This paper focuses on the turn-of-the-month (TOM) and intramonth anomalies in government bond returns. In particular, we examine whether the TOM and intramonth effects exist in government bond markets, and moreover, whether these anomalies are related to the release of macroeconomic news as suggested in recent stock market studies. Using data on the 2-year and 10-year US Treasury Notes and German government bonds, we document a modest TOM effect in government bond returns. This effect does not disappear after controlling for the release of macroeconomic announcements, thereby suggesting that the origin of the TOM effect is not necessarily the same across asset classes.
The turn-of-the-month (TOM) and intramonth effects are two of the many seasonal anomalies that have been documented in the existing literature. Considerable empirical evidence suggests that stock returns are positive at the turn-of-the-month and that returns are positive during the first half of the month (see e.g. Ariel, 1987, Lakonishok and Smidt, 1988, Agrawal and Tandon, 1994, Booth et al., 2001, Kunkel et al., 2003, McGuinness, 2006, Nikkinen et al., 2007 and Gerlach, 2007).3 Although the TOM and intramonth effects have been examined extensively in stock markets, surprisingly little is known about the TOM and intramonth effects in bond markets.4 This paper aims to contribute to the literature by focusing on the turn-of-the-month and intramonth effects in government bond returns.
Previous studies have proposed several alternative explanations for the TOM and intramonth anomalies. These include the “window dressing” (see e.g. Thaler, 1987) and the turn-of-the-month liquidity hypothesis (see e.g. Ogden, 1990 and Booth et al., 2001). Most recently, Nikkinen et al. (2007) and Gerlach (2007) postulate that the TOM and intramonth anomalies in stock markets are caused by the clusterization of US macroeconomic news releases. The empirical findings documented in Nikkinen et al. (2007) and Gerlach (2007) indicate that both the TOM and intramonth effects disappear after controlling for the clustered release of macroeconomic news.
The existing literature shows that most of the value-relevant information for government bonds is related to macroeconomic fundamentals (see e.g. Balduzzi et al., 2001, Ahn et al., 2002 and Andersson et al., 2009). Given that macroeconomic news releases are shown to explain the TOM and intramonth effects in stock markets (Nikkinen et al., 2007 and Gerlach, 2007), it is of interest to examine whether the clustered release of US macroeconomic news causes turn-of-the-month and intramonth effects also in government bond returns.
Using data on 2-year and 10-year US Treasury Notes and German government bonds, we document a modest TOM effect with positive bond returns on the last trading day of the month. We find no evidence of intramonth effects in government bond returns. Furthermore, our empirical findings indicate that the observed turn-of-the-month effect is not caused by the clustered release of macroeconomic news. In general, these findings indicate that the origin of the turn-of-the-month effect is not necessarily the same across asset classes.
The remainder of this note is organized as follows. Section 2 presents the data, while Section 3 describes the methodology used in the empirical analysis. Our empirical findings are reported and discussed in Section 4. Finally, Section 5 provides concluding remarks.
This paper examines the turn-of-the-month and intramonth effects in government bond markets, and furthermore, examines whether these anomalies are related to the clustered release of macroeconomic news as suggested in recent stock market studies (Nikkinen et al., 2007 and Gerlach, 2007). Given that government bond markets are significantly affected by macroeconomic announcements, it is of interest to examine whether the clustered release of macroeconomic news leads to turn-of-the-month and intramonth effects in government bond returns.
Using data on 2-year and 10-year US Treasury Notes and German government bonds, we document a modest TOM effect with positive returns on the last trading day of the month. We find no evidence of intramonth effects in government bond returns. Furthermore, our empirical findings indicate that the turn-of-the-month effect, or the observed last-day-effect, is not caused by the clustered release of macroeconomic news. These findings are somewhat inconsistent with the results previously documented in stock markets. In general, the findings reported in this paper indicate that the origin of the turn-of-the-month effect is not necessarily the same across asset classes.
