تأثیر دیدگاه خبر ورود جریان پول، حق بیمه ارزش و قیمت گذاری دارایی بر ادغام بازار سهام اروپا
|کد مقاله||سال انتشار||مقاله انگلیسی||ترجمه فارسی||تعداد کلمات|
|15736||2010||18 صفحه PDF||سفارش دهید||محاسبه نشده|
Publisher : Elsevier - Science Direct (الزویر - ساینس دایرکت)
Journal : Journal of International Money and Finance, Volume 29, Issue 7, November 2010, Pages 1406–1423
The decomposition of national CAPM market betas of European countries’ value and growth portfolio returns into cashflow and discount rate news driven components reveals that i) high average returns on value portfolios are associated with disproportionately high sensitivity to national cashflow news which corroborates recent evidence for the U.S. and ii) two-beta variants of national CAPMs capture the cross-sectional dispersion in European stock returns. The latter finding is suggestive of relatively well integrated stock markets among the core European countries and reflects basic asset pricing theory. One (national) discount factor should price any (international) asset.
If stock markets are perfectly integrated, then they should be driven by the same factors. Harvey, 1991 and Campbell and Hamao, 1992 and Ferson and Harvey (1993) document the importance of global risk factors for the predictability of national stock market returns and explanations of their cross-sectional differences. However, if capital markets are sufficiently integrated, then cross-sectional dispersion in international asset returns should be explained by national risk factors as well. It is this latter line of thought that this paper pursues. This argument follows immediately from the basic pricing equation for asset returns. equation(1) View the MathML source1=Et(Mt+1Rt+1i) Turn MathJax on with Mt+1 the stochastic discount factor and View the MathML sourceRt+1i the gross return on asset or portfolio i. In words, an expected asset return should be constant once discounted with the stochastic discount factor (SDF) that is the same for all assets. Since equation (1) should hold for any asset from a national investor’s point of view, it requires sufficiently integrated financial markets when confronted with foreign asset returns. It is natural to test these implications for European countries which experienced a convergence process in terms of monetary and fiscal policy in the course of the establishment of a common currency area. The empirical exercises conducted in this paper could thus be interpreted as an assessment of the integration of the core European stock markets in the sample period from the first quarter of 1975 to the fourth quarter of 2007. From an asset pricing perspective, the choice of European countries is guided by two additional considerations. First, I would like to minimize the impact of exchange rate risk and focus on purely stock market based explanations of cross-sectional differences in stock returns. Various versions of international asset pricing models show that exchange rate risks are an important factor in explaining the cross-sectional dispersion in international stock market returns (e.g. Dumas and Solnik, 1995, de Santis and Gerard, 1997, Harvey, 1991 and Solnik, 1974). Since this paper concentrates on the core EMU countries plus Switzerland, the impact of foreign exchange risks on cross-sectional stock returns is likely to be relatively small. Secondly, Lane and Milesi-Ferretti (2005) show that European investors predominantly invest into euro-area equity. This paper applies equation (1) from a national investor’s point of view to explain the cross-sectional dispersion in European stock returns. Of course, the empirical analysis conducted in this paper relies on the choice of the stochastic discount factor, i.e. the asset pricing model. I focus on the two-beta variant of the Sharpe (1964) and Lintner (1965) capital asset pricing model (CAPM), recently proposed by Campbell and Vuolteenaho (2004) to explain the value premium on U.S. stock markets. Given this choice of the pricing kernel, it is natural to additionally examine the value premium in the European context. Value stocks, defined as stocks with high book value relative to market value (B/M), high earnings-to-price ratio (E/P), high cashflow-to-price ratio (C/P) and high dividend-to-price ratio (D/P) receive a lot of attention by practitioners as well as academics since they offer higher average returns than expected from their market betas in a Sharpe and Lintner CAPM. Conversely, growth stocks (stocks with e.g. low book-to-market value ratio) promise lower returns than predicted by the CAPM. This finding is not a unique observation on U.S. stock markets but by now well documented in international data (e.g. Capaul et al., 1993, Chan et al., 1991 and Fama and French, 1998). The Sharpe–Lintner CAPM assumes the existence of a so called market portfolio comprising all risky assets. The excess return on this market portfolio is a measure of all systematic sources of risk. Differences in the sensitivity to the market return (“betas”) should thus explain differences in average asset returns. In empirical work the market