Housing Finance Policy

house in hands

It is unclear whether the United States should continue to subsidize the purchase of homes. While U.S. subsidies may have helped expand homeownership, they also may have made the economy more susceptible to shocks and more prone to sluggish recovery, as well as led households and financial institutions to take on risks that can jeopardize the economy at large.

Last updated May 18, 2015

•  The United States’ approach to homeownership subsidies appears to have had harmful side effects.

•  High rates of homeownership can make the economy more vulnerable to shocks. In addition, the benefits of homeownership are likely smaller than is often assumed.

•  Indirect homeownership subsidies through the government-sponsored enterprises (GSEs) have arguably encouraged greater risk-taking in mortgage markets at taxpayer expense. Many of the benefits have not been passed on to lower-income households.

•  Withdrawing GSE subsidies could raise the cost of the popular 30-year fixed-rate mortgage, but it is not clear that this effect would be large.

•  The mortgage interest tax deduction has likely benefited wealthy households most. Moreover, research suggests that it encourages people to purchase larger homes than they were already planning to buy, rather than encouraging greater homeownership.

•  If the United States continues to pursue homeownership subsidies, policymakers should consider efforts that subsidize the accumulation of housing equity rather than housing debt, as has been done in other countries.


The United States has a century-long history of promoting homeownership. It does so through subsidies that are both direct (to homeowners) and indirect (through mortgage markets) to make it cheaper for households to finance the purchase of a home (for a brief history, see Slivinksi 2008). The U.S. government subsidizes homeownership far more than any other developed country.

This policy stance has been called into question after the events of the last several years. The unprecedented rise in house prices during the early 2000s reversed sharply starting in 2006, which was followed by a record wave of foreclosures and a related financial crisis. For millions of people, homeownership has not paid off. Moreover, the overall economy has arguably been harmed because of widespread house price declines and mortgage defaults.

In considering optimal housing finance policy, there are two important questions to answer: Should we subsidize homeownership? And if so, how? This essay addresses each of these questions in turn.

Should We Subsidize Homeownership?

Economists argue that one of two specific tests must be passed to justify government subsidies to a given activity. One is that positive "externalities" are present — that is, the activity in question has beneficial side effects on other parties not directly involved. Those undertaking the action do not experience firsthand its full benefit to society, so too little is undertaken. A subsidy creates an additional incentive for these individuals to take these actions, potentially leaving society as a whole better off. (Similarly, taxes can offset negative externalities.) The second possible justification for a subsidy is the desire to redistribute resources to assist certain groups, in this case to expand access to owner-occupied housing. In either case, it is important to ask if current policy is a productive way to achieve society's goals.

A substantial amount of effort has gone into measuring positive externalities to housing, and whether they exist on a scale that warrants policy intervention. Many studies have documented, for example, that neighborhoods with a large proportion of owner-occupied homes tend to have better upkeep, higher-performing schools, lower crime, and greater civic involvement (see a summary provided by the National Association of Realtors in 2012). However, it is difficult to say with certainty that homeownership is responsible for creating these benefits. Individuals likely to purchase homes may have innate characteristics that make them more likely to invest in their communities in other ways. People induced through policy to enter homeownership will not necessarily share these characteristics. Recent work has made progress toward answering this question by exploiting a "natural experiment" in the context of an urban revitalization effort, finding that efforts to improve properties can increase the value of surrounding properties (Owens, Rossi-Hansberg, and Sarte 2010). This suggests that improvements in the quality of housing can benefit surrounding communities that did not themselves pay the full cost of making them. As a result, subsidies to encourage such improvements may be warranted. It is important to keep in mind, however, that this does not necessarily establish the case for the approach chosen in the United States, which has been to subsidize homeownership rather than improvements in housing quality.

By definition, a subsidy makes its recipient, in this case home-buying households, better off, all else equal. However, just as market prices of homes can be too high in the absence of subsidies to yield the efficient level of homeownership, misplaced or overly large subsidies can make the private cost of homeownership lower than the social cost. Recent events suggest in fact that there may be negative externalities to high rates of homeownership that accrue to society as a whole, in particular when home prices decline.

