The drive to securitize mortgages combined with deregulation were key triggers of the credit crisis, says Wharton finance professor Susan Wachter. She is one of seven Wharton professors interviewed by Knowledge@Wharton for this special report on the credit crisis.
An edited transcript of the conversation follows.
Knowledge@Wharton: The blame for this crisis has been put on all sorts of factors from the rise of securitization to low interest rate policies of the Federal Reserve, homeowners’ greed, short-term mentality on Wall Street, federal regulators with their heads in the sand, and all sorts of things. What do you think were the major factors of the last 10 or 15 years have contributed to the sub-prime crisis?
Wachter: Well, guilty as charged, all of the above were part. They all contributed. It would not have happened but for securitization. It would not have happened but for deregulation, and deregulation securitization came together. I don’t know about the greed. There certainly are investors out there who are speculating.
But a large part of it is the prices rising — the affordability crisis — and pushing on the part of both borrowers and lenders for more affordable products which also … let more people into home ownership [and] let more people refinance, but the standards eroded over time.
Knowledge@Wharton: Now why was that? This is what we call partly the underwriting question — that the lenders were simply willing to lend to people that they had less likelihood of getting paid from than they were before. What led them to do that?
Wachter: Well, I want to be clear too. There are two ways that standards erode over time. One, the underwriting standards literally started to come into play in 2006, where people could basically say what their income was on many of these loans. And that’s underwriting eroding over time.
But also the standards [themselves] — the LTVs (consolidated loan to values) — went up with piggyback loans. So the lending standards that existed became more liberally offered. So as a consequence, supply function shifted. There was more supply. Now, you ask me why.
Knowledge@Wharton: Yes, why?
Wachter: Well, it is an interesting question. That is going to be a question for economic historians to ask and answer. But in models that my colleague Andrey Pavlov and I work [with], it is the natural competitive. Moving for market share, you compete not just on price, but you compete on market. You compete for market share by competing with your fellow lenders by undercutting them. And you can undercut them on rate [and] you can undercut them on standards to increase market share, especially so if the ones who originated making the loans don’t have any exposure to the risk on the other side.
Knowledge@Wharton: Now there also is a change in the kinds of institutions that were making mortgage loans. And back in the days when we think of it all as being Freddie Mac and Fannie Mae, there were certain standards that some of these newer firms didn’t have to abide by. Is that correct?
Wachter: Absolutely correct. And the firms who were gaining in market share were not as exposed to risk, so that is part of the story of how this happened. Fannie and Freddie’s market share, FHA’s market share [in] particular — the Federal Housing Administration government insured — dramatically declined as sub-prime increased. We saw that starting in 2002 through the end of 2006.
Knowledge@Wharton: So Fannie and Freddie were simply not able to make the kinds of loans that some of these other firms started to make?
Wachter: Fannie and Freddie, and I want to also include FHA because there is really a one-for-one. Fannie and Freddie and FHA were not able to make these loans because they by law [they] needed to invest in “investment grade,” which excluded sub-prime.
Knowledge@Wharton: Now, why were these other types of firms able to get into the market at this point rather than 15 or 30 years ago or some other time?
Wachter: This is a private label securitization which did not exist 15 or 20 years ago. What existed was securitization, which was Fannie and Freddie and jumbo securitization. Jumbo securitizations are loans that are investment grade prime, but over the conforming loan limit of $417,000 until recently.
So those were the markets. That was it. And that’s all that there could be because there was really no way of evaluating risk, charging VARs for different risks. So with risk-based pricing came the incentive to lend to riskier borrowers and to charge them perhaps more for the loan.
Knowledge@Wharton: And was this a technical advance or research on what risks are and that sort of thing? What were the elements that went into that?
Wachter: This is technical. Automated underwriting was first developed by Freddie Mac in the mid-1990s.
Knowledge@Wharton: That means computerized, basically, right?
Wachter: Correct, for the risk of the borrower using credit card FICO score type of information that used to be used in consumer credit that migrated over to [the] mortgage market.
Knowledge@Wharton: And the purpose of the automated underwriting is to give the lender an idea of the prospects that a particular borrower is likely to default on the loan based on past history, patterns of other people in that category?
Wachter: Based on past history and their characteristics.
Knowledge@Wharton: And I take it that one of the problems in the sub-prime situation is that now you are dealing with a group of borrowers and a group of products for which the track record is not as extensive, is that right?
Wachter: That’s simplified, that’s really too simplified.
Knowledge@Wharton: Well let’s hear the complicated version.
