Market response to analyst recommendations in the “dartboard” column: the information and price‐pressure effects

Abstract

The market response to securities selected by analysts featured in the “Dartboard” column of The Wall Street Journal (WSJ) is the subject of this study. Prior studies have reported significantly positive abnormal returns on the WSJ issue date, followed by a partial price reversal, suggesting that the initial response is partially attributable to price pressure. If we use a post‐event instead of a pre‐event estimation period, we find a similar event day response but no significant reversal. Though we are unable to reject the possibility of a price pressure effect, our results are generally consistent with the existence of an information effect. We also detect a differential response to securities recommended by analysts invited back to the contest and those recommended by newcomers, suggesting that the information effect is related to the market following recommendations of potentially superior analysts.

Do Investment Newsletters Move Markets?

“…On both sides of the recommendation date, excess relative daily turnover declines almost monotonically. These results are consistent in pattern and are slightly larger in magnitude to the excess volume documented in many of the studies on the effect of second‐hand news such as Liu, Smith, and Syed (1990), Barber and Loeffler (1993), Albert and Smaby (1996), Chang and Suk (1998), Mathur and Waheed (1995), Sarkar and Jordan (2000), and Neumann and Kenny (2007) 7…”

“…Note that in all of these studies, the news is semi‐private information that is re‐released in a public venue. Event study results for analyst buy recommendations highlighted in the Wall Street Journal's “ Dartboard ” column are as follows: Barber and Loeffler (1993)– 4.06% for t = 0 to t + 1; Metcalf and Malkiel (1994)– 3.0% for t = 0 to t + 1; Griffin, Jones, and Zmijewski (1995)– 1.10% for t + 1, Albert and Smaby (1996)– 3.92% for t = 0 to t + 1; Beltz and Jennings (1997)– 0.52% for t + 1; Liang (1999)– 3.52% for t = 0 to t + 1. Pruit, Van Ness, and Van Ness (2000)– 3.66% for t = 0.…”

Fifty-Fifty. Stock Recommendations and Stock Prices. Effects and Benefits of Investment Advice in the Business Media

“…Dimson and Marsh (1986), 135 and passim; cf. also Albert and Smaby (1996), 60ff. Salinger (1992) shows that an insufficient consideration of standard deviations can lead to distorted results of event studies and a false assessment of the level of significance.…”

“…The information content of the recommendations only plays a marginal role -if any role at all. Albert and Smaby (1996) are the only ones to find no indications of a return reversal. They do not find any sign of significant returns either, though.…”

Market Reaction to Second-Hand News: Attention Grabbing or Information Dissemination?

“…Moreover, Albert and Smaby (1996) document that the post-publication price reversal shown in the PPH studies concerning "Dartboard" has a methodological explanation: most stocks recommended by analysts in the column tend to be momentum stocks; therefore, a pre-event estimate of the market model generates inflated alpha values that bias upward the post-event normal return, leading to negative abnormal returns in the days after the publication.…”