 41

Appendix I Objectives, Scope, and Methodology

Appendix II Program Information in State TAA IT Systems

Appendix III State TAA Administrative Allocations, Fiscal Years 2005-2006

Appendix IV Comments from the Department of Labor

Appendix V GAO Contact and Staff Acknowledgments

GAO Related Products

Tables  
              Table 1: TAA National Performance Goals and Outcomes, Fiscal 
                                 Years 2004 and 2005                       10 
            Table 2: Percentage of Participants Included in States' TAPR   
                          Submissions Who Received Training                15 
                Table 3: How TAA Performance Measures Changed under Common 
           Measures                                                        36 
           Table 4: Site Visit States and Local Areas                      45 
           Table 5: Program Information for Other U.S. Department of Labor 
                           Employment and Training Administration Programs 
                        Contained in State IT Systems Used to Compile TAPR 
           Submissions                                                     46 
               Table 6: State TAA Training and Administrative Allocations, 
                                                                    Fiscal 
           Years 2005-2006                                                 48 
Figures                                                                 
           Figure 1: Participants Who Received Training Are Most Likely to 
                       Be Reported in States' TAPR Submissions             14 
           Figure 2: Many States Are Using WRIS                            18 
                Figure 3: Many States Are Not Using Supplemental Data      19 
             Figure 4: Number of States with TAA IT Systems That Currently 
                     Have or Will Be Adding Certain Capabilities           21 
              Figure 5: Period between When a TAA Participant Exits the    
                     Program and When Data Are Reported to Labor           26 
              Figure 6: Fiscal Year 2004 TAA Performance Data Represent    
                      Exiters from as early as Fiscal Year 2002            27 
            Figure 7: Many States Found Labor's Performance Information to 
                                Be Moderately Useful                       29 
             Figure 8: Most States Reported That Additional Performance    
                        Information from Labor Would Be Useful for Program 
           Management                                                      31 
             Figure 9: States' View of How Labor's Data Validation Efforts 
                                                                      Have 
           Helped Them                                                     32 
               Figure 10: Most State Expressed Interest in Having More     
                     Opportunities to Share Lessons Learned on Data Issues 35 
               Figure 11: Most States Viewed Changes Needed for Common     
                               Measures to Be a Burden                     38 
           Figure 12: Coordinating Exit Dates Was Change States Were Least 
                     Likely to Expect to Have Completed in 2006            39 

Abbreviations

ATAA                Alternative Trade Adjustment Assistance                
ES                  Employment Service                                     
HCTC                Health Coverage Tax Credit                             
IT                  information technology                                 
NAFTA-TAA           North American Free Trade Agreement Transitional       
                       Adjustment Assistance                                  
OIG                 Office of Inspector General                            
OMB                 Office of Management and Budget                        
PART                Program Assessment Rating Tool                         
TAA                 Trade Adjustment Assistance                            
TAPR                Trade Act Participant Report                           
TRA                 Trade Readjustment Allowance                           
UI                  Unemployment Insurance                                 
VETS                Veterans Employment and Training Program               
WIA                 Workforce Investment Act                               
WRIS                Wage Record Interchange System                         

This is a work of the U.S. government and is not subject to copyright
protection in the United States. It may be reproduced and distributed in
its entirety without further permission from GAO. However, because this
work may contain copyrighted images or other material, permission from the
copyright holder may be necessary if you wish to reproduce this material
separately.

United States Government Accountability Office Washington, DC 20548

April 25, 2006

The Honorable Charles Grassley Chairman The Honorable Max Baucus Ranking
Minority Member Committee on Finance United States Senate

More than 150,000 manufacturing workers lost their jobs in fiscal year
2004 due to international trade. Most of these workers were eligible for
services under the Trade Adjustment Assistance (TAA) program, the primary
federal employment and training program serving trade-affected workers.
TAA, funded at about $1 billion in fiscal year 2005, provides workers with
a variety of services, including training and income support after they
exhaust their Unemployment Insurance (UI) benefits. The U.S. Department of
Labor (Labor) oversees the program, and states and local areas administer
services to TAA participants. Despite the role the TAA program plays in
helping trade-affected workers transition to new employment, program
outcomes have shown mixed results. Labor collects some key performance
information from states and uses it to track program performance against
three national goals in the areas of employment, job retention, and wages.
Since 2001, the TAA program has exceeded the national goal for job
retention in every year but one. However, it has failed to meet at least
one of the national goals each year, and in fiscal year 2003, the program
failed to meet all three. Further, the Program Assessment Rating Tool
(PART)-a diagnostic tool that the Administration uses to help formulate
the budget-rated the program ineffective, in part because it failed to
meet its national goals. It is unclear, though, whether the outcomes
reported in the goals accurately reflect the program's achievements. Our
previous work and work by Labor's Inspector General have identified
problems with the performance data states submitted to Labor-information
was often incomplete, and many states did not validate the information
they reported to Labor. While Labor has taken some steps to improve the
performance data, it is not known whether the data currently collected on
TAA performance are reliable.

In the current tight budgetary environment, program performance is likely
to be an increasingly significant factor used to help policymakers assess
programs and determine funding levels. Given concerns over the quality of
TAA data and the importance of having meaningful information to assess

                                Results in Brief

program performance, we examined (1) whether the TAA performance data
provide a credible picture of the program's performance, (2) what TAA
performance data Labor makes available to the public and states and the
usefulness of the data for managing the program, and (3) what Labor is
doing to address issues with the quality of TAA data submitted by states.

To address these questions, we conducted a Web-based survey of workforce
officials in the 46 states that were allocated TAA funds in fiscal year
2005, and we obtained a 100 percent response rate. 1 In addition, to
gather in-depth information about how states manage their TAA performance
data, we conducted site visits in California, Iowa, Ohio, Texas, and
Virginia, where we interviewed state officials and visited at least one
local area in each state. We selected these states because they represent
different TAA data collection approaches (that is, states where data are
entered into information technology [IT] systems at the local level and
those where data are entered at the state level), received a relatively
large share of TAA funds in fiscal year 2005, and are geographically
dispersed. In addition, to gather insights into data management strategies
and requirements, we interviewed Labor officials in headquarters and in
the regional offices and experts on data quality. We also reviewed
legislation, guidance, summaries of state TAA performance outcome
information, participant data submitted by states to Labor, and other
relevant reports and literature related to data quality. Our work was
conducted between December 2004 and March 2006 in accordance with
generally accepted government auditing standards. (See app. I for a
detailed discussion of our scope and methodology.)

The performance information that Labor makes available on the TAA program
does not provide a complete and credible picture of the program's
performance. Only half the states reported that the data they submit to
Labor for determining progress toward national performance goals include
all TAA participants who stop receiving benefits or services-that is, exit
the TAA program-as Labor requires. Other states reported that they are
more likely to include in their performance data participants who received
TAA-funded training rather than those who received other TAA benefits but
did not receive TAA-funded training. Labor does not currently have a
process in place, such as a standard

1

Delaware, Hawaii, North Dakota, and Wyoming were not included in our
survey because they were not allocated TAA funds in fiscal year 2005.

Page 2 GAO-06-496 Trade Adjustment Assistance

monitoring tool, to ensure that states are including all exiting
participants in their data. As a result, the performance data are
incomplete and may be skewed. Further, not all states obtain the
documentation they need to verify that TAA participants' exit dates, used
to develop the performance data, are accurate. Without such documentation,
there is no assurance that the right participants are included and,
therefore, that the employment outcomes are reliable. Some states are not
using all available data sources to determine TAA participants' employment
outcomes. This may result in lower reported outcomes because states may be
inaccurately recording some workers as unemployed who actually have jobs.
To compile TAA data, some states have IT systems with limited capabilities
that may hinder their ability to ensure that the TAA data are complete and
accurate. Nine states are using manual rather than automated processes to
compile their TAA performance data, increasing the data's vulnerability to
data entry errors, and only about half of the states' TAA IT systems can
perform edit checks to help minimize errors. Many states are planning to
make improvements to their TAA IT systems' capabilities this year. State
TAA officials said that resource shortages contribute to their data
problems.

Labor reports data on TAA petition and certification activity, program
participation, and key performance measures, but this information may not
be useful for gauging current program performance. This performance
information may be helpful in providing a long-term national picture of
program outcomes, but it does not represent a current picture of program
performance. UI wage records-the primary data source for tracking TAA
performance-provide a fairly consistent national view of TAA performance
and allow for tracking outcomes over time. At the same time, the UI wage
records suffer from time delays and, together with the use of longer-term
outcome measures, such as employment retention, affect the timing of
states' performance reports to Labor and, subsequently, the information
that Labor makes publicly available. Most of the outcome data reported in
a given program year actually reflect participants who left the program up
to 2 years earlier. In addition, Labor does not consistently report TAA
data by state or by services or benefits received-a step that would make
the data more useful to policymakers. Most states reported that they find
the TAA performance information they receive from Labor to be at least
moderately useful, but many states would like additional information from
Labor, such as how their TAA performance compares to that of other states
and other federal employment and training programs.

While it has limited authority to hold states accountable, Labor has taken
steps to improve the quality of TAA data submitted by states, but these
steps do not fully address all issues. Labor has no mechanism to sanction

                                   Background

states for poor performance or poor quality data because the law and
current regulations do not provide one. However, beginning in 2003, Labor
implemented a new data validation process for its TAA performance data
that requires states to review a sample of participants' records and
compare the data recorded for certain data elements to source files. While
it is too soon to fully assess whether Labor's efforts have improved data
quality, most states reported that Labor's new requirements have improved
data accuracy and increased awareness of data quality at the state and
local level. Labor has taken other steps to improve data quality such as
providing additional guidance, training, and technical assistance. Yet
most states reported that they do not currently have opportunities to
share lessons learned with other states on various topics related to TAA
data quality and expressed interest in having such opportunities. In
fiscal year 2006, Labor is requiring changes in some TAA data-reporting
requirements in order to implement a set of common measures for federally
funded job training programs that share similar goals. Many states
reported that the changes are burdensome, and some states are experiencing
delays in implementing the changes.

