The evidence increasingly points in another direction. Housing markets are connected through household mobility. When new units open, they set off a sequence of moves that reshuffles demand across price levels.
The Atlantic recently published a piece on a recent study of a 43-story condominium tower in Honolulu that tracked this process in detail. Researchers followed buyers into 512 newly constructed units and traced their previous addresses. They then identified who moved into those vacated homes and, in many cases, followed that chain further. The result: the 512 new units generated at least 557 vacancies across the city. A new home at the top tier created openings in older and typically less expensive units elsewhere.
The mechanism is straightforward. A household moving into new construction leaves behind an existing unit. That unit becomes available to another household, often one moving from a somewhat lower-priced home. The process repeats through multiple rounds. Economists refer to this as a migration chain or vacancy chain, but it reflects a basic feature of urban housing markets: households tend to move incrementally across neighborhoods and price points rather than making large jumps.
Earlier research by Evan Mast reached similar conclusions using national data. In a 2019 study of 686 new multifamily buildings across 12 large U.S. cities, Mast used address-history data to track 52,000 movers and subsequent occupants. He found that new market-rate construction is linked to lower-income neighborhoods through a sequence of moves. In simulation, 100 new market-rate units ultimately created the equivalent of 70 openings in neighborhoods below the metro-area median income and 39 openings in neighborhoods in the bottom fifth of the income distribution. These effects occurred within roughly five years of completion.
The pattern depends on connectivity between segments of the market. Mast’s analysis of Chicago-area moves shows that while households rarely leap from the lowest-income neighborhoods to the highest-income ones, they frequently move one or two “steps” up or down the income distribution. Those incremental moves are sufficient to link new construction to substantially lower-income areas over several rounds.
Research outside the United States has found similar dynamics. A study of centrally located market-rate construction in Helsinki observed that new units were associated with openings in the city’s poorest neighborhoods. Swedish evidence points in the same direction. While local conditions vary, the underlying mechanism — sequential mobility across connected segments — appears consistent.
The Honolulu study adds nuance. Vacancy chains there were shorter than theoretical maximums. On average, each market-rate unit produced about 1.6 vacancies. Some moves did not free up a unit at all, such as when a buyer moved out of a shared household. In a tight market, new supply may relieve overcrowding before generating a long string of vacancies.
That finding is not a contradiction. It reflects initial market conditions. In cities with high rates of doubling up, the first effect of new housing may be to reduce crowding rather than to cascade through multiple vacancy rounds. But even shorter chains indicate that new construction absorbs demand that would otherwise compete for existing units.
The research also suggests limits. Mast notes that in neighborhoods where rents are already close to the marginal cost of providing housing, reducing demand may not produce large additional price declines . Vacancy rates, rent burdens, and local cost structures all shape outcomes. Migration chains diffuse benefits across a metropolitan area rather than concentrating them in one block or district.
For policymakers, the central takeaway is empirical rather than ideological. New market-rate housing is not isolated from the rest of the market. It is connected to middle- and lower-income neighborhoods through observed patterns of mobility. When new units come online, they reduce demand pressure in the units households leave behind. Over time, that effect extends across multiple segments of the market.
This does not imply that any single project will transform affordability. The magnitude of the effect depends on scale, baseline vacancy rates, and regional mobility patterns. But the consistent finding across multiple studies is that building additional housing — including at the higher end of the market — increases openings in lower-priced segments within a relatively short time horizon.
Cities confronting high housing costs face a structural imbalance between households and homes. The evidence indicates that adding supply at one tier does not remain confined to that tier. Housing markets adjust through movement.
If the objective is to reduce pressure across the system, sustained construction at multiple price points — including market-rate units — is part of how metropolitan housing markets rebalance over time.
