Interview with Lee Ohanian

Interview with Lee Ohanian (University of California, Los Angeles)

Paper: “Neoclassical Models of Aggregate Economies”

The paper’s main takeaways, according to Ohanian:

  • The big-picture summary is that a lot of our economic historical record can be reasonably well understood within the fairly simple mechanics of supply and demand, and there are times when various forces interfere with supply and demand. Sometimes forces depress output or depress competition, which results in depressing output and trade between people. There are some government forces that might impede productivity growth, such as restrictions on immigration of highly skilled workers.
  • We can learn a lot about our economic record in terms of its success and its limitations based on a careful reading of the historical record of economic policies. We can put those through the lens of pretty simple economic models and learn a lot about that.

The Federal Reserve Bank of St. Louis hosted its 40th Annual Fall Conference on Oct. 15-16, 2015. David Andolfatto, a vice president and economist with the St. Louis Fed’s Research division, sat down with each of the conference presenters and discussed their work in plain English. The content below is from those interviews. All interviews have been edited for clarity and length, so the content below should not be considered a transcript. 


Why don’t you tell us what are you talking about here in your paper, “Neoclassical Models of Aggregate Economies?” What’s this paper about?


Lee OhanianIt’s actually a really simple idea. The idea is that economies are always being hit by various kinds of shocks. And two of the biggest shocks that we can think about are shocks to technologies and shocks to government policies—fiscal policies, government spending, tax rates, regulatory policies. The technology shock could be something like the development of the microprocessor or the development of the Internet. It could be something like a change in the weather, such as the drought in the western part of the United States.

Economies evolve and respond to these shocks through market processes. And in doing so, we see movements in how many people are working, how much people consume, how much firms invest and how much is produced. The idea in this paper is to look at just two factors: changes in technologies, which have been such a big part of our lives, and changes in government policies. And we examine only those factors that are really long lived, such as the Internet or income taxes, which began in 1914. These are large, permanent changes in government policies and technologies.

How we learn about the actual economy, as you know, is we simulate model economies. We simulated a variety of model economies in which markets function very well. That’s what we mean by a neoclassical model.

Then we asked, “How did those model economies respond when we look at changes in technology and changes in government policy that we can measure using data?" And then we asked, “To what extent do the movements in the model economy correspond to actual movements in the data?”


I gather from the approach you take here that this is a bit in contrast to the traditional business cycle approach, where at least in the last 20 or 30 years, we were more interested in what we call higher frequency movements in the data. It seems here what you’re trying to do is try to take the same modeling framework and see to what extent it can account for broader swings, lower frequency. Is this a big departure from what has traditionally been done? Is this something that’s important, to look at the lower frequency movements?


It is a departure. As you mentioned, the traditional approach was to assume there were two pieces to economic data: There was a very smooth trend, and there was what we call the business cycle—movements that were very, very short-lived.

But what we show in this paper is that there’s a big part in the middle that’s not very short run and that’s not super long run. And there’s a lot of movement and data that correspond to what we call these middle frequencies.

To us, it makes a lot of sense, because when you think, “OK, the Internet’s invented. There’s a new technology,” or, “The microprocessor’s invented,” those innovations are going to take a long time to evolve and disperse through the economy. We should expect to see things like the development of the Internet or the development of the microprocessor, the development of smartphones, to have an impact on the economy that lasts five, 10, 15, maybe 20 years or even longer.

What we see in the data is that the movements that correspond to those 10-, 15-, 20-, 25-, 30-year cycles are very big. In fact, they’re bigger than the movements at the business cycle frequency. What really surprised us is that, today, the business cycle frequency movements are almost all gone. The remaining movements are these 10-, 15-, 20-, 25-year movements.


Another characteristic or benchmark of your modeling approach is that you are beginning with the hypothesis that markets work relatively well on their own. That might strike a lot of people as odd, especially at higher frequencies where we think that the business cycle is often, at least in part, a manifestation of markets not working so well at times.

