I certainly don't think the choice of 1926 was blatant cheating. And as will be seen the first presentation of the data gave results for 22 selected pairs of endpoints
. The "selected endpoints" is a bit weird, but to me the actual choices do have the flavor of "good-faith effort," not "cherry-picking."
But I still think there are questions to be asked. Despite the integrity of researchers and the peer review process, there is
such a thing as funding bias. The CRSP's own description of their history
talks about Engel telephoning Lorie, but omits the salient details that Engel wanted (and ultimately used) the information for an ad--and that the SEC had played a role in the story.
"Rates of Return on Investments in Common Stocks" L. Fisher and J. H. Lorie, The Journal of Business, Vol. 37, No. 1 (Jan., 1964), pp. 1-21, says that "This work is the first to emerge from the Center for Research in Security Prices (sponsored by Merrill Lynch, Pierce, Fenner & Smith Inc.)" and opens with a description of CRSP that states
The sole purpose of the Centerfor Research in Security Prices is to conduct research and to disseminate the results throughout the academic and financial communities.This is what Merrill Lynch, Pierce, Fenner & Smith Inc. had in mind when they provided the funds to establish the Center, and this is what the Graduate School of Business had in mind when it sought support.
Methinks the lady doth protest too much; the prompt dissemination of those results to retail investors via the medium of a Merrill Lynch ad is not quite consistent with the CRSP's "sole purpose." I haven't found a high-res image of the newspaper ad, but if you squint at table 1 from the paper below, and the ad above, it sure looks as if the 1965 Merrill Lynch ad reproduced this and two other tables from the paper. That is, there was a direct connection between paper and ad.
Now, as to 1926, The paper says "Monthly closing prices from the New York Stock Exchange from January, 1926 through December, 1960 have been placed on tape." That's really all there is. As nearly as I can tell, the paper does not give any explanation at all
of the starting date.
The paper--yes, a genuine research breakthrough--presented data from equally-weighted investments because "a policy of allocating funds in proportion to shares outstanding or according to any other criterion implies less neutrality of judgement in making investments." A footnote says that they tried cap-weighting and some results were higher, some lower, and there was no systematic difference and the yields were "generally similar." OK.
To their great credit, they
- included reinvested dividends
- included expenses (showing results with and without expenses)
- included taxes, three different brackets.
Inflation and real returns were not discussed, which seems like an oversight given that the postwar inflation, one of the highest in U.S. history, ought to have been a recent memory.
The results were presented according to a set of chosen pairs of endpoints
As to the endpoints, their explanation is "The periods were chosen for obvious reasons." (!) In full:
The periods were chosen for obvious reasons. The period from 1926 to 1960 is a long span with booms and depressions--prime examples of each!--and war and peace. The periods beginning in September, 1929 were included to indicate the experience of those who invested at the height of the stock-market boom of the 1920s. The periods beginning in June, 1932 were included to show the results of investing at the nadir of this country's worst depression. The numerous brief, recent periods were included to bring details of postwar experience into sharp focus. Aside from most periods ending in 1932 or 1940, the rates of return are surprisingly high...
Annual income twenty pounds, annual expenditure nineteen nineteen and six, result happiness; Annual income twenty pounds, annual expenditure twenty pounds ought and six, result misery.