** Why One Should Sell Bank Stocks **
September 15, 1999 - 5:48pm
Credit Suisse First Boston Corporation
CREDIT SUISSE FIRST BOSTON CORPORATION
Equity Research
Americas
U.S./Financial Services/Banking
Ten Reasons to Sell Bank Stocks
Summary
Ten reasons to sell bank stocks include:
Efficiency
Capital Markets
Asset Quality
Y2K
Consolidation
Less Earnings Certainty
Less Catalysts
Valuation
Housecleaning
Capitulation
Ten Reasons to Sell Banks
Efficiency - In the first quarter, First Union announced an earnings
shortfall because it needed to develop an enhanced banking business model. In
the second quarter, State Street indicated that expenses would likely exceed
revenue growth because it needed to continue spending on its processing
business. In the third quarter, Bank One indicated that it would cost more to
make its credit card business more competitive. These are not isolated events
but are indications that banks need to spend more merely to maintain existing
market share, consistent with the theme of our March 1999 report Banks and
the Red Queen Effect. The industry stopped setting new efficiency records as
of midyear 1998. Much of the low hanging fruit has been picked. For instance,
it is easier to downsize inefficient businesses than it is to generate
revenues at lower marginal cost. Also, Y2K has pushed off productivity
initiatives. New investments can eventually help efficiency get back on track
but involve uncertain back-ended benefits in exchange for tangible up-front
costs.
Capital Markets - Managements of several multinational banks (Citigroup, Chase,
J.P. Morgan, Nat West and others) recently indicated that capital markets
business will likely slow in the quarters around year-end because of Y2K
factors. Indeed, Citigroup specifically indicated that analysts should not
raise estimates given an uncertain fourth quarter, but this did not stop
several analysts from raising their estimates anyway following second quarter
earnings-i.e., don't blame the company if it falls short. Banks need to
reduce risk profiles for Y2K. Since the date of the event is known (i.e., in
contrast to the emerging markets crises of 1998), there are few excuses for
losses. A below-normal level of activity, which we expect for the two
quarters around year-end 1999, would be in stark contrast to the above-normal
levels of the first half 1999. A slowdown can hurt banks, especially since
banks are so leveraged to the capital markets. Banks own half of the ten
largest brokerage firms. Fees, as a percentage of revenues, have about
doubled in the past 20 years (close to 40%), but perhaps half of this
increase is due to market-sensitive items, such as brokerage, trading,
venture capital, and mutual funds, thereby leaving banks vulnerable to a stock
market or other capital markets disruption.
Asset Quality - The level of domestic business problem loans, while still in
a good range, showed the largest increase in the first half of 1999 than at
any time since 1991. New bankruptcies, such as Iridium and Planet Hollywood,
have the potential to hurt banks' portfolios, too. Given the 4%-plus GDP
growth, the increase in problem loans raises an issue if the economy slows
even modestly. While consumer credit quality is improving in some areas
(credit cards), a stock market correction could have a negative impact.
Foreign loan quality has also improved, but there are still areas that are
vulnerable, such as China (devaluation is not out of the question), Ecuador
(defaulted on Brady Bonds), Indonesia (political upheaval), and elsewhere.
Instances of fraud are also more rampant, partly reflecting a time of high
leverage and market participants' ability to make so much money so quickly.
Also, securities losses are possible, given substantial unrealized securities
losses at several banks.
Y2K - Most of the 15 largest U.S. banks (80%) have increased estimates for
Y2K spending over the past year, including Citigroup (up $50 million) and
Chase (up $31 million) as recently as August. In addition to higher-than-
budgeted expenses, another threat is the potential for loan losses if
borrowers have Y2K problems. The large German bank, Commerzbank, set aside
specific Y2K loan loss provisions. Another issue is third-party risk, which
affects banks relatively more than other industries given their multitude of
connections. On a recent internal CSFB conference call, 20 bank analysts
worldwide were unanimous in the belief that Y2K third-party risk poses a
threat to earnings. A bank could suffer longer-term reputational harm if it
fails to handle any problems as well as its peers do. At a minimum, Y2K poses
an additional risk without a commensurate return.
Consolidation - Perhaps since many of the most willing sellers have sold,
banks seem to be stretching terms of agreements to get mergers done, such as
offering dual management posts, taking abnormally large restructuring charges,
and at times catering more to social issues than shareholder value. The
elimination of Glass-Steagall would encourage more mergers among banks,
brokers, and insurance companies, but as reflected by NatWest's recent
purchase of Legal General (the stock declined over 10% after the announcement),
the result can be a substantial decline in the stock price of the acquirer,
which may well be a bank. Also, the potential elimination of Glass-Steagall
at the same time as the elimination of pooling-of-interests accounting
creates the potential for a merger boom (deals done soon to get pooling
treatment), but it could lead to a merger bust given a rush to combine. Also,
mergers make less sense for the period around Y2K. System conversions get
delayed. The mere announcement can exacerbate third-party Y2K risk if, as often
happens, employees leave.
