John C. Haltiwanger, Distinguished University Professor in the Department of Economics at the University of Maryland, is known for his research on the determinants of firm-level job creation, job destruction and economic growth the United States. He has created the statistical measures many agencies around the world use to measure firm dynamics.
Consequently, it was no surprise when The Wall Street Journal sought his views for an article on how America’s “risk-taking spirit appears to be fading” (“Risk-Averse Culture Infects U.S. Workers, Entrepreneurs,” June 3. 2013). Prof. Haltiwanger responded that “The U.S. has succeeded in part because of its dynamism, its high pace of job creation and destruction, and its high pace of churning of workers…The pessimistic view is we’ve lost our mojo.”
So naturally Innovation Management Report’s Editor Michael F. Wolff began his subsequent interview with Prof. Haltiwagner by asking:
Are you one of those pessimists? Has the U.S. lost its Mojo? And what is the principal evidence for that?
I think the evidence is clear that the United States has benefited substantially over the last 30 to 40 years from a highly dynamic economy with lots of business entries and lots of business exits. Year in and year out we typically get a large number of entrants that add a lot of jobs when they come in. Although most of them fail, among those that survive are very-fast-growing firms that add lots of jobs and lots of productivity.
The concern is that the pace of entrepreneurship has been steadily declining over the last 30 years, especially since 2000. In some ways we’ve never recovered from the last two recessions. Business startups took a dip in the 2001 recession and never recovered. There was actually a high pace of entrepreneurship back in the late 1990s with the dotcom boom.
What exactly do you mean by the pace of entrepreneurship? The number of companies that started up?
I’ve been working with the United States Census Bureau to help develop the Business Dynamics Statistics database. It tracks every firm in the private nonfarm sector since the 1970s, so we can actually track when a new business comes into existence. We track not only the number of such businesses but the amount of employment associated with those businesses. So the declining pace of entrepreneurship means a decline in the entry rate, which is the number of new businesses divided by all businesses. That number has fallen steadily.
Another statistic is the number of jobs created by new businesses as a fraction of overall employment. That has also fallen.
One statistic I find useful is to not just think about the contribution of young businesses at entry, but their contribution to employment. Thirty years ago, 20% of workers worked at businesses that were less than five years old. Now only about 10% do. Unfortunately, there was a substantial decline in that number in the last five years because of the 2007-09 Great Recession.
So your evidence is these statistics, which show the pace of entrepreneurship declining.
Yes. But the question is, why are we seeing this decline in entrepreneurship? There’s no doubt there is a decline.
One concern is that somehow the United States is a less attractive place for entrepreneurship, or, could I say, less propensity. That’s sort of the take the Wall Street Journal article took: less propensity to engage in entrepreneurship. Being an entrepreneur is very risky and most entrepreneurs fail.
I don’t know that we’ve got any strong evidence that that’s what is going on, but I do know that entrepreneurship has contributed substantially to both job growth and productivity growth historically. So it’s hard to view this decreasing rate as good news
Let me segue to startups. Your recent paper, “The Secular Decline in Business Dynamism in the U.S,” stated that business dynamics in the U.S. have declined, and firm startup rate is an important component of that. So I would ask you, how important?
Startups are just a part of this dynamism. What’s remarkable about the U.S. economy is that it’s constantly reinventing itself so that in any given year some businesses are starting up and/or growing rapidly while other businesses are shutting down and contracting. The pace of this is quite high.
We’ve found that restructuring and reallocation — which is obviously costly — is also productivity-enhancing. The U.S. is a place with lots of experimentation; we try different things, the things that work take off and the things that don’t work contract.
Relative to the rest of the world, the U.S. has been very good at moving resources away from less-productive activities to more-productive activities. That’s been a key strength of the United States.
So a related statistic, and you could say a related concern, is that the pace of this reallocation—the contraction of some businesses and the expansion of others— has been declining. So you ask yourself, why has the pace been declining?
Contraction of existing businesses?
It’s both. If you go to businesses that exist in March, which is usually the focal point for administrative data, and you ask what fraction of those jobs didn’t even exist one year previously, historically in the United States it’s about 17%. That’s a big number.
Then, if you ask the flipside question: In any given March what fraction of the jobs that are there today won’t be there next year, it’s about 15%. That suggests the U.S. is constantly moving resources away from some businesses toward other businesses.
