Holtz-Eakin Testimony Before Senate Banking Committee on Housing Finance Reform

Chairman Johnson, Ranking Member Crapo, and Members of the Committee, thank you for the opportunity to appear today. In this testimony, I wish to make three basic points:

  • The legislation being worked on by this Committee represents a necessary and long-overdue effort to address damaging weaknesses in the U.S. housing finance system. The status quo is unsustainable and undesirable.  I applaud the Committee for moving forward;
  • While helping low-income households afford necessary shelter should be a policy priority, and such support should be appropriated by Congress and on the budget.  Budgetary integrity requires that Congress balance the full range of policy priorities; and
  • The affordable housing goals imposed on the housing GSEs were flawed, and should not be repeated

The Need for Reform

Housing finance was at the center of the 2008 financial crisis that visited substantial economic stress on Americans and spawned dramatic government intervention. Yet, over five years later the central actors in the crisis and response – Fannie Mae, Freddie Mac, and the Federal Housing Administration (FHA) – remain essentially unchanged. While the task is politically daunting and mechanically difficult, I applaud the Committee’s desire to undertake these reforms.

Fannie Mae and Freddie Mac

Fannie Mae and Freddie Mac need to be wound down and closed as a matter of both policy and politics. From a policy perspective, the government-sponsored enterprises were central elements of the 2008 crisis. First, they were part of the securitization process that lowered mortgage credit quality standards. Second, as large financial institutions whose failures risked contagion, they were massive and multidimensional cases of the too big to fail problem. Policymakers were unwilling to let them fail because financial institutions around the world bore significant counterparty risk to them through holdings of GSE debt, certain funding markets depended on the value of their debt, and ongoing mortgage market operation depended on their continued existence. They were by far the most expensive institutional failures to the taxpayer and are an ongoing cost.

There is vigorous debate about how big a role these two firms played in securitization relative to “private label” securitizers. There is also vigorous debate about why these two firms got involved in this problem. In the end, this debate need not be fully resolved to recognize that while Fannie Mae and Freddie Mac did not by themselves cause the crisis, they contributed significantly in a number of ways.

The mortgage securitization process turned mortgages into mortgage-backed securities through the government-sponsored enterprises, as well as Countrywide and other “private-label” competitors. The securitization process allows capital to flow from investors to homebuyers. Without it, mortgage lending would be limited to banks and other portfolio lenders, supported by traditional funding sources such as deposits. Securitization allows homeowners access to enormous amounts of additional funding and thereby makes homeownership more affordable. It also can diversify housing risk among different types of lenders. If everything else is working properly, these are good things. Everything else was not working properly.

There were several flaws in the securitization and collateralization process that made things worse. Fannie Mae and Freddie Mac, as well as Countrywide and other private label competitors, all lowered the credit quality standards of the mortgages they securitized. A mortgage-backed security was therefore “worse” during the crisis than in the preceding years because the underlying mortgages were generally of poorer quality. This turned a bad mortgage into a worse security. Mortgage originators took advantage of these lower credit quality securitizations standards and the easy flow of credit to relax the underwriting discipline in the loans they issued. As long as they could resell a mortgage to the secondary market, they didn’t care about its quality.

In addition to feeding poorly-originated mortgages into the system, Fannie Mae and Freddie Mac proved to be so deeply interconnected with the broader financial system that policymakers were forced to step in to prevent their failure. In September 2008, the Federal Housing Finance Agency (FHFA) put Fannie Mae and Freddie Mac into conservatorship. Policymakers in effect promised that “the line would be drawn between debt and equity,” such that equity holders were wiped out but GSE debt would be worth 100 cents on the dollar.