I suppose, given your emphasis on the longer horizon, do you think that this view might be more acceptable and it’s more justified or something like that? Oftentimes people think markets don’t work well in the short run, but in the long run they work things out.


A common view toward thinking about business cycles is that there are changes in monetary policy, and the dollar price of goods and services doesn’t respond immediately in response to those changes in monetary policy. The fact that the dollar prices of goods and services don’t change immediately perturbs the economy in particular ways. By looking at these 15-, 20-, 25-year movements, that’s plenty of time for dollar prices of goods and services to change, for the wage rates of workers to change and for search behavior of workers to sort itself out.

When you think about the market process in advance, like the United States, you look at these long-run episodes we used to have: 80 percent of the workforce was in agriculture in the 19th century. At that time, we needed 80 percent of our workers in agriculture in order to feed ourselves. Today, it’s about one-half of 1 percent. The market processes evolved to move those workers out of agriculture to produce manufactured goods as well as services.

So when we look at these long-run issues, we think market process is incredibly powerful for understanding movements like from agriculture to manufacturing and services. We think that these longer-run movements are exactly what the doctor ordered for thinking carefully about the usefulness of modeling markets that function well.


There are many episodes—not just in the United States, but in other countries—where the countries enter into these slowdowns or these phases or even great depressions or lesser depressions. It’s probably a bit difficult to reconcile with the notion of new technologies, though not impossible, I suppose, if the new technologies are disruptive. But there must be some other hypotheses at work here to account for these episodes. What was responsible for the recent crisis through the lens of your modeling framework?


Let’s start with Europe. We look at data from Europe, and again, we use this particular decomposition in which we separate the data into these different pieces. Some pieces are what we call very short movements associated with the business cycle, and then these longer movements are 15-, 20-, 25-year movements.

In Europe, what we find is almost all of the movement is due to these 15-, 20-, 25-, up to 50-year movements in the data. There’s hardly any movement whatsoever from what we would call the business cycle. This is really important, because it is at the business cycle that we think monetary policy would be the most important. That’s a case in which markets that don’t work perfectly, such as wages or prices that don’t adjust instantaneously, are going to be relevant. In Europe, you see virtually no business cycle fluctuations. It’s all due to these long-run movements. Then you combine that statistic with the fact that many European countries—including Spain, France, Germany, Italy—have had almost no productivity growth in the last 35 years, and that’s a remarkable fact, because you think about the Internet, microprocessor, all of these things have transformed our lives.

I think what may be going on there is that Europe has adopted a lot of regulations, a lot of restrictions, that actually ironically make markets work less efficiently-restrictions on labor and labor mobility and hiring restrictions and firing restrictions. Venture capital is very difficult to access in Europe.

A lot of regulation protects incumbent businesses at the expense of entrance. A great example is in France. France banned Amazon from free shipping. We have Amazon Prime, we click on the icon and make our order, we get the product in two days, and there’s no shipping charge. You can’t do that in France. Why is that? The local bookstores lobbied the French government and said, “Hey, we can’t compete.” So France passed a law saying, “OK, Amazon, you can’t have free shipping.” So you see lots of these regulations in Europe. There’s some research I’m doing with Jesus Villaverde, who’s at the University of Pennsylvania. We’re looking at this issue.

What I conjecture is that these types of government policies that impede and suppress the competitive forces, which are so important in a market economy, are responsible for why productivity growth has been so low in Europe. When viewed through that lens, the crisis in Europe almost looks like a manifestation of what’s been brewing for a long, long time. Economists and policymakers look at Europe and focus on debt crises, such as the problems that Greece is having, but that didn’t come out of nowhere. I think that’s been a long time brewing.


Your model ascribes absolutely no role to financial market frictions or debt overhang. Oftentimes, we abstract in our models to get at what we think are more important forces. But do you think that—in light of the evidence, say, by Carmen M. Reinhart who suggests that following major financial crises, economies stagnate for quite a bit of time—there is an effect through deleveraging and things like that? Would it be fair to say that your model is identifying different forces at work that has either nothing or very little to do with debt deleveraging or financial market frictions?


