Establishments ≠ Firms

This article originally posted at the Updated Priors blog

I recently read a paper that took an innovative approach to at least partially answer a question that is boggling the minds of many economists and other observers: why has the firm formation rate declined precipitously the last three decades? I think it’s one of the most important topics in the economics profession today, and warrants a great deal of continued research in the coming years.

Aside from liking what was new about this paper, something else stood out to me—something I’ve seen before. The paper itself isn’t important here, so I’m not going to reference it explicitly—I’m not one to needlessly criticize someone else’s hard work, particularly when doing so isn’t central to the argument I’m trying to make.

The paper analyzes the relationship between, let’s say variable X, and the “startup rate,” defined as the rate of new establishment formations, not firm formations. This was done, presumably, because data on the former are much easier to obtain. But firms and establishments are not the same, and evidence suggests it has increasingly become important to distinguish between the two.

In research published last year by Mark Schweitzer, Scott Shane, and myself, we showed that the source of new business establishments is increasingly coming from existing firms, or what we call “new outlets”:

Figure 1: Distribution of New Establishments by Type (1978-2012)

Furthermore, because new outlets are generally larger than are new firms in terms of employment, economic activity at new business establishments (as measured by employment) occurs in no small part in new outlets:

Figure 2: Distribution of Employment at New Establishments (1978-2012)

This is something that is not occurring in isolation. The data show that this trend has happened in each broad industrial sector (9 SICs sectors) and across each state (data are not available at the metropolitan area level) during the last three and a half decades:

Figure 3: Share of New Establishments that are new Firms by Sector (1978-2012)

Figure 4: Ratio of New Firm Share of New Establishments in 2012 over 1978 by State

Among the sectors, the shift from new firms to new outlets as the source of new business establishments was greatest in finance, insurance, and real estate, the very broad transportation and utilities (which includes communications) sector, and in construction. Agriculture, retail trade, and mining saw the smallest changes over the 34-year period—but changed nonetheless.

For the states, the chart shows the 2012 new firm share of new establishments over that same share in 1978. Since the data are sorted by largest ratio at the top, states experiencing the smallest shifts from new firms to new outlets as the source of new business establishments are at the top, while those witnessing the most change are at the bottom.

In short, the importance of new outlets to the formation of new establishments has grown steadily and significantly during the last few decades, and this shift has occurred in each of the broad industry sectors and US states. As a result, studies that substitute new establishments for new firms—including when exploiting industrial and geographic differences—are increasingly using less precise estimates over time.

Even piecemeal immigration reform could boost the U.S. economy

This was originally posted on Fortune

(with Robert Litan)

With President Obama expected to make a statement about immigration tonight, Washington is gearing-up for a fight over the President’s seeming willingness to exercise his executive authority to prevent the deportation of primarily low-skilled immigrants. While that’s worth watching, the more important economic picture risks getting lost: the impact that immigrants can bring to the American economy in the long-term. It’s the immigration discussion we ought to be having, and if the rumors are correct as of this writing, it’s the one the President will at least partially extend in the executive action that he will outline. 

High-skilled immigrants are good for America, and we should encourage more of them to come here given recent trends in entrepreneurship, where more firms are dying than being created every year. But high-skilled immigrants could help turn that trend around -- they are twice as likely to start businesses as native-born Americans. This is especially true in high-tech sectors, where immigrants are not only more likely to start firms, but also to patent new technological discoveries. Giving green cards to foreign students completing STEM degrees at U.S. universities, and to many more immigrant entrepreneurs, would increase income and employment opportunities for American workers across the board. 

We have recently published new evidence at the Brookings Institution that supports the link between immigration reform and economic growth. Our research brings a new perspective -- the importance of entrepreneurs -- to an older idea: the link between greater population growth and economic expansion.

This connection was at the heart of concerns expressed by Harvard economist Alvin Hansen in his address, “Economic Progress and Declining Population Growth," before the American Economic Association in 1938. Hansen worried that what he saw as falling rates of population growth and technological advance portended a slump in investment, which would lead to persistently low employment and income growth. Hansen’s forecast never came true, thanks to post-war booms in innovation and fertility rates.

Three-quarters of a century later, another Harvard economist, Larry Summers, has updated Hansen’s concept of “secular stagnation” to describe post-recession slow growth across much of North America and Europe. In the year since Summers first made his remarks, economists have been debating whether and to what extent his thesis will hold in the future. 

