Listen to the podcast:
Does competition lead to better outcomes for government sponsored enterprises (GSEs) like Fannie Mae and Freddie Mac? Not necessarily, says Andrew Davidson, president of Andrew Davidson and Company in New York City. His firm specializes in the application of analytical tools to investment management, and he has worked extensively on developing, valuing and hedging mortgage-backed securities. In an interview with Knowledge@Wharton, Davidson discussed the motivation and the main thesis behind his recent paper titled, “Is There a Competitive Equilibrium for Government Sponsored Enterprises?” Davidson notes, “We think the solution is to not encourage more competition, but something more along the line of a regulated utility.”
An edited transcript of the conversation follows.
Knowledge@Wharton: This month, Fannie Mae and Freddie Mac began their ninth year in conservatorship. That seems like a timely occasion for a paper titled, “Is There a Competitive Equilibrium for Government Sponsored Enterprises,” or GSEs. What motivated you to write the paper and what is your main thesis?
Andrew Davidson: There have been many discussions of different forms for GSE reform during this nine-year time period. And recently, there has been more and more discussion of the idea that there should be several firms in the place where we now just have Fannie Mae and Freddie Mac to increase competition. We wanted to look at whether or not that made sense as a solution. Our main thesis, unfortunately, is that it probably doesn’t make sense. That the characteristics of the market that Fannie and Freddie operate in probably are not such that where competition will lead to better outcomes.
Knowledge@Wharton: What conditions do you think are required for a successful competitive market and do these conditions apply to the mortgage market in the U.S.?
Davidson: Economists over time have thought deeply about competitive markets and they’ve come up with many different requirements that they think make competitive markets more successful. We looked at about eight of those, including property rights, excludability, rejectability, diminishability, information symmetry, accessibility, no externalities and immediacy. For us, the ones that weren’t met are the most relevant, they are those last few. Accessibility, no externality and immediacy.
Knowledge@Wharton: Your paper points out that some segments of the mortgage market meet some requirements for successful competitive markets. But there have been some successes as well as failures across the segments. What do you see as the biggest source, or the biggest cause, of failures in the mortgage market?
“The characteristics of the market that Fannie and Freddie operate in probably isn’t one where competition will lead to better outcomes.”
Davidson: When we look at the segments of the mortgage market where it’s the borrowers interacting with mortgage banks, that’s a fairly competitive market. When we look at the investing side of the market, that’s also fairly competitive. But when we look at the part where the GSEs are involved on the origination and portfolioing of mortgages, we’ve seen failures over and over again. We think it’s related to those last three factors. But probably the most important one is that a lot of the risks of mortgages are really hidden.
If you originate a mortgage of poor quality, you don’t often find out about that until sometime later. It takes changes in unemployment or home prices or movements in interest rates before the true risks of the mortgages are exposed. This means it’s fairly easy for a player who wants to dominate the market by increasing market share to start making compromises on quality or in some other aspects of their origination process. Then they can gain more and more business, and we won’t really see the negative outcome of that until some later time.
Knowledge@Wharton: What are some of the remedies that economists have found to address these market failures, and what’s your assessment of those remedies?
Davidson: Each failure has different types of remedies. For the immediacy, it’s really one of the hardest ones to solve. For accessibility, there are ways of setting up utilities that can create accessibility for more people. For externalities, economists usually require either some form of regulation or taxation. We are talking about carbon tax or things like that to take care of pollution, externalities. And so any of these solutions could be a good solution and you really have to make sure you take the solution that’s appropriate to the problem.
For example, one of the problems in the mortgage market has always been information asymmetry. What does the borrower know versus what does the investor know. Over time, a tremendous amount of data transfer has taken place, and we have excellent systems in place to reduce that asymmetry. During the crisis, people stopped doing the basic underwritings. They lost the ability to take care of that problem. But that’s a problem that can be solved through more information.
Knowledge@Wharton: What are the implications of your analysis for GSEs and what kinds of interventions are required to achieve results that are best for society?
Davidson: As I said, one of the things that’s been proposed is the idea that we should have many GSEs, create a competitive market. The idea is competitive markets are more efficient at allocating resources. Most people recognize that there is difficulty in access, because the mortgage market is so large and Fannie Mae and Freddie Mac were so central that they are really providing the plumbing that keeps our system going. So one of the proposals was to separate out the plumbing from the other functions of Fannie Mae and Freddie Mac through the creation of this common securitization platform.
It’s a good idea, but it really only addresses one aspect. It only addresses the accessibility aspect of the obstacles to a competitive market, and doesn’t really do much for the issues about immediacy and externality. A real solution has to address all of those.
Unfortunately we think the solution is to not encourage more competition, but something more along the line of a regulated utility. Then the question is, what is the governance for that utility, and there’s really three possibilities. One is that the government could own and run that. Two is that it can be a publicly owned utility like a power utility, or three, the one I favor most, is that we use a cooperative structure to own and manage that utility.
Knowledge@Wharton: What would be some of the implications of the solution that you’re proposing?
Davidson: It turns out that the solution I’m proposing is actually not much different than the situation we’re in currently. We currently have two large GSEs that are heavily regulated. And so the main addition is how do you bring in the private capital to support them so the government has less risk exposure. And then we could use a system much like the federal home loan bank system, which also most of these lenders participate in. So it would be taking something from another area of the housing market and imposing it on this portion.
“We think the solution is to not encourage more competition, but something more along the line of a regulated utility.”
Knowledge@Wharton: Based on what you’re saying, how would you say the mortgage ecosystem has fared during the past almost decade that the GSEs have been under conservatorship? And do you think that a high degree of regulation should continue? Or is more competition a possible solution?
