Research

What has happened to the labor market in the Great Recession?

Lisa B. Kahn — September 2011

With 14 million people out of work in the U.S., labor markets are receiving a lot of attention. Yale SOM's Lisa Kahn did groundbreaking work on the impact of graduating into a bad economy. She offers her take on what's happening now and what to expect.

Q: You were the first person to systematically study the career impact of college graduates entering the labor market in a recession. The work has gotten a great deal of attention. Could you explain your findings?

We intuitively understood that students graduating into a bad economy have a tough time getting that first job. Looking at data to figure out whether there are long-lasting effects to entering the labor market in a recession, it turns out there are very long-lasting negative effects. I took existing panel data for a nationally representative sample of people who graduated from college in the 1980s were followed for the next 20 years. I could see that those graduated in better economies (for example, 1987 and 1988) had very good opportunities and the ones who graduated into the recession of 1980 and 1981 had very bad opportunities. And down the line, they didn't catch up. I found that for a one-percentage-point increase in the unemployment rate, new graduates earn 7% less at the start of their careers. This effect dissipates some as workers gain labor market experience. But even almost 20 years later, those who graduated with a one-percentage-point higher unemployment rate are earning about 2% less. When you compare the unluckiest (the 1980 and 1981 grads) to the luckiest (1988 grads) the earnings gap, almost 20 years later, is 10%. And earnings losses over the entire career add up to more than $100,000 in present discounted value. It's not entirely clear why there would be such long-lasting effects. Here are some possible factors: It's very typical for a young worker to have four or five jobs in their first five to ten years of experience. Most career advancements happen by moving across firms. Workers who enter firms in recessions don't move as often. The jobs they accept might not be exactly in the field that they'd be best at or wanted to be in. And the firm is paying less because of the weak market. What ends up happening is these young workers take a less-than-ideal job at less pay and stay there for longer. It's counterintuitive, but could be contributing to the long-lasting impacts.

Q: What should individuals do to give themselves the best chance in a down market?

In order to get around this effect of graduating in a recession, people get an extra year of school on average. It makes sense: if you don't have good labor market opportunities, the opportunity cost is low. With that extra degree or professional certification, they do about as well as the typical college graduate in a better time. My advice beyond that would be, don't accept the status quo. People should be asking, "Is this the job I want?" If you find the best job match, you're going to be the most productive. They should be searching very hard for the right job. Do they have the right skills? Are they in the right location? Are they making the right connections? They should be willing to accept less compensation, with so many people looking for jobs and firms being really reluctant to hire. At the same time, if things aren't working out, people have to realize, it's not their fault. It's a recession.

Q: What should we be watching in the current market and going forward?

We have 9% unemployment. There are 14 million people looking for work. By far the most fundamental issue is that we need to create jobs. We need firms to want to hire.

Q: What will get companies to start hiring?

I wish I knew. We've had two false starts. In early 2010, things looked great. We saw signs that this was a pretty normal, but deep, recession and we were going to have a pretty normal recovery. We started growing jobs, and then the stimulus ran out. The stimulus was supposed to boost the economy so that it would take off. Instead, it boosted and then nothing. There are lots of short-term factors shaping the situation. Right now, uncertainty at home and in Europe makes firms less likely to hire, makes people less likely to spend money. Rising gas prices earlier in the year didn't help. But those are just small things. The question is, when are firms going to start hiring? And they're going to start hiring when they think that there's going to be demand for their products, but there's going to be demand for their products when people think they're going to have jobs. Bigger picture, we have these long-run shifts in the American economy. We have an aging workforce. We have a shift from manufacturing to services. We have a huge expansion in healthcare and in innovation-type firms—the Googles, Yahoos, Microsofts. As cheap labor became available overseas, the U.S. decided its comparative advantage was in high-skilled labor. We started making goods and services that required a more skilled workforce. Where do we go from here? Basically, the U.S. has to keep evolving and innovating. Going back to the question of what to look for going forward, there's an issue to keep an eye out for once we do have hiring again. If Sony accidentally produced 10 times as many TVs as they were supposed to, it wouldn't be too much of a problem. They lower the price of the TV and clear the inventory. If you accidentally produced 14 million people looking for jobs, it takes time to find the right person for each job. There's going to be a huge bottleneck.

Q: How do labor markets differ from other markets?

A labor market is a market like anything else. But there are some important features of a labor market that make it stand out. It is a two-sided matching problem. Employers are looking for workers, and workers are looking for firms. And there are lots of informational problems. Those informational problems are like getting a used car. You don't know why the person wants to give up the car. Maybe it's a lemon. Except with the labor market you have the uncertainty on both sides. That creates some very interesting features of how to allocate all of these workers to all of these employers in the most efficient way possible. Problems exist because of frictions; for instance, it's impossible for me, as an employee, to search for every possible employer on the planet and know exactly how I would fit at each of those firms. Similarly, it's impossible for each of those firms to know exactly how they would feel about me if I were working for them. Those issues continue to be extremely important. The sorting and matching process that goes on is a very long process. That's why workers move around a lot, until they find the right fit for them. The information structure separates the labor market from some other markets. I've been looking at one aspect of information asymmetry. When you work for a firm, they learn a lot about you, but I've shown that outside firms are not learning as much. That has consequences for whether you want to move to another firm. Let's say you've already proven yourself to your current firm. You haven't yet proven yourself to the outside market, because, in some instances, they can't really see how well you've been doing. So you might be less likely to move. You might have worse opportunities because of it. There might be some matches that don't take place, even though they would be efficient, because they don't have the information. It's an issue that's very hard to measure and quantify

Q: How effective are companies in evaluating their current employees?