Much less publicized is another debate that concerns the current and future state of U.S. economic growth—that of a pervasive decline in the rate of firm formation and economic dynamism. We and others have documented this decline in a broad range of sectors and regions throughout the United States during the last few decades—a decline that even reached the high-tech sector and so-called high-growth firms, the small group of (often young) businesses that are responsible for creating the bulk of U.S. jobs.

And the U.S. wasn’t alone—this decline was observed in other advanced economies of the OECD—suggesting that large global factors are at play.

This evidence runs counter to the narrative that an entrepreneurial renaissance is sweeping the globe, and the seeming endless technological change disrupting the economy. We also see plenty of those signs around us, but their impacts haven’t yet been reflected in the data. Even if they were, their effects would be following more than three decades of persistent decline.

Our most recent Brookings research suggests that slowing population growth hurts entrepreneurship. This was particularly true beginning in the 1980s in America’s West, Southwest, and Southeast regions of the United States—once places with the highest rates of new firm formation as population surged in the 1970s. During the three decades that followed, however, the formation of new businesses fell partly because population growth slowed.

In other words, population growth matters.

George Mason University Economist Tyler Cowen recently warned that the “relatively neglected field” of population economics could hold answers to the period of slow growth facing much of the developed world. For Cowen, one solution is obvious: absorb more immigrants. We couldn’t agree more.

Given the declining rate of growth of native-born Americans in the decades ahead predicted by official federal forecasters, the only other sure way to boost our work force is through more legal immigrants. Welcoming more foreign-born workers makes both economic and political sense, and in an ideal world, Congress and the President would agree on a comprehensive reform package. Since that’s not likely in the cards for now, let’s at least begin with high-skilled immigrants, who can help reverse our nation’s falling startup rate and provide a boost to our innovative capacity.

The Aging of American Businesses

This article originally appeared at Harvard Business Review

The year was 1995.

Tom Hanks was awarded an Oscar for Forrest Gump, Coolio’s “Gangsta’s Paradise” was number one on Billboard, and a young undergraduate named Monica Lewinsky began a summer internship at the White House.

Elsewhere, Timothy McVeigh murdered 168 people in Oklahoma City, three years of war in the Balkans came to an end, O.J. Simpson was acquitted of double homicide, and the World Trade Organization was formally launched on New Year’s Day.

A lot has changed since then—the iPhone, Tivo, the Toyota Prius, Google, Facebook, YouTube, human genome sequencing, GPS navigation, Skype, mobile broadband, just to name a few.

And yet, despite all of that change and memories of historic events that now seem ever distant, America’s business sector might be less dynamic than ever. That’s the major takeaway of research I co-authored with Robert Litan of the Brookings Institution and published this week. The evidence is pretty overwhelming.

Our research documents a steady rise in economic activity occurring in mature firms during the last two decades, and declining in new, young, and medium-aged firms, or what we dub “The Other Aging of America.” Mature firms (those aged 16 years or more) comprised 34 percent of U.S. businesses in 2011—up from 23 percent in 1992, for an increase of half in just under two decades.

The situation is more pronounced with employment. By our estimate, about three-quarters of private-sector employees and nearly 80 percent of total employees (private + public) work for organizations born prior to 1995. This is especially remarkable considering the volume of product innovations and household-name businesses that have emerged in the last two decades.

Further, we found that the aging of the business sector during this period has been universal across the American economy—occurring in every state and metropolitan area, in every firm size category, and in each broad industrial sector.

The evidence suggests that a decline in entrepreneurship is playing a major role; perhaps the largest. As we and others demonstrated in previous research, the rate of new firm formation has fallen by half during the last three decades, and has contributed to the decline of American “business dynamism”—the productivity-enhancing process of firm and worker churn. Fewer firm births means fewer young and medium-aged firms. It’s a matter of simple arithmetic.

Compounding this, we document a sharp uptick in early-stage firm failure rates and believe it might be playing an increasing role over time. The failure rate of firms aged one year—the youngest firms in our data outside of freshly launched ones—increased from a low of 16 percent in 1991 but rose steadily and persistently to reach 27 percent by 2011. It is possible that the increased likelihood of failure in the first year is holding back would-be entrepreneurs from forming businesses at all. In the last decade, failure rates have also increased for all firm age categories except for one—mature firms, where failure rates have held steady.

In short, fewer firms are being born, and those that are born are increasingly likely to fail very early on, as are firms that survive into young- and medium-aged years. Those that are old, on the other hand, tend to persist, allowing them to constitute a larger share of economic activity in the United States over time.

