Job Loss: Causes, Consequences, and Policy Responses*

Thank you, especially to Rich Hobbie, for inviting me to address this meeting of the National Association of State Workforce Agencies. Last fall, we held a conference that we called "Job Loss: Causes, Consequences, and Policy Responses" at the Federal Reserve Bank of Chicago. Rich helped us with that conference by asking provoking questions in his role as discussant, and through that interaction, we learned that we have a common interest in the issue of job loss.

First, let me tell you about why the topic of job loss is of particular interest to me. Then I will tell you about why we thought this was an important topic on which to hold a conference, and some of the ideas that were salient at the conference.

As you know, in addition to my current job as president of the Federal Reserve Bank of Chicago, I have also served as Under Secretary of Labor at the U.S. Department of Labor and Deputy U.S. Trade Representative. So, I have been involved with the causes, the consequences, and the policy responses to job loss dating back over 30 years. To me, the fact that I have been involved with these topics for so long highlights their persistence.

In the United States, we all enjoy low prices, higher productivity, and a higher overall standard of living thanks to our openness to competition and changes in technology. However, some people pay significant costs as the economy adjusts to these changes.

Perhaps the largest costs are paid by the workers who have their careers upended. According to the most recent Displaced Workers Survey, from 2001 to 2003 more than 5 million workers lost a job that they had held for 3 years or longer. Most of these workers lost their jobs through no fault of their own. For many, the arrival of a new technology simply rendered their job unnecessary.

Even at the Federal Reserve, we aren't immune from the effects of technological improvements. Twenty years ago, the Chicago Fed had nearly 3,000 employees; today, we have almost 1,800. Much of the recent reduction was in staff working to process paper checks, and it has been the direct result of new payments technologies. As a society, we all gain from the convenience of electronic payment methods, and we all benefit from the quick and efficient processing of payments. However, the workers who were doing a good job processing checks have to go find something else to do. This is a difficult transition for the Federal Reserve, and I know it is very painful for those most directly affected.

But the Chicago Fed experience is just a small example of how technology has affected many parts of the economy. Over the past decade, we've seen enhanced competition and large improvements in technology lead to much faster growth in productivity. From the medium- and long-term perspective, this has been unequivocally good news for the economy overall. When productivity grows at a faster rate, the economy can grow faster, resulting in higher incomes and producing more goods and services for all of us to enjoy, without generating inflationary pressures. This ultimately makes our job at the Federal Reserve easier, because our mandate is to set monetary policy to support maximum sustainable economic growth and price stability. But, in the short run, there are some complex judgments to make.

Recently, labor market conditions have improved: The unemployment rate has fallen and the economy has added 2 million jobs over the past year. However, rates of job displacement are relatively high, as high as in some periods when the unemployment rate was significantly higher than it is now. This suggests that the pace of change in the economy has increased the risks that workers' skills will become obsolete, and they will lose their long-held jobs. It also means that a larger-than-normal fraction of the unemployed are the kind of displaced workers who often face difficult challenges in finding new employment.

This increased rate at which experienced workers have been losing long-held jobs is one of the factors that monetary policy makers have to think about in judging the extent of inflationary pressures currently facing the economy. This judgment is complicated because an increase in job displacement could have two opposing effects on the labor market. On the one hand, with an increase in the fraction of unemployed workers likely facing difficult job transitions, there could be somewhat less slack in the economy than the current unemployment rate would ordinarily suggest. With a higher fraction of the unemployed lacking the needed skills to fill available jobs, there could be more shortages of certain kinds of workers, leading to upward pressure on labor costs. On the other hand, an environment in which job displacement is more common may make workers reluctant to press for large wage increases, which would tend to restrain labor cost pressures. What all of this means for monetary policy is that we need to keep an eye out for changes in the relationships between labor costs and our usual measures of resource slack in labor markets. So far at least, wage pressures have not been higher than one would expect on the basis of the usual measures of labor market slackness.

