Advisor Perspectives welcomes guest contributions. The views presented here do not necessarily represent those of Advisor Perspectives.
Peter Diamond is a professor emeritus at MIT and the winner of the 2010 Nobel Prize in Economics for his work on unemployment and labor market policy. Dan Richards spoke with him last week at the annual Meeting of the American Economic Association held in Chicago.
A video of this interview appears here.
We have just seen the announcement of what is generally viewed as good news on the US unemployment rate, down to 8.5%. But it is still historically quite a high level. How should the government go about determining the priority to stimulating employment and helping accelerate the reduction in the unemployment rate given where we are right now?
The debate over the vocabulary of cyclical versus structural unemployment is a debate that shows up every time unemployment is high. In the 1930s there were people saying the problem we've got is a structural unemployment problem.
If it is structural then people don't have the right skills, and stimulus isn't going to help in that case.
Exactly. Stimulus will cause inflation and we shouldn't stimulate. Of course in the 1930s it was nuts, and also in the late 1950s, early 1960s, the same debate emerged. Bob Solow gave lectures around this issue of how do you differentiate structural from cyclical unemployment, and whether there any reason to believe that in the US economy in the early 1960s, before any stimulus came along, there was completely a structural problem.
Now we have high unemployment and we obviously have a world were globalization, the computer, the Internet and all sorts of things are affecting the whole process of the labor market adapting to change.
Change is always there. Manufacturing companies have been moving to lower labor for more than a century. Blacksmithing died; there was a big drop in what they could do. All sorts of things happen and the economy adapts. That is part of the efficiency of a good capitalistic economy that people move across jobs, although it can be very hard on the people involved. Unemployment insurance and retraining programs help people deal with that.
The key issue right now is that we have some structural unemployment. It's never zero. Maybe because of what has been going on and things have been slow for a while, we have a bit more structural unemployment than if we took a 20-year average or something like that. But all the people who have attempted to split up structural and cyclical unemployment say the bulk of what we've got now is cyclical. It will respond to good stimulus.
Is there any form of stimulus you think makes special sense right now?
I'm particularly in favor of major infrastructure spending. The forecasts and the Reinhart-Rogoff book say we've got a number of years until we get back to a tight labor market. We don't need shovel-ready projects. We can do the serious large projects, and maybe only part of them will happen with high unemployment, but we need them. It will help. We can do more R&D. The drop in support for R&D is plausibly connected to the slowdown in technical progress that we've seen compared to the golden days earlier on. Of course, the US is in serious need of education reform. Some of that involves money. There are lots of issues around education reform.
Right now we have such slack labor markets and output markets, there's no reason to be concerned about inflation and we should be marching ahead with major fiscal stimulus at this point.
The trade-off right now is a concern about an increasing deficit from an already historically high level. So a higher deficit is something you would be prepared to live with to create those infrastructure jobs to reduce the unemployment level?
First of all, let me clarify. It's not just to create the jobs; it's to create the infrastructure. So I view us as having an unemployment crisis. The long-term unemployed have a hard time getting back in the labor market at productive and well-paying jobs. Young people are having delays in the start of a career and are not only losing out on earnings now, but they are not accumulating the experience that leads to higher wages later. They will be hurt for years and years. So we do have an unemployment crisis.
And how would you characterize the issue of the government deficit issue?
We have a debt problem. The US's ability to pay off its current debt is there. We can be very upset at Washington’s shenanigans during the debt-limit fiasco and not coming to grips with Social Security, for example. There is no question that there is inability there. The US could stay on its current track for at least a decade, maybe even two before we'd see the bond market starting to look for higher interest rates, because of the concern about the ability to pay back. There isn't a concern.
As for infrastructure, there are things that we need to do sooner or later anyway. Doing them sooner is going to save money on the debt down the road, even though it is costing us now. So it's important to not behave as if we have a debt crisis and an unemployment problem, but to make sure we've got the timing of those priorities right. The debt problem can wait, particularly if you put in place something like a Social Security fix. It doesn't harm the short run economy. It will have a big hit 20, 30 or 40 years from now.
And if that is in place, the bond market will see that.
I’d like to discuss some of the research for which you received the Nobel Prize, related to labor markets, and specifically to market frictions.
Market frictions are more than just in the labor market. You can go on the Internet, with all that advertising, and it can be hard to decide what you want to buy even when you know what the prices are. It is an oversimplification, but it applies to the problem of finding a job, the problem of finding a loan, or finding what you want to buy. Think of the complexity of switching from one cell phone provider to a different provider and what's involved in sorting that out.
The way introductory economics start – and it is full of great insights for lots of questions – is we have demand and supply and we have a price the clears the market.
And perfect information and perfectly efficient decision-making.
It all fits. It hangs together. We think of the economy as a whole. We have demand for everything. We have supplies of everything. We have a vector of prices, and that is going to determine your incomes, what you'd be able to sell as well as what you can buy.
With that as your starting place and the blatantly obvious failure of everybody who wants a job at the going wage to have one, particularly now, but recurrently, it's not a mindset that works for lots of questions. Yes, it is still works if the demand for labor goes up. You'll expect the wages then to go up. It's still something that works for some questions. But models aren't reality. Models are a simplification to illuminate certain causal and correlation links. You leave out what you believe is not the important part and you put in the rest.
So where does your thinking on this issue start?
The starting place for my thinking is the market clearance story. The claim that the wage doesn't clear the market wasn't a satisfactory story for me, because wages are set by people. In some cases wages are posted by an employer, and if he makes you an offer that's the wage. Sometimes you can negotiate and it's a bargained wage. Sometimes you move from job to job, and sometimes there's unemployment in the way.
The question was how to think through this as a dynamic process. So you look at what it's worth to the firm to hire a person and what it's worth to the worker to take a job. In each case they are thinking about the alternatives. This is particularly true here at the AEA meetings where all the interviews are going on for the new assistant professors for next year.
Firms are thinking, “Would I do better waiting for somebody else? Would I do better because I had some number of applicants, and this is the first interview I've done, or maybe this is the one I think right now is best?” The worker thinks, “Should I take this job? Should I wait, stay unemployed say and look for better job?”
So given all that, how did you approach the question of decision making by buyers and sellers when it comes to labour markets?
First you have to model their individual thinking. Then you have to look at this process and figure out how frequently new job offers come along. How frequently do new suitable workers come along? The whole process of working out that equilibrium is what markets with search frictions, or search equilibria are about.
The prime focus of this has been the labor market, although there have been quite a few studies of the housing market, which has the same characteristic. The economy goes bad, the houses sit on the market, prices don't move very much. They don't move not because there is some force stopping it from moving. It's that people are deciding on prices.
Or are people married to a certain price even though perfectly rational behavior might reduce their price?
Some of it is what we now think of as behavioral economics. Some of it is saying, “Well, I can lower the price now, or I can wait. And maybe it makes sense to wait. Maybe the market will get a bit better and it's not such a good idea.” But that's just a part of the economy.
In order to round out the story, the critical part is to think what is it worth to the firm to open up another vacancy and start searching for another worker. That is going to depend critically on what the firm thinks will be the profit for either alternative. That is going to depend on the nature of that market. Particularly for expansions and startups, it is on the ability of the firm to borrow to finance an expansion for which it would look for jobs.
Peter, thank you very much.
Read more articles by Dan Richards