Asymmetrical Information - Megan McArdle

06.05.13

Obamacare: Job Killer or Entrepreneurial Turbocharger?

A new report says Obamacare will help boost firm formation. Don't break out the champagne just yet.

Is Obamacare about to unleash a wave of entrepreneurship?  A National Journal article, based on a new report from the Robert Woods Johnson suggests that it will.  The researchers estimate that as many as 1.5 million new people may enter self-employment now that they can count on getting insurance.  If they're right, we could certainly use the boost: young firms are responsible for a disproportionate share of job creation, and they're also the engine of creative destruction in the American economy.  And since the financial crisis, new firm creation has declined.

But will it actually happen?  A reader asks, in re the National Journal piece "What should we all think about this?"

That it's certainly possible, in theory.  Obamacare is supposed to do away with what economists call "job lock": people who stay in a steady corporate-type job because of the health benefits.  Though it's hard to estimate how big the effect is, most of the researchers who have investigated the question have ended up concluding that it's substantial.  

I mean, it makes intuitive sense, right?  You can imagine a guy who's got a great idea for a startup, but whose wife has multiple sclerosis, making it imperative for him to maintain health coverage.  So long startup, then.  You can easily picture that guy waking up in 2014 and deciding that now that he no longer needs to worry about health insurance, it's time to take the plunge.

Unfortunately, I can also imagine that guy waking up in 2014 and realizing that it would be really risky to found a startup which might require him to work 16 or 18 hour days in the middle of one of his wife's health crises.  Thought experiments usually just end up validating the prior beliefs of whoever is constructing them.  

But what about the research?

The research is certainly suggestive.  But it's easy to overinterpret.  Most of the studies estimate the extent of job lock using one (or more) of three methods:  

Chronic illness:  if you or a family member have a chronic illness, are you less likely to own a business?

Spousal coverage:  If you have access to health insurance through a spouse, are you more likely to own a business?

Natural experiments:  When you become eligible for Medicare, do you become more likely to own a business?

All three measures are, unfortunately, flawed.  Not through any fault of the researchers; this is what we have.  But it's problematic.  

As I noted above, people with chronic illnesses have to worry about more than just health insurance.  If your spouse or child is sick, you're going to value employment stability--and having others around to pick up the slack--more than a healthy person with a healthy family.  At a job you've had for a while, you know that they'll tolerate the occasional emergencies.  They're invested in you.  Switching jobs, or striking out on your own, is very risky, simply because a new situation may not tolerate your needs as well as a group of people you've known for a while.  If you are the sick one, your illness may prevent you from working the kind of hours that a new enterprise requires.

A chronically ill family member probably also means that your spouse is less likely to have a steady and remunerative job, either because they are sick, or because they are caring for a sick family member.  (Labor force participation among the mothers of special needs and chronically ill children is shockingly low.)

Measuring spousal coverage involves a similar problem.  In America, "has access to employer health insurance" is strongly associated with "has a steady, well paying full time job".  You can see how having a spouse with a steady, well paying full time job might make it somewhat easier to leave your job and start your own business.  

I think the evidence from Medicare is strongest, albeit with a caveat: I don't think what we're seeing with Medicare is "entrepreneurship" as we commonly understand it, which is to say, founding a business at which you work full time, and eventually expect to employ others.  Although it's never too late--Colonel Sanders founded Kentucky Fried Chicken at 65--this is not a normal entrepreneurial profile.   Few seniors can (or want to) work "entrepreneurial" hours.  Everything I know about entrepreneurship, and retirement, tells me that what we're seeing in that discontinuity is people retiring, not leaving the workforce to become entrepreneurs.  The businesses they're starting are disproportionately part time ventures designed to keep their hand in, or provide a little cash.

We can also look at Massachusetts, where Romneycare gives us a preview of Obamacare.  And there the evidence is a little bit encouraging--entrepreneurship there was substantially higher in 2009-2011 than it was in 1999-2001.  But again, there are caveats.  First, what we'd really like to look at is the period from 2005 to 2008, not the wider interval where a lot might have happened*.  And second, while Massachusetts' performance was very good over the period--their 10 year increase put them in the top 20% of states--there were lots of other states that did just as well, without a statewide health insurance program.  

I think overall, there's reason to expect that some people will leave their jobs to start businesses as a result of Obamacare.  But I'd be pretty cautious about claiming that we'll see a big boost to overall entrepreneurship--1.5 certainly million seems implausibly large.  At least, for actual entrepreneurship.  The report's authors bundle entrepreneurship in with self-employment, and I'm entirely willing to believe that self-employment will rise that much, or even more, as employers seek to shed health care obligations by shifting to contractors from full time employees.  

One reason we might not expect the effect on actual entrepreneurship to be large is that people who currently have good employer health coverage already have the ability to buy health insurance when they leave to start a business.  After the Supreme Court's Obamacare cases came down last summer, I wrote that though I opposed the decision, it was ironically probably quite good for me, because I've got asthma and an autoimmune disease, which probably makes me uninsurable.  Virginia Postrel, a breast cancer survivor who buys insurance in the private market, wrote to tell me that I was wrong: under HIPAA, a law passed in the 1990s, even cancer survivors are able to buy insurance as long as they maintain continuous coverage.  In theory, we've already fixed this problem, so it's not clear just how much additional benefit we'll get from Obamacare.

Another reason is that people who have good insurance coverage often have other things--like mortgages--that make it hard to take risks with your salary.  

But the biggest reason is that while Obamacare may provide some incentives to start your own business, it will also provide some disincentives to the young and healthy, who are the most likely to start businesses.  We just made their insurance more expensive, and slapped a tax on them if they decide to go without, as a lot of young, healthy self-employed people evidently do.  How much does this matter?  Impossible to say.  But the marginal impact is unlikely to be zero.  

An even bigger impact is the effect on the decision to grow a company.  Once you hit 50 employees, you now have a big added cost for heatlh insurance.  For a high-tech startup that's growing like a bacteria colony (and likely already provides health insurance anyway), this probably doesn't matter.  But for more run-of-the mill businesses in retail, manufacturing, food services, and so forth, this could be a big barrier to expansion.

You can also, to return to the thought experiments, imagine a guy who gets laid off at 55 and goes to work building a business because he needs some way to buy health insurance for him and his wife.  And then imagine that guy realizing that since he can now buy cheap insurance through the exchanges.  Like I said, you shouldn't rely too heavily on thought experiments.  The point is that Obamacare is going to have a lot of different effects at once, and we don't have any good way to measure them.  Here's a starter figure, though: the CBO expects Obamacare to reduce labor force participation by about 800,000.  

So I'm not quite ready to predict a big entreprenurial boom.  Job lock is intuitively plausible, and there is some evidence for it, but it's far from proven.  

The good news, as I keep saying about all of these Obamacare conjectures, is that we're about to find out.  Think it will reduce mortality?  We should know by 2016.  Think bankruptcy will fall precipitously?  We should know by 2016.  Think it will boost the rate of firm formation . . . well, you know what I'm going to say next.  We can argue about whether Obamacare will be good for America.  But it is indisputably a great boon to social scientists.



* But isn't it suggested that it rose from 2000, which was the height of the dotcom bubble?  Actually, not really.  It rose across the US, not just in Massachusetts.