Somewhat surprisingly, we were unable to find evidence of a direct link between business consolidation and an aging firm structure. Though we do find a substantial rise in consolidation during the last few decades—confirming the widely suspected belief—it doesn’t appear to be a major contributor to business aging specifically, which has been occurring across all firm size classes, and most of all in the smallest of businesses. If business consolidation were a driving factor, we wouldn’t expect this to be the case.

This leaves some questions unanswered and some future areas for research—most notably the cause of declining business formation. But whatever the reason, our research clearly establishes that it has become increasingly advantageous to be an incumbent, particularly an entrenched one, and less advantageous to be a new and young competitor—regardless of business size, location, or broad industry group.

Declining Business Dynamism: It's For Real

This originally appeared on the Brookings Institution website

(with Robert Litan)

Two weeks ago, we wrote a paper for Economic Studies at Brookings that documented the decline in business dynamism in the United States during the last few decades. We of course feel that this is an important issue, but it would be an understatement to say that we have been surprised at the level of interest our work has received.

While we’re happy to have invoked a spirited debate, we also don’t agree with some of the analysis we’ve seen of the research. We’d like to use this opportunity to respond to some of it.

What We Found

To recap, our report showed a persistent decline in the rate of new firm formations and in the job reallocation rate—a broad measure of labor market churn that results from firm formations, expansions, contractions, and failures (what we and other economists call “business dynamics”). We also showed that these declines were nearly universal across the U.S. states and metros during the 30 year period between 1978 and 2011, as well as across a broad range of industries and firm sizes.

In short, the decline in dynamism and entrepreneurship doesn’t appear to be isolated to any one segment of the economy or region, but instead has been a widely shared experience.

What we did not advance was a cause. Getting at why individuals are choosing not to launch their own companies and instead opt for employment at incumbent firms is an inherently tricky proposition. But we, and others, are making advances at getting to an answer, and hope to have some new findings to report in the future. For now, it is important to document what’s happening—the response from our report seems to confirm that.

Unfortunately, not advancing a cause also left the door open for a lot of speculation as to what is driving the decline in business dynamism—some entirely valid, some without merit in our view, and some in-between. We address in the next sections several of the more important questions or critiques.

Retail and Services

One criticism is that the decline in dynamism has a lot to do with the combined effects of (i) a shift of economic activity and employment into retail and services, and (ii) the well-known consolidation of firms in those sectors. In other words: our report simply reflected already well-known declines in firm formation and job reallocation in the sectors of the economy that have experienced the most growth. So why worry?

This critique oversimplifies. For one thing, it ignores the fact business dynamism and firm entry are both down across each of the broad industry groups, as shown in Figure A2 from our report (and reprinted here as Figure 1). Put differently, the decline we documented is not limited to retail and services.

To amplify our point, consider some additional data from the same Census Bureau series that we used in our initial report, which allow us to calculate firm entry and job reallocation rates for the retail and services sectors versus the rates for the remaining sectors as a group.

As shown below in Figure 2, the firm entry data tell two stories. First, excluding the retail and services sectors pushes the firm entry rate lower, not higher as the critique seemingly would imply. This is especially true when retail firm entrants are excluded, since the entry rate in this sector is above average throughout the three-decade period we examined. In contrast, the entry rate for services largely tracks or slightly underperforms the overall rate until around 2001, after which it increases as the non-Retail and non-Services sectors display linear declines (in fact, this decline starts in the mid-1990s).

The patterns for job reallocation rates are slightly different. As shown in Figure 3 below, the drop in the job reallocation rate (partially a function of firm entry), is significantly more pronounced for Retail relative to Services (before the late-1990s) and relative the rest of the economy prior to the early 2000’s. From there, all trends move down in a similar manner.

Services on the other hand, perform about the same as the job reallocation rate overall. The pattern in this part of the economy suggests that the decline in dynamism in Services is less driven by firm entry than is the case for other sectors.

So, we’re back largely where started and initially reported.  It is true that retail and services constitute a greater share of the economy than in the past, which means any decline in dynamism in these sectors would drive down the overall rates of dynamism and firm entry disproportionately. But the evidence is more complicated. Both Retail and Services pushed up the firm entry rate, while for job reallocation, Retail pushed it down significantly while Services was about average.