For example, homeownership can affect the labor market. Homeownership makes it hard to relocate since selling a home entails significant transaction costs. This can lock people into one geographic location, which can affect the health of the local job market by, for example, making it costly for firms to find the right mix of workers if technological or business conditions change. As a result, unemployment rates may be affected as firms find their search for employees less effective and post fewer vacancies. In fact, some recent research suggests that U.S. states with high homeownership have much higher unemployment rates in the long run, possibly due to the lower labor mobility that exists in those states (Blanchflower and Oswald 2013). The cost of low mobility can heighten in a bad economy, when job opportunities are scarce, yet this is precisely when mobility is likely to fall because house prices may be low. Following the recent housing downturn, a record number of homeowners — as a many as a quarter, by some estimates — had negative equity in their homes (they owed more on the mortgage than the home was worth), a situation that may force a homeowner to accept a loss in order to sell. Households with negative equity are 30 percent less likely than other homeowners to relocate, according to one study (Ferreira, Gyourko, and Tracy 2011). In the initial years of the downturn, states with the greatest incidence of negative equity experienced greater declines in residential mobility (Caldera, Sanchez, and Andrews 2011). Low labor mobility — which could be exacerbated if homeownership subsidies result in an underdeveloped rental market — stops workers from finding the jobs at which they are most productive, in part by affecting the decisions of firms to search for workers. The inability or unwillingness of some households to relocate, while sensible for them individually in the circumstances, may impede the recovery of the overall labor market from a recession, thereby limiting opportunities for potentially all other workers, although the evidence is not conclusive on whether negative equity has contributed to the high unemployment rates of recent years. For example, recent theoretical research has found that a decline in house prices reduces migration and causes unemployment to rise by different degrees in different locations, but the effect on aggregate unemployment is moderate (Karahan and Rhee 2013).

Widespread homeownership can also impair the opportunities for households if house prices fall. Owner-occupied housing is typically the single largest investment one makes in a lifetime, and so at any time it constitutes a large share of a single homeowner's total portfolio. While households can diversify their financial investments relatively easily — for example, by holding an indexed security that tracks the overall market — there are relatively few methods available to the average homeowner for protecting their portfolios against the risk that their housing asset will fall in value. In fact, local house prices – the ones homeowners actually face — are much more volatile than national averages, especially in large coastal cities. Thus, housing subsidies encourage households to adopt a risky investment strategy. Once again, individual households may willingly accept these risks in exchange for the private benefits of homeownership. However, when many households have undiversified portfolios centered on housing, a widespread decline in house prices could translate into potentially steep declines in wealth, and therefore to declines in household spending and overall employment. For example, a San Francisco Fed study shows that developed nations that experienced greater increases in household leverage in the recent episode experienced faster house price appreciation and larger subsequent declines in household consumption following the housing bust, a phenomenon that other research has found to also exist across states (Glick and Lansing 2010).

In short, subsidizing homeownership may have made the economy more susceptible to shocks and more prone to sluggish recovery. And since the externality-based case for subsidizing homeownership is not overwhelming, it is reasonable to view subsidies for homeownership as being motivated less by the desire to correct externalities and more by distributional goals. However, as discussed later in this essay, the specific types of subsidies the U.S. government has traditionally offered may not benefit targeted groups as much as hoped.

If We're Going to Subsidize Homeownership, How?

Irrespective of the motivation for a subsidy, whether externalities or redistribution, it is important to ask whether the specific form of the subsidy is effective at achieving society's goals, and whether there are unintended consequences, especially for the economy as a whole.

In the United States, subsidies to housing finance take two dominant forms:

  • Indirect subsidies through the mortgage market, most notably through the government sponsored enterprises (GSEs) Fannie Mae and Freddie Mac; and
  • Direct-to-consumer incentives, most notably through tax deductions that homeowners can claim for mortgage interest payments on owner-occupied homes.

There is reason to question each of these avenues, as discussed below.

The GSEs

Since their creation as private entities in 1968 and 1970, respectively, Fannie Mae and Freddie Mac have existed to make mortgages cheaper and more widely available. At inception, they did so by purchasing mortgages from lenders with knowledge of local markets, thereby allowing them to make loans with the option to avoid having a balance sheet full of geographically concentrated risks. The GSEs perform the same function today even though lenders are much more geographically diversified. With the mortgages they buy, the GSEs create and guarantee mortgage-backed securities (MBS), some of which they hold as investments and some of which they sell to investors. Empirical studies have found that the actions of the GSEs have tended to push down interest rates on mortgages that are eligible for GSE purchase, although this doesn't seem to explain the entire spread between eligible mortgages and other mortgages (Passmore, Sherlund, and Burgess 2005).

