Evan Mast and his colleagues at the W.E. Upjohn Institute have become rising stars in the growing field of housing economics research. In a new working paper “Supply Shock versus Demand Shock,” Asquith, Mast, and Reed (2019) make a compelling case for new housing supply as a tool for mitigating rent increases and absorbing demand shocks, rather than causing them. “Contrary to common concerns,” they conclude, “new buildings slow local rent increases rather than initiate or accelerate them.” So what does the data say?
We can break it down into three key findings:
- By comparing low-income neighborhoods with new market-rate development to nearby neighborhoods without it as a rough control condition, the authors found that new development can keep rents lower by 5-7% less than what they otherwise would be. Note that this doesn’t mean rents go down; rather, rent increases, on average, are slowed. (One might even speculate that the latter is a precondition of the former.)
- Reinforcing results from Mast’s previous paper, the data shows that the “amenity effect” of bringing in new market-rate supply, higher-income earners, restaurants, SoulCycle—you name it—this effect is vastly smaller than the benefit that new supply brings in mitigating rents.
- While new development increases in-migration to the neighborhood, newcomers primarily arrive to live in newly-built market-rate housing. In-migration to older housing stock actually decreases on average.
In parsing this study, it’s important to note that, as the authors argue, new construction is an after-effect of gentrification rather than its harbinger. Naturally, if rents weren’t already rising, new market-rate development would not be financially feasible or attractive to private investors.
But therein lies a methodological conundrum: given these patterns, new development is hardly random, and thus developing a true control condition for the empirical study is more difficult. Asquith et al develop some novel analyses with rental prices, individual addresses, and construction data from 11 major cities. First, a “near-far” specification compares the area closest to a new building with an area slightly farther away. As the authors explain: “The idea is that frictions in the land assembly and development approval processes lead to random variation in building placement and timing at the hyper-local level, making the outer area a good control for the treated inner area.”
Another control method is a “near-near” specification, comparing rental listings near buildings completed in 2015-2016 and listings in 2019, after their survey concluded. “The underlying logic is that developers choose sites in both groups for similar reasons, but one building is completed before the other for idiosyncratic reasons, such as the timing of when sites are available for purchase,” the researchers explain. The third tool is to simply combine these effects in a “triple-difference” analysis.
By running Zillow listings from 11 major cities ranging between 2013-2018, the authors found the same result in every control condition: new market-rate housing reduces rents by 5-7% relative to what they would be if the housing hadn’t been built. That’s a difficult counterfactual to posit statistically, but we have yet to see a better effort at estimating it.
Finally, since this new construction typically only occurs in a low-income neighborhood when it is already gentrifying, the researchers explore the in-migration effects this housing has on in-migration to the area. We often hear this question posed as “housing for whom?” This paper poses a simple answer: housing for people who would otherwise bid up the price of older housing stock. The benefits, when placed in context, are dramatic.
By tracking individual address changes, the authors found that new construction increases decreases the “average origin neighborhood income” of new arrivals by around 2%, and increases arrivals from other low-income neighborhoods by 3%. The authors have a hypothesis explaining these results that may provide fruit for further research, namely “that new buildings reduce costs in lower segments of the housing market, not just in the high-end units that are the most direct competitors of new buildings.”
“These results suggest that the effect of new buildings on rents is not driven entirely by the high-end listings that are their closest competitors,” they further speculate. “Instead, the increased migration from low-income areas is consistent with rent decreases among relatively cheaper apartments.” These results could also demonstrate that fewer low-income households are moving out of a gentrifying neighborhood thanks to the benefits of new supply.
California is in the throes of a crisis-level housing shortage. We need every tool at our disposal if we are going to see a state that guarantees housing as a human right, including policies that simply allow more housing to be built. What are we waiting for?