A group of LSE economists have published a paper which makes a strong effort to actually work out the damage regulations do to economic efficiency. Luis Garicano, Claire Lelarge, and John Van Reenen hit upon the method of looking to France, where there are sharp increases in the regulatory burden when firms employ 50 or more workers.

Seemingly as a result of those burdens, a noticeable number of firms appear to "stick" at 49 employees even when they may hire that extra marginal person. The piece's key chart is a killer:

Notice how the number of companies with 49 employees is actually higher than the number with 45 – and also that there's a precipitous drop between the number with 49 and the number with 50.

The paper tries to estimate, to a preliminary level, the regulatory cost of this burden. They estimate that 0.05 per cent of firms are distorted, and that the total output lost by those distorted firms is about 35 per cent – meaning that GDP is lowered by 0.5 per cent.

As the New York Times' Casey Mulligan writes, however, we shouldn't confuse the direct cost with the total cost of the regulations:

This is not to say that the regulations imposed on 50-employee companies are necessarily excessive, because they can create public benefits that more than justify their net costs for an employer and his employees, just as taxes and government spending can. For example, an air-pollution regulation might kick in at 50 employees that creates a significant cost for the employer and little aggregate benefit for his employees but creates a significant benefit for the people of France.

The employers also miss another transfer of wealth that might be just as important. Matt Yglesias covered it in another context last week:

One very plausible consequence of this would simply be to strongly discourage the owners of small firms from pursuing growth. And the big winners from that kind of disincentive to firm growth will be the owners of other small firms that simply aren't as lucky or well-managed as the growing ones.

In other words, it's a transfer of wealth from a company which may expand by a few employees to the companies which aren't going to have their lunch eating by a growing competitor. In the French context at least, that transfer will be relatively small. While companies are disinclined to grow from 49 to 50 employees, they may well be happy to leapfrog from 49 to 51 or higher; and the impact on creation of massive companies will be small indeed. But it will have an impact.

Fundamentally, it's an example of why it's best, where-ever possible, to target margins rather than absolutes. In taxation, for example, there's rarely this sort of adverse incentive, because there are few margins where earning more affects the taxes you pay on money already earned (where that does happen – between £100,000 and £150,000 of income, for instance – it is still carefully planned so that there is a positive value for every extra pound earned).

The problem is that that's harder to do for regulation. You can't really tell a company that they have to provide health insurance for the 50th employee but not the first 49, for instance. It would be unfair, not to mention probably even more impractical.

Better answers may be to more gradually phase in the burdens, so that at no point is there a leap in regulation big enough to dissuade too many companies from expanding; to stop fetishising small businesses, and make them subject to the same regulations as every other company (which would also force regulations to be easy to comply with, of course); or to make such regulations more explicitly support entrepreneurship rather than merely being small by imposing them a set period after a company has been founded, rather than basing them on growth.

More research, please!