return is typically proxied by broad stock market indexes. While this practice can be criticized on various grounds (e.g. Roll, 1977, Campbell, 1996, Jagannathan and Wang, 1996 and Lettau and Ludvigson, 2001b), Davis et al. (2000) show that the CAPM works well when confronted with U.S. value and growth stock data in the sample period from 1929 to 1963 but works poorly in the modern time period from 1963 to the present. Campbell and Vuolteenaho (2004) explain the difference in the performance of the CAPM in the two sample periods by decomposing CAPM market betas into a cashflow (“bad”) and discount rate (“good”) variety. Intuitively, bad news about the market’s future cashflows reflect a decrease of wealth and hence lead to a fall in the value of the market but leave future investment opportunities unaffected. The value of the market portfolio could also decline because investors increase the discount rate applied to cashflows, which at the same time mirrors better future investment opportunities. Furthermore, the intertemporal CAPM of Merton (1973) suggests that the receptiveness to innovations in dividends (cashflows) should be rewarded with a higher price of risk than sensitivity to discount rate news. Campbell and Vuolteenaho (2004) show that value stocks’ market betas in U.S. post-war data contain a substantially higher cashflow component than growth stocks’ market betas which explains seemingly abnormally high average returns on value portfolios. This paper shows that European value stocks offer higher excess returns than their growth portfolio counterparts. In line with the findings of Campbell and Vuolteenaho, 2004 and Cohen et al., 2008 and Campbell et al. (2009), high average returns are associated with disproportionately high cashflow betas. From the perspective of a national investor, differences in the exposures to the cashflow component of the national market return explain the cross-sectional dispersion in value and growth stock returns across the European countries under study. One exception is Spain for which different exposures to the discount rate component of the national market return rationalize cross-sectional dispersion in value and growth portfolio returns. Taken together, however, the implications of capital market integration on asset pricing theory seem to be reflected in the sample of European countries under consideration. In addition, this evidence corroborates and extends the findings for the U.S. that differences in the sensitivity to the market return’s cashflow news component help to explain the value and growth anomaly. The main results of this paper suggest that this explanation is not only true for national but also for foreign countries’ value and growth stock portfolio returns. Finally, this paper extends the set of test assets to a cross-section of international stock market returns that comprise also countries outside Europe. This exercise could thus be interpreted as a test of world market integration. This paper presents the respective results from the perspective of a German investor as these are representative for the other European countries. It turns out that the national two-beta CAPM explains the cross-sectional differences in these asset returns only to a modest extent compared to the results obtained with European value and growth stocks. In general, its performance is slightly worse but still comparable with the Fama and French (1993) three-factor model. The remainder of this paper is organized as follows. In Section 2, I sketch the framework of Campbell (1991) and Campbell and Vuolteenaho (2004) used to identify cashflow and discount rate betas. Thereafter, I briefly discuss the choice of state variables in Section 3 and provide details of the data employed in this paper in Section 4. Section 5 discusses the empirical evidence. Finally, Section 6 concludes.
نتیجه گیری انگلیسی
Employing the framework of Campbell (1991) and Campbell and Vuolteenaho (2004), this paper shows that high average returns on European value portfolio returns can be reconciled with the two-beta variant of the Sharpe–Lintner CAPM from a national investor’s perspective. High returns on value stocks are associated with relatively high cashflow betas compared to the respective discount rate betas. With the exception of Spain, this finding is a salient feature of the data irrespective if one takes the stance of a Belgium, French, German, Italian, Dutch or Swiss investor. The main results of this paper thus corroborate Campbell and Vuolteenaho, 2004 and Cohen et al., 2008 and Campbell et al. (2009) who explain differences in returns on U.S. book-to-market ratio sorted stock portfolios with different exposures to the U.S. market’s cashflow news. Since two-beta versions of national CAPMs capture the cross-sectional dispersion in European value and growth stock returns, the findings of this paper could be interpreted as supporting evidence for the implication that a national asset pricing model should explain cross-sectional dispersion in national and foreign asset returns if national capital markets are sufficiently integrated.