The debate around whether building more market rate housing will alleviate the housing affordability crisis can sometimes feel like it degenerates into a false dichotomy: “YIMBY! It’s simple supply and demand!” vs. a host of excuses for why supply and demand doesn’t neatly apply to the housing market. A more nuanced economics perspective might be that yes, supply and demand does apply to the housing market, but it is a heterogeneous market, so we must also consider the cross-price elasticity of demand. As Rick Jacobus's piece in Shelterforce explains, the cross-price elasticity of demand is “a measure of how readily people switch from one submarket to another,” and accordingly, “how a change in the price of one product impacts the level of demand for another product.”
The fact that housing markets are heterogeneous is pretty intuitive. Most obviously there’s location; NYC is a separate market from LA. Even within a geographic location there are distinct submarkets, like the student housing submarket called out by Jacobus or the yuppie submarket of new construction, amenity-rich, one-bedroom apartments. These submarkets may have blurry edges, and units can change submarkets over time or with investment. For example, is Newark a separate submarket from NYC? Or, a luxury unit could slide into the next lower price range as it ages, or a student house could be remodeled and upgraded into a unit for young professionals or families. Similarly, housing in different markets are substitutes for residents, meaning a student may be able to find a mid-range apartment if they can afford it, instead of a student apartment.
The question then is not whether new market rate housing will affect rents at the bottom of the market, but by how much. This illustration of the concept shows how lower prices might trickle down to the next market tier.
It makes it pretty easy to imagine that prices in the "Very Low Cost" tier will be little affected by changes at the top. However, according to the California Housing Project, this UC Berkeley study found “New Affordable Housing Twice As Effective at Combating Displacement of Lower-Income Families.” Turn that around, and we learn that building market rate units would have an equivalent effect on the bottom of the market as building half as many affordable units. Which ain’t nothing!
Indeed, the cross-price elasticity of demand in any housing market is certainly not 1, as it would be in a completely homogeneous market. All of the usual arguments for why supply and demand don’t apply simply translate to a lower elasticity of demand. Supply in the high end goes up, but that just attracts foreign investors and most units stay vacant? That’s just that many more units that won’t trickle down to the next housing bracket. Increase in the mid market range allows some roommates to uncouple and get their own places? Same thing. None of these arguments cause supply and demand to completely fall apart. They just suggest that the effect of increasing supply in one part of the market will have less effect in another part of the market. It doesn’t mean that effect is zero.
The fact that housing markets are heterogeneous is pretty intuitive. Most obviously there’s location; NYC is a separate market from LA. Even within a geographic location there are distinct submarkets, like the student housing submarket called out by Jacobus or the yuppie submarket of new construction, amenity-rich, one-bedroom apartments. These submarkets may have blurry edges, and units can change submarkets over time or with investment. For example, is Newark a separate submarket from NYC? Or, a luxury unit could slide into the next lower price range as it ages, or a student house could be remodeled and upgraded into a unit for young professionals or families. Similarly, housing in different markets are substitutes for residents, meaning a student may be able to find a mid-range apartment if they can afford it, instead of a student apartment.
The question then is not whether new market rate housing will affect rents at the bottom of the market, but by how much. This illustration of the concept shows how lower prices might trickle down to the next market tier.
It makes it pretty easy to imagine that prices in the "Very Low Cost" tier will be little affected by changes at the top. However, according to the California Housing Project, this UC Berkeley study found “New Affordable Housing Twice As Effective at Combating Displacement of Lower-Income Families.” Turn that around, and we learn that building market rate units would have an equivalent effect on the bottom of the market as building half as many affordable units. Which ain’t nothing!
Indeed, the cross-price elasticity of demand in any housing market is certainly not 1, as it would be in a completely homogeneous market. All of the usual arguments for why supply and demand don’t apply simply translate to a lower elasticity of demand. Supply in the high end goes up, but that just attracts foreign investors and most units stay vacant? That’s just that many more units that won’t trickle down to the next housing bracket. Increase in the mid market range allows some roommates to uncouple and get their own places? Same thing. None of these arguments cause supply and demand to completely fall apart. They just suggest that the effect of increasing supply in one part of the market will have less effect in another part of the market. It doesn’t mean that effect is zero.
