On Friday the 13th, Barack Obama was campaigning in Roanoke, Virginia—the Star City of the South, hometown of Wayne Newton and, broadly speaking, myself—and said this:

If you were successful, somebody along the line gave you some help.  There was a great teacher somewhere in your life.  Somebody helped to create this unbelievable American system that we have that allowed you to thrive.  Somebody invested in roads and bridges.  If you’ve got a business — you didn’t build that.  Somebody else made that happen.  The Internet didn’t get invented on its own.  Government research created the Internet so that all the companies could make money off the Internet.

Quick! Guess the antecedent to "that"!

It's become a rorschach test for partisanship. If you don't like Obama, or the political philosophy he putatively represents, the antecedent is "business." If you're more temperate towards the idea of government, the antecedent is "roads and bridges." The Romney campaign immediately jumped on the phrase, generously using the red pen to put it in their preferred context. It's also been interpreted as "you didn't build your business, someone else did," by members of the media: "'If you've got a business, you didn't build that. Somebody else made that happen,' Obama said Friday…. [A]t its core — you didn't build that, somebody else made that happen — the president's remarks rang as an astonishing dismissal of the hard, often lonely work of American business people."

I was astonished, but more as someone who has to edit prose for a living. More interesting than the partisan point-scoring is the actual status of entrepreneurship in America. And as far as that goes, there have been a couple good reads recently. First, Ezra Klein:

A recent paper by the Kauffman Foundation — a nonprofit devoted entirely to encouraging entrepreneurship — looked at some of these issues, and they began with a fact that should force both sides to moderate their rhetoric: We don’t really know what leads to more firm creation. In fact, firm creation is something of a mystery, as it’s been eerily stable over the past 50 years, despite the radically different policy and economic environments we’ve had over that time.

That's not quite how I read the whole timeline (PDF):

[I]n the individual ten years included in the dataset from 1960 to 1978, new business incorporations averaged 323,000 per year. From 1979 to 1994, they doubled, averaging 640,000. It is entirely possible that the U.S. economy did indeed enter a new era of entrepreneurial activity, not only increasing in volume but also maintaining that volume over a certain threshold. There are good reasons to accept this story and, indeed, several changes did occur in the late 1970s that seem to have led to higher rates of firm formation, including the advent of the personal computer revolution and the rapid increase in venture capital investments.

In 1960, according to the authors' figures, 183,000 new businesses were incorporated. In 1970, 264,000, not a huge gain. In 1980, 532,000 were started, a doubling of that number over the decade. In 1990, it was 647,000; the growth of new businesses slowed, but the increase was maintained. Policy may have played some role in that, but there was also a substantial increase in the ratio of working-age population (15-64) from 1960-1980, which leveled off afterward. And, of course, there's only so much national-level policy can do for business creation; the local fights over food trucks and shared kitchens emphasize how much local regulations, particularly in large cities, effect entrepreneurs.

But "business incorporations" are only one way to measure entrepreneurship. At the Washington Monthly, Barry Lynn and Lina Khan used some more research by the Kauffman Foundation as a jumping-off point, looking into the creation of "employer businesses," ones with at least one paid employee, and adjusted for population growth. What they found was more worrisome than "eerily stable":

In 1977, Americans created more than thirty-five new employer businesses for every 10,000 citizens age sixteen and over. By 2009, however, Americans were annually creating fewer than eighteen such businesses, a 50 percent drop. While the Great Recession clearly cut into new business creation, the decline was clear well before 2007. The averages across decades capture that decline: between 1977 and 1989 Americans created more than twenty-seven new businesses for every 10,000 working-age citizens. This compares to fewer than twenty-five in the 1990s and around twenty-two in the 2000s.

[snip]

If anything, there’s good reason to believe that this decline in entrepreneurship is even steeper than government data shows, thanks to what appears to be systematic miscategorization by the government of what counts as a true independent company.

As an example, Lynn and Khan give people who would otherwise be employees, but are forced to become independent contractors.

The authors favor two major explanations: consolidation in the financial industry, and consolidation among large companies more generally. Lynn is particularly interested in the latter; in 2010, he discussed his book Cornered: The New Monopoly Capitalism and the Economics of Destruction with Chris Hayes:

Anti-monopoly law, which is really a better way to understand what antitrust is, is fundamentally a set of political laws. The purpose of these laws was originally to ensure that we didn’t have concentrations of political power—people using their control over something we need (e.g., grain, transportation, etc.) to enrich themselves and raise themselves up above their fellow citizens politically.

More recently, antitrust law has been used mostly as a tool to keep companies from price-gouging; an important part of the laws' purpose in protecting citizens from the effects of monopoly, but not, historically, the only one. Antitrust law was also used to keep large retailers like A&P from undercutting smaller competitors: "In November 1942, the Antitrust Division filed a Sherman Act case against the retailer, one section of which detailed how the A&P had used 'several turns of the screw' to coerce Ralston Purina into granting it a discount three and a half times what the cereal packer offered any other firm." This was meant to prevent the "waterbed effect," in which "better supply terms for powerful buyers can lead to a worsening of the terms of supply for smaller or otherwise-less-powerful buyers, which might then have an adverse consequence for consumers if downstream competition is lessened." If, say, you were running a small grocery store, antitrust law prevented A&P not just from undercutting you, but even passing on the actual costs of those cuts to you as well.

Lynn's argument will undoubtedly cause consternation—is it wrong to prevent a major retailer from getting whatever price it can? But it does raise good questions about what freedom and competition mean in the economy, what balance we want, and how much enforcement we want to get it. And it's more interesting than soundbite telephony.

Photograph: The White House