Controversy Over an Important Article Finding Large Negative Effects of Zoning

Chang-Tai Hsieh and Enrico Moretti's 2019 article, "Housing Constraints and Spatial Misallocation" is a much-cited and highly influential paper in the literature on the economic effects of zoning. It finds that exclusionary zoning restrictions in several major US metro areas had large negative effects on the economy. Between 1964 and 2009, the authors concluded, they lowered GDP to the point where it was 8.9% lower in 2009 than it would have been if these jurisdictions had only average levels of zoning restrictions.

In 2021, economist Bryan Caplan pointed out methodological errors in the authors' calculations, which—when corrected—showed they greatly underestimated the negative effects of zoning.  The authors acknowledged they had erred, and the apparent net effect was to further strengthen the case against zoning.

Recently, economist Brian Greaney of the University of Washington posted a critique the Hsieh-Moretti article in which he argues they made serious methodological errors that, when corrected, severely weaken their conclusions. Unlike in the case of Caplan's critique, this time the authors have not admitted error. Hsieh has posted a response to Greaney (drawn from his referee report on Greaney's paper), and Greaney, in turn, has posted a rejoinder. Housing policy expert Salim Furth has a helpful overview of the issues raised by Greaney (though he only briefly notes Hsieh's response).

This debate is ongoing, and it's not yet clear to what extent the Hsieh-Moretti article will end up being discredited. The issues involved are technical in nature, so I cannot easily summarize them here. Interested readers will have to read the article, critique, response, and rejoinder for themselves, to get a sense of what is going on. And, unfortunately, much of it is not easily accessible to those with no background in econometrics. Greaney's paper has not—so far—been published, and it's not entirely clear whether it will survive peer review, and what it will look like in its final form (assuming publication).

Nonetheless, my tentative judgment is that Greaney has at least raised serious questions about the article's model and use of data. At the very least, scholars and policy analysts should be more cautious in citing the Hsieh-Moretti study unless and until these doubts are resolved.

In addition, the fact that the article had already been shown to have one significant methodological error (the one Caplan found) should, in retrospect, have led more of us to wonder whether there were others.

I feel an obligation to emphasize these points because I am a longtime critic of  zoning and have cited the Hsieh-Moretti study in various writings of my own. I also highlighted the error identified by Caplan, pointing out how it strengthened the case against zoning. Intellectual honesty requires pointing out new analysis that cuts the other way, too.

The Hsieh-Moretti study is far from the only one that finds large negative effects of zoning on the economy, housing availability, and opportunities for the poor and minorities. I cited others in Chapter 2 of my book Free to Move. As Furth notes in his commentary on this debate, economists Gilles Duranton and Diego Puga have reached very similar conclusions to Hsieh and Moretti's in a study that avoids the possible errors Greaney argues undermine the latter.

Thus, I continue to believe that zoning has large negative effects, and that most restrictions on housing construction should be abolished. But I do have to admit the case for that position is weaker at the margin than it would be if there weren't serious questions about the validity of the Hsieh-Moretti article.