The government's role in housing finance was large before the recent crisis. At their pre-crisis peak in 2003, the two GSEs purchased half of all new mortgages being originated. Their role grew even larger following the recent housing bust. For example, according to the Federal Housing Finance Agency, the GSEs' regulator, the GSEs issued $488 billion or 69 percent of all mortgage-backed security volume in the first nine months of 2014 (an additional 30 percent of the total volume was issued by Ginnie Mae, another quasi-government entity).

GSEs raise funds in the market by selling debt and issuing debt. Though the GSEs are private corporations, markets have, for decades, widely expected that the government would provide the GSEs with emergency funding in the event of financial trouble arising from the need to repay debt. This implicit promise of support gives creditors good reason to expect repayment. As a result, investors have historically been willing to lend to Fannie Mae and Freddie Mac at lower interest rates than most other corporations, which has helped them remain bigger than they otherwise would be and has hindered the ability of private mortgage institutions to compete on an even playing field.

The expectation of government backing was proven correct in September 2008, when the GSEs were seen to be headed for failure and were placed into government conservatorship. Conservatorship means the GSEs' operations were allowed to continue under government control, and with large government financial support, with the FHFA responsible for allocating losses among the GSEs' creditors. The federal government remains in this role today. (See Federal Housing Finance Agency: Conservatorship). Thus, the extraordinary role that the GSEs have played in the mortgage market since the housing downturn is entirely dependent on taxpayer support.

The government's implicit subsidy to the GSEs arguably led to increased risk-taking in mortgage markets. With an assumed government backstop, the GSEs and the investors that funded them had reduced incentive to ensure that the mortgages they financed could withstand a nationwide decline in house prices, since such an environment was most likely to elicit government support. Given the large impact of GSE activities on the U.S. housing market, the GSEs' bias toward risk helped to similarly distort the incentives of other housing market participants. These risks were spread throughout the financial system as mortgage-backed securities were traded, repackaged, and resold among investors worldwide. In the view of the Richmond Fed, the government's implicit financial safety net was a key component of the financial crisis of 2007 and 2008. (See Our Perspective: Too Big to Fail.)

U.S. taxpayers fund these risks. Federal Reserve Bank of Richmond researchers estimate that Fannie Mae and Freddie Mac constituted one-fifth of the total federal financial safety net — which itself includes 61 percent of the entire financial sector — at the end of 2014 (Marshall, Pellerin, and Walter 2016). Under conservatorship, taxpayers have purchased more than $187.5 billion in GSE stock, and the Treasury has agreed to purchase up to $274 billion additional capital if necessary under the most recently amended conservatorship agreement.

Before conservatorship, the government's support of the GSEs was implicit, meaning it did not appear on congressionally approved budgets. Since the GSEs' September 2008 conservatorship, the government's support of the GSEs is now arguably explicit (See Bailout Barometer: Frequently Asked Questions). Although the assistance given to the GSEs since September 2008 does appear in the federal budget, the total liabilities of the GSEs continue to be omitted from the federal budget. If the full scope of the GSE safety net were made more visible in the budget, taxpayers might very well view its size as unacceptable. The combined liabilities of Fannie Mae and Freddie Mac were $5.2 trillion as of December 2014, roughly 40 percent of the amount of outstanding U.S. Treasury debt held by the public.

Meanwhile, not all the gains of taxpayer support to the GSEs have accrued to homeowners. Economist Wayne Passmore of the Federal Reserve Board of Governors estimated the market value of the GSE subsidy to be between $122 billion and $182 billion (Passmore 2005). Passmore concludes that more than half of those gains are likely retained by GSE shareholders rather than passed through to homebuyers in the form of more affordable mortgage financing, though other estimates vary.

One common argument made in favor of the GSEs is that removing mortgages from the balance sheets of mortgage lenders enables those lenders to offer the 30-year fixed rate mortgage, which is generally viewed as a popular, consumer-friendly way to purchase a home. Historically, branching restrictions on banks made it difficult for them to diversify, so the GSEs offered a way to help banks offload geographically concentrated risks as well as interest rate risk. Since branching restrictions were lifted in 1994, the GSEs now mainly provide relief from interest rate risk, and their comparative advantage relative to private investors in doing so stems from the implicit government subsidy. Thus, if the function of the GSEs were to go away, the 30-year fixed rate mortgage could become more expensive, reflecting the removal of the government's implicit subsidy. At the same time, given the findings of Passmore and others that much of the subsidy has not been passed on to homeowners, it is not clear whether that effect would be large. It is also worth noting that the average duration of homeownership is much shorter than 30 years, so a mortgage of that term may unnecessarily increase the cost of the loan. The vast majority of other developed nations rely on mortgages of much shorter duration, often with adjustable interest rates, and often with higher homeownership rates than the United States (Haltom 2011).

