Changes in the Composition of Publicly Traded Firms: Implications for the Dividend-Price Ratio and Return Predictability

Abstract

Standard-Nutzungsbedingungen:Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden.Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen.Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Changes in the composition of publicly traded firms: Implications for the dividend-price ratio and return predictability Terms of use: Documents in EconStor may S. JankChanges in the composition of publicly traded firms: Implications for the dividend-price ratio and return predictability *Stephan Jank * * November 16, 2012Abstract This article documents how the changing composition of U.S. publicly traded firms has prompted a decline in the long-run mean of the aggregate dividend-price ratio, most notably since the 1970s. Adjusting the dividend-price ratio for such changes resolves several issues with respect to the predictability of stock market returns: The adjusted dividend-price ratio is less persistent, in-sample evidence for predictability is more pronounced, there is greater parameter stability in the predictive regression (particularly during the 1990s), and there is evidence of out-of-sample predictability.

Beating the Average: Equity Premium Variations, Uncertainty, and Liquidity

“…The predictive regression model is motivated by Merton’s (1973) ICAPM, which shows that the expected equity premium is affected by market variance as well as covariance between returns and other financial and macroeconomic state variables (see, Zhu and Zhu, 2013). We generate out‐of‐sample forecasts by applying a commonly used rolling window approach (see, for example, Welch and Goyal, 2008; Rapach et al ., 2010; Ferreira and Santa‐Clara, 2011; Zhu and Zhu, 2013; Jank, 2015). This means that for a given window of constant length we use the first observations, where , to estimate …”

“…(2010), Ferreira and Santa‐Clara (2011), Zhu and Zhu (2013), Jordan et al . (2014), Jank (2015), Çakmaklı and van Dijk (2016), Algaba and Boudt (2017), Kolev and Karapandza (2017), Gupta et al . (2018), Stivers (2018), Yin (2019), Cao et al .…”

“…However, the TED spread and MU measure show significant positive utility gains compared to the historical average, which means that despite being not profitable, both predictors produce a lower loss than the historical average. In a situation where all investors lose money, 3 Studies analyzing the out-of-sample forecasting power of economic indicators are, for example, Goyal and Welch (2003), Campbell and Thompson (2008), Welch and Goyal (2008), Rapach et al (2010), Ferreira and Santa-Clara (2011), Zhu and Zhu (2013), Jordan et al (2014), Jank (2015), Çakmaklı and van Dijk (2016), Algaba and Boudt (2017), Kolev and Karapandza (2017), Gupta et al (2018), Stivers (2018), Yin (2019), Cao et al (2020), andPan et al (2020).…”