How Banks Work 5: Selling Loans
March 9, 2008
In our previous discussions about how banks work, we noted that the presence or quantity of impaired loans on banks’ balance sheets tends to get way more attention than it deserves. The problem is generally not loans that have already had issues, but rather the flow of new bad loans, in other words the continuing deterioration of credit conditions, which typically reflects broader economic factors that banks themselves can’t do much about. Thus, rather than having bankers do something about their bad loans, it is best for politicians to do something about the economy.
February 24, 2008: How Banks Work 4: Banks and the Economy
February 17, 2008: How Banks Work 3: More Elephant Poop
February 10, 2008: How Banks Work 2: Shitting Like an Elephant
February 3, 2008: How Banks Work
This can help even those loans that have already had issues, as an improving economy can increase recoveries for loans, or allow borrowers to become current on their payments, thus reducing banks’ losses.
Usually there is a pretty clear economic impairment — very often high or rising taxes, or monetary instability — which can be solved by a policy change. The present situation is somewhat different. There have always been credit busts, but the present one is rather special both in its extremity and extent around the world. (I have heard that, to some extent, the US housing boom in 2001-2002 was centrally engineered to compensate for the recession of that time — and I’m not talking about the Fed funds rate target alone.) Thus, the present situation is a bit of an exception to the general rules I’m presenting here. The government should do what it can to promote economic health, via a better tax system or monetary stability, or at the very least not make things worse. However, to the extent that the present situation does not reflect broader economic issues (high or rising taxes, or monetary instability), then the sloppy lenders and the sloppy borrowers have to take their losses. There is a responsibility for the government to keep this process from getting too far out of hand, which is why I focused on bank recapitalization a few weeks ago.
We’re now hearing that our little adventure in Iraq may end up costing the US taxpayer in the neighborhood of $2.4 trillion, or more (probably). Thus, the government has more than enough resources to take care of the banking system, if necessary, if it wasn’t expending those resources to blow stuff up and kill people in the desert.
Typically, as bad loans at banks pile up, there is a suggestion that they “solve the bad loan problem” by selling the loans. This doesn’t make the bad loans disappear, of course, it merely transfers their ownership. As far as the broader economy is concerned, it generally matters little who owns the loan. (These days, most borrowers probably don’t even know who owns their loan.) As loan losses mount, banks’ capital bases become strained. There are two immediate ways to deal with this: get more capital, or shrink the balance sheet, possibly by selling loans. But, a bank could just as well sell performing loans rather than non-performing. What difference does it make? Indeed, a bank might be better off selling some “good” loans for $1.02, when the bank thinks there might be problems that will make their ultimate value more like $0.96, instead of selling some “bad” loans for $0.20, when their ultimate value might be $0.65. Indeed, to sell “bad” loans for a deep discount can create more losses for a bank. If a loan is on the books at $0.65, and this is a reasonable figure for ultimate value, but the bank sells it for $0.20, then the bank has to take an additional $0.45 loss on the loan, which means more capital impairment.
If a bank can sell good loans to others, then it can continue to make loans, and thus we see that the bank’s capital base is not really a restraint on lending. Indeed, banks have been getting more and more into this business model over the past twenty years, packaging loans and passing them along to other investors in the form of asset-backed securities, etc.
We can also see that, if there is a buyer for a loan (good or bad), then in effect more capital comes into the banking system. It may not be on banks’ balance sheets, but effectively there is some entity willing to make loans, and that entity has capital.
Typically, banks recognize that in an environment where there is lots of supply of impaired loans, but not very many buyers, they are better off holding the $0.65 loan and getting $0.65 for it eventually, or even $0.50, rather than selling it off for $0.20 and taking another $0.45 loss. That’s when you start to hear the economists of the big brokerages start to make all sorts of arguments that banks need to “take care of the bad loan problem” by selling off their bad loans en masse. Of course the journalists pick up on it, and the politicians start to hear this from everywhere that “banks could take care of the bad loan problem, but they are being stubborn and holding on to them.”
This is often driven by the fact that it is quite a wonderful business to buy a $0.65 loan for $0.20. So, the bad-loan vulture investors start to have private conversations with the politicians, about how they could Do Wonderful Things for the National Economy if the stubborn banks would Just Solve the Bad Loan Problem by selling off their bad loans. Often, the politicians relent and actually force banks, through legislation if necessary, to liquidate their inventories of bad loans. Since all the banks are then selling all at once, as mandated by the government, the supply of loans is extreme, and there are not very many buyers, certainly very few domestic buyers, but only the foreign vulture-type investors who have been quietly encouraging this whole process. The banks thus sell the $0.65 loans to the vulture types for $0.20, or $0.10 or $0.02 or whatever. This causes catastrophic losses to the banking industry, at which point the banks themselves are then bought out by other foreign vulture investors, including the big banks of the U.S. or Europe. Thus, the process of economic colonization continues.
Now, selling a loan simply changes ownership. It doesn’t make the bad loan disappear. The loan still needs to be “worked out” in some way, which can take several years. The original bank might have been very careful about this process. They have a multi-year relationship with the borrower. They understand the business, the possibilities, the markets, and so forth. They make careful plans to get the $0.65 value out of the loan, or maybe more than $0.65. The new foreign vulture investor is often a lot sloppier about this process. If they bought the loan for $0.10, it doesn’t really matter if they get $0.20 or $0.30 for it. Either one is a huge profit. They like to do thinks quick and sloppy, book the profits, collect their bonuses, and buy another house in Sun Valley. This liquidation itself often has rather poor economic consequences.
The foreign vulture investor types probably haven’t really thought all this out. Maybe some of them have. Most probably believe the myth that “taking care of the bad loan problem” is the path to economic recovery. They would rather not think about it too much. That could be inconvenient. In the end, they’ll say anything, and believe anything, if it helps them buy a $0.65 loan for $0.02.
