How High Leverage Has Brought Down the Whole Banking Industry

by: Thomas Tan, CFA

On Tuesday, as the S&P 500 briefly touched 1,200, the banking sector represented by KBW Bank Index [$BKX] (or other similar indices) went down to 47 (which had been range bound and traded around 75-90 early this year), and VIX reached 30, it seemed that the stock market was under capitulation similar to the March plummet. The dropping of $BKX  was so severe that it broke my technical target of 55 by 8 points.

Now we may see a bear market rally lasting for a few months, similar to post-March capitulation. I still feel Jeremy Grantham's target of 1,100 will be reached sooner than 2010, probably later this year or early next year. I also doubt that funds which have withdrawn money from the market this year, especially in last several weeks, will re-enter the market anytime soon. The worst case scenario is that they may never re-enter the market at all, if these are funds for the baby boomers.

There is an interesting book "Bringing Down the House", which was turned into a movie called "21". It is about a group of MIT students who were trained as card counters to play blackjack at casinos in Las Vegas. They acted as a group, and played small when the odds were not clear or not in their favor. But after receiving a signal from his companion, one of them acting as a super-rich guy entered into the table and played huge stakes when the odds were in their favor. This is a typical example of using leverage against casinos.

It turns out to be that this highly leveraged technique is also used by banks, especially investment banks. No one is trying to bring them down as in the case of "Bringing Down the House". All the current troubles for this industry are of their own making. They are the ones who choose to gamble with their own capital, unlike the casinos. They can't blame anyone else but themselves. However, now with the help of their friends in the government, they are arguing that they need unlimited funding protection and bailout from the government, or more accurately, U.S. taxpayers.

To understand leverage, just look at this WSJ article from Wednesday. Lehman's (LEH) market cap of $9B is only 40% of their book value of $23B, and it sounds very cheap. But then look at their assets: They have $160B hard-to-value Level 2 assets and $41B impossible-to-value Level 3 assets. The WSJ article applies a 5% haircut on Level 2 and 25% on Level 3 to come up with  $19B future write-offs. However, based on analysis from many other public sources, most of the Level 3 assets are MBS CDOs, even if they are AAA rated, the recovery rate is only about 50%. For Level 2 assets, 10% haircut is actually a conservative estimate. The combination of both more realistic haircuts will result in $36B additional losses, which would more than wipe out their book value of $23B plus their market cap of $9B. This is leverage in the working, unfortunately at the down side.

Let us also look at Fannie Mae (FNM) and see what the level of leverage they are using. FNM has long-term liabilities of about $580B, according to Yahoo Finance. It has a negative duration mismatch of 14 months between its assets and its liabilities. Due to this mis-match, a 1% drop in interest rate will cause roughly 14/12 or 1.17% loss in value, or $7B (1.17% x $580B). And their market cap is only $10B. We are only talking about pure interest rate risk, not even losses of credit risk from lending practices, delinquency, foreclosure, etc., which will be much larger than the interest rate risk.

People have drawn parallels between the current failure of IndyMac and the failure of Continental Illinois Bank in 1984, with the expectation that IndyMac will cost FDIC about $4B to $8B, when FDIC has only $52B in its insurance funds. But this comparison has missed the whole S&L crisis, in which losses were much larger than one commercial bank, and FDIC was actually not quite involved. For S&L, the Federal Savings & Loan Insurance Corporation (FSLIC) was the show, and it had $5.6B in 1984 to pay claims; by 1989 its balance had turned into an $87B deficit. The total number of failed S&L institutions is estimated to be around 1,000, and GAO (US General Accounting Office) has estimated the total losses for S&L to be at $166B for taxpayers.

Today people are trying to estimate how many banks will fail this time. The number probably won't get to the 1,000 mark as in the S&L case, and the figure of a few hundred banks has often been mentioned. At the end of this crisis, FDIC will be most likely in the red, similar to FSLIC.

A week ago, Bridgewater Associates issued a report saying that the banking system losses will likely hit $1.6 trillion, but didn't give any breakdowns. This is more than the $1 trillion estimate in my previous article Will CDS Replace Subprime To Cause $1 Trillion Total Loss For This Credit Crisis? in January this year. I actually tried to give a breakdown at that time as follows: $500B for over the counter credit default swaps (CDS), $250B for subprime, and $250B for everything else such as commercial real estate, leveraged loans, credit card losses, auto loans, etc. Due to the further deterioration of the real estate market and continued losses, I think I underestimated the subprime by about $150B, also I should have included some losses for Alt-A and prime mortgages since the losses have already cut into them, especially Alt-A products. In addition, CDS has become a larger, deeper and wilder threat to the whole banking system day by day, maybe $1 trillion is a better estimate now. Overall, $2 trillion losses for this credit crisis are really not stretching at all.

So far this year, some financial institutions have touched the single-digit territory, such as Fannie Mae, Freddie Mac (FRE), Wachovia (NASDAQ:WB) and Bear Stearns. After the current bear market rally for the banking sector, who will be the next round of candidates to join the single-digit club? Investment banks are still the usual suspects, such as Lehman, UBS (NYSE:UBS), and possibly even Merrill Lynch (MER) and Citigroup (NYSE:C) too.

In order to avoid the domino effect of the current credit crisis rippling through the whole banking industry, the Fed is currently holding the bag by bailing out everyone in trouble. And so far we are only talking about residential mortgages and their CDO derivatives. If the next wave of the credit default swap crisis hits, it will be much more complicated than LTCM, much worse than S&L, and much deeper than subprime. With raising capital becoming more difficult these days, banks have to rely more and more on the Fed with balance sheet of only $800B.

Can the Fed handle the worst monetary crisis in 60 years? Should taxpayers save and bail out the financial institutions in trouble? Did we ask them to use high leverage initially? Have they ever shared the profit with the public at the time when they were making tons of money by using leverage? Why should the whole society have to pay for the bad decision of a few banks at the down side, but  never be allowed to participate at the up side?