During the latter half of the debate over the health care bill, you started to hear an intriguing argument about the effects of the bill.  Where conservatives claimed that it would stifle the economy, liberals countered that in fact, it would unleash the entrepreneurial power of people "locked into" their jobs by fear of losing their health coverage.  David Leonhardt rehearses this argument in his most recent column:


It's easy to look at the current debate and see an unavoidable trade-off between this country's two economic traditions -- risk-taking and security. But I don't think that's quite right. I think it is ultimately as misplaced as those worries about Social Security and Medicare equaling Bolshevism.

Guaranteeing people a decent retirement and decent health care does more than smooth out the rough edges of capitalism. Those guarantees give people the freedom to take risks. If you know that professional failure won't leave you penniless and won't prevent your child from receiving needed medical care, you can leave the comfort of a large corporation and take a chance on your own idea. You can take a shot at becoming the next great American entrepreneur.

With every previous major expansion of the safety net, history has had a chance to prove the naysayers wrong. It may yet in the case of universal health coverage. But the decision now seems to rest with the nine members of the Supreme Court.

This is the sort of argument I am fond of because it's so intuitively counterintuitive--it gives you one of those, "A-ha! Of course!" moments that are the main pleasure of  writing about economics.

But while that a-ha! reaction means a theory is plausible, it doesn't actually mean it's true.  And while I think that Leonhardt's narrative about freedom from fear is certainly one possible outcome of passing health care reform, there are also quite a few reasons to think that it might not be true.  I worked through a bunch of them earlier in the year, and still believe, as I noted then, that "the number of people who wish to start businesses, but are held back by the health insurance problem, cannot be zero."

But that is a far cry from saying that on net, creating this new safety net will help spur labor mobility and entrepreneurship.  Creating the safety net imposes costs, as well as benefits.  Some of those costs come in forms that discourage entrepreneurship.

Net cost or net benefit?  How do we know?

A lot of self employment is what economists call tax and regulatory arbitrage.  In plain words, companies often employ contractors because the overhead cost is cheaper, since contractors aren't subject to the same tax and work rules.  They can pay the contractors higher hourly rates, because their overhead is lower.

It's not surprising that people are more willing to be contractors or consultants when they have access to benefits through another company.  But this is not really the same thing as the entrepreneurship that Leonhardt is describing.  

Nonetheless, they tend to get conflated when people are looking for the effect of the safety net.  It's easier to measure "people who are working for themselves" than "people who are starting a company that will do something new and innovative and improve the economy's net productivity".  Tthe latter number, the one we really care about, is certainly much smaller than the former. 

Let's look at another attempt to quantify the problem: net jobs created.  Economist Scott Shane took a crack at calculating the number of people who will start companies that eventually employ someone else.  This is not a perfect proxy for entrepreneurship and dynamism--very small operations can add to an economy's growth, as I've written elsewhere.  But net jobs created is a pretty powerful way to approximate what Leonhardt is talking about.  Here's Shane's math:

Small Business Economy an annual publication of The Office of Advocacy of theSmall Business Administration explains that 60.7 million people have employer-sponsored health insurance from their employment (people covered by their spouses don't matter here because they don't face job lock). Therefore, 607,000 additional people per year would begin the entrepreneurial process if we eliminated health insurance job lock.

But not everyone who begins the process of starting a business manages to get one up and running. In fact, analysis of the Panel Study of Entrepreneurial Dynamics data by Paul Reynolds shows that a new business results from about one-third of startup efforts. So we will get about 200,000 new businesses if we can eliminate the job lock that comes from employer-sponsored health insurance.

This same research shows that only about 19% of new businesses employ someone other than the founder. Because people who leave jobs to start nonemployer businesses don't generate any net new jobs, it's the 38,000 additional new employer businesses that would be created if we eliminated the health insurance job lock that would be the source of any additional jobs.

Data from the Small Business Administration's Web site reveals that the average number of employees in a new employer firm is 5.6. Therefore, we will create an estimated 213,000 annually if universal health care eliminates the problem of job lock.

It actually might be slightly smaller than that--people currently covered by their employer might still have access to spousal coverage if they decided to leave (or so I read the data).

Meanwhile, as Shane points out later, the bill will also cost jobs--Doug Elmendorf, the head of the CBO, has estimated that PPACA will cause employment to fall by about half a percent, or about 700,000 people.  The job losses come from two sources: the taxes and penalties that make hiring more expensive for a company, and the Medicaid expansion and the subsidies making it more attractive to leave the workforce altogether.

Knowing that Elmendorf is a very smart and thorough man, I presume that he has already accounted for the positive effects of PPACA on entrepreneurship.  But even if he hasn't, that's a net loss of about half a million jobs.

Of course, maybe we're trading 700,000 terrible jobs for 200,000 awesome ones.  That could be a net gain.  But to me, we just don't have a lot of good evidence that generous social safety nets somehow spur labor mobility or entrepreneurship.  For every "just so" story you can tell about health insurance or generous welfare benefits making someone willing to leave their cushy corporate job in order to risk it all founding a company, you can tell an equal and opposite "just so" story about someone who doesn't bother to become an entrepreneur because it's so comfortable there on the dole, someone discouraged from pursuing a promising idea because taxes reduce the potential payoff or make it harder to turn a profit, or someone who isn't motivated to work harder because they're not so worried about security.

Looking at the Big Picture

In fact, probably both effects exist.  The prophets of dynamism tend to focus only on the studies that show the benefit side--mostly studies that show people are more likely to become self-employed if they have access to alternative health insurance, such as a spouse's insurance, COBRA, or Medicare.

But when we try to look at the net effect, it's hard to see much dynamism coming out of the places with a generous safety net.  Rates of entrepreneurship and labor mobility are far lower in Europe than they are in America--and one of the many factors restraining European labor mobility is thought to be the social safety net, both because of the difficulty of moving between benefit system, and because the benefits lessen the urgency of say, relocating in order to find a job.

Nor do US states show much of a pattern of dynamism created by generous benefits.  Unemployment in Massachusetts after RomneyCare passed does not drop (or rise) noticeably compared to the rest of the US.  High-tax, high-benefit states like New York don't seem to generate a lot of new businesses--in fact, New York routinely ranks last or next to last on surveys of business climate.  Then there are states like California, with high rates of innovation in entertainment and tech--but even before the state's budget crisis, it was losing manufacturing jobs and experiencing net out-migration to other states.  It's hard to know how to score that one.

And of course, there are other things going on in all these states, so it's not as simple as drawing a straight line from the government safety net, to the entrepreneurship rate--one way or another.  

Nonetheless, I'd say that if the safety net made a significant positive difference, we should see a clearer pattern than we do in state and national comparisons.  Instead, if anything, the pattern points in the opposite direction.  So I'm pretty skeptical that on net, PPACA is going to build us a better economy.

However, unless the Supreme Court strikes it down, we have an excellent test.  If the law really makes a different on net, then by late in this decade, we should see significantly rising rates of entrepreneurship and labor mobility, and at least a slight uptick in trend productivity growth and GDP.  I'll certainly be watching avidly.