Max Tabarrok’s paper on Endangered Species Act and Housing (WP) has just been published in the Journal of Public Economics! Smart paper: Max noted that as soon as an animal is listed as endangered, developers face increased compliance costs and liability risk. But what is important to the experienced economist is that this increased control is not universal—it is limited to where the species live. Therefore, the ESA creates many environmental tests, areas where it binds and nearby areas where it does not and the list changes over time – there were 82 lists in 1970 and about 1500 today – and there are even listings that reduce regulation.
Here, for example, is a photo of the habitat (red) and control areas (blue) when the Northern Long-eared Bat was listed as endangered.
The lower left panel measures annual housing permits per 1000 treatment areas (red) versus the control (blue). At the bottom right are the coefficients for the event study. After the bat was listed as endangered, the number of new housing permits decreased in areas where the bats were likely to live compared to control areas.
Here’s what happens when the Peregrine falcon is delisted. Before the delisting, housing permits were lower in regions (red) where the falcon had habitat compared to control areas but after delisting treatment areas were held in control areas.
Overall:
…this paper provides evidence that some endangered species listings reduce the annual housing permit flow by 0.5 permits per 1,000 residents in 1980, about 10% of the average local permit flow. Accounting for wasted money and reduced costs, my estimates suggest the net effect of the ESA has been to reduce the national housing stock … about 6.3 million homes missing between 1980-2024, about 4 percent of the total housing stock in 2025.
Now, you might say, well this shows that the ESA has a cost. What about ESA benefits? It is difficult to measure the benefits, of course, or to know whether the ESA is working. But Max shows using satellite data that there are a few places where the ESA binds fill in development.
…at the deep end of housing production, new development often replaces existing buildings or fills space in an overdeveloped area that cannot host endangered species even if no new construction is done. In the deep end, the tradeoff with species protection is not binding, and it may be a good sum as it replaces the development of less green space. Therefore, to what extent the ESA hinders development in the extensive versus intensive range is related to the trade-off we face between housing production and species protection, and thus related to the overall welfare effects of the law.
In this section I extend the main specification of the paper to land use satellite data from the National Land Cover Database (NLCD) (Multi-Resolution Land Characteristics Consortium, 2025) and the diversity within the Building Permit Survey to examine where the impacts of the Endangered Species Act are increasing.
NLCD is a collection of satellite images of the United States compiled and classified in advance by the US Geological Survey. They divided the 30-square-meter pixels into one of fifteen land use groups, including four levels of development, three types of forests, and two types of wetlands. NLCD has annual files going back to 1985. I pass these pixels and map the permit areas to BPS using 2024 fixed boundaries, and track the changes in pixels within each area over time. The risk level of broad margin or green space development is measured by the flow of undeveloped pixels (eg, forests or wetlands) at any of the four levels of developed land use, divided by the total area of green land use.
He concludes:
The 15% most urban areas are responsible for 90% of the permit flow, and the area with the highest number of endangered species is outside these developed areas. The Endangered Species Act appears to limit unfilled development in these dense areas as it prevents green space development from outside sprawl (Table 9, Table 10, Table 11). Relaxing the Endangered Species Act’s legal approach to developed areas could increase the flow of permits to dense, energy- and land-efficient cities in California and the East Coast at the expense of sprawling areas in the Sun Belt, increasing both the supply of housing and habitat for endangered species.
The Trump administration is trying to limit the ESA, many lawsuits have already been filed. So Max’s paper is timely and points to a fix that would satisfy housing advocates again environmentalists: relax the ESA on non-filling and redevelopment in already built-up areas, where the exchange of housing against housing is not binding, rather than integrated everywhere.
Addendum: Obviously, I am glad to see this paper in print. Max started writing the paper before graduating—he just finished his freshman year. He was fortunate to get a lot of great advice along the way, especially from the excellent pre-doc he did at Dartmouth under Heidi Williams.
