Marvin Bower, the great managing partner of McKinsey & Co. in the 1950s, offered this advice: Direct every action toward building a stronger position for the long term, but take the action now. I think we can apply this advice to addressing the government-sponsored enterprises, Fannie Mae and Freddie Mac, and the huge distortions and losses they have caused.
We should aim in the long term for a housing finance sector which is principally a robust private market, and one in which you can be either a private company, or you can be a government agency, but you can't be both. In other words, in the long run, there should be no GSEs.
As we all know, the GSE charters meant that Fannie and Freddie were private corporations with an implicit, but absolutely real, taxpayer guaranty; they were granted many special privileges and very large economic subsidies, which privatized profits and socialized risk, and as a result, have passed giant losses to the taxpayers. The GSEs were highly politicized, exercised duopoly market power, and transferred a portion of their subsidies to politically directed housing programs, thereby escaping the democratic discipline of appropriations. They were described as "Masters of Beltway Capitalism." Fannie, in particular was genuinely feared as a hard-ball political operator. They had especially high leverage and low capital ratios, because their real capital was known by the bond market to be the credit card of the U.S. Treasury.
A private company is subject to market discipline; a government agency is subject to government discipline: GSEs escape both. Virtually everyone now agrees that they must be fundamentally reformed. Of course, prescriptions differ, but no mortgage sector reform which fails to address Fannie and Freddie can be meaningful.
My view is that in the long run, Fannie and Freddie need to be divided into a "bad bank," a "good bank" and a government agency. The bad bank should be put into a liquidating trust, the good bank should be privatized, and the governmental activities of delivering subsidies and nonmarket loans should be merged into the structure of the Department of Housing and Urban Development.
Such a restructuring would certainly be complex, and while we are still mired in the aftermath of the bubble which Fannie and Freddie did so much to help inflate, probably can't be done yet. But there are a number of focused, specific actions we could take now, consistent with our long-term aim. I suggest a dozen of them.
1. Enable Covered Bonds as an Alternate Long-Term Mortgage Funding Option
A principal challenge of mortgage finance is developing long term funding for mortgages, and for fixed rate mortgages, in particular. To help develop a more robust and competitive housing finance sector, alternatives to Fannie and Freddie securitization should be encouraged.
This should include creating a statutory basis for a U.S. covered bond market, as has been done in many other countries, in some cases of very long historical standing. Legislation is required, I believe, to protect the covered bondholders' rights to the relevant collateral; regulatory policies are insufficient, subject to change, and suffer from a conflict with the self-interest of the managers of the deposit insurer.
A lesson everyone has learned from the bubble is the importance of whether mortgage lenders retain "skin in the game" for the mortgage credit they originate. With covered bonds, the issuing bank has 100% "skin in the game" credit responsibility-a major advantage over the GSE "originate and sell" model.
2. Set a Sunset Date for Fannie and Freddie's Charters
Granting perpetual charters to GSEs was a major historical mistake-all GSEs should have limited life charters, so the Congress must periodically re-examine their structure, privileges, risks and usefulness. I suggest a five-year sunset be put on Fannie and Freddie's charters, thus having them expire in 2016. Before then, we will be ready for their long term restructuring and being divided into a truly private business, a fully governmental agency, and a liquidating trust.
3. Bring GSE Capital Requirements Up to Those of National Banks
Although Fannie and Freddie's own capital (excluding taxpayer contributions) is hugely negative, we should formally end their history of being allowed to run with much less capital and much more leverage than was prudent, as events have conclusively demonstrated. Very low capital requirements also gave them a major competitive advantage vs. private actors, thus causing risk to become concentrated in the GSEs.
Congress should instruct the GSE regulator to set Fannie and Freddie's capital requirements at no less than those applied to national banks for the same assets and risks. This is basic financial prudence, and would also help prepare for the ultimate privatization of their "good banks."
4. Mandate the Run-Off of the GSEs' Investment Portfolios
The GSEs' investment portfolios in loans and securities serve little purpose except to arbitrage the government's credit. This is particularly true of their investing in their own MBS (with "Kafkaesque circularity," to use of phrase of former full committee chairman Jim Leach), by issuing unsecured, taxpayer guaranteed debt. This debt was sold all over the world as "Treasury credit with a higher yield," which indeed it was, just as the bond salesmen said. Now ordinary Americans are being taxed so that the foreign bondholders can be paid at par.
