Induced demand has to come from somewhere

Amenity effects reallocate demand, they don't create it

The literature on local effects of new housing uses the framing of supply effects vs. amenity effects. A new apartment building adds housing supply, which decreases rents. But it also adds new amenities (e.g. nicer stores and cafes), which can induce new demand and raise rents. So the effect of a new building on local rents is the net effect of these two channels. (And empirically, the supply effect dominates.)

This analysis is looking only at the neighborhood around the new building. But induced demand means drawing in new residents from elsewhere: when demand increases in this neighborhood, it decreases in the neighborhoods that the new residents moved out of.

Upshot: induced demand through amenity effects is a wash. It increases demand in one neighborhood, but reduces it in another. So even if residents near the new building are worse off due to higher rents, residents in the origin neighborhoods of the movers are better off. A full welfare accounting needs to reflect this.

Read my full literature review on local supply effects and local demand effects here:

Gentrification and displacement