Banks: Hedgeless Horsemen, VAR, & Bear Stearns

Mr. Dimon Wishes He Hadn’t Bought Bear Stearns and So Do I

For many people, the point of no return in policy in this country was when JPM bought Bear Stearns in a hastily constructed deal, motivated as much by fear of the unknown as an rational business argument.

Short dated Bear Stearns bonds were trading wildly the day before that weekend at prices around 50 cents on the dollar. Many in the market were expecting Bear to fail. Some were expecting Bear to be saved. Almost no one expected Bear to be bought for $2 a share. That headline was stunning, people were tapping their screens to see if a zero was missing.

The deal had many unique features, not the least of which was that JPM would immediately and irrevocably guarantee all the Bear Stearns derivatives contracts. Some haphazard guarantee was crafted, but in the end we were told by government officials to not worry about it. That JPM would back it all and trade appropriately. No real transaction would have been designed in such a way and it highlighted that regulators and politicians were horrified of the derivative books.

In many ways this was the point of no return. I have always believed we would have been better off letting Bear Stearns fail. It wasn’t that big of a bank. It was big, but I believe manageable. In the end we won’t know whether or not I was correct, because they did bail it out.

The focus on the derivatives book at the time is interesting on two fronts.

The first is that the policy makers did nothing about derivatives after that. Within weeks the derivative markets were full on business as usual. In fact when people had threatened to pull lines to places like Lehman, the government stepped in and total companies that wasn’t in the interest of the markets. So rather than encouraging weak banks to get out of the business, good banks to tighten their standards, and demanding actions to shrink the notional amount outstanding, the policy makers actually encouraged bad practices. The immediate relief wasn’t used to fix core problems, it was used to mask it further and hope it goes away. We have seen this behavior over and over during the crisis, most recently with the story that Basel III will be delayed in Europe.

The second is that for all the focus on derivatives, the big problem when Lehman went under was much simpler. The amount of short term debt was the real problem. Investors in short term debt tend to rely on it being paid back (no offence to money market funds). These investors aren’t compensated enough to bear any loss whatsoever, so that froze money markets. The repo business took far longer to unwind and led to far more lawsuits than the derivatives business. To my knowledge there was not a single lawsuit specifically tied to Lehman’s involvement in the credit derivatives market, but the simple repo business, ignored by regulators, caused far more problems.

Lehman was let go, which surprised many, especially those that bought bonds on the Friday hoping for another Bear Stearns like announcement. Merrill was saved. Morgan Stanley managed to remain independent with some help from the Fed, the FDIC, and Treasury. AIG FP and the holding company were bailed out by the government as well. We will never know what would have happened if Bear Stearns had failed, but I remain convinced we would have been better off today had it not. Heck, we would be better off had someone done something in the 6 months leading up to Lehman to fix the core problem rather than pretending one didn’t exist.

Sowing the Seeds for the Next Crisis

For the first time since the Fed began its new programs for intervention in the markets, affectionately known as Quantitative Easing, I think the banks may actually be frustrated by it. Wells Fargo so a drastic decline in Net Interest Margin. With borrowing costs pretty much stuck at zero, they cannot benefit much from improvement in their cost of funds, without creating non FDIC insured deposits (don’t laugh, since the FDIC fees may be higher than the rate paid to actual depositors). But with their cost of funding floored, the impact of ever lower yields on the loans they make is taking its toll.

So banks will do what they can do keep profits high, lend more and lend riskier.

Hedgeless Horsemen

The banks will do what they can to protect their NIM. They will employ fewer if any hedges. They will look for riskier deals to finance (this isn’t just occurring at banks as the number of PIK and dividend deals getting done in the high yield bond market has spiked recently).

While the economy is grinding along, these riskier and under-hedged lending portfolios will do just fine, but if we don’t see stabilization in housing or the economy, then banks will be left with too much exposure once again. This isn’t an immediate problem, but it is happening already and may explain why the market reacted so negatively to WFC’s earnings.

Bigger for Less – The Problems with VAR

JPM has once again changed their VAR calculation. Regardless of the specifics at JPM we are likely to see the VAR for most portfolios reduce.

On a very basic and simplistic level, VAR is meant to give some indication of how much risk a bank (or fund) is taking. How much can a portfolio move in a given time period. There is a “confidence interval” around this. Say 99% of the time, the portfolio won’t move more than X.

The VAR model is often reliant on certain inputs, which change over time. Most models that I know of use a variety of methods to calculate the inputs, some blend of historical rates, levels that can be determined by the market etc. In general models are more focused on being conservative and not adjusting down as quickly as they may have back in 2005 to 2008.

In any case, cross asset correlation is a big input. As correlation between assets drops, it usually requires less VAR to support the same size portfolio. So if a portfolio produced $100 million of VAR under a high correlation assumption, it might produce only $75 million of VAR under a reduced correlation assumption. What often happens, especially in low return environment, is that the institution increases the size of the portfolio so that it still uses $100 million of VAR. That is the key problem, in low return environments, the portfolio is increasing, the opposite of buy low sell high.

As we move farther away from the U.S. financial crisis, we lose some periods of high correlation, and recently correlation has dropped. Whether any individual firm’s model is impacted in such a way, I don’t know, but I have reason to believe it is.

Volatility is another key input. If volatility is decreasing, then institutions can run larger portfolios for the same amount of VAR. Again, this is happening at a time of record low yields, encouraging risk taking at the wrong time.

I struggle to name a single bank that got in trouble because they bought too much CCC paper at the lows. I can name institution after institution that got in trouble loading up on “safe” paper at the tights.

I don’t like the banks right now, but that is more a question of valuation and my belief that the European Spanish bailout, when it arrives, will be a dud. It would also be just about perfect timing if some more LIBOR headlines hit the tape. I would think democrats are busy doing all they can to make some noise about LIBOR and bank punishment ahead of the elections. I am not yet worried about any more severe stress for U.S. banks, but heading back down a path of bigger, riskier banks is concerning.