Last month the head of a Washington D.C.-based affordable housing organization asked me to attend a meeting in the Hart Senate Office Building to give my views on “Corker-Warner 2.0” and its potential effect on low- and moderate-income mortgage borrowers. As I presented my perspective and arguments, a Senator with whom I had not met before seemed interested in and intrigued by them, and asked me if I could write them up. I did, and sent the resulting paper to the Senator’s staff. Since affordable housing has been a key sticking point for reform legislation in Congress, I thought readers of this blog would be interested in the text of the paper, which I’ve reproduced below.
* * *
The need for reform
For the past several decades the U.S. has had two general sources of financing for conventional (i.e., non-government guaranteed) home mortgages: depository institutions and the international capital markets. Depository institutions finance home mortgages using consumer deposits and purchased funds; capital markets investors finance them by purchasing mortgage-backed securities (or MBS).
In the mid-1970s three-quarters of all home mortgages were made and held by depositories, mainly thrift institutions. Two successive thrift crises—the first in the late 1970s triggered by deposit deregulation and the second in the late 1980s triggered by asset deregulation—caused that system to collapse. Thrifts’ share of home mortgages financed plunged from 57 percent in 1975 to just 17 percent in 1995. Financing by Fannie Mae and Freddie Mac, sourced from the international capital markets, filled almost the entire gap. At the end of the 1990s, Fannie and Freddie either owned or guaranteed more than 40 percent of all home mortgages, up from less than 5 percent 25 years earlier.
In the early 2000s, during a time when the Federal Reserve and Treasury were embracing free market principles and declining to regulate new lending practices or mechanisms, private-label securities (PLS) emerged as an alternative to mortgage securitization by Fannie and Freddie. PLS placed few limits on the riskiness of the loans they would accept, and as a consequence issuance of PLS increased rapidly. In 2004, more new mortgages were being financed by PLS than by Fannie, Freddie and Ginnie Mae MBS combined. Easy, ample and low-cost borrower access to mortgages through unregulated PLS issuance fueled an unsustainable housing boom that continued until the fall of 2007, when the PLS market finally collapsed amidst an avalanche of delinquencies and defaults.
We now have nearly ten years’ worth of performance data on mortgages financed just prior to the crisis by PLS, FDIC-insured depository institutions, and Fannie and Freddie. These data leave no doubt as to where the flaws in the mortgage finance system were. Cumulative loss rates on the PLS that were outstanding at the end of 2007 exceed 25 percent, and loss rates on home loans made and held by depository institutions as of the same date exceed 12 percent. In contrast, loss rates on year-end 2007 loans held or guaranteed by Fannie and Freddie barely exceed 4 percent.
Capital markets investors and Congress understood what had happened in the PLS market—where an absence of regulation allowed both primary market lenders and Wall Street securities issuers with no capital at risk to make money off mortgages that had little chance of being repaid—and they reacted accordingly. Congress responded with the 2010 Dodd-Frank Act, which required lenders to apply an “ability to repay” rule to mortgage borrowers and, through the “qualified mortgage” (QM) standard, effectively prohibited the most risky mortgage products and loan features that proliferated during the PLS bubble. Investors simply abandoned the PLS market due to its inherent conflicts of interest, and they have not come back
From 2004 through 2006, the PLS market had been financing nearly two of every five new home mortgages in the country. When it suddenly imploded, that left the commercial banks, Fannie and Freddie, and the FHA to pick up the slack. All were under stress because of the overheated housing market, calling into question how, and even whether, the mortgage market could continue to function.
What happened at this point is disputed, but compelling evidence supports the notion that Treasury Secretary Henry Paulson made a policy decision to use what he termed “the awesome power of the government” to force Fannie and Freddie into conservatorship—without statutory authority, against their will, and while they remained in compliance with their regulatory capital requirements—because, as he told the Financial Crisis Inquiry Commission, “[Fannie and Freddie] were the only game in town,” and with mortgage credit drying up, they “more than anyone, were the engine we needed to get through the problem.”
At the same time, Treasury seems to have found in the turmoil of the financial crisis a unique opportunity to strengthen the competitive position in the $10 trillion home mortgage market of the commercial banks it regulates, at the expense of Fannie and Freddie and the capital markets investors to whom they sell their securities.
As soon as the companies were put in conservatorship, Treasury required them to shrink their combined $1.6 trillion mortgage portfolios by 10 percent per year (later increased to 15 percent per year), even though the spread income from those portfolios helped offset their credit losses. Shortly afterwards, Treasury and the companies’ conservator, the Federal Housing Finance Agency (FHFA), used highly pessimistic estimates of future losses to justify recording $320 billion in non-cash expenses that exhausted the companies’ capital and caused them to draw $187 billion in senior preferred stock they did not need and were not allowed to repay. Then, just as the artificial losses that caused this alleged “bailout” were about to reverse (as they did, in only 18 months), Treasury and FHFA agreed to a “net worth sweep” that required Fannie and Freddie to pay all of their net income to Treasury in perpetuity, ensuring that they would remain in conservatorship until Treasury and advocates for the banking industry could come up with a plan to replace them.
