Editor’s note: This is the second in a two-part series. See Part 1 here.
The first article in this series examined the two obvious strategies for expanding private sector lending as a replacement for Fannie Mae and Freddie Mac. The first, expansion of portfolio lending by depository lenders, is not an option because home mortgages have become increasingly unattractive to them. This is mainly a consequence of the financial crisis, which showed that home prices didn’t always rise, prices could also decline, and when they did, losses from mortgage defaults would balloon. These losses included heavy financial penalties imposed by government on the largest lenders, who were considered responsible for what went wrong.
The second option, revival of the private secondary market that collapsed during the crisis, is not feasible either — it may take a generation for investors to forget how badly they were burned. But even if they forgot more quickly, we should not attempt to revive that market because it is a "fair-weather model" — it cannot withstand stress.
The major problem is that each individual mortgage security is a stand-alone entity that is nobody’s liability. If the reserves or insurance protections embedded in a security are inadequate, that security will default — never mind that 100 other securities have more protection than they need. What is needed is a more robust secondary market.
The good news is that a tested model exists in Denmark. In the Danish model, every mortgage security is a bond that is a liability of the firm issuing it. The Danish mortgage banks issue separate bonds for each type of mortgage, and if one of them experiences a high loss rate, all the resources of the bank are available to deal with it. There has never been a default on a Danish mortgage bond in more than 200 years. During the very worst months of the financial crisis, it was business as usual in the Danish mortgage bond market. And more recently, during the worst months of the eurozone crisis, it has been business as usual in the Danish market.
The core of the Danish system is an institution called a "mortgage bank," which originates mortgage loans and simultaneously funds each one in the bond market. The mortgage banks also service the loans they make on behalf of the bondholders. We have no such firms. Our mortgage banks originate loans that they may or may not service, and sell them. When they did securitize loans before the crisis, it was on a block (rather than loan-to-loan) basis, and the security was a stand-alone for which the mortgage bank assumed no liability.
To convert some of our existing mortgage banks into Danish-style mortgage banks (henceforth DMBs) may require enabling legislation at the federal level, but that is the least of it. A DMB will require more capital than any existing mortgage bank has, and to generate the margins required to earn an adequate return on capital, the bonds it sells must be priced as very low-risk investments. To meet these requirements, DMBs will need a helping hand.
Fannie and Freddie should provide the needed assistance, but under the supervision of the Federal Housing Finance Agency, which will monitor the development of DMBs and the associated phaseout of Fannie/Freddie. To provide the additional capital required by newly chartered DMBs, the agencies could purchase non-interest-bearing capital notes from them, which must be paid off over a specified period. The capital note repayment period would be long enough for the DMB to generate the retained earnings needed to retire the notes.
The initial series of bonds issued by DMBs would carry federal guarantees, but these bonds would be open-ended only for a specified period. Beyond the bond guarantee period, new mortgages would be funded by bonds that are guaranteed by the DMB, but not the government. The bond guarantee period would be long enough for the DMB to demonstrate that its bonds were safe and sound investments without a federal guaranty.
Existing mortgage banks are the most logical candidates to become DMBs, and the most logical constituency for moving the development through the political process. In our political system, nothing much gets done without a push by an entity with some political clout that has a strong interest in getting it done. Nonetheless, DMBs could also be formed de novo, or as affiliates of depositories or other types of financial firms.
The phaseout of Fannie/Freddie would be tied to the growth of DMB bonds without federal guarantees. As DMB lending increases, Fannie/Freddie lending can decrease.
This proposal is not a quick way to get rid of Fannie/Freddie, but there is no quick way that would not seriously disrupt the market. The alternative to a slow fix is no fix.
Note: My thanks to Alan Boyce for helpful comments. Boyce suggests, as an alternative to purchasing capital notes from DMBs, that FHFA could provide them with full faith and credit reinsurance. This is an alternative worth considering.