The GSEs have not been reformed since the recent crisis, so the problems caused by their implicit subsidy, which is now arguably explicit, remain intact. The major financial market reform legislation passed in 2010, known as the Dodd-Frank Act, did not include GSE reform. However, there is a consensus among policymakers — illustrated by a February 2011 proposal from the U.S. Treasury and the Department of Housing and Urban Development (U.S. Treasury and HUD 2011 ) — that at least some of the support the two GSEs provide to the housing market must be unwound, potentially by dissolving Fannie Mae and Freddie Mac. A key principle of these discussions has been that private capital must have a larger role in housing finance, as the implicit subsidy provided to the GSEs has historically hindered the ability of private parties to compete.

In the Richmond Fed's view, limiting or eliminating the type of subsidy that the GSEs provide to mortgage finance would be the most effective way to ensure a stable market for home mortgages. Many existing reform proposals, even if they significantly wind down or dissolve the GSEs, retain a role for government guarantees to promote affordable housing (for a recent example, see a February 2013 report produced by the Bipartisan Policy Center). As reform proceeds, a key risk is that an entity or function could emerge in place of the GSEs, that is, one that similarly enjoys the market perception that it would be rescued in the event of failure. Policymakers should work to prevent these expectations from arising. To the extent that the government continues to support housing finance, the terms should be made explicit. Firms receiving support should be charged premiums to offset the subsidy provided to them, and they should be regulated appropriately (Lacker and Weinberg 2010).

Direct Subsidies to Homeowners

There is somewhat less consensus about how to reform other aspects of housing policy. The federal government has a long history of supplying direct-to-consumer support of housing finance, most notably in the form of favorable tax treatment of owner-occupied housing investments relative to financial investments. A key example is the long-standing mortgage interest tax deduction, which has existed since 1913 and was strengthened in the mid-1980s. The president's 2016 budget estimates that this benefit will cost the government $69.5 billion in 2015 and is projected to be the government's second-largest category of income tax expenditures over the coming decade.

A common argument in favor of homeownership, especially for lower-income households, is that it promotes wealth-building since households can build equity by making monthly mortgage payments. However, most of the government's policies, especially the deductibility of mortgage interest, subsidize the accumulation of housing debt rather than equity. This comes with unique risks. While debt may be beneficial to a homeowner if house prices rise, homeowners can find themselves with negative equity if house prices fall, making default and foreclosure more likely.

The mortgage interest tax deduction has also led to unintended distributional effects. The deduction is regressive in nature since wealthier households are more likely to itemize their tax deductions. The wealthy are also more likely to have larger homes and, consequently, larger mortgage interest payments to deduct. In 2006, households in the lowest 60 percent of the income distribution received 3 percent of the mortgage interest deduction proceeds, while the top 20 percent of filers received more than 80 percent (Gale, Gruber, and Stephens-Davidowitz 2007). In addition, theoretical (Cho and Francis 2010) and empirical research (Glaeser and Shapiro 2002) suggests that the mortgage interest deduction doesn't have a large effect on homeownership rates. These findings, combined with the fact that most of the gains from the mortgage interest deduction go to higher-income households, suggest that the tax benefits tend to go to individuals who were already willing to buy a home. In other words, rather than pushing households into homeownership, the mortgage interest deduction may simply encourage people to purchase larger homes than they were already planning to buy.

A better strategy might be to encourage the accumulation of home equity, as through tax-preferred savings vehicles that can be used for a down payment. Other developed countries have had success with savings programs (Haltom 2011). Such programs provide households with the incentive and means to build equity, hence creating the right conditions for the beneficial spillover effects of homeownership. Greater equity also reduces the likelihood of negative equity when house prices fall, thus better insulating the macroeconomy from house price changes.


In the Richmond Fed's view, it is not clear that the United States should continue to devote substantial resources toward subsidizing homeownership. Promoting homeownership, especially in the ways currently done, appears to have the potential for substantially misallocating resources and increasing the risks and difficulties faced by households, putting the overall economy's resilience at risk.

