Advisor Perspectives welcomes guest contributions. The views presented here do not necessarily represent those of Advisor Perspectives.
Dale Mortensen is an economist and a professor at Northwestern University. He was a co-winner of the 2010 Nobel Prize in Economics. Mortensen's research focuses on labor economics, macroeconomics and economic theory. Dan Richards spoke with him in early January at the annual Meeting of the American Economic Association held in Chicago.
A video of this interview appears here.
There are many public policy issues that policymakers around the world are grappling with today. One of them relates to unemployment. In the United States, unemployment is 8.5% which is less than what it has been, but still high by historic standards. There is a view in fact that that 8.5% under-represents the real number. Based on your work, how serious a problem is the level of unemployment in the US and other advanced economy's today?
It's a very important problem. To give some notion of its magnitude, at the moment in the United States there are six million fewer people employed than there were at the beginning of the recession. And that doesn't account for the fact that the size of the labor force has risen in those four years. That's huge. It's the biggest number we've had.
There are some who claim that we are in a new equilibrium, that the problem is a mismatch between the jobs that are available and workers’ qualifications.
Is it a structural problem?
I don’t agree that it is a so-called structural problem. There is some of that, and there is even more potential for it in the future, because what we do know from past research is that long periods of unemployment adversely affect people’s ability to become reemployed for a variety of reasons. But if unemployment is four percentage points higher now than it was, I would attribute only a quarter of that to structural issues.
From a public policy perspective, there are two broad views on what's to be done. One of them is to let the business cycle work its normal course. We have been through periods in the past of economic slowdown, so just let the normal mechanisms work. The other view is to say that we need to accelerate that process of recovery through a significant fiscal stimulus and other policies to create employment. Where do you stand on that?
The answer is obvious. By letting things go as they are we are throwing away close to a trillion dollars a year in terms of goods and services. What do you do when private demand falls? Well, particularly when wages fall and prices fall, it may be a good time to substitute public demand. It isn't that we don't have needs for that. So I'm of the view that a much more aggressive fiscal policy is in order.
Of course, that raises a whole set of other problems about the current political situation.
There is obviously a politically charged environment around budget levels and debt.
That's right. There is a short-run and a long-run problem and it is hard for politicians to distinguish the two.
Let's get your view on this. The short run problem is?
You are not going to be able to conduct fiscal stimulus in an effective way without increasing the deficit in the short run.
What is the long run issue on which we should focus?
The long-term issue we all know, at least we should know, began with a couple of things – the Bush tax cuts, two wars that for one of the first times in history were not financed with a surcharge tax, and a growing commitment to social programs that probably need some reformation.
Coming back to the structural unemployment issue, there are problems with skills, educational levels, and in some cases attitudes toward work that need to be addressed as well.
These are secular problems, or a long-run problem that is always with us. Technology is always a step ahead in some respects with respect to education, so you are always fighting an uphill battle on that score. The underlying unemployment numbers are always there, but there is no evidence that suddenly it became a huge problem that accounted for the increase in unemployment. For example, the unemployment rate went up almost four percentage points in one year. That didn't happen because of structural issues.
Let’s turn to your current field of research You received the Nobel prize for your work on search frictions in labor markets. Can you define what search frictions relate to?
It refers to the fact that in many markets, but not all, there is a matching process, a process of learning about where the jobs are, what the jobs involve, what's available on one hand, and on the other side of the market, employers trying to find an appropriate worker for their job. It refers to that process, and how that process affects unemployment, and how other relevant variables interact.
In theory you'd have perfect information on the part of sellers and buyers, and there would be a match between them.
That is the classic theory. Consider a market like a commodities exchange where there are professional traders. Although this is not explicitly discussed in the model, the traders provide that information in terms of matching bid and ask prices.
What are the implications and applications of the work you’ve done in terms of government policy and labor markets?
There are a broad range of applications. Some of the policies that were addressed very quickly after the creation of our rudimentary model were issues like what affects how long people are unemployed. The issue was whether there would be some influence, for example, of unemployment insurance. The model explicitly said yes, there are disincentive effects. There is an obvious purpose in unemployment insurance, but with every insurance program there are also side effects which we tend to call moral hazard. The fact you are insured affects your behavior.
Your work goes back a fairly lengthy period of time. Have more recent phenomena, like the Internet, changed the dynamics of the interaction between buyers and sellers with search frictions?
Peter Kuhn at Santa Barbara has been researching this area. He conducted a study 10 years ago, although the data was older than that. The data probably pertained to the early- to mid-1990s, just at the beginning of the Internet era. At that time they were trying to measure the possible effect.
Let me first discuss the model. The model says under certain circumstances the Internet should reduce the frictions. That should shorten unemployment durations. However, the more sophisticated version of the model says maybe that's not the only answer, because it may change behavior. People are going to realize if it is easier to find a job, maybe they will raise their standards for an acceptable job, and that could wash out the duration effect. So there's a quandary.
The study using the early 1990s data found no effect. Kuhn was comparing those who at the time were using the Internet with those that were not. He has since replicated that study with new data. The new study says there has been a big effect.
There is an issue how you explain that. One is that the technology’s matured.
And the adoption rate of the Internet accelerated.
Absolutely. There were very few people who knew how to use it back then on either side of the market, and now it has become standard. If you can find somebody who doesn't use the Internet and compare that with somebody who does, then it turns out there is quite a difference.
Early on in the Tech boom, in late 1990s, there was a view that the Internet was going to make everything a commodity, and that entities like monster.com would match buyers and sellers and create a much more perfect and efficient method of matching demand and supply for labor. Has that happened based on the work that you've done?
The commodity trend is well defined, particularly in some financial markets. Let's take a futures option, which is a contract between a buyer and seller. It is well understood. It is well written down. In that particular case, the matching process can be particularly mechanical.
But for other commodities where what is being traded is not well understood, such as the labor market and the housing market, frictions come into play. The internet can have an impact as I mentioned before but it is mainly at the stage of contacting. It provides a mechanism by which applications can be made and responded to in an efficient way. But beyond that there is much more information that both sides need to know before a transaction occurs.
It's not purely a commodity.
Yes, you just can't commoditize because the heterogeneity is implicit in the individual and also the jobs that vary a lot, as in the housing market. So the internet helps, but it doesn't solve the problem of search frictions.
Read more articles by Dan Richards
