The Fed’s Bank Stress Tests: What Do the Results Mean for Companies Investing Their Balance Sheet Cash?

First, a bit about the tests (officially known as the Federal Reserve’s Supervisory Capital Assessment Program) and their results:

There are a few facts that give might be cause for some concern about the reliability of these tests.

1.      The test assumed U.S. unemployment would not reach ten percent until the end of 2010, twenty months away. Unemployment currently stands at 8.9% and has been rising about a quarter of a percentage point a month since the start of the recession in December, 2007. Over the last six months, the average monthly increase in unemployment has been just under four-tenths of a percent. If that pace persists, in about three or four months we will be at ten percent unemployment. To reach ten percent unemployment in twenty months, monthly job losses would have to average just over 100,000, rather dramatically reversing the recent trend of about 600,000 jobs lost a month. That is a very rosy outlook for the pace of increases in unemployment, one that most would consider implausible, at least in the near term.

2.      The stress tests made assumptions about bank profits going forward. Projected profits were calculated by essentially annualizing Q1 2009 profits. Positive first quarter U.S. bank profits surprised many, except for those that foresaw a jump as a result of modifications to (manipulations of?) mark-to-market accounting of bank assets. Further deterioration in the U.S. economy will directly diminish the valuation of bank-held investments and banks will need to raise meaningfully more capital than the large amounts the government has today indicated.

3.      The stress tests were conducted over a ten week period and the Federal Reserve Bank described the program this way: “The analysis was a comprehensive one, which included an exhaustive review of loan portfolios, investment securities, trading positions, and off-balance sheet commitments.” Exhaustive implies that not a single loan, security, trading position, or commitment went untested. Not only is this very difficult to imagine, it can be suggested without much debate that this would be an impossibility.

But the program did involve the contributions of the Office of the Comptroller of the Currency (OCC.) The OCC’s job normally comprises the following:

“Examiners analyze a bank's loan and investment portfolios, funds management, capital, earnings, liquidity, sensitivity to market risk, and compliance with consumer banking laws, including the Community Reinvestment Act. They review the bank's internal controls, internal and external audit, and compliance with law. They also evaluate bank management's ability to identify and control risk.”

That sounds fairly exhaustive, too. If the speed of the stress test can be explained by the fact that the OCC already had the data, as it certainly should have, then why the need for the tests at all?

The last concern aside, if lowered estimates of losses on investment assets turn out to have been a bit too optimistic and profitability weakens as a result of asset valuation re-adjustments and/or just plain weak bank business, the stress test model will be seen as having been rather materially flawed (I think it can already be seen as such.)

Now, what do the test results mean for companies investing balance sheet cash? Given that the stress tests may not have actually applied sufficient stress or that banks may come up short in efforts to raise additional capital, there could well be further downgrades to bank debt and other bank obligations. Corporations investing cash will need to be excruciatingly diligent in minding their exposure to bank-related investments.

How can they do this? U.S. corporations investing balance sheet cash expose themselves to bank-related credit risk, directly or indirectly, when they invest in any of the following:

1.      Bank deposits

2.      Bank debt obligations

3.      Sweep accounts that sweep to the bank’s balance sheet

4.      Asset backed commercial paper (ABCP) that relies on bank liquidity facilities (all ABCP programs do)

5.      Variable Rate Demand Obligations (VRDOs) that rely on bank LOC’s or stand-by purchase agreements (only a tiny fraction don’t)

6.      CDARS programs

7.      Negotiable certificates of deposit

8.      Bankers acceptances

9.      Any money market mutual fund that invests in any of the above (prime money funds can invest in all of these and many are also eligible for purchase now by government, treasury, and municipal series of money funds.)

A far higher proportion of corporations make use of money market funds now than even two years ago as they have been forced to abandon other traditional, direct investment choices like commercial paper and auction rate securities. These companies can unwittingly face inappropriately high exposure to bank credits.

Within a typical prime money market fund’s underlying portfolio, one will find a number of the variety of money market instruments described above. Fund portfolios are governed by certain S.E.C. diversification requirements (under Rule 2a-7 of the Investment Company Act of 1940), the most basic of which specifies that a fund may not hold the obligations of any single issuer in excess of five percent of the portfolio’s total value. But 2a-7 does not specify concentration limits for bank support, like supporting Letters of Credit, liquidity facilities for ABCP, or other indirect bank enhancements. It is very common to find fund portfolios that, when all such cross-over obligations are added up, have exposures to single banks as high as ten or twenty percent. Such concentration is not only imprudent in good economic times, but it is downright foolish in times such as these in which the need to stress test banks at all is seen as a necessity.

Now that the Treasury is temporarily backing short term bank debt through FDIC insurance and the Federal Home Loan Bank and Fannie Mae are supporting some VRDO’s, many government and even treasury funds have seen fit to significantly grow their holdings of these. While no one seems to be at all concerned about investing in the FDIC, it should be remembered that before you need to get a check from the FDIC, you must first endure a default. Should the rate at which U.S. banks fail increase and the line for FDIC claims grows longer, the government may need to slow the process down so as to be able to sort things out (this happened in the nineties) and buy time in which they can raise additional Treasury funds to replenish the FDIC’s fire bucket (a process already underway.)

In 2007, three U.S. banks failed. Last year (2008), twenty-five U.S. banks failed or about two a month. In the first seventeen weeks of 2009, thirty-three banks have failed, about two banks every week. Annualized at that rate, we can expect one hundred banks to fail this year. (For historical perspective, in 1988 and 1989, over one thousand banks failed, or two every business day. From 1980 to 1995, 2,368 U.S. banks failed, or three banks a week.) In 1989, the Federal Savings and Loan Insurance Corporation (FSLIC) ran out of money after repeated taxpayer-funded replenishments and was dissolved and replaced by the Resolution Trust Corporation. Can anyone tell us with certainty if this scenario is likely to play out this time around? No, but the more important question is “As a corporate cash investor, do I even want to have to worry about this?”

The greatest threat from investing in banks is probably not principal loss. Even if FDIC pay-outs take longer than investors would like, they will eventually get their money back. The likelier threat that corporations investing in banks face will be the diminution of the value of investments should banks suffer rating downgrades or other market-price-unfriendly events. In such circumstances, corporate balance sheets will weaken or, if unrealized loss conditions can be seen as “other than temporary”, earnings will take a hit, too.