To help ensure that TAA participant data reported by states is consistent,
complete, and accurate, we recommend that Labor clarify reporting and
documentation requirements, provide guidance to regional offices for
assessing states' data collection processes, and create opportunities for
states to share lessons learned on data quality issues. Finally, to make
TAA performance information more useful for program management, we
recommend that Labor provide performance information that is more
detailed. In its written comments on a draft of this report, Labor did not
disagree with our findings and recommendations and said the report will be
helpful in its continuing efforts to improve the quality of TAA
performance data.

To assist workers who are laid off as a result of international trade,
Congress passed the Trade Expansion Act of 1962 and created the Trade
Adjustment Assistance program. Historically, the main benefits available
through the program have been extended income support and training.
Participants are generally entitled to income support, but the amount of
funds available for training is limited by federal statute. Labor
certifies groups of laid-off workers as potentially eligible for TAA
benefits and

Services Available under the TAA Program

services by investigating petitions that are filed on the workers' behalf.
2 Workers are eligible for TAA if they were laid off as a result of
international trade and were involved in making a product or supplying
component parts to or performing finishing work for directly affected
firms. Workers served by the TAA program have generally been laid off from
the manufacturing sector.

Congress has amended the TAA program a number of times since its
inception. For example, in 1974 Congress eased program eligibility
requirements, and in 1988 Congress added a requirement that workers be in
training to receive income support. In 1993 Congress created a separate
North American Free Trade Agreement Transitional Adjustment Assistance
(NAFTA-TAA) program specifically for workers laid off because of trade
with Canada or Mexico. 3 The most recent amendments to the TAA program
were included in the TAA Reform Act of 2002 (Pub. L. No. 107-210), which
was signed into law in August 2002. The Reform Act consolidated the former
TAA and NAFTA-TAA programs into a single TAA program, doubled the amount
of funds available for training annually, expanded program eligibility to
more workers, extended the time periods covered by the program, and added
new benefits.

Under the current TAA program, eligible participants have access to a
wider range of benefits and services than before, including:

Training. Participants may receive up to 130 weeks of training, including
104 weeks of vocational training and 26 weeks of remedial training (e.g.,
English as a second language or literacy). On-the-job training is also
available under TAA. Participants in TAA-approved training must attend
training full-time.

2

Not all workers covered by an approved TAA petition are individually
eligible for TAA benefits and services. Individual eligibility also
depends on factors including the timing and duration of a worker's layoff.
In this report, when referring to workers eligible for the TAA program, we
generally mean workers who have been certified as potentially eligible for
the program.

3

For more information on the earlier TAA programs see GAO, Trade Adjustment
Assistance: Trends, Outcomes, and Management Issues in Dislocated Worker
Programs, GAO-01-59 ( Washington, D.C.: Oct. 13, 2000); GAO, Trade
Adjustment Assistance: Experiences of Six Trade-Impacted Communities ,
GAO-01-838 ( Washington, D.C.: Aug. 24, 2001).

Extended income support, or Trade Readjustment Allowances (TRA).
Participants may receive up to 104 weeks of extended income support
benefits after they exhaust the 26 weeks of UI benefits available in most
states. This total includes 78 weeks while participants are completing
vocational training and an additional 26 weeks, if necessary, while
participants are completing remedial training. The amount of extended
income support payments in a state is set by statute at the state's UI
benefit level. 4

During their first 26 weeks of extended income support, participants must
be enrolled in training, have completed training, or have a waiver from
this requirement; to qualify for more than 26 weeks of extended income
support, participants must be enrolled in training. The TAA statute lists
six reasons why a TAA participant may receive a waiver from the training
requirement, including that the worker possesses marketable skills or that
the approved training program is not immediately available. 5 States must
review participants' waivers at least every 30 days and if necessary may
continue to renew participants' waivers each month throughout the initial
26 weeks of extended income support.

Job search and relocation benefits. Payments are available to help
participants search for a job in a different geographical area and to
relocate to a different area to take a job. Participants may receive up to
a maximum of $1,250 to conduct a job search. The maximum relocation
benefit includes 90 percent of the participant's relocation expenses plus
a lump sum payment of up to $1,250.

Health Coverage Tax Credit (HCTC). Eligible participants may receive a tax
credit covering 65 percent of their health insurance premiums for certain
health insurance plans. To be eligible for the credit, trade-affected
workers must be either receiving extended income support payments, or they
must be eligible for extended income support but are still receiving UI
payments, or they must be recipients of benefits under the wage insurance
program. As a result, trade-affected workers who are still

4

Extended income support payments may be reduced based on other income and
training allowances.

5

The four other reasons listed in the TAA statute are (1) worker will be
recalled by former employer, (2) worker is within 2 years of retirement,
(3) worker is unable to participate in training because of health
problems, and (4) approved training is either not available or not
available at a reasonable cost, or no training funds are available.

Page 6 GAO-06-496 Trade Adjustment Assistance

               Certification Process and Eligibility Requirements

receiving UI rather than extended income support may register for the HCTC
only if they are in training, have completed training, or have a waiver
from the training requirement.

Wage insurance. The wage insurance program-known as the Alternative TAA
(ATAA) program-is a demonstration project designed for workers age 50 and
older who forgo training, obtain reemployment within 26 weeks, but take a
pay cut. Provided the participant's annual earnings at his or her new job
are $50,000 or less, the benefit reimburses 50 percent of the difference
between the participant's pre- and postlayoff earnings up to a maximum of
$10,000 over 2 years.

The process of enrolling trade-affected workers in the TAA program begins
when a petition for TAA assistance is filed with Labor on behalf of a
group of laid-off workers. Petitions may be filed by entities including
the employer experiencing the layoff, a group of at least three affected
workers, a union, or the state or local workforce agency. The TAA statute
lays out certain basic requirements that all certified petitions must
meet, including that a significant proportion of workers employed by a
company be laid off or threatened with layoff. In addition to meeting
these basic requirements, a petition must demonstrate that the layoff is
related to international trade. The law requires Labor to complete its
investigation, and either certify or deny the petition, within 40 days
after it has received it. When Labor has certified a petition, it notifies
the relevant state, which has responsibility for contacting the workers
covered by the petition, informing them of the benefits available to them,
and telling them when and where to apply for benefits.

Workers generally receive services through a consolidated service delivery
structure called the one-stop system, where they can access a broad range
of services beyond TAA, including the Workforce Investment Act (WIA)
Dislocated Worker program, the Wagner-Peyser Employment Service (ES)
program, and services funded by the WIA National Emergency Grants. 6
Training for trade-affected workers may be funded by TAA or by one of the
WIA funding sources. Workers often meet one on one with a case manager who
may assess worker's skills and help decide what services

6

State officials generally have responsibility for approving training and
determining eligibility for extended income support, while local one-stop
centers are the main point of intake and actual delivery of TAA services
and benefits.

Page 7 GAO-06-496 Trade Adjustment Assistance

                                  TAA Funding

they need. Because the TAA program has limited funds that can be used for
case management and program administration, these case management services
are often performed by ES or WIA Dislocated Worker program staff. When
this occurs, participants are often co-enrolled in WIA or ES as well as
TAA.

About $750 million was appropriated for income support for trade-affected
workers for fiscal year 2005, while another $259 million was appropriated
for training, job search and relocation allowances, and administrative
costs. 7 Of the $259 million, $220 million is set aside for training, and
Labor allocates 75 percent of it to states according to a formula that
takes into account each state's previous year allocations, accrued
expenditures, and participant levels. Labor holds the remaining 25 percent
of training funds in reserve, to distribute to states throughout the year
according to need. To cover administrative costs associated with training
under the TAA program, Labor allocates additional administrative funds to
each state equal to 15 percent of its training allocation. 8

                        TAA Performance Reporting System

Labor is responsible for monitoring the performance of the TAA program. In
fiscal year 1999, Labor introduced a new participant outcomes reporting
system, the Trade Act Participant Report (TAPR), that was designed to
collect national information on TAA program participants, services, and
outcomes. States are required to submit TAPR reports to Labor each
quarter, with data on individuals who exited the TAA program. The TAPR
data submitted by states are used to calculate national and state outcomes
on the TAA performance measures for each fiscal year, which include

(1) the percentage of participants that found jobs after exiting the
program (reemployment rate), (2) the percentage of those participants who
were employed after exiting the program who were still employed 9 months
later (retention rate), and (3) the earnings in their new jobs compared to
prior earnings (wage replacement rate).

7

Unlike other federally funded employment and training programs that
operate on a program year basis, the TAA program operates on a federal
fiscal year basis, that is, fiscal year 2004 ran from October 1, 2003, to
September 30, 2004.

8

To cover the cost of administration, states also receive an additional 15
percent of any additional funding they receive from the reserved training
funds or the job search/relocation allowances.

Page 8 GAO-06-496 Trade Adjustment Assistance

Labor's guidance requires states to include in their TAPR submissions all
TAA participants who exit the program, that is, stop receiving benefits or
services. Under Labor's guidance, a participant is defined as any
individual who receives any TAA benefit or service, including extended
income support payments, training, or job search and relocation
allowances. According to this definition, participants would include those
who, for example, received only extended income support and a waiver that
allowed them to forgo training.

TAPR reports include data on each exiter's characteristics, services
received, and employment outcomes. Data on characteristics, for example,
should include the worker's date of birth, gender, ethnicity, educational
level, and layoff date. Data on services received should include data on
training (such as dates the participant entered and completed training,
and the type of training received), on other TAA benefits received (such
as extended income support, job search allowance, and relocation
allowance), and on co-enrollment in WIA or other federal programs. Data on
outcomes should include the date the worker exited the TAA or other
federal program, whether the worker was employed in the first full quarter
after exit, whether the worker was employed in the third full quarter
after exit, and the worker's earnings in these quarters. Where possible,
outcome data are to be obtained from state UI wage records.

Labor uses the TAPR data to track TAA program outcomes against national
goals. Unlike the WIA programs, however, TAA has no individual state
performance goals, and states do not receive incentives or sanctions based
on their performance levels, nor are they otherwise held accountable for
their performance. At the national level, the TAA program has failed to
meet at least one of its performance goals each year since 2001, the first
year for which goals were set. Table 1 shows goals and outcomes for fiscal
years 2004 and 2005.

