The Unintended Consequences of Affordable Housing Policy
by Carl Bonham, UHERO, September 18, 2013
Honolulu City Council member Ron Menor has introduced a resolution to amend city policy on household income thresholds for affordable for-sale and/or rental housing. The purpose is to increase affordable housing opportunities for Oahu households in the lower income ranges. Resolution 13-168 would amend the percentages of affordable housing units that developers are required to provide to receive authorization for housing. Current city housing policy requires that 10% of a development's units must be affordable for households earning no more than 80% of the HUD median income for Honolulu. Another 10% of units in a development should be affordable for families earning between 80 and 120% of the median income, and 10% for families earning between 120 and 140% of the median income. The resolution proposes that the mix of affordable housing required of developers be reconsidered. The intent is to change the requirements in a way that leads to more affordable housing for those who need it the most.
While requiring developers to set aside a fraction of a project to be sold at below market prices may seem like a reasonable way of dealing with the problem of affordable housing, economic theory and years of experience suggest exactly the opposite. Such requirements, known as inclusionary zoning (IZ), act as a tax on developers with the proceeds used to subsidize housing for gap income households earning between 80 and 140 percent of the median income. But that tax reduces incentives for developers to produce all forms of housing, and will reduce the overall supply of housing units and increase the price of housing.
In 2010, UHERO conducted a comprehensive review of studies that analyzed IZ policies across the United States.1 Approximately 90% of the studies concluded that IZ increases the market price of housing and decreases housing units available in the market. Of the 18 studies that were able to quantify the effect of inclusionary zoning on housing market outcomes, 13 found that IZ policies both increased the market price of housing and decreased housing units available in the market, and three more studies found evidence of at least one of those effects. UHERO’s report concluded that “Inclusionary Zoning policies have failed in other jurisdictions, and are failing on Oahu.” Such policies have not delivered substantial numbers of affordable housing units to households the programs were designed to help.
The undersupply of housing services relative to household formation on Oahu is a chronic problem. While IZ policies are politically appealing, they mistakenly tax housing to encourage more of it! The effect of a tax on the production of any product, housing included, is relatively straightforward. The extra tax imposed by IZ increases the cost to developers and limits the supply of housing provided. Facing the additional cost, developers will build fewer housing units, all else equal. In the worst case scenario, if the expected loss on the affordable units does not allow developers to meet their required rate of return, then projects will never get off the ground. The primary means of insuring the project is viable is to produce more upscale, higher priced homes to offset the loss on the subsidized housing.2 So, the IZ tax not only reduces the overall supply of housing, it also changes the mix of housing by encouraging higher end and more expensive housing developments.
“Low-cost housing is usually produced through a process called filtering where existing housing units drop in cost as their relative quality falls, rather than through construction of new, lower-cost units.” (Feldman, 2002 p. 9) Over time, the existing stock of housing depreciates and declines in quality relative to new amenity rich units. For example, new housing often includes central air-conditioning and energy saving appliances, whereas twenty years ago few housing units would have such amenities. The construction of new housing units also increases the overall supply of housing, which increases the supply of lesser quality units to those with lower incomes as owners of older homes “trade up”. (Feldman 2002, p. 10). Malpezzi and Green (1996, p. 1811) also found that “high-quality new construction is associated with growth in the low-quality stock as well... [T]o the extent that a city makes it easy for any type of housing to be built, it will also enhance the available stock of low-cost housing.” By discouraging construction of market priced housing, IZ reduces filtering and leads to less low-cost housing than without these well-intended policies.
On Oahu IZ policies also miss the cyclical nature of the affordability problem. The figure below shows the maximum affordable mortgage3 for four Honolulu income groups from 1990 to 2012 and median resale prices of single-family homes and condominiums from the Honolulu Board of Realtors.
From 1990 to 1995, at the end of the last Oahu housing cycle, single-family homes remained out of reach of households earning 140% of the HUD median family income (MFI). Yet from 1996 to 2003, households in the 120-140% MFI bracket could afford the median priced home, and all gap income households could afford a median priced condo from 1996 through 2005. From 1992 to 2012, all income groups except the 80% group could afford the median priced condo.
The recent peak in unaffordability occurred in 2007 at the height of the last housing cycle. The declining price of homes through 2009 and the record low mortgage rates since then has once again brought single-family homes within reach of the top gap income group. And, condos are again within reach of all gap income levels. Imposing more costly affordability taxes on developers at this stage of the home building cycle will result in some small increase in the number of new affordable housing units. But, the tax will also accelerate the end of the construction cycle by leading to a more rapid increase in home prices and eventually closing off demand for the higher priced units developers need to build to cover their costs. IZ policies will endanger project viability by squeezing profit margins, especially as other construction costs rise and home prices flatten out. The result is that less housing will be produced than otherwise would be the case.
Reducing or eliminating overly burdensome regulation on development, including inclusionary zoning, will increase affordability of housing for two reasons. First, it will encourage building, increasing the overall stock of housing, which will help hold down the market price of housing. Second, removing IZ will facilitate the natural “filtering” process, with newer units going to higher income households and older depreciating units being increasingly occupied by lower income households. Finally, IZ policy misses the basic fact that affordability problems arise not just due to the high cost of housing. The affordability problem is a dual problem of high prices and low incomes. Just as is done in Federal Housing Assistance programs, affordability could be addressed through housing subsidies that help households purchase or rent housing units.
1 See Bonham, Burnett, and Kato, “Inclusionary Zoning: Implications for Oahu’s Housing Market”, February 2010.
2 Of course, this only works when there is a demand for higher priced units.
3 UHERO calculates the Affordable Mortgage as the maximum mortgage a household earning the HUD Median Family Income for Oahu could afford after a 20% down payment and assuming the household spends no more than 30% of its monthly income on the mortgage.
Bonham, C., Kimberly Burnett, and Andrew Kato. 2012. Inclusionary Zoning: Implications for Oahu’s Housing Market. UHERO Project Report, accessed at http://www.uhero.hawaii.edu/assets/UHEROProjectReport2010-1.pdf.
Feldman, R. 2002. The Affordable Housing Shortage: Considering the Problem, Causes and Solutions. Federal Reserve Bank of Minneapolis. Banking and Policy Working Paper 02-2.
Malpezzi, S. and R. K. Green. 1996. What has Happened to the Bottom of the U.S. Housing Market? Urban Studies. 33(10): 1807-1820.