In a University of Maryland working paper, economists there and from the Census Bureau took a closer look at the decline across sectors. While they make stronger comments about the decline in Retail and Services than we do, even they conclude:

“But we note that even in those sectors, we cannot account for most of the decline by taking into account simultaneously the interaction of firm age, firm size, detailed industry, geographic location and indicators of the firm operating in multiple locations. Ultimately, most of the decline in these sectors is in the “unexplained” within component.”

In other words, declining dynamism is real and the causes may be difficult to detect, but the spread of large retail chains like Walmart, Starbucks, Costco, and Walgreens does not contradict our basic finding.

High-Growth Firms

A second criticism has been that by focusing on entrepreneurship in all sectors, we aren’t separating the one’s that “matter” from those that “don’t.” In other words, by including the mom-and-pop shops (predominantly in retail and services, but also in areas like professional services and construction) along with high-growth potential firms, we are masking the true amount of entrepreneur-driven innovation occurring in the economy—the type of entrepreneurship that will lead to income growth and job creation.

While we disagree with normative judgments about which types of entrepreneurs matter and which ones don’t, there is evidence about the importance of high-growth firms. In fact, a very small percentage of firms account for the substantial majority of new jobs created in a given year.

The problem is that it is very difficult, if not impossible, to know at the time of founding whether or not firms are likely to survive and/or grow. This is true even with venture-capital backed firms, which are presumably the ones some of our critics believe that matter, but which in fact are a tiny fraction of all firms launched in any year. Conversely, research conducted at the Kauffman Foundation (where one of us used to work) has shown that the large majority of firms that do grow rapidly – making it to the Inc 500 list of fast-growing firms in any given year – never received VC money.

The key point about the entrepreneurship rate is that it represents the commercial equivalent of hockey’s “shots on goal.” The more firms started, the more rapidly growing firms are likely to emerge. For that matter, and to use another sports analogy, the more at bats we get the more doubles and triples will be produced –not just the home runs. That is a major reason why the secular decline in the new firm rate is so worrisome.

Policy and Politics

We also want to caution against those who might use or have used our findings to advance a particular partisan political agenda.  

First, we have seen a number of comments in the conventional media and social media that somehow the fact that firm exits now exceed firm births, or that the decline in firm entry generally, is President Obama’s fault.  We understand in this highly partisan environment, the temptation to advance such claims, but we do not believe they have merit. For one thing, it is doubtful that Presidents have much to do with the rate of firm entry. But even if they did, our data span a period of 30 years, and include periods when Presidents from both political parties have governed. And yet dynamism has continued to decline over the entire period.

Second, it is difficult to pin the blame for the secular decline in dynamism on high personal income tax rates (which matter for many startups, which increasingly are pass-through entities), either at the federal or state level. At the federal level, the top marginal tax rate has fallen since the Reagan tax cut of 1981 (the beginning year of our three decade span), and yet business dynamics have declined. Meanwhile, at the state and local levels, it is difficult to detect from the data we looked at that differences in state and local taxes have accounted for declining dynamism, since the decline has occurred in every state and virtually every metro area.

Third, we have speculated in comments since our report that mounting regulation – from all levels of government – could be one factor frustrating job reallocation while tilting against entrepreneurship. Younger, smaller firms do not have the resources that larger, more established firms do to hire full-time attorneys or compliance officers, which should put them at a progressively larger competitive disadvantage as regulations continue to grow in number and complexity. In other words, we think a more relevant conversation isn’t a generic one about regulation per se, but one that considers if our regulations systematically disadvantage entrepreneurs vis-à-vis incumbents and larger firms. We want to stress that this “cause” for declining dynamism is only a hypothesis at this stage, and awaits confirmation by future researchers, but it seems to be a plausible contributing factor.

Fourth, notwithstanding the possibility that mounting regulation in general may be one factor driving the decline in dynamism, it is also possible that the much-criticized Affordable Care Act could eventually help to begin reversing the secular decline in firm formation. This is because “guaranteed issue” of health insurance without discrimination on the basis of preexisting conditions, in principle, could loosen “job lock” that can inhibit employees with ideas for new businesses from leaving more established firms to take the entrepreneurial plunge. It will take several years, however, to see whether the data confirm or refute this plausible hypothesis. In addition, any benefits the ACA may entail in this regard say nothing about other critiques of the Act and proposals for how it may be improved.

