The stress test results will be released to banks today. The banks will then have until Tuesday to try to persuade the government to change them.
Meanwhile, Oppenheimer analyst Chris Kotowski explains why Wells Fargo CEO Dick Kovacevich was right when he called the tests “asinine.”
The bottom line?
In this endeavour, one size just doesn’t fit all.
First, the results: Everyone’s a winner:
According to a CNBC report (4/22/2009), the banks will need to maintain a 3% TCE/RWA
ratio at the end of 2010 in order to be considered adequately capitalised. As shown in
panel 5, just about everyone passes that test… The only major bank holding that does not “pass” the test based on this cookie cutter methodology is Regions Financial, and there is clearly a major risk that this company will be among those that need to raise additional equity. However, the market had figured that
out some time ago.
Next, why the tests are asinine:
Analyses like this are useful to run through because they can stimulate ways of thinking
about the differences between companies, but investors should not pay too much
attention to the results, for a number of reasons:
First, not all loans in a given asset category are created equal. For example, a working
capital line at a highly capitalised defence contractor that is well secured by inventory and
receivables is classified as a “commercial loan” in just the same manner as the leveraged
buyout loan of a struggling retailer. The loss potentials of the two loans are vastly
different. Another good illustration of this are the differing loss characteristics of consumer
loans made to deposit relationship customers versus those made without such a
relationship. At a recent analyst day, JP Morgan disclosed that its home equity loan
losses were 0.9% on loans made to deposit customers but 4.2% on customers without a
deposit relationship. This methodology treats them equally.
Second, the analysis penalizes banks that underwrite well. US Bancorp, for example, has
consistently shown among the best credit trends among the major bank holding
companies. However, the stress test methodology automatically bakes in the assumption
that their losses will revert to the mean of lesser underwriters. Thus, their capital ratios at
the end of 2010 look worse than average, even though in our assessment they are likely
to emerge from this current crisis as one of the stronger institutions because their asset
quality has in fact been a lot better than average.
Apparently in response to some of these complicating factors, Bloomberg now reports that
“Regulators conducting the stress tests are increasingly focusing on the quality of loans
banks made after finding wide variations in underwriting standards, a regulatory official
said earlier this week.” (U.S. Weighs Revealing Bank Capital Needs After Tests,
Bloomberg.com, 4/23/09.) This revolutionary insight seems to just further underscore Mr.
Kovacevich’s characterization of the whole testing process and the way that it has been
presented to the public. It seems that a good old fashioned bank examination with a
rigorous deconstruction of underwriting processes, testing of assumptions, and
examination of the actual quality of actual assets might in fact produce better results than
simply processing a bunch of generic inputs in a macro blender called the “stress test.”
And, finally, one of the biggest problems–which is that the loan losses aren’t the biggest factor in whether a given bank will survive. The biggest factor is revenue.
Finally, the test is flawed because the single biggest unknown is not so much the loan
losses (there is a recession going on, and thus it is obvious that loan losses will rise on
some curve), it is the pre-provision earnings. While so far, pre-provision earnings have
held up remarkably well, we see both significant risks and some opportunities on this front.
The chief risks are: (1) Shrinking loan volumes may pressure net interest income. (2) The
low interest rate environment may pressure net interest income. (3) Rising NPAs may
pressure net interest income. (4) For both market and political reasons, banks may lose
the capacity to charge fees aggressively. (5) Customers may act more defensively to
avoid overdraft and late payment fees. (6) At banks with large investment banking
operations, the fear of compensation restrictions may cause a flight of talent and their
Conversely, there are opportunities to increase pre-provision earnings by: (1) Cutting
expenses. (2) Earning wider spreads on new extensions of credit. (3) Merging or acquiring
businesses at distressed valuations and realising merger synergies. It is our belief that
the best indicator of banks’ ability to survive and thrive without mass dilution in the next
two yeas will be those that have effectively focused on customers to maintain and grow
their base of pre-provision earnings.