One of the most striking and troubling features of today’s economy is the slowdown on innovation and economic dynamism. Former U.S. Treasury Secretary Larry Summers argues that the American economy has slipped into a period of low growth, low innovation, and secular stagnation.
A new study by Michael Mandel of the Progressive Policy Institute finds some evidence of a rebound in economic dynamism.
For one, he finds that, over the past 25 years, young firms—that is, companies five years old or younger—substantially outperform older firms in job creation, based on data from the U.S. Bureau of Labor Statistics. In 2014, the latest year data available, young firms generated 2.2 million jobs, compared to just 450,000 from firms that are more than five years old.
That said, the rate of job creation by younger firms has slowed since the Great Recession. From 2010 to 2014 it averaged roughly 1.9 million jobs per year, compared to 2.8 million jobs between 1996 and 2000.
The chart above shows the uptick in dynamism, as Mandel defines it, in the wake of the Great Recession. It charts the two-year growth in private sector business establishments for the leading tech hubs of San Francisco, San Jose, New York, Seattle, Boston, and Austin, the top 35 metros, and the rest of the nation. In all three cases the lines fall downward with the Great Recession, but then turn upward in recent years. As Mandel points out:
Between 2008 and 2013, the number of establishments in the top 35 metro areas grew by only 3 [percent]. Even worse, the number of private-sector establishments outside of the top 35 metro areas actually fell slightly, by 0.3 [percent].
Meanwhile, the major tech hubs—the San Francisco Bay Area, New York City, Seattle, Boston, and Austin—continued to grow, although at a reduced rate. Between 2008 and 2013, these tech hubs added 67,000 new establishments, more than half of the national total during that time period.
Starting in 2015 and continuing in 2016, establishment growth accelerated not just in the tech hubs and the large cities, but also in the areas that had been lagging. Indeed, most of the establishment growth gap between the tech hubs and the rest of the country disappears. Between 2014 and 2016, the areas in the U.S. outside the top 35 metro areas accounted for almost half of net new establishments.
The geography of startup hubs
Mandel also develops novel data on the nation’s leading startup hubs. To do so, he uses what he calls “organic data” on the startup economy from job postings on webpages and job boards such as Indeed.com. The study captures and organizes job posts that use the word “startup” as part of the company description for the United States as a whole and its 100 leading tech hubs, gathering data once in October 2016 and then in March 2017 and averaging the two.
Mandel’s startup index for metros is based on the percentage of job postings that contain the word “startup” normalized according to the median percentage across all metros. The table below shows America’s top ten tech hubs on this metric.
Unsurprisingly, San Francisco and San Jose (Silicon Valley)—top the list. Indeed, San Francisco proper tops the Silicon Valley on Mandel’s startup index, mirroring my own findings. Seattle and New York are next, followed by Boston and Austin, with Provo, San Diego, Chicago, and LA rounding out the top ten.
The next 25 leading metros on Mandel’s startup index are a more interesting bunch (see the table below). They include some well-known knowledge and tech hubs, including Washington, D.C., Atlanta, Denver, Salt Lake City, Portland, Dallas, and Raleigh in the North Carolina Research Triangle. But there are also Rust Belt metros like Detroit, Pittsburgh, Cleveland, and Cincinnati. Southern and Sunbelt cities like Phoenix, Charleston, Miami, and New Orleans also make appearances.
Based on this index, Mandel finds that job growth was strongest among the top 25 startup hubs, averaging 11.9 percent total from 2007 to 2016, compared to roughly half that for the rest of the top 100 startup hubs. Even taking into account the ”superlative economic performance” of the nation’s leading startup hubs, he also finds a positive correlation between his startup index and job growth. That holds even when he excludes the top ten hubs from his analysis.
Mandel sees this as evidence of the spread of startup hubs. He makes a strong case that restoring the United States’s economic dynamism and generating new good jobs rests on encouraging startups and building stronger startup hubs across the nation. He goes so far as to advocate for developing a startup policy to do so. His reasoning here echoes Steve Case’s intriguing “rise of the rest” initiative to strengthen startup hubs across the country.
While the spread of startup hubs would do much to bolster lagging regions and to take the pressure off the nation’s leading startup hubs, they run an uphill battle against some very strong and deeply ingrained economic forces. The clustering effects of talent and skill are what drive high-tech innovation and startups. Economists have documented this “great divergence,” and the same phenomenon is happening with the preponderance of high-skill, high-tech jobs clustered in a small set of U.S. cities and metro areas. The startup economy is spiky—and it appears to be getting spikier all the time. Behind the issue of economic dynamism and stagnation lies the even-more-pressing issue of geographic divergence and inequality.