Wachter: OK. The automated underwriting is quite successful at predicting based on the risk of the borrower. There’s another component, which is the value of the home. So any incentive to default on a mortgage or eventually foreclosing their losses, that series of events happens because of the borrower risk, their willingness to pay, etc., which is predicted by credit score. But in addition, … the most key factor for foreclosure prediction is loan to value.
So if values deteriorate and if loans exceed values, then it is very difficult to sell the home, therefore foreclosure is almost inevitable. That is where we are today. And the models — the automated underwriting — were really not developed to predict what values would be under extreme circumstances.
Knowledge@Wharton: Let’s look at that how extreme they really were. When you look back at the period when many of these loans were being written — earlier to middle part of this decade — interest rates were extraordinarily low, unusually low at that time. And what happened afterward that seemed to have been the trigger event for many of these defaults and then foreclosures appears to be that interest rates simply rose to normal levels. The Fed Funds rate went from 1.0% to, I think, 5.25%, something like that. Was that really so unpredictable that rates would go back to the normal range from an abnormally low range?
Wachter: Yes. First, we can’t predict rates, but certainly the rates as they were in 2006 were not historically high. In fact, they are historically in a moderate range. So it is not as simple, again, [as] say[ing] it was the rate rise that occurred in 2006, nor is it [as] simple [as] say[ing] that [it] was the rate decline in 2001 to 2003 which set us up. Those, I think, were contributory factors, but they were insufficient to explain the sharp debacle that we are in now.
Knowledge@Wharton: Because rates certainly haven’t gone back to levels we saw 20 or 30 years ago when there were double digit rates for many mortgages, is that correct?
Wachter: Exactly. And at that point, obviously, we didn’t have anything like we have today. We did not have this kind of crisis.
Knowledge@Wharton: Now the second element of this was the falling of home prices, which as you said, left lots of people under water. They can’t sell or they may feel inclined just to walk away, as people say. Many people tend to think of their homes as the perfect investment that always raises in value, but there have been periods when they have retrenched. And it doesn’t seem, again from the professional’s point of view, all that surprising that home prices would level off and perhaps drop after the extraordinary gains from the ’90s into the early part of this decade.
Wachter: This is the first time in U.S. history since the Great Depression that we have had a national decline in home prices. So it is exceptional. And part of the story of the extraordinary collapse of 2007 — and current 2008 continuing — is the 2006 explosion of credit. So that, on one hand, we did have rates go up in 2006. On the other hand, standards eroded dramatically in 2006. And it’s the 2006 book of business that is under most stress. That’s where the foreclosures are coming from as of now.
Knowledge@Wharton: So why are we seeing this sort of unprecedented nationwide decline in housing prices?
Wachter: So it is not local markets. It is not unemployment. It is not simply expectations reversing. It should be something…
Knowledge@Wharton: It is not the local factory shutting down or something like that.
Wachter: And that indeed has been what previous housing price recessions have been attributed to. There is a critical factor that all of this misses, and it is not interest rates either. … It is that the standards [are] eroding, especially in 2006 when there [was] a dramatic decline in standards, [which] was unsustainable and that retrenched.
We had a sharp reversal, starting in 2006, of these unsustainable standards which were artificially inflating housing prices. So over this period standards eroded, artificially inflating house prices. Housing price increases and even levels could only be sustained with this unsustainable, overly liberalized credit being manufactured, which directly caused the price rise of 2006.
Knowledge@Wharton: And basically, that boils down to making money so easily available that people have more to spend, and they bid up prices. Is that the mechanism?
Wachter: Absolutely. It’s a credit induced bubble. People are able and willing to pay more when you have zero down payment, negatively amortized. They can get into the home and the bet is not on them. The bet is on someone else [as to] what’s going to happen in the future.
So it’s affordability, and it also is simply that these are loans without money going into them in some cases. That’s the investor side of the story.
Knowledge@Wharton: Now, there was during this period an increase in the use of adjustable rate mortgages. This has been one of the problems for borrowers who now are facing resets. The interests rates have gone up, and that’s causing their monthly payments to rise to points where they can’t pay it or don’t want to pay it.
First, is it correct that the use of adjustable rate mortgages increased? And what caused that?
Wachter: Not just adjustable rate mortgages. More to the point: teaser rate adjustable rate mortgages and option ARMs. All of these latter are negatively amortizing instruments. That is, you get into the loan and then you borrow more money over time.
That drives the loan-to-value ratio up, which again is going to be the key to the next part of the story which is: Now what do we do? Our loan is worth more than the home. Maybe we walk.