                                Common Measures

Table 1: TAA National Performance Goals and Outcomes, Fiscal Years 2004
and 2005

Fiscal year 2004 Fiscal year 2005 Measure Goal Outcome Goal Outcome

                        Wage replacement 90% 73% 80% 76%

                       Reemployment rate 70% 62% 70% 70%

                         Retention rate 88% 86% 89% 91%

Source: Department of Labor.

Note: Bold denotes TAA program goal met or exceeded.

In addition to submitting TAPR data, states also submit data to Labor on
TAA services and expenditures each quarter through the Form 563. Form 563
includes counts of participants receiving TAA services, while TAPR
includes individual-level data on former participants who have exited the
program. States are required to submit each quarter's Form 563 data about
1 month after the end of the quarter. Form 563 includes data on services
such as the number of new training participants (by type of training-
occupational, remedial, and on-the-job), the number of workers in training
at the end of the quarter, the number of training waivers issued, and the
number of recipients of job search and relocation allowances, and
expenditures on extended income support.

In response to an Office of Management and Budget (OMB) initiative, Labor
recently began requiring states to implement common performance measures
for WIA programs. 9 OMB established a set of common measures to be applied
to most federally funded job training programs that share similar goals.
Labor further defined the common measures for all of its Employment and
Training Administration programs and required states to implement these
measures beginning July 1, 2005. Because it operates on a fiscal year
rather than a program year basis, Labor required the TAA program to
implement the measures by October 1, 2005.

In addition to standardizing the performance measures, the common measures
guidance also standardizes the definition of exiters across all programs.
An exiter is defined as any participant who has not received a

9

See GAO, Workforce Investment Act: Labor Should Consider Alternative
Approaches to Implement New Performance and Reporting Requirements,
GAO-05-539 (Washington, D.C.: May 27, 2005).

Page 10 GAO-06-496 Trade Adjustment Assistance

Flow of Data for Collecting and Reporting TAA Data

service funded by the program or funded by a partner program for 90
consecutive calendar days and is not scheduled for future services. The
exit date is defined as the last date of service. For TAA participants,
the exit date may be the training completion date, but if additional
services are provided after training is completed, or if the participant
is continuing to receive TRA, he or she would not be exited from the
program. Some services are not significant enough to delay exiting,
however. These include receiving UI benefits, some case management
services, and postplacement follow-up. 10

The process of collecting and reporting TAA performance data involves all
three levels of government. Participant forms and case files are generally
collected and organized by frontline staff in local areas, usually at the
onestop. In some states, local staff may enter some of the information
into an IT system that is either integrated with the state's IT system or
able to create an electronic file to transmit to the state. In other
states, paper case files are physically transferred to state officials for
data entry.

At the state level, TAA data are often maintained in more than one IT
system. For example, benefit payment information is usually in the same IT
system that houses Unemployment Insurance payment information. However,
information on participant characteristics and services (including status
of training and whether or not the individual has exited) resides in one
or more other systems. In some states, this participant information
remains as a paper case file until it is determined that the participant
has exited, and it is time to include him or her in the TAPR submission.

To compile the TAPR submission, state agencies administering TAA typically
match participant records to their state's UI wage record system to
determine whether these former participants are employed and, if so, the
wages they are earning. In some states, staff must manually enter
information obtained from the UI wage record system into the TAPR file,
while other states have IT systems capable of automatically matching UI
data with participants' records. States may also use the Wage Record
Interchange System (WRIS) to match participant records to other states'

10

Labor distinguishes between UI benefits and TRA benefits in determining
exit because TRA benefits are tied to continuous participation in skills
training or having a waiver of that requirement.

Page 11 GAO-06-496 Trade Adjustment Assistance

  TAA Data Do Not Provide a Complete, Credible Picture of the Program's
  Performance

UI wage records for participants who found jobs in other states. Some
states may link participant records to other partner programs' IT systems
to track activities across programs or to determine if the participant has
exited all programs. Once Labor receives the TAPR data, officials perform
edit checks and calculate performance levels at the national and state
level.

TAA performance data are incomplete and may be inaccurate. States report
that they are not including all TAA participants in their TAPR performance
data, despite Labor's requirement that all participants be included after
they exit the program. In addition, some states may not have documentation
to verify the accuracy of participants' exit dates in TAPR and are not
using all available data sources to determine TAA participants' employment
outcomes. Furthermore, 1 state in 5 is using manual rather than automated
processes to compile TAPR data, and others have IT systems with limited
capacity to control for errors. Having such IT systems could hinder
states' ability to ensure that the data are complete and accurate.
However, many states are planning to make improvements to their TAA IT
systems' capabilities this year. Some state TAA officials said that
resource constraints have made it difficult to ensure their data are
complete and accurate.

TAA Performance Data Are Incomplete and May Be Inaccurate

Many states are not including all exiting participants in the TAPR
submissions that Labor uses to calculate performance outcomes for TAA
participants, such as the reemployment and retention rates. Participants
who received training were most likely to be included in states' TAPR
data, but those who had training waivers and had not received training
were least likely to be included. Only 23 of the 46 states we surveyed
reported that they are including in their TAPR submissions to Labor all
exiting participants, regardless of the type of benefit or service they
received. Fourteen states reported that participants who received waivers
but did not receive training were unlikely to be included in the TAPR (see
fig. 1), and 3 states reported that they do not include any participants
unless they receive training. This finding is consistent with a review by
Pennsylvania's state auditor that found that participants who received
waivers from training were not included in their TAPR submissions. 11

11

Commonwealth of Pennsylvania's Office of the Budget and Comptroller
Operations, Single Audit Report for Fiscal Year Ended June 30, 2003
(Harrisburg, Pennsylvania: Aug. 31, 2005).

Page 13 GAO-06-496 Trade Adjustment Assistance

Figure 1: Participants Who Received Training Are Most Likely to Be
Reported in States' TAPR Submissions

Number of states

                                   gd trainin

                                       ut

                                gfunded trainin

                                     ut no

b

,

                                     h and

gAA-funded trainin

                                   , bswance

c

                                     saiver

earsJob

                                     -funde

allo

                                       W

                                      AA-A

ion

A

no T

                                       T

T

relocat

    Types of TAA-funded services

Report few or no participants

Report some participants

Report all or most participantsSource: GAO survey of state TAA officials.

Note: States that did not respond or that responded "uncertain" to the
specific survey questions analyzed for the figure were not included in the
analysis.

Our review of the TAPR data states submitted to Labor during fiscal year
2005 confirms our survey results-some states appear to be excluding some
of their participants in their TAPR data files. For example, 9 states only
included in their TAPR submissions participants who received training.
Another 12 states had TAPR submissions composed almost exclusively (97 to
99 percent) of participants who received training (see table 2). However,
several states did include relatively more of the participants who had not
received training. For example, for 6 states, under 60 percent of the
participants reported in the TAPR had received training. We have no other
reliable source of data to help us assess what proportion of participants
nationwide actually receive training and, therefore, what the proportion
in the TAPR should be. In a recent study that examined services and
outcomes for five trade-related layoffs, however, we found that between 9
and 39 percent of potentially eligible TAA participants enrolled in
training. 12 Excluding certain participants from the TAPR could skew the
TAA performance outcomes calculated by Labor because the outcomes may be
disproportionately based on participants who received TAA-funded training.

Table 2: Percentage of Participants Included in States' TAPR Submissions
Who Received Training

                  Percentage of participants included in TAPR

Received other benefits Number of states or services but no training
Received training

                                   9 None 100

0.1-3 97-99.9

4-20 80-96

21-40 60-79

Over 40 Under 60

Source: GAO analysis of TAPR data submitted to Labor during fiscal year
2005.

Labor does not have a process in place to ensure that states are including
in their TAPR submissions all exiting TAA participants. Labor's regional
offices may review whether states' TAPR submissions are complete during
their state monitoring visits. However, because Labor has not had a
standard monitoring tool, there has been no assurance that the regional
offices were consistently reviewing whether all exiting participants are
reported in states' TAPR data. Labor officials tell us that they are
currently developing a core monitoring guide, but it is not clear if the
guide will address this issue.

Despite the importance of accurately identifying exiters, the exit dates
themselves may not be accurate because some states do not consistently
obtain proper documentation to verify the dates. Accurate exit dates are
critical to TAA performance data for two reasons. First, a participant's
exit

12GAO, Trade Adjustment Assistance: Most Workers in Five Layoffs Received
Services, but Better Outreach Needed on New Benefits, GAO-06-43 (
Washington, D.C.: Jan. 31, 2006). For this study, potentially eligible
participants are defined as all workers covered under the certification,
whether or not they enrolled in the TAA program.

Page 15 GAO-06-496 Trade Adjustment Assistance

date determines if the individual should be included in the state's TAPR
submission to Labor. Second, the timing of the date of exit determines
when a participant's employment outcomes will be assessed. Labor's
guidance requires that states have documentation for participants' exit
dates but does not specify the type of information that needs to be
included in the documentation. For example, for participants who received
training, it does not specify that the documentation should demonstrate
that training was actually completed. Such documentation could include
certificates of training completion, attendance records, or reports from
training providers.

TAA officials in 4 of the 5 states we visited said they had a process for
obtaining documentation to show that participants completed training, but
it is not clear whether such processes are uniformly followed by states.
Officials in 3 states said that they receive training certifications,
either from participants or from trainers, that show that training was
completed. In another state, a TAA official said that the state sends
participants a follow-up survey after training to verify that the training
was completed, but some participants do not return the survey. Officials
in 1 of the 5 states we visited said they did not have a process for
certifying or documenting that participants completed training.

A recent review in 4 other states by Labor's Office of Inspector General
(OIG) confirmed that states do not have effective processes for verifying
exit dates. 13 In its review of 150 TAA case files, the OIG found that
there was no documentation in any of the reviewed files to verify that the
participants had completed the program on the recorded date of exit. OIG
reported that states often recorded an anticipated date of exit when
participants first entered the program, but did not collect any further
documentation to confirm that participants had completed the training, and
if so, whether they had completed training on the originally recorded
date. The OIG recommended that Labor ensure that states collect and record
TAA participants' actual date of exit, maintain the source documentation
for such exit dates, and make the documentation readily available for
review. According to an OIG official, Labor had not implemented these
recommendations as of January 2006.

