President and Chief Executive Officer
Federal Reserve Bank of Atlanta
Creative Leadership Summit
Louise Blouin Foundation
New York, New York
September 23, 2013
My thanks to the organizers of this leadership summit for inviting me to be part of this important event.
Today I want to talk about a trait that has characterized U.S. economic life, often with positive comparisons to other advanced economies. That trait is economic dynamism. This is a broad concept with dimensions such as innovation, entrepreneurship, new business formation, risk tolerance and risk taking, labor mobility, and job creation. The question I want to address is whether the economic dynamism of the United States is declining. Is America losing its economic mojo?
On this question, there are many opinions out there and no scarcity of anecdotes supporting those opinions. I will try to step beyond the world of anecdote and, to the extent possible, look at the evidence. I will explore what the data tell us about this question.
A couple of caveats up front: I won't be able to give comprehensive and exhaustive treatment to this subject in the few minutes I have this morning, so I plan to look at three interrelated aspects of the subject—job reallocation (job creation and destruction), productivity growth, and new business formation. And I won't offer a definitive answer. My remarks are intended to raise awareness and contribute to an ongoing conversation.
As always, I am not speaking for the Federal Reserve System or the Federal Open Market Committee (FOMC). I'll present my personal views. My colleagues may not agree.
The economic context for this discussion
I'll start with a very brief economic context for thinking about economic dynamism. As you may already know, the Federal Reserve pursues two main longer-run objectives through its economic policies: price stability and maximum employment. In the postrecession recovery period, the employment objective has been the more challenging of the two.
The recovery has been under way for more than four years now. Unemployment at its peak reached 10 percent. The official unemployment rate now stands at 7.3 percent—somewhat higher than the 5.3 to 6 percent range that many economists have historically referred to as "full employment." We've made a lot of progress, but there's a way to go before the Fed can claim that the maximum employment objective has been achieved.
Significant progress has been accomplished just in the past year. The official unemployment rate has fallen from 8.1 percent to 7.3 percent, and payroll employment has grown by about 2.2 million jobs. Payroll job gains for the past 12 months have averaged 184,000 per month, but recently there appears to have been some slowing. The monthly average for the most recent three months is 148,000.
This progress on the Fed's employment objective has been recorded in spite of rather weak economic growth. Over the 17 quarters of recovery since the recession ended, gross domestic product (GDP) growth has averaged a little above 2.2 percent. This is a lackluster pace of growth, and the progress we're enjoying on the jobs front makes sense only in the presence of low labor productivity growth. That, in fact, is the reality. Labor productivity growth is averaging significantly below historic norms. I'll return to this concern in a moment.
Employment dynamics and job reallocation
Let me introduce the concept of employment dynamics. Employment in America is in a constant state of churn. Workers, or potential workers, are entering and leaving the labor force. Employees are moving from one firm to another. Firms are changing the composition of jobs in their workforces. Firms are adding jobs in one category while eliminating jobs in another. All of this movement is a good thing, on balance. It is part of an economy that is continuously adjusting to shifting reality by reallocating resources from sector to sector, industry to industry, firm to firm, and job type to job type.
A particular type of churn is job reallocation. The rate of job reallocation across firms can be measured.1 During the 1990s, job creation at expanding firms and destruction at shrinking firms, taken together, typically amounted to about 16 percent of total private-sector employment per quarter. To put that figure in context, private-sector payroll employment today is around 113 million. A 16 percent job reallocation figure would amount to roughly 18 million jobs created and destroyed per quarter. This is a picture, clearly, of a very dynamic economy.
Since the early 2000s, however, the job reallocation rate has slowed. The most recent data for 2012 suggest that the quarterly reallocation rate is now closer to 12 percent, or 14 million jobs, per quarter.
Interpreting this apparent loss of dynamism across firms is not straightforward. It could be that these slowing dynamics are not a bad thing if we get the same growth of productivity we've enjoyed in the past without so much churning of jobs. But that doesn't quite fit with the low productivity growth we've experienced during this recovery.
At the level of the business firm, we observe similar trends in measures of internal employment flows. An expanding firm is one whose hires exceed its separations (meaning quits, retirements, and layoffs). A firm that is expanding its payroll must hire employees both to replace some separations and to fill the newly created jobs. A firm that is shrinking its workforce has more separations than hires. The data show that both the rate of hiring and the rate of separations have also been declining since the early 2000s.
Some of this decline reflects the rate at which people are voluntarily quitting their current jobs to do something else, including going to work for a different employer. On this dimension, too, there has also been a decline, with the drop most evident during the last recession.
Putting these various pieces together, we see a picture in which fewer firms are expanding employment, and each expanding firm is adding fewer new jobs on average than in the past. At the same time, fewer firms are shrinking, and each is downsizing by less on average. Fewer people are being laid off or are quitting their job, and firms are hiring fewer people. In other words, the employment dynamics of the U.S. economy are slower.
The low rate of hiring has contributed to an unusually weak flow of unemployed people into the workforce. Even though the unemployment rate has declined from 10 percent at peak to 7.3 percent in the most recent reading, the flow rate from "U" (unemployed) to "E" (employed) dropped from close to 30 percent a month prior to the recession to 16 percent during the recession and has improved only marginally over the past four years. Low job-finding prospects are almost certainly contributing to a larger share of unemployed workers leaving the labor market. While boomer retirement demographics are a partial explanation, this outflow of potential workers from unemployment is a factor behind the sharp decline in labor force participation we've witnessed in recent years.
