The news of no news in stock markets

Click to increase image sizeClick to decrease image size

Acknowledgements

We thank Vedat Akgiray, Celal Aksu, Noyan Arsan, Sezai Bekgoz, Paul Bennett, Christian T. Lundblad, William L. Megginson, Dan Palmon, Dilip Patro, Ephraim Sudit, Celali Yilmaz, Kun Yu, several Istanbul Stock Exchange specialists, the 2002 METU conference, and particularly the three anonymous referees for extremely helpful comments and suggestions on earlier versions. Finally, we thank the Whitcomb Center for Research in Financial Services for providing research support through the use of the WRDS system. All remaining errors are the authors' responsibility.

Notes

†We are indebted to an anonymous referee for suggesting this reference.

‡By unfavourable news we refer to worse than expected news.

†Please see ‘Selective Disclosure and Insider Trading’ (http://www.sec.gov/rules/final/33-7881.htm—last accessed on February 20, 2008) for more information on Regulation FD and public disclosure requirements in the US.

‡We are grateful to the ISE specialists and Sezai Bekgoz, for their insights on this issue.

§Please see Healy and Palepu (2001) and Verrecchia (2001) for surveys of the empirical and theoretical literatures on corporate disclosures.

¶Interviews with ISE officials suggest that the ISE, in most cases, receives a response from firms the same day.

∥Istanbul Stock Exchange: http://www.ise.org (in English) and http://www.imkb.gov.tr (in Turkish).

⊥Capital increases, dividend and/or price data for firms with the following tickers were unavailable: AGIDA, AKFIN, ARDEM, CUMRA, EGBRA, ERCYS, ESBNK, GUNEY, HLKSG, ISGEN, KAVOR, NIGDE, OSGYO, PENGD, PETUN, and firms with the following tickers were eliminated due to insufficient data for the estimation period: PNUN, RANTL, TURCS, YABNK, YASAS, AKTAS, BJKAS, EGFIN, ESCOM, FVORI, GSRAY, IPMAT, METUR, VKFRS, TEKTU, SELGD, SERVE.

†There are only six events with pre-event window CARs between −1% and 1%. Since this is a very small sample to analyse we exclude this sub sample from our analysis.

‡Our empirical results remain identical when we use (−10, 0), (−9, 0), (−7, 0) or (−3, 0) to divide the sample into three.

§We replicated our analysis using comparison period mean adjustment, market return adjustment, and Ibbotson RATS methods. Our results remained identical.

¶Fama and French (1993) propose the inclusion of size and book-to-market factors as an addition to the standard market model. We do not anticipate the inclusion of those factors to qualitatively change our results because this study only deals with the analysis of short-term returns. Brown and Warner (1985) examine daily stock return properties and how they affect event study methodology. Based on simulations, Brown et al. (1988) find very little difference in test power between simple and sophisticated abnormal return generating models.

∥We require that each event firm have at least sixty observations during the estimation period.

⊥ISE-100 is a value weighted index computed by the Istanbul Stock Exchange.

ΔBrown and Warner (1985) and Berry et al. (1990) document evidence that suggest a nonparametric test assuming an excess return median of zero to be misspecified.

†For brevity, we only report test statistics for selected event windows. Daily abnormal returns and their test statistics are available from the authors upon request.

†We are indebted to an anonymous referee for making this remark.

‡In this regression and later regressions we check for normality of residuals (Shapiro–Wilk W test), homoskedasticity (Cook–Weisberg), multicolinearity (variance inflation factor) and model specification (Ramsey RESET test). None of the regression diagnostic tests suggest any violations of the assumptions of ordinary least squares.

†Please see Farmer et al. (2004) and Weber and Rosenow (2006) for an investigation of the relationship between liquidity and large price changes.

‡Our choice of news dummy variables is arbitrary. However, we recomputed our results using an alternative specification; the periods (+1, +15), (+1, +30) and (+1, +60) and found similar results.

†Several index securities and index future contracts based on ISE have recently been launched. However these were not in existence during our sample period. Furthermore these securities mimic mainly large cap stocks and the future contracts pose difficulties due to their rigid maturity structures. Hence, reaping the returns of a hypothetical market portfolio may pose to be much more difficult in Turkey than in well developed markets.