Another housing myth debunked: Neighborhood price effects of new apartments

New research shows new apartments drive down rents in their immediate neighborhood, disproving the myth of “induced demand” for housing

If you’re a housing supply skeptic, there’s one pet theory that you’ve been able to hang your hat on, in the face of a barrage of evidence that increasing the supply of housing helps hold down, or even drive down rents. It’s the theory of “induced demand”–that building nice new apartments (or houses) in a neighborhood, so changes a neighborhood’s attractiveness to potential buyers that it drives up prices. It’s a plausible sounding argument, but in our view a wrong one, meaning it’s time for another episode of City Observatory’s own “Myth Busters.”

Regular readers of City Observatory will immediately recognize the term “induced demand” because we talk about it frequent in the context of transportation.  When a highway department widens a road (invariably, in a vain attempt to reduce traffic congestion), it tends to quickly attract new traffic and becomes just as congested as before (a phenomenon so common and well-documented that it is now termed “the fundamental law of road congestion“).

The induced demand theory applied to housing then, is that building new housing somehow signals a big change in the neighborhood’s amenities and livability and the new supply of housing triggers an even bigger increase in demand, such that any beneficial effects of added supply that would occur in the textbook model are more than offset.

A slightly more nuanced version is a claim that while it may help with supply regionally, in may trigger a change in a neighborhood’s relative perception as a desirable place to live, and while the supply effect may help lower rents regionally, it drives up rents locally:  Rick Jacobus makes this argument

In other words, the demand for housing in any neighborhood is highly variable and can switch from very low to very high quickly. But the supply is almost entirely fixed. In established neighborhoods, no matter how much building is going on, the new supply will be small relative to the overall market so increased supply will have almost no impact on rents. It might theoretically drive rent down some tiny amount but, in practice, the impact of new development in a neighborhood is usually the opposite because it increases demand (for that neighborhood) by more than it increases supply. Partly this is true because any new development is visible, new and exciting. Developers push this process along with marketing campaigns that invariably promote not just one building but the surrounding neighborhood (even if they have to coin a new name for the neighborhood). The result is that—on the neighborhood level—adding supply may not lower rents. It may raise them.

While it’s a semantically appealing analogy, it makes little if any economic sense.  A key difference between roads (nearly all roads, anyhow) and houses, is that we charge positive prices for housing in a way we don’t for freeway capacity. The reason expanding capacity induces demand in the case of roadways is that we charge road users a zero price. Thus capacity (and willingness to tolerate delay) are the only things regulating demand, and when capacity is expanded, demand responds quickly.  As we’ve shown time and again, as in the case of the Louisville I-65 Ohio River Crossing, when you actually price new capacity at even a fraction of its cost, demand evaporates.

Crying “induced demand” seems to be an increasingly popular gambit from housing supply skeptics.  The Southern California Association of Governments deployed it as part of its “Regional Housing Needs Assessment” a state mandated document to calculate how many housing units the region needed to add. SCAG argued that adding more housing to improve affordability would be futile because it would induce additional demand, as on freeways.  UCLA Professor Paavo Monkkonen, challenged that analogy:

The package compares housing supply and affordability to induced demand on freeways (page 23), which they properly note is unlikely to alleviate congestion in the long run. This comparison is not apt, because freeway access is free and housing is not. Congestion occurs when the absence of prices causes a shortage. A housing crisis occurs when a shortage of housing causes high prices. This crucial difference means that new supply is almost useless in the former and incredibly important in the latter.

Still, in the abstract, its possible to imagine that the construction of a new apartment building is some sort of watershed event that triggers a mass re-appraisal of the attitudes of potential renters.  There is evidence at least some externalities and positive feedbacks in development.  The empirical question is whether the size of these effects is enough to swamp the downward effect on rents from expanding the supply of housing in the neighborhood.

Until now, that’s been a factual void–one which lets supply side skeptics assert the induced demand hypothesis for housing.  But it is a void no more:  A forthcoming paper–“Does Luxury Housing Construction Increase Rents in Gentrifying Areas?” from Brian Asquith, Evan Mast, and Davin Reed, explores this question in detail.  Asquith and Mast are economists with the Upjohn Institute; Reed is an economist with the Philadelphia Federal Reserve Bank. Their paper, available in preliminary draft form here, uses very geographically detailed data on apartment rents and new apartment construction in gentrifying neighborhoods to see whether a new building drives up prices nearby (the induced demand theory) or whether it depresses them (the supply side theory).  Earlier, we reviewed an paper from Mast on the chain reaction of migration triggered by the construction of new buildings; it showed that constructing new market-rate buildings triggered a series of moves that produced significant numbers of vacancies in lower income neighborhoods.)

Asquith, Mast, and Reed gathered data for 100 new apartment buildings in each of eleven cities around the country. They identified geographically isolated buildings, and then gathered data on rental listings for apartments in the area surrounding the new building. They analyzed the data to see whether rents went up or down closer to the new building in the year after it was constructed.  If the “induced demand” theory were correct, one would expect the rental prices of existing apartments close to the new building to rise more or faster than other apartments located further away.  This chart summarizes rents by distance (in meters) from the newly constructed building.  The dashed line is the “before” showing the level of rents prior to construction, while the solid line is the “after” showing rents after construction.  (And to be clear:  the rents shown are for apartments in the area excluding the apartments in the newly built building.)

This chart shows that rental prices for apartments close to the new building fell relative to the prices for apartments located further away. The dashed “before” line has a negative slope, suggesting that prices declined the further you got from the site of the new building.  The solid “after” line has a positive slope (prices increase the further you get from the new building).  Overall, prices are higher (the solid line is above the dashed line), but prices actually went down next to the new building, and increased far less than in the area further away from the new building.

These data are a strong challenge to the induced demand theory.  If a new building made an area more attractive, one would expect the largest effect in the area very near the building. But, consistent with the traditional “more supply reduces rents” view, the addition of more units in an area seems to have depressed rents (or at least rent increases) compared to buildings in the surrounding area.

While this is still a preliminary paper, and has yet to be published, it does offer the best evidence yet presented on the theory of induced demand. We’ll reserve a final judgment until after review and publication, but based on the data presented here, this myth is busted.

Addendum: A hat tip to Trulia economist Issi Romem for flagging this study.

 

 

By Joe Cortright