The model we have as of now is one in which financial market imperfections and debt deleveraging is not playing a role. You’re absolutely right: It identifies different forces and identifies forces that are moving in big ways that economists and policymakers haven’t discussed very much. One of these issues is productivity growth, which is really what this model is all about, and it has been slowing down for a long time. Productivity growth seems related to entrepreneurship activity and startups.

If I can have a moment: If you think about job creation in the United States in an average year—not a recession, but in an average year—if you just mechanically take away the startups and add up all the jobs created and all the jobs lost, then the U.S. economy actually loses jobs just from all the incumbents. If you look at gross job creation, startups and rapidly growing young businesses account for about 60 percent of job creation. So that’s really where the action is in terms of a growing economy.

But entrepreneurship rates are down 35 percent today relative to where they were in the 1980s. In the U.S., a lot of what was going on in the financial crisis, I don’t think it’s just like we snapped a finger and suddenly Lehman Brothers came down and we had the problem with AIG and then all heck broke loose. I think that the underlying economy was weakening long before that, reflecting a significant drop in new business formation.  

That’s not to say financial issues were unimportant. But the longer our economy continues to be below its pre-2007 trend and hasn’t made much, if any, recovery whatsoever to that trend, I think the Great Recession is going to fade out of view. And the big question becomes: Why aren’t we recovering? It’s been seven years now, and productivity growth is 0.9 percent as opposed to 2.5 percent average. Employment population rate is way, way down.


Is it a bit of a tough sell for people out there? Your model emphasizes productivity, and I think most people out there would accept that productivity movements influence the longer-term movements of the economy. But people must come up to you and ask you, “Are you really serious?”

Do you really believe that it was a productivity event or accumulation of development of either sectorial or aggregate productivity that was somehow responsible or perhaps in conjunction with a set of government policies that was responsible or could account for the recent Great Recession, the financial disruptions? Everything that the Fed worries about and that government policymakers worry about really suggest that they truly believe something is still amiss in terms of financial markets. How do you really persuade people about this stuff?


There’s really two issues. One is the crisis and the rapid drop in employment that occurred in the first three or four months of 2009. Other things are going on, absolutely. But what I’ll ask people is to step back and say, “Okay, what are by far the most important determinants of long-run living standards?” And people will say productivity, absolutely, no question whatsoever. I said, “Well then, let’s work ourselves backwards. How about at 30-year levels productivity?”

Then we can have that conversation. At some point you’re going to say, “Well, okay. Productivity’s important. It’s always important. Maybe it’s more important at 10- and 20-year horizons than one- or two-year horizons. But it’s there.” Then I’ll tell people, “Let’s take a look at what’s happening to information technology productivity growth.”

That’s been really such an important force in driving the economy in the last 30 or 35 years. You think about Apple, Google, Microsoft really transforming the world and the economy. What we’re seeing is far fewer startups in IT, far fewer successful startups in IT, successful startups meaning businesses that grow at least 25 percent per year in terms of employment for at least five years.

Of those that have that rapid growth rate, there are far, far fewer that are really taking off the way Apple did or Microsoft did or Google did. You look at that statistic, and that’s a grim statistic. I think a lot of people who have thought this was a financial crisis, and even with Reinhart-Rogoff, thought we should be having some recovery by now. We’re not seeing recovery.

You look at businesses, and as you know, they’re cash rich. The corporate sector has enough cash on hand or more cash on hand to finance all the investment that they need to do. So the current economy is a puzzle, I think, from a lot of perspectives. I think it’s particularly a puzzle for those who look at financial forces as being the be all and end all. If that was the case, then you’d say, “Well, why haven’t we gotten back? These businesses have tons of cash. Interest rates are close to zero, at least on government securities. Why aren’t they investing?” My response would be: “It’s an underlying economy that’s not nearly as healthy as one we used to have.”


