A few days ago The Times published a very interesting column by my colleague Ezra Klein about America’s peculiar lack of progress in the art of building things. Drawing on a recent paper by Austan Goolsbee and Chad Syverson, he noted that at least according to official statistics, we’ve gone a half century without any rise whatsoever, and maybe even a decline, in construction productivity — basically the number of person-hours it takes to build a house or other structure of a given size.
What makes this strange is that there have been many technological advances since 1970 that should have made it easier and cheaper to build stuff. But none of these advances seem to have paid off.
Klein suggests that the problem may be overregulation in the broad sense, that there are too many “veto points” where vested interests can block construction unless their demands are met. And he may well be right.
But his discussion had me thinking about a debate in economics that I’m old enough to remember taking place in real time: the attempt to explain the drastic economywide slowdown in productivity growth in the 1970s. This debate had a lot in common with the current discussion of construction productivity. And it also raises some questions about whether productivity is the right measure of economic success.
The Bureau of Labor Statistics has a nice graphic dividing up modern U.S. productivity growth into various eras:
Productivity grew rapidly for several decades after World War II, doubling in a generation. Then it slowed drastically for many years. The revival of growth after 1990 — probably driven by information technology — and its more recent stagnation are interesting stories too, but they are not my subject today.
The question is: What happened to productivity during that dip in the 1970s? One popular theory at the time, with some empirical backing, was that at least some of the slowdown reflected increased government regulation. The Environmental Protection Agency came into existence in 1970, and the Occupational Safety and Health Administration in 1971. Both imposed an array of new rules on businesses, and it’s not difficult to imagine that these rules had some adverse impact on worker productivity.
But does that mean that increased regulation was a bad thing? Not necessarily.
In 2020 the Bureau of Labor Statistics released a 50-year retrospective on OSHA, which contains among other things this remarkable chart:
It turns out that American workplaces in the early 1970s were very dangerous places by modern standards. And I don’t know about you, but a greatly reduced probability of getting injured or sick on the job seems like progress to me.
It is not, however, progress that shows up in measures of real gross domestic product, and hence in productivity data. So productivity numbers show only the costs, not the benefits, of safety regulations.
The same is true for environmental regulations. On bad days, New York City used to look like this:
It doesn’t look like that anymore. And the E.P.A. has done systematic studies of the costs and benefits of the Clean Air Act, which find that the benefits, many of them in the form of improved health, have greatly exceeded the costs.
Again, however, the benefits don’t show up in measured productivity, except possibly with a long lag (because healthier workers are presumably more productive).
So part of the productivity slowdown during the 1970s probably represented not so much a loss of dynamism as a shift in priorities — deliberate choices to make workplaces safer and skies cleaner, even at the expense of production.
Were these choices defensible? Definitely yes. Could the policies have been better applied? Of course — but when isn’t that true?
Now, I am quite willing to believe that the trade-offs on construction have been much worse than average, without equivalent social benefits to those 1970s policy choices. The problems with NIMBYism are huge and obvious, and they’re presumably part of a larger picture in which too many interest groups have the power to make construction difficult, even when those projects would be very much in the public interest. Nonetheless, it’s important to realize that making it easier for businesses to do what they want isn’t always a good thing.
And the broader lesson is that measured productivity isn’t the only thing that matters. What, after all, is the economy for? The goal is to improve people’s lives. This is often achieved by increasing gross domestic product per capita, but G.D.P. is an indicator, not an ultimate goal. We could have a bigger economy if we were willing to have filthy air and a lot more injured workers, but that’s not a trade-off we want to make.
The troubling resurgence of black lung disease.
America used to have safe roads. Not anymore.
Why did it take so long for information technology to pay off?
Are ideas getting harder to find?
Facing the Music
Honestly, New York air is fairly decent these days.