Reforming Trade Adjustment Assistance: Keeping a 40-Year Promise
Paper presented at the Institute conference on "Trade Policy in 2002"
© Peterson Institute for International Economics
The author served as a consultant to the Senate Finance Committee in the spring of 2001 and contributed to drafting S 1209. He is grateful to Shara Aranoff, Steve Beckman, C. Fred Bergsten, I. M. Destler, Kim Elliott, Gary Hufbauer, Lori Kletzer, and David Richardson for their constructive comments.
"Those injured by (that) trade competition should not be required to bear the full brunt of the impact. Rather, the burden of economic adjustment should be borne in part by the federal government .... (T)here is an obligation to render assistance to those who suffer as a result of national trade policy."
This statement, made by President John F. Kennedy in 1962, could have easily been made by any number of Democratic and Republican Presidents, Cabinet members, Senators, and Congressmen and women over the last 40 years. And yet, the US government continues to do very little toward meeting this objective.
When President Kennedy made this statement, the United States had a trade surplus and imports amounted to less than 5 percent of GDP. Manufacturing employment was 30 percent of total employment. Over the last 40 years, despite strong employment growth, millions of US workers have lost their jobs. Imports as a share of GDP have tripled. The trade surplus has turned into a huge trade deficit. The manufacturing share of total employment has fallen to 13 percent.1 It appears that President Kennedy's comment is more relevant today than it was 40 years ago.
International trade benefits an economy by lowering prices, encouraging higher productivity, and improving consumer choice. But these gains from trade are "net" gains. On the way to realizing these net gains, an increase in imports usually contributes to plant closings and worker layoffs. The gains from international trade tend to be very large and are widely distributed throughout an economy. By contrast, the costs associated with liberalizing trade tend to be smaller, relative to the benefits, but they are heavily concentrated by industry, location, and worker demographics. The fact that the gains from international trade almost always outweigh the costs does not mean that the costs are any less real. The costs can be very significant for individual workers and their families. In addition, the costs can potentially undermine efforts to further liberalize trade.
In a recent book published by the Institute for International Economics, Lori G. Kletzer finds that about 17 million US workers lost their jobs from manufacturing industries between 1979 and 1999.2 Of those, 6.4 million, or close to 40 percent, were employed in import-competing industries. Over this same period, total US employment increased by approximately 39 million. In other words, between 1979 and 1999, for every job lost in the manufacturing sector, more than 2 jobs were created in the economy.
So What's The Problem?
Some might argue that the ability of the US economy to create jobs suggests that it is not necessary to be overly concerned about workers who lose their jobs. This simplistic argument misses several important points:
- Job creation does not always take place at the same location where jobs are lost. This is particularly relevant with regard to trade-related dislocations, since they tend to be highly concentrated by location. For example, in the 1980s, large layoffs in the steel and auto industries occurred in the "rust belt" states-Pennsylvania, Ohio, Michigan, and West Virginia. In the 1990s, trade-related layoffs occurred in the textile and apparel industries and along the US-Mexican border-North and South Carolina, Texas, New Mexico, and California.
- The new jobs are in different industries than the jobs lost. The vast majority of trade-related job losses are in the manufacturing sector. Almost all the net new jobs created in the United States over the last 20 years have been in the service sector. The skills required by the new jobs are different from those employed in the lost jobs.
- The jobs lost associated with increased import competition and shifts to overseas production, in many cases, pay higher wages and provide better benefits than in the new jobs being created. Unfortunately, the burden of adjustment does not end when the worker gets a new job. Lori Kletzer reports that the average worker who lost his/her job from an import-competing industry experienced a 13 percent decline in average weekly earnings. Thirty-six percent of the workers found jobs, which paid the same or more than their previous jobs. Twenty-five percent of workers laid off from import-competing industries experienced earnings losses of 30 percent or more.3
Workers who lose their jobs associated with increased import competition are faced with the challenge of finding new jobs, acquiring the skills necessary for those jobs, and sometimes uprooting their families in order to take those jobs.