Startups are only one part of this process, because some of this reallocation is going on among existing businesses. But I would say that startups play a particularly critical role because the startup process doesn’t stop at the year of entry.
In a typical year, let’s say 2006 — the last year where we had really robust net
job creation in the United States — net job creation in the U.S. private nonfarm sector was four million net new jobs. Startups alone accounted for 3.5 million jobs. That’s a big number!
Of those jobs typically created in a new cohort, about 50% are gone five years after entry. But even so, the surviving businesses grow so fast that five years later that cohort actually has about 85% of the number of jobs it started with. That says that after five years each cohort hands off to each generation a substantial chunk of jobs.
If you think about what I’ve just said, what’s going on with entrants is not just at the point of entry but there’s an enormous amount of dynamism in the first five and actually ten years. This reallocation is disproportionally high among young firms.
Nevertheless, it’s not enough to stop this decline in entrepreneurship.
In 2006 we were getting 3.5 million new jobs out of startups in their first year, but by 2009 we were only getting 2.4 million jobs. That’s 1.1 million jobs per year down from startups alone. Had we been back at 2006 levels with startups we would have been up around 200,000 to 300,000 jobs a month. That would be an incredibly robust economy. Now, you’ve got to be careful with this kind of simple accounting, but nevertheless the point is that startups fell dramatically and they have only recovered a little.
What you’re telling me is surprising to someone who sees the news with exciting stories of new startup activity, not just in Silicon Valley, but in New York City, in Austin, in Boulder, Boston, and elsewhere. Techstars, for example calls itself a mentor-driven accelerator model and says there are now 1,000 programs modeled after it. And there’s Dreamit Ventures, a startup accelerator from which 112 startups have graduated so far. But all this activity is apparently not making much of an impact?
I’m talking about declining, not disappearing. We are still getting 400,000 startups a year, and they’re adding 2.5 million jobs a year. That’s a huge number relative to the net changes in the U.S. economy. But the point is that we are still down.
The U.S. is still a very entrepreneurial economy. But I say the loss of a million new jobs a year is a big number.
Let’s talk about entrepreneurs. Hurst and Pugsley have observed that many U.S. entrepreneurs have little intention of becoming high-growth firms. Do you share that finding?
I don’t disagree at all with their facts although we may have some subtle differences in interpretation. We agree that most businesses that start up fail and that even most of those that survive don’t grow. So it’s a mistake to say that all we need to do is to target startups because that’s where the action is.
What’s pretty remarkable about the United States economy is that a relatively small fraction of the startups grow very rapidly, and they’re the ones that are accounting for all these jobs. And so success is very rare in the U.S.
Can you put a number on “relatively small”?
The typical young business has a growth rate of zero. But the top 10% are growing at more than 50% a year. They are the ones adding lots of jobs.
These are the people you call transformational as opposed to subsistence entrepreneurs?
Exactly. However, I don’t think even the individuals themselves know that; and I don’t think we know enough about that process.
The Kauffman Foundation recently called entrepreneurship “one of the fastest growing subjects in today’s undergraduate curricula.” It seems every campus has courses or departments or Master’s degrees on entrepreneurship. What sort of impact do you see from them?
The statistics agencies have traditionally focused almost all their attention on large, mature businesses because that’s where most of the activity is. If you’re only trying to measure the level of employment, or even the level of GDP, go ask the big guys, because they’re producing most of it. But, if you want to track who’s producing most of the change, you want to go to the young guys.
I think the statistics agencies have finally figured out that we are actually interested not only in level but in change. So there’s been a revolution in the amount of data and, therefore, studies available on entrepreneurship.
The Kauffman Foundation has been a major supporter of that activity. They’ve taken a kind of a “let a thousand flowers bloom” approach of supporting research and related activities. It has contributed to a burgeoning area of research in economics, and indeed that’s translated into both graduate and undergraduate programs.
As people graduate from these programs, do you think they will have much of an impact in turning around this decline?
That’s a really good question. Part of my answer is I just don’t feel I really understand why we’ve seen this decline. I think it’s good to document it, and certainly I’m working on it, trying to figure out what it is in the economic environment that has led to this.
There’s lots of evidence, and not just my own, that entrepreneurs and business dynamism have contributed substantially to U.S. economic job and productivity growth. So it makes sense that we should be spending time and resources trying to understand those dynamics.
I think we’ve made great progress, and some of the work we’ve done has put new facts on the table, but it has also raised new questions.
