- Focus on long-term performance. In no event should you pick an adviser on the basis of less than five years of past performance.
- It is much more preferable to focus on returns over 10 or more years.
- It is better to focus on risk-adjusted returns as opposed to raw performance.
The Price-Value Relationship: Do You Get What You Pay For?
by Mark Hulbert
If I told you that this column was about the relationship between price and value, chances are youd think it is going to be yet another discussion about the power of various stock valuation ratios such as price-to-earnings and price-to-book.
But its not. Instead, my column is about value and price among investment newsletters.
Specifically: Is a higher-priced newsletter really worth the extra money?
I was prompted to ask this question by the Hulbert Financial Digest (HFD) having compiled a years track record for a new newsletter edited by Tom and David Gardner, the founders of the Motley Fool financial Web site (www.fool.com).
Unlike the model portfolios that, up until February of this year, were provided at their Web site, the Motley Fool Stock Advisor (the name of the Gardners new newsletter) is not provided free of charge.
Over the 11 months for which the HFD has track records for both services, the Gardners for-pay newsletter did better. Between March 31, 2002, and February 28, 2003, according to the HFD, the newsletter produced a 3.4% return, while the several portfolios recommended at the Motley Fool Web site produced an average loss of 22.1%.
Cost vs. Performance
Is this a fluke? Or is it indicative of a larger pattern in which more expensive newsletters provide better returns?
These are the questions I set out to answer.
I turned to the HFDs database containing investment newsletter track records. I retrieved their returns over both the five- and 10-year periods through March 31 of this year, calculated on both an unadjusted and a risk-adjusted basis. If a newsletter recommended more than one portfolio during either of these periods, I calculated a composite that reflected an average of its several portfolios (including those that may have been discontinued along the way).
For each of the 104 newsletters for which I obtained five-year returns, and the 77 newsletters for which I had 10-year returns, I then obtained their current advertised subscription rates for a yearly subscription. Inevitably in such cases, there was a lot of noise in the data. To reduce the role played by that noise, I conducted several analyses on the data to see if there were any statistically significant relationships between the price of a newsletter and its performance.
Much to my surprise, there was a relationship between price and performance. But the relationship was the reverse of what occurred in the case of the Motley Fool: There is an inverse correlation between a newsletters current price and its trailing five- and 10-year returns.
The results were mixed, however. The regression analyses found a statistically significant relationship only when relating subscription price to raw performance (that is, performance that is unadjusted for risk). When regressing price and risk-adjusted performance, in contrast, no detectible correlation emerged.
What might explain these mixed findings?
My guess is that it has a lot to do with the fact that I was correlating newsletters past performances with their current subscription rates. This is a crucial point, because newsletters do not keep their subscription rates constant.
This especially is the case for newsletters that have a year or two of spectacularly good performance. Their editors often use such occasions to jack up their prices. To the extent luck played a role in those spectacular returns, however, it is quite likely that these newsletters performances will subsequently regress to the average.
In my experience, furthermore, newsletters subscription rates are less likely to come down than they are to go up. This means that a newsletters current high subscription rate may be a relic of a particularly good burst of performance a number of years ago.
If my hypotheses are correct, then the group of most-expensive newsletters will include a disproportionate number of high-risk services. While such services rarely produce good long-term returns, this will especially have been the case recently, given that we are in the fourth year of a bear market.
My hypotheses in this regard are reinforced by the analyses involving newsletters risk-adjusted returns. By design, luck plays less of a role in a newsletters risk-adjusted returns than it does in its raw returns. A newsletters risk-adjusted returns will be less volatile than its unadjusted returns. Furthermore, because potential subscribers have a more difficult time wrapping their minds around the concept of risk adjustment, newsletter editors are less likely to jack up their subscription rates because of an increase in their risk-adjusted returns.
Because of these factors, it makes sense that subscription price has a different statistical relationship to risk-adjusted performance than it does to raw returns.
My hunch, therefore, is that the relationship between price and raw returns would be weaker, if not disappear altogether, if my regressions had focused on newsletters subscription rates five and 10 years ago. Unfortunately, because the HFD database has not captured that data, verifying that hunch will have to await further study.
Even if the inverse correlation between price and raw returns persists after additional study, however, my advice to you probably would remain the same: Dont use subscription price as a basis for picking a newsletter.
This may seem obvious, but youd be surprised how many investors dont heed the advice. Direct mail marketers tell me that, not infrequently in head-to-head tests of two advertisements that differ only in price, the higher-priced one receives more responses. This is the same result that liquor stores find when raising the price of Scotch.
To pick an investment newsletter, or any adviser for that matter, heres my advice:
In the final analysis, of course, a successful newsletter is worth a hefty subscription price, while an unsuccessful one is worth nothing.
This column appears quarterly and is copyrighted by HFD and AAII.
Mark Hulbert is editor of the Hulbert Financial Digest, a newsletter that ranks the performance of investment advisory newsletters. It is published monthly and is located at 5051B Backlick Rd., Annandale, Va. 22003; 703/750-9060; www.hulbertdigest.com.
This column appears quarterly and is copyrighted by HFD and AAII.