Davidson: It’s sort of surprising, but the last 10 years have probably been the most stable time period in the mortgage market history, at least from the time I’ve been involved, which is from the early 1990s or late 1980s, and the 10 or 20 years of history I know before that. Some of that may just be the hangover from the crisis, and people are more conservative. But I think a lot of it has to do the way that Fannie Mae and Freddie Mac have been regulated to stick to their knitting and provide a consistent set of services across the mortgage market.
It’s important to keep a fair amount of that regulation in place, to provide that stability. But it would be very good to start bringing in additional capital. Right now, you and I as taxpayers are taking the risk of further losses if we go through another down cycle. It would be nice to get private capital into taking that position.
Knowledge@Wharton: Do you think the GSEs that get too big and get into trouble should be allowed to fail, or like some banks, are they too big to fail?
Davidson: The problem is that it’s very hard to create a system where entities such as the GSEs can fail. They essentially are acting as sort of exchanges in our system, more sort of too-big-to-fail providers of infrastructure. We can’t allow the infrastructure to fail. We’ve seen what happens in the past. Whenever there’s a crisis in the housing market, the government’s going to step in anyway, because we need to maintain the housing finance system in order to keep the economy going.
We should also recognize that this just isn’t a market where if you have a competitive failure, if you have a firm that fails, you can say, “oh, just let them fail and other players will come in.” The amount of disruption is just too great. It’s also important to point out that if one firm fails, it’s very likely all of them will be failing at the same time. It might be nice to say, “Well, we have 10 firms and if one failed, we still have 90% to the system.” But as we’ve seen in the mortgage market, usually when you have one failure, you have a cascade of failures, because they’re all exposed to the same risks.
Knowledge@Wharton: Some experts have proposed that a common securitization platform should be created, and their argument is that it will help create a level playing field for multiple competitors by providing equal access to the mortgage market, or the mortgage-backed securities market. Do you think this is a good idea?
Davidson: Well, it might be a good idea to have a common securitization platform and to provide some additional stability to the infrastructure. But it’s naive to think that a competitive solution will be stable. What we’ve seen in the mortgage market is that generally one of two things happen. Either a dominant firm emerges, because it’s not just the infrastructure, but there are other pricing or service aspects that they can use to grow their market share. Or two, there’s a race to the bottom where people take advantage of the lack of immediacy that we talked about in order to cut corners, quality, improve pricing slightly and take on more and more risk. So it’s very easy to grow your market share at a time when things are good, and hide the underlying problems. That’s unfortunately my fear of what would happen in a more competitive market.
Knowledge@Wharton: The other idea that has been proposed, I think this is by the Mortgage Bankers Association, is that the GSEs should be transformed into utilities. And that more GSEs should be created to encourage competition. What’s your view of that?
“I think the problem is that it’s very hard to create a system where entities such as the GSEs can fail.”
Davidson: Well, certainly I favor the idea of them being turned into utilities, which means that if you’re going to have some dominant firms that basically have monopoly power, they need to be regulated as to what prices they can charge and what activities they can engage in. So the utility part, I’m certainly supportive of. Once again, on the competitive side, the question is what are you gaining through that competition? If you’ve already created a utility that limits excess returns, what are you gaining from having more competition, and then what are you risking?
And once again, my view is that you’re risking a race to the bottom or a dominant firm emerging anyway. And so why not just face up to the fact that this just isn’t a market that meets the requirements that economists say are needed for a competitive market and use one of the other types of solutions.
Knowledge@Wharton: Coming back to the solution that you propose in your paper, what do you see as the opportunities and risks inherent in your solution?
Davidson: Whenever you make changes, there are always risks that things won’t work out correctly. But the opportunity is really to recognize that what we have now is a system that’s working fairly well. Fannie and Freddie are providing services. They’ve already started to shed some of the risk through risk sharing transactions. And they can probably be transformed into the system I would like with very little change, other than bringing in a new governance structure. The biggest risk really is whether that governance structure will work.
Say I favor a cooperative system, and that means large lenders and small lenders would be in the governance position, and it may be difficult to keep both the interests of the large institutions and the small institutions balanced. Once again, the regulator would have to focus on that. And then the other question is whether or not you’d raise enough capital. Well, my view is the capital should be activity based. If people want to use the system, they’ll have to come up with the capital. If they don’t, then that will just shrink the system, which may or may not be a bad thing.
Other systems, which involve bringing in external capital, may need to work much harder or create much higher returns in order to attract that capital, and then bring in different types of instability, because we now have outside players who are going to want to expand the role of Fannie and Freddie to improve their returns and growth opportunities.
Knowledge@Wharton: What would it take to change the governance structure along the lines that you are proposing?
Davidson: Somewhere along the line we’re really going to need Congress to act. Congress needs to do two things. One, it needs to change the charter to allow this type of governance structure. And two, it needs to move to a situation where there’s an explicit guarantee on the mortgage-backed securities so that the market can continue to function the way it is right now. Right now, we’re relying on treasury providing capital to Fannie and Freddie in a very indirect fashion, and an explicit guarantee would really help to move the whole process forward. One other thing is that we need to have a capital rule to say how much capital these entities really need to protect the taxpayers.
Knowledge@Wharton: Given what’s going on in Washington, D.C., these days, do you see a possibility of this happening?
Davidson: During the spring and even early summer, there was a tremendous amount of interest from Congressional staff and Congressmen, senators, in trying to do something on GSE reform. Unfortunately, with all of the turmoil over the last few weeks, between the hurricanes and issues of the debt ceiling, talking about tax reform, it just seems like there might not be enough attention to go around to really do anything soon. But even if it doesn’t, I think some of these discussions will be going on in the background.