They made this decision because banking regulators (and others) treated Fannie and Freddie debt as equivalent to Treasuries. A bank cannot hold all of its assets in debt issued by General Electric or AT&T, but can hold it all in Fannie or Freddie debt. The same is true for many other investors in the United States and around the world—they assumed that GSE debt was perfectly safe and so they weighed it too heavily in their portfolios. Policymakers were convinced that this counterparty risk faced by many financial institutions meant that any write-down of GSE debt would trigger a chain of failures through the financial system. In addition, GSE debt was used as collateral in short-term lending markets, and by extension, their failure would have led to a sudden massive contraction of credit beyond what did occur. Finally, mortgage markets depended so heavily on the GSEs for securitization that policymakers concluded that their sudden failure would effectively halt the creation of new mortgages. All three reasons led policymakers to conclude that Fannie Mae and Freddie Mac were too interconnected with the system to be permitted to fail.

As a matter of politics, Fannie Mae and Freddie Mac are extremely unpopular and the public supports winding them down. (This section draws on a poll commissioned by the American Action Forum.[1]) The polling shows that a large majority of the voters have a “hard ID” of Fannie and Freddie. They are viewed favorably by only 20 percent and unfavorably by 52 percent.

This is related to another finding, namely that 52 percent of the voters said that their greatest concern is either no accountability of banks and Wall Street or that Wall Street banks are so big that if they fail the taxpayers will have to bail them out again. By a small margin (11 percent) voters are still unfavorable toward the bank bailouts and TARP. Likely for this reason, a majority favor (52 percent) phasing out both Fannie and Freddie.

Greater information sharpens these views. When informed that Fannie and Freddie played an instrumental role in the housing bubble and received nearly $200 billion in a bailout, voters’ opposition to Fannie Mae and Freddie Mac moves to 59 percent. Additionally, the notion that Fannie and Freddie could require more public money in future bailouts is unacceptable to a sizable majority of the voters.

Using Housing Reform to Address Affordability

As the Committee considers housing finance reform, the affordability of shelter naturally arises as a policy concern.  How should this be addressed?  First, the affordable housing goals of Fannie Mae and Freddie Mac were both unsuccessful and dangerous, and should not be repeated. The Housing Finance Reform & Taxpayer Protection Act (S. 1217), being considered by this Committee, does not authorize HUD to set similar goals, but it does aim to preserve them in other ways.

Affordable Housing Goals. Previously, HUD was the mission regulator of the GSEs, setting minimum percentage-of-business housing goals for mortgage purchases. The goals were set to support low-income lending and lending in underserved geographic regions. Specifically, the GSEs had three annual affordable housing goals:

  1. Low- and Moderate-Income Goal: targets borrowers with income no greater than area median income (AMI)
  2. Special Affordable Goal: targets very low-income borrowers (less than 60 percent of AMI) and low-income borrowers (less than 80 percent of AMI) living in low-income census tracts
  3. Geographically-Targeted or Underserved Areas Goal: targets low-income and high-minority neighborhoods

Overall, Freddie Mac consistently lagged the conventional market in funding the three groups targeted by HUD’s housing goals. Fannie Mae, while performing better than Freddie Mac, performed on average below market levels except over the 2001-2003 period when it only lagged the market in funding underserved area loans.[2][3] Yet despite failing to meet their goals, HUD added subgoals and consistently raised the purchase percentages of existing goals to ever-higher levels.

Though inclusion of affordability goals did not singularly undermine the GSEs’ profitability or cause their demise, the goals complicated their mission, blurring public purpose and pursuit of private profits. More importantly, the goals were not successful. Evidence shows the other lenders were buying more loans that met the goals than the GSEs themselves. Furthermore, if the purpose of the goals was to provide shelter, it did not have to take the form of owner-occupied homes.  If the goal was to promote homeownership, the GSEs failed by concentrating more on homebuyers with homes already (not first-time homebuyers) and on homeowners refinancing.[4]

Funding the National Affordable Housing Trust Fund. Provisions under Title IV of the proposed legislation would replace the failed affordable housing goals of the GSEs with a user fee of 5-10 basis points on FMIC-insured securities to fund the Housing Trust and Capital Magnet Funds. Instituting that surcharge presents three main problems:

  • The cost of contributions may be passed on to mortgage borrowers, raising the cost of mortgage rates even further than the proposed legislation likely will already.  That is, it works against the goal of affordable mortgage finance;
  • Fund advocates have an increased incentive to encourage the FMIC to grow beyond its chartered purpose; and
  • Revenues from the surcharge and the effectiveness of the ways they are spent will not be subject to the scrutiny and approval inherent in the budget and appropriations process.