As a way of addressing the adjustment burden, Congress established Trade Adjustment Assistance (TAA) in the Trade Expansion Act of 1962.4 The program provides income maintenance payments and reimbursement for training. Very few workers received assistance during the program's first 12 years, due in large part to rigid eligibility criteria. In 1974, Congress eased the eligibility criteria and the program began to assist thousands of workers in every state throughout the country.5 Since 1975, over 3 million American workers have been certified as eligible for assistance under TAA, and approximately 1.9 million workers have received assistance. (See figure 1)
The most frequently asked question about TAA is, do trade-related dislocated workers deserve to receive any more assistance than workers who lose their jobs for other reasons? At the time TAA was established, trade-related dislocated workers tended to be older, women and minorities with less education than other dislocated workers.6 Increased imports from abroad were forcing workers who were already disadvantaged to adjust to job loss. It was argued that trade-related dislocated workers faced a greater adjustment burden than others and this warranted providing them more assistance.
Kletzer's research, based on more recent data, finds that difficult adjustments are found throughout the manufacturing sector. In addition, workers who lose their jobs from import-competing industries are still slightly older, less educated and more tenured than workers displaced from nonmanufacturing industries. These characteristics are still associated with costly job loss.
Since the establishment of the program, the unions and others have also argued that TAA is a means of compensating workers who lose their jobs due to no fault of their own, but rather associated with a change in government policy. In that regard, TAA is similar to efforts to assist workers adversely affected by various environmental regulations, like the Clean Air Act and legislation to protect the Redwood trees.
With the growth in the importance of trade to the US economy, it has become increasingly difficult to disaggregate the causes of worker dislocation. As a result, from an economic perspective, the argument for a separate program to assist workers whose job loss can be attributed to increased imports has weakened.
From the outset, the primary motivation behind a special program to assist workers who lose their jobs associated with increased imports was based on political considerations. TAA was a part of a package to win AFL-CIO support for the Trade Expansion Act of 1962. It was also believed that TAA would make it easier for members of Congress to support efforts at trade liberalization.
Recently, there has been a breakdown in bipartisan support for further trade liberalization. Congressional support has weakened, as the magnitude of trade-related dislocations has increased. At the same time, the unions have been pursuing other trade policy changes and have widened their concerns beyond domestic dislocations to include labor standards in other countries.
A poll conducted by the Program on International Policy Attitudes suggests that Americans tend to be more willing to support "free trade," if the government assists workers who lose their jobs. Sixty-six percent of respondents agreed with the statement, "I favor free trade, and I believe that it is necessary for the government to have programs to help workers who lose their jobs." Only 18 percent of respondents said that they favor free trade and do not think it is necessary for the government to help workers who lose their jobs. By contrast, 14 percent of respondents did not favor free trade. 7
In contrast to these trade-specific arguments, there are some who believe that existing US programs to assist all dislocated workers are inadequate, and therefore support TAA as a model for a general program that covers all workers. They argue that given the difficulties in attributing causality to dislocation, all workers should be afforded the kind of assistance provide under TAA. This group does not believe that TAA is too generous; on the contrary, they believe that assistance for non-trade related dislocations is inadequate and should be expanded along the lines of TAA.
Assistance Under TAA
Under the existing TAA program, workers can receive up to 52 weeks of income maintenance (beyond the standard 26 weeks of unemployment insurance), training, and job search and relocation assistance. Income maintenance is an entitlement, i.e. Congress must appropriate sufficient funds to provide payments to any worker who is eligible and participates in the program. Training is a "capped entitlement," i.e. the total amount of funds appropriated for training is fixed by Congress, but every eligible participant is entitled to enroll in some government-sponsored training.8 TAA also provides funds for job search and relocation assistance, although not many workers request this assistance.
In 1981, Congress greatly curtailed the provision of assistance under TAA. First, the program was changed to require workers to enroll in training in order to receive income maintenance.9 Second, the amount of income maintenance, which had been set at the average manufacturing wage, was significantly reduced to match each state's unemployment insurance payment. This constituted more than a 50 percent cut in assistance.
The average weekly payment in FY2000 was $217 under TAA and $212 under NAFTA-TAA, less than half of the total average weekly earnings, which was $474.38, and considerably less than the average weekly earnings in manufacturing, which was $598.21. Contrary to what some people believe, workers cannot become rich living on TAA income maintenance payments. In constant terms, the average weekly payment has been rather constant since 1983.
