Viv Davies interviews Viral Acharya and Matthew Richardson for Vox

May 2011

Transcription of a VoxEU audio interview [http://www.voxeu.org/index.php?q=node/6510] 

Viv Davies:  Hello, and welcome to Vox Talks, a series of audio interviews with leading economists from around the world. I'm Viv Davies from the Center for Economic Policy Research. It's the 10th of May, 2011, and I'm speaking to professors Viral Acharya and Matthew Richardson, about their recent book, titled, Guaranteed to Fail: Fannie Mae, Freddie Mac and the Debacle of Mortgage Finance. The financial collapse of Fannie Mae and Freddie Mac precipitated one of the biggest government interventions in private financial markets in history. The bailout has already cost US taxpayers around $150 billion. The authors of the book maintain that these government-sponsored enterprises were guaranteed to fail.

I began the interview by asking Viral why he thought the book needed to be written.

Viral Acharya:  So, as you know, the book is titled Guaranteed to Fail. This refers to Fannie Mae and Freddie Mac in the United States in particular, but more broadly to a host of government-sponsored enterprises in the United States and perhaps elsewhere in the world, too. The idea is that these institutions are guaranteed on their downside. There is either an explicit or an implicit government guarantee that when they fail to deliver on their liabilities, the government balance sheets will backstop them. But very often their incentives in good times are, however, not sufficiently ring fenced. And so, they are allowed to speculate and take gambles in good times, essentially at the expense of the taxpayers in the end.

Now, one would think that, if there was one place in the financial sector where moral hazard would be rife, it would be these government-sponsored enterprises. And yet, when you look at the aftermath of the crisis, what gets discussed, it is true that government-sponsored enterprises get a mention. But as we ask in the book, where is the outrage? We should be absolutely outraged that institutions like these have been designed in the form they are, where essentially profits are privatized but losses are socialized.

So, this book is written primarily to fill in that void. We are asking the question, let's get outraged about this. Let's do something about this.

Viv:  So, Matt, Fannie Mae and Freddie Mac are prime examples of US government-sponsored enterprises or GSEs. Why were they established as such in the first place? And to what extent is government intervention necessary in the mortgage markets?

Matthew Richardson:  Yes, they came about in the wake of the Great Depression. So, in 1938, Fannie Mae was created, and the reason was because banks who were making mortgage loans, there was no secondary market for these loans. There was no way for banks to get them off their balance sheets, there was no way trade could take place, and so the whole, sort of, market kind of collapsed. And they were created there as kind of a backstop, as a provider of liquidity, so that, in times of stress, the market could continue going. That's in the 1930s. We can understand that, that was a stressful period.

Move forward, and then in the late '60s, so with Fannie Mae part of the public sector, what happened, the Johnson administration sort of decided it wanted to make them a private entity. Now normally, when we think about governments taking public institutions and making them private, we think it's because of some kind of need for efficiency, need for better allocation of capital.

Here, it was really very different. The Johnson administration wanted them off the balance sheet because they added to the US debt. So, Fannie Mae was made private, literally, so they would no longer show up as a balance sheet item. And as a result, what we ended up was with these private entities, but having all this public support.

The thing that Viral just talked about, we had this bad mix of the private sector and public sector in one place, and Freddie Mac joined shortly after.

Viv:  Viral, perhaps you could explain the difference between a deposit-based system of mortgage finance, as opposed to a securitization system, as in the US. And maybe also explain the advantages of the securitization system.

Viral:  A deposit-based system of mortgage finance would be like more the traditional banking has been. Banks raise deposits, they may hold long-term deposit insurance. They are holding some capital against the risk of losses on these deposits. But on their asset side, they originate and essentially hold these mortgages on their balance sheets. What are the pros and cons of the system? The pro is that the originator owns the entire mortgage risk, so as long as they're well capitalized that they would have the incentives to ensure that these mortgages are of right quality.

What is the con? The con is that the mortgages by themselves are quite illiquid. Each individual mortgage would be very hard to describe to someone else and sell it on a piecemeal basis in the markets.

Now, to the extent that bank balance sheets may need to raise capital to meet losses from time to time, if they're not able to sell these mortgages in the markets to anyone, so there's no secondary market for these mortgages, sometimes solvent banks could basically just face an illiquidity problem and they could wind down completely.

