New Paper: Demand-Side Business Dynamism


By Mike Konczal, Marshall Steinbaum |

We—Marshall Steinbaum, who has recently joined the Roosevelt Institute as a visiting fellow, and Mike Konczal—have a new working paper out titled Declining Entrepreneurship, Labor Mobility, and Business Dynamism: A Demand-side Approach. We hope you check it out! We think it adds some important evidence on an unfolding debate. Here is a great write-up by Anna Louie Sussman in the Wall Street Journal this morning.

A lot of people have recently been focusing on a worsening economic phenomenon commonly referred to as the decline in “business dynamism” or “labor market fluidity.” There are a lot of ways to measure that, but it’s generally defined as the declining rate at which new businesses are formed and grow, or a decline in overall labor market mobility, which includes job transitions, quitting a job, and geographic migration for work. This has sparked a new literature and debate among people studying and reporting on the economy. Notable examples of coverage include Derek Thompson at The Atlantic Monthly, Ben Casselman at FiveThirtyEight, Will Wilkinson at VoxJason Furman and Peter Orszag, current and former senior economic policy officials in the Obama administration, the Obama administration itself, Ryan Nunn at the Hamilton Project, and many others.

Most of these analyses stress supply-side factors such as excessive occupational licensing, restrictions on building new housing, and regulations more broadly. However, recent investigations haven’t found evidence for these supply-side factors as drivers of the decline. Measures of dynamism that don’t require geographic mobility are also falling, so it’s not driven by housing. Goldschlag and Tabarrok (2015) found that industry-level regulations didn’t match with industry-level declines in fluidity, pointing away from a simple story about government regulation. Molloy et al. (2016) catalogued much of this, showing that the labor force age distribution, overall industry composition, zoning and housing requirements, and many other supply-side factors don’t explain the changes.

This paper provides an alternative explanation for the recent trends of declining entrepreneurship, falling labor mobility, and rising concentration of employment in old firms and large firms. Our explanation focuses on weakening demand, especially during the slow recovery from the last two recessions. That demand slowdown should, in turn, be investigated further, keeping in mind both secular stagnation and how power is shifting in favor of the owners and managers of incumbent firms alongside rising profits and inter-firm inequality. The key findings in this report are:

Wages Are Falling the Most Where Dynamism Has Fallen the Most.

If labor mobility or “dynamism” were declining due to excessive regulation and the increasing cost of job-switching or starting a new firm, then standard economic theory predicts wages and earnings would increase. But the data show that earnings have declined most where the declines in dynamism and mobility have been worst. This is true across metropolitan areas and industries as well as across new hires and all workers; in the chart below, each dot refers to a metropolitan area.


To step back and interpret what’s going on through the lens of an economic model of supply and demand, the supply-side story implies that the supply curve for labor has shifted left, at least in the increasingly regulated sectors and occupations. Supply shifts in the labor market would be expected to manifest as wages and employment moving in opposite directions. In other words, if the supply-side were driving these changes, we would expect to see wages go up and employment decline due to the scarcity of labor. Instead what we find is wages and employment moving together, which supports a demand-side story. (For those who want the more formal bells and whistles, we present a Diamond-Mortensen-Pissarides search model in an appendix, which formalizes this story.)

Those Who Do Switch Jobs Have Flat, or Even Declining, Wage Increases.

The supply-side theory also implies that wages for workers who do manage to switch jobs should be going up. If the market is defined by increasing barriers, say from an increase in required credentials and licenses, those who are capable of hurdling those barriers to take advantage of new job opportunities should see their wages bid up further. Instead, the data show that percentage wage increases for job-switchers have either stagnated or declined. There have been several estimates of this particular phenomenon, such as Wiczer (2016), and while results vary across datasets and specifications, none find an increase in percentage wage changes from job-switching.

Other evidence shows that employed workers are getting fewer offers to work at other firms in recent decades, which reduces their leverage to demand raises from current employers and leads to wage and earnings stagnation on the job.

If the problem were on the supply side, firms should be trying desperately (but without success) to recruit workers. But this wage stagnation applies even within employment matches because wages are renegotiated less frequently now and are less sensitive to outside job offers, which seldom occur. Other evidence, like Hyatt and Spletzer (2016), shows that the wage-tenure relationship within employment spells has flattened, meaning that spending longer in a given job no longer generates the wage gains it once did.

Dynamism and Fluidity Are Procyclical

The quit rate (or the rate of workers moving from one job to another) and the hiring rate from non-employment tend to move up and down in tandem. We look back to the 1990s, whereas the existing literature tends to start in 2000. We find that quits and hires trended positively in the boom years of the 1990s, before the 2000 reversal also highlighted in the business dynamism literature. In short, these labor mobility measures are observations on the state of the labor market, like the employment-to-population ratio, wage growth, and the labor share of national income. Other recent research has shown that businesses started in a downturn start smaller and remain smaller, even years later, thanks to deficient demand during their formative years (Moreira 2016). It’s unlikely that variation in the intensity of regulation over time followed exactly this pattern—coinciding perfectly with the Great Recession and its aftermath, for example.


This paper argues that the decline in mobility, dynamism, and entrepreneurship is a result of declining labor demand since 2000. When it is hard to find another job, employed workers stay at the jobs they have, impairing their ascent up the job ladder and the accompanying wage growth over careers that historically led to the middle class. Declining entrepreneurship can also be explained by workers’ reluctance to leave large, stable incumbents to start their own firm or to work at a start-up when they cannot be assured that they will have a more stable job to return to. Thus, we find that the concentration of employment in old firms and in large firms mirrors the timing of declining labor mobility due to declining demand.

Regardless of what you make of the merits of policy debates over occupational licensing, housing restrictions, and regulation, these items are not the driver of the aggregate decline in labor market fluidity. Our alternative analysis suggests future research should investigate potential policy-related causes of those trends in demand and market structure—such as declining effective marginal tax rates on high earners and a permissive environment for inter-firm mergers—that deemphasize full employment and market competition and enable secular stagnation.

We hope you check out the paper!

Mike Konczal is a Roosevelt Fellow working on financial reform, unemployment, inequality, and a progressive vision of the economy. His blog, Rortybomb, was named one of the 25 Best Financial Blogs by Time magazine. Follow him on Twitter @rortybomb.

Marshall Steinbaum is Roosevelt's Senior Economist and Fellow. Follow him on Twitter at @econ_marshall.