Recent election-season talk about economic growth and employment got us thinking again about topics which we return to regularly: job creation, productivity growth, and business dynamism, as well as the overall business environment these trends reveal, and what they tell us about the direction of the U.S. economy and its longer-term trend. We will lay our cards on the table at the outset and say that the trends suggest headwinds for U.S. economic growth moving forward, and sluggish growth will be challenging for stocks, particularly if they are too highly valued.
The GDP Picture
In 2014, former Treasury Secretary Larry Summers famously referred to an era of “secular stagnation,” and since then economists and pundits have been arguing about whether such stagnation is occurring, and if it is, what is causing it.
A simple graph of real U.S. GDP growth illustrates the phenomenon:
Data Source: Bureau of Economic Analysis, Wall Street Journal Economic Forecasting Survey
Looking at U.S. GDP growth from World War II to the present (starting in 1947), we see an annual mean growth of 3.5 percent from 1947 to 2000, and an annual mean growth of 1.8 percent from 2001 to 2015. Ultimately, that gap is the reason that middle-class voters are now sufficiently angry that they’re contemplating the election of candidates on both sides of the aisle who make their party leaders very nervous. That gap also illustrates in the simplest way what the “secular stagnation” discussion is about.
On the face of it, GDP growth in the U.S. has slowed significantly from its long-term postwar trend. We also note that for the first 53 years after World War II, economic contractions were followed by strong growth snap-backs, back to the trend or significantly above it. After 2001, the snap-back was weak, and after 2008, it was weaker still.
What’s Going On?
The question is why the shift has happened.
One major and often-discussed explanation focuses on public debt levels. Several economists have shown that above a certain threshold (about 90 percent of GDP in the work of Carmen Reinhart and Kenneth Rogoff), government debt begins to crimp economic growth. The size of the effect is debated -- the reality is not.
Our recent reading about “business dynamism,” though, has crystallized our thought about another extremely significant factor.
“Business dynamism” is one of the key characteristics of the U.S. economy that has long impressed analysts and economists. Joseph Schumpeter had the same phenomenon in mind when he coined the phrase “creative destruction” in 1942. It refers to the “churn” that the U.S. economy exhibits, with the rapid birth and death of firms.
That churn turns out to be particularly important for an economy’s long-term strength -- simply because new businesses turn out to be disproportionately important for job creation and productivity enhancement. This leads us to believe that lying behind the decline in trend GDP growth for the U.S. may be a decline in churn -- the births and deaths of businesses.
Until the past few years, government data didn’t track “firm age,” so analysts had to rely on firm size as a proxy (since new firms tend to be smaller than older firms), and emphasized how smaller firms are disproportionately responsible for job creation. Now, however, data on firms’ ages is available from the U.S. Census Bureau, and analysis of these data is revealing.
New Firms, Not Just Small Firms, Are the Engines of the Economy
Economic research shows first, that startups are extremely important engines of job growth. Startups account for less than 10 percent of all firms, but generate 20 percent of firm-level gross job creation. But all startups are not the same. MIT finance professor Antoinette Schoar distinguishes two basic types of entrepreneurs, “subsistence” and “transformational.” Subsistence entrepreneurs aim to provide employment for themselves, and possibly for a few family members, and are often motivated to be entrepreneurs not because they aspire to a high-growth business but for reasons of personal flexibility or ideals. Transformational entrepreneurs, on the other hand, start their businesses with aspirations for rapid growth and innovation. Their startups can be distinguished by the rapidity with which their workforces grow. Firms that expand their employment by more than 25 percent a year account for about 15 percent of all firms -- disproportionately startups -- but about 50 percent of gross job creation.
Not only do such firms provide an engine of employment growth, they are also substantially more productive than incumbents by a variety of measures, according to Lucia Foster, chief economist at the Census Bureau’s Center for Economic Studies.
Young firms are, of course, not just more likely to hire and grow -- they are also more likely to die. However, the U.S.’ long-term status as the world’s best example of business dynamism has rested on the fact that more businesses are born than die, and thus the “pool” is skewed towards younger, more productive businesses -- the kind that produce a big share of productivity improvements and new jobs. That also means that the whole ecosystem of young companies is an environment in which innovation occurs -- and even if young firms die, their innovations live on in the new firms that their founders and employees raise from the ashes.
However, the surplus of business births over business deaths was true only until recently. Our first clue: in mid-2007 deaths began to exceed births, and by 2010 the trend was continuing to deteriorate.
Source: Bureau of Labor Statistics (BLS)
The trend of business deaths is up… the trend of business births is weak, or down.
