Economists and land use scientists across the political spectrum have long known that restrictive zoning is keeping millions of people away from housing and employment. It is one of the major barriers to increasing economic growth and promoting opportunities for the poor and disadvantaged. However, new evidence suggests that the problem is even worse than previously thought.
The work of economists Chang-Tai Hsieh and Enrico Moretti is perhaps the most influential in the literature that documents the effects of zoning on economic growth. Recently, my colleague at George Mason University, economist Bryan Caplan, discovered some significant computational errors in his seminal 2019 article, Housing Constraints and Spatial Misallocation. Hsieh and Moretti graciously acknowledged the mistake.
When scientists are wrong, it often results in their arguments appearing stronger or better supported than they actually are. In this case, however, the mistake of Hsieh and Moretti made the deleterious effects of zoning appear much less than it actually is. Bryan explains:
Where did HM go wrong? On pages 25-6 of their article, they write:
Assuming perfect mobility, the second row in Table 4 shows how the change in the regulation for housing supply only in New York, San Jose and San Francisco changes to that in the mid-US city. This would increase the total output growth rate from 0.795 percent to 1.49 percent per year – an 87 percent increase (column 1). The net effect is that below this counterfactual figure, US GDP in 2009 would be 8.9 percent higher, which translates into additional average wages of $ 8,775 for all workers.
On the next page, they reassess the results with incomplete mobility:
Table 5 shows that changing the regulation of housing supply in New York, San Jose, and San Francisco to that in the mid-US city would increase the rate of total production growth by 36.3 percent (second row). The net effect is that US GDP in 2009 would be 3.7 percent higher under this counterfactual factor, resulting in additional average wages of $ 3,685 for all workers or an increase in wage costs of $ 0.53 trillion.
Both tables show that HM covers the period from 1964 to 2009. How can these huge changes in annual growth rate, which add up over 45 years, lead to relatively modest changes in total GDP? Answer: You can’t!
The correct estimate from Table 4 is that GDP will be 1.0149 ^ 45 / 1.00795 ^ 45 = + 36% higher, not + 8.9% higher.
Similarly, the correct estimate from Table 5 is that growth will be 1.084% per year (0.795% * 1.363), so GDP 1.0108 ^ 45 / 1.00795 = + 14% higher, not + 3, Will be 7% higher.
Bryan also describes some similar errors elsewhere in the article.
The result of all is that the restrictive zoning reduces GDP many times what the original calculations by Hsieh and Moretti imply. And I would add that a high percentage of this loss is attributable to the poor and disadvantaged (who for obvious reasons, due to artificial supply shortages, are disproportionately represented among those who are priced out of desirable housing markets and the jobs available there through zoning).
I cite some of Hsieh-Moretti’s earlier estimates in my own work, including Chapter 2 of my book Free to Move: Foot-Voting, Migration, and Political Freedom. I plan to update as soon as I get the chance!
All of us who work on these topics have a duty to Bryan Caplan to discover this problem in the data and to bring it to our attention. In the United States at least, the exclusion zone is the greatest property rights problem of our time. It severely restricts the rights of millions of property owners. It is also one of the major barriers to increasing economic growth and fighting poverty.
Bryan’s discovery shows that the latter aspect of the problem is even worse than we thought. He is currently working on a book on housing and zoning, which I am looking forward to with great interest.
In recent years, important advances have been made in reducing zoning in several parts of the country. The Biden government has included some useful incentives for state and local governments to cut zoning into their otherwise mostly dire infrastructure bill. But much more needs to be done.