Outside of the realm of monetary policy, there are many other important policy issues directly related to the experience of job loss. If we want a dynamic economy, where technology is changing and markets are increasingly open to competition, some existing jobs will lose their value and will disappear. This clearly causes disruptions to many lives and deserves careful attention. We need to pay special attention to the workers affected by such change, because this dynamic process helps generate ever higher standards of living. We need policies to address their job loss, and we need to regularly rethink our policies as different industries and different types of workers are affected by change.

But, how should we compensate the workers who lose their jobs due to the very changes we seek to encourage? This is a difficult task, one without easy answers. That is why I was very supportive when our research department proposed that we hold a conference on job loss. The conference brought together researchers, policymakers, and practitioners to discuss a range of topics related to job displacement.

One topic addressed by the conference was the changing characteristics of workers who are displaced. A more diverse group of workers is experiencing job displacement. There have been striking increases in the fraction of displaced workers who are female, well-educated, white collar, and who were displaced from the service sector. This has happened in part because technological change and foreign competition are affecting a broader group of workers.

Another topic discussed was the costs of job loss for displaced workers. Displaced workers often take a longer time to find a new job, which translates into larger chunks of lost income. Moreover, the recent Displaced Workers Survey showed that, on average, displaced workers earn about 17 percent less than they earned in the job they lost. Other research has shown that this gap in earnings tends to persist for up to 10 years. Many displaced workers have skills that are highly specific to their employer or industry, and the value of those skills often disappears along with their jobs. The most recent Displaced Workers Survey has another striking bit of information: From 2001 to 2003, job losers with education beyond high school suffered greater earnings losses on their new jobs than in any previous period for which we have data.

Next let me highlight some of the policy ideas that emerged from the conference. I will also try to point out areas where more research might be useful, and how interaction between policymakers and practitioners can facilitate that research.

Our unemployment insurance system is the main policy program to help workers who have lost their jobs. However, the system is better structured to assist those for whom unemployment is expected to be of short duration than those suffering longer spells, such as displaced workers. How might the unemployment insurance system be designed to better address the needs of these workers? First, it would probably need to last longer than 26 weeks. But as you know, that may create an incentive problem. If unemployment insurance benefits are high relative to the new, lower wages that these workers face, they have an incentive to remain unemployed for as long as the program allows. So, we may need to create inducements to get these workers back into the labor market. For example, the government might allow workers to keep some unemployment benefits if they return to work early, or provide a "wage subsidy" that ensures the worker is better off taking a job than collecting unemployment benefits. Research suggests that such programs can shorten the length of time workers search for new jobs.

Second, an unemployment program aimed at displaced workers would probably need to help workers acquire new skills. We have implemented and experimented with a wide range of retraining programs for as long as I can remember. Unfortunately, many of these programs have not been beneficial to the trainees when you conduct a cost/benefit analysis. However, we have had better news recently when assessing the retraining conducted by community college systems. Of course, not all courses are equally valuable. For example, training to be a nurse is likely to be much more valuable than learning about the history of photography. So, there also may be a role for programs to help inform displaced workers about which fields would constitute their best bets.

The Bush Administration has proposed "re-employment accounts." The proposal contains elements of the employment bonus and re-training policies I've just discussed. Unemployed workers would have a sum of money that they could use to get training, and they could keep the balance of the money in the account if they took a job within some specified period of time. While this proposal is still under consideration, seven states have been chosen as the sites for a $7.9 million demonstration project.

Trade Adjustment Assistance for workers displaced by trade is another example of a policy that is more tailored to the needs of those who have permanently lost a job. Benefits are paid for longer than the typical 26-week period, and workers may receive assistance with retraining and purchasing health insurance. Older workers who are less likely to benefit from retraining may be eligible for wage subsidies. However, as was pointed out repeatedly at our conference, Trade Adjustment Assistance is really a component of trade policy, and not a comprehensive labor market policy. Assistance is only available for workers who are certified to have lost their jobs through trade and only available for those displaced from manufacturing. But from the point of view of a worker, it does not matter if his or her job disappeared because of competition from New Delhi or New Jersey. We can learn from the Trade Adjustment Assistance about what programs can effectively help displaced workers. But we need to apply those lessons to help all workers displaced—no matter what caused their job loss and regardless of the sector where they had been employed.