* * *
One concept that is blowing up in people’s faces right now is that of “mark to market.” Experienced investors know that, often, “market” prices diverge to an extreme extent from underlying value. They assume that the market is wrong all the time. The only questions are: is the market wrong enough to be useful to me, and can I come to a reasonable determination of how wrong it is? Often, markets are very wrong, but it is difficult to tell how they are wrong. For example, the market price of Google is probably grossly different than its actual economic value over the next thirty years. However, it may be hard to say just how wrong the market is, or even the direction in which it is wrong! It takes experience in investing to become an experienced investor. Most people, as they do for any topic that comes their way, look for the consensus. A market price appears to be a consensus, although that may not actually be the case at all. Maybe the reason that security A is falling in price is due to the selling of Fund X. Maybe the selling in Fund X is due to redemptions from investors, caused by Fund X’s losses in security B. It may have nothing at all to do with security A. In any case, the fact that most people consider a market price to be a consensus creates a consensus that the market price is a consensus.
Hedge funds and the like are in the mark-to-market business. It’s the game they play. If they buy WMT at $15 and it goes to $10, then they deal with that on that basis, even if WMT stock is really worth $80.
That’s why we see, for example, that it is very difficult to use lots of leverage in stocks (excepting some hedged strategies), because of the daily volatility of stock markets. Just try to buy-and-hold the S&P 500 at 3x leverage. You would be liquidated on the first 15% decline. However, private-equity firms regularly use 4x or 5x leverage, because they are subject to the underlying business, and its cashflows, rather than the silly market.And these are individual companies. The S&P 500, as a whole, due to diversification, has much more stable operating characteristics than virtually all individual companies.
However, there are more and more companies, which are really operating companies rather than investment funds, that are being treated like investment funds. This is getting dangerous. Take the strange example of Thornburg Mortgage. This is basically a mortgage bank, which makes loans to upper-income homebuyers. Their clientele is gold-plated. Here is the average borrower:
47 years old
$435,000 annual income
Thornburg has had total, cumulative credit losses of $174,000 in the last ten years — on a $30 billion book! At present, only 78 of 38,000 loans (0.2%) are 60 days delinquent. It made $286 million in 2006. Thus, it could take a loss of $286 million, and still be in reasonably good shape. If it took a total loss on all 78 of its delinquent loans (not likely as they are well collateralized), that would cost about $50m. Operationally, Thornburg looks fine.
However, Thornburg has become a victim of the “mark to market” mentality. The “market” has said that loans like Thornburg’s are worth perhaps $0.95. These “losses” made Thornburg’s balance sheet look iffy. Some of Thornburg’s lenders demanded that Thornburg sell some of its loans to reduce leverage. Thornburg was thus forced to sell $22 billion of its mortgages for about $0.95, resulting in a realized $1.1 billion loss. Instead of just sitting on the mortgages, all of which were paid in full, and maybe, perhaps, experiencing such a disaster that it recovered only $0.95 on its mortgages, Thornburg ended up taking an actual cash loss in the market! And, because of the selling of Thornburg and others like them, the market price fell further! Producing even more problems for Thornburg!
Let’s think about what would have to happen for Thornburg to actually take a 5% loss on its loan portfolio. Let’s assume that the price of the houses that serves as collateral for Thornburg falls by a whopping 50%. Remember, the LTV is 67%. So, if the house was originally appraised at $1m, then there would be about a $670,000 mortgage on it. (Which is just about in line with the actual figures.) Then, if the market value of the house fell by 50%, the house could be sold for $500,000. Remember, this is an average, not just the single worst house in the bunch. Thus, Thornburg would have a $670,000 mortgage and a $500,000 house, for a loss of $170,000. That would be about a 25% loss on a $670,000 mortgage. Now, people don’t just stop paying their mortgages just because their house is worth less than before. Certainly not people making $435,000 a year. But let’s say there is mass job loss among these borrowers. 20% of them lose their jobs and are unable to pay their mortgage. 20% — that is a huge figure. That’s 6,707 mortgages — compared to 78 presently. And, none of these 20% has any other assets that could be used to pay off the mortgage. (Remember, these people are making $435,000 a year.) It’s just a total disaster — which doesn’t really happen in real life. Some of the borrowers pay the mortgage by selling their yacht, or their vacation house, or some stocks, or their third and fourth cars, or whatever. Or the loan terms are changed so the monthly payments are less, so the borrower keeps paying and there is only a partial loss. Or maybe they get another job. But if all these terrible things happened, the loss would be about 25% * 20%, or 5%. Or $1.1 billion. And, that would be spread out over a couple years, so that the losses could be matched against income from performing debts. You can see why Thornburg has had so few realized losses over the years. It takes absolute Armageddon for them to suffer a 5% loss.
And what happened to the people who bought mortgages from Thornburg? After all, if the mortgages were basically OK, then they should have done well, right? No — they blew up too!
These were smart guys — and they bought best-quality mortgages, like those of Thornburg. They bought with only 5x leverage, which is about half that of a normal bank. However, with all the other sellers hitting the market — Thornburg-like entities — prices collapsed further in February. Now Peloton has to sell, into the collapsing market, causing — you guessed it — more problems for Thornburg.
Thornburg is now in default. It may go bankrupt and be liquidated.
Operationally, everything is fine. People are paying their mortgages on time. However, due to today’s rather extreme allegiance to the “mark to market” concept, entities are blowing up right and left.
* * *
This is wonderful. Video of Richard Nixon announcing the end of the gold standard in 1971. Be sure to listen for all the goofy rationalizations.