Skip to page content
Office of the Secretary
Bookmark and Share

OSEC Congressional Testimony

Statement of Robert B. Reich Secretary of Labor before the Committee on Finance, United States Senate [05/26/94]

Chairman Moynihan, Senator Packwood, and distinguished Members of the Committee:

I am pleased to have the opportunity to discuss with this Committee the role of the Reemployment Act of 1994 (S. 1951) in equipping all Americans to prosper in today's challenging new economy, and thus cementing a broad-based coalition in favor of open markets and continuous economic adaptation.

Briefly put, the Reemployment Act represents the key component in an ongoing effort to transform America's unemployment insurance system into a reemployment system. This is an urgent enterprise, because the current system is simply not working as it should for America's workers, businesses, or taxpayers.

The programs that make up our existing unemployment system were designed in an earlier time to meet the needs of a simpler economy. Their main purpose was to cushion the employment impact of cyclical and seasonal downturns. The foundations of the current system were laid nearly sixty years ago by the Committee on Economic Security established by Franklin D. Roosevelt and led by my legendary predecessor, Frances Perkins. As Chairman Moynihan observed upon the introduction of S. 1951, Secretary Perkins and her colleagues saw their institutional design as experimental. Their report, and the then-revolutionary initiatives it inspired, were expected to "secure the much-needed experience necessary for the development of a more nearly perfect system."

The unemployment insurance system launched during the New Deal was a remarkably durable success for so sweeping an experiment. But the economy has changed, and the system has not. It is time for us to return to Frances Perkins' experimental ethic, to learn from the evidence that six decades of experience offer us, and to take up anew the challenge of developing a more nearly perfect system.

While cyclical and seasonal unemployment still exists, the problem of structural unemployment has grown in importance as technological progress, corporate restructuring, the integration of the world economy, and defense downsizing have accelerated the pace of fundamental economic change. A smaller fraction of lost jobs can be expected to return with the next upturn in the business cycle. A growing number of unemployed workers need to equip themselves for new jobs, often in new industries. The 1935 report by Frances Perkins' Committee on Economic Security noted that "normally the insured worker will return to his old job or find other work before his right to benefits is exhausted." Last year, however, fewer than one-fourth of unemployed job losers expected to be recalled to their old jobs. The broad statistical trends are mirrored in the anxieties of individual Americans. The Family and Workplace Institute found that 42 percent of surveyed workers reported recent job cuts at their workplaces, and 17 percent felt it was likely or very likely that they themselves would permanently lose their jobs within a year.

Because our current system is not geared to assisting this transition to new work, we face a serious problem of long-term unemployment. Even in this increasingly solid jobs expansion, almost 1.8 million Americans have been jobless for more than 26 weeks. In fact, the share of long-term unemployed as a percent of total joblessness has been rising over the last 25 years. During the 1970's, about 11 percent of the jobless were long-term unemployed. So far in the 1990's, the average has been about 16 percent. The most recent figure, for April of this year, is over 21 percent of the total unemployment accounted for by people who have been without work for more than 26 weeks.

As long-term unemployment rises, so does the number of unemployment insurance recipients who run through their full entitlement to benefits without returning to work. In 1993, 39 percent of those who collected regular unemployment benefits ended up exhausting their eligibility. Over the past quarter-century only in 1983, when the unemployment rate averaged 9.6 percent, did a higher fraction of UI recipients fail to find work before their eligibility for benefits expired.

We would face a different problem, and in some ways a simpler one, if long-term unemployment simply reflected a lack of jobs. Yet while workers are suffering high rates of unemployment, underemployment, and long-term joblessness, some employers are having difficulty finding the workers they need. This mismatch between labor supply and demand is a tragic waste for both workers and employers. But it also complicates macroeconomic policy. When workers are skilled and flexible, labor shortages in a particular industry can be quickly remedied. But when unemployed workers are unable to fill new jobs as the economy creates them, labor shortages can persist and kindle inflationary pressures, despite high levels of unemployment.

Economists are fond of debating the "natural" rate of unemployment--the level of joblessness that is required to keep inflation in check. This is not the time or place for engaging that debate, but I would like to point out that whatever that rate might be, it is not an immutable constant, but can be lowered. If we can arrange to keep inflation under control while suffering less joblessness, we can spare America both the lost production and the social damage that needless unemployment entails. And among the most promising ways to lower the natural rate of unemployment is to ease the transition of American workers from job to job. Workforce flexibility can help alleviate the painful paradox of simultaneous skill shortages and joblessness.

We already know a great deal about how to boost workforce flexibility. In preparation for drafting the Reemployment Act proposal, the Labor Department conducted an exhaustive review of the evidence. We were determined to learn what works, and what doesn't work, for moving people quickly into new jobs. Wherever possible we have relied upon the kinds of experimental studies, with random assignment between participant groups and control groups, that can generate conclusive results.