Office of Inspector General, U.S. Department of Labor, GPRA Data
Validation Review: Trade Adjustment Assistance Program, (Department of
Labor Office of Audit Report Number 22-05-007-03-330, Sept. 15, 2005).

Page 16 GAO-06-496 Trade Adjustment Assistance

Some States Are Not Using All Available Data Sources to Determine
Employment Outcomes for TAA Participants

Some states are not using all available data sources to determine TAA
participants' employment outcomes. Labor requires states to use UI wage
records to determine the employment outcomes of participants reported in
the TAPR. However, each state's wage record database includes only wage
data on workers within the state and does not have data on participants
who found employment in another state.

To help track employment outcomes of TAA participants across state lines,
states can obtain their employment and earnings information using other
methods. Labor encourages states to use WRIS, a data clearinghouse that
makes UI wage records available to participating states seeking
information on TAA participants who may have found employment outside
their state. 14 Thirty-four of the 46 states we surveyed reported that
they routinely use WRIS to obtain employment outcome data on former TAA
participants (see fig. 2). Three states reported that they do not use WRIS
but instead routinely use interstate agreements with individual states to
obtain employment outcome data. Opting to use interstate agreements with
individual states instead of using WRIS is likely to result in access to
fewer states' UI wage records than states would have if they used WRIS and
may result in lower reported outcomes. Seven states use only their own
states' UI wage records to determine participants' employment outcomes.
State TAA officials cited several reasons for not using WRIS, including
that it took too long to receive the needed information and it was not a
priority for the state. Six states that do not currently use WRIS said
that they plan to begin using this system in the future.

14

According to an official involved in the operation of WRIS, all states
except Hawaii currently provide UI wage data to WRIS.

Page 17 GAO-06-496 Trade Adjustment Assistance

Figure 2: Many States Are Using WRIS 
Number of states50 45 40

                             sreementgtate asInter

e

                                state'UI recorsn

)

                                       s

                                      ain

                                   St not WRI

                                 ganStem (WRIs

h

c

                                       d

                                     Uncert

d Inter

                                       w

                                       o

                                       u

                                       y

                                       b

y

                                      Onl

e RecorgWaS

      Type of method

Source: GAO survey of state TAA officials.

Nearly half of the 46 states are not routinely using other supplemental
information sources even though it may be the only way to collect outcome
information for certain participants. UI wage records, which cover about
94 percent of workers, do not include some categories of workers, such as
self-employed persons, most independent contractors, military personnel,
federal government workers, and postal workers. To document the employment
status of these workers in the TAPR, states can use supplemental data,
such as pay stubs and follow-up surveys sent to participants after they
leave the program. Using supplemental data is likely to provide a more
complete picture of participant outcomes because it helps states avoid
inaccurately recording participants as unemployed in the TAPR. In an
earlier report on WIA performance data, 23 of the 50

Some States' TAA IT Systems Have Limited Capacity to Help Ensure Data
Quality

states told us they needed to use supplemental data in order to meet their
expected performance levels for the reemployment measure under WIA. 15

Twenty-two states reported that they rarely if ever collect supplemental
data to obtain outcome information on TAA participants (see fig. 3). State
TAA officials said that they did not collect supplemental data because

     o states' TAA IT systems lacked the capacity to record supplemental
       data;
     o they judged data collected through UI wage records and WRIS as
       sufficient, or collecting supplemental data was not required; and
     o they lacked sufficient resources.

Figure3: Many States Are Not Using Supplemental Data Frequency of use
Rarely or never

22

                                       10

    Sometimes

14

    Always or often 0 5 101520 Number of states

Source: GAO survey of state TAA officials.

Some states reported IT system limitations that could hinder the states'
ability to ensure their TAA data are complete and accurate.

Use of manual processes for compiling data. Nine states reported that they
compile their TAPR report manually by entering data from various sources
into a spreadsheet or other database. For example, 1 state reported that
its TAPR submission was compiled every quarter by culling data from the
actual case files housed in a state official's office. Of the 9 states
that reported compiling their TAPR manually, 4 keypunch their UI

15

GAO, Workforce Investment Act: States and Local Areas Have Developed
Strategies to Assess Performance, but Labor Could Do More to Help,
GAO-04-657 (Washington, D.C.: June 1, 2004).

Page 19 GAO-06-496 Trade Adjustment Assistance

data on employment outcomes into their TAPR data rather than
electronically transferring the data from the UI wage record file. Using
manual rather than automated processes increases the opportunity for
errors to be introduced into the data through data entry. Six states
responding to our survey expressed concern that errors in data entry may
be one of the main causes of incomplete or inaccurate TAA data.

One state's process for manually compiling the TAPR illustrates
opportunities to introduce errors into the data. In this state, staff at
the state level enter data on TAA participants' training contracts into a
contract database. To compile the TAPR, they identify participants in the
contract database whose training was scheduled to be completed during the
quarter covered by the TAPR, and they enter data on those participants
into a new spreadsheet. To identify employment outcomes for the TAPR, the
staff look up the exiting TAA participants on printouts from the state's
UI wage record system and manually enter data on the participants'
employment status and wages into the spreadsheet. The data from the
spreadsheet are then converted into the TAPR reporting format and sent to
Labor.

Limited IT system capabilities. Many states' IT systems for compiling TAA
data do not have certain IT system capabilities, such as performing edit
checks, that help a state report complete and accurate data to Labor. Only
15 of the 46 states we surveyed had TAA IT systems with each of three such
capabilities:

     o Performing edit checks to prevent data errors: Edit checks aid in
       identifying invalid data, such as an entry in a date field that is not
       a date.
     o Identifying dates TAA participants completed WIA-funded services: The
       ability to identify when TAA participants complete WIA-funded services
       can help ensure that TAA participants are not counted in the TAPR
       report while they are still receiving services under WIA. If
       participants are still receiving services, then it is too soon to
       assess their employment outcomes in the TAPR report.
     o Allowing staff to query the system to assess data reliability and
       completeness: Queries allow staff to pull certain information out of
       the system to answer questions, and without this capability, staff may
       not be able to properly assess data quality and diagnose data
       problems. For example, in one local area we visited, a TAA specialist
       who is responsible for reporting on numerous TAA participants
       described

having great difficulty determining if training completion dates had been
entered for participants as appropriate because the specialist could not
query the system to get a list of participants and their training status.

More than half of the states told us they had plans to make at least one
of these system improvements during the next year. For example, 17 states
reported plans to improve their TAA IT systems' capability to perform edit
checks (see fig. 4).

Figure 4: Number of States with TAA IT Systems That Currently Have or Will
Be Adding Certain Capabilities Number of states50 45 40 35 30

                                       25

25 24 2320

17

17

16

15

10

5

0 Perform edit Identify WIA Allow staff checkscompletion datesqueries Type
of capability

Have capability Adding capability

Source: GAO survey of state TAA officials.

Some states with electronic systems may have the capability to track TAA
participants across other programs serving them, but several do not. Out
of the 37 states that told us they had electronic systems to compile their
TAPR data, 29 states said this same system captures program information
for WIA programs, and similarly, 29 said the system captures information
for Employment Services. Thirteen states said that these capabilities
extended to all six programs and benefits that we examined, which

Some States Told Us Limited Resources Contributed to TAA Performance Data
Problems

includes, in addition to WIA and ES, Trade Readjustment Allowance,
National Emergency Grants, Veterans Employment and Training Program, and
UI. See appendix II for a complete listing of states' systems linkages.

In addition, several states commented on our survey that they have planned
enhancements to their TAA IT systems that may help coordinate across
programs and increase the likelihood of capturing more outcomes:

     o Improving coordination of data across programs: Six states reported
       planning changes, such as developing a single case management system
       for several programs that would allow more coordination of data across
       programs.
     o Transitioning from manual to electronic processes: Two states that
       have been using manual processes reported plans to develop electronic
       interfaces to capture needed data for the TAPR.
     o Adding capacity to record supplemental data: Two states reported that
       their IT system changes will enable them to begin recording
       supplemental data for use in determining TAA participants' employment
       outcomes.

Some states reported that limited TAA administrative funds hindered their
ability to ensure the quality of the TAA performance data they collect and
maintain. To cover their TAA program's administrative costs, states
receive an allocation each year equal to 15 percent of their TAA training
allocation. In fiscal year 2006, 9 states received less than $100,000 in
TAA administrative funds, and another 10 states received between $100,000
and $300,000. These funds are used to cover all the administrative
activities of the program, such as reviewing waivers and training plans,
processing applications for job search or relocation allowances, and any
associated data collection and reporting. Some states also use these funds
for direct case management services to participants because they are the
only TAA funds available to provide these services. However, we recently
reported that state officials told us the TAA administrative funds were
often insufficient to meet the case management needs of the program and
they relied on other programs to provide those services. 16 (For a
complete listing of each state's TAA training and administrative funds,
see app. III.)

16 See GAO-06-43.

  Labor Makes Some TAA Performance Data Available, but Its Usefulness Is Limited

State and local TAA officials said that resource shortages contribute to
difficulties in identifying exit dates, using supplemental data sources,
and entering data in a timely manner. For example, one state official
commented on our survey that TAA case managers often do not have enough
time to follow up with participants to learn about their status after they
have been sent to training. Another state official said that insufficient
case management can delay the identification of participants exiting the
program. Officials in 2 other states told us that supplemental data were
too time-consuming and burdensome to collect, given the program's current
funding levels. Officials said that resource limitations also presented
challenges in entering the data in a timely manner. An official in one
local area we visited reported a tremendous backlog in entering TAA
participant data into the IT system because there were just two staff to
handle approximately 1,000 TAA cases. Similarly, in another local area,
the office manager told us that TAA staff were spread too thinly, a
condition that adversely affected the collection and entry of TAA data.

Labor reports data on TAA petition and certification activity, program
participation, and key performance measures, but this information may not
be useful for gauging current program performance. The information may be
helpful in providing a long-term national picture of program outcomes, but
it represents past, rather than current, performance. UI wage records-the
primary data source for tracking TAA performance- provide a fairly
consistent national view of TAA performance and allow for tracking
outcomes over time. At the same time, the UI wage records suffer from time
delays and, together with the use of longer-term outcome measures, affect
the timing of states' performance reports to Labor and, subsequently, the
information that Labor makes publicly available. Most of the outcome data
reported in a given program year actually reflect participants who left
the program up to 2 years earlier. In addition, Labor does not
consistently report TAA data by state or industry or by services or
benefits received-a step that would make the data more useful to
policymakers. States responding to our survey reported that they would
like additional information from Labor, such as how their TAA performance
compares to the performance of other states and other federal employment
and training programs.

