When the U.S. housing market cratered in 2007 and 2008, it exposed the weaknesses of Fannie Mae and Freddie Mac, the two government-sponsored entities that underpin the nation’s mortgage market. After Washington was forced to bail them out — lest their failure do even more economic damage – an array of experts, economists, politicians and pundits suggested major changes for the institutions. Most common was the idea that it was time to simply shut them down.

However, none of the plans offered found enough political support to advance, and today, Fannie Mae and Freddie Mac today are buying and securitizing home loans just as they did before. But they still need to be reformed. Andrew Davidson, the head of mortgage industry consultancy Andrew Davidson & Co., recently published a paper with a somewhat new solution for the GSEs. He sat down with Knowledge at Wharton to explain how we can save not just those institutions, but our nation’s homeownership culture, and the classic 30-year fixed-rate loan, without putting the taxpayer totally on the hook the next time catastrophe strikes the real estate sector.

You can listen to the interview using the player above. An edited transcript of the conversation appears below.

Knowledge at Wharton: To set the stage here, we’re going to talk about the state Fanny Mae and Freddie Mac, which are the two huge government-sponsored entities that help to securitize mortgages, and pass them on into the investment market. They provide liquidity to the residential mortgage market, and a huge percentage of U.S. home loans pass through those two entities hands at some point.

Davidson: That’s correct.

Knowledge at Wharton: They blew up in 2008, along with a lot of mortgage-related things, and the federal government bailed them out. As I understand it, most of that bailout has been repaid, or more than repaid. But nevertheless, it still required an investment of almost $200 billion. Naturally, there is a lot of interest in preventing that from happening again.

Now, while they were put into conservatorship by the government after the meltdown, they were quasi-government entities before: They had a government charter, but really they were run as private companies. In the wake of all this many had suggested that someday that would get reversed. There have been lots of plans put forth about how to do that but none of them seem to have worked out.  Now there are a couple of new papers out — from Mark Zandi, who’s with Moody’s Analytics, and who you were just involved in a seminar with. So you’ll be fresh on this topic. And I understand you have a new paper coming out.

“If you look at almost any of the proposals from 2008 through last year, they all started with either, ‘Wind down the GSEs,’ or ‘Shut down the GSEs,’ or ‘Eliminate the GSEs.’”

What might be different about these papers is that there’s starting to be a realization that maybe these entities can’t be run as private companies. Maybe the government really needs to be involved. And I’ll let you take it from there.

Davidson: Sure. So in 2008, when these entities were set down, Secretary Paulson said there should be a time-out. And I think to have a time-out seems like the game, instead of the time-out.

He was sort of hoping that people could reexamine how these entities were run and ask, “How do we make some adjustments?” We went through a process after that where different people had a wide range of proposals on how to change Fannie and Freddie. Some said to make them more private; some said to make them closer to the government. But none of these proposals really could capture a large percentage of Congress supporting them.

In the meantime, the regulator of Fannie and Freddie, the Federal Housing Finance Agency, had actually made a number of changes to these organizations. And I think some people are starting to realize that the changes that have been occurring may be GSE [government-sponsored enterprise] reform — that this is a pathway forward. That one answer is to look at what’s happened already, and say, “Can we do something to continue to change the entities, rather than throw them out and start all over?”

Knowledge at Wharton: Because that’s been one of the suggestions – to say that they’re irretrievable, and we just have to start from the ground up.

Davidson: Yes. I think if you look at almost any of the proposals from 2008 through last year, they all started with either, “Wind down the GSEs,” or “Shut down the GSEs,” or “Eliminate the GSEs.” I think this new round of proposals — the Zandi proposal as well as my proposal — say, “Let’s start with what we have. How do we take what we have now and turn it into what we want?”

Knowledge at Wharton: It’s worth pointing out that GSEs, which Fannie and Freddie are or were, are this hybrid thing. Even back before 2008, before the financial crisis, although they were considered private companies, most people thought, “Yes, but if they get in trouble, we all know the government’s going to bail them out.” To which other people would say, “No, no, no. They’re private. The shareholders will take the hit.” But when push came to shove and things were collapsing — not just Fannie and Freddie, but so much of the financial system — the government did come to the rescue. Those who thought that would happen ended up being correct. Is what you’re talking about really so different from the way things were run for decades?

Davidson: The thing to think about is that the shareholders actually did lose almost all their money — or all their money, depending on how litigation comes out. So the government did not rescue the shareholders. But what the government did do is step in and protect the bondholders. Those people who were investing in the mortgage-backed securities…

Knowledge at Wharton: On the secondary markets.

Davidson: Right. For the debt of these entities. I think almost everyone thought that the government would step in on that.

Knowledge at Wharton: OK.