Advocates for a national affordable housing trust fund – e.g., the National Low Income Housing Coalition – have stated the importance of “[a] dedicated source of funding not subject to the annual appropriations process.”[5] Why?  Despite the woes of the congressional budget process, it would fundamentally diminish the accountability of the Housing Trust and Capital Magnet Funds to which the fee is allocated if they stay outside of the regular appropriations process. A fundamental lack of transparency and accountability underscored the GSE model and continues to trouble the FHA. It would be a core policy error to make a similar mistake in the funding of these trusts included in the proposed legislation.

Preserving Support for Small Institutions & Underserved Communities. Support for all geographic regions of the country, whether an underserved rural or urban community, has been a mainstay of bipartisan proposals for housing finance reform. Additionally, equal access to the secondary market for institutions of all sizes is important to widespread availability of affordable credit.

S. 1217 contains several provisions that aim to preserve support for small institutions and underserved communities. First, the legislation mandates in Title II that the FMIC “help ensure all geographic locations have access to mortgage credit.” The FMIC is also charged with facilitating securitization for credit unions and community and mid-size banks. In developing products, structures, contracts, etc., the FMIC must take into account how their actions would affect small financial institutions, the availability of credit, and equitable access to secondary mortgage market financing for lenders of all sizes and locations including those in rural communities. The legislation also requires fee uniformity for institutions purchasing insurance from the FMIC regardless of location or size of the institution. Finally, if concerns exist about the viability of those aforementioned duties, those concerns must be submitted in a report to Congress.

Reforming the FHA. Housing finance reform, whether S. 1217 or another piece of legislation, should consider undertaking needed reforms to FHA. While the GSEs supported affordable housing through goals sets by HUD, those goals worked to undermine the housing finance system. Following the housing bust and entry of the GSEs into conservatorship, significant market share shifted to FHA, causing unprecedented financial losses for the agency.

Some of the potential reforms – for example, as embodied in the PATH Act – would narrow FHA’s mission to support only low-income and first-time homebuyers. With a mix of income-based borrower requirements and revised loan limits, the FHA would more adequately address a demonstrated need among low-income communities and first-time homebuyers while also enhancing the role of the private market. Addressing FHA in conjunction with the wind down of the GSEs is preferable because of how easily misaligned prices, limits, and standards can shift market share between government-backed programs. S. 1376, the FHA Solvency Act of 2013, considered by this Committee, does not include similar income-based borrower requirements, fundamental changes to the administration of FHA, fair-value and GAAP accounting requirements, or tightened loan limits present in the House legislation.

The Dodd-Frank Act & Affordability. Housing finance reform must also take into account how Dodd-Frank regulations deeply affect affordability.  Eligible mortgages under the S. 1217 are limited to qualified mortgages (QM) as defined by regulators. A recent analysis of Home Mortgage Disclosure Act (HMDA) data by economists at the Federal Reserve found that 22 percent of borrowers in 2010 had debt-to-income ratios above the 43 percent standard set by the QM rule.[6] 70 percent of borrowers above that threshold had FHA, VA, or RHS insured loans. Those borrowers with debt-to-income ratios above 43 percent were also most likely to be lower-income, black, or Hispanic-white. This exemplifies just how easily regulations enacted from the Dodd-Frank Act can severely can impact affordability among lower-income communities and communities of color. Under the new regulatory regime, it will be either costly or impossible for private capital to back high LTV or non-QRM lending to fill the government’s role.