Publicly Available TAA Data Do Not Provide a Clear Picture of Current
Program Performance

Labor makes some TAA statistics available through postings on its Web site
and through published reports, but they do not provide useful information
on current performance. 17 Labor provides some TAA activity and
participant data by fiscal year including

     o number of petitions received, certifications issued, and denials by
       state;
     o distribution of certifications by industry;
     o number of new participants receiving extended income support payments
       or training; and
     o summary statistics on former TAA participants (such as race, education
       level, and benefits and services received).

In addition to reporting on TAA activity and participant data, Labor also
reports on three key TAA performance measures. The TAPR data submitted by
states are used to calculate national and state outcomes on the TAA
performance measures-wage replacement, reemployment, and retention-for
each fiscal year. In 2005, Labor made state-by-state TAA outcome
information publicly available for the first time. According to Labor
officials, making this information public represents an effort to
emphasize performance, and they intend to post state-by-state outcome
information on the Web site for all future fiscal years. Labor's regional
offices directly provide states with information on their TAA performance
relative to the program's national goals. Some regional offices also
provide states with reports showing the performance of all states in the
region, according to officials we interviewed.

However, the information Labor makes publicly available may not provide a
clear picture of current TAA performance because, in addition to being
incomplete and perhaps inaccurate, the data represent past performance and
are not consistently reported by type of service, state, or industry.

Data represent past performance. Because TAA performance is measured using
UI wage records and long-term performance measures such as employment
retention, the most up-to-date TAA performance data currently available
may represent performance from several years in the past.

17Labor's Web site for TAA statistics can be found at
http://www.doleta.gov/tradeact/taa_stats.cfm.

Page 24 GAO-06-496 Trade Adjustment Assistance

Use of wage records: Using UI wage records to measure outcomes provides a
common yardstick for making long-term comparisons across states because
they contain wage and employment information on most workers. At the same
time, these files suffer delays between the time an individual gets a job
and when this information appears in wage records. State procedures for
collecting and compiling wage information from employers can be slow and
time-consuming. Data are collected from employers only once every quarter,
and employers in most states have 30 days after the quarter ends to report
the data to the state. For example, the wage report for the last calendar
quarter of the year (ending on December 31) is due to the state on January
31. We previously reported that for the majority of states, the delay
between the time an individual gets a job and the time this information
appears in wage records is up to 4 months. 18

Design of measures: In addition to using a job placement measure, Labor
also uses two longer-term measures to gauge TAA performance-an earnings
measure and a job retention measure. These measures may be useful for
assessing how well the program is meeting its long-range goals to increase
the employment, retention, and earnings of participants. However, the use
of these measures requires states to wait from one to three quarters after
participants exit the TAA program before measuring the outcomes. For
example, although states record whether participants entered employment in
the first quarter after exit, two more quarters must elapse before
employment retention is measured. Participants who exit the TAA program
have their outcomes assessed in the first, second, and third quarters
after exit. However, data to measure all outcomes are not available until
the fifth quarter after exit, and the outcomes are not submitted to Labor
until midway through the sixth quarter. Figure 5 illustrates the time it
takes before a TAA participant would be included in performance outcome
calculations.

18GAO-04-657.

Figure 5: Period between When a TAA Participant Exits the Program and When Data
                             Are Reported to Labor

Source: GAO analysis of Department of Labor guidance and Art Explosion.

Note: The outcome measures shown in this figure are those used prior to
implementation of common measures in fiscal year 2006. Under common
measures, retention is measured in both the second and third quarters
after exit.

Labor posts TAA performance data on its Web site on an annual basis, and
because of the time required to collect outcome data on exiters, the
performance data that Labor makes available each year include workers who
exited the program up to 30 months earlier. For example, states were
required to submit performance data to Labor on individuals who exited the
program in the July-to-September quarter of 2002 by February 15, 2004.
However, these data were not publicly released until early 2005, when they
were included in the annual data report covering those who exited the
program between July 2002 and June 2003 (see fig. 6). Furthermore,
individuals who exited the TAA program in July 2002 and who received
training could have entered the program 2 years earlier, or July 2000- 4
1/2 years before their outcome data were included in the annual reporting
of outcomes.

Figure 6: Fiscal Year 2004 TAA Performance Data Represent Exiters from as early
                              as Fiscal Year 2002

                                                                 Labor receives final
                                                                 quarterly report and
                                                                 compiles fiscal year
                                                                   2004 annual report
                         ...are included in fiscal year 2004
                           quarterly reports to Labor for
                            these quarters                                Labor posts
                                                                          fiscal year
Participants who exit in                                                2004 outcomes
                                                                           on Website
these quarters...                                                       
                                                                             Early    
                                                                        Nov  calendar 
Jul-Sep, Oct-Dec, Jan-Mar, Apr-Jun, Oct-Dec, Jan-Mar, Apr-Jun, Jul-Sep, 15,  year,    
2002     2002     2003     2003     2003     2004     2004     2004     2004 2005     

Source: GAO analysis of Department of Labor guidance.

Categories used to display outcome data are not sufficiently informative.
The categories Labor uses to report program outcomes are unlikely to
provide policymakers a broad understanding of different aspects of the
program's performance. Labor combines all participants in the publicly
available outcome information, regardless of the types of services and
benefits they received. As a result, there is no information available to
the public on how the outcomes of participants who received TAA-funded
training differ from those of participants who received other benefits or
services, such as job search assistance or extended income support. In
addition, Labor does not consistently report TAA participant and activity
data by state or by industry. This makes it difficult for policymakers to
accurately assess program activities and performance and determine future
needs.

Examining data by different categories may also allow Labor to recognize
and address problems related to performance and data quality. We and other
experts have found that reporting performance information in smaller,
meaningful categories can help identify and resolve problems. For example,
comparing the performance of states on different TAA measures may draw
attention to low-performing states. Furthermore, careful analysis of
disaggregated data could uncover data quality issues. For example, if
Labor analyzed each state's data by the different benefits and services
TAA participants received, then it may be able to identify those states
that are excluding some exiting participants in their TAPR submissions by
looking at which states reported few or no participants in certain
categories.

Ongoing study will assess program impact. Labor has funded a longterm
study to assess the impact of TAA program services such as training on
participants' employment and earnings that will provide policymakers a
broader understanding of the program's effectiveness. The goal of the
study is to determine not only the outcomes achieved by TAA participants,
but also the impact of TAA program services-that is, whether participants
had better outcomes as a result of the program than they would have if
they had not received program services. Labor last completed an evaluation
of the TAA program in 1993, but methodological issues resulted in
inconclusive findings from that study. According to Labor officials, the
methodology used by the new study is an improvement over the methodology
used by the 1993 study and should provide more conclusive findings. The
new study will compare the outcomes for a treatment group (TAA
participants in 25 states) and a comparison group (UI claimants in the 25
states who are similar to the TAA participants in a number of observable
characteristics). It will examine, for example, the workers' job search
methods, their training outcomes, and their employment history before and
after being laid off. This methodology will likely allow an assessment of
the impact of the TAA program, rather than just outcomes. Data collection
began in 2005 and will continue until 2008, and a final report is
scheduled to be issued by the end of 2008.

Most States Find TAA Performance Data Useful, but Many Would Like
Additional Information

While approximately one-third of the states found TAA performance
information they currently receive from Labor to be greatly useful, some
would like Labor to provide them with additional information to help
manage their program. Nearly half of the 46 states we surveyed told us
that they find the performance information they receive from Labor to be
moderately useful (see fig. 7), and 8 states reported that Labor's TAA
performance information is of little or no use for program management.

Figure 7: Many States Found Labor's Performance Information to Be
Moderately Useful

Number of states25

22

20

15

15

10

8

5

0

                                       y

                                       y

                          ModeratelLittle or noGreatl

    Reported usefulness

Source: GAO survey of state TAA officials.

Note: One state responded "uncertain" and was not included in this figure.

Nearly half of the states we surveyed told us that they routinely develop
information on their own performance beyond what they submit to Labor. For
example, an official in 1 state reported that it calculates its own
outcomes before receiving them from Labor in order to make managers and
executives aware of the state's performance, and it uses this information
to engage state and local TAA staff in making program adjustments. In
addition, approximately one-third of the states routinely develop
information on their local areas' performance. Labor does not provide
analysis of local area performance to states because it does not collect
this type of information in the TAPR.

While many states provide the performance information of their own state
and that of other states to their local area TAA staff, few states provide
information on their local areas' performance to local TAA staff. Only 27
of the 46 states in our survey reported that they share information from
Labor with local area staff on how their state's performance compares to
national TAA performance goals. In addition, only 7 of the 16 states that
generate additional performance information for local areas reported that
they share this information with local TAA staff. One expert we spoke with
told us that regularly sharing performance information with local program
staff enables them to understand how the data they collect are being used
and the importance of complete and accurate data for producing reliable
performance information. Our recent report on performance measurement also
noted that frequent and routine communication of performance information
helps program staff and stakeholders use such information to accomplish
program goals as they pursue day-to-day activities. 19 These practices
could lead to better program management and produce more reliable
performance data to assess TAA performance in the future.

States said that they would like to receive additional performance
information from Labor to help manage their TAA program. Thirty-four
states would like more information than they currently receive on their
own state's performance, and 39 states reported they would like
information comparing their states' TAA performance to their WIA
Dislocated Worker performance (see fig. 8). 20 According to one state
official we spoke with, receiving additional TAA performance information
that is displayed by type of service and by state would enable officials
to respond more effectively to performance problems and to learn what
strategies states with similar TAA populations are using to achieve
different performance outcomes.

19See GAO, Managing for Results: Enhancing Agency Use of Performance
Information for Management Decision Making, GAO-05-927 (Washington, D.C.:
Sept. 9, 2005), and GAO, Program Evaluation: An Evaluation Culture and
Collaborative Partnerships Help Build Agency Capacity , GAO-03-454
(Washington, D.C.: May 2, 2003).