Less Earnings Certainty - While we have a downward bias to estimates, we do
not know the timing of the decline. Sometimes it seems at the discretion of
the companies. First Union's estimates declined by one-fifth following its
guidance on two separate occasions. Indeed, several managements acknowledge
that earnings are as good as they are going to get. The problem is that when
earnings improvement starts to decelerate, the pace of this deceleration can
gain speed-i.e., the first downward revision is often not the last. Banks
have been showing more one-time items in results. Indeed, both First Union
and Bank One very clearly had lower earnings quality after backing out one-
time items, such as above-normal venture capital, securitization, securities
gains, and other items.
Less Catalysts - One sign of this sentiment is National City. In September
1999, the stock hit new 52-week lows even though its progress was generally
on track. When investors are buying stocks, the group tends to move more
together. This is not the case when investors are selling, since only those
names owned by a particular seller decline. Generally, when investors look to
sell stocks, there do not seem to be many investors looking to step up to the
plate. In any event, even if the stocks were inexpensive, there are still a
lack of catalysts. One reason that it was relatively easier to be bullish
during 1994-98 was that there were several answers to the question, "Why
should we own these stocks now?" These included record consolidation
stemming from new 1994 national banking legislation, dramatic restructurings
as banks looked to redirect de facto subsidies from employees and customers
to shareholders, and large, new buybacks. When stocks declined, investors
could rely on new consolidation, buybacks, and restructurings to help. In a
way, the entire bank story has been one of better optimizing franchises.
These factors are no longer as strong. In comparison, Y2K presents a very
tangible negative catalyst.
Valuation - Bank P/Es are expensive relative to the S&P 500 after backing out
new economy companies such as Dell, Microsoft, and America Online from the S&
P 500. The relative P/E of banks (estimated 2000 P/E) increases from 68% to 74%
after backing out 20 new economy companies, above a three-decade average of
68%. Even if P/Es are inexpensive for certain individual banks, we have less
confidence in the "E" component than in times past. Some investors want to
buy banks given their underperformance since midyear 1998. Yet for those
investors who previously owned banks for structural improvement (i.e., since
1995 or earlier), it is still time to take profits. Bank stocks have
increased almost 3.5 times since the start of 1995, significantly
outperforming the S&P 500 (about 50 percentage points). Also, the so-called
"smart money" has sold financial stocks, including the co-CEOs of Citigroup,
Warren Buffett, partners of brokerage firms, and others.
Housecleaning - In our view, the tougher decision is not whether bank stocks
should be reduced but at what level investors should get interested again. We
generally think that the group can decline by one-third. However, even after
the declines that have occurred at certain companies (like Bank One) since
our May report, we are not interested, given our expectations for a final
housecleaning of earnings. This could involve taking one-time charges for new
systems, resolution of lingering asset quality problems, and the use of Y2K
as a scapegoat to throw in the kitchen sink later this year. Our best guess
for the timing of such a move would be concurrent with the reporting of
fourth quarter 1999 earnings in January 2000, though it could drag out another
quarter or so.
CREDIT SUISSE FIRST BOSTON CORPORATION
Equity Research
Americas
U.S./Financial Services/Banking
Ten Reasons to Sell Bank Stocks
Capitulation - Another reason not to jump back into bank stocks too soon is
because of expectations for a capitulation-if earnings are weaker than
expected for at least the two quarters around year-end. In particular,
ratings on the top 25 banks are still roughly two-thirds "buy" or "strong buy"
(per First Call). The reaction to Bank One's August announcement is a
reminder of the safety in numbers mentality that surrounds analysts' stock
ratings and, therefore, to some degree, stock ownership. Following Bank One's
announcement about credit card weakness, 12 analysts downgraded the stock
(despite many recommending it for the Internet strategy) and the stock opened
down more than 20%. It is too late to react when problems become too obvious.
One analyst said to a well-read newspaper about Bank One that "if the company
could not see it coming, then how could we?" This type of attitude (in which
the analyst disavows direct responsibility) is probably only a dress
rehearsal for reactions by analysts if and when additional problems occur.
Overall, if our analysis plays out as expected over the next six to nine
months, we would look for the housecleaning and capitulation stages to pass
before considering whether banks are again an attractive investment.
N.B.: CREDIT SUISSE FIRST BOSTON CORPORATION may have, within the last
three years, served as a manager or co-manager of a public offering of
securities for or makes a primary market in issues of any or all of the
companies mentioned.
America Online
Bank One
Chase Manhattan
Citigroup
Commerz Bank
Dell Computer
First Union
Iridium
J.P. Morgan
Legal General
Microsoft
National City
National Westminster
Planet Hollywood
State Street
|