With the approval of the North American Free Trade Agreement (NAFTA) in 1993, Congress established a separate program for workers who lose their jobs from industries facing increased imports from and/or relocates production to Canada and Mexico. The assistance provided workers under NAFTA-TAA is almost identical to that provided under the general TAA program, but there are some differences in coverage. In addition to covering workers who lose their jobs from import-competing industries, NAFTA-TAA provides assistance to workers who lose their jobs from plants that move overseas. Over the last few years, the Department of Labor has been assisting "secondary workers"-workers who lost their jobs because they worked for suppliers or downstream producers for firms that faced increased import competition from Canada or Mexico, on a discretionary basis.10
Since the establishment of the separate NAFTA-TAA program, there has been a lot of confusion and arbitrary discrimination between workers. For example, a worker who loses his/her job from a plant that shifts production to Mexico is eligible for assistance, yet a worker laid-off from a plant that moves to China is not eligible. There are also several technical differences between the two programs.
Evaluating TAA is highly sensitive to the measures used. Those who believe that TAA was designed to reduce opposition toward trade liberalization argue that the program has failed. They point to the increasing difficulties in winning Congressional support for fast-track legislation as evidence that TAA is ineffective. Workers who actually receive assistance under the program do not share this view. For them, TAA may not erase all the economic pain caused by dislocation, but it has made the adjustment to a new job a little easier.11 They also believe that TAA is a small compensation for the losses they and their families have experienced.
The main handicap in evaluating TAA is that the Department of Labor does not collect the necessary data. The Department of Labor has been very slow to address this deficiency, despite consistent calls to do so by the General Accounting Office (GAO). Part of the problem lies in the fact that states were given the responsibility for data collection. The states do not have the resources or interest to monitor the program's effectiveness.
Based on the limited data available, there is evidence that TAA and NAFTA-TAA did meet performance goals established by the Department of Labor in 1999.12 Seventy-five percent of participants for whom the states provided data, found employment. Only 56 percent of them obtained jobs with earnings at least 80 percent of their previous earnings.13 These data also suggest that those participants who were enrolled in training tended to have higher post-employment earnings than those participants who did not.
Unfortunately, the lack of data makes it impossible to perform a sophisticated evaluation of the program. A proper evaluation would include a comparison between those workers who received assistance under TAA and a suitable control group.14 Without this comparison, it is almost impossible to pass judgment on the effectiveness of TAA and NAFTA-TAA.
Pressure on the US manufacturing employment has intensified throughout much of the last 25 years. During that time, approximately 2 million workers have received assistance under TAA. More than half of these workers were laid off from the auto, textiles and apparel, and steel industries. (See Figure 2) In FY 2000, 33,000 workers received assistance under TAA and 2,000 workers received assistance under NAFTA-TAA. The average worker received weekly income maintenance payments, equal to approximately $220, for about 35 weeks. The total budgetary cost of both programs in FY 2000 was about $400 million.
The number of workers receiving assistance pales in comparison to the total number of workers who lose their jobs, and even by comparison with trade-related dislocations.15 In 1999, the last year for which data are available, 3/4 million workers lost their jobs from the manufacturing sector. Of those, approximately ¼ million lost their jobs from industries facing heavy import competition (as defined by Lori Kletzer). Of those, only 30,000 workers, or less than 10 percent, received assistance under TAA.
Efforts to reform and expand TAA have been debated since 1974. Interest in reforming the program is highly correlated with congressional consideration of trade liberalizing legislation. Trade Adjustment Assistance is often seen as a quid pro quo for support on trade liberalizing legislation. Significant weaknesses in the program have depreciated its value in "buying" that support.
Despite periodic debate to expand and improve the program, the only changes to TAA since 1974 were in 1981, when TAA criteria were tightened (income support was made conditional on enrollment in training), and in 1994, when Congress created a separate program related to NAFTA.
Two years ago, congressional authorization for TAA and NAFTA-TAA expired. Senators Roth and Moynihan, then Chairman and Ranking Member of the Finance Committee, and both long-time supporters of TAA, proposed to reauthorize TAA and NAFTA-TAA for 2 years, as opposed to the normal 5-year reauthorization. They also asked the GAO to study the program and make recommendations for improvements. Their intention was to reform the program when the short-term authorization expired.
