Does Upzoning Reduce Housing Prices?
A recent careful study says no, but it suffers from unavoidable data and conceptual problems.
A new study on upzoning is out from the highly-respected Urban Institute, and it doesn’t have great news for YIMBYs:
We find that reforms that loosen restrictions are associated with a statistically significant 0.8 percent increase in housing supply within three to nine years of reform passage, accounting for new and existing stock. This increase occurs predominantly for units at the higher end of the rent price distribution; we find no statistically significant evidence that additional lower-cost units became available or became less expensive in the years following reforms. However, impacts are positive across the affordability spectrum and we cannot rule out that impacts are equivalent across different income segments. Conversely, reforms that increase land-use restrictions and lower allowed densities are associated with increased median rents and a reduction in units affordable to middle-income renters.
All that Sturm Und Drang for a 0.8% increase in housing supply? And no lower-cost units? That’s it?Well, yes. How-EVAH, we should know a few things about this study. Here is how the researchers did it:
[W]e generate the first cross-city panel dataset of land-use reforms that increase or decrease allowed housing density and estimate their association with changes in housing supply and rents. To generate reform data, we use machine-learning algorithms to search US newspaper articles between 2000 and 2019, then manually code them to increase accuracy. We identify 180 reforms increasing allowed density (upzoning) or reducing it (downzoning) across 1,136 cities in eight metropolitan regions. We merge these data with US Postal Service information on per-city counts of addresses and Census data on demographics, rents, and units affordable to households of different incomes. We then estimate a fixed-effects model with city-specific time trends to examine the relationships between land-use reforms and the supply and price of rental housing.
Getting good data on this question is fiendishly difficult, and the authors deserve credit for using AI to get it (as well as doing a lot of robustness checks). But still, what isn’t here impresses me a lot more than what is.
For example, the study includes “upzoning,” but as far as I can tell from the pre-print, it does not assess the extent of the upzoning. It matters a lot depends upon the extent of the upzoning, both in depth (how many more units per lot?) and breath (how many lots overall were upzoned?). That’s not a design flaw in the study: it just represents how hard it is to get good data.
In addition, it measures changes only nine years out, which is good, but does not really get at the affordability question. The vast majority of low-income people do not live in dedicated, deed-restricted affordable units: they live in units that used to be market-rate, and then filtered down over time: that time is usually around 30 years or so. So we shouldn’t be surprised that there are very few affordable units – and that is especially so since deed-restricted units pop up most frequently when subsidy is available, not when there is upzoning.
Most importantly, “upzoning” is itself quite a crude measure, as the authors acknowledge. They carefully distinguish between different sorts of land-use reforms: FAR increase, ADUs, height/setback changes, etc. But the problem is that in order for any one reform to work, it must be accompanied by others. For example, Minneapolis ended single-family zoning a few years ago, but it did nothing because the city still adhered to antiquated height and setback requirements. That’s not a knock on upzoning: it’s a demonstration that it must be done right. Similarly, as we have seen in California, the California Legislature ended single-family zoning through SB 9, allowing duplexes by-right – but it did nothing because of municipal implementation resistance piling on other requirements, necessitating a cleanup bill (which will likely be one in a series). If a city allows more multifamily buildings but does not change parking minimums, then it will have a minor effect.
That said, with all this, I believe that the study is basically right in the way it presents the context for reform. It concludes that if we are looking for short-term solutions to affordability problems, relying solely on upzoning cannot do the job by itself. I am deeply sympathetic to the YIMBY movement, but we must remember that we cannot hang everything on regulatory reform. Inclusionary policies, preserving expiring affordable units, greater demand-side help (such as making Section 8 an entitlement) and many other tools must be used.
And it will take a while, as I alluded to above. A friend in a planning department once acidly remarked to me, after a story on a particularly nice overhaul of a street, “yes: it’s the typical overnight success that took a couple of decades.” California clamped down on housing production in the 70’s: one can’t expect everything to change overnight.