Comparative study of stock analysts raises doubts about $450-million program to improve research
August 1, 2004
For more information, contact: Benjamin Haimowitz, HHaimowitz@aol.com
"Investment banks are not the problem they have been made out to be"
On July 27 ten major investment banks began to implement a $450-million program designed to improve stock research for investors and prevent a recurrence of the inordinate hype and optimism that swept the markets in recent years.
Agreed to last year under pressure from law enforcement officials, the program responds to what a new study calls "the popular explanation" for those recent excesses -- namely, that "analysts working for investment banks have either been compromised by the hefty bonuses that they can earn from writing positive reports on investment-banking clients, or have been pressured to write favorable reports by investment bankers at their firms."
The study then proceeds to cast doubt on this explanation -- and with it the worth of the new program, which requires the banks to provide three independent research reports along with their own report for every company they cover.
Presented at the annual meeting of the Academy of Management (New Orleans, August 8-11), the research finds that, whatever the misdeeds of particular individuals or institutions, the analysis emanating from investment banks has been, on the whole, less optimistic than that issued by other types of financial firms.
"Our findings suggest that, as a group, investment banks are not the problem they have been made out to be," comments Paul Healy a professor at Harvard Business School, who carried out the research there with Amanda Cowen and Boris Groysberg. "Unless analysts outside the investment banks experience a material decline in optimism, the new system seems as likely to reduce overall research quality as to increase it."
The findings are based on comparisons of forecasts made from 1996 through 2002 by analysts at four different types of financial firms:
--full-service investment banks, 150 in all, which fund research through lead underwriting and distributing new issues;
-- syndicate member banks, 131 in all, which fund research through modest fees from distributing new issues and from trading;
-- brokerage firms, 84 in all, which do not participate in investment banking and rely in whole or in part on trading commissions to fund research;
-- research firms, 18 in all, that sell equity research to their clients and provide neither investment-banking or trade-execution services.
The study measures differences in forecast optimism among the four groups with respect to near-term earnings, long-term earnings growth, and target prices. Measures of optimism are obtained through a formula that compares each analyst's forecasts with those of all other analysts covering the same companies. The near-term earnings sample consisted of about 6,500 analysts, while the other two samples consisted of about 4,000 analysts. The near-term and long-term earnings forecasts were issued from 1996 through 2002, and the target-price forecasts from 1999 through September 2002. All told, close to one million forecasts make up the final sample, with investment banks accounting for about 86%, syndicates for 9.2%, brokerages for 4.2%, and pure research firms for 0.6%.
Close to 90% of the forecasts were for near-term earnings, where investment-bank analysts were, on average, significantly less optimistic than their counterparts in non-underwriter firms. In contrast, analysts of brokerages tended to make the most optimistic forecasts, leading the study's authors to observe that "analyst optimism is more closely linked to trading, the sole source of research funding for brokerage firms, than to underwriting."
Investment-bank analysts were also significantly less optimistic, on average, than other analysts in their predictions of long-term earnings growth (three to five years in the future). The pattern was similar for target price forecasts, where again brokerage analysts were conspicuous for their high degree of optimism.
"Given the concerns expressed by regulators, practitioners, and academics about the impact of underwriting on analysts' incentives, our results are somewhat surprising," the study's authors acknowledge. In trying to account for them, they examine five potential explanations.
The first is the possibility that the low degree of optimism of investment-bank analysts was true of the post-boom period but not of the boom period itself. The investigators find, however, "that the patterns observed...during the full sample period also hold for both the period prior to April 2000 (when the NASDAQ was at its peak) and the subsequent period (when it crashed)."
A second possible explanation is that the low optimism of investment banks is largely owed to "bulge" banks, the six biggest, whose high status obviates the need for optimistic predictions. But, while there is some evidence of lower optimism in bulge banks than in the non-bulge group, "it is unlikely to explain the difference in relative forecast optimism for underwriter and brokerage firms."
A possible explanation for the high degree of forecast optimism among brokerages is that it derives from firms that serve retail clients as distinct from institutional investors, retail customers allegedly being less discerning. But, when the investigators test this hypothesis by dividing brokerage firms into two groups based on their clientele, they find only scant evidence of greater optimism among firms that serve retail clients.
Healy and his colleagues also test the possibility that, hidden in the results for underwriting banks, are a subset of highly optimistic forecasts for companies that are most likely to make new issues. But even in rating such companies, they find, the analysts of investment banks proved to be less optimistic than those of non-underwriter firms.
Finally, the investigators ask whether lower optimism of the investment-bank analysts means greater forecast accuracy. They find it does.
In conclusion, the authors write, their findings "suggest that the primary force underlying analysts' forecast optimism is not analysts' underwriting incentives but the trading incentives faced by most analyst firms, and brokerage firms in particular."
"Recent attempts to eliminate banking conflicts," they continue, "may not reduce analyst research optimism, since analyst bias is even more prominent for pure brokerage firms than for firms that also perform underwriting...Unless the additional resources provided by the ten sanctioned banks leads to the formation of new research firms that will provide unbiased research, or to a material decline in the optimism of analysts at brokerage firms, which are stepping up their research offerings to take advantage of this opportunity, this requirement seems as likely to reduce overall research quality as to increase it. Further, by requiring investment banks to eliminate underwriting as a source of research funding, the Settlement implies that banks will have to rely more on trading to support research, increasing the likelihood that bank analysts will face the same incentives as brokerage analysts."
The Academy of Management, founded in 1936, is the largest organization in the world devoted to management research and teaching. It has about 15,000 members in over 90 countries, including some 10,000 in the United States. The academy's 2004 annual meeting drew about 7,000 scholars and practitioners to New Orleans for more than 1,000 sessions on a host of issues relating to corporate organization, the workplace, technology development, and other management-related subjects.
- Media Coverage:
- Reuters. Investment banks not to blame for research hype: study. (Sunday, August 08, 2004).
- Reuters. Long-time analysts more optimistic: Harvard study. (Monday, August 09, 2004).
- The Wall Street Journal. Picking Your Analyst Gets Messier. (Friday, August 06, 2004).