ABSTRACT
This paper provides new evidence on the pricing of market skewness risk by incorporating investor sentiment in the relation between sensitivity to innovations in implied market skewness and expected stock returns. Using both univariate and multivariate specifications, we conduct an extensive series of asset pricing tests on the cross-section of stocks during high and low sentiment periods separately. We find that market skewness risk carries a negative premium that cannot be explained away by known risk factors when sentiment is low. In contrast, the results are not conducive to a risk explanation when sentiment is high.
Disclosure statement
No potential conflict of interest was reported by the author(s).
Notes
1 An extensive review of the literature on the impact of skewness on asset returns can be found in the paper by Adcock, Eling, and Loperfido (Citation2015).
2 Ince and Porter (Citation2006) show that after a careful screening, which mainly consists in excluding securities that are not common shares, inferences drawn from Thomson Reuters Datastream data are similar to those drawn from CRSP (Center for Research in Security Prices) data.
3 Our sample was constrained by the availability of the sentiment proxies. The data was collected before the last update of Baker and Wurgler (BW) sentiment index on August 21, 2022. The index is now available until June 2022.
4 Ken French’s data library can be found at < www.dartmouth.edu/~kfrench/>.
5 Hou-Xue-Zhang q-factor’s data can be found at <globalq.org/factors.html>.
6 This approach is the most commonly used method in the literature as it mitigates the issue of measurement error in skewness associated with historically based estimates (Merton Citation1980; Harvey and Siddique Citation2000).
7 The AAII sentiment index is the result of a weekly survey conducted by the American Association of Individual Investors. The IIA sentiment index reflects the outlook of over 130 independent financial market newsletter writers. Both sentiment indices are computed as the spread between the percentage of bullish investors and the percentage of bearish investors (Bull – Bear).
8 Baker and Wurgler (Citation2007) (BW) sentiment index is computed as the first principal component of six proxies for sentiment: the trading volume as measured by the NYSE turnover, the dividend premium, the closed-end fund discount, the number and first-day returns on IPOs and the equity share in new issues. BW Index is obtained from Jeffrey Wurgler’s homepage < www.stern.nyu.edu/jwurgler/ >.
9 Huang et al. (Citation2015) (HJTZ) sentiment index is a modified measure of the Baker-Wurgler sentiment index using the partial least square approach to minimize noise in the index’s input variables. HJTZ Index is obtained from Goufu Zhou’ webpage: http://apps.olin.wustl.edu/faculty/zhou/PLS_BW_ investor_sentiment_indexes.xls.
10 To maintain parsimony, we keep the number of factors to a minimum. Hence, in our pre-formation regressions, we employ the market factor along with the market skewness risk, but we are careful to control for other known pricing factors later in our post-formation regression tests.
11 SKEW is highly serially correlated with a first-order autocorrelation of 0.789. ARMA (1,1) model is necessary to remove the autocorrelation from the data.
12 For , taking the first difference removes most of the autocorrelation in the data.
13 Value-weighted portfolios are obtained by weighing each stock in the quintile by its relative market value at the end of the beta estimation period.
14 See for example Duchin, Ozbas, and Sensoy (Citation2010); Garcia-Appendini and Montoriol-Garriga (Citation2013); Wang et al. (Citation2016).
15 Following numerous economic reports and papers such as NBER WP, 2010, Philippas et al. (Citation2013), Bekiros et al. (Citation2017) among others, the significant effect of the crisis remained until at least mid-2009.
16 The results of the chow test are available upon request.