Although the legislative history of the Securities Exchange Act and Securities Act never cites Graham, the validity of his theory is pretty well assumed in that legislation. That's one reason that we have all of those financial comentators speculating on how the earnings reports for the most-recent quarter resulted in changes of today's price of a given stock issue. The major problem with this theory is that it depends upon honest data. The most obvious problem with the data is purposefully deceit, as Ken Lay and John Rigas allegedly committed (well, no longer "allegedly" for Rigas). This is indeed a serious problem, and one that can be attacked by attempting to intimidate executives into providing clear, unvarnished descriptions.
However, they are not the real problem. There's a serious GAAP in the logic of fundamental analysis: the assumption that the accounting system is itself both accurate and honest. This problem is magnified in entertainment- and publishing-industry royalty statements, so what follows is far from the worst case!
- The distortions of our tax system make comparisons of companies in different industries, or even just with different product mixes in the same industry, virtually impossible to validate. Sometimes even location is enough; a "tax-free enterprise zone" may result in an improving business climate in that zone that encourages marginal businesses to take unjustified risks, or at least risks that are not consistent with a hardheaded accounting analysis. One obvious example of this in the arts is the tax-free status accorded writers residing in the Republic of Ireland. (Eligibility is a little more complex than just moving there, though.) Bluntly, the quality of work for most writers who develop a career and then move to Ireland drops off substantially and quickly, perhaps and at least in part because they don't need to work as hard to have the same or greater income.
- Even within the thousand pages of the GAAP (Generally Accepted Accounting Principles), there remains substantial discretion to call a given expenditure (or revenue) any of several things, each of which has a different effect on the final balance sheet. In other words, even without conscious effort preconceived notions of what management wants can cause accountants to handle money flows in ways that are inconsistent with "industry standards" and/or reality (not always the same thing).
- The definitions of "earnings" has little or nothing to do with reality. Sorry, guys, but amortization of a "capital expenditure" over several years means nothing if in reality the cash for that expenditure came out up front. On the back end, choices of when to "recognize" income don't help either. Then there's the question of calculating the NPV of a future lump-sum payment… I'd go on, but the problem is that the math is all internally consistent; it's only when one examines the fundamental assumptions (pun intended) behind the math that these questions become relevant.
I'm not saying that one should never pay attention to a company's balance sheet. I'm only saying that worship of the balance sheet in the investing community is the real cause of Malkiel's apparent success—because the game by its nature isn't completely honest, and understanding the bias is probably at least as important as knowing the numbers. None of this is to say that fundamental analysis is completely worthless; it is only to say that it's not nearly as important—or, based on actual results, accurate—as qualitative analysis and judgment.