How much does regulation crimp innovation? Not very much, according to a new study from the U.S. Census Bureau’s Nathan Goldschlag and George Mason University’s Alex Tabarrok. They find that “Federal regulation has had little to no effect on declining dynamism.” In other words, fewer businesses are starting up today than in previous years, but the authors don’t think federal regulations are among the major causes (see also Tabarrok’s summary over at Marginal Revolution).
That said, the authors are unsure of what else might be responsible: “The subsequent analysis will be unable to address the declining share of employment for young firms as evidence for the secular decline in dynamism and entrepreneurship (p.9).”
They base their regulation exoneration on a dataset called RegData, put together by analysts at the Mercatus Center (disclosure: one of whom is a former professor of mine). It is the best dataset yet devised for quantifying federal regulatory burdens. I’ve cited it before in some of my own work, and will very likely do so again. RegData works by counting the number of times the terms “shall,” “must,” “may not,” “prohibited,” and “required” appear in the Code of Federal Regulations. These individual restrictions are then broken down by industry and over time, going back to 1997. The total number of such restrictions currently in effect is more than one million.
But RegData has limits, and Goldschlag and Tabarrok have exceeded them. RegData counts the number of burdens, but does not estimate how much each one costs. These costs are over the map. One “shall” burden may be nearly costless, such as requiring a business to post a notice of local labor practices in the break room. Given the cost of printing posters and the minute or two of staff time required to hang it up every year, this may or may not cost a business a dollar per year. Another “shall” requiring power plant scrubbers may cost billions of dollars per year. Even though those rules both count as one restriction, they have very different costs.
RegData is state-of-the-art. But the art needs to improve its state before one can convincingly argue that the Code of Federal Regulations doesn’t harm economic dynamism.
For example, regulations tend to help incumbent firms, and give newcomers an artificial disadvantage. Tabarrok and Goldschlag agree with this point, finding that “job creation appears slightly positively correlated with regulation at the industry level (p.10).” And every worker who gets a job at an established firm is one who isn’t working at a startup. The labor force is fluid enough that this isn’t a zero-sum game, but regulations helping incumbent firms still have a crowding-out effect on startups. Regulations still hurt dynamism.
There are two points in Goldschlag and Tabarrok’s favor. One is big, and one is small. The small one is that federal regulations are only part of the story. They also acknowledge “other sources of regulation such as state legislation and judicial regulation through the common law (p.21).” We can add to this local-level and international-level regulations, as well as other countries’ domestic-level regulations that affect U.S. businesses.
Innovation does not necessarily respect boundaries. No doubt some companies offshore many R&D activities when domestic regulations make them too difficult. The companies then bring the fruits of those innovations to the U.S. later, so consumers still benefit, if more slowly. In these cases, the amount of innovation occurring inside the United States decreases, but the overall amount of innovation is only lightly affected, if at all.
The big point is barely alluded to in the paper. It is that the underlying institutions of social cooperation, market exchange, and dynamism are strong enough that federal regulation has, according to Goldschlag and Tabarrok’s analysis, so far been unable to squelch them. Just as a balloon pressed on one end pushes air to the other end, people will still find ways to cooperate and exchange with each other even when regulations push down on them. This inner strength of human cooperation is my great source of optimism, and Tabarrok draws on similar themes in his excellent 2011 e-book Launching the Innovation Renaissance.
Goldschlag and Tabarrok’s coda is less sensationalist than their lede, but probably more accurate: “global dynamism… may be increasing even as measured national dynamism decreases.” We shall see how things go, but on balance, look for human lives to continue to improve in the future. This will happen despite growing federal regulations, not because of them.