Without government intervention, the proportion of American households that enter into homeownership may be lower than it is today. Housing affordability may be a goal with widespread societal approval, but care is warranted, especially with respect to the empowerment of large, private entities presumed to benefit from a government backstop. Support of the GSEs has directed taxpayer dollars toward subsidizing risk-taking that is likely socially excessive. In addition, it is likely that mortgage tax policy has encouraged consumers to accumulate more mortgage debt and less-diversified portfolios of wealth than they might otherwise hold, while transferring resources regressively. Both types of subsidies, while possibly having helped expand homeownership, have led households and financial institutions to take on risks that can jeopardize the economy at large.

Richmond Fed Research and Speeches

Federal Housing Finance Agency: Conservatorship.

Haltom, Renee. "Foreign Housing Finance." Region Focus, Second Quarter 2011, pp. 12-18.

Karahan, Fatih, and Serena Rhee. “Geographical Reallocation and Unemployment during the Great Recession: The Role of the Housing Bust.” Federal Reserve Bank of New York Staff Reports, No. 605, March 2013.

Lacker, Jeffrey M. "Innovation in the New Financial Regulatory Environment." Speech to the 2011 Ferrum College Forum on Critical Thought, Innovation & Leadership, Roanoke, Va., April 7, 2011.

Lacker, Jeffrey M. and John A. Weinberg. "Now How Large is the Safety Net?" Federal Reserve Bank of Richmond Economic Brief No. 10-06, June 2010.

Marshall, Liz, Sabrina R. Pellerin, and John R. Walter. "Bailout Barometer: How Large Is the Federal Financial Safety Net?" Federal Reserve Bank of Richmond Special Report, February 2016.

Owens, Raymond E., Esteban Rossi-Hansberg, and Pierre-Daniel G. Sarte. "Housing Externalities." Journal of Political Economy, June 2010, vol. 118, no. 3, pp. 485-535.

Slivinski, Stephen. "House Bias: The Economic Consequences of Subsidizing Homeownership." Region Focus, Fall 2008, pp. 12-15.

Other Research

Blanchflower, David G. and Andrew J. Oswald. "Does High Home-Ownership Impair the Labor Market?" Peterson Institute for International Economics Working Paper No. WP 13-3, May 2013.

Caldera Sanchez, Aida and Dan Andrews. "To Move or Not to Move: What Drives Residential Mobility Rates in the OECD?" OECD Economics Department Working Papers No. 846, February 18, 2011.

Cho, Sang-Wook and Johanna Francis. "Tax Treatment of Owner Occupied Housing and Wealth Inequality." Journal of Macroeconomics, March 2011, vol. 33, no. 1, pp. 42-60. (Abstract available online.)

Federal Housing Finance Agency News Release, June 13, 2013.

Ferreira, Fernando, Joseph Gyourko, and Joseph Tracy. "Housing Busts and Household Mobility: An Update." Federal Reserve Bank of New York Staff Reports No. 350, August 29, 2011.

Gale, William, Jonathan Gruber, and Seth Stephens-Davidowitz. "Encouraging Homeownership Through the Tax Code." Tax Notes, June 18, 2007, pp. 1171-1189.

Glaeser, Edward L. and Jesse M. Shapiro. "The Benefits of the Home Mortgage Interest Deduction." National Bureau of Economic Research Working Paper No. 9284, October 2002.

Glick, Reuven and Kevin Lansing. "Global Household Leverage, House Prices, and Consumption." Federal Reserve Bank of San Francisco Economic Letter No. 2010-01, January 11, 2010.

"Housing America's Future: New Directions for National Policy." Bipartisan Policy Center, February 2013.

Passmore, S. Wayne. "The GSE Implicit Subsidy and the Value of Government Ambiguity." Real Estate Economics, Sept. 2005, vol. 33, no. 3, pp. 465-486. (Earlier working paper version available online.)

Passmore, S. Wayne, Shane M. Sherlund, and Gillian Burgess. "The Effect of Housing Government-Sponsored Enterprises on Mortgage Rates." Real Estate Economics, Sept. 2005, vol. 33, no. 3, pp. 427-463. (Earlier working paper version available online.)

"Reforming America's Housing Finance Market: A Report to Congress." U.S. Department of the Treasury and U.S. Department of Housing and Urban Development, February 2011.

"Social Benefits of Homeownership and Stable Housing." National Association of Realtors, April 2012.

Contact Us


Jim Strader
(804) 697-8956
(804) 332-0207 (mobile)

Laura Fortunato
(804) 697-8196
(804) 698-0927 (mobile)