A securitization system partly came about to essentially get around the liquidity problem. There are lots of long term capital providers in the financial intermediation system, such as pension funds and insurance funds, maybe, to some extent, even hedge funds which may be somewhat more short in their horizons but are willing to take riskier assets on their balance sheet.

So, the idea of securitization was that individual banks or even pools of mortgages from different banks could be put together. And then, once you take a first loss piece, what is often called as an equity crunch, the remaining piece essentially looks like a very safe bond.

It's a safe bond in the sense that it's basically going to get hit only if you have a very secular house price decline, so that the results on the pool of mortgages aren't worse enough, even though the pool is reasonably diversified. These safe assets, like a AAA mortgage‑backed security, could be sold to these pension funds, hedge funds and so on.

What are the advantage of the securitization system, as I said, the advantages that it grants liquidity to the bank balance sheets to creation of a secondary market for mortgages. What is the disadvantage? The disadvantage that gets stressed is the fact that this is a departure from the originate-and-hold model. Now, it's an originate‑to‑distribute model, where banks that are originating mortgages may not necessarily be owning the entire risk.

Note, however, that in the way the traditional securitization is supposed to be designed, at the time, banks want to sell the mortgages in the secondary market, they really have to hold the first loss. Because this is the way you give credibility to the secondary market buyers that are not actually just giving you the lemons, while I'm actually designing this pool of mortgages or a mortgage‑backed security.

Now, the bigger disadvantage of securitization system, and the one that we have witnessed in the kind of world we had so far, was that there were agencies such as Fannie Mae and Freddie Mac that essentially were taking on the risk, credit risk, from the securitization, basically retaining the mortgages for themselves, but they had that implicit government guarantee. And the capital requirements to do this were very low.

There were also players in the private sector who were basically guaranteeing the mortgage risk on some of these pools. These were the, what I called as the private label securitization, or private guarantors. Their prudential regulation was also inadequate, so that many of them, effectively, themselves also became too big to fail or too systemic to fail.

So, the disadvantage of securitization that we saw, according to us in the book, was not that of the failure of originate‑to‑distribute, but it was really the failure that those who were owning the tail risk of the mortgage market, which is the risk that you would get a secular house price decline, were extremely poorly regulated. So they were thinly capitalized, and therefore their incentives to take on the tail risk, make the premiums in good times, but then, sort of, default ex post, leaving the losses with the taxpayer.

We think both the GSEs and the private sector came together to concoct a really toxic mix in the securitization market.

Viv:  So Matt, why did the market break down and require such massive government intervention and support? Aren't GSEs systemically risky institutions? Or are there also wider political issues at play? Is the intractability of the issue the result of it being a political football in the US?

Matthew:  Most of the risk in the mortgage market--take the Fannie Mae, Freddie Mac. They controlled about 50 percent of the risk in the mortgage market by the time the crisis hit. That's about five trillion dollars. And it was under circumstances under which they had the government guarantee. If you entered into the mortgage market with a transaction with Fannie Mae, Freddie Mac, your overall capital in the whole financial sector could be about half as much. It could be as twice as leveraged. And through the 1992 Housing Act, thereafter, the GSEs were encouraged to take on riskier and riskier mortgages. So, by the type of the collapse, we had, sort of, five trillion dollars with a pretty risky asset base. So, the moment you're going to get a housing collapse, whether it's in 2006, or it could be in 2002, or 2010, didn't matter when it would occur, you would have had these large losses on the GSEs.

And the reason the government had to go in, you've got an institution that has 3.5 trillion dollars of insurance with the rest in the financial markets that are highly interconnected. You had 1.5 trillion dollars of portfolio positions by that entity, so you'd have a huge problem of fire sales if it failed. You had a problem that they were the only player, because they crowded out the private market for prime mortgages, so they were the majority of the market.

So, the housing system would collapse without them there. And you also had a problem that their own debt was widely held by the banking sector and some large foreign governments, so those losses would have been imposed elsewhere which could even put the US sovereign debt at risk. So, there was no, probably the most systemic institutions out there, so there's no choice but to go in and support them.

The question, of course, is why did we let them grow to be this way? And I think now the issue is, we know it's a big, big problem. Both the left and the right parts of the US political spectrum sort of generally agree that these guys were a bad idea. Now the issue is, how do you go from where we are now to this broader issue of getting rid of these GSE institutions?