Firms need time to learn about workers even as workers' abilities are evolving. Firms have to understand a moving target. With new technologies and computing power, firms are collecting more data on performance and have increasing power to sift through all the signals coming in about people. At the same time, jobs are getting more and more complex, so even as managers have more and more information at their fingertips, they now have to be predicting something much more complex. On top of that, it's very hard to predict how well you're going to fare at each new task a firm assigns to you. Sometimes you're promoted and don't do well with what seemed a logical next step. Sometimes you're assigned to a random project and excel. Firms make a lot of mistakes and workers are placed in not quite efficient areas. All of this explains why firms invest very heavily in HR departments, which do regular performance evaluations of their workers. It just makes sense that they need to be continually appraising what's going on, because workers are going to be different in this task than that task.

Q: Is there a market-based reason for why top executives are earning larger and larger multiples of what the average worker is paid?

There are lots of efficiency reasons why top executives should be making so much money. For one thing, it's seen as an incentive for them to work harder from early on in their career. It's like a tournament. Those in the first round are not making very much money or even getting a trophy at the end of the day, so why bother? Well, it's because someday you may advance to the later rounds of the tournament. You may win the tournament and the trophy. That chance can motivate you early on. It's actually a way workers would be willing to accept less money at a very entry-level position. Law firms are a classic example, where associates come in and work so many hours that, per hour, they're not making very much money. Then they either get promoted or have to leave. Why would you ever want to be in a system like that? Because, once you get promoted to partner, you get to share the profits with the firm. It's a great deal. It's just how many markets function. Another part of this is that as a top executive, you're regularly making choices to either press Button A or Button B, and each decision could have millions of dollars of impact on the profits of your firm. And so when you think in terms of people being paid their marginal product, as we often do in economics, the CEOs have the most worth. They have the most ability to impact the revenue and the profits of the firm; they have a huge scope to influence how the company does. They're getting compensated for that scope.

Q: Does the growing economic inequality in the U.S. tell us anything important?

Inequality, on the face of it, is not a bad thing, because inequality is really just incentives. It's saying what you do matters. If everybody earns the same amount of money, no matter what, then no one has any incentive to try. Where negative side effects of inequality come in is the fact that households at median earnings on down have not seen their real income rise in recent decades. In fact, for those below the median, their incomes have fallen in real terms. That's a big problem. The costs of education and healthcare are rising faster than real terms. How does the average family cope with this kind of thing? If you're in a poorer situation, your kids are going to have worse opportunities and then they're going to end up maybe worse off than you were, and that's just going to breed more inequality. To me, the most important goal of society should be that every person has the tools to be able to reach their full potential, whatever that may be. In a very unequal society, you're not going to get that, because you have clusters of bad schools, a lack of mentors and opportunities. And I think inequality tends to perpetuate that. This issue is closely related to the problem of polarization: employment has been shifting to either high-skilled, high-paying, complex jobs or low-skilled, low-paying jobs. David Autor, from MIT, has done a lot of fascinating work in this area. We're going to see demand for sales people and scientists and opticians, nurses and doctors. Those are complex jobs that a machine can't do and you can't outsource. We're also always going to need janitors and dry cleaners. Services are in high demand in the U.S. The middle has almost disappeared. It used to include tasks that now are either automated or outsourced. The people who used to do the middle-skill jobs, they've got to go to one end or the other. They're more likely to go to the lower-skilled end, because that's an easier transition to make. That suppresses the wages on the low end. So in the U.S., those with a high school diploma or less can't find work that lets them stay in the middle class anymore. That is probably part of why we've been having this huge inequality problem in the last 20 or 30 years. Certainly, Wall Street, the tax code, and any number of other things play in too, but the economy has undergone a long-run structural change, where we don't need as many strong, white, male high school graduates as we did before. Compounding that, educational attainment among white men is going down. Women have passed them. What they should be doing is getting more schooling, particularly specialized technical skills. But they're not. It's not clear why.

Q: Are there things to be learned from how the labor market was impacted by this recession?

The U.S. lost an astonishing number of jobs, even compared to similar economies in Europe. But we did not fare the worst in terms of GDP; and, productivity actually went up. Germany's unemployment rate is lower than at the start of the recession, but their productivity is relatively low. That suggests that the recession, as with everything else in the U.S., hit workers very unequally. There are some workers bearing too much of the burden of the recession, because while some companies are doing fine—they're making profits, they're productive—some workers are out of work and we have worse safety mechanisms for them. Our typical unemployment insurance is atrocious, compared to Europe. I think everyone would have been better off, as a whole, if we could have more evenly distributed the burden.

Q: You've recently returned to Yale SOM after a stint as senior labor economist on the president's Council of Economic Advisors. Did being in the role of an advisor change how you work? Does the political feasibility of an idea play into making recommendations?

I still have an order of things that I think would work better or worse. It's just there are some things that I know won't ever actually happen. It doesn't really shift the way I fundamentally think about these issues. My time in Washington may shape my academic work. Needing to consider the bigger-picture policy and politics made me think, what research topics should I be thinking about? If I think optimal corporate taxes will never be on the table, why would I study it? Whereas, optimal unemployment insurance schemes are a very hot policy topic, and that relevance makes it more interesting to me.


Interview conducted and edited by Ted O'Callahan.