Knowledge@Wharton: And there’s lots of research that people who don’t have “skin in the game,” as they say — that is, equity in the house — are more likely to walk away from it than people who do.
Wachter: Absolutely. That is a key constant in all of our research. Foreclosure loan is driven by high loan to value, upside-down LTVs greater than 1.
Knowledge@Wharton: And one of the factors here is that lenders stop requiring the kind of down payments that many people were accustomed to years ago: 10% or 20% of the sales price. Why did they stop demanding those?
Wachter: Well, in part it was the ability to expand the market, this natural competition. If one guy’s doing it, the other guy’s doing it. That’s how you get market share.
And it’s not simply the LTV of one lender. These are consolidated loan to value ratios. So they were allowing piggybacks to come on, which were very popular to allow people to afford their home. Good business, good fees.
Knowledge@Wharton: Now, there’s lots of debate about what to do now that this has happened. It seems to me there are two broad areas: What to do about the people, the borrowers and lenders who are in trouble right now, and then what to do to prevent a recurrence of this kind of thing in the future.
Let’s break it down. What do you think should be done for the borrowers who are finding themselves underwater today? This has been a big issue in Washington recently.
Wachter: Yes. On the one hand there’s moral hazard involved. There’s the story [that] rescuing people who make bad decisions only supports bad decisions going forward. And there were bad decisions not only by borrowers, but by the underwriting agencies … and by investors. So let’s let everyone take their medicine, [become] educate[d] and go forward. Learn by doing.
There’s an argument for that, unless it actually leads to a serious recession which is a destabilizing one where in fact we have a free fall in prices which feeds back to the overall economy. We don’t know that that’s going to happen, but we don’t know that it’s not going to happen.
Knowledge@Wharton: So, it’s one thing to say that this is a homeowner who took a chance and it didn’t work out. That’s their tough luck. It’s another thing to say to that person’s next door neighbors that it’s too bad that your home prices are falling because there’s a glut of houses on the market in your neighborhood due to this. That’s really what we’re concerned about — the collateral damage and the rest of us suffering in some way. Is that one of the issues?
Wachter: And the overall economy. And what’s the medicine? What’s the response to that? The Fed is out of quivers at that point. So preventing that kind of free fall in housing prices has to be out there.
We have to have some tool to do that even if we don’t implement it immediately. Because, in fact, the need for it depends on the severity of the slowdown or recession that we may be in.
Knowledge@Wharton: And what do you think is the tool for dealing with that?
Wachter: Well, the concept has been supported by the Fed Chair Bernanke. There’s a bipartisan legislation on [Capitol] Hill that has the FHA coming in and putting a floor on the market.
There’s something to be said for that kind of response where the lenders take a haircut and the investors take a haircut. But they do it, and then in return for it the federal government then takes the risk of further price declines, which hopefully then will be slowed down.
Knowledge@Wharton: So you make both parties suffer enough that they wouldn’t want to do it again, and at the same time limit their losses to the point where their disaster doesn’t spill over and affect everyone else.
Wachter: Exactly. A floor for security for all of us.
Knowledge@Wharton: How do we distinguish between the borrower who is just an ordinary homeowner with a primary residence who either made a mistake or was duped into a mortgage that wasn’t a good idea, and somebody who is a speculator or somebody who was buying more house than they really could afford?
Wachter: Well, that’s a simple one. There are owner-occupants versus investors. You have to go down that line.
Knowledge@Wharton: Again talking about the institutions, the ones that lent money, how big a haircut do they need to take to discourage this kind of thing in the future?
Wachter: Well that of course is going to be questioned. But there are already firms out there that are pricing. There is some sign, and it looks like right now there are some trades at about a 20% to 30% haircut.
Knowledge@Wharton: Now, this is dealing with the current crisis. Looking down the road, I think most people would like to not see this kind of volatility in these markets in the future. Are there things that can be done to tweak the way the mortgage markets and the securities markets work to prevent the kind of excesses that led to this?
Wachter: “Tweaks” is a small word. There are going to have to be major changes to the way securitization works.
This exposes the Achilles’ heel of securitization, which is that it leads to a pro-cyclical impact of housing on the overall economy. Housing brings down the overall economy, and then the overall economy interfaces with housing. Then we’re in a free fall.
We need to have a response. I think it’s early in the game, certainly early in the game to put on major changes in regulatory responses because right now we have to be very concerned about liquidity. That’s the number one issue. In the sub-prime market, we have to be very concerned about liquidity in the overall mortgage market. It’s major.
Looking down the line, we are going to have to question the underlying basis for where we are here today, which is risk-based price securitization. That’s what’s new this time.
Knowledge@Wharton: And can you explain that?
Wachter: What’s new this time is that unlike the securitization of the past, the securitization is tranching of risk in very complicated CDO’s, CLO’s, SIV’s — instruments which do not trade.
So we do not have market discipline. Although the price of the loan may be varied by risk, … the price of the mortgage instrument and the securitization of the mortgage instrument, these securities did not trade. Therefore, there wasn’t a market discipline to price the risk and give the signal that these were extraordinarily risky instruments.
They were marked to model, not to market. There were lots of fees up front across the board. But the ultimate risk was unknown, because in fact they weren’t priced to the risk.
In some sense, it’s a disaster when we have potentially the foreclosure rate that we have of 2-3%. It could bring down the economy with it if there’s a recession that’s serious. But on the other hand, if this is priced and if the rate of return on the instruments that Wall Street pays is high, at least we’re sending the right signal. That did not happen.
In our research, what we’ve seen is that the standards eroded, but the rate did not go up. So this is a failure of pricing. It’s not surprising because these instruments did not trade.
Knowledge@Wharton: Well, if they don’t trade and you don’t have examples of recent sales to look at to set values, how will you set values? They tried to use models and that clearly doesn’t work. These are theoretical bases for setting value.
Knowledge@Wharton: What’s the alternative?
Wachter: If we are going to have securitization, it’s going to have to be securitization with trading, or with some indicator that these are extraordinarily risky securities because they don’t trade — some kind of red flag. … There is discussion [and] the SEC is moving along these lines. That is one answer.
The other answer is — and these are not either/or — … [that] we have been in an extreme deregulated environment where basically anything goes. And in that environment, even the competitors can race to the bottom. So there maybe [is] some need for the return of prudential regulation, especially if it is the banks themselves who are engaging in this race to the bottom competition, because then all of us are exposed to demand deposit insurance.
Knowledge@Wharton: Now in order to encourage trading, would you have to have some form of standardization so that people would have a fairly good understanding of what this instrument is? I gather that in this recent event, many were sort of custom-made and each one was, in some respects, unique to all the others.
Wachter: Which of course means you can’t trade the liquidity, etc.
Knowledge@Wharton: So would you standardize them?
Wachter: Without standardization, there is no trading. So that’s an open question. There are ways, and we are just beginning to think about how that would actually come about. It could be totally market-based incentives for standardization, [and the] SEC might have to have a role, but as you say, without standardization, there is no liquidity, there is no trading.
Knowledge@Wharton: Another proposal that has been floating around is to make sure that the participants in these markets all keep skin in the game, as they say. That is that instead of packaging up or bundling up one of these securities and then sending it on its way into the market and forgetting about it, the creators would have to maintain some sort of position or some sort of guarantee so that if things went wrong, they would pay a price. And that would give them an incentive to produce things of a higher quality. Does something along those lines make sense?
Wachter: It does. The basic research that my colleague Andrey Pavlov and I have done, which in fact helped explain the Asian banking crisis, points to the incentive to produce … short-run fees. And if there is nothing on the other side about what the consequence is of these fee-driven returns, then you are going to have a race to the bottom. It is under certain circumstances inevitable.
So this is not the first time that we have had a real estate and banking crisis that occurs together. In fact throughout history — we just look at … at Japan over the last 20 years. So yes, [having] skin in the game so that decisions are not just short-term [and] fee-driven is absolutely critical in every step of the way.
Knowledge@Wharton: Looking at the other part of the equation, which is the home buyer who will continue to want to get mortgages. You get the impression that many people are still going to want to use adjustable rate mortgages. For some types of borrowers, they may be a better choice than the classic 30-year fixed rate mortgage. Do you agree that they should still be part of the market?
Wachter: Absolutely. There is nothing wrong with adjustable rate mortgages. We have had adjustable rate mortgages for tens of years, and in the rest of the world they are the most common, they are the standard mortgage. It is not adjustable rate mortgages that are problems, the problems are negatively amortizing teaser rate option ARMs, which are predictably subjecting borrowers to payment shock at the same time that they are predictably artificially boosting the market, so that at the other side the market prices will fall. A recipe for exactly where we are.
Knowledge@Wharton: Now, when people take out adjustable rate mortgages, one of the appeals is that with the teaser rate, you qualify for the loan based on the low rate with a smaller payment that will fit a more modest income. But there has been some talk of having to look more deeply into the borrower’s future or estimate what it is going to be and see what will be the borrower’s ability to pay a higher payment if interest rates cause a reset that is much higher. Is there some way to do that?
Wachter: Well certainly, very recent votes in the Fed proposed rules that indeed you underwrite to the rate that that teaser rate adjusts to, which of course would have avoided a lot of the damage that we have. The other issue of adjustable rate mortgages [is that] you can’t have an adjustable rate mortgage which underwrites to any interest rate. But adjustable rate mortgages, as I said before, are not the cause of the problem that would occur in the end. They are a well wedded instrument. We can have them.
We can have adjustable rate mortgages. We can have fixed rate mortgages. What we can’t have is an explosion of credit that induces price rises artificially and then induces the pullback.
[There are] two things which you didn’t ask me about, but that I want to quickly address. One [is that], early on, you said, “What caused this?” One of the causes was [that] default rates did not rise immediately, and of course they don’t rise as long as the values are driven up. But as long as credit standards erode, and so as more credit is pushed out there, that is unsustainable [and] prices will increase. So the signal to pull back is not in the defaults. If you’re looking there, you are not going to find it. So you need to have market discipline, which is looking to the long-term and being able to identify this artificial credit induced bubble and price it. That’s one way out.
The other way out is a prudential way out, which is if the banking sector is part of this … artificial boom. By the way, it was tangential this time around in the U.S., but it isn’t in the rest of the world. We have seen these kinds of booms and busts brought down [in] Japan for 20 years, which came from the banking sector.
So in this case, there has to be prudential supervision of the banks themselves because banks too can engage in [a] fee-driven race to the bottom. Even if they have skin in the game, it happens and there is no reason why it wouldn’t happen.
Knowledge@Wharton: I want to make sure we understand what prudential means in this context. Can you explain that a little further?
Wachter: Thinking further… Further down for the long run.
Knowledge@Wharton: Further down the road.
Wachter: And that, of course, is what the regulators must do if, on the one hand, they are also giving out demand deposit insurance, which basically takes out a big part of the providers of the funds to the banks from making these decisions. Someone has to have the long-run concern. Demand depositors are neither in place nor do they have the incentive for their insurance to do so.
Knowledge@Wharton: Finally, what is your general assessment of the way the Federal Reserve has handled this so far?
Wachter: Bernanke has really pulled this off. It is really incredible. Nonetheless, even he is quite concerned with going forward. There may need to be more that happens and it’s out of his hands at this point.
Knowledge@Wharton: And looking at the things they’ve done, it’s a whole range of things from opening the discount window to lending out treasuries and taking mortgage securities as collateral to helping out with the Bear Stearns situation. Which of these do you think have been the most useful?
Wachter: Well, we have been so exposed — not just in the U.S. but worldwide to a potentially worldwide crash — that there are two of these that are absolutely critical. One was the decline in short-term rates, the increased liquidity — which is not just the Fed but worldwide — which keeps ARMs lowered short-term rates, which allow these teaser rates to adjust to a lower rate than they otherwise would adjust to.
Secondly, [and] equally important, is the historic Bear Stearns intervention, which is going to be very controversial historically and brings on its own questions going forward now [that] we have a whole other part of the financial system that is going to be rescued to some degree. Therefore, there is a very increased moral hazard there as well. [It's] a very controversial decision.
Knowledge@Wharton: And have you got a conclusion of your own about that?
Wachter: Well, at the time, I think without that we would have had a bank run in the non-financial sector, no doubt about it.
Knowledge@Wharton: And it does seem that although people have called it a bailout or a rescue, there certainly was a lot of suffering on the part of Bear Stearns’ employees and shareholders and executives.
Wachter: The shareholders were indeed disciplined and I am sure that was part of the planning of the event.
Knowledge@Wharton: Let me just finish by asking, do you think we are closer to the end of this whole process or still in the beginning?
Wachter: We are not in the beginning. We are not at the end. It is really this critical middle piece of how does the overall economy perform? And how does that subvert the potential recovery in the housing market or, in fact, cause the housing market to further unravel? There is no answer. We don’t know. That’s uncertainty that’s simply going to play out in the next three to six months.
Knowledge@Wharton: So this really is very much not just a routine crisis or cycle, but a kind of uncharted territory.
Wachter: For the next six months, we are in uncharted territory. We simply don’t know how far the economy is going to fall, if it’s going to fall, and how it’s going to interact with falling home prices.
Knowledge@Wharton: We’ll be watching. Thank you very much.
Wachter: My pleasure.