Viv:  So, you outline some proposed solutions in the book. Maybe you could briefly describe what these are and, given that some time has elapsed since the book was published, to what extent your recommendations have been taken on board or not by the US government in trying to resolve the issue.

Viral:  Our proposed solution in the book really consists of two parts. The first part is where we want to be in the long run. And we think that, really, in the long run, we want to have a private mortgage finance market. It may partly consist of mortgages just being held as they were originally on bank balance sheets, and it may partly consist of a private securitization market. The mortgages securitized et cetera would be decided by the private sector, but as we've discussed, with securitization, there is the risk that private guarantors would want to take on the tail risk of a secular house price decline, collect premiums in good times, but then default ex post.

So, I think, what they securitized would have to be tightly ring fenced, in the sense of having some strict criterias, such as loan-to-values being no greater than 80 percent, and they would have to have prudential capital requirements which are adequate with whatever risks are being taken on.

Now of course, given this long run view, the idea would be that institutions of the size of Fannie Mae and Freddie Mac, which is a huge imprint of the government in mortgage finance in the United States, would not be present. But the big question is, how do you get there? Right now, practically 90, more than 90 percent of the mortgages being originated in the United States are, in one way or the other, backed by government-sponsored enterprises, be it Fannie Mae, Freddie Mac, Ginnie Mae, FHA, whatever it is.

So, to go from here to a fully private solution is quite a daunting task. In the book, we call it the problem of putting the genie back in the bottle. Which is, every time there's a crisis, it's tempting to use the government-sponsored enterprises to prop up the housing markets, get consumption to be boosted, it's a convenient political tool. We've done it, now, quite a few times, but it is very clear that it's emerging out to be a Ponzi scheme in the end. So, we need to put an end to this.

And so, our solution is not to attempt to do this in a very abrupt fashion, which is to just come out one day and let Fannie and Freddie fail. As Matt highlighted, they are too interconnected, too systemic, too huge to fail all at once. So, the idea would be to phase them out over a period of seven to ten years, and the phasing out would occur in three steps.

First, every year you would reduce the size of the mortgages where they are in the securitization market. Say, you reduced the size by $75,000 every year, currently for the 700,000 within one year, bring it down to 625, and so on. So, this way, you give credibility to the private sector, bring your capital in, because Fannie and Freddie are going to be out of this market in seven to 10 years time.

Second, the biggest problem with the way the mortgage subsidy is being provided in the United States is that, because it's being done through the cheap guarantees being offered by Fannie Mae and Freddie Mac, it's basically making mortgages cheaper, its boosting house prices a lot.

This means that the low income households are really not getting a net transfer, because they have a cheap mortgage, but a very expensive house compared to what they could get otherwise. The real benefit is just going at the very top to, sort of, the very wealthy households by them consuming more housing, them owning extra houses, and so on.

So, to get around this problem, and simultaneously to ensure that whatever limited presence Fannie, Freddie have over the phase out period, they don't crowd out the private players. The second step is to ensure that their guarantees are at the same price as that of the private sector.

So, it won't be the GSEs who decide going forward what gets privately securitized. It would actually be the private sector. But for every dollar on a pool that's guaranteed by the private sector, only 25 cents of capital will initially come from the private sector, 75 cents will come from the government, say, through this new GSE, so to speak.

So, this is basically a passive GSE. It lets the private sector decide what's being securitized. It lets the private sector set the price. It charges exactly the same price for the guarantee. All that it's providing is to get around the possible market failure that private capital may not enter at a fast rate in the next five to ten years.

But note that over 10 years, they are not going to be there in the first place. They have a current portfolio and a current set of guarantees. After the next seven phase out years, they will have another portfolio and guarantees, another set of guarantees, but that would also be phased out.

The third point is that currently, the GSEs have an investment function, which is, they actually just own some portfolios and mortgage backed securities themselves. We think this is basically the largest hedge fund being run on the planet with a government guarantee. We think it should just be shut down right away, no more owning of portfolios by GSEs going forward.

So, we think these are the three critical steps. Maybe I should let Matt respond to the issue of what's being taken on board and so on.