A consequence can be seen in the graph that follows, which charts trends of job creation and job destruction from 1980 to 2011.
Source: Decker, Haltiwanger, Jarmin, and Miranda. 2014.
“The Role of Entrepreneurship in US Job Creation and Economic Dynamism.” Journal of Economic Perspectives 28:3:3-24.
Both job creation and job destruction show a long-term decline, accelerating after 2001. This in itself signifies decreasing business dynamism, or “churn.” However, the acceleration after 2001 is primarily due to the fact that job creation began to decline more rapidly. That, in turn, means that the weakening representation of young, dynamic firms in the U.S. economy, which we saw in the BLS data above, may be responsible for exactly the negative effect on employment that we would anticipate.
What else can we surmise? We saw that dynamic young firms run by “transformational” entrepreneurs are the engines not just of employment, but of economic growth. It is not a stretch to believe that the shift in trend of U.S. GDP growth also correlates with the declining representation of younger firms in the U.S. economy. While we believe that employment trends and new business formation trends have improved since 2011 when the last data were analyzed, it is obvious that the U.S. has not recovered its pre-2001 GDP growth rate, and that in great part is a result of poor net new business formation.
Why the Decline in Startups?
So the data allow us to pinpoint a very interesting question: why are fewer transformational entrepreneurs starting new businesses?This question, we believe, is critical to any discussion of job creation, wage and productivity growth, and living standards for the U.S. middle class -- exactly the issues that are pushing voters of all political stripes towards disruptive, upstart candidates in 2016.
Last year, the center-left Brookings Institution published a white paper on “Declining Business Dynamism in the United States” (showing that concern over this issue is bipartisan, as it should be). Brookings drew on research from University of Maryland economist Ryan Decker and his colleagues; Decker ably summarizes an answer to the question of declining entrepreneurship:
“The trends suggest that incentives for entrepreneurs to start new firms have diminished over time… One possibility is that the business climate, broadly defined, has changed in ways that impede job reallocation — that is to say, by impeding entry, exit, expansion, and contraction. Moreover, if the cause of the decline is an increase in the costs of adjustment on one or more of these margins, this can imply adverse consequences for growth, productivity, and welfare… If an economy experiences an increase in adjustment costs for job destruction (for example, due to increased regulation), then not only will there be a decline in job destruction but also a decline in job creation (including a decline in startups) and ultimately in both productivity and welfare… The same logic applies to changes in regulations or institutions that affect the costs of starting up or expanding a business, including regulations that raise the costs associated with expanding beyond some threshold of size… If the more sluggish pace of adjustment is due to increasingly burdensome regulation and institutions, this has potentially large adverse consequences for intermediate and long-run U.S. job and productivity growth.”
In short, entrepreneurs are like all economic actors (no surprise): they respond to incentives and disincentives. Regulatory burdens are both reducing incentives to start new, transformational businesses, and increasing disincentives to do the same.
If this is true, the decline in business dynamism is not primarily a secular phenomenon, driven by some unavoidable long-term shifts in technology or demographics. Nor is it a cyclical problem that will inevitably right itself after reaching an unpleasant bottom. Rather, it is a structural problem, one that is tractable to policy changes. With renegade candidates making inroads on both left and right, it seems to be a year when the middle class -- who have borne the primary brunt of declining living standards that have resulted from declining dynamism -- will make their pain felt to the political elites.
In the long run, however, the outcome of one election is not what matters -- what matters is whether or not we see a business climate in which transformational entrepreneurs are being more incentivized to start businesses, in which case we could anticipate a revival of startups. Should current trends continue -- and there is nothing foreordained to suggest that they must -- we would expect to continue to operate in an environment with sluggish GDP growth, with all that implies for stock valuations and for investment strategy.
Investment implications: Since 2014, investors have been talking about “secular stagnation.” U.S. GDP growth shows a significant trend decrease when we compare the 1947-2000 and 2001-2015 time periods. One explanation focuses on the negative effect of rising government debt levels on GDP growth, demonstrated by several economists. We see another reason as well: declining business dynamism. Young businesses are decisive sources of job creation and productivity growth, and rates of new business formation have been in decline for some time, turning down more sharply after 2001. We think that incentives to start new businesses have been declining, and disincentives have been increasing: simply put, regulations have made it more difficult and expensive to start new businesses. As long as that environment continues, we will expect to see the trend of sluggish GDP growth in the U.S. continue. In that environment, overvalued stocks will tend to struggle and be vulnerable. If the trend continues, it will lead us to focus on cheaply valued stocks of companies with excellent growth prospects.