While high-quality experimental answers do not exist for every relevant question, on a large number of the key reemployment issues we have been able to assemble quite robust findings. We have extracted from the evidence several core lessons. And we have attempted to parlay these lessons, as quickly and directly as possible, into program reforms--in some cases through incremental changes that have already been accomplished, but more often through the new authority the Reemployment Act would provide.

  1. The first lesson concerns the importance of early identification and action for dislocated workers. This makes it possible to reduce unemployment through simple, low-cost services. Some of the best data come from well-designed experiments in five states--New Jersey, Nevada, Minnesota, South Carolina, and Washington. When workers first applied for unemployment insurance, state agencies used labor market information, work history and educational attainment information to determine which of the newly jobless were at risk of long-term unemployment. These at-risk workers then received counseling and job-search assistance.

    A scientific assessment of these experiments found that at-risk workers who got this kind of basic reemployment assistance found new jobs earlier than they otherwise would have--averaging between one-half and four weeks earlier--with no loss in the quality of the employment they found. Quicker reemployment meant lower unemployment insurance costs; on average, government saved two dollars for every dollar invested in targeted job search assistance. And workers, of course, were better off with a shorter period without work or wages.1

    This evidence was sufficiently convincing to warrant our proposal and Congress's enactment of a special provision attached to one of the extensions of Emergency Unemployment Compensation last year. Under this provision, unemployment insurance offices will take steps to identify workers who are unlikely to get their old jobs back, and then link those workers to all available reemployment services.The Reemployment Act will deepen this reform, increasing the proportion of workers who benefit from prompt reemployment services.

  2. A second lesson shows the potential payoff to letting unemployment insurance recipients start their own businesses. Labor Department demonstration programs in Washington State and Massachusetts gave unemployment insurance claimants a self-employment option. Jobless workers interested in starting their own businesses were given training and support through the UI system. A systematic evaluation found that participants were about twice as likely as comparable non-participants to successfully launch a business. They also had higher average subsequent employment rates and earnings than the control group.2

    The solid results from state demonstrations of self-employment assistance inspired a provision in the legislation implementing the North American Free Trade Agreement. Under this provision, all states have the option of creating self-employment programs of the sort that have been shown to pay off in Washington and Massachusetts.This authority is only temporary under current law; the Reemployment Act would make it a permanent option.

  3. A third lesson concerns reemployment bonuses paid to workers who find jobs quickly. Random-assignment experiments in Illinois, Pennsylvania, and Washington showed that offering such bonuses cut the average length of unemployment among eligible populations by one-half to one week--even though only ten to fifteen percent of eligible clients actually made use of the bonuses. The programs paid for themselves from the government's perspective; savings in benefit payments offset the cost of the bonuses. And since they meant higher earnings for participants, they generated net benefits overall.3 The evidence is robust enough to convince us to propose, in the Reemployment Act, a new reemployment bonus option for all States.

  4. Beyond reemployment bonuses and self-employment options, a proven provision to boost the flexibility of the unemployment insurance system concerns short-time compensation. This option allows states to offer partial unemployment benefits to employees who are working reduced hours, because their employer is seeking to reduce costs to avert a shutdown, or is spreading work reductions throughout the labor force in an attempt to avoid layoffs. About one-third of the States now operate short-time compensation programs. The Reemployment Act would encourage more States to offer this option.

  5. A central lesson concerns the importance of integrating skill training with unemployment insurance. While only a minority of job losers require new skills for reemployment, for those who do there are great advantages to integrating training and unemployment benefits. At present, too many workers who need and can benefit from training have their first exposure to training programs only after using up their unemployment insurance. (A 1988 study found that only one percent of the long-term unemployed had attended training programs by the time their eligibility for jobless benefits expired.4)

    Yet there is a good deal of solid evidence on the payoff to longer-term post-secondary education, including training at community colleges.5 While there are no good data on dislocated workers specifically, the findings on post-secondary education generally show a clear pattern of less joblessness and higher earnings as skill levels increase. Importantly, the payoff from education seems to be largely due to the new skills acquired, rather than the "credentialing" effect of graduation: Workers with college experience earn five to ten percent more than do comparable high-school graduates per year of courses completed, whether at four-year or community colleges, and whether or not college education leads to a degree.6

    By contrast, the evidence suggests that many forms of short-term training for dislocated workers are not effective. In three studies--two of them featuring randomized control-group methods--workers offered short-term training plus job search assistance showed no significant increase in earnings or employment over workers offered job-search assistance alone.7 While the Reemployment Act encourages rapid reemployment through improved job search assistance, it also incorporates three provisions to make longer-term training a practical option for workers who need it: First is the emphasis on early identification and intervention, to get dislocated workers into training before their regular unemployment benefits are exhausted. Second is the Act's provision for extended unemployment benefits for workers who need it to complete appropriate retraining programs. Third is access to student financial assistance, including income-contingent loans, to fund especially lengthy or expensive training regimes, or for workers who are ineligible for income support delivered through the unemployment insurance system.