How do you think we could turn this decline around? You suggested in your May 2013 paper, “Who Creates Jobs?”, that we should refocus the debate about private sector job creation. “Policies targeting firms based on size without taking account of the role of firm age are unlikely to have the desired impact on overall job creation… we have to identify the specific market failures that prevent entrepreneurs from starting and growing new businesses.”
I actually think we’ve been spending too much time on small businesses, and this is no offense to small businesses because they play an important role. But if we’re looking for where there may be market failures, we ought to be looking for where there is this dynamism and this contribution from job creation, and where it’s declined.
Actually, there’s not much decline in the share of activity by small businesses but there’s a huge decline in the share of activity by young businesses.
I have great respect for the Small Business Administration, and think they have done lots of good work in at least advocating for better data on business dynamics, but I am skeptical of a policy approach that says what we need to do is to help small businesses.
Behind your question is, what do we need to do for young businesses? I don’t know at this point, because I don’t fully understand what’s causing this decline. If it were me, I would change the SBA to something closer to a Young Business Administration. How could that help? Instead of subsidized loans, which the SBA currently does, the right policy prescription might be to reduce some of the barriers that young businesses face.
Barriers such as?
We actually know more about this around the world than we know in the United States. The United States is often perceived as a place where regulatory barriers are not especially high, where problems of infrastructure aren’t particularly difficult, and so on. But if you look at studies from the World Bank and their Doing Business project, there is lots of evidence that countries that make it difficult for a business to start up really restrict the kind of dynamism I’m talking about, and restrict the role of entrepreneurship in those countries. That’s partly why countries like India have struggled.
Now that’s the extreme. If you really hammer the startup process, if you make it really difficult to open up and expand, and be over a certain size, that can really squelch an economy. I don’t think the U.S. is anywhere close to that. I’ve been asked whether the U.S. has become so over-regulated that this is what’s stifling the process. I don’t think we have any smoking gun about that, but I think it’s also naïve of us not to at least think about that.
You have to ask yourself, is it the case that young small businesses just aren’t doing what they should be doing because of the business climate? I don’t have any evidence to suggest that’s what is going on, but I think we ought to be asking the question, if for no other reason than to reassure ourselves that that’s not what is going on.
I do have some evidence, by the way, that the Great Recession of 2007-09 was especially hard on young businesses, and financial markets played a role there. We saw a long decline even before we hit the Great Recession; we were already down in 2006 relative to 1999. But then it fell enormously in 2008. I think I understand that a little better in the sense that in the states where there was the largest decline in housing prices we saw the largest decline in entrepreneurship. That makes sense for all kinds of reasons.
For one, there’s some evidence that these startups used a form of collateral home equity. So that’s part of it.
I also think part of what’s going on is that it was in those states where financial institutions got themselves into the most trouble. California was one of the states with the largest decline in housing prices and it saw a tremendous decline in the pace of entrepreneurship during the Great Recession.
One interesting question is whether as the housing markets and the local financial institutions, which are important for young small businesses, recover, will we indeed see some recovery? This isn’t about the trend but about the cycle. I don’t have enough data yet to be able to tell you whether that’s actually going on.
Is that your latest research project?
I have two projects that are closely related. One is just what I talked about, where we discovered the decline in entrepreneurship was especially large in the states with the largest declining housing prices. We’re trying to dig into that and understand why is it that house values play such an important role and are so highly correlated with entrepreneurship. The second project is again related. I’ve emphasized that this reallocation process is costly but productivity-enhancing. But there’s some theory out there that says that in recessions this process can get disrupted, especially if it’s a financial market disruption.
That’s because what’s normally happening is the less-productive businesses are contracting and shutting down while the more-productive businesses are expanding. But if financial markets break down, what might actually be the case is the businesses that are contracting and shutting down aren’t shutting down because they’re the low-productivity businesses but because they can’t get credit. We have found some evidence that that’s the case.
What exactly is it you’re trying to confirm or understand better?
The early 1980s recession was very tough but coming out of that recession the U.S. grew rapidly. From 1983 through 1986 there was very rapid job and productivity growth. That’s because the recession was a time when there was lots of restructuring and reorganization. You could say it accelerated that process of creative destruction.
However, in the Great Recession that didn’t seem to happen. The process of creative destruction, of reallocation, actually slowed down, and the reallocation that occurred was less productivity- enhancing. That’s the fact.
And you’re trying to understand why.
Thank you, John