At the same time, there were a large number of plant closings around the country resulting in massive worker layoffs. In response to a number of plant closings in New Mexico, Senator Bingaman stepped up his efforts to find assistance for the unemployed workers and the financially troubled communities. Senator Bingaman assisted workers and communities to navigate the maze of Federal and state assistance programs. There was an effort to identify shortcomings in existing programs and try some things that were not part of the standard response to a plant closing.16 This effort became one of the foundations for the current effort to reform TAA.
In November 2000, the bipartisan Trade Deficit Review Commission, whose members included Donald Rumsfeld, Robert Zoellick, George Becker, and Carla Hills, issued its final report. The Commission was unable to reach consensus on anything directly related to the trade deficit. The only thing all Commission members were able to agree on was the need to reform and improve assistance to workers who lose their jobs from import-competing industries.
The Trade Adjustment Assistance Reform Act of 2002 (S1209)
The GAO recently completed several studies of TAA and NAFTA-TAA. These studies identified the following problems with the programs:
1. Confusion between TAA and NAFTA-TAA
2. Inadequate training funds
3. Inconsistency between the time for income maintenance and training
4. Lack of effective performance measurements
5. The absence of any assistance to communities with large number of trade-related dislocations.
In June 2001, Senators Bingaman, Baucus, and Daschle introduced legislation to reauthorize and reform TAA. The proposals in their bill (S1209) are based on the GAO findings and Senator Bingaman's on-the-ground experiences in several communities coping with large plant closings and worker layoffs, including in Roswell and Las Cruces, New Mexico, and El Paso, Texas. Most of these proposals have been debated over the last 25 years.17 The proposals include:
Harmonize TAA and NAFTA-TAA: The existence of TAA and NAFTA-TAA has resulted in confusion, duplication, and an increased administrative burden. S1209 combines the two programs into one, using NAFTA-TAA eligibility criteria.
Extend TAA to secondary workers: Under the current programs, in order to be eligible for TAA, a worker has to be laid off from a firm that faces import competition. These are not the only workers who can be adversely affected by increased imports and shifts in production to overseas. For example, a worker laid off from a plant that produces cars may be eligible for TAA, yet a worker laid off from the firm that produced the tires that go into those cars is not eligible. Furthermore, workers laid off from a steering wheel plant which is owned by an auto company facing import competition may be eligible for assistance, but not so if the plant is independent.
In order to reduce these artificial discriminations, Senators Bingaman, Baucus, and Daschle propose to extend TAA to cover all "secondary workers"-defined as workers employed by firms that are suppliers and downstream producers of firms that are certified as eligible for TAA.
GAO estimates that there may be up to 95 secondary workers for every 100 workers currently eligible for TAA. Thus expanding TAA to cover secondary workers will almost double the size of the current program.
Expanding eligibility criteria to include secondary workers is an idea that has been debated since 1974, and has been included in legislation several times over the last 25 years. Proponents argue that only a subset of trade-affected workers are covered. Critics are concerned about the administrative burden of identifying secondary workers and the cost of assisting them.
During the Clinton Administration, the Department of Labor began providing assistance to secondary workers adversely affected by trade with Canada and Mexico. This assistance is not part of NAFTA-TAA. Congress has not appropriated any money for this effort, and instead it has been financed out of the Secretary's discretionary funds. S1209 would formalize the coverage of secondary workers and establish specific procedures for providing them assistance.
Expand eligibility criteria to include shifts in production: One of the major differences between NAFTA-TAA and TAA is that only the former covers workers who lose their jobs associated with a shift in production to Canada and Mexico. It is not clear that this actually results in covering more workers in the end, since the shift in production overseas may lead to an increase in imports. The more immediate problem is timing. By the time imports rise, workers may no longer be able to meet the time requirement under TAA. Accordingly, Congress decided to add shifts in production to the import trigger in NAFTA-TAA.
This additional trigger in NAFTA-TAA has also resulted in discriminating between workers. Workers who lose their jobs because their plant moved to Mexico are eligible for assistance, yet workers who lose their jobs as a result of a plant moving to China do not receive assistance. S1209 proposes to cover all workers who lose their jobs associated with a shift in production, regardless of which country.
Lengthen period of income maintenance to match training: One of the complaints the GAO heard from workers was that TAA and NAFTA-TAA will pay for training for up to 78 weeks, yet the programs provide income maintenance for only 52 weeks. This seems contradictory to one of the motivations for providing income maintenance-namely to support workers while they are enrolled in training.