AAF previously estimated that the bottom line effects of proposed Dodd-Frank and Basel III regulations may include 20 percent fewer loans, resulting in 600,000 fewer home sales. In turn, the resulting tightened lending and reduced sales were estimated to cost up to 1,010,000 housing starts, 3.9 million fewer jobs, and a loss of 1.1 percentage points from GDP growth over the next three years.[7] While some of those regulations have been revised, the reality of tightened credit and its potential effect on the economy remain largely the same.

Another study by the National Association of Realtors estimated that Dodd-Frank regulations could raise mortgage rates by 75-125 bps for non-QRM, high LTV borrwers and Basel III could raise rates by 80 bps.[8] These estimates of the effect regulations have on mortgage rates are equivalent or higher than the effect economist Mark Zandi estimated S. 1217 would have on mortgage rates.[9] To not take into account how Dodd-Frank regulations affect affordability in a future system would be a mistake.

Redundant Government Support for Affordable Housing

The federal government provides multiple avenues of support for the construction of affordable housing and assistance for low-income renters and homebuyers. While the success of these programs can be debated, support exists and does so in many forms; primarily the federal government provides appropriated funding through more than 30 programs within the Department of Housing and Urban Development (HUD), tax credits and deductions for both corporations and individuals, housing programs for veterans through the Department of Veterans’ Affairs (VA), rural housing programs through the Department of Agriculture (USDA), the mortgage insurance programs of the Federal Housing Administration (FHA), and government corporation Ginnie Mae. In attempting to overhaul the housing finance system, it would be wise to address the issue of affordability through the reform or increased support of existing programs and not through legislation addressing the mechanics of the government’s role in the secondary market.

Last year the Government Accountability Office (GAO) concluded that the federal government “incurred about $170 billion in obligations for federal assistance and estimated forgone tax revenue in fiscal year 2010.”[10] The GAO found housing assistance fragmented across 160 programs and activities. Table 1, which does not include any tax expenditures, provides an example of the number of programs that exist across agencies. (Note: housing programs are not limited to those in Table 1.)

According to the Joint Committee on Taxation, tax expenditures subsidizing housing, for individuals and corporations, will cost the federal government $868 billion in foregone taxes from 2012 to 2017 (See Table 1). While many of these credits, deductions, and exemptions favor owner-occupied housing over rental housing, that problem can and should be addressed through tax reform.

The redundancy of federal support of housing calls into question the efficiency and effectiveness of government efforts to assist low-income renters and homebuyers and maintain affordability. Tackling affordability requires both a dedication to greater job and wage growth, as well as a renewed look at what improvements can be made to existing federal housing programs.

The effects of reform legislation on affordability will ultimately depend on the system put in place; whether the new system has many players, maintains the flow of mortgage credit and protects taxpayers from liabilities is more important than any fund or single program. The most important components of housing finance reform are the mechanics of a new system and how to transition to it.


Tenant Based Rental Assistance



Project-Based Rental Assistance



Community Development Fund



Public Housing Operating Fund



Public Housing Capital Fund



Homeless Assistance Grants



Home Investment Partnership Program



Housing for the Elderly



Housing Certificate Fund



Native American Housing Block Grant



Other Assisted Housing Programs



Housing Opportunities for Persons with AIDS



Housing for Persons with Disabilities



Emergency Homeowners’ Relief Fund




Self-Help Homeownership Opportunity Program



Indian Housing Loan Guarantee Fund Program Account



Rural Housing & Economic Development


Permanent Supportive Housing


Manufactured Housing Trust Fund



Native Hawaiian Housing Block Grant



Choice Neighborhoods Initiative


Native Hawaiian Loan Guarantee Program

Housing Trust Fund

Capacity Building

Rental Housing Assistance Fund


Flexible Subsidy Fund


Nehemiah Housing Opportunity Fund

Rural Housing Insurance



Rental Assistance Program



Multifamily Housing Revitilization Program



Rural Housing Assistance Grants



Mutual & Self-Help Housing Grants



Grants to State for Low-Income Housing Projects in Lieu of Low-Income Housing Tax Credit Allocations