  6. A sixth lesson is the importance of integrated, accountable delivery of reemployment services. Here, it requires no sophisticated analysis to pinpoint the problem: The current system is confusing and complex, balkanized into a profusion of narrow programs. We clearly need to streamline the delivery of employment and training services. The Reemployment Act, which will combine all six programs for dislocated workers into a single system, represents an aggressive step toward program consolidation. And this is only our first step. The Reemployment Act is crafted to encourage additional waves of consolidation, and set in motion a continuous campaign of streamlining and improvements.

Program consolidation is clearly important for efficiency and accountability. An equally central tactic for accomplishing the same key goals, however, is integrated one-stop delivery at the point where the customer encounters the system. One of the cornerstones of the Reemployment Act is the progressive construction of a network of one-stop career centers that will unite under one roof access to a wide range of workforce services--not just for dislocated workers, and not just programs based at the Labor Department. Fortunately, there is a wealth of pioneering models on which to build. In New York, for example, the GATEWAY program organizes access to a rich array of reemployment services, including job data banks, skills assessment, counseling, and training. GATEWAY uses up-to-date information technology to minimize red tape and maximize customer service, and has paid off for hundreds of workers--including John Hahn, who spoke at the White House ceremony launching the Reemployment Act proposal. In Maine, the Career Advancement Center links dislocated workers with a wide range of funding sources and services to accelerate their transition to good new jobs. Like GATEWAY, it is founded on the three pillars of comprehensive services, modern information technology, and a culture of commitment to the customer.

In Brookings, South Dakota, the Career Learning Center offers an integrated, customer-oriented approach to delivering information, job-search assistance, and training. In Blaine, Minnesota the Anoka County Human Service Center brings the services of 23 public agencies and numerous private programs together under one roof, giving customers one-stop access to the whole range of help they need to improve their labor-market prospects. Oregon has a statewide program called Choices and Options, which helps dislocated workers assess their options and make informed decisions before their eligibility for jobless benefits expires. And in Michigan, the famous Focus:HOPE program delivers high-quality training, carefully connected to jobs in demand in the local labor market.

I can't expect to even mention all of the promising approaches under way out in the states: A St. Louis program for accelerating the transition of McDonnell Douglas workers displaced by defense downsizing; a worker transition program, tightly linked to economic development efforts, in Davenport, Iowa; the new Resource Room initiative launched by Oklahoma's employment service; the list could go on and on. The Reemployment Act takes the best ideas developed in these "laboratories of democracy"--one-stop service delivery, market-driven training, customer orientation, computer-driven labor market information systems, and so on--and builds them into the national reemployment system.

The incremental reforms to the unemployment insurance system--reemployment bonuses, short-time compensation, self- employment options, universal profiling--and the local innovations occurring throughout America will make a modest but real difference for dislocated workers, and for public budgets. But modest changes are not enough. If we are to do right by American workers faced with a fundamentally different economy--if we are to honor through action the pragmatic, experimental tradition of my predecessor Frances Perkins and your predecessors in the Senate who shaped America's original unemployment insurance system--we must commit to a fundamental transformation in policy and programs.

The mismatch between the current system's structure and the needs it is pressed to serve burdens federal and state taxpayers, and especially businesses, with enormous costs. State unemployment compensation payments have exceeded $13 billion in each of the last seven years, and in 1993--even as recovery set in--states paid out nearly $22 billion in regular unemployment benefits. Federal spending on administrative costs for regular unemployment insurance totalled $2.5 billion last year. But unemployment insurance is not designed to help speed workers into reemployment. So despite these huge outlays, the predicament of the long-term unemployed led to the repeated provision of federal Emergency Unemployment Compensation payments, costing $24 billion over the past two years.

Consider this: in 1993 alone, combined federal and state unemployment compensation costs exceeded $38 billion. This money was not wasted, to be sure; it spared jobless workers from the worst forms of financial hardship and helped shore up consumer spending, which are important accomplishments. But because the system is not geared to reemployment, we have only limited long-term results to show for the $38 billion spent last year--few investments in new skills, little growth of flexibility, no coordinated attack against the problem of structural unemployment.

We can do better. In the current budgetary environment, we can and must deploy the resources of the unemployment insurance system to greater effect. By gearing jobless benefits to rapid reemployment, and by integrating cash benefits and skill training programs, we can respond more effectively to the risks and opportunities of today's economy.