S1209 extends the period of income maintenance to 78 weeks, to match the period for which training is provided under TAA. This provision does not add a lot to the cost of the program, since very few workers remain in the program that long. In FY 2000, the average duration for assistance was approximately 35 weeks.
Provide assistance for health insurance: Unemployed workers face three difficulties concerning health insurance. First, they have to be eligible for health insurance. Second, they need to afford the coverage. Third, they need the cash flow to enable them to make payments when they are due.
By far, the largest financial burden on a worker who loses his/her job, after the home mortgage, is maintaining health insurance coverage during periods of unemployment. Under COBRA, workers with employer-provided health care insurance can pay to continue their coverage for up to 18 months after layoff.18 Although this addresses the coverage issue, it does not address the cost issue. The average unemployment insurance payment is approximately $250 per week. The average COBRA premium payment for a family of four is approximately $500 a month. In other words, half of a person's unemployment insurance is needed just to maintain health insurance coverage. There seems to be only two ways out of this problem-raise the weekly unemployment insurance payment or somehow lower the cost of maintaining health insurance coverage.
In addition to the severe burden on workers, the lack of health insurance has other consequences for the economy. Families not covered by insurance tend to postpone attention to their health, usually resulting in the need for greater and more expensive care at a later date. They also tend to use hospital emergency rooms as their primary health care provider. This is the most expensive and least efficient form of care and places an additional burden on an already overburdened health delivery system.
Initially, Senators Bingaman, Baucus, and Daschle proposed a 50 percent refundable tax credit for COBRA premium payments. They proposed to provide Medicaid-like coverage to workers not eligible for COBRA. The bill considered by the Senate Finance Committee included a 75 percent subsidy for COBRA premium payments and maintained the Medicaid-like coverage for workers not eligible for COBRA. This issue will likely be settled on the Senate floor or in the House-Senate conference on the bill.
Establish wage insurance: One of the criticisms of the current TAA programs is that it focuses almost exclusively on assisting workers during the period of unemployment and does little to help workers get back to work. In addition, for most workers, the adjustment burden does not end when they find a new job. Lori Kletzer reports that the average trade-dislocated worker experiences a 13 percent drop in earnings. Twenty-five percent of them experience earnings losses of 30 percent or more.
The legislation would provide for reimbursing up to 50 percent of the difference between a worker's new and old wage for two years, for workers over the age of 50 years, earning less than $50,000 annually. The reimbursement would be limited to $10,000 per worker, per year. In addition to reducing the earnings loss experienced by workers, wage insurance can also serve as an incentive for employers to hire older workers and for older workers to remain in the labor force. In order to be eligible to receive wage insurance, a worker has to be reemployed within 6 months after being laid off. Wage insurance would be provided in place of the more costly income maintenance and training.19
Performance measures: In response to repeated criticism by the GAO, the legislation proposes to shift the responsibility for data collection and the development of performance measures to the federal government. The legislation includes a rather detailed list of data to be collected according to specific performance measures.
Community adjustment program: In addressing any job loss, the primary objective should be to get the person back to work, as soon as possible, with the least amount of financial loss. TAA only takes a small step toward helping workers meet that objective. TAA is primarily a training program. It needs to be a reemployment program. The most important ingredient of any reemployment program is the availability of jobs, preferably high-paying jobs.
The proposal for a community adjustment program is based on the growing awareness that the effectiveness of training is limited by the availability of jobs that utilize the skills acquired in that training. Either jobs are available locally or the worker must move. To attract jobs locally, the program borrows from two existing programs-the Department of Defense base closing program and the Community Adjustment and Investment Program (CAIP) established under NAFTA.20
Communities found to be trade-impacted, using the criteria developed in CAIP, would be assigned a technical advisor to help them identify and apply for relevant federal assistance. The communities could also apply for funds to develop a strategic plan to help them respond to the situation. Communities would also be able to apply for federal programs that provide grants, loans, and loan guarantees.
