President Trump has proposed to privatize the two housing finance agencies Fannie Mae and Freddie Mac, now in public ownership after their collapse in 2008. While the fisc might benefit in the short term from such a deal, the risk to taxpayers of a second collapse and bailout is high, for risk correlation reasons I will explain. The Fannie/Freddie structure is a relic of the dozy socialist New Deal; it is unnecessary and imposes excessive hidden risks on taxpayers, who would always be forced to bail them out in a crisis, as in 2008. Much better to kill them off and allow home mortgage finance to be provided by the market, as it should be.

Fannie Mae is a true child of the Great Depression, founded in 1938 (Freddie Mac is a mere clone, founded in 1970, by which time policymakers should have known better). Home mortgages in the 1920s had typically been short-term, coming due in a mere 5 years, with a substantial balloon payment at the end which generally required refinancing. The one genuflection to credit quality was that the loan to value ratio even on a new loan was normally only around 50%. This made housing difficult to afford for ordinary Americans, though when combined with the almost total lack of building regulation and zoning, it kept house prices down.

A typical middle-class detached home, albeit smaller and much less well fitted than today, sold for perhaps $6,000 — about $114,000 in today’s money according to the Bureau of Labor Statistics price index, but that index is fudged, and the fudging becomes very noticeable over long periods like a century – probably the true equivalent today is about $200,000. (In terms of gold, even at last year’s $2,500 per ounce, it would be 289 ounces, or $723,000!) The difficulty was then finding the 50% downpayment – even at the BLS’s today equivalent, $3,000 or $57,000 in today’s money is a lot for the Bank of Mum and Dad from a working-class family.

Then the Great Depression came, and the average house price declined by 43% in nominal terms (27% adjusted for the period’s deflation). People having 5-year mortgages with a large balloon found it very difficult to refinance, when a third of the nation’s banks had gone bust and the remainder were being far more stringent in their credit conditions. Even those who had kept their jobs had huge difficulty, and of course those out of work had no chance of refinancing and lost their homes. Thus, the early and mid-1930s saw a huge surge in foreclosures, and house prices remained depressed, far below 1920s levels, for the rest of the decade, with no partial recovery as occurred in the stock market.

Part of the problem was that the Glass-Steagall Banking Act of 1933, splitting commercial and investment banks, had severely decapitalized investment banking. There were thus almost no equity new issues during the decade, and certainly no ability to invent innovative capital market solutions to the home financing problem. The foolish Banking Act by itself bore considerable responsibility for the prolongation of the U.S. Great Depression, which did not occur elsewhere. Britain, which did not break up its merchant banks until later, had mutually owned mortgage lending institutions, the building societies formalized by the Building Societies Act 1874, that financed homes on a longer-term (10-20 year) fully amortizing basis, with no balloon refinancing needed.

The New Deal’s solution to the home finance problem was the Federal Housing Act of 1934, which set up the Federal Housing Administration to micro-manage terms and interest rates in the mortgage market – a typically socialist solution, that should not have survived the 1930s. The FHA established 30-year fixed rate mortgages, longer than in Britain and impossible to finance in the 1930s capital markets, and in 1938 founded Fannie Mae to guarantee home loans so that they could be bought and sold by unsophisticated investors. The FHA’s sponsorship of Fannie Mae (and later Freddie Mac) gave both agencies an implied Federal government guarantee that was called upon in the 2007-08 financial crisis. Since 1938, the U.S. mortgage market has been dependent on these agencies, which inevitably have followed political objectives, for example in their early decades classing black and mixed neighborhoods as “high-risk,” therefore ruling African-Americans out of the market for loans.

The economic effect of this peculiar, thoroughly politicized system has been pernicious. For one thing, making mortgages equivalent to each other (because of the guarantee) and having credit risk assessment done by two centralized behemoths has caused the banking market to agglomerate. That has produced huge banks with poor customer service and no local knowledge. In a traditional mortgage system without guarantees the local bank with local knowledge of the real estate market and its customer’s business has a substantial advantage over the behemoths in making credit decisions. In a guarantee system the local bank is merely a box-ticker and adds no value. The borrower is subjected to two levels of bureaucracy and cost, to no purpose.

Furthermore, as we saw in 2002-06, the current system excessively rewards aggressive mortgage brokers, who being pure salesmen push unsophisticated, generally impoverished borrowers to take on subprime mortgages they cannot afford, with the knowledge that the credit risk of those mortgages will be passed off either on Fannie/Freddie or through securitization on dozy German landesbanken. Those shady operatives quickly go out of business in a localized system, because they are left with many of the loathsome credit risks that they source.

The real problem is that if Fannie/Freddie are privatized, they will not stay privatized but will come back zapping taxpayers as they did in 2007 with credit losses far in excess of any short-term profit from privatizing them. This is because, as Kevin Dowd and I explained in our “Alchemists of Loss” (© Wiley – 2010) the risk of a large portfolio of home mortgages is not “Gaussian” but has much fatter “tails” that make Wall Street’s conventional risk management methodology simply wrong.

Home mortgages are highly correlated, both locally and across an entire housing market. When the Youngstown, Ohio steel industry collapsed in 1977-82, the entire local housing market was devastated and defaults spiraled, if only because so many wage-earners had lost their solid, well-paid union jobs. In the subprime mortgage crisis of 2007-08, securitization tranches rated AAA, supposed to be exceptionally safe, still suffered massive defaults because the downturn in the housing market especially affected the low-quality broker-sold mortgages that made up the pools. Gaussian methodology completely fails to account for these correlations, because it assumes the chance of multiple defaults is vanishingly low. What is needed is a fuzzy logic methodology, in which the probability of multiple defaults is the minimum of the mortgage default probabilities, not their product, so perhaps 10% rather than 0.000001%. Thus, the next housing downturn after privatization, with losses once again highly correlated, would almost certainly result in the two agencies’ collapse, with subsequent Federal rescue at astronomical cost to long-suffering taxpayers.

Privatizing Fannie and Freddie presumably requires legislation; so would closing them down. If we can get rid of these relics of a failed New Deal socialist ideology, the housing finance market can revert to its natural state, albeit with some modern financing techniques added. For example, local mass bankruptcies can be avoided by the use of credit swaps, so that a collapse in our local market of Poughkeepsie need not take out the Ulster Savings Bank, our local lender, because Ulster would have laid off much of the specific Poughkeepsie housing market risk in the same way that insurance companies use reinsurance.

The home mortgage market would fluctuate, like all markets, and those fluctuations would immediately be reflected in the pricing and availability of mortgages, with no backstop central bureaucracy providing guarantees at a fixed cost. That would reduce the probability of a nationwide home mortgage crash like 2007-08, because the market would signal tightness long before the crash became inevitable. (It would be helpful here if the Fed could avoid pumping out tsunamis of cheap money whenever it felt like it, but maybe that is too much to ask.)

President Trump is doubtless surrounded by overpaid lobbyists telling him to privatize Fannie Mae and Freddie Mac. By now, he should know that the smart move is to do the opposite of what the lobbyists tell him!

-0-

(The Bear’s Lair is a weekly column that is intended to appear each Monday, an appropriately gloomy day of the week. Its rationale is that the proportion of “sell” recommendations put out by Wall Street houses remains far below that of “buy” recommendations. Accordingly, investors have an excess of positive information and very little negative information. The column thus takes the ursine view of life and the market, in the hope that it may be usefully different from what investors see elsewhere.)