The WIA Dislocated Worker program serves dislocated workers in general and
is not reserved for those affected by international trade. States and
local areas are required to monitor the outcomes of WIA Dislocated Worker
program participants on measures including job placement, job retention,
and earnings change.

Figure 8: Most States Reported That Additional Performance Information
from Labor Would Be Useful for Program Management Number of states50 45 40
39 35

34

35

                                      3333

                                                           30 25 20 15 10 5 0

                                   AA to WIA

         instatesOther iongthe resLocal areaide ofsoutstateSiongthe re

                                tatesditional s

                                      rker

                                       s

                                       i

                                       y

                                     analo

                                  located wsdi

                                   TgComparin

                                       Ad

      Type of performance information

Source: GAO survey of state TAA officials.

While it has limited authority to hold states accountable, Labor has taken
steps to improve the quality of TAA data states submit, but these steps do
not fully address all issues. Labor has no mechanism to sanction states
for poor performance or poor-quality data because the law and current
regulations do not provide one. However, Labor has begun an initiative
that requires states to review a sample of their data for accuracy. It is
too soon to fully assess whether Labor's efforts have improved data
quality, but most states reported on our survey that Labor's new
requirements have increased awareness of data quality at the state and
local levels. States also report that they would like more opportunities
to share lessons learned about issues related to data quality. Labor is
requiring changes in some TAA performance measures to align them with
measures for other federally funded job training programs. Many states
reported that the

  Labor Has Taken Steps to Improve the Quality of Performance Data, but Issues
  Remain

Data Validation Is Having Positive Effects but Does Not Address All
Concerns

changes are burdensome, and some states are experiencing delays in
implementing the changes.

To address data quality concerns, Labor developed a process for states to
use to validate the TAPR data they submitted to Labor. Starting with data
submitted in fiscal year 2003, Labor required states to review a sample of
participants' records and compare what was reported for certain data
elements to data in source files. State staff review the source files and
record whether each data element is supported by source documentation and,
therefore, passed data validation. If the source files show a data element
was incorrect or was not supported with documentation, the data element
fails. States use Labor's software to calculate error rates, and they
submit the results to Labor.

While it is too soon to assess whether Labor's data validation efforts
have improved data quality, many states said that the efforts are having a
positive effect. Thirty-five states reported that efforts have improved
the accuracy of the data. Thirty-seven of the 46 states told us they have
helped increase the awareness of data quality at the state level, and 25
states told us they have improved awareness at the local level (see fig.
9).

Figure 9: States' View of How Labor's Data Validation Efforts Have Helped
Them State's awareness of

37

data quality

    Data accuracy

35

28Data collection process

Local's awareness of data quality                   25  
                0       5             10       15  20  25  30  35  40  45  50 
                              Number of states                             
Source: GAO survey of state TAA officials.                              

Until recently, Labor has not had a standard process for ensuring that
states performed data validation correctly. Labor officials tell us,
however, that beginning in 2006, regional offices are conducting data
validation compliance reviews of a subsample of validated records to
ensure that the

Labor Has Taken Several Other Steps to Help Improve TAA Data, but States
Want More Opportunities to Share Lessons Learned

records were accurately validated and the files contained all required
source documents.

While states report that Labor's data validation requirements are having
some positive effects, Labor's data validation efforts do not address two
key problems. First, guidance for data validation defined for the first
time the type of source documents needed to validate TAPR data elements,
including exit dates. However, the guidance does not specify that the
source documents for training completion dates should show that
participants actually completed training. Second, data validation does not
provide for assessing whether TAPR submissions are complete. Because the
data validation process only covers participant records included in
states' TAPR submissions for the year, it does not look beyond those
records to determine whether all exiting participants were included.

In addition to implementing data validation, Labor has taken various
actions to better instruct states and to provide tools for improving the
data they submit to Labor.

Technical assistance and training: In 2005 and 2006, Labor brought
together state TAA staff for training conferences on the new data
requirements for implementing common measures. According to Labor
officials, Labor's regional offices periodically hold roundtables with
states to discuss issues that sometimes include data quality. Labor
provides technical assistance, as needed, to states through telephone
calls and e-mails. According to Labor officials, Labor is planning to
start holding quarterly conference calls with states about TAA issues,
including data quality.

Guidance on data reporting: Labor issued guidance and instructions for TAA
data reporting, such as instructions defining how "date of exit" is to be
determined under common measures. In May 2005, Labor issued a guidance
letter to states addressing several issues with data quality, such as the
use of WRIS and supplemental data to determine employment outcomes. In
general, states reported that the guidance and training they had received
from Labor provided a clear understanding of certain data requirements,
such as the requirements for data validation and for using UI wage
records. States were somewhat less likely to say that Labor had provided a
clear understanding of the documentation needed for the date of exit and
how supplemental data could be used to document TAA employment outcomes.

Monitoring: Labor's regional offices conduct monitoring visits to review
states' TAA programs. In the past, Labor did not have a standard protocol
for these monitoring visits, and the monitoring did not always cover the
quality of the TAA data being submitted by states. However, as of March
2006, Labor was developing a standard monitoring guide for its regional
staff.

Pilot project on federal employment data: Labor collaborated with the
Office of Personnel Management, the U.S. Postal Service, and the
Department of Defense to create a pilot data exchange system to provide
states access to wage record information on federal and military
employment. The system that began operating in November 2003 can help
states obtain more complete employment outcome data on participants who
exited job training programs because it provides information on federal
employment that is not available in state UI wage records. Many states are
using the system to help determine employment outcomes for job training
programs, such as those funded under the Workforce Investment Act.
However, only 3 of the 46 states we surveyed reported that they were
routinely using this system to obtain employment outcomes for the TAA
program.

Despite Labor's efforts to improve data quality, most states would like
more help. Most states reported that they do not currently have
opportunities to share lessons learned with other states on topics related
to TAA data quality, such as how to use supplemental data, and they
expressed interest in having such opportunities. For example, 29 states
told us they do not currently have opportunities to share lessons learned
on data validation, and 44 states told us more opportunities to do so
would be helpful (see fig. 10).

Labor's Common Measures May Address Some Concerns, but States Are Still
Struggling to Implement Changes

Figure 10: Most State Expressed Interest in Having More Opportunities to
Share Lessons Learned on Data Issues

Number of states 50

45 44 40

40

38

35

33

                                       29

30 29 25 20 15 10

5 0 Data validation Use of Use of WRIS supplemental data Issues for
sharing lessons learned

     Do not currently have opportunitiesMore opportunities would be helpful

                   Source: GAO survey of state TAA officials.

In response to an OMB initiative, Labor made changes to some of the TAA
performance measures and to TAA reporting requirements in order to
implement common measures (see table 3). OMB established a set of common
performance measures to be applied to most federally funded job training
programs that share similar goals. Labor further defined the common
measures for all of its Employment and Training Administration programs
and required states to start reporting TAA data under the revised
requirements in fiscal year 2006.

       Table 3: How TAA Performance Measures Changed under Common Measures

                          Past measure Common measure

Wage replacement: earnings in the Average earnings: total earnings in the
second and third quarters after exit divided second and third quarters
after exit divided by earnings in the second and third by the number of
exiting participants quarters prior to dislocation

Reemployment rate: percentage of Entered employment rate: same as the
program exiters employed in first quarter past reemployment rate measure
after exit

Retention rate: percentage of exiters Retention rate: percentage of
exiters employed in the first quarter after exit that employed in the
first quarter after exit that are still employed in third quarter after
exit are still employed in both the second and

third quarters after exit

Source: Department of Labor guidance.

Moving to common measures may increase the comparability of outcome
information across programs and make it easier for states and local areas
to collect and report performance information across the full range of
programs that provide services in a one-stop system. Prior to common
measures, many federal job training programs had performance measures that
tracked similar outcomes but had variation in the terms used and the way
the measures were calculated. For example, the programs used different
time periods to assess whether participants got jobs. Under common
measures, the time period used to assess employment outcomes is uniform
across all covered programs.

Implementation of common measures involved some changes in the data states
collect for the TAPR:

Standardized exit definitions: Labor's guidance on common measures
provides for a clearer understanding of when TAA participants should be
exited from the program than did earlier TAA guidance. Under Labor's
guidance, states must wait 90 days after participants receive their last
service or benefit-from TAA, WIA, or other related programs-to record them
as exiters. Prior to this change, states could exit participants without
waiting 90 days. Most states reported that the guidance and training they
received from Labor provided a clear understanding of the definition of
exit under common measures, but 7 states disagreed.

Coordination of exit dates: Under common measures, states are encouraged
to establish a common exit date for each participant who is co-enrolled in
more than one program. For example, if a participant receives services
under TAA and under WIA, then the two programs should use the same exit
date for the participant. Coordinating exit dates improves data quality by
avoiding the problem of counting a participant as unemployed in the
program's performance measures when, in fact, the participant is still
receiving services in another program and is not ready to be counted in
the performance measures.

Changes in IT systems: A number of data fields were added or changed in
the TAPR as part of the new common measures policy, requiring states to
add or change data fields in their IT systems and to instruct staff on
changes in data to be collected on participants and employment outcomes.
Most states reported that the guidance and training they received from
Labor provided a clear understanding of the changes needed in the TAPR to
implement common measures; however, 7 states disagreed.

Although moving to common measures may ultimately make it easier for
states to collect and report performance information across programs, most
states reported that making changes to implement common measures had been
a burden in terms of time and cost (see fig. 11), and often viewed
coordinating exit dates as burdensome. States were nearly evenly divided
in their views, however, on whether they had been given sufficient time by
Labor to complete the changes. Nineteen states said they had not been
given sufficient time, while 18 states said they had. Twentysix states
reported that they will have provided guidance to staff or changed data
elements in their IT systems by the time the first quarterly TAPR is due
in fiscal year 2006 (see fig. 12). Other states reported that they would
have these changes completed sometime later in 2006, while some states
said they could not estimate when they will complete the changes.
Coordinating exit dates was the change that states considered the most
burdensome. Seventeen states were unable to estimate when they would be
able to coordinate exit dates across programs.