Davidson: The rating agencies for a long time have said that due to government sponsorship, they view the debt instruments of Fannie and Freddie to be of the highest credit quality. Based on the capital that those entities had, there’s no way that they would have thought that otherwise. So I think you have to separate out the equity investors from the bond investors.

Knowledge at Wharton: OK. Fair enough.

Davidson: One of the important things in GSE reform is to just be explicit about that, rather than have a “wink” relationship. “Oh, no, we don’t guarantee any of those assets.” You know, it’s in big, bold letters on the front page of the Fannie Mae and Freddie Mac bonds: “These are not government-guaranteed bonds.” But privately, Washington is saying, “Oh, sure. We’re not going to let those fail.”

Knowledge at Wharton: And the rates were set in a way that reflected what would happen if they were guaranteed, correct?

Davidson: Exactly right.

Knowledge at Wharton: So the markets said, “They’re really guaranteed.”

Davidson: Yes. So, rather than live in that world, why not live in a world where you say, “Yes. The government will guarantee certain instruments created by these entities, because we recognize that in order to have a giant real estate market and housing market like we do in the United States, we’d like to have a 30-year fixed-rate mortgage. And a 30-year fixed-rate mortgage functions much better with a government guarantee than without one.” And so if you believe that, then you say, “We need the guarantee. And so let’s start from the idea that we will have a guarantee, and it will be explicit.” But we still have to figure out what the entities are that are going to create these loans.

Knowledge at Wharton: Which is where your paper probably starts. Could you tell us the title, where people can find it, and a bit about what’s in it, please?

Davidson: I don’t think it’s up yet, but it will be on the Urban Institute website, and also be on the website of my company, Andrew Davidson & Co. — www.ad-co.com. The paper’s called “Four Steps Forward.” The four steps are to streamline, share risk, wrap, and mutualize.

“If they had $200 billion more in capital than they had before, there wouldn’t have been a need for the bailout of the entities.”

By streamline, it means, let’s make these entities smaller than they are now.

By share risk, it says that we don’t actually have to put all of the capital for bearing credit risk into these two entities. They’ve actually started to sell off their credit risk through credit-risk-sharing transactions, which are basically ways where Fannie and Freddie buy insurance from the market. They’re also buying insurance from reinsurers, and maybe some from the mortgage insurance industry. That program is now $30 billion or so of money that actually stands in front of the taxpayers if there are credit losses on Fannie and Freddie mortgages. That program should be expanded.

The third step is “wrap,” and that’s what we were talking about before, that there needs to be an explicit government wrap, and it has to be structured in such a way that the government only is liable if there are catastrophic losses. You’d have to have very bad performance in both the housing market and unemployment before the government would need to pay something. But the government will know that’s when they’ve got to put the money in.

Knowledge at Wharton: Right. But that would be a taxpayer-funded bailout, right?

Davidson: Yes, but … you know, I don’t know if I’d consider it a taxpayer bailout. The government would be running an insurance program. It’s going to be collecting premiums for that. You know, when the FDIC makes good on deposit insurance, we don’t call that a bailout. When they make good on saving a company maybe that should have failed, maybe that’s a bailout. But this is an insurance program, and they’d be paying the money that they’ve been earning premiums on.

Knowledge at Wharton: Under that plan, the government would be creating a reserve that would cover a certain percentage up front. But if there’s an economic catastrophe, the reserve would only cover things up to a certain percentage, correct?

Davidson: The reserve is only going to cover so much in losses. Then the government can either recover or not recover that afterward. But the point is to structure it so it’s much further out on the probability curve than we stand now, with this current implicit guarantee. Most people are talking about Fannie and Freddie having either capital or reinsurance that covers 4% to 5% of losses, versus before, when they only needed to hold 45 basis points, or less than half a percent of coverage in case of losses.

If they had $200 billion more in capital than they had before, there wouldn’t have been a need for the bailout of the entities. And that was definitely a bailout, because, as I say, there was no mechanism in place to say that they were buying some insurance.

Knowledge at Wharton: If what you’re proposing had been in effect in 2008, I think what you’re saying is, things would not have happened the way they happened. There would have been enough of a cushion, enough of a shock absorber there, to prevent at least the failure of Fannie and Freddie.

Davidson: The sub-prime market had its own problems, and this wasn’t going to help that. But Fannie and Freddie would have had enough capital, or enough reinsurance, to cover the losses. So most people are talking about loss coverage probably about double the experience of 2008.

“Creating the government wrap is really the essential ingredient in order to continue the 30-year fixed-rate loan. There just really aren’t investors who are willing to invest in that kind of interest rate risk, and also take on credit risk.”

So that covers “wrap.” Then, the fourth step is mutualize.

