نقش اتصال کوتاه، اندازه شرکت و زمان بر ناهنجاری های بازار
|کد مقاله||سال انتشار||مقاله انگلیسی||ترجمه فارسی||تعداد کلمات|
|13086||2013||27 صفحه PDF||سفارش دهید||18193 کلمه|
Publisher : Elsevier - Science Direct (الزویر - ساینس دایرکت)
Journal : Journal of Financial Economics, Volume 108, Issue 2, May 2013, Pages 275–301
We examine the role of shorting, firm size, and time on the profitability of size, value, and momentum strategies. We find that long positions make up almost all of size, 60% of value, and half of momentum profits. Shorting becomes less important for momentum and more important for value as firm size decreases. The value premium decreases with firm size and is weak among the largest stocks. Momentum profits, however, exhibit no reliable relation with size. These effects are robust over 86 years of US equity data and almost 40 years of data across four international equity markets and five asset classes. Variation over time and across markets of these effects is consistent with random chance. We find little evidence that size, value, and momentum returns are significantly affected by changes in trading costs or institutional and hedge fund ownership over time.
The returns to portfolios based on firm size, value, and momentum have presented a challenge to asset pricing theory since their discovery.1 The pervasiveness, robustness, and magnitude of the return premia associated with size, value, and momentum has made them the focal point for discussions of market efficiency as well as critical inputs for describing the cross section of expected returns. These market anomalies have been shown to be robust in other stock markets, other time periods, and other asset classes (Chan et al., 1991, Hawawini and Keim, 1995, Fama and French, 1998, Fama and French, 2012, Rouwenhorst, 1998 and Griffin et al., 2003; Asness, Moskowitz, and Pedersen, forthcoming) and have motivated the use of empirical asset pricing models that incorporate their returns (Fama and French, 1993, Fama and French, 2012 and Carhart, 1997; Asness, Moskowitz, and Pedersen, forthcoming). The vast literature on these anomalies has generated a wide debate as to the underlying explanations for these return premia, which generally fall into two categories: rational risk-based models or behavioral theories. Also, a lack of consensus exists on the implementability of these strategies in practice. Both of these issues are paramount to discussions of market efficiency with respect to these anomalies. Given the disparate views in the literature, we take stock of the empirical evidence of these market anomalies, to shed some light on these issues. We investigate three questions. First, how important is short selling to the profitability of these strategies? Second, what role does firm size play in the efficacy of these investment styles? Third, how have the returns to these strategies and the role of size and shorting varied over time? The importance of size, shorting, and time to the profitability of these strategies helps identify possible explanations as well as implementation costs associated with each anomaly. Without having to specify a trading cost model, which is investor specific, we acknowledge that small stocks are more costly and more difficult to trade and that shorting is more costly and more constrained. Arbitrage activity and capital are, therefore, likely to be more limited in small stocks and when shorting. Consistent with this view, many behavioral theories suggest stronger returns among smaller, less liquid securities and when there is negative news (Hong and Stein, 1999, Hong et al., 2000 and Lee and Swaminathan, 2000). How these effects have evolved over time could also help reveal what drives these anomalies. We examine how these effects have varied with changes to trading costs and institutional ownership over time, including the surge in hedge fund activity over the last two decades. We examine the role of shorting from two perspectives: the value added from short selling of assets in a long-short portfolio and the value added from underweighting stocks relative to a benchmark (e.g., the market portfolio). Because short positions are generally more costly to maintain than long positions and because some investors are restricted from taking short positions (e.g., mutual funds and institutions) the net of trading cost returns could be substantially lower and not accessible to many investors, if shorting is an important driver of the profits to these strategies. The role of firm size also plays a dual part in our study. First, we examine the return premium associated with size. Second, we examine the interaction between firm size and the return premia to value and momentum, including the interaction of firm size with the importance of shorting for these strategies. If the bulk of the returns to these strategies is concentrated among small or micro-cap stocks, then the fraction of the market affected by these anomalies could be small. Moreover, trading costs are typically highest among the smallest stocks, and small stocks are the most difficult and costly to short. Hence, the interaction between firm size and the other anomalies provides insight into the implementation costs of these strategies. Using data over the last 86 years in the U.S. stock market (from 1926 to 2011) and over the last four decades in international stock markets and other asset classes (from 1972 to 2011), we find that the importance of shorting is inconsequential for all strategies when looking at raw returns. For an investor who cares only about raw returns, the return premia to size, value, and momentum are dominated by the contribution from long positions. Shorting only matters if investors care about returns relative to a benchmark, such as the market portfolio. Looking at market-adjusted returns (market alphas), long positions comprise the bulk of the size premium, capture about 60% of the value premium, and comprise half of the momentum return premium. Long-only versions of value and momentum deliver positive and significant alphas relative to the market. Looking across different size firms, we find that the momentum premium is present and stable across all size groups. Little evidence exists that momentum is substantially stronger among small cap stocks over the entire 86-year US sample period. The value premium, meanwhile, is largely concentrated only among small stocks and is insignificant among the largest two quintiles of stocks (largest 40% of NYSE stocks).2 The contribution to value and momentum profits from shorting varies with firm size. Shorting becomes less (more) important for momentum and more (less) important for value strategies as firm size decreases (increases). However, across all size groups, we cannot reject that the abnormal profits to value and momentum trading are generated equally by long and short positions. We examine the robustness of these findings over time and in relation to time series variation in trading costs, institutional ownership, and hedge fund assets. First, we find that significant momentum returns are present across size categories in every 20-year subsample we examine, including the most recent two decades that followed the initial publication of the original momentum studies. Moreover, the findings of Hong, Lim, and Stein (2000) and Grinblatt and Moskowitz (2004) that momentum is markedly stronger among small cap stocks and on the short side seems to be sample specific. Outside of the samples studied in those papers [most notably, 1980 to 1996, the sample period covered in Hong, Lim, and Stein (2000)], we find no evidence that momentum is stronger among small stocks or from shorting, and, over the entire period that includes those samples, no significant difference emerges in momentum returns across size groups or from shorting. Returns to value investing, however, are consistently stronger among small cap stocks in every subperiod and are largely nonexistent among large cap stocks in three of the four subperiods we examine. We find no evidence that returns to these strategies have changed over time or that the contribution from long versus short positions has changed over time. Second, we examine how the returns to size, value, and momentum have varied with changes in trading costs and institutional ownership over the last century, including hedge fund participation over the most recent two decades. While we find some weak evidence that momentum returns rise as trading costs rise, particularly among small stocks, little evidence exists that trading cost changes have had a material effect on these return premia. Likewise, we find little to no relation between institutional ownership or hedge fund growth and these return premia, other than a decline in the size premium with increased institutional ownership, consistent with Gompers and Metrick (2001). Finally, we also examine these strategies across four international stock markets and five other asset classes over a 40-year period. Additional evidence from these other markets confirms the existence of value and momentum return premia and similarly finds an equal contribution of long and short positions to those returns. Our findings shed new light on the vast literature on size, value, and momentum effects in asset pricing. Hong, Lim, and Stein (2000) and Grinblatt and Moskowitz (2004) also examine momentum returns across firm size and the long versus short side contributions to momentum profits. Their conclusions that momentum is stronger among small cap stocks and that two-thirds of the profits come from shorting are not robust in our larger sample. We find that momentum returns are largely unaffected by size and that the short side is no more profitable than the long side over our longer 86-year sample period and in eight other markets and asset classes. Fama and French (2012) examine value and momentum in international stock markets from 1990 to 2009 across size groupings. They find that both value and momentum premia are present in all markets, with the exception of momentum in Japan, and that value and momentum premia exist in all size groups, with return premia being stronger as size decreases.3 Over our longer 86-year sample period, we find no evidence that momentum declines with firm size, but we do find consistent evidence that value returns are weaker for larger stocks. The rest of the paper is organized as follows. Section 2 describes our data and portfolios. Section 3 analyzes the importance of shorting and time on size, value, and momentum strategies. Section 4 examines the role of firm size on value and momentum return premia as well as the interaction between firm size and the importance of shorting. Section 5 analyzes how the importance of shorting and firm size varies over time and whether variation in trading costs or institutional ownership and hedge fund participation has impacted these strategies over time. Section 6 examines other equity markets and asset classes. Section 7 concludes.
نتیجه گیری انگلیسی
We examine the role of shorting, firm size, and time on size, value, and momentum strategies over the last century in US data and over the last four decades in international stock markets and other nonstock asset classes. We find that the returns to value decrease with size over our sample period and are insignificant for the largest stocks. Momentum premia are present in every size group and do not vary reliably across size groups over the entire sample period. We find no consistent evidence that momentum returns decrease with firm size as suggested by the literature. We also find that about half of value profits and half of momentum profits come from the long side. We find no evidence that shorting profits are more important for momentum, in contrast to claims in the literature. The role of shorting and size on momentum profits previously shown in the literature are not a robust feature of the data. Using a much longer time series and looking across other equity markets and asset classes, we find no reliable size effect in momentum or stronger role for shorting. However, the contribution of long versus short positions does vary with firm size. Short selling profits become more important for momentum strategies and less important for value strategies as size increases, and they become less important for momentum and more important for value as size decreases. These patterns are generally consistent over time. Long-only versions of value and momentum also consistently yield positive alphas across size groups, across markets and asset classes, and across time. Overall, the premium for momentum, whether long-short or long-only, appears to be consistently higher than that of value, especially among large cap stocks in which the value premium is weakest. Finally, we examine whether profits to these strategies have changed significantly over time or in relation to changes in trading costs or institutional and hedge fund investment over time. We find little evidence that size, value, or momentum premia have changed over time or are affected by changes in institutional or hedge fund participation in markets and find only mild evidence that trading costs have any relation to the profitability of these strategies. Our results have implications for understanding the nature of size, value, and momentum return premia and for implementing size, value, and momentum portfolios in practice.