- This summer has brought the U.S. two bills in Congress at last addressing the long-festering problem of what to do with Fannie and Freddie.
- The existence of a government guarantee will cause excess debt and excess leverage flow to real estate.
- In the intense politics of American housing finance, both bills agree on ending the disastrous political experiment of Fannie and Freddie.
The summer of 2013 has brought the U.S. two bills in Congress, one in the House and one in the Senate, at last addressing the long-festering problem of what to do with Fannie Mae and Freddie Mac: both propose an American housing finance sector without them. This would be a big change, since Fannie and Freddie are involved with about $5 trillion of the $10 trillion of U.S. first mortgage loans.
The Senate bill, called after Senators Corker and Warner, who introduced it, is self-consciously bi-partisan, and sets out with impressive success to give the maximum number of relevant interest groups something they want. The House bill, passed by the Financial Services Committee with almost all Republicans voting for and all Democrats voting against, focuses on bringing much more robust private markets to American housing finance.
The historical context of these bills is critical: the American housing finance system has collapsed in disgrace twice in three decades, in the 1980s and again in the 2000s. This is a terrible record for the world’s largest economy and largest housing finance market.
That we have perhaps learned something from these remarkable failures of former housing finance structures is suggested by the number of essential provisions that the two bills have in common:
1. Both bills would end Fannie Mae and Freddie Mac as government-sponsored enterprises or “GSEs”. Both have concluded that the fundamental GSE model, which means trying to pretend that you are simultaneously a private company and part of the government, and which results in privatized profits and socialized losses, is insidious and needs to be definitively terminated.
2. Both also understand that this large change must be phased in, so they phase out Fannie and Freddie over five to seven years.
3. Both understand that we must have much more private capital bearing the credit risk of the mortgage market, that is, more private “skin in the game” or “SITG”. Both go in the same direction here, although the House bill is better in that it goes further. I estimate that the House bill would take the U.S. to a housing finance sector which would be about 80% private and 20% government—about the optimal mix, in my opinion. But both bills unambiguously recognize that to have built a huge government-guaranteed system with so little SITG, as the U.S. did in the 1990s and then inflated into the bubble, was really stupid.
4. Both would end the “affordable housing goals” of the GSE system, and recognize that these “goals,” actually lending quotas, were a mistake, just as the fundamental GSE structure was a mistake.
5. Both bills propose, sensibly in my view, to use the Federal Home Loan Banks [FHLBs] as mortgage aggregators and securitizers for the small and mid-sized banks and credit unions, which in the U.S. banking system, number in the thousands. These institutions in general originate the highest credit quality mortgage loans. (I must confess that as the inventor of the FHLBs’ mortgage programs, which are built on emphasizing originator SITG, I have a sentimental preference here.)
6. Neither of the bills has the least interest in having the hedge funds, which have become speculative investors in the old preferred stock of Fannie and Freddie, rewarded by political action. As put in the form of a sharp-pointed rhetorical question by Senator Corker: Who would pay Fannie or Freddie a penny of fee to guarantee mortgages based on Fannie or Freddie’s own financial merit, if the government were not completely propping them up? Correct answer: Nobody–not one penny.
A critical shortcoming of both bills is that neither reflects an essential lesson of the housing bubble and collapse: the need for counter-cyclical standards which constrain the overleveraging of real estate in bubbly markets. The most efficient way to do this is by having counter-cyclical loan-to-value limits [CCLTVs], so that as house prices soar to unsustainable levels, the allowable leverage is reduced. CCLTVs will moderate both bubbles and busts, and both bills would benefit from having this core concept added.
The most prominent difference between the two bills is that the Senate bill includes a full faith and credit government guarantee of mortgage-backed securities by a “Federal Mortgage Insurance Corporation”. We can agree, as most observers do, that an explicit government guarantee with an explicit price is better than the implicit, but real, guarantee Fannie and Freddie got for free. So this proposal is better than and, in particular, more honest than, the former foolish claims that Fannie and Freddie were not really guaranteed by the government. But the taxpayer guarantee is still, in my opinion, unwise and unnecessary.
Proponents of the bill argue for putting in this taxpayer risk as “balanced"–by which they mean it makes happy many parties who enjoy subsidies from such guarantees. Proponents also say the government guarantee is only for “catastrophes”–and point out the bill’s requirement that credit losses up to the first 10% of principal must be covered by private capital. This 10% SITG requirement is distinctly better than Fannie and Freddie. Nonetheless, the existence of such a government guarantee will cause excess debt and excess leverage to flow to real estate. It will create a lot of creditors who do not care if they are financing increased asset price inflation and increased systemic risk. It will thereby make the very catastrophes it is supposed to protect against instead more likely to occur.
Catastrophes in financial markets are not outside, exogenous events that just happen, beyond any human control, like hurricanes or earthquakes. Financial catastrophes are caused by human financial behavior–they are endogenous to the financial system. That is why so-called “100-year floods” in financial markets happen about once every ten years. The much discussed “tail risk” of financial markets is not an objective, stable parameter: it is created by optimistic financial behavior. As any credit-driven bubble inflates, the “tail risk” of a collapse finally becomes 100%. In short, government guarantees encourage catastrophe-causing financial behavior and increase the tail risk.
A second key difference between the bills is that the House bill addresses the government’s Federal Housing Administration [FHA], now struggling with possible insolvency, as a central part of the problem. It proposes the excellent idea of removing the FHA from the control of the political Department of Housing and Urban Development, and making it into a separate government financial corporation, with an arm’s-length regulator.
The House bill also provides for establishing the legal framework for an American covered bond market. Covered bonds would give the banking sector an additional alternative for long-term financing of mortgages, on a 100% SITG basis.
Overall, at this point we can say that in the intense politics of American housing finance, it is a good start to have bills in both the House and Senate which agree on the main point of ending the disastrous political experiment of Fannie and Freddie.