Considerable attention has been focused on the role of Wall Street's "sell-side" analysts in generating so much of the hot air that inflated the late market bubble. In the desire to attract new investment banking clients and retain old ones, so-called security analysts put forth glowing recommendations that were long on concepts and potentials, but short on facts and figures. Of 6,000 recommendations made by Wall Street firms during 1999, according to one study, only eight recommended "sell."
While an obvious conflict of interest exists when an analyst is rewarded for his help (or punished for hindrance) to the investment banking department, that conflict has existed for a long time, with no obvious alternative. But even if we eliminate the conflict by a complete separation of the financial analysis of investment banking deals from analysis for brokerage clients, the challenge of providing truly objective research would remain. Why? Because Wall Street is a selling machine -- a giant and complex marketing system for securities distribution.
The incentive for the brokerages is to sell a product. If investors don't buy stocks, those of outstanding issues and IPOs alike, brokerage firms won't have much in the way of revenues. So the balance is one-sided in favor of buys. What is more, much -- perhaps too much -- research is based on interviews with company officials, so it takes some courage to risk their ire and recommend sales. Further, every single one of a firm's clients can act on a buy recommendation, but only clients holding the particular stock can act on a sell recommendation. There is no obvious way around these roadblocks.
Nonetheless, the recent requirement that firms disclose their recommendations seems to have redressed the imbalance, though the terminology has changed. One major firm, for example, reports that so far this year 32% of recommendations are to "overweight," 47% are to "equal-weight," and 21% are to "underweight." I'm not at all sure, however, that this "enhanced indexing" terminology doesn't create more problems then it resolves. Further, it seems unlikely that either building a Chinese Wall between brokerage and investment banking or setting up a Byzantine new system in which giant Wall Street underwriters would subsidize independent research firms would work with much effectiveness. Perhaps the best remedy is the traditional one: Disclosure. "Sunlight, sunlight, sunlight" on the source of funding for research, on analysts' compensation, and on performance appraisals of prior recommendations.
But it's time to face reality: There is no evidence that research -- even the research of the Institutional Investor all-stars -- adds value. Academic studies only confirm what we all believe: The stock market is highly efficient, and that stock prices incorporate virtually all information. As I've often said, "Never think you know more than the market. Nobody does."
So if widely-disseminated sell-side Wall Street research -- whether independent or a product of investment banking and brokerage firms -- has little if any demonstrable economic value, it should command relatively small cash revenues. And that's the case: It is currently paid for largely by "soft-dollars," with the commissions paid for research bundled up with the price of execution and block positioning (and, for that matter, for other services such as mutual fund marketing). When a mutual fund manager buys research with commission dollars, there's far less attention paid to the cost/value equation than would be the case were it paid for out of the management company's profits. The more interesting issue is why users of research, largely the buy-side analysts of the institutional investing community, place much stock in street research in the first place.
The near-universal consensus among research providers and users alike is that if Street research could be purchased only with hard (i.e., real) dollars, the amount spent on it would plummet. Yet it does not necessarily follow that all research lacks intrinsic value. While the value of an original, comprehensive, and insightful research study becomes zero at the moment it becomes available to all market participants, the value of the same study by the research department of a single institution remains as long as the information remains proprietary.
If the value of mass-marketed Street research has a half-life measured in moments, and if proprietary institutional research has at least a potential value, the direction of change should be obvious. In their search for the Alpha of higher risk-adjusted returns, the firms in the mutual fund industry should beef up their own proprietary portfolio management, security analysis, and internal research capabilities.
Of course it would take real dollars, but they would be a drop in the bucket of revenues of mutual fund managers. Last year, for example, some $71 billion in revenues was generated by mutual fund advisory fees, sales loads, and out-of-pocket fees. Yet, according to my estimates, only about $4 billion of that amount, less than 6%, was spent on investment supervision and research, that very service which is said to constitute the prime reason that investors select funds -- "professional management."
In a business in which the profit margins of managers can exceed 50%, even after the huge marketing costs fund firms expend to bring in new money, increasing expenditures on investment management would hardly seem a tall order. And if fund performance improved even modestly, additional fee revenues on assets that were enhanced by higher returns -- and by the cash inflows that tend to accompany superior performance -- would more than repay the extra costs.
Indeed, if fund managers were willing to put their money where their mouth is and move to incentive-penalty fee schedules, the additional investment in research could be accompanied by far higher profits, for managers and fund owners. Under this scenario, the responsibility for most security analysis and research would gradually shift from the sell side -- with its present conflicted motives and unsatisfactory outcomes -- to the buy side, independent and proprietary. Only time will tell whether this tectonic shift would lead to a larger research community or a smaller one. But the reality that makes the dilemma of research -- who provides it, how much it costs, who bears that cost -- so interesting is that there can be no seller without a buyer, and no buyer without a seller. One wins, the other loses.
As a result, for the market as a whole, research is a dead-weight cost that turns a zero-sum game into a loser's game. While the billions spent by Wall Street and institutional managers on research doubtless elicits useful information, stimulates trading activity, and fosters liquidity, its costs, along with all of the other -- and even higher -- costs of financial intermediation, guarantee that for investors as a whole, beating the market will remain a loser's game.
So where does this scenario leave the individual investor? My advice is to follow these rules: 1) Recognize the fallibility of research and be skeptical, even cynical, that it will enable you to earn superior returns. 2) Demand disclosure -- of conflicted Street relationships and information flows, and of the amounts and effectiveness of your fund manager's portfolio supervisory and research services. 3) Understand the nature of the game, and diversify as broadly as you can. 4) If all else fails, skip the research and buy an all-market index fund. Then hold it for Warren Buffett's favorite holding period: Forever.
Note: The opinions expressed in this article do not necessarily represent the views of Vanguard's present management.
Return to Op-Ed Articles in the Bogle Research Center
©2010 Bogle Financial Center. All Rights Reserved.