I will mention one final dimension of employment dynamics, and that is a shift in the mix of jobs from full time to part time. The number of people who report they are working part time when they would prefer to work full time (what economists call "part time for economic reasons") rose sharply during the recession and remains high. You can conclude that the unemployment rate does not fully capture the true underutilization of labor resources today in our economy.
Job reallocation and productivity growth
I'd like to return to the subject of productivity growth, which is the second piece of the economic dynamism question as I've framed it. Recently, the rate of productivity growth has been significantly below its historical norm.
In my official economic outlook submitted a week ago as part of the FOMC's forecasting exercise, I have assumed we are experiencing a temporary spell of low productivity growth that will correct itself. I am assuming this will happen as demand kicks into higher gear and as businesses expand production somewhat faster than they expand their payrolls.
The likely direction of productivity measures in our economy is the subject of considerable debate. Another, more concerning possibility is that the slow advance of productivity reflects a fundamentally less dynamic economy. There are innovation pessimists, for example, who question the potential for transformative technological change in coming decades. I'll leave that for others in this conference to weigh in on.
At ground level, the productivity question is a bit of a puzzle, in my view. In many conversations with business leaders in recent years, I've heard accounts of lots of productivity-oriented investment—investment in software, automation, and robotics, for example. This would have resulted in enhanced productivity within a firm. But there is a between-firms aspect as well. Historically, the process of shifting resources from less-productive sectors and firms to more-productive activities has been an important driver of overall productivity growth. But—as already noted in my discussion of employment dynamics—resource reallocation appears to have been on a declining trend in recent years.
New businesses, employment dynamics and productivity
Finally, let me comment on the third component of my "mojo triad": new business formation. The American economy benefits from an extraordinary ongoing dynamic that is, well, dynamic. New businesses are launched. Some survive the start-up phase, but it's just as likely that they fail in the early years. Some of the survivors grow rapidly. Much job creation and destruction accompanies this process of start-up and failure or survival. Young firms account for a significant share of both job creation and job destruction. In that sense, young firms are a key factor in the cycle of job reallocation in the economy.
The vast majority of new businesses do not have highly ambitious growth aspirations. They start small and stay small. However, a few new firms desire to grow substantially, and an even smaller fraction of these will actually succeed and take off. These high-growth survivors generate a lot of new jobs and contribute to raising overall productivity, often by taking market share from less-productive incumbents and through product innovation.
The data on new business formation show a slowdown in recent years—paralleling the decline in overall job creation and destruction. The rate of new business formation took an especially large hit during the last recession and has been slow to recover. The data suggest that there has been very little growth in the number of businesses being launched since the end of the recession, and those that are starting are creating fewer new jobs on average.
Why is that? There has been a general atmosphere of uncertainty in the wake of the financial crisis and recession. Also, reduced access to start-up financing and tighter credit have constrained the flow of new firms. Until recently the collapse in real estate values all but extinguished an important source of start-up financing for many entrepreneurs. Similarly, tighter standards on personal credit cards limited another important source of start-up funds. This situation is improving as real estate values rise and credit standards ease somewhat. However, it's also possible that the sluggish pace of new business formation reflects other, possibly more entrenched factors such as regulatory barriers to starting a business. Demographics may also play a role. An older population is not as ambitious and risk-taking as one that is on average younger.
We at the Atlanta Fed have been trying to learn more about aspects of the entrepreneurial ecosystem that have been particularly important in fostering high-growth businesses. Borrowing a term coined by the Kauffman Foundation, my research staff is undertaking what we call the "Gazelle Project." We've been looking at the early-stage development stories of a number of successful fast-growing young businesses (so-called gazelles) in the Southeast. One of the interesting preliminary findings of this research is that knowledge of and access to high-quality talent, social networks, and local support resources are critical. That is, businesses start in a particular location for a reason—because entrepreneurs know the place has the key resources they need. It's encouraging that so many local communities have taken up the challenge of strengthening their entrepreneurial environments. I don't think it's an overstatement to say there is a nascent entrepreneurship movement underway led by community leaders.
So, against this backdrop of declines in the rates of job and worker reallocation, slower new business formation, and a subpar rate of productivity growth, is there a call to action? I think yes.
Can public policy do anything to foster greater dynamism? At a high level, I think public policy can do three things.
First, public policymakers can try to remove obstacles to growth and entrepreneurship—obstacles such as policies that discourage new business formation and disincentives to invest.
Second, the public sector can contribute with positive, pro-growth actions that address economic fundamentals, such as investment in human capital and critical, productive infrastructure.
And finally, I think there is a role for monetary policy. Monetary policy can help deliver appropriately favorable interest rate conditions that can promote a faster economic recovery, always in a context of low and stable inflation. Monetary policy can also play a critical role in creating the most favorable conditions for other policy actions to do their work.
The question I posed at the outset was whether the economic dynamism of the United States is declining. Is America losing its economic mojo? There is some evidence to the affirmative. I believe some of what we observe can be explained by the recent recession and frustratingly slow recovery. There are reasons to believe that some of the decline is cyclical in nature and will reverse itself over time. Anything more will call for creative leadership—the subject of this conference. My congratulations to the Louise Blouin Foundation for convening this important event.
1 The data referenced on job creation/destruction and hires/separations are measured at the establishment (business unit) level within a firm. Because some reallocation occurs across establishments within the same firm, the data overestimate the total amount of churn at the firm level. However, the vast majority of firms have only a single establishment, and the trends are similar at both the firm and establishment level.