The shares of the top five banks could be in for a tough time over the next three to six months. According to a report in the Financial Times last week, banks with a Wall Street presence are preparing for a decline in trading profits during the second half of the year. The third quarter has gotten off to a very slow start, with trading activity across asset classes at its lowest level of the year. Add financial reform to the mix, and you have all the ingredients for earnings disappointments. In that context, which of the banks are most vulnerable?

Investors would rather stand pat right now
Reasons behind the slowdown include May's "flash crash," concerns about the health of the global economy and uncertainty linked to the stress tests of European banks. I expect all of these factors to continue to weigh on the market throughout the current quarter. Admittedly, there has been some pickup in volumes since the release of the stress test results at the end of last month; however, given that many investors felt the tests were unrealistic, it's safe to assume that all doubts concerning the strength of the European financial system have not been erased.

The valuations
In order to analyze how this trading slump could the shares of the top five banks, let's take a look at how their valuations stack up:


P/E Multiple* Consensus Estimate, Next 12 Months' EPS

P/E Multiple* Lowest Estimate, Next 12 Months' EPS

Bank of America (NYSE: BAC)



Citigroup (NYSE: C)



Goldman Sachs (NYSE: GS)



JPMorgan Chase (NYSE: JPM)



Morgan Stanley (NYSE: MS)



Source: Author's calculations based on data from Capital IQ, a division of Standard & Poor's.
*Based on closing prices on Aug. 4, 2010.

Cheap or expensive?
This table clearly demonstrates that some of the banks -- I'm looking at you, B of A and Citi -- look reasonably cheap on the basis of the consensus estimate but could, in fact, appear very expensive if earnings disappoint. You might object that it is very unlikely that any of these banks will achieve only the lowest analyst estimate over the next 12 months, and I would agree, but that's hardly necessary for the market to penalize a stock with a sharp downward revaluation. Investors typically overweight the most recent information; extrapolating a single bad quarter into the future is enough to prompt them to sell the stock mercilessly. A single quarter at or below the lowest forecast is well within the realm of possibility.

Throw the historical benchmarks out
Taking a longer term view, none of the banks look particularly expensive: Based on estimates for 2011, all but Citigroup are valued at less than 9 times' earnings (even Citigroup's multiple is below 10). However, one must be a bit wary here; indeed, what is an appropriate multiple for these institutions? In a world of higher capital requirements, heavier regulation, and lower leverage, historical multiples lose their relevance as benchmarks; the new environment dictates structurally lower multiples.

Attractive from two angles: Morgan Stanley & JPMorgan Chase
All told, I think the bank shares that are most likely to perform well over the long term while experiencing less volatility in the short term are those of Morgan Stanley and JPMorgan Chase. Meanwhile, although Bank of America and Citigroup could also do well over a longer period, the stocks look most vulnerable to earnings disappointments in the near term -- just look at the spread between the "consensus" and "lowest estimate" multiples!

The Goldman conundrum
Goldman Sachs, true to form, is in a category of its own with regard to this discussion. On the one hand, as with Morgan Stanley and JPMorgan, the gap between the consensus and the lowest earnings estimate is relatively narrow, which suggests the price risk linked to potential earnings volatility is limited. However, the bank has been unusually profitable over the past 18 months, with net income margins exceeding even the inflated figures achieved in the boom years of 2006 and 2007. As competitors close the distance that Goldman opened up on them during the financial crisis, it's no sure thing that Goldman will be able to maintain recent levels of profitability and earnings growth.

Furthermore, Goldman is the least diversified of among the top five banks, with no commercial banking business, nor even a retail brokerage. As such, it has the highest earnings exposure to financial reform.

The cost of reform
A prominent example of this concerns proprietary trading, which is banned under the Volcker rule. Goldman is the most heavily reliant of all five banks on this business, which, along with private equity investments, represent approximately 10% of the firm's profits (compared to just 2% at Morgan Stanley). Goldman is currently examining two options to comply with new regulations: Spin out its proprietary trading desk as hedge fund, or transfer the business to its asset management arm. Either way, we should expect this activity's contribution to firm profits to shrink.

An alternative strategy: Own the whole segment
I've singled out Morgan Stanley and JPMorgan Chase as having reasonably attractive long-term prospects and less potential for price volatility over the next few quarters. However, perhaps trying to pick one or two among these five banks is wasted effort. An alternative strategy consists of owning the entire group. One super-investor is doing just that (or close enough): Bruce Berkowitz owns four of the top five banks (JPMorgan Chase being the exception) in his Fairholme Fund, and all four are among his top 10 positions. Furthermore, these are mostly new positions: Apart from Citi, they were initiated in 2010.   

At the height of the crisis, Warren Buffett invested $5 billion in Goldman preferred shares. By virtue of Berkshire's size, he is now restricted large- and megacap stocks, but he wishes he could buy these stocks instead.

Fool contributor Alex Dumortier has no beneficial interest in any of the stocks mentioned in this article. Try any of our Foolish newsletters today, free for 30 days. The Motley Fool has a disclosure policy.