Our conference also considered the policy issues from the firm's perspective. For example, does the current unemployment insurance system affect how firms decide to lay off workers? Firms pay an unemployment insurance (UI) tax rate that increases with the number of workers it has laid off in the past. However, the tax rate is capped: Once a firm has reached the maximum rate, it can lay off additional workers without raising its UI tax rate. Research suggests that temporary layoffs would be reduced by moving toward full experience rating of the unemployment insurance system, so that each additional laid-off worker raises the tax rate.

Several panelists at the conference were intimately involved in the inner workings of firms, and they painted an interesting picture of how firms decide to reduce their workforce. One panelist pointed out a seeming paradox: At the same time that firms are struggling to lay off workers in the face of new technologies and competition, they are also struggling to retain and hire skilled workers. He pointed out that there may be unintended consequences associated with firms' ability to so easily lay off workers, namely that it creates an environment where it is also hard to retain workers.

This is because the surviving workforce takes a great deal of notice of how laid-off employees are treated. If a firm gives longer notice of impending lay-offs or offers outplacement services, then it may have an easier time retaining other workers. But if the firm treats laid-off employees poorly, then other workers are more likely to leave as soon as they get an attractive offer.

These provocative ideas lead to several interesting research questions. For example, do workers who use outplacement services do better than those who do not? It's difficult to answer this question since the firms that provide outplacement services are likely to be different from those that do not. Similarly, the workers who choose to use these services are likely to be different from those who do not. But, by using sophisticated research methods, we may able to disentangle the causal effects of outplacement services. And, if private outplacement services affect the speed of finding a new job and the quality of subsequent jobs, these types of services could be made available on a wider scale.

As is usually the case, many of the program and policy recommendations at our conference would require additional resources. If policymakers are going to ask the public for more funds, they should subject their policy initiatives to objective evaluation. One of the best things I did at the Labor Department in the early 1970s was to establish a team of well-trained economists to examine the impact of employment training programs. The research conducted by this group at the Labor Department continues to influence both labor market policies and how researchers think about policy evaluation.

In some ideal instances, researchers, policymakers, and practitioners have been able to cooperate and set up gold-standard randomized controlled experimental evaluations. But these are very costly and not always feasible. In recent years, researchers have found it fruitful to use administrative data, along with sophisticated statistical techniques, to investigate these types of questions. As administrators of state unemployment insurance and employment and training services, you are viewed by many labor economists as the keepers of the holy grail: data. I encourage you, whenever possible, to try to bridge the gap between research, policy, and practice by allowing your data to be used to examine questions of interest.

This might involve thinking creatively about how to preserve confidentiality of individuals and firms. But many government agencies have found ways to protect confidentiality in order to share data with researchers. And this work can have big benefits. For example, one of the researchers at the Chicago Fed, in collaboration with other economists, used unemployment records from Pennsylvania to examine the earnings of workers before and after they were laid off. The study helped shape our current understanding of the wage losses suffered by displaced workers. This same group of researchers has used earnings records from the unemployment insurance system in the state of Washington and transcripts from the state community college system to examine the benefits to displaced workers of receiving training. We appreciate very much the cooperation of Pennsylvania and Washington in this important research.

In closing, let me say again that an important reason that the U.S. economy is dynamic and productive, and generates a higher standard of living for us all, is that we embrace change. Now there will be people who suffer from such changes. But because these losses are part of the dynamic process that makes us all better off, society can and should help these people find their place in the new workforce.

However, there is a great distance to go from "can" and "should" to precisely "who" and "how." As the pace of technological change quickens, and as more markets open to global competition, job displacement is affecting an increasingly diverse group of workers. If we do not attend to the needs of displaced workers, concern about their future may block policies that help improve our overall standard of living. And this could have costly consequences for the economy as a whole if it were to stifle technology and competition. So this is another reason we need to design policies that support openness to change by supporting those who are adversely affected.

I hope that my invitation to speak here is a sign that our conference was successful in encouraging further interaction between researchers, policymakers, and practitioners around the crucial issue of job displacement.

Thank you again for inviting me.

---

*The views presented here are my own, and not necessarily those of the Federal Open Market Committee or the Federal Reserve System.