TARP Housing Programs


Capital Magnet Fund


Veterans Housing Benefit Program



Native American Veteran Housing Loan Program



Source: OMB

TABLE 2. TAX EXPENDITURE ESTIMATES, FY 2012-2017 (Billions of Dollars)



Deduction for Mortgage Interest on Owner-Occupied Residences


Deduction for Property Taxes on Real Property


Exclusion of Capital Gains on Sales of Principal Residences


Exclusion of Interest on State & Local Government Qualified Private Activity Bonds for Owner-Occupied Housing



Deduction for Premiums for Qualified Mortgage Insurance

> 0.5

Exclusion of Income Attributable to the Discharge of Principal Residence Acquisition Indebtedness


Credit for Low-Income Housing



Credit for Rehabilitation of Historic Structures



Credit for Rehabilitation of Structures, Other than Historic Structures



Exclusion of Interest on State & Local Government Qualified Private Activity Bonds for Rental Housing



Depreciation of Rental Housing in Excess of Alternative Depreciation System






Source: Joint Committee on Taxation[11]

[1] American Action Forum, “AAF Releases New Poll of Public Attitudes on Fannie, Freddie Mac, & Housing Reform,” (July 2013); http://americanactionforum.org/topic/aaf-releases-new-poll-public-attitudes-fannie-mae-freddie-mac-and-housing-reform

[2] Bunce, Harold L., Office of Policy Development & Research, U.S. Department of Housing & Urban Development, “Working Paper No. HF-018. The GSEs’ Funding of Affordable Loans: A 2004-2005 Update,” (June 2007); http://www.huduser.org/portal/publications/pdf/workpapr18.pdf

[3] FHFA, “Mortgage Market Note 10-2 – The Housing Goals of Fannie Mae & Freddie Mac in the Context of the Mortgage Market: 1996-2009,” (February 2010); http://www.fhfa.gov/webfiles/15419/Housing_Goals_1996-2009_02-01.pdf

[4] John C. Weicher, “The Affordable Housing Goals, Homeownership and Risk: Some Lessons from Past Efforts to Regulate the GSEs,” Conference on “The Past, Present & Future of the GSEs,” Federal Reserve Bank of St. Louis, (November 2010); http://research.stlouisfed.org/conferences/gse/Weicher.pdf

[5] National Low Income Housing Coalition, “National Housing Trust Fund;” http://nlihc.org/issues/nhtf

[6] Neil Bhutta & Glenn B. Canner, Division of Research & Statistics, Federal Reserve, “Mortgage Market Conditions and Borrower Outcomes: Evidence from 2012 HMDA Data and Matched HDMA-Credit Record Data,” (September 2013); http://www.federalreserve.gov/pubs/bulletin/2013/pdf/2012_HMDA.pdf

[7] Douglas Holtz-Eakin, Cameron Smith & Andrew Winkler, “Regulatory Reform and Housing Finance: Putting the ‘Cost’ Back in Benefit-Cost,” (October 2012); http://americanactionforum.org/sites/default/files/Regulation_and_Housing.pdf

[8] National Association of Realtors, “Recent Lessons for the QRM,” (December 2011); http://economistsoutlook.blog.realtor.org/2011/12/08/recent-lessons-for-the-qrm/

[9] Mark Zandi & Cristian deRitis, “Evaluating Corker-Warner,” (July 2013); http://www.economy.com/mark-zandi/documents/2013-07-08-Evaluating-Corker-Warner.pdf

[10] Government Accountability Office, “Housing Assistance: Opportunities Exist to Increase Collaboration and Consider Consolidation,” (August 2012); http://gao.gov/assets/600/593752.pdf

[11] Joint Committee on Taxation, “Estimates Of Federal Tax Expenditures For Fiscal Years 2012-2017,” (February 01, 2013); https://www.jct.gov/publications.html?func=select&id=5