Matthew:  OK, let me add to that one more thing to what Viral said. So, I think part of the motivation is that there's a big investor class out there that invests in mortgage‑backed securities with some type of guarantee associated with it. So, we still in that transition period and that investor class has to switch from investing in guaranteed securities to more of a credit based mortgage‑backed security. That would take a while. So, that's the motivation behind having this sort of joint public-private partnership on the guarantee front. So, that's sort of, you know, the standard approach to try and develop that in the investor class.

In respect to what the government taking on board, it's always unclear whether they listen to you or how much, what they do when you throw out these suggestions. I mean, I will say that in the terms of the long term solution, both the administration's plan, and the plans from Congress are fairly consistent with each other and also with what we have in the book. And that's for, basically, it to be a private market. The question is the transition.

The administration doesn't really address that issue too much. Congress has just put forth eight bills, a bit of a hodgepodge of bills. But part of, in those bills, are the idea that the imprint of GSEs will decline through time, though they haven't really addressed how to get from A to B. Or as we call it, how to put the genie--the genie's out of the bottle, it's been out of the bottle, really, for 25 years now. So, it's not straightforward to put it back. And so, we're a little bit concerned that, you know, even though both sides kind of agree to the general solution, they're not that focused on it.

And, you know, part of the reason is, right now, the GSEs are providing lots of freebies. They're providing the equivalent of Percoset to the US financial system. They're a painkiller, but it's not solving the problem, it's just making it worse. So, it's an issue, and it's not clear to me it's going to get resolved in the near future.

Viral:  I just wanted to add, sort of, three small points to that. One, some people argue when they look at our plan that, can a purely private securitization market with guarantees from some private guarantors really function? We think that's really something for us to see organically in the markets. If it doesn't function, we should not try to push securitization at our rates. There are many banking systems in the world where there's no securitization market for mortgages and the home ownership rates in those countries are no less than in the United States, and in many cases, they are, in fact, higher.

Second, a question that often gets raised is, what about the affordable housing goals of the United States government? Should they continue to provide loans through FHA and for government workers and veterans through Ginnie Mae?

These programs have been much smaller in size. Their incentive structure has not been as bad, partly because their debt is explicitly recognized on the government balance sheet. And what they can do is, right now, very tightly ring fenced.

But one thing we've observed, studying the mortgage finance in the United States history, is that there is a tendency of a mission creep. Which is, you have an entity set up to do X, over time it becomes convenient to ask it to do Y.

And so, we are not convinced that subsidizing housing through cheaper mortgages is the best way of making a transfer, tax transfer, to the low income households. Maybe a direct rental subsidy, or it could just be a direct tax transfer. Let them decide what they want to consume. Do they want to spend on a car? Do they want to spend on their children's education? Or do they want to buy a house? It might be much better to do a direct transfer of this form.

The third point I would make is that, we have focused in the book, pretty much, for most part, on Fannie and Freddie. But it's very clear that the landesbanken in Germany, the cajas in Spain, they have pretty much served exactly the same purpose. They maybe, sort of, captured by local municipal governments rather than the federal government at large.

It's also clear that in many countries in Asia that have state-owned banking sectors and they often get used to do very strong directed lending in the midst of a crisis, and so on.

So we think that the lessons of how to organize financial sector and if at all you want to have a government imprint, in what form you should have, extends beyond just the GSEs. The idea is, address the market failure, but ensure that the imprint is not a long lasting one. Get back to a private market as soon as you can, once the crisis role of the GSEs has been delivered.

Viv:  So finally, in your opinion, who's essentially at fault for allowing this situation to arise? Is it the various US governments who've expanded their guarantees to these institutions over many years? Is it the result of bad governance and the institutions themselves? Or is it about the middlemen who overstated the creditworthy credentials of home buyers, and subsequently, walked away richer and scotfree? Or is it simply all about market failure?

Matthew:  That's a tough question, because I will say everyone's a little bit at fault. But, one way we like to describe, when we're asked this type of question, we sort of like to go to a favorite quote of the crisis, as respect to the mortgage market. And that's from David Frum, who is a former speechwriter of former President George Bush. And it goes something like this. He said, "The shapers of the American mortgage finance system hoped to achieve the security of government ownership, the integrity of local banking, and the ingenuity of Wall Street. Instead, they got the ingenuity of government, the security of local banking, and the integrity of Wall Street." I think that kind of sums up where the problem is.

Viv:  Viral Acharya, Matt Richardson, thanks very much for talking to us today.

Viral:  Thank you, Viv.

Matthew:  Thank you.