حجم بالای حق بیمه بازده : شواهد کشور صلیبی
|کد مقاله||سال انتشار||مقاله انگلیسی||ترجمه فارسی||تعداد کلمات|
|19352||2012||25 صفحه PDF||سفارش دهید||محاسبه نشده|
Publisher : Elsevier - Science Direct (الزویر - ساینس دایرکت)
Journal : Journal of Financial Economics, Volume 103, Issue 2, February 2012, Pages 255–279
We examine the high volume return premium across 41 different countries and find it to be a phenomenon found in both developed and emerging markets. The premium is not caused by systematic differences in risk or liquidity. Using Merton's (1987) investor recognition hypothesis as a guide, we find the magnitude of the premium is generally associated with country and firm characteristics hypothesized to affect returns subsequent to a change in a stock's visibility. We also characterize the time-series properties of the premium and consider economic trading strategies.
The high volume return premium, that is, the excess market-adjusted return that occurs after a stock receives a substantial positive volume shock, has been found to be an intriguing component of financial markets in the United States (Gervais, Kaniel, and Mingelgrin, 2001). In this paper, we take the high volume return premium to cross-country data and examine two major issues. First, we examine whether the premium holds across diverse stock markets to the degree it holds in the United States. Second, we take advantage of differences in market, investor, and firm characteristics across countries to examine the determinants of the high volume return premium. We investigate the hypothesis as to whether the high volume return premium is associated with changes in investor visibility for a stock, as would be predicted by Merton's (1987) investor recognition hypothesis and as suggested by evidence presented by Gervais, Kaniel, and Mingelgrin. Examining data from 41 countries that vary in their market structure, investor composition, and constituent firm characteristics, we first confirm that the high volume return premium is pervasive, occurring in almost all developed countries and in many emerging market countries as well. We further show that differences in risk or liquidity cannot explain these return premiums. In addition, we characterize their time-series properties. We then turn to the question of how the existence and the magnitude of the high volume return premium are affected by different characteristics of the firm, its market, and its potential investors. As a guide to determining which characteristics would be expected to be related to the high volume return premium, we employ Merton's (1987) investor recognition hypothesis. This theory implies that investors' incomplete information affects their trading behavior and the resulting security values. That is, because of the incomplete nature of their information, some investors may not become aware of certain securities and, consequently, do not hold those securities in their portfolios. In such a case, Merton shows that investors will be inadequately diversified and will demand a premium for taking on nonsystematic risk, causing a stock's required rate of return to depend on the size of its investor base. The main idea in Merton's (1987) theory relies on an information environment that limits the investors who are aware of a firm's securities to a subset of the potential investing population. The stock's limited visibility among investors means that if the stock achieves increased visibility and consequently increases its investor base, there should be a reduction in the cost of capital and a concomitant increase in the firm's market value. Thus, the implications of the investor recognition hypothesis should vary across firms with different market, demographic, and firm characteristics as these characteristics of the information environment might affect the costs of being informed, the level of a stock's visibility, and an investor's decision on whether to purchase the stock. In our empirical tests we identify potential determinants of the high volume return premium and in so doing also test predictions derived from the investor recognition hypothesis using both country-level and individual firm analyses. In the country-level analyses we provide evidence that is consistent with most, but not all, of the derived predictions from Merton's (1987) model. In particular, we show that the magnitude of the high volume return premium is associated with country characteristics that are expected to be related to the importance of a stock's visibility, such as investor demographics, the extent of information dissemination, the country's stock market composition, and investor confidence in the country's markets. Consistent with the implications of Merton's (1987) hypothesis, we find that the return premium on a stock following a volume shock is increasing in the extent to which the stock is less visible, a priori, to investors. We find this result with several measures of visibility. That is, we find greater high volume premiums in countries that are more developed, countries with more listed companies per urban population, and countries with more dominant stocks (either through large size or industry domination) in their stock markets. We also find that the high volume return premium is decreasing in the market's aggregate risk aversion (as reflected in the degree of investor confidence in the market). In the individual firm analyses, we find mixed evidence on whether the high volume return premium is associated with the firm-specific variables that would be predicted by Merton's (1987) hypothesis according to our interpretation. Consistent with the visibility argument, we find that the high volume return premium is decreasing in a firm's size relative to other firms in the domestic market and it is also smaller if the firm is a member of the FTSE All-World index. However, not all of the expected predictions are supported by the data. For example, we find that the high volume return premium is increasing in the existence of analyst coverage and inclusion in the Standard & Poor's (S&P) Transparency and Governance Index, but it is not affected by the magnitude of the analyst coverage or the S&P Transparency and Governance ranking for the firm. We go further by examining the effects of the firm-specific determinants of the high volume return premium within each of the G-7 countries separately and obtain results that are consistent with our cross-country findings. We next consider the viability of economic trading strategies for retail and institutional investors in different countries, particularly given the previously documented variation in transaction costs across these two types of investors and countries (e.g., Lesmond et al., 1999, Domowitz et al., 2001, Chiyachantana et al., 2004, Lesmond, 2005 and Eleswarapu and Venkataraman, 2006). We assume large institutional investors are likely to face transaction costs distinct from those of retail investors. Specifically, beyond the explicit trading costs required of both retail and institutional investors, the institutional investors also face implicit trading costs (e.g., bid-ask spread) due to the large size of their transactions. Consequently, we differentiate between these classes of investors by considering variations in the impact of explicit and implicit transaction costs on the viability of trading strategies. In tests employing estimated transaction costs, we first show that the high volume return premium remains significant in the G-7 stock markets after controlling for the explicit transaction costs retail investors would face. However, once we include the implicit trading costs that large investors would face in the G-7 markets, we do not find that the premium remains significant. We also find that in developed markets other than the G-7 and in emerging markets, even the estimated explicit trading costs are too high for the retail investors to profit from the high volume return premium, on average. Overall, our results are generally consistent with previous empirical studies that provide support for the implications of Merton's (1987) investor recognition hypothesis in that changes in stock visibility are an important aspect of investor decision-making.1 A key distinction between these previous studies and ours is that their findings are confined to a single within-country sample while our results merge within-country results for multiple countries with cross-country evidence. Our paper proceeds as follows. In Section 2, we present the data and the methodology for measuring the high volume return premium. We then provide the results from the empirical tests of the premium. We examine the determinants of the high volume return premium using tests involving characteristics across and within countries in Section 3. We consider the questions of time variation in Section 4 and economic trading strategies in Section 5. Finally, we provide our concluding comments in Section 6.