Congressional Testimony on the Effectiveness of Trade Adjustment Assistance and Suggestions for Improving the Program

In this testimony, I will sketch key findings from research that I conducted with others, bearing on the justification and effectiveness of trade adjustment assistance. Most economists agree that protecting workers from the negative consequences of the North American Free Trade Agreement is appropriate, since we believe that a measure is unambiguously good only when some benefit, but no one is hurt. Even if trade liberalization is only a minor source of job loss, when many workers are hurt by imports, trade often is restricted. Thus, an effective assistance program can counter a drift toward protectionism that would be far more costly than compensating job losers. For example, in the early 1970's, quotas cost steel users $11 billion, while removing those quotas would have cost workers, at most, $1.5 billion. A key problem is that gains per person are small, while the losses per person are large. Even worse, many workers not adversely affected feel threatened. Thus, losers have strong incentives to restrict trade, compared to incentives gainers have to make trade freer. This is a situation where providing all threatened workers with insurance would increase the general welfare at a modest price. To reduce workers' fears, the TAA program must effectively compensate losers. Because the cost of job loss is high, benefits per person need to be substantial. Four years after displacement, 60 percent of high-tenure job losers typically have not found similar work and have earnings reduced by 50 percent. Even job losers who find similar work suffer reductions in earnings from 12 to 22 percent. Job loss is common even in prosperous times, but paradoxically, even in the worst of times, roughly two-thirds of laid-off workers are recalled. In short, most workers adversely affected by trade liberalization need protection for less than 2 years, while a minority face permanent earnings reductions. TAA extends UI payments by 26 weeks. Even in the early 1980's, UI plus TAA virtually eliminated the incidence of large losses among trade impacted workers for the first two years following jobs loss given the workers were covered by TAA. That is the good news. The bad news is that many workers with large losses did not return to similar jobs within two years, and therefore, were not adequately protected even if covered by TAA. TAA now provides training vouchers often worth $12,000 or more. Unfortunately, training programs targeted on dislocated workers have been shown to have very limited success. Even under highly optimistic assumptions, an investment in training of over $100,000 would be needed to offset the costs of job loss. The GI bill is a precedent for providing assistance of that magnitude. In 1967, the GI bill provided inflation-adjusted benefits of $80,000. High-tenure dislocated workers are unlikely to benefit from a similar program, however. They are older than vets, already have marketable skills, and sometimes are lacking basic education. The bottom line is that training is not a panacea for dislocated workers. In fact, placement services, other job search assistance, and counseling have been shown to be as effective as training, just much less expensive. My research strongly supports that view. The huge earnings difference between finding and not finding similar work suggests helping job losers to locate similar jobs could be equivalent to providing a $100,000 annuity. Equally important, plenty of comparable jobs are available. In Pittsburgh, between 1978 and 1982, one-quarter of all manufacturing jobs were filled by new hires. Even though net employment fell by 15 percent, quits and retirement opened enough job slots, so every displaced worker could have remained in that sector. In addition, the employment service has been shown to help dislocated UI exhaustees return to work 9 weeks sooner. Initially pay is below predisplacement levels, but pay grows substantially. Yet, without outside help, dislocated workers find accepting pay cuts difficult, particularly while receiving UI. Thus, my central policy conclusion is that the combination of UI and TAA does a good job in protecting the majority of trade impacted workers. But available funds would be better spent by helping dislocated workers find jobs, particularly by adequately funding the employment service. Improving the coordination between EDWAA and the ES also would help, as would improving the monitoring of UI claimants' job search. A more innovative approach, with a good chance of success, is offering wage subsidies or reemployment bonuses. Such measures should encourage more rapid return to work and at least partially offset wage reductions. Finally, we should remember that high-tenure workers need protection against sources of job loss other than government actions. Workers should be willing to pay for that protection, and many unionized workers have done just that. The government can play a key role, by encouraging labor and management to adopt measures to deal with job loss. As many workers learned in the early 1980's, by the time a threat is apparent, it is too late to solve the problem. Improving TAA is difficult, but fairness suggests the threat to workers livelihoods inherent in the NAFTA should be reduced, and common sense suggests reducing the threat is the key to maintaining free trade. I hope this testimony will contribute to making a good program better. I am very grateful for this opportunity to discuss trade adjustment assistance with you.