Blog post

How Filtering Increases Housing Affordability

Good research indicates that building middle-priced housing increases affordability through "filtering," as some lower-priced housing occupants move into more expensive units, and over time as the new houses depreciate and become cheaper.
Todd Litman | August 27, 2018, 10am PDT
Share Tweet LinkedIn Email Comments

My recent column, "Affordability Trade-offs," advocates a broad definition of affordability that considers middle- as well as low-income households, transportation as well as housing costs, and future as well as current cost burdens (see table below).

Narrow Versus Broad Affordability Analysis




Target populations

Homeless and low-income households that spend more than 30% of their budgets on housing

Low- and moderate-income households that spend more than 45% of their budgets on housing and transportation

Time perspective


Current and future

Costs considered

Rents or mortgages

Rents or mortgages, heating/cooling, maintenance, property taxes and basic transportation


Preserve and build cheap housing through regulations and government subsidies

Build lots of moderate-priced housing units in walkable urban neighborhoods.

A narrow definition favors policies that preserve and subsidize cheap housing. A broader definition tends to supports policy reforms allow much more development of moderate-priced housing ($200,000-600,000 per unit) in walkable urban neighborhoods. Even if the new units are initially too pricey for lower-income households, they increase affordability through filtering, as some lower-priced housing occupants move up to the moderate-priced units, and over time as they depreciate and become cheaper.

There is solid research indicating that filtering occurs, including Stuart Rosenthal's 2014 study, "Are Private Markets and Filtering a Viable Source of Low-Income Housing? Estimates from a 'Repeat Income' Model," published in the American Economic Review, and Miriam Zuk and Karen Chapple’s 2016 study, Housing Production, Filtering and Displacement: Untangling the Relationships [pdf], published by the Berkeley Institute of Government Studies. Both studies indicate that increasing housing supply does tend to reduce housing prices, particularly over the long-run, although Zuk and Chapple emphasize that subsidies are often needed to deliver enough lower-priced housing to avoid lower-income household displacement. Similarly, recent research by the City Observatory shows that displacement rates are lower in neighborhoods that increase housing supply and therefore reduce competition for available housing units.

Three studies suggest that increasing housing supply is ineffective at increasing affordability. Richard Applebaum and John Gilderbloom’s 1983 study, "Housing Supply and Regulation: A Study of the Rental Housing Market," published in the Journal of Applied Behavioral Science, found that cities that had more rental housing construction do not have lower rents or higher vacancy rates. Similarly, Andrejs Skaburskis’ 2006 "Filtering, City Change and the Supply of Low-priced Housing in Canada," published in Urban Studiesusing 1996 Canadian census data, found that the filtering process is too slow to significantly reduce low-income households housing burdens. 

However, these studies are old and did not account for significant confounding factors. For example, rental housing growth tends to be greatest in attractive and economically successful cities that also have more population growth and higher wages. They also focus on low-income affordability, overlooking impacts on middle-income housing costs. Another problem with these studies is that they ignore additional costs of older houses. Older houses and apartments often have the lowest rents, which leads some experts to recommend policies that preserve older housing stock, such as restrictions on renovations and demolitions. However, their low prices partly reflect their higher operating costs, including greater maintenance and utility expenses, and other undesirable factors such as poor seismic and fire safety. Older housing stock can be rehabilitated with weatherization and structural reinforcement, but this is costly; replacement is often more cost effective overall, particularly if the new buildings have more units.

As a result, it is unsurprising that areas with more housing supply growth also have less very-low-priced housing, but that does not really mean that in attractive cities, increasing moderate-cost housing supply is ineffective at increasing affordability for moderate- and low-income households.

A study by Dr. John Rose, The Housing Supply Myth analyzed the ratio of household growth to housing unit growth in 33 Canadian metropolitan regions and found that in some areas (notably, Toronto, Vancouver and Victoria), prices  skyrocketed although housing units grew faster than households. However, Rose's methodology has been criticized by Professor Nathan Lauster and statistician Jens Von Bergmann for failing to account for data problems, particularly changes in data collection methods between census, which counted many previously overlooked housing units, and so exaggerated the growth in housing units. Another major weakness with Rose's study is that it aggregates all housing in a metropolitan region, which in cities like Toronto, Vancouver and Victoria includes many time-shares and student housing. As a result of these problems, Rose's  study does not really prove that increasing middle-priced housing supply is ineffective at increasing middle- and lower-priced housing affordability. 

Recent experience indicates that even in high growth cities like Seattle and Portland, increasing rental apartment supply does drive down prices. During the last year these cities have built tens of thousands of new rental apartment units, and Seattle experienced a 2.4% decline, and Portland a 2.6% decline, in average rents, despite strong population and economic growth.

Seattle Rent Growth Over Past 12 Months

Seattle Rent Price Trends

Such one-year reductions are insufficient to make these cities truly affordable to lower-income households, and much of these new housing consists of small, city-center high-rises that are unsuited to many households, particularly families with children, but it demonstrates proof of concept: building lots of moderate-priced housing increases middle-income affordability, and if maintained for several years, can help increase future lower-income affordability. Here in Victoria, where our population grows about 1.5% annually, we call this the 1.5% Neighborhood Affordability Solution [pdf]

A responsible affordable housing policy allows older housing stock to be replaced when appropriate, particularly if that allows more infill, for example, if an older single-family house can be replaced by multiplexes, or a two-story apartment can be replaced by three or four stories. Although these units will initially be too costly for lower-income households, they expand housing supply, which drives down prices, and become more affordable over time.

Such policies do not eliminate the need to subsidize some housing units for low-income and special needs households, but subsidies can only serve a small portion of the demand for lower-priced housing units. Communities that want true affordability, including middle- as well as low-income households, transportation as well as housing affordability, and future as well as current affordability, must expand infill housing supply faster than population growth in walkable urban neighborhoods. 

Let thousands of new houses bloom! Accomplishing this will require zoning code and development policy reforms that make it easy to build lower-priced housing type—multiplexes, townhouses and low-rise apartments—in residential neighborhoods that are currently only allow single-family houses. These policies help reduce housing prices and increase the number of affordable units that can be built with a given housing subsidy. I'll describe these strategies in a future column.

For More Information

Dan Bertolet (2018), Want Less Expensive Housing? Then Make it Less Expensive to Build Housing Sightline Institute.

Filtering YIMBY Wiki.

Daniel Hertz (2015), What Filtering Can and Cannot Do, City Commentary. 

Josh Lehner (2016), Housing Does Filter, Oregon Office of Economic Analysis.

Legalizing Affordable Housing, Sightline Institute.

Stuart Rosenthal (2014), "Are Private Markets and Filtering a Viable Source of Low-Income Housing? Estimates from a 'Repeat Income' Model," American Economic Review.

Old Urbanist (2015), When the Market Built Housing for the Low Income.

Miriam Zuk and Karen Chapple (2016), Housing Production, Filtering and Displacement: Untangling the Relationships, Berkeley Institute of Government Studies.

Share Tweet LinkedIn Email