Banks need more skin in the housing finance game
We all know it was a really bad idea in the last cycle to concentrate so much of the credit risk of the huge American mortgage loan market on the banks of the Potomac River — in Fannie Mae and Freddie Mac.
But the concentration is still there, a decade later.
The Fannie- and Freddie-centric U.S. housing finance system removes credit risk from the original lenders, taking away their credit skin in the game. It puts the risk instead on the government and the taxpayers.
Many realized in the wake of the crisis that this was a big mistake (although a mistake made by a lot of smart people) in the basic design of the inherently risk-creating activity of lending money. Many realized after the fact that the American housing finance system needed more credit skin in the game.
Skin-in-the-game requirements were legislated for private mortgage securitizations by the Dodd-Frank Act, but do not apply to lenders putting risk into Fannie and Freddie. Regulatory pressure subsequently caused Fannie and Freddie to transfer some of their acquired credit risk to investors — but this is yet another step farther away from those who originated the risk in the first place.
That isn’t where the skin in the game is best placed. The best place, which provides the maximum alignment of incentives, and the maximum use of direct knowledge of the borrowing customer, is for the creator of the mortgage loan to retain significant credit risk. No one else is as well placed.
The single most important reform of American housing finance would be to encourage more retention of credit skin in the game by those making the original credit decision.
In this country, we unfortunately cannot achieve the excellent structure of the Danish mortgage bond system, where 100% of the credit risk is retained by the lenders, and 100% of the interest rate risk is passed on to the bond market. The Danish mortgage bank that makes the loan stays on the hook for the default risk and receives corresponding fee income. The loans are pooled into mortgage bonds, which convey all the interest rate and prepayment risk to bond investors. This system has been working well for over 200 years.
There are clearly many American mortgage banks that do not have the capital to keep credit skin in the game in the Danish fashion. But there are thousands of American banks, savings banks and credit unions that do have the capital and can use to it back up their credit judgments. The mix of the housing finance system could definitely be shifted in this direction.
If you are a bank, your fundamental skill and your reason for being in business is credit judgment and the managing of credit risk. Residential mortgage loans are essential to your customers and are the biggest loan market in the country (and the world). Why do you want to divest the credit risk of the loans you have made to your own customers, and pay a big fee to do so, instead of managing the credit yourself for a profit? There is no good answer to this question, unless you think your own mortgage loans are of poor credit quality. For any bankers who may be reading this: Do you think that?
But, it will be objected: The regulators force me to sell my fixed-rate mortgage loans because of the interest rate risk, which results from funding 30-year fixed-rate loans with short-term deposits. True, but as in Denmark, the interest rate risk can be divested while credit risk is maintained. For example, the original 1970 congressional charter of Freddie provided lenders the option of selling Freddie high loan-to-value ratio loans by maintaining a 10% participation in the loan or by effectively guaranteeing them, not only by getting somebody else to insure them.
Treasury Secretary Steven Mnuchin recently told Congress that private capital must be put in front of any government guarantee of mortgages. That’s absolutely right — but whose capital? The best solution would be to include the capital of the lenders themselves.
In sharp contrast to American mortgage-backed securities in this respect are their international competitors, covered bonds. These are bonds banks can issue, collateralized by a “cover pool” of mortgage loans that remain assets of the bank. Thus covered bonds allow a long-term bond market financing, but all the credit risk stays on the bank’s balance sheet, with its capital fully at risk.
American regulators and bankers need to shake off their assumption, conditioned by years of Fannie and Freddie’s government-promoted dominance, that the “natural” state of things is for mortgage lenders to divest the credit risk of their own customers. The true natural state for banks is the opposite: to be in the business of credit risk. What could be more obvious than that?
The housing finance system should promote, not discourage, mortgage lenders staying in the credit business. Regulators, legislators, accountants and financial actors should undertake to reform regulatory, accounting and legal obstacles to the right alignment of incentives and risks. The Federal Housing Finance Agency should be pushing Fannie and Freddie to structure their deals to encourage originator retention of credit risk.
The result will be to correct, at least in part, a fundamental misalignment that the Fannie and Freddie model foisted on American housing finance.