Fannie, Freddie, Loans & Layer Cakes

I have a modest proposal. If we are concerned that mortgage giants Fannie Mae and Freddie Mac are getting the profits and the taxpayers are taking the losses why not just do without; Fannie and Freddie that is. The problem seems to be that banks and other loan originators want to make more loans than they want to keep. We all applaud this impulse. It results in more property taxes for municipalities, a stronger real estate market for established home owners [at least in normal times], work for home builders, real estate agents, appraisers, home inspectors, title companies, real estate lawyers etc. But can this all happen without Fannie and Freddie. I suggest that it can. First though, we have to get over the idea that all regulation is bad. After all no one would ever play a game of base ball if the rules of the game were not settled in advance. We are not talking here about getting into everyone’s intimate business; we are just trying to decide the broad rules of the game so we can tell whether we are playing base ball or cricket.

So if the rules we are playing under are giving us the sort of game we would really rather not play, why not try these on for size. First, if loan originators make better loans if they keep them why not simply require that they do, keep them that is. Not all the loans, just a portfolio of them that would be big enough to sink their company if they went bad in too large of numbers. You could tweak this rule so that the through-put of loans depended on the size of the loans that are retained, such that the originator was really in the business of holding the same kind of assets they were selling. You could require them to stratify their portfolio by quality. If they wanted to sell say one percent of their loans they could create one hundred portfolios of equal average quality and let the buyer choose one at random. If the quality of loans were like the layers of a cake, then every slice would have an equal number of loans conforming to every stratum of quality. The loans remaining would be identical in composition to the ones sold. A little thought here would be required. Quality would have to include all factors that could be used to differentiate loans in favor of the originator to the detriment of the buyer. You might even require that the originator guarantee the performance to be equal to the loans retained that were originated at the same time. A statistician may be required to tell you if there was a real violation here or not but even so I like the notion of keeping the risk in every slice uniform.

If this were done, it would not be too long before different originators would themselves be distinguished for the quality of the loan “cake” [if you will] that they serve. It would be in the best interest of all the competing companies to maintain and build the best quality portfolio they can for the yield they are trying to achieve. This of course would rules some levels of yield out of the question as too risky to build a business around. But why shouldn’t it, after all -- even poker players try to stay in the game.

I can almost hear a hedge fund manager objecting somewhere, but we would not be able to pursue out sized returns the same way we do now, if you did that. Good! I’m neither against hedge funds, nor their managers but there is something more at stake here. One of the primary reasons that people were willing to buy mortgages was that things have to be pretty bad for people to be turned out of their homes in large numbers. Well, here we are. We are discovering that there is plenty wrong with turning large numbers of people out of their homes beyond the damage done to the people most directly affected. Oops! We’re finding it is hurting the rich folks too. Who would have thought that you can only throw so many burning embers in your neighbor’s lawn before you run the risk of burning your own house down too?

However that may be, I would suggest that the greatest virtue of my “layer cake loan origination rules” is the fact that there are perhaps thousands of loan officers and mortgage brokers that have the skills to manage risk at this level. There could be ten thousand times ten thousand little companies that could engage in this kind of market. What a great diversifier of risk. No behemoths, just a lot of people with skill, doing a job that is very important for the economy. This should work great so long as no Ivy League MBA decides that a risk manager in Calcutta can evaluate a loan in Peoria better than a loan officer in Illinois can. Destroying a lot of little hands-on-companies in deference to their larger and “betters,” on the misguided belief that they will be better managed, doesn't seem to be working too well right now.

Before the banking consolidations of the 1990’s the risk in the banking system was widely diversified. Now it is highly concentrated. Why is that better? I for one think, and have thought for a long time, that it is substantially worse. In years past you couldn’t go to a social gathering without running into the president of the local bank. Now you won’t meet that person unless you go to New York or San Francisco. For every thousand bank presidents in the early 1990’s there are now 999 retired bank presidents in Florida and one guy in New York. I’m sure that Realtors of Florida are grateful, but is this really best for our communities? You can judge for yourself, but I believe I've drawn my own conclusion.

Layer cakes? It is at least, food for thought.