The Reemployment Act is designed with a keen awareness of fiscal limits. Most of the financing is discretionary, and the majority is financed through consolidating separate dislocated worker training programs into this integrated system. Additional discretionary funds come from reductions in other Federal programs. The modest mandatory component, limited to retraining income support, is initially financed in part by offsets from consolidation, and in part by applying to a jobs-related purpose Unemployment Insurance revenues now routed elsewhere. One of the programs to be consolidated into the Reemployment Act system is Trade Adjustment Assistance. We have attempted to ensure a smooth transition from this categorical adjustment program to the Reemployment Act's comprehensive approach. Virtually all trade-displaced workers will receive equal or superior benefits under the new system. One of the main distinctions is that extended unemployment benefits will be targeted to workers who need that support to complete longer-term training programs.

Beginning in FY 1999, income support for all workers who need long-term training will be financed by a permanent extension of the 0.2 percent surcharge first levied in 1977 under the Federal Unemployment Tax Act. This FUTA extension is a small but crucial component of the systematic reform envisaged by the Reemployment Act. Since some dislocated workers will inevitably need long-term training to equip themselves for reemployment, a reliable source of funding to make this training possible is essential to the integrity of the reemployment system. Income support is available only to the minority of workers who participate in the training for reemployment. By integrating training with the unemployment insurance system, this provision reinforces the system-wide emphasis on preparation for new jobs that is central to the overall reform effort.

The mandatory component is also carefully crafted to meet the imperative of fiscal discipline. It is tightly capped. Spending for income support cannot exceed 20 percent of FUTA receipts. While some workers will need training that goes beyond their term of eligibility for income support, they will not be left on their own. Income-contingent loans and other forms of financial aid delivered under the Higher Education Act of 1965 are intended to extend, supplement, or in some cases substitute for income support.

Since the income-support financing is linked to FUTA receipts, and since the wage base against which that charge is levied is limited to the first $7,000 paid to a worker, the maximum annual cost of retraining incoming support is about $11 per employee. I believe this is a small price for business to pay for a reform that will boost the supply of skilled, flexible labor.

We expect that the Reemployment Act will also pay off in narrow budgetary terms. The decreases in unemployment and increases in earnings we anticipate from the Reemployment Act's implementation will boost tax revenues and shrink governmental payments for unemployment insurance and other forms of public support.

Finally, American business and the economy at large can expect an even more fundamental payoff from the Reemployment Act. Putting in place a sturdy system of reemployment services will give all American workers greater cause for confidence that they will benefit from the economic changes that will continue to transform American industry. By broadening the coalition in favor of change, the Reemployment Act will reduce the risk of the kind of backlash against change that would imperil our common agenda of open markets, technological dynamism, and structural transformation.

The Reemployment Act of 1994 is informed by systematic attention to empirical evidence, and a deep commitment to what works. Through respect for the evidence, and through persistence in pursuit of the American tradition of broadly-shared middle class prosperity, we can prepare every American to succeed in the skill-based economy taking shape all around us today. There is no excuse for leaving a single citizen behind.

This concludes my prepared remarks. I would be glad to answer any questions.

1Bruce Meyer, Policy Lessons from the U.S. Unemployment Insurance Experiments, National Bureau of Economic Research Working Paper #4197, 1992

2Jacob Benus et al., A Comparative Analysis of the Washington and Massachusetts UI Self-Employment Demonstrations, Abt Associates, Bethesda, Md., November 1993

3Paul Decker and Christopher O'Leary, An Analysis of Pooled Evidence from the Pennsylvania and Washington Reemployment Bonus Demonstrations, Unemployment Insurance Occasional Paper 92-7, U.S. Department of Labor, 1992; Bruce Meyer, Policy Lessons From the U.S. Unemployment Insurance Experiments, National Bureau of Economic Research Working Paper #4197, 1992

4Philip Richardson et al., Referral of Long-Term Unemployment Insurance Claimants to Reemployment Services, U.S. Department of Labor Occasional Paper 89-2, 1989

5A good overview of the growing role of education levels in income differentials is Frank Levy and Richard Murname, "U.S. Earnings Levels and Earnings Inequality: A Review of Recent Trends and Proposed Explanations," Journal of Economic Literature, September 1992

6Thomas J. Kane and Cecilia Rouse, Labor Market Returns to Two and Four-Year College: Is a Credit a Credit and Do Degrees Matter? Working Paper #311, Industrial Relations Section, Princeton University, January, 1993

7The studies are summarized in Duane Leigh, "An Overview of Existing Evaluation Evidence for the U.S.," in Assisting Workers Displaced By Structural Change: An International Comparison, Upjohn Institute, Forthcoming 1994

History Home Page

In-Depth Research

Annals of the Department

History eSources

Departmental Timeline

Historical Office

 Century of Service  

Wirtz Labor Library