Figure 11: Most States Viewed Changes Needed for Common Measures to Be a
Burden

27

Coordinating exit dates

                                       9

18

      IT system changes

16

15

      Providingguidance to staff

16

05 10152025 Number of states

Great or very great burden Moderate burden

                   Source: GAO survey of state TAA officials.

Figure 12: Coordinating Exit Dates Was Change States Were Least Likely to
Expect to Have Completed in 2006

Number of States 30

26 26

25

20 19 17

15

13 12

10

10

8

7

5

0 CoordinatingIT system Providing exit dateschangesguidance

Types of changes to implement common measures

By first quarterly report in fiscal year 2006

Later in 2006

Unable to estimate Source: GAO survey of state TAA officials.

In a previous study, we cautioned that rushed implementation of reporting
changes may not allow states and local areas enough time to fully meet the
requirements and could negatively affect the data quality of the
information reported. 21

Since the passage of the TAA Reform Act of 2002, the TAA program has

Conclusions

evolved to become one of the most important means to help the workers
affected by our nation's trade policies rejoin our nation's workforce. The
program has seen substantial increases in the population it serves and in
the funds available to serve them. Unfortunately, efforts to monitor the
program's performance have not kept pace with the program's

21GAO-05-539.

development. Four years after the passage of the reforms, we still do not
know whether the program is achieving what lawmakers intended.

The TAA program has suffered a history of problems with its performance
data that have undermined the data's credibility and limited their
usefulness. And while we see that Labor has taken some steps aimed at
improving the performance data, the data remain suspect. They fail to
capture outcomes for some of the program's participants, and many
participants are not included in the final outcomes at all. These failures
may have contributed to the program's poor performance in achieving its
national goals. Labor lacks the authority to hold states accountable for
their outcomes or for the quality of their data, and as a result, some
states may not see the value of investing more effort to ensure their data
are complete and accurate. In truth, officials tell us the funding to
support their efforts is small, and it fluctuates from year to year,
making such an investment difficult to sustain. But the success of the
program is being judged by the outcomes the program achieves and whether
or not it meets its goals. The current budgetary environment makes it
risky not to take all necessary steps to ensure that the outcomes are an
accurate and credible reflection of the program's performance.

Labor has taken a major step toward improving the quality of its
performance data through its new data validation requirements. States
report that these requirements have significantly raised the awareness of
data quality at the state and local levels-an essential component in any
effort to improve the accuracy of the data. But these efforts do not fully
address all issues. No steps have been taken to ensure that all
participants are included in the TAA performance data or that exit dates
are adequately documented. Monitoring can help address data issues, but
Labor is just now developing a standard monitoring guide that would help
ensure that key problems are identified during monitoring visits. Until
these steps are complete, the data can not be verified and may remain
incomplete. Providing opportunities for states to share lessons learned
may make states more aware of effective approaches for ensuring data
quality, and several states expressed an interest in more such
opportunities.

Labor has recently improved the availability of TAA performance
information by posting the information on its Web site and by making some
state-by-state performance data available. However, the performance data
are not as informative as they could be because they aggregate all
participants and do not show the outcomes of participants based on the
types of services they received. As a result, policymakers

  Recommendations for Executive Action

lack the information they need to understand program participation and
performance and to assess future needs.

While Labor has taken steps to share information with states and to
improve data quality, more work is needed.

To help ensure that TAA participant data reported by states are
consistent, complete, and accurate, Labor should

        * clarify through guidance and other communications with states
             o that all participants who exit the program should be included
               in the TAPR and
             o the documentation needed to verify the training completion
               date;
     o ensure that the core monitoring guide currently under development for
       regional office site visits includes guidance for assessing whether
       states' data collection processes for performance reporting capture
       all participants; and
     o provide states with opportunities to share lessons learned with other
       states on issues that may affect data quality.

To make TAA performance information more useful for program management,
Labor should provide this information by the type of services received by
TAA participants.

                                Agency Comments

We provided a draft of this report to Labor for review and comment. In its
comments, Labor did not disagree with our findings and recommendations and
said the report will be helpful in its continuing efforts to improve the
quality of TAA performance data. Labor noted that the issues raised in the
report about administrative costs and the burden of new reporting
requirements are compounded by having a workforce investment system that
is duplicative in its service delivery design, resulting in separate
record-keeping and reporting systems. Labor also identified a number of
actions that it is taking to ensure that performance accountability is an
expectation of the program. A copy of Labor's response is in appendix IV.

As arranged with your offices, unless you publicly announce the contents
of this report earlier, we plan no further distribution of this report
until 30 days from the date of this report. At that time, we will send
copies of this report to the Secretary of Labor, relevant congressional
committees, and others who are interested. Copies will also be made
available to others

Page 41 GAO-06-496 Trade Adjustment Assistance

upon request. The report is also available on GAO's home page at
http://www.gao.gov.

If you or members of your staff have any questions about this report,
please contact me at (202) 512-7215 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
this report are listed in appendix V.

Sigurd R. Nilsen

Director, Education, Workforce, and Income Security Issues

Appendix I: Objectives, Scope, and Methodology

We examined (1) whether the Trade Adjustment Assistance (TAA) performance
data provide a credible picture of the program's performance, (2) what TAA
performance data Labor makes available to the public and states and the
usefulness of the data for managing the program, and (3) what Labor is
doing to address issues with the quality of TAA data submitted by states.

To learn more about the factors that affect TAA data quality and to learn
what states are doing to ensure data quality, we conducted a Web-based
survey of state TAA officials and conducted site visits in five states,
where we interviewed state officials and visited local areas or one-stop
centers. We also collected information on the quality of TAA data through
interviews with Department of Labor officials in headquarters and all six
regional offices, nationally recognized experts, and reviewed relevant
literature. Our work was conducted between December 2004 and March 2006 in
accordance with generally accepted government auditing standards.

To determine the factors that affect the quality of TAA performance data,
we conducted a Web-based survey of workforce officials in the 46 states 1
that were allocated TAA funds in fiscal year 2005, and we obtained a 100
percent response rate. These officials were identified using Labor's list
of state TAA officials. We e-mailed the contacts, and they confirmed that
they were the appropriate contact for our survey or identified and
referred us to another person at the state level. Survey topics included

(1)
           the current status of TAA data collection and reporting systems,

(2)
           implementation of the U.S. Department of Labor's data validation
           requirements, (3) state and local efforts to ensure the quality of
           TAA data, and (4) the implementation of common measures. The
           survey was conducted using a self-administered electronic
           questionnaire posted on the Web. We contacted respondents via
           e-mail announcing the survey, and sent follow-up e-mails to
           encourage responses. The survey data were collected between
           November 2005 and January 2006. We received completed surveys from
           all 46 states that were allocated TAA funding in fiscal year 2005
           (a 100 percent response rate). We did not include Washington,
           D.C., and U.S. territories in our survey.

1

Delaware, Hawaii, North Dakota, and Wyoming were not included in our
survey because they were not allocated TAA funds in fiscal year 2005.

Page 43 GAO-06-496 Trade Adjustment Assistance Appendix I: Objectives,
Scope, and Methodology

                                Web-Based Survey

We worked to develop the questionnaire with social science survey
specialists. Because this was not a sample survey, there is no sampling
error. However, the practical difficulties of conducting any survey may
introduce errors, commonly referred to as nonsampling errors. For example,
differences in how a particular question is interpreted or in the sources
of information that are available to respondents can introduce unwanted
variability into the survey results. We took steps in the development of
the questionnaire, the data collection, and data analysis to minimize
these nonsampling errors. For example, prior to administering the survey,
we pretested the content and format of the questionnaire with four states
to determine whether (1) the survey questions were clear,

(2) the terms used were precise, (3) respondents were able to provide the
information we were seeking, and (4) the questions were unbiased. We made
changes to the content and format of the final questionnaire based on
pretest results. We also performed computer analyses to identify
inconsistencies in responses and other indications of error. In addition,
a second independent analyst verified that the computer programs used to
analyze the data were written correctly.

Site Visits

We visited five states-California, Iowa, Ohio, Texas, and Virginia-and
traveled to local areas or one-stop centers in each of these states. We
selected these states because they represent different TAA data collection
approaches (that is, states where data are entered into information
technology [IT] systems at the local level and those where data are
entered at the state level), received a relatively large share of TAA
funds in fiscal year 2005, and are geographically dispersed. From within
each state, we judgmentally selected local areas to visit (see table 4).
In each state, we interviewed state TAA officials about their collection
and use of TAA data, IT systems used to compile TAA performance data, and
efforts to ensure the data are complete and accurate. Similarly, we
interviewed local area officials about their collection and use of TAA
data.

Information that we gathered on our site visits represents only the
conditions present in the states and local areas at the time of our site
visits, from January 2005 through October 2005. We cannot comment on any
changes that may have occurred after our fieldwork was completed.
Furthermore, we cannot generalize the findings from our site visits beyond
the states and local areas we visited.

Appendix I: Objectives, Scope, and Methodology

                   Table 4: Site Visit States and Local Areas

                                State Local area

                           California Alameda County

                                 Orange County

                                  Iowa Newton

                                   Burlington

                              Ohio Franklin County

                               Texas San Antonio

                                     Dallas

                             Virginia Harrisonburg

                                    Danville

Source: GAO.

Appendix II: Program Information in State TAA IT Systems

In our survey, states were asked whether the IT system they use to compile
data for the Trade Act Participant Report (TAPR) currently captures
program information for certain other Labor programs or benefits.

Table 5: Program Information for Other U.S. Department of Labor Employment
and Training Administration Programs Contained in State IT Systems Used to
Compile TAPR Submissions

Trade

Readjustment Workforce

Allowance Investment Act State (TRA) (WIA)

Veterans National Employment and Unemployment Employment Emergency
Training Insurance (UI) Services (ES) Grant (NEG) Program (VETS) Program

                               Alabama X X X X X

                                  Alaska X X X

                                Arizona X X X X

                               California X X X X

                              Colorado X X X X X X

                              Florida X X X X X X

                               Georgia X X X X X

                                   Idaho X X

                                  Illinois X X

                               Indiana X X X X X

                                 Iowa X X X X X

                                Kansas X X X X X

                              Kentucky X X X X X X

                                Maine X X X X X

                                Maryland X X X X

                           Massachusetts X X X X X X

                                 Michigan X X X

                                  Minnesota X

                            Mississippi X X X X X X

                               Missouri X X X X X

                             Montana

                           North Carolina X X X X X X

                                New Hampshire X

                              New Jersey X X X X X

                              New York X X X X X X

                                   Oklahoma X

                                Oregon X X X X X

                              South Carolina X X X
							  
								South Dakota X X X X X
								Tennessee X X X X X X
								Texas X X X X X X
								Utah X X X X X X
								Vermont X X X X X X
								Virginia X X
								Washington X X X X X
								West Virginia X X X X X X
								Wisconsin X X X X X X

Trade

Readjustment Workforce

Allowance Investment Act State (TRA) (WIA)

Veterans National Employment and Unemployment Employment Emergency
Training Insurance (UI) Services (ES) Grant (NEG) Program (VETS) Program

Source: GAO survey of state TAA officials.