OK, now we have these entities. We have a functioning system. But there’s still a conservatorship. How should we spin these back out into the market? And so in my proposal, I think they should be turned into mutual companies, owned by the originators. And the mutual should own Fannie and Freddie. They should put up some capital based on how much they use those entities. And that those entities will basically fill the same functions that Fannie and Freddie are doing now. But they wouldn’t become stockholders of a company. They’d become mutual-owned companies.

The Zandi proposal is almost exactly the same as mine up to this point. Except that instead of being a mutual, he and his co-authors, believe that it should become a government corporation.

Knowledge at Wharton: I see.

Davidson: It wouldn’t be a government agency like FHA. It would be more like a government corporation. The FDIC is an example of a government corporation. Then, other people have said these entities should become utilities. Or if you go back to the Johnson-Crapo housing reform bill, it said these entities should have become competitive guarantors. At this point, we’d really be talking about differences in governance of this residual entity. But I think any discussion along that line is a big step forward from saying, “Let’s shut them down and start all over again.”

Knowledge at Wharton: What are the big benefits to all of this, from your point of view? There’ll be more of a cushion if there was a meltdown, which seems like it’s a public good. But is this also going to preserve the 30-year mortgage? I think Americans would be very disappointed if somehow market conditions didn’t allow that to continue. Does this accomplish that?

Davidson: Yes. Creating the government wrap is really the essential ingredient in order to continue the 30-year fixed-rate loan. There just really aren’t investors who are willing to invest in that kind of interest rate risk, and also take on credit risk. What the guarantee does is just splits those two risks apart. Certainly, that’s one important advantage.

Another important advantage is that a lot of people are very concerned about maintaining access to credit, in two different ways. One is that every borrower who is a qualified borrower and can afford a loan over time should be given the opportunity to have a loan. By having a national entity, you can have it be responsible for making sure there are lenders operating in every market.

Knowledge at Wharton: Preventing discrimination of one sort or another?

Davidson: That’s right, and to keep them from saying, “Oh, we don’t serve this particular market, because it’s too far out of the way.” A national entity can either do it themselves, or encourage the banks and its members to say, “Someone needs to serve this market.”

Knowledge at Wharton: Is this addressing the idea of affordability? Is that part of the point?

Davidson: Yes. There’s access, and then there’s affordability as well. Affordability is handled in several different ways. Right now, Fannie and Freddie don’t price each loan individually. There’s some averaging of the risk. By averaging the risk, people with better credit actually subsidize, to some degree, people with worse credit, which helps with affordability. The other thing that most of the proposals include is a method by which some of the money the government earns from providing the wrap goes into a fund that also directly supports affordable lending.

Knowledge at Wharton: I see. Affordable in the sense of subsidized rates?

Davidson: Some sort of subsidy.

Knowledge at Wharton: Some kind of subsidy. OK.

Davidson: Another important part of these proposals is what happens to the profits that are made by Fannie and Freddie. Fannie and Freddie made a lot of profits when they were private entities. One of the advantages of them being mutual companies is that if they’re making a lot of profits, those profits go back to the owners, who are also mortgage originators. And they’re competitive, which can feed back into lowering prices, or lowering rates for borrowers.

If you move from a competitive market where you have stockholders outside of the mortgage market, to the stockholders inside the mortgage market, even if you do end up with monopoly profits, they recycle back into the system.

Knowledge at Wharton: Is there a problem with diversification though, if everyone invested in these entities is in the same family of the mortgage world?

Davidson: One of the things that would make this work is the existence of these credit-risk-sharing deals. I think about 75% of the risk that’s created should be moved outside of this system. By using different bond structures or reinsurance structures, the credit risk wouldn’t all sit inside Fannie and Freddie. It doesn’t actually have to sit with those originators. They’re going to get other people to take that risk. The diversification happens through the capital markets.

Otherwise you would have this situation where it’s like, $200 billion, $300 billion worth of capital sitting at one or two entities, and you would need to find a way of diversifying the shareholders.

Knowledge at Wharton: A little too incestuous that way.

Davidson: The industry couldn’t take on that much risk. On the other hand, we do believe in skin in the game. A big part of having the members also be part of this mutual is that they are responsible jointly and separately for the activities that they’re creating.

Knowledge at Wharton: What else is it important to know about your proposal?

Davidson: I don’t know whether or not it would happen, but I think that there’s a pretty clear pathway to get there. As I say, it doesn’t require creating all new entities. It just requires changing the ownership structure of existing entities. By doing that, it takes us out of having a transition plan that lasts 10 or 20 years.

Knowledge at Wharton: Or longer.

Davidson: Or longer. Who knows what would happen over that time horizon?