Thursday, July 8, 2010

Now for the Biggest Financial Regulatory Reform: Residential Mortgages

If we likened the efforts to prevent another financial crisis to the government's program to combat the flu, here's what it would look like. The financial reform legislation now working its way through Congress changes structure and process. Regulators will operate in a new structure, with an overarching systemic risk council and a new consumer protection bureau. Processes for trading in derivatives would change, as would proprietary trading by banks. Heightened capital requirements would be imposed on banks, and troubled financial institutions would be subject to seizure and wind down by the government. Comparable changes in government programs to combat the flu might consist of structural change at the CDC and FDA, together with heightened reporting requirements for hospitals and medical professionals concerning all actual or potential cases of the flu.

But what about the vaccine? The government's primary weapon against the flu isn't structure or process. It's fostering the manufacture of vaccine. Vaccines do the most to prevent illness. New flu vaccines are produced every year. Where's the vaccine for the financial crisis?

The biggest single reason for the financial crisis of 2007-08 was the way residential real estate transactions are financed. The large majority of purchases were (and still are) made with long term mortgages--those that contemplate a 30-year or comparably long repayment period. Although many of those mortgages had adjustable rates, the amortization schedules for those mortgages (i.e., the anticipated repayment of the principal of the loan) involved only gradual reductions of principal or none whatsoever (in the case of interest only loans). Whether or not the loans had fixed or adjustable rates, they were all structured with long repayment periods and very gradual amortization of principal, in order to make home purchases more affordable.

Long term obligations are subject to a high degree of interest rate risk. Stated otherwise, when interest rates change, the obligation imposes large burdens on one party or another. If the debt has a fixed rate, the creditor takes the loss. If the debt has an adjustable rate, the borrower takes the loss. It the loss has been transferred by means of a derivatives contract, the counterparty bears the loss. The important thing to understand is that losses are sustained and someone must bear those losses. The losses cannot be made to go away. This is a crucial reason why long term mortgages are so risky.

The long repayment schedules of these mortgages were meant to serve a governmental purpose: making it easier to buy a home. The private mortgage market didn't develop a 30-year mortgage on its own. Government policy, beginning in the 1930s and 1940s, created it. To accommodate financial institutions that didn't want to hold these puppies (and there were many), the government created first Fannie Mae, and then Freddie Mac. Fan and Fred bought the hot tamales, in effect passing the long term risks onto taxpayers. However, the federal government's numerous subsidies for housing and the nation's recovery from the Depression caused home prices to rise steadily on a national level. The risk to taxpayers seemed theoretical. So more and more 30-year mortgages were written, prices rose more, and demand for housing grew apace. The secondary market for mortgage-backed securities blossomed, supported by a seemingly golden asset that never fell in value. Institutional investors far and wide piled into the mortgage markets, believing they'd found the pot of gold at the end of the rainbow. But as the mortgage markets ballooned, so did the systemic risks presented by the ever increasing number of outstanding mortgage loans. The sheer quantity of loans (encouraged by the government to increase home ownership) puffed prices up into a bubble, and that meant systemic risk was skyrocketing.

Certainly, bad lending practices made things worse. The plethora of subprime, Alt A, no money down, no doc, no verified income, more stupid than stupid loans that came into fashion added a shipload of credit risk to the the interest rate risk inherent in long term obligations. But much of the reason these loans were so toxic is they too were structured along the same lines as the traditional 30-year mortgage: a long repayment period, with amortization of principal almost imperceptible in the early years of the loan. They presented the same risks as traditional 30-year fixed rate loans, along with a big helping of credit risk. And all this was possible because of the government policy of sponsoring and subsidizing long term real estate lending.

Once the real estate bubble burst, prices fell and numerous homeowners went underwater on their loans. Many defaulted due to unemployment. Increasing numbers are defaulting for strategic reasons. The real estate market remains in the septic tank for now. When it will recover is anyone's guess, because the sheer quantity of defaulting mortgages from the 2000s weigh heavily on the financial system and the economy.

It is unrealistic to believe that changing the structure of financial regulation and the processes of the financial system are sufficient to prevent another crisis. We need to reduce the levels of systemic risk. The 30-year mortgage and its first, second and other cousins create systemic risk of the first order. Because this risk was the biggest factor leading up to the credit crisis, federal regulators should treat it as their highest priority. Structure and process won't suffice. No package of new rules, more thorough examinations, heightened capital requirements and other measures can shield the financial system from the uncontrolled growth of risk from mortgage markets gone wild. The creation of risk must be moderated, and that means limiting access to long term mortgage finance.

Credit standards are already tightening. But perhaps not enough. There is still no meaningful secondary market for home mortgages without a federal guarantee, which means that the private market adjudges them to be overpriced compared to the risks they present. When long term mortgages can be sold to private investors without a federal guarantee, then we can be comfortable that their downpayment requirements, interest rates and other terms are reasonably priced in relation to the risks they present. That they cannot now be sold signals that taxpayers are still subsidizing real estate purchases to a significant degree.

There is little indication that the mortgage markets will be privatized. Talk now is of nationalizing Fannie Mae and Freddie Mac, which would formalize the federalization of housing finance. And why not? They're functionally nationalized already, and any privately owned-publicly backed hybrid would only perpetuate the outrage of the 2000s: private profit at public risk. But nationalizing them could contribute to the problem--the opacity of governmental accounting (which would apply to the nationalized Fannie and Freddie) would make the extent of taxpayer subsidies, already in the hundreds of billions, even less clear and therefore subject to abuse. As it is, Americans suffer grievously from taxpayer abuse syndrome in subsidizing residential real estate. One can only rationally assume the abuse would continue unabated and rarely seen if Fannie and Freddie are nationalized.

Nationalizing Fannie and Freddie may, in the end, be how the government controls the systemic risk presented by the gargantuan hordes of long term mortgages rampaging around the financial system. That would be a poor outcome, since it would continue costly policies of taxpayer subsidies, with most of the benefit probably going to the higher income brackets. And if the housing market makes another gigantic u-turn, as it did in recent years, the risks of taxpayer support of the real estate market would be realized. While most likely the federal government would simply increase the federal deficit to finance these costs in the near term, they remain real costs that eventually would be visited on taxpayers.

It's important to have a way to measure these potential costs. One way would be for federal authorities to require Fannie and Freddie to offer mortgages for sale in the private market without a federal guarantee, simply to see what price they would command. The discounts that private investors would demand could be used as a proxy to calculate the extent of taxpayer exposure. The numbers would probably be Brobdingnagian. But ignorance won't be bliss. Such a calculation would be a useful way to measure how much systemic risk is quietly building up in housing finance. Large figures would trigger alarms--and that's the idea.

No comments: