American Enterprise Institute
March 28, 2007
[Edited transcript from audio tapes]
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Nouriel Roubini, New York University |
|
|
|
R. Christopher Whalen, Institutional Risk Analytics |
|
|
|
Thomas Zimmerman, UBS Investment Bank |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Proceedings:
Alex J. Pollock: Welcome to our discussion of “Mortgage Credit and Subprime Lending: Implications of a Deflating Bubble.” The conference is co-sponsored by the American Enterprise Institute and the Professional Risk Managers’ International Association. On behalf of both, let me say how pleased we are to have you all here today. I am Alex Pollock, a resident fellow here at AEI. We have an excellent panel whom I will introduce in a moment.
Briefly to set some context, we had a big housing boom with unprecedented increases in real house prices, facilitated by a mortgage lending boom and notably, a subprime mortgage lending boom. As we all know, the subprime boom is over, and the bust is here. Former enthusiasm has been replaced by large financial losses, the bankruptcy or closing of numerous subprime lenders, lay-offs, accelerating defaults and foreclosures--what Moody’s has just referred to as “stunning erosion in mortgage credit quality,” a liquidity squeeze, and of course, escalating political recriminations.
What was not so long ago praised as a creative or innovative expansion of home ownership is now described as an irresponsible and culpable activity. How much more bust will we have? Is this problem confined to the subprime sector, as a whole panel of regulators told Congress yesterday, or will problems spread to other parts of the mortgage market? What are the implications for house prices, that is, the market value of the most important asset for households by far? What are the implications for the market of mortgage related securities and the housing industry? And are there wider implications for monetary policy and the overall economy?
Of course, and needless to say, booms and busts are hardly new. As they recur in financial history, they differ in detail but display the same general patterns. You would think that we would learn, but we do not. I can never think about this subject without recalling the marvelous Victorian prose of Walter Bagehot in Lombard Street, published in 1873. Some of you have heard this before, and you should enjoy it again as I quote, “The mercantile community will have been unusually fortunate if during the period of rising prices, it has not made great mistakes. Such a period naturally excites the sanguine and the ardent. They fancy the prosperity they see will last always, that it is only the beginning of a still greater prosperity. They altogether overestimate the demand for the article they deal in, or the work they do. They all in their degree--and the ablest and the cleverest the most--trade far above their means. Every great crisis reveals the excessive speculations of many houses no one before suspected.”
Bagehot continues, “The good times of too high price almost always engender much fraud. All people are most credulous when they are most happy, and when much money has just been made, when some people really are making it, and when most people think they are making it, there is a happy opportunity for ingenuous mendacity. Almost everything will be believed for a little while.” Alas, how true in 1873 and in 2007.
We have a distinguished panel to discuss the implications of the deflation of our most recent bubble. Their detailed biographies are in your conference materials.
We will hear first from Desmond Lachman, who is a resident fellow at AEI, having been previously a Wall Street economic strategist. His research includes global currencies in emerging market economies, multilateral lending institutions, and lately, the housing bubble. Desmond and I have been reinforcing each other’s bearish outlook on this for some time now.
Next will be Nouriel Roubini, who is a professor of economics at the New York University Stern School of Business and is chairman of Roubini Global Economics. He has also served as the senior economist for International Affairs at the White House Counsel of Economic Advisers, among many assignments, and has written provocatively on today’s topic.
Our third speaker will be Chris Whalen. He is the senior vice president and managing director of Institutional Risk Analytics to which he brings experience as an investment banker, research analyst and a journalist, including working in equity and fixed income, as well as risk management. Chris has been my excellent partner in organizing this conference. Thank you again, Chris.
Tom Zimmerman will be our last speaker, bringing us a firsthand securities market perspective. Tom is a managing director at UBS Investment Bank where he manages the firm’s mortgage credit and asset-backed securities research. His research has appeared in numerous fixed-income reference works, and he is a member of the UBS team voted first in the latest Institutional Investors survey of fixed-income analysts.
And now Desmond, you have the floor.
Desmond Lachman: Thank you very much. Before I make my remarks, I should commend Alex for the timing of this seminar. When we discussed this sometime in December, most people had not heard about subprime mortgages. Alex’s timing was so great that even this morning he managed to orchestrate Ben Bernanke to give testimony before Congress indicating that subprime mortgages might be a bigger problem than the Fed originally thought.
What I want to do is put the subprime mortgages mess, as I would call it, into context and talk about other things going on in the housing market. I do this not because I think that the subprime mortgage problem is a minor problem. After all, it was something like 20 percent of all loan originations in 2006, and we have a mere $1.2 trillion of subprime mortgages outstanding, which looks rather dubious. To put that in perspective, that sum is 10 percent of United States GDP. But the reason that I put it in context is that we will see that there are just a whole bunch of adverse factors operating in the housing market. And when you connect the dots, it is very difficult to buy the “sell side” line that has been given on Wall Street and elsewhere that the worst of this crisis is behind us.
My view is that we are in the early innings of a long drawn-out process in which we will see housing prices drop by a considerable amount before this is all over. We will see this has been a big drag on the United States economy. Before one prognosticates, it is always good to know where one is coming from, so the way in which I’m going to organize my remarks is by first asking where we came from. Next, where are we? And then finally, where are we going?
Now, I think, what best encapsulates where we are coming from is this chart by Robert Shiller looking at housing prices over the last 100 years in constant dollar terms. What is really very striking is that we see between 2000 and 2006 something with absolutely no historic precedent in the United States. We see house prices increasing by something like 80 percent in constant dollar terms. If we look just at the post-war period since World War II, house prices generally oscillate in a very narrow range and do not really move much in constant terms, as I say, until you get to the last several years.
This has pushed housing affordability way beyond the reach of most, and the point that I would emphasize is that here we have a boom of incredible proportions, so I’m not too sure that previous busts are that great a guide. I think that we are in new territory, that this is a bust that is going to be of a great magnitude, and therefore, you should be expecting the fallout to be all the greater.
The second chart is somewhat a mirror image of the first, which encapsulates what has been going on. Home ownership, which used to hover around about 64 percent--the rate of home ownership in the United States for much of the period between the 1960s and the mid-90s--suddenly moves up to pretty close to 70 percent. So, essentially what has occurred is a whole lot of people have been brought into the market, driving the prices to very high levels.
I would submit that demographic factors or fundamental factors of that sort cannot explain the tremendous boom in house prices in the last six or seven years. One really has to look towards financial factors, and I think that there is a confluence of four factors that contribute to this. I do not think that the Federal Reserve Board comes out very well in this procedure, not looking at asset prices. I think that they were pretty much asleep at the wheel as this was all occurring.
The first factor driving the house prices was the easing of monetary policy in the wake of the dotcom bust in 2001. Interest rates were reduced to one percent and kept at very low levels. They were kept at one percent for a long while. And then, the move to neutral policy was very gradual.
So, we had a bunch of liquidity created by the Fed that I think started the problem. But then what occurs at the same time--and I’m not quite sure what the regulators were thinking or what the regulators were doing when this was going on--was an explosion of subprime lending; lending to people who really are not creditworthy, and by the time we get to 2005 and 2006, subprime lending is something like $600 billion. If you add Alt-A lending, which is not a whole lot better, you have something like a trillion dollars each year being lent, and this is something like 35 percent of the market. So I just do not see this as a niche problem, I think that this has been a very big driving force.
The other two factors that I would mention were the exotic instruments, whether they were adjustable-rate mortgages, whether they were interest-only loans, whether they were negative amortization mortgages, and these effects go beyond the subprime universe such that something like a third of all loans now are adjustable-rate mortgages.
The final factor that I would like to emphasize because I do not think it gets sufficient attention in all of this: people observe there is a certain amount of speculation in the market. I think that might be an understatement. If I look at these numbers by the National Association of Realtors, they are suggesting that investment purchases of houses has been something like 25 percent. To put that in perspective, historically the figure has been something closer to 12 percent. So we got a lot of froth, and what bothers me about that is not simply that this will dry up going forward, but you will get unwinding of positions that can really complicate matters.
Just to round out where we have come from, the legacy of all of this is that the households in the United States are now stuck with a lot of obligations, and Alex is very fond of telling me that one of the things that he learned as a banker was that liabilities generally do not shrink. It is the assets that shrink, and people are going to be stuck with these obligations.
The other thing from a macro point of view that is dramatic is that the boom in housing prices, coupled with mortgage equity withdrawal and all the like, has induced American households to reduce their saving rates, which used to be something like seven percent of disposable income. We now have negative rates of saving. So, we really have quite a party going on. I do not want to get into external imbalances, because I think the story is depressing enough as it is.
Where are we now? In a word, I would say that we are in a situation of housing oversupply. Whatever indicator you want to look at, there is too much supply, and that is causing prices to begin to weaken. If you have an overhang of inventory, what you expect is prices to weaken. I think that vacancy rates might be the thing to look at. These are houses that are sitting vacant; nobody is renting or living in them. This used to run at something like 1.5 percent. It has suddenly gone up by 50 percent and it is now 2.7 percent. So, it is telling you that there are too many houses on the market.
If you do not like that indicator, you can look at housing inventory relative to sales. It has now gone up to seven months sales, and we are back to the levels where we were in 1995.
Another side is housing completions still run at a high rate, more than the fundamental demand, which could weigh on prices. Finally, I realize that there are many indicators of prices, and whether you have smoothed them out or not, they all seem to be telling the same story that house prices that were increasing towards the end of 2005 at a 15 percent annual rate are now actually declining.
That brings me to where might we be going. I took Economics 101 and the idea there was that if there is more supply than demand, generally prices fall. I have already mentioned that we were in a situation of excess supply. So, the question you have to ask yourself is what is going to happen to demand and supply? Is demand going to operate in a way that is going to reduce that excess supply? Are we going to get supply being reduced? And the answer that I come to is no.
What is really going to be occurring is that as the factors that drove this process go into reverse, we are going to get demand shrinking because of foreclosures and because of unwinding speculative positions; we are going to have supply increasing. So, I find it difficult to see how this is not going to get worse. So, the factors that go into reverse are, first, the Fed has brought interest rates back to a more normal level, so affordability issues are going down.
The mess in the subprime sector reflected in rising default levels just means that what you are going to get is either credit crunches developing or the incentive of people making subprime loans is drying up. We have already got something like 30 of these subprime lenders going out of business. There are many more down the track. Just to put it in perspective, if subprime and Alt-A is something like 35 percent of all originations in 2006, it does not take much of a decline in that segment of the market to produce sales dropping by perhaps 10 percent just from that factor.
Then there is the issue of the bank credit standards generally tightening. The regulators suddenly realize what they are supposed to be doing. In typical fashion they are operating in a pro-cyclical manner, so the last time that we need banking standards to be tightened is now. It is just making sure that the demand goes down. I have mentioned that we have adjustable-rate mortgages with something like $400 billion of resetting at two percentage points higher. So, that is another factor restricting demand. And then finally, I would just say the speculative side really does that in.
Let me conclude by giving my two cents worth on the macroeconomic impact of where this goes. I’m not talking about other problems that we have; high oil prices, or dollar weakness or factors like that. Just looking at the housing side, the way I would see it is, you have to look first at the direct impact of the housing market. If you have lower construction, that means that you are going to have fewer jobs in that industry, and you just take a look that housing is something like six percent of GDP. So, if you have a 30 percent decline in housing activity in 2007 that is one-and-a-half percent shaved off GDP right away.
I think, though, that the more important factors are going to be the indirect impacts. That is a little bit more difficult to quantify because there is a lot of psychology involved. American households, as I mentioned, ran down their savings from seven percent of disposable income to zero because housing prices were rising at 10, 15 percent a year. The question you now have to ask is, is that not going to operate in reverse when households figure out that their houses are not appreciating but in fact are declining? Is that not going to freak them out? They do not really have savings. Somehow they have to restore their savings, and the various wealth effects would tell you that we are shifting from a 15 percent increase in house prices to a 15 percent decline--we could easily be talking about one-and-a-half to two percent further shaving off GDP growth. So when Greenspan’s talking about a 25 percent probability of recession, I think he might be being a little modest. But then, I’m no Greenspan fan. I think that most of this problem really lies at his door.
Just finally, I should say that the other indirect impact that I find very difficult to quantify is that when I look at financial markets, risk is not priced anywhere. And I have no way of quantifying that when we get losses all over. How many dead bodies are going to be floating on Wall Street, and what are the ramifications of credit derivatives and the like, but I really do not need that in my case just to say that we are in for a pretty rough ride. And I would be so bold as to say that I think that this issue is going to be the issue in the 2008 election campaign.
Alex J. Pollock: Thank you, Desmond. I thought securities were floated in Wall Street, not bodies. Nouriel.
Nouriel Roubini: Thanks. Like Desmond, I would like to discuss how the fallout of the subprime meltdown is going to have implications for the economy. I think that is one of the crucial things, because right now there is this debate ongoing between the consensus that says the economy is going to experience a soft landing and the alternative view that actually we might experience a hard landing in the form of either a growth recession or an actual recession. I’m certainly of the latter view.
I find it a little bit curious that we are only talking about subprime mortgages, because if you think about it, we are in a subprime economy. I’m not talking about it just metaphorically. Think about it: we have dozens of millions of subprime credit cards, and even before you default on your subprime mortgages, you are going to default on your credit cards. We have dozens of millions of subprime auto loans, and today those are important.
Bloomberg and the S&P say that the subprime auto loans have sharply increasing default rates. And there is another piece today I thought was interesting, which suggests that 20 percent-plus of all the loans that financed the purchase of Harley-Davidson Hogs are also subprime, and the delinquencies on them have gone up since last year from two-and-a-half percent to five percent today. So the point is we are talking about subprime mortgages, but it is also auto loans, credit cards, and other things. There is a whole economy that is subprime. The spillovers are going to go to the other parts of the mortgage market, other parts of consumer credit, and also to corporate credit risks.
I think that the consensus view has been faulty since last summer. The consensus was wrong then, and then they had to admit there was a subprime problem. Now what they are telling you is that it is just a niche problem; there is no contagion from housing to the rest of the economy, no contagion from subprime mortgages, and so on.
So, I would like to address some of these consensus views and try to make some points on why they were wrong then, and they are going to be wrong again now.
First, a consensus has said since last fall that the housing recession is bottoming out.
I have a long paper on this. I’m not going to go into details, but essentially it says that we are nowhere near close to the bottom of the housing recession. In the typical housing recession, housing starts falls by 50 percent approximately, and in some of the deeper ones by over 60 percent. We are down only 30 percent. It has a long way to go. Consider any indicator you have right now from the housing market, whether it is building permits, whether it is housing starts, whether it is construction, whether it is completion, whether it is the demand for new homes--it is just heading south. The glut of existing and new homes is becoming worse by any standard, and the price pressure is downwards.
The Case-Shiller index that came out yesterday is already now showing falling prices, but a lot of it is actually seller side incentives that are not measured. When you get a $40,000 free swimming pool when you buy a $400,000 dollar home, that is a 10 percent price cut that does not show up in any one of the official numbers, so home price are already falling today at the rate that is closer to 10 percent, even if the official number is telling you otherwise. So if you look at any indicator of the housing market, before we even talk about subprime mortgages, it is a disaster. This is going to be the worst housing recession since 1960.
Second, the consensus now says, “We have a subprime problem.” Now whenever they say the word “subprime” they attach to it meltdown or carnage. This has become almost automatic, when three months ago they were not even talking about the problem. But now the consensus is that it is just a niche problem, it is only subprime; it is not the rest of the mortgages. But think about the reckless lending practices that were essentially being used for the last four, five years. You have zero-down payments, no documentation of assets or income--what people refer to as “liar loans”--interest-only mortgages, teaser rates, negative amortization, option ARMs.
Was it only subprime? Look at the numbers, it was subprime, Alt-A, piggyback loans, home equity, and also a good chunk of the option ARMs, subprime, near prime, prime. If you look carefully at the numbers, I would argue that about 50 percent if not more of all the origination of mortgages for the last couple of years were things that I would consider reckless. It is just toxic waste.
So that is what is happening. Of course, the rate at which Alt-A and the other stuff is going to start defaulting and getting into trouble is going to be later. It is going to start with subprime; it is going to go to all the other mortgages. But the idea that this is just a niche, that subprime is only 10 percent of the stock of mortgages and therefore is not a problem is just nonsense.
Now that people are recognizing we have a total mess in subprime, there is also a credit crunch in subprime. About 30 of these lenders have already gone bankrupt in the last three months. But again, the problem, they say, is only “a niche problem.” “It is only going to be a mini-credit crunch only for the subprime section.” The reality is otherwise. When you look at the whole series of indicators, the chart that Desmond showed about how banks are getting more worried and tightening standards, they are not tightening only for subprime. They are doing it across the board. The borrowers are now facing a credit crunch.
The regulators were asleep at the wheel for six years under the ideology they should not regulate markets and they let this thing fester and grow. So, now they are cranking it on the other side. We have seen it in every cycle before. There was a nasty credit crunch and we got a recession in 1990. This time around it is going to spread from subprime to other mortgages. It is going to spread to consumer credit and to the rest of the economy.
A fourth point: people say the residential mortgage-backed security market is still kind of okay. As long as it is okay then it is going to be financing all the rest. I think there is already evidence that there are massive losses in the CDO market. A recent study by Rosner and Mason shows that if you are going to have a significant disruption in the CDO market, then the whole financing base for the residential MBS market which peaked at about $1.3 trillion of issuance of new residential mortgage-backed securities last year is going to falter. That is the kind of thing we are facing. Securitization helped the growth of this credit boom and bubble, and the squeeze now on the other side is going to create a mess on the securitization side.
People say there is no contagion to corporate credit risk. Those spreads are still relatively low. Guess what? In a matter of two weeks the CDX spreads for firms such as Goldman-Sachs, Merrill-Lynch, Morgan-Stanley, went from triple-A to near junk rate. There is a study by a colleague at Stern, Ed Altman, who is the world’s leading expert on corporate defaults, based on fundamentals, default rates for corporate debt today should be around two-and-a-half percent. Historically, they are three-and-a-half percent. Last year, they were only .6 percent, 20 percent of what they should be given current fundamentals. Why? With just a massive amount of liquidity coming from private equity, levered institutions, lots of firms that are under distress are being refinanced out of court. They do not go through Chapter 11 but under serious distress there are tons of junk being issued right now. Once the party is over, and I would say the party is going to be over soon, all these problems are going to come to the surface. You are going to see massive increases in corporate defaults, back to normal and worse, and then the spreads are going to go through the roof. You will see the contagion is going to take a few months. Until now people said this is just a housing recession and is not affecting the rest of the economy. That is actually incorrect. We do not just have a housing recession. We have an auto sector recession. We have a manufacturing recession. We have every single component of investment that has been falling since Q4.
People said we never have a hard landing until the consumer is faltering--since consumption is 70 percent of GDP--and the consumer is resilient. The trouble is that consumption depends on four things. It depends on job and income generation, interest rates, wealth, and the debt servicing ratio.
We are already seeing losses of jobs that are going to accelerate in housing and manufacturing, and that is going to slow down job and income generation. You have now a credit crunch, and once there is a credit crunch, there rationing, so the price does not go up, but the quantity gets cut. So the quantity of credit is shrinking. That’s what houses are facing right now--a credit crunch.
Until now, the households were consuming more than their income--negative savings--because they were using their homes as their ATM machines. As long as home prices were going up, you could continue this. Now home prices are falling, and home equity withdrawal was at a $700 billion annual rate in 2005; Q4 is down to $270 billion. In the meanwhile, the debt servicing ratio is going up. This year alone, you are going to have something like $1 trillion of ARMs that are coming to being reset at much higher interest rates. So the issue is that the consumer is on the ropes, is being squeezed, and now, we are having two consecutive months of retail sales that are pretty much flat.
So the argument that this is just a small housing recession and is mostly a subprime problem does not have a good basis. What we are facing right now is a serious situation in which the economy is heading into a recession. A good chunk of the economy is already in recession.
Now, the consensus has it that this quarter has at least two-and-a-half percent GDP growth. But how could it be? The 2.2 percent number for Q4 was before the subprime collapse, before we had the lousy number of consumption, before we had the collapse of capital spending and investment by firms. How could the consensus tell you that the growth rate this quarter is going to be better than last quarter? It just does not add up.
Will the Fed come to the rescue in spite of what they say? They will try to come to the rescue. Is it going to make a difference? No difference at all. Once you have a glut of capital goods, what the Fed does will not make a difference.
In 2000 the Fed was behind the curve. They worried until November of 2000 about inflation rather than growth. Like today, there was a tightening bias. Then from November to December they went from tightening to easing and two weeks later on January third, NASDAQ opened after New Year and collapsed. Then they cut rates in between meetings and they slashed rates very aggressively. Did they avoid the recession? No, they put a floor on it and the simple reason why is that the Fed Funds went from six-and-a-half to one; and real investments fell by four percentage points as a share of GDP between 2001 and 2004.
Then you had a glut of capital goods and tech goods, and this time around it is housing. Until you work out this glut--and it is going to take years to work out the glut of housing, the same way it took five years to work out the glut of tech--we are not going to see a recovery. So the Fed is going to cut rates. Are we going to avoid a hard landing? My answer is no.
Alex J. Pollock: Thank you, Nouriel, for that incisive outlook. The co-author of the paper you cited, Josh Rosner, is with us today. Josh, thanks for coming. Let’s go on to Chris.
R. Christopher Whalen: Thank you, Alex. Especially on behalf of Professional Risk Managers’, I want to thank AEI for putting on this event with us.
I would like to spend a few minutes talking specifically about where we have been, where we are, and where I think we are going to be with respect to financial institutions. I want to cover a lot of the same ground that you have already heard, but by looking at a couple of bank business models, making some observations on their default experience. This is also referred to as “charge offs.” Then I will make some comments about the asset quality of US banks and the major sources of revenue that they have enjoyed over the last couple of years because of the boom in mortgages, where I see those factors are going.
First, how far down is down in terms of banks? My firm spends a lot of its time benchmarking bank financial performance and credit quality. The real peak for credit losses was in 1991, and I would like to use Citibank NA, the lead bank of the Citigroup organization as my surrogate.
What is interesting is even half a decade into the rate increases by the Fed, default rates still remain very low. In fact, charge offs in 2006 may have been the trough for the last 10 years.
Now, as I mentioned, Citibank is one of my favorite surrogates. Citibank credit loss peaked at 330 basis points in the fall of 1991; that is 3.3 percent of total loans and leases. That was a pretty big hit. That was the time when people worried about the solvency of major banks. There were rumors of rescues by the Fed. You had events like the insolvency of the Bank of New England and de facto rescues by the regulators. What is interesting though is the recent near-term loss peak in the year 2002 was only two-and-a-half percent losses or charge offs on total loans and leases. That was bad, but it certainly was not as bad as the early 1990s.
Now this chart you see up here, this is Citibank NA versus a large bank; peers going back 18 years to 1989. What I would have you notice is it kind of looks the way you would expect. In other words, if you were looking at a GDP chart or a chart about interest rates, it would follow basically the same form in the terms of the timeline. You have in the early 1990s very high default rates at Citibank, as you can see well above peers which are typical for them. They generally have a more subprime or a more high-risk profile in terms of their lending compared with, say, JP Morgan or particularly, Bank of America.
Then we have this period, the 1990s, very quiet, especially Citibank when losses go down below 100 basis points or one percent default per year. That is actually extremely good for them. Then you can see again 2001, 2002 period. A pretty serious spike, but notice how long it lingered. There actually were four years where Citi’s default rate was just above or below 200 basis points.
Now, the difference, of course, in the next bank example we are going to look at is the mortgage sector. Since the early 1990s, defaults in mortgage products generally have been minor. They have been so low for so long that most of the mortgage bankers I know have not even bothered to update their default studies. They just have not had enough events to put into the work. So why bother? You just buy the vendor default calculator and everything is fine. Some of the bigger, more pure mortgage models like World Savings, which was acquired by Wachovia last year, actually reported zero or negative defaults for consecutive quarters.
Now this is the portfolio data from the FDIC. The dark line is Washington Mutual. The pink line is World Savings as you can see kind of crawling along the bottom of the chart. And then the peer average. This is about 70 members of the mortgage specialization peer group that the FDIC maintains. What I would ask you to notice is the way defaults have fallen off in that period basically from the mid-1990s onwards--just absolutely extraordinary. Defaults here are very low. You are talking about for Washington Mutual, for example, they were reporting only 10 or 15 basis points of defaults on their mortgage portfolio.
Now a couple of observations about subprime. I totally agree with what Nouriel was saying about subprime. There are many different kinds of subprime credit. The way we find them in our work is to look at the yield on the loan portfolio. If the yield is above 10 percent or mid-teens, that is a subprime portfolio, at least in my opinion, as well as when you see high loss rates, such as over 300, 400 basis points (or three or four percent) loss rates in a given year.
The “loss-given-default rate” is a Basel II term. What is your loss rate after default? Most credit card banks, most subprime lenders, they will be in the 90 percent range. If you default on the loan, they just take it and sell it to a collection agency or a hedge fund for a couple of cents on the dollar. So, those are the kind of indicators we look for to identify subprime lending.
There was a great piece in the Journal a couple of weeks ago about a guy who buys homes from banks out of foreclosure. It was a very interesting business model. He would pay $8,000-$10,000 thousand for a house out of foreclosure, mark it up to $30,000 or $35,000, and put it out there in the market with a very minimal mortgage payment. He would typically hold on to the title for at least a year before he gave the title to the buyer, just to see how they did. Then he still had a 30, 35 percent default rate. That is a tough business. That is a business where you got to have at least 15, 20 percent capital to total assets to see you through the tough times because you know that portfolio is going to throw off 10, 15, sometimes 20 percent defaults in a given year.
Now one of the interesting things is that if you look at the different lists HUD has available on their website, they have six years of lists of subprime and Alt-A lenders. Early on, these are small banks, but none of the big guys really. This was basically a private sector business with a couple of regulated financial institutions involved, but you did not see the big guys. Today, you do. Today, when you look at the list you see Citigroup, you see Wells Fargo, you see a number of large bank holding companies who have all gotten into what used to be, as I say, a private cash business. This was a tough business to finance--not anymore. You have hedge funds who want to get into this business now.
Now, this is a chart of a very interesting non-bank, GE Money Bank, one of GE’s credit card subsidiaries. What I would have you notice about this series is how it does not look like that original series for Citibank, which had some rough correlation to the economic and interest rate cycles. You see default experience ranging between five and 10 percent a year and then this big anomaly. Now that anomaly is all GE Money Bank. And I left it in for a reason, which is that there tend to be a lot more idiosyncratic events with subprime lenders.
If you go through them as a group and look at each one, almost every one has an event like this, which has nothing to do with interest rates. They screwed up. They had to essentially write off 25 percent of their portfolio in a single year. That is pretty tough. But that describes what kind of business this is. When you talk about subprime, you are really talking about something that is a triple-C, double-C bond equivalent rating.
The thing that scares me the most as an analyst is that mortgage loss trend. This really frightens me because when you see a group of financial institutions essentially telling you that there is no risk on their portfolio for years at a time, that should put up red flags everywhere. It did, I can tell you, among the supervisory community because they have been out trying to head this off for the last couple of years, pulling people out of retirement and placing them as consultants with financial institutions. There are a couple of proposals that have come out through www.fbo.gov in the last couple of months. They all have a very clear supervisory focus. In fact, we will probably be going to hear the final release within a month or so of the revisions to the shared national credits.
That is again going to be focused on CDOs, syndicated risk enhancement, even hedge fund reporting. You may see some credit rating aggregated for hedge funds if they are a significant part of a bank’s risks. But all of it is trying to get their hands around the risk that this chart shows you, which is essentially that the risk is not priced. As Desmond was saying before, when you were doing these deals a year and two years ago in the CDO market, they were going to hedge funds and saying, “Hey, we will pay you a point-and-a-half or two points per year to guarantee a portfolio” that in normal times should be throwing off 10 percent defaults a year.
What is wrong with this picture? This is what scares regulators. It is not that the banks have kept a lot of this risk; it is that they have sold it to somebody else. Now the million dollar question in Washington is, where is all of this risk that has been packaged and sold by the sell-side to buy-side investors?
So, I think that the performance of the mortgage industry, if you look at the default rates of the banks, is a pretty good surrogate for the marketplace generally, especially for things like bond market spreads and credit derivative swaps. My personal view is 2006 was probably the low in terms of bank default experience and it is going to be up from here. I think that for a number of reasons, we are several standard deviations from the mean even today. So, if the Fed starts easing interest rates later this year or early 2008, I still think defaults may rise for quite a number of years just because of what is baked into the pie.
It has been mentioned about weakening home prices; I totally agree with that. I live in the New York area. I have a lot of family in the real estate business, and things are quiet. The great fiction that they are trying to maintain up in New York is that the New York City market is okay, but that is not true. The flow in New York City has dribbled to about nothing, and the story that I have heard a couple of times now are sales have failed because people in the business have lost jobs. I do not like hearing that stuff on weekends from family members who make their living in the real estate market in New York City, but it is getting to be an increasingly common problem.
The last issue I would leave you with is this. Too often when people react to subprime stories they make the mistake of thinking just of about mortgages. But I think the scope of the problem that is embedded in our financial system right now is almost unknowable at this point. Why? Imagine I’m a trader in a hedge fund. I own a bunch of CDOs that I have been buying for the last three years from our prime broker and from other brokers who service us. I get my valuations for these products from those dealers. There are no public prices for these securities. I could call two dealers, and I’ll get two different prices that will be widely disparate in many cases.
There is absolutely no way for an auditor or manager of a bank or a mutual fund or a hedge fund to verify the value of these securities, until you try to get a bid for them. When you go back to the broker-dealer, and you say, “You know that tranche you sold us for 115 over--what is your bid in five or 10?” You are going to find out that it is probably 200 or 250 over, because the stuff is totally illiquid and if you think of it from a broker-dealer’s perspective, what is he going to do once he buys it from you? It is just going to become dead inventory on his book. So I think the issue of liquidity is the issue I would want to leave you with today because it really affects this entire market, from the homeowner on through the end-investor in the asset.
Alex J. Pollock: Thank you, Chris. We hope you will remember Chris’s question, if the banks sold all this risk to somebody else, where did it go? We are hoping someone in the audience will tell us the answer to that. We ended up on liquidity which sounds a lot like the securities market, so Tom, you get the last word.
Thomas Zimmerman: Well, there are always problems going last. Most of your thunder has been stolen one way or another, including Alex’s first comment, because that was what I was going to start with. Bull markets are bull markets. We see a lot of them and we always know what happens when they end. A lot of people love it when they are going up and then when they come down, they do not like them. In the subprime world there are going to be a lot of victims; it is not just the homeowners. There are a lot of guys who will lose their jobs in these subprime originators. They are all going out of business. They are not going to be around.
You are going to see all the Wall Street operations shrink in size. The subprime market going forward will be half of what it was in the past few years. We do not need so many bankers or research people. So, a lot of people will get hurt in this thing, and not just the homeowner. It is classic bull-market enthusiasm and classic bear-market pain.
In terms of this topic of contagion, it is a funny world. For the last two or three years, when I have appeared on mortgage panels around the country, I have always been way over there on the bearish side. As time went by I got to the middle, and now I come here to this group and I’m on the other side. I agree with a lot of the comments made here today. I really think that this is a big, big problem, and I have been saying that for some time. I do not know if it was quite as big as has been said here, when we talked about some of the numbers. But clearly, this is not a small issue. It is a big issue.
I would like to spend a little time with my slides and go and drill down in some of the details of this subprime world itself and then talk about extrapolations from that. Most of my presentation is focused on the subprime market and describing how I saw it unfold.
Now, here is a slide that I gave at a previous seminar about what was happening in the subprime markets. So, if you read those questions you would say, “Yes, there is obviously something going on here.” The investors were asking these questions. They are not naïve. Isn’t this fast growth unsustainable? Aren’t these standards dropping? Aren’t the new loan types going to crater? This was in January 2005. So, this is not a new story. The story has been out there for quite some time. We knew it was coming. We knew it was a problem and it was going to happen.
Here is my September 2005 forecast of what was going to happen when the market slowed down. This scenario is like our average subprime world of the last seven or eight years, not the last two or three years, but the previous seven or eight years. Housing appreciation is five to seven percent. The total loss over the life of these loans is four percent.
Now of course, when the housing market slows down, here is what we thought was going to happen. These numbers would go up by factors of two or three. No question about that. This is subprime.
The next couple of charts go beyond subprime and take a look at Atl-A subprime and prime. These next three charts show--these are default rates and they are linked to housing price appreciation or HPA--the higher the HPA, the lower the defaults--right across the board. Loss severities are dramatically linked to HPA. That is almost a no brainer.
You take how many times people default, times the loss severities, you get actual losses. If you slow the housing market down, losses are going to go up. That is almost physics. So, we knew this was going to happen.
Now take a look at this subprime number--this is historic data at zero to five percent HPA. We estimate eight percent losses. That is an ugly number to a lot of CDO investors. The old numbers were three and four percent. So, if the market slows down, the housing market slows down to a flat market, and there are a lot of problems. This is what we were talking about for the past couple of years.
Now, the longer history. These are the cumulative defaults in the subprime market for the last 10 years, and the real bad years were 2000-2001, really ugly stuff. And 2003-2004 looks really great. You take these curves and you can standardize them and normalize them. What percentage of these losses occur when? By the end of the second year, maybe 15 percent of the losses have occurred. So, we knew this train wreck was coming for some time. So, we figured, out here a year or two or three years from now, we are going to see lots and lots of problems. That is what always happens. Well, we got surprised. We got surprised like everybody else did.
The surprise was the thing called “early pay defaults,” or EPDs, out of the blue. No one expected they were coming. I did not. Nobody that I know in the industry did. We were out talking to some subprime guys early in 2006 and they were saying, “You know, one thing we are worried about is EPDs are really creeping up fast. We do not understand what it is.”
An early pay default is when the person takes out a loan and does not make the first payment or maybe the second or the third payment. This used to be a small niggling little problem, say less than one percent of the loans. These things soared to six or seven or eight percent by the middle of 2006. That is what knocked the subprime business out. It was not the normal default curve. It was these EPDs that knocked them out, not the regular default statistics.
What happened when people originate these loans, typically they will sell half of them to the Street. The ones they sell to the Street, they sell them with reps and warranties, which typically say, “When I sell this loan to you, if within the first two or three months there is an early pay default, you can put that loan back to me.” That is what took down all these subprime shops. They were selling loans to Wall Street. Wall Street took a look at them and says, “Hey, these are EPDs. Take them back, guys.” “But wait a minute; we do not have the money to take them back. We securitize loans. We sell them. We do not take them back.” “Okay, and now you are taking them back.”
So, a lot of the smaller capitalized shops -- a lot of these small guys are gone. They just went out like that. These EPDs also really hammered the big guys: New Century or Fremont or even WaMu and Countrywide. All got hit with this stuff. But the bigger guys could survive it. The smaller guys, they were going.
What is interesting to me, and we still do not know what caused it exactly and who these people are. Now we know part of what went on was just underwriting standards. We talked about how loose underwriting got in this industry. These are statistics for the subprime world for the last five or six years. You can take a look at the couple of things that really look pretty ugly like these second liens. They went higher and that is really bad. These interest-only loans come up at 30 or 40 percent in the market. Take a look at debt to income. It is increasing. What is really bad is that the full-doc loans went from 73 percent down to 56 percent. And on the far right are these 20 percent seconds associated with the 80 percent firsts. So, by the end of 2006, 25 percent of these loans out there were these 80-20s where you actually were giving people a loan with no money down. This is what was creating this perfect storm, and we are now seeing the fruits of it.
Now, here is the breakdown of how many 2/28s and 3/27s there were in this subprime product. About 60 percent are 2/28s and 12 percent are 3/27s. Now, what is 2/28? A 2/28 is a hybrid ARM. The first two years are fixed. Usually, at some rate linked to two-year swap rates. It runs maybe 200 basis points above a 30-year fixed rate loan. After the two-year period, it will automatically reset to an ARM, and that ARM is six-month Libor plus six percent. That is now about 11 percent.
So, these 2/28s that were being created for this industry probably were the wrong product for a subprime borrower who has little margin anyway in terms of income. Now, you have got him into 2/28 and now you are going to give him an 80/20. These are 80 percent firsts, 20 percent seconds, no money down, and now you are going to give him a low-doc loan where he does not have to identify who he is or what his income is. He just sort of says what he has. So, this product by the end of 2005, early 2006, was really, really a very scary product waiting to blow up. It is not surprising it blew up. I still say it is surprising that it blew up so fast.
This is a good chart because it shows what these 2/28 individuals are faced with. The blue line, the most volatile line, is six-month Libor plus six percent. The pink line is the actual loan rate for the two-year part of the 2/28. That is the loan rate these guys were paying for the first two years. And the yellow line is the pink moved forward two years. So, if you take a look at the yellow line it tells you where you are at today. So, for instance, in January, in March of ’07, we are at about seven-and-a-half percent. That means if our loans were getting ready for a two year reset, I have seven-and-a-half percent loans the last two years; and now, my choice is to reset to this current pink line, which is up about nine percent, or go up to about 11 percent. That is the refinance choice these people have.
Now, in addition to the lousy underwriting we just talked about, we know what happened. The housing market has slowed down. These are just the data showing the housing market is slowing down. We all know this.
The question is who are these EDP people, this six percent of this billion-dollar loan packages? Who are these six percent people who took out a loan and then a month later or two months later did not make the first payment? I do not know, and no one knows exactly. My guess is some of them were speculators, who for the last several years have been given zero percent down loans and go out and speculate on houses. They tried it one more time. Wait a minute. The market is not going in my direction so I'll just walk out of here.
Or it could be one of these serial re-fi people. If you were back in ’02, ’03, ’04, and you had a subprime loan and you were in that yellow line, hey, you can re-fi out, go down on to that pink line, lower your loan rate by 200 basis points, and by the way, take cash out because the market is going up 20 percent a year. That was a great trade for all these poor people we are now worried about.
They had a pretty good time for a couple of years. They were doing very well. They were rolling out of those loans into lower-rate loans and taking cash out of their houses. It was, as we say, a win, win, win for everybody during that period. So, I think some of those people got caught, they just re-fied one or two many times. And who knows, if you do not charge anybody a down payment, maybe many young persons decided to just take out the loan to live in the house for a year and then walk away. I do not know who these people are. But there are a lot of them, and that is what put the industry down. It is not just the default rate we expected to see in the coming years.
Now, the other thing that hurt the industry, of course, was that the gross profitability that they have for the past four or five years disappeared. When Libor dropped dramatically in ’01-’02, and their loan rate, which is this 2/28 initial rate on the top light blue line, the net margin they had just exploded during this period of ’02, ’03, and ’04. There were companies like Ameriquest who were taking a billion dollars a year out of this market. It was just boom, and there were enormous amounts of money being made in this industry during that period. As Libor rises, their rates rise a little bit, but not nearly as fast. You go back to normal environment or close to a break-even kind of a situation.
Now these early pay defaults really pushed them over the edge. So, by early 2006, their margins which had been enormous had shrunk to pretty small amounts, and these early pay defaults were always there, but at a very low number. Now, you go by early 2007 or late 2006. If you calculate that those early pay defaults are eight percent of your loan packages, you are gone. So it is not a profitable industry.
How do you get rid of the early pay defaults that are sinking you? Well, you’ve got to reduce what you are doing. You’ve got to shrink. One of these companies reduced their production by 25 percent, and their early pay defaults went from six to three percent. They are still getting killed.
So, you have to really change the product that you are producing to get rid of these early pay defaults, and when you do it you shrink your volume dramatically. The problem is that the last five years, you built overhead dramatically. So now, you’ve got a really lose-lose situation, and that is where they are at right now. That is why these guys are going down.
So, the early pay defaults were the mechanism that triggered it, but it was a larger problem of having underwritten these lousy loans and having a profit margin already squeezed. If you put them together, you got massive defaults across this industry.
what is going to happen? The industry is going to create a new product; they have to. It is not going to be 2/28. It may be a 5/1. It may be a 15-year fixed-rate loan. I do not have a clue what it will be, but they are right now trying to figure it out. We are in the middle of the transition. No one knows what it is going to be. But it is not going to be a 2/28, 80/20 low-doc. We know that. The industry is in shambles; the only people who will survive are the very large companies who are a part of a major commercial bank or investment bank. The rest of them will be gone.
Now, the government steps in. In December 2005, it came out with the new rules about how you should underwrite certain types of loans. There is big discussion to include language that says, “Every loan you write must qualify the borrower at the fully-amortized, fully-indexed rate.”
There is some logic to that. If you are going to qualify a person for a loan, once you fully indexed it and you fully amortized it, they should be able to carry that loan. But the first set of rules that went out only said if it is an IO or an option ARM, those rules apply. So, the subprime industry thought, “Whew, we are off the hook. They did not get us.” But a month later, comments start to be seen in the press: some people think that maybe a 2/28 is also a pretty risky loan. Now, we know what is going to happen. Within a month or two, the new rules will be out and the 2/28s and the 3/27s will now have to be qualified at a fully-indexed, fully-amortized number, which is 11 percent now. As one executive from WMC said in a recent conference, “That is 75 percent of my business.”
So, it is a whole new ballgame. In six months, we will have a new industry with half the volume or 60 percent of the volume, a new loan type, and the same players, damaged a little bit, beaten up a little bit. Countrywide will be there. WaMu will be there, other banks will be there. But the problem are the guys who took out the loans in the last two or three years--and this is back to some of the questions and comments made earlier--these are the guys we are concerned about and who Congress is concerned about. These are the people who took out the 2/28s.
Now nobody wants these loans. The bank raters would not let them do it. So, what is going to happen is these guys in the 2004-’05-’06 vintages, mostly 2005 and ’06, when it comes time for them refinance in two years, there is nowhere to go.
It is going be a very different world. There is going to be a higher loan rate, by far, 100, 200 basis points--I do not know what the number is but it is going to be big. A lot more than these old 2/28s, and we will have to put some money down. No more of this zero down financing. That is gone. That is history. So, many borrowers may be able to deal with this higher loan rate, but they are not going to be able to deal with this 10 percent down, which brings us back to the Congressional question.
The question is, do we have a bail-out for these subprime borrowers? I do not know. How are you going to deal with it and how do you figure out whether or not the person who you bail out is like the people who you see on the front pages of the Wall Street Journal and the New York Times, or like those six percent EPDs who decided to take a flyer on the housing market and then decided to walk away. Are you going to save them all together? How do you decide which one to save and which one not to save, or where the money comes from to do the saving? That is beyond me. That is a trap that the government is right now trying to deal with, and it is not easy to try to figure out how to solve this problem, but it is a big problem.
These are some data that show what percentage of the market was subprime. Twelve percent by this measure, 13 and some percent by the MBA’s number of loans. They really underestimate it because they are only covering about 60 percent of subprime markets. So, the numbers may be 17 or 18 percent.
Let's say 15 percent of the market is subprime. Let's say that the old numbers where 15 or 20 percent default have become 25 or 30. Let's say it is 20 percent. So three percent of the loans in the country go bad, something like that. That is why I will say it is not a great big problem because we are talking about three percent of the mortgages. Not everybody has a mortgage. So, it is less than three percent of the homes in America are affected by this; two or three percent are affected.
By that measure, yes, it is going to hurt those people in a certain section of the market, but it is not a great big deal.
These are the delinquencies; subprime delinquencies are going up. Look at the prime numbers. They are going up just as fast. Everything is going up. It is just that in the subprime world, you have delinquencies and default rates that are four or five or ten times what they are in the prime world.
I think I agree here with some of the comments made earlier, even though only three percent of the houses might be at risk of a default from this process in the subprime world. In the last three or four years, Alt-A and subprime accounts for 40 percent of the market. You take that out of the housing market and that is a big vacuum, and that is really going to hurt the new home sales, existing home sales, and the whole housing market.
We will just begin to see the pain from this part of it. It is not really the subprime default rates that are going to kill us. It is going to be the lack of subprime funding that has been going on the past two or three years. What that number is and how that impacts the overall economy--you are a better judge of that than I.
So, Wall Street is going to take some hits here, and their trading desks will get hit, but it is not going to be a serious systemic problem, because only a small part of their revenues come from this business.
Who owns this stuff? First of all, 75 percent of the subprime securitized product is triple A. You are home free, not a problem. I can stress these numbers as much as I want; that triple A guy is not going to get hit. His spreads might wind down a bit, but he is not going to lose a dollar of principal.
That is good because Freddie and Fannie own about half of that triple-A. 75 percent of the subprime market is not a problem. The lower rated tranches, triple B minus--those guys have problems, and they will get hit. Some of them will go down, half of them go down, the people that own the equity pieces will go down, but that is spread out all around the world. That is spread out into private money coming out of Central America. It is state money coming out of China. It is hedge fund money coming out of Europe and America.
It is all over the place. So, the guys are going to get hit with this all around the world. This is really spread out, and it is back to your question about who’s got the risk. It is just all over. In a way, that is good, because it is not concentrated.
So, I do not see that as a systemic problem. If the corporate side starts to get the same thing and we suddenly realize that corporate risk is of a different magnitude, and if it all happened together, yes, then it will be a different issue, I agree. But just the subprime world itself and its Alt-A cousin together, are not going to produce the kind of numbers that would create a systemic financial crisis. It is going to weigh heavily on the housing market for the next few years.
Alex J. Pollock: On that relatively hopeful note, Tom, thanks very much and thanks to the panel for excellent presentations. Let me give the members of the panel a minute for any further comments or responses to other speakers.
Desmond Lachman: I was just a little bit surprised why Tom, after making a presentation like that, describes himself as being relatively optimistic. I thought that what he is telling us is that there are going to be a lot of defaults, that there are going to be a lot of foreclosures, that more of these houses are going to be turning onto a saturated market, that prices are going to decline, and that his HPA ratios are going to go into the wrong direction, and we are just going to have more defaults. I would be a lot more worried about more downward pressure on house prices going forward.
Nouriel Roubini: It may not be just subprime. It might be spreading to other parts of the economy. I think that what is going to happen to home prices is essential because if prices are falling, equity is falling, and equity withdrawal is falling, and then you have a vicious circle.
I think that the crucial question is what is going to happen to the excess supply or the glut of inventory of new and existing homes--it is just going to get worse. It is true that housing starts have fallen, but in the last few months new home sales have fallen even more. So, the ratio of unsold homes is going up. There is a government study that suggests the shrinkage of the subprime market might reduce new home sales by 200,000 this year alone. If you get another 200,000 fall in demand, that is an excess supply versus demand of new homes on the market.
And that is only one channel. We have three other ones on existing homes. Houses are going to foreclosure, and then the bank owns the thing and six months down the line they are going to dump them on the market. That is an excess supply of existing homes. You have the people that did speculative buying and now are seeing their home equity shrinking. And therefore, they have to sell as fast as they can before they get a loss. So that is an excess supply. And you are going to have lots of people that are having these resetting ARMs and they have to decide, what do I do? If I can afford it, I try to sell the home.
So, if you think about the channels through which you are going to have an increase in the supply of homes, both existing and new ones on the market, I think that the situation is that the excess supply that we see today is going to get worse. If that is going to happen, home prices will fall even more.
[Question and answer session not included]
Alex J. Pollock: Let’s show our appreciation for our speakers. Thank you all for coming.
