Centre for Economic Policy Research

2.4
Analysts’ ‘excessive’ optimism
In most popular accounts of the stock market boom of the late 1990s, analysts
played an important role, promoting stocks of technology, media and 
telecommunications companies, helping the firms to raise capital. Stories after the
crash of the stock market suggest that pressures on analysts and misaligned 

---

Figure 2.2 Initial public offerings, 1980-2001
Source: Ritter and Welch (2002).
18 Conflicts of Interest in the Financial Services Industry
Figure 2.1 The boom and crash, 1995-2003
700
600
500
400
300
200
100
0
1995 1996 1997 1998 1999 2000 2001 2002 2003
S&P500
DowJones
Nasdaq
80
70
60
50
40
30
20
10
0
1980
1982 1984
1986
1988
1990
1992
1994
1996
1998
2000
1981
1983
1985
1987
1989
1991
1993
1995
1997
1999
2001
IPO Gross Proceeds
Money left on the table

---

Investment Banking: Conflicts of Interest in Underwriting and Research 19
incentives were greatest in the last few years of the boom. Even with the decline
in the market, some analysts appeared to be cheerleading, giving very positive
reports and recommendations to investors. Complaints by investors prompted
New York State’s Attorney General Eliot Spitzer to use his powers under the 1921
Martin Act to investigate and bring charges against any individual or firm
involved in the fraudulent purchase or sale of securities. As a result of his 
investigation, a sweeping settlement of the leading investment banks with the
SEC, the New York Attorney General, NASD, NASAA, NYSE, and state regulators
was announced on 20 December 2002 whose purpose was to reform the abusive
practices that had been uncovered.
One of the common complaints was that analysts made far more buy than sell
recommendations. For example, Shiller (2000) viewed the predominance of buy
recommendations and the optimistic forecast bias as obvious evidence for 
conflicts of interest. At the peak of the market in March 2000, 73% of recom-
mendations were to buy, 27% advised holding, and only 1% counselled sale. A
year later, these shares had changed little: 69, 30 and 1% respectively. In spite of
the bear market, by May 2002, they still stood at 62, 35 and 3% (Anderson and
Schack, 2002). Research directors at investment banks are known to dislike offer-
ing negative judgements; many prefer to drop coverage of companies rather than
continue to follow them with a sell recommendation. This aversion arises in part
because they do not want to anger investor clients, especially institutional
investors and their internal analysts. These analysts also make recommendations,
help to direct trading, and vote for the Institutional Investor polls. Similarly,
changing a recommendation to a sell may risk angering issuing companies and
losing their business (Pratt, 1993). 
The high percentage of buy recommendations looks like obvious evidence for
excessive optimism. Yet, many research-only houses also have far more buy than
sell recommendations. Furthermore, the predominance of buy recommendations
and positive earnings forecasts may not be the result of over-optimism but of 
censoring. If analysts censor by discontinuing coverage of a stock or failing to
update their forecasts, then the observed average buy recommendations and 
earnings forecasts will be higher than the unobserved means. This censoring
behaviour, rather than some bias in their true beliefs, may explain some of the
observed over-optimism in analysts’ forecasts and recommendations (McNichols
et al., 1997). This does not explain, however, the differences in optimism between
analysts working for underwriting and non-underwriting banks and the 
optimistic trend in the stock market boom (Hong and Kubik, 2003).
The perception of conflicts between research and underwriting in investment
banks is longstanding. There is considerable anecdotal evidence suggesting that
investment banks have not sought to separate the activities of their analysts and
investment bankers if only to present a consistent face to the public. The Wall