Farmer/fishermen programs: Farmers and fishermen who are adversely affected by import competition cannot take advantage of TAA or NAFTA-TAA since they do not technically "lose their jobs" like other program participants. Furthermore, farmers and fishermen are not covered by federal unemployment insurance, a prerequisite for eligibility on TAA and NAFTA-TAA. This necessitates developing a separate program for farmers and fishermen. Under the proposal, if the price of an import-competing commodity falls by more than 20 percent below the previous 5-year average price, then the program will provide the farmer or fisherman 50 percent of the difference. Assistance is limited to $10,000 per year.21
The total cost of TAA and NAFTA-TAA was approximately $400 million in FY 2001.22 The Congressional Budget Office (CBO) estimates that the proposals included in S1209 will add $8.6 billion over 10 years to the cost of the program. On average, the annual cost of the program would increase by approximately $800 million to $1.2 billion.
CBO estimates that expanding the eligibility criteria to include secondary workers and shifts in production overseas would cost $300 million per year and the wage insurance program would cost $85 million per year. The community adjustment program is estimated to add $5 million to the annual cost of TAA. The farmer and fishermen programs are capped in the legislation at $100 million per year. CBO estimated that providing workers a 75 percent subsidy for their COBRA premium payments and enrolling other workers in a Medicaid-like program would cost $260 million per year.
Congressional Activity on TAA reform
On December 4, 2001, the Senate Finance Committee approved S1209 on a voice vote. Since then, Senators Baucus and Daschle have both publicly stated that Senate passage of legislation to provide trade negotiating authority to the President depends on expanding and reforming TAA along the lines of S1209. Senator Baucus has stated that the two bills were "inextricably" linked.23
Immediately prior to voting on the fast-track bill in December 2001, the House of Representatives almost unanimously approved a straight 2-year extension of TAA with no major changes. On February 4, Congressman Bentsen and Congresswoman Eshoo introduced HR 3670 in the House of Representatives, which is identical to S1209.
It is estimated that international trade increases the income of the average American household by somewhere between $1,500 and $2,000.24 It is not too much to ask each household to contribute $12 a year to provide assistance to those workers adversely affected by that trade.
The United States engagement in the global economy is likely to continue and intensify over the coming years. The nation's social and economic infrastructure needs to be updated in order to keep pace with this trend, so that no one is left behind, and so one group does not bear an inequitable share of the burden associated with this trend. Reforming TAA is a start in that direction. (See table 1 for a comparison of TAA programs)
5. The eligibility criteria were liberalized, so that imports had to "contribute importantly" to job loss. In other words, the increase in imports had to only be one of several contributing factors to the job loss.
7. Eighty-seven percent (56 percent strongly) of respondents agreed with the statement, "I would favor more free trade, if I were confident that we were making major efforts to educate and retrain Americans to be competitive in the global economy. Only 11 percent disagreed. See Program on International Policy Attitudes, October 1999.
8. By contrast, the Workforce Investment Act (WIA)-the program, which provides assistance to all dislocated workers regardless of cause-is not an entitlement. Workers only receive training if there are adequate funds available. Most states exhaust training funds under WIA well before the end of the year, denying workers the ability to enroll in training. In addition, states can deny training, if it is determined that a worker can find a job that pays a subsistence wage.
10. A downstream producer is defined in the bill as "a firm that performs additional, value-added production processes, including a firm that performs final assembly, finishing, or packaging of articles produced by another firm."
15. This is due in part to the fact that workers find new jobs soon after their lay off. Other workers do not want to enroll in training, thereby making them ineligible for TAA assistance. A third group of workers are concerned about the stigma associated with accepting government assistance.
16. For a complete description of this effort, see Rosen, Howard, "A New Approach to Assist Trade-Affected Workers and their Communities: The Roswell Experiment," Journal of Law and Border Studies, Volume 1 Number 1, 2001.
19. A similar program for all dislocated workers is discussed in Lori G. Kletzer and Robert E. Litan, "Prescription to Relieve Worker Anxiety," Institute for International Economics Policy Brief PB01-2, Washington, DC, Institute for International Economics, 2001.
20. Under the base-closing program, the Department of Defense provides communities adversely affected by the closing of a military base with technical assistance and funds to prepare strategic plans. CAIP provides loans, loan guarantees and grants to communities adversely affected by trade with Canada and Mexico. See General Accounting Office, Trade Adjustment Assistance: Opportunities to Improve the Community Adjustment and Investment Program, September 2000.
24. Brown, Deardorff and Stern estimate that a one-third cut in the post Uruguay Round tariff and non-tariff protection would increase US GDP by $177 billion. There are approximately 100 million American households. CBO estimates that the revised TAA program would cost approximately $1.2 billion per year.