A lot of people point to those same statistics as evidence that the financial markets are impeded in some manner. Firms don’t typically hold that much cash unless they want to build up a precautionary saving. We know that imperfect financial markets are what cause a buildup in precautionary saving.

So the evidence that they have cash is actually indicative that they’re fearful of something: future borrowing constraints or future finance conditions. You mentioned real interest rates on U.S. short-term Treasuries are –2 percent. Inflation is very, very low. It just sounds very strange. It may be coincidental or symptomatic of a deeper problem is what you’re suggesting?


Take a parallel of something like Japan. Japan’s had very slow economic growth for a long, long time. They’ve had such low inflation, occasionally deflation. Not particularly severe deflation, but some deflation. Interest rates are near zero. There are some economists, I think, who have the view that, if we could just pop up inflation from zero to 3 percent in Japan and get them off that zero lower bound, that would pretty much bring back prosperity to Japan.

I look at Japan very differently. I look at Japan as a country with an aging labor force, very low birthrate, no immigration, very low entrepreneurship. Low entrepreneurship means few startups. Few startups means few successful startups. Few successful startups means very few new engines of economic growth. So when I look at Japan, I think going from zero to 3 percent inflation is not their panacea.

So the data that you mention can be interpreted in different ways. For those who think financial markets are really what’s holding the U.S. economy back, I would say the questions have become: “Why? And when will it turn around?” The Reinhart-Rogoff evidence doesn’t say this continues forever, but I don’t see much signs of recovery. And I do see very little productivity growth.


The same question could be asked of you: When will productivity growth turn around? We can’t make those forecasts. We can only remain hopeful that it will, I suppose. And I guess government policy in your model, strictly speaking, takes quite a laissez-faire approach.

What sort of practical policy implications could you recommend—at the federal level or lower levels—in order to get out of this funk that we’re in?


We talked about entrepreneurship and startups. Successful startups are disproportionately done by immigrants. 45 percent of the Fortune 500 was founded by an immigrant or the child of an immigrant. And we have severe restrictions on immigration, a lot of which go back to the 1980s, many of which are country-specific quotas.

So I think the simple answer is reforming immigration, particularly for highly skilled workers. We have these people that come here and that study at Stanford and MIT and Caltech, and they get PhDs in computer science and electrical engineering.

They want to start businesses here, but we make it hard for them to do that. Congress understands this, and there are bipartisan-supported bills in both the Senate and the House aimed at bringing in more immigrants who want to start businesses. Based on my reading of the data, if I could think of one policy, it would be that one. And it’s heartening for me to see that Congress understands that and they’re trying to move forward on it.


Very good. But I’m going to ask just one question. Is there a role for government in any dimension in terms of doing good things for building—not just promoting growth by removing restrictions that they’ve imposed? Perhaps infrastructure spending or whatever. I’m just curious to know what your thoughts are.


If you went back to President Obama’s American Recovery and Reinvestment Act, a part of that was for infrastructure investment. The president talked about shovel-ready projects, and there’s not so many shovel-ready projects on at any point in time. Only about 3 percent of the ARRA went towards infrastructure investment.

Now you look at the American Society of Civil Engineers. They grade the country’s infrastructure every year. The grade they give it is D.

You think about all the resources that we have in this country, and our infrastructure is crumbling. So massive investment in infrastructure, based on what civil engineers think about how adequate our bridges and airports and roads are, I think would be called for.

Government also plays an incredibly important role in trying to make sure the playing field is level and in trying to make sure that we have competitive markets where everyone has the chance to compete. This goes right to the heart of criticisms about “crony capitalism.” A lot of what I think people sometimes worry about involves them saying, “That industry or that guy is getting a subsidy or benefit, and I’m not getting that subsidy or benefit.” So I think governments that level the playing field enhance competitive forces, and those are really important things for governments to do.