Affordable housing borrowers are critically dependent on a low-cost and efficiently functioning secondary market. Prior to the 2008 financial crisis, Fannie and Freddie had been able to support these borrowers by using cross-subsidization to price their credit guarantees more advantageously, and by purchasing nonstandard affordable housing loans for portfolio. Today, Fannie and Freddie’s portfolios are less than one- third their former size—because of Treasury’s directive to shrink them—and they operate with guaranty fees set artificially high by their conservator, FHFA, not based on the riskiness of the loans they guarantee but instead, in the words of the agency, to “reduce their market share,” and “encourage more private sector participation.”
Since the mid-1990s more funding for mortgages has been provided by capital markets investors than by depository institutions, so primary market lenders use the cost of selling into the secondary market to determine their quoted mortgage rates. Increases in the cost of the credit guaranty process, i.e., MBS guaranty fees, therefore get passed on to all borrowers, whether their loans are sold or not.
Dodd-Frank reformed both the PLS market and primary market mortgage lenders through the QM standard and the ability-to-repay rule, and regulators have raised banks’ capital requirements. Only Fannie and Freddie remain unreformed, because advocates for the banking industry insist not on reform that benefits the financial system and homebuyers but on reform that benefits the competitive position of primary market lenders. With the PLS market dormant, Fannie and Freddie are virtually the sole means of tapping the capital markets for money to finance conventional mortgages. Holding this financing channel hostage to banks’ demands that it be restructured in a way that makes it more costly and less efficient—for the patently false reason that its current structure is a “failed business model” that caused the financial crisis—has had severe negative consequences for the mortgage system as a whole, and for affordable housing borrowers in particular. Congress and Treasury are the only entities that can remedy this situation.
Support for affordable housing
The government can increase the availability and lower the cost of mortgages for affordable housing in three ways: (a) fostering the development of a secondary market system that provides the lowest-cost mortgages to the widest range of borrower types, consistent with an agreed-upon standard of taxpayer protection; (b) setting mandatory affordable housing goals, and (c) devising and implementing subsidy programs that assist underserved populations. Taking each in turn:
A credit guaranty mechanism that expands borrower access
Advocates for commercial banks have sought legislative replacements for Fannie and Freddie for nearly a decade. Their idea in the 2013 Corker-Warner bill was a cumbersome bureaucracy called the Federal Mortgage Insurance Corporation; today the new version of Corker-Warner calls for multiple credit guarantors with explicit government guarantees on the securities they issue. Both versions, however, offer self-serving solutions to invented problems, and fail because Fannie and Freddie’s business model, with its proven track record of providing efficient and low-cost access to capital markets funding, works far better than the proposed alternatives.
Yet as the 2008 financial crisis revealed, there are weaknesses in this model that can and should be remedied. The companies need a new and more effective capital standard; they need better (and less adversarial) regulation, and they perhaps also should have utility-like limits on their returns, as a risk-control measure. These reforms can be done either in legislation or administratively. Whichever course is followed, however, the most critical aspect of Fannie and Freddie reform for access and affordability will be the capital standard.
The capital standard imposed upon Fannie and Freddie, or any credit guarantor, must strike a careful and deliberate balance between a very high level of taxpayer protection on the one hand and the cost and breadth of access to mortgages on the other. Too little capital will expose the taxpayer to potential losses; too much capital, or a standard that does not tie capital to credit risk, will distort guaranty fee pricing and impede the flow of capital market funds into mortgages, for affordable housing borrowers in particular.
Fortunately, Congress already has addressed the issue of credit guarantor capital, and come up with the right answer. Section 1110 of the Housing and Economic Recovery Act (HERA) of 2008 states, “The Director [of FHFA] shall, by regulation, establish risk-based capital requirements for the enterprises to ensure that the enterprises operate in a safe and sound manner, maintaining sufficient capital and reserves to support the risks that arise in the operations and management of the enterprises.” A true risk-based capital standard for Fannie and Freddie, consistent with HERA, will result in the least amount of unnecessary capital, the lowest average guaranty fees, and the greatest flexibility to use cross-subsidization to support the affordable housing community, while still providing an extremely high level of taxpayer protection.
FHFA has not yet implemented the capital provision in HERA, likely because it and Treasury have chosen to manage Fannie and Freddie in conservatorship in a manner consistent with ultimately winding them down and replacing them. That is a serious policy mistake that needs to be corrected by Congress, or by Treasury itself.