Note: Nine states that reported compiling their TAPR manually were
excluded from the table: Arkansas, Connecticut, Louisiana, Nebraska,
Nevada, New Mexico, Ohio, Pennsylvania, and Rhode Island.

Appendix III: State TAA Administrative Allocations, Fiscal Years 2005-2006

The TAA Reform Act authorizes up to $220 million per year for training
under the TAA Program. Labor allocates 75 percent of the training funds to
states according to a formula that takes into account each state's
previous year allocations, accrued expenditures, and participant levels.
Labor holds the remaining 25 percent of training funds in reserve to
distribute to states throughout the year according to need. To cover
administrative costs associated with training under the TAA program, Labor
allocates to each state additional administrative funds equal to 15
percent of its training allocation. Table 6 shows Labor's initial 75
percent allocation for training and associated administrative expenses.
States also receive an additional 15 percent of any reserve (25 percent)
funding and job search/relocation allowances for program administration.

Table 6: State TAA Training and Administrative Allocations, Fiscal Years
2005-2006         
                   2005 Allocation                 2006 Allocation
States                Training Administration      Training Administration 
Alabama             $2,468,374     $370,256     $2,642,640        $396,396 
Alaska                 398,625      59,794          429,982         64,497 
Arizona              2,358,372     353,756       2,440,988         366,148 
Arkansas             2,059,660     308,949       1,750,711         262,607 
California           6,180,645     927,097       6,642,537         996,380 
Colorado             1,678,693     251,804       1,426,889         214,033 
Connecticut          1,765,584     264,838       1,500,746         225,112 
Delaware                     0        0                   0              0 
District of                  0        0                   0              0 
Columbia                                                    
Florida              3,941,816     591,272       3,350,544         502,582 
Georgia                854,284     128,143       1,559,104         233,866 
Hawaii                       0        0                   0              0 
Idaho                2,332,696     349,904       2,390,380         358,557 
Illinois             4,294,247     644,137       4,696,350         704,452 
Indiana              4,432,026     664,804       4,780,198         717,030 
Iowa                 3,336,400     500,460       2,835,940         425,391 
Kansas               3,265,572     489,836       2,775,736         416,360 
Kentucky             2,998,984     449,848       3,705,162         555,774 
Louisiana              594,658      89,199          612,573         91,886 
Maine                3,674,863     551,229       4,021,621         603,243 
Maryland               482,983      72,447          525,184         78,778 
Massachusetts        5,473,152     820,973       5,600,876         840,131 
Michigan             5,559,171     833,876       5,774,380         866,157 

Appendix III: State TAA Administrative Allocations, Fiscal Years 2005-2006

                  2005 Allocation                      2006 Allocation
States               Training Administration       Training Administration 
Minnesota           3,824,119     573,618         4,005,739        600,861 
Mississippi         1,909,216     286,382         2,076,016        311,402 
Missouri            4,993,894     749,084         4,244,810        636,721 
Montana             1,054,844     158,227         1,109,440        166,416 
Nebraska              469,538     70,431            480,298         72,045 
Nevada                298,265     44,740            253,525         38,029 
New Hampshire         600,301     90,045            510,256         76,538 
New Jersey          1,545,011     231,752         1,698,502        254,775 
New Mexico            444,554     66,683            377,871         56,681 
New York            2,496,152     374,423         2,642,798        396,420 
North Carolina      8,174,834    1,226,225        9,918,421      1,487,763 
North Dakota                0        0                    0              0 
Ohio                4,226,657     633,998         4,579,676        686,951 
Oklahoma            1,440,538     216,081         1,523,960        228,594 
Oregon              5,116,592     767,489         5,242,514        786,377 
Pennsylvania       17,538,533    2,630,779       14,907,751      2,236,165 
Rhode Island          690,084     103,513           734,856        110,228 
South Carolina      5,137,159     770,574         4,366,585        654,988 
South Dakota          341,148     51,172            371,610         55,741 
Tennessee           2,464,473     369,671         2,681,734        402,260 
Texas              10,638,355    1,595,753       11,149,519      1,672,428 
Utah                2,134,549     320,182         1,814,367        272,155 
Vermont               287,696     43,154            296,965         44,545 
Virginia            5,222,843     783,426         5,712,451        856,868 
Washington         13,920,774    2,088,116       14,357,300      2,153,595 
West Virginia         770,639     115,596         1,038,332        155,750 
Wisconsin          11,108,427    1,666,264        9,442,163      1,416,324 
Wyoming                     0        0                    0              0 
Total            $165,000,000   $24,750,000    $165,000,000    $24,750,000 

                          Source: Department of Labor.

Appendix IV: Comments from the Department of Labor

                 Page 50 GAO-06-496 Trade Adjustment Assistance

Appendix IV: Comments from the Department of Labor

                 Page 51 GAO-06-496 Trade Adjustment Assistance

Appendix IV: Comments from the Department of Labor

                 Page 52 GAO-06-496 Trade Adjustment Assistance

Appendix V: GAO Contact and Staff Acknowledgments

Sigurd R. Nilsen, Director (202) 512-7215

GAO Contact

                                Acknowledgments

Dianne Blank, Assistant Director Kathy Peyman, Analyst-in-Charge

In addition, the following staff made major contributions to this report:
Vidhya Ananthakrishnan, Melinda Cordero, Laura Heald, Adam Roye, and
Leslie Sarapu served as team members; Amanda Miller and Carolyn Boyce
advised on design and methodology issues; Rachael Valliere advised on
report preparation; Jessica Botsford advised on legal issues; Lise Levie
verified our findings.

GAO Related Products

Trade Adjustment Assistance: Most Workers in Five Layoffs Received
Services, but Better Outreach Needed on New Benefits. GAO-06-43.
Washington, D.C.: January 31, 2006.

Workforce Investment Act: Labor and States Have Taken Actions to Improve
Data Quality, but Additional Steps Are Needed. GAO-06-82. Washington,
D.C.: November 14, 2005.

Workforce Investment Act: Substantial Funds Are Used for Training, but
Little Is Known Nationally about Training Outcomes. GAO-05-650.
Washington, D.C.: June 29, 2005.

Unemployment Insurance: Better Data Needed to Assess Reemployment Services
to Claimants. GAO-05-413. Washington, D.C.: June 24, 2005.

Workforce Investment Act: Labor Should Consider Alternative Approaches to
Implement New Performance and Reporting Requirements. GAO-05-539.
Washington, D.C.: May 27, 2005.

Trade Adjustment Assistance: Reforms Have Accelerated Training Enrollment,
but Implementation Challenges Remain. GAO-04-1012. Washington, D.C.:
September 22, 2004.

Workforce Investment Act: States and Local Areas Have Developed Strategies
to Assess Performance, but Labor Could Do More to Help. GAO-04-657.
Washington, D.C.: June 1, 2004.

National Emergency Grants: Labor Is Instituting Changes to Improve Award
Process, but Further Actions Are Required to Expedite Grant Awards and
Improve Data. GAO-04-496. Washington, D.C.: April 16, 2004.

Workforce Investment Act: Improvements Needed in Performance Measures to
Provide a More Accurate Picture of WIA's Effectiveness. GAO-02-275.
Washington, D.C.: February 1, 2002.

Trade Adjustment Assistance: Experiences of Six Trade-Impacted
Communities. GAO-01-838. Washington, D.C.: August 24, 2001.

Trade Adjustment Assistance: Trends, Outcomes, and Management Issues in
Dislocated Worker Programs. GAO-01-59. Washington, D.C.: October 13, 2000.

  (130437)

  GAO's Mission

The Government Accountability Office, the audit, evaluation and
investigative arm of Congress, exists to support Congress in meeting its
constitutional responsibilities and to help improve the performance and
accountability of the federal government for the American people. GAO
examines the use of public funds; evaluates federal programs and policies;
and provides analyses, recommendations, and other assistance to help
Congress make informed oversight, policy, and funding decisions. GAO's
commitment to good government is reflected in its core values of
accountability, integrity, and reliability.

The fastest and easiest way to obtain copies of GAO documents at no cost
is through GAO's Web site ( www.gao.gov) . Each weekday, GAO posts GAO
Reports and newly released reports, testimony, and correspondence on its
Web site. To

have GAO e-mail you a list of newly posted products every afternoon, go to
www.gao.gov and select "Subscribe to Updates."

                             Order by Mail or Phone

The first copy of each printed report is free. Additional copies are $2
each. A check or money order should be made out to the Superintendent of
Documents. GAO also accepts VISA and Mastercard. Orders for 100 or more
copies mailed to a single address are discounted 25 percent. Orders should
be sent to:

U.S. Government Accountability Office 441 G Street NW, Room LM Washington,
D.C. 20548

To order by Phone: Voice: (202) 512-6000 TDD: (202) 512-2537 Fax: (202)
512-6061

  To Report Fraud, Contact:

Web site: www.gao.gov/fraudnet/fraudnet.htm

  E-mail: [email protected]

Federal Programs Automated answering system: (800) 424-5454 or (202)
512-7470

Gloria Jarmon, Managing Director, [email protected] (202) 512-4400 U.S.
Government Accountability Office, 441 G Street NW, Room 7125 Relations
Washington, D.C. 20548

Paul Anderson, Managing Director, [email protected] (202) 512-4800

  Public Affairs

U.S. Government Accountability Office, 441 G Street NW, Room 7149
Washington, D.C. 20548

    PRINTED ON

RECYCLED PAPER
*** End of document. ***