ارزش بازار ویژه سرمایه انسانی و بازده سهام
|کد مقاله||سال انتشار||مقاله انگلیسی||ترجمه فارسی||تعداد کلمات|
|4815||2009||14 صفحه PDF||سفارش دهید||محاسبه نشده|
Publisher : Elsevier - Science Direct (الزویر - ساینس دایرکت)
Journal : Journal of Banking & Finance, Volume 33, Issue 9, September 2009, Pages 1610–1623
We investigate whether and how well firms’ stock market valuations reflect their employees’ collective skills and effectiveness relative to that of their industry peers and competitors. We devise a relative stock market valuation measure of human capital intangibles (EVHC) and find that portfolios of low EVHC firms systematically outperform portfolios of high EVHC firms by an average 1.34% per month. However, this is primarily a small firms effect, because for large firms the excess returns of the arbitrage portfolio that is long on the low EVHC stocks and short on the high EVHC stocks is zero. Our results suggest that reliance on human capital intangibles may proxy for risk not fully accounted for by conventional asset pricing models, or alternatively, that the market cannot correctly price human capital intangibles for small size firms.
Human capital comprises the physical and intellectual skills and capabilities acquired through education and training that enable an individual to perform tasks effectively and to be productive. As such, human capital is at the core of several economic theories that provide explanations for the link between individuals’ skills and earnings. Companies compete in the labor market