To implement the risk-based standard in HERA, Congress or Treasury first would pick the stress environment it wants the guarantors to be able to protect against. (The 25 percent nationwide decline in home prices used in the Dodd-Frank stress tests for banks would be a likely candidate). FHFA then would use Fannie and Freddie’s historical data to project default rates and loss severities during this environment, by risk category—at a minimum, combinations of loan-to-value ratios and credit scores—and the resulting stress loss amounts would be the basis for setting the companies’ new capital requirements, at the risk-category level.
Fannie and Freddie’s experience during and after the 2008 financial crisis suggests how they would have to capitalize against a future 25 percent home price decline. With their loans and MBS at the end of 2007, and the (low) guaranty fees they then were charging, 2 percent capital would have been more than enough to absorb all of the credit losses on those books of business. FHFA of course will set the actual stress capital levels. It also should impose a minimum capital ratio, which will benefit affordable housing borrowers by making it impossible for Fannie or Freddie to drive their required capital below that minimum by unduly favoring pristine credits.
A true risk-based capital standard for Fannie and Freddie, endorsed by Treasury as meeting a rigorous stress standard for taxpayer protection, would obviate the need for an explicit government guaranty on their mortgage-backed securities. Moreover, if catastrophic loss insurance were necessary to enhance Fannie and Freddie’s credit quality (to hold down their MBS yields), private insurers could provide it; with true risk-based capital, insurers would know the precise thresholds, by risk category, at which their loss coverage would kick in, and could price for it accordingly.
Mandatory affordable housing goals
The potential impact of mandatory affordable housing goals for Fannie and Freddie, or any credit guarantor, is limited by the fact that these companies do not originate mortgages; they only can guarantee the loans primary market lenders make. Even so, housing goals can be used to create incentives for secondary market companies to channel the benefits of capital markets financing to subcategories of underserved borrower. A regulator should have the authority to impose penalties on a credit guarantor if, in any year, its percentage of affordable housing business—either overall or in an underserved subcategory designated by Congress—falls short of the percentage originated in the primary market that year.
Housing goals also can be used to promote cooperative initiatives with state and local governments, nonprofit institutions and community housing groups to create custom-tailored products for targeted housing needs on a smaller scale, although for this to be effective a credit guarantor will need to be able to hold some number of these loans on its balance sheet, since many will not be eligible for securitization.
HERA contains a robust set of housing goals for Fannie and Freddie, which would be maintained in administrative reform of the companies. Mandatory credit guarantor housing goals may be more difficult to include in new legislation, given the political disagreements over their appropriate role.
Subsidy programs that support underserved populations
The bank-promoted bill currently being drafted in the Senate Banking Committee relies almost exclusively on direct subsidy programs to assist affordable housing borrowers, funding these programs with a ten basis point Mortgage Access Fee.
Supporters of this mechanism are correct that direct subsidy programs are more efficient, on a dollar-for-dollar basis, than the cross-subsidies historically done in the secondary market through charging lower-risk borrowers higher guaranty fees so that guaranty fees for higher-risk borrowers can be reduced. But the draft Senate bill errs in presenting direct subsidies as an alternative to cross-subsidies, rather than as a complement to them. The two are independent; direct subsidy programs do not in any way lessen the importance of doing as much as possible to support affordable housing through an efficient credit guaranty mechanism.
The draft Senate bill also proposes to assess the Mortgage Access Fee only on loans that have secondary market credit guarantees. This creates two problems, both of which can be fixed by assessing the fee on all new mortgages. The first is transmittal leakage. Because a credit guarantor’s ten basis point Mortgage Access Fee will be added to its guaranty fee, the same amount also will be added to the mortgage rates quoted by primary market lenders, for the reason noted earlier. But these ten basis points paid by the borrower only will enter the affordable housing subsidy pool if the loan receives a credit guaranty; a lender that keeps a loan in portfolio also keeps the borrower’s ten basis points. A Mortgage Access Fee on all new mortgages will avoid this leakage (and windfall for primary market portfolio lenders), and ensure that all monies collected from the fee go to their intended beneficiaries.
The second problem with a Mortgage Access Fee charged only on loans financed in the secondary market is that it unnecessarily ties the subsidy programs that benefit from such a fee to legislative mortgage reform, which will be challenging to enact. A Mortgage Access Fee on all new mortgages would be less controversial as a stand-alone measure, and thus more likely to be implemented.
* * *
How best to address the needs of the affordable housing community in mortgage reform is no mystery. But that hasn’t been the goal of secondary mortgage market reformers—it’s dominance of the $10 trillion residential mortgage market by large primary market banks. With Fannie and Freddie banned from political activities, these banks have been free to distort the record, define the objectives of secondary market reform, and put forth reform proposals that benefit themselves. Advocates for affordable housing, as well as community banks, realize this, and are responding with fact-based rebuttals and alternatives of their own. That’s a very good thing. If large primary market lenders are allowed to dictate how the capital markets-based financing channel is structured and operates, homebuyers in general, and affordable housing borrowers in particular, will pay a steep price in both cost and access.