Finally, as we noted in our initial report, the lowest hanging “fruit” for reversing the decline in startups is to increase the numbers of permanent work visas for immigrants coming here to earn advanced degrees in STEM fields and to establish businesses in the United States. Such measures would build on the well-established fact that immigrants generally are more likely to establish businesses than native-born Americans. Importing more individuals with technical knowledge, in particular, should help ensure that many of the new businesses that immigrants launch here introduce the kinds of new scientifically advanced products, services, and methods of production that are especially likely to enhance productivity growth, and thus faster growth in U.S. living standards.

With Fewer New Firms, the High-Tech Sector is Losing its Dynamism

This article originally appeared in Harvard Business Review

Believe it or not, America’s high-tech sector has become less dynamic and less entrepreneurial in the last decade. That’s the key takeaway of a recent Kauffman Foundation report I co-authored.

Despite the fanfare this vital segment of the economy and its start-ups have received in recent years, the high-tech sector is experiencing a consolidation of activity away from young firms into more mature ones, and the pace of job creation has been on a persistent decline. While it’s true that high-tech companies have been well-represented among the fastest growing firms in the past few years, the high-tech sector–like the rest of the economy–is less dynamic overall.

What do I mean by “dynamic”? The study of business dynamism involves measuring the flows of firms and workers underlying the private economy. Businesses are constantly being formed, growing, shrinking, and closing. Labor markets reflect this churning: some jobs are created while others are destroyed, and some workers move into new roles as others seek to replace them. New and superior ideas replace existing and inferior ones, while more productive firms usurp less productive ones.

A particularly important component of this dynamic process is the entrepreneur, who starts a venture to create a new market or to replace incumbents in an existing one. Entrepreneurs also play an outsized role in new job creation. While older and larger firms account for the substantial majority of employment levels, new and growing young firms drive net new job creation overall.

The process of business and labor market churning is a messy one. But it’s also fundamental to modern economies. Research has firmly established that this process of “creative destruction” fuels productivity growth, making it indispensable to our sustained economic prosperity. In other words, a more dynamic economy is a key to higher growth.

But business dynamism is breaking down.

Forthcoming research from economists at the University of Maryland and the Census Bureau shows that business dynamism has been declining across a broad range of sectors during the last few decades–and the single biggest contributor is a declining rate of entrepreneurship. A host of indicators point to a workforce that has become more risk-averse, and therefore less likely to change jobs or start a new venture.

I recently teamed up with two authors of the aforementioned research to produce the Kauffman report, John Haltiwanger of the University of Maryland, and Javier Miranda of the Census Bureau. We surveyed how these trends might apply to the high-tech sector, looking at data through 2011 and using a broader definition for high-tech that stretches beyond software and Internet companies to include things like computer hardware, life sciences, aerospace, and scientific research. What we found surprised me.

Though the high-tech sector was particularly dynamic and entrepreneurial during the 1980s and 1990s–a period when the same was not true across the economy–all that changed in the 2000s. The job creation rate (representing expanding firms) has been on a sharp decline since the beginning of the last decade, while the job destruction rate (representing contracting firms) has held about steady–squeezing net job growth in the process. By 2011, the rate of overall labor market churning in high-tech had converged with the rate for the total private sector.

Even more striking was the declining entrepreneurship. Young firms that I’ll call “start-ups”– those aged five years or less–comprised 60% of all high-tech firms in 1982. That figure fell to 38% by 2011. About half of this decline took place after the dot-com bust dissipated. The decline in both entrepreneurship levels and rates during the period associated with the Great Recession were sharper in high-tech than for the rest of the economy.

A decline in high-tech dynamism might be particularly problematic for future growth. Aside from the direct impact on productivity this sector has on technology-adopting segments of the economy, the high-tech sector itself plays an outsized role in income, employment, and productivity growthoverall. Of the job-creating young firms, high-tech start-ups are particularly dynamic–growing at twice the rate of a typical young business, and high-tech firms account for an outsized share of America’s fastest growing businesses.

How does this analysis square with talk of a “tech bubble”? One answer is that this activity, which is concentrated in web and mobile, represents only a subset of the broader high-tech sector. There’s good reason to believe that the last two years have ushered in a new wave of these typically leaner start-ups, and early-stage VC and angel investment data would back that up.

On the other hand, it may genuinely reflect slower growth across the broader high-tech sector. That appears consistent with one recent analysis that shows high-tech job growth slowing in 2013 after a rapid expansion the two years prior. Taken in the context of a longer-term decline, segmented green shoots wouldn’t reflect a robust recovery.

While this work may pose more questions than answers, the results make it clear that we must do more to promote a dynamic and entrepreneurial high-tech sector right now.