5. Set a Regular, Predictable Reduction in GSE Conforming Loan Limits
No private entity can compete with the government-granted advantages of the GSEs, which are even greater now that they have no capital of their own left to worry about. But the future American mortgage finance sector should have a robust private market for its largest segment: prime, middle class, conforming loans. In this segment, private investors should put private capital at risk, and prosper of lose as the case may be. Risks here are fully manageable; no taxpayer subsidies or risk exposures are needed or desirable.
A private secondary market for prime, middle class mortgages would have been a natural market development, but cannot develop while the GSEs wield market power derived from their government-granted advantages.
6. Mandate Clear Federal Budget Accounting for Fannie and Freddie
At present, as we all know, Fannie and Freddie are wards of the state, owned overwhelmingly and totally controlled by the government. Their debt is debt of the government; their obligations are obligations of the government. This ought to be accordingly accounted for on the government's books.
As proposed in the "Accurate Accounting of Fannie Mae and Freddie Mac" bill introduced in 2010, the GSEs should be accounted for on-budget and be subject to the Gramm-Rudman-Hollings At of 1985 and the Credit Reform Act of 1990.
Obviously, this would no longer apply to the "good banks" after they are privatized.
7. Prohibit Lobbying by Fannie and Freddie Until Privatization
The prohibition of GSE lobbying should be made statutory until they are privatized. The biggest mistake we could make would be to recreate the "masters of beltway capitalism" with their former political clout.
8. Eliminate All GSE Affordable Housing Goals and Transfer Such Goals to HUD
Public subsidies for affordable housing and nonmarket, higher-risk lending should be explicitly governmental activities, carried out by the Department of Housing and Urban Development, and subject to the governmental disciplines of oversight and appropriations. They should be not financed by taxpayer subsidies or, as it is now, taxpayer capital, moving thorough the GSEs.
9. Prohibit Double Leveraging of GSEs by the Banking System
One of the big mistakes made by bank regulation was to use the banking system to increase the systemic risk of the GSEs. Banks and thrifts were encouraged to hold disproportionate amounts of GSE credit and debt, without exposure limits, and moreover to invest in the equity of Fannie and Freddie in the form of GSE preferred stock. The banks could leverage this equity investment 60-to-1 for risk-based capital (that is, it had a 20% risk weighting), which is equivalent to buying stock on 98% margin. The result was that insured deposits were used to leverage GSE equity. The GSEs then again leveraged the preferred stock 60-to-1. The actual leverage of the combined preferred stock structure was thus 60-squared-to-1.
This was double-dipping on the government guaranty, and more importantly, was double leverage of the GSEs. In other words, if you combine the capital of the system of GSEs plus banks to understand the systemic risk, most of the GSE equity held by the banks disappears as a consolidating elimination. Thus, the GSE system was even more leveraged than it appeared to be because of how it used the banks.
10. Require a One-Page Mortgage Information Form for All Loans Guaranteed by GSEs
Everyone agrees about one improvement sorely needed by American mortgage finance: the provision of clear, simply stated, straightforward key information to the prospective borrowers about the debt commitments they are considering entering into.
This can and should be accomplished in one page, with the addition of some avuncular explanations of the vocabulary of mortgages. The form should help the borrower answer this fundamental question: "Can I afford the payment commitments I would be making?"
It has been hard to get this obviously good idea implemented, but the GSEs offer us a direct way to do so: Mandate that no loan can be guaranteed by Fannie or Freddie which has not provided the borrower with the appropriate one-page information form.
11. Consider Requiring GSE Approval for the Addition of Second Liens
Second mortgages have made coping with the collapse of the housing bubble very much more difficult, in addition to having contributed to its hyper-leveraged expansion.
12. Remove the Taxpayer Guaranty of GSE Subordinated Debt
An outrageous part of the GSE bailout was the full protection, at the expense of the taxpayers, of the holders of Fannie and Freddie's subordinated debt. This subordinated debt was sold with the explicit purpose of introducing market discipline into the debt behavior of the GSEs. The theory obviously did not work, but the risk of the subordinated status was fully articulated to the buyers of this debt. They need to experience the market discipline intended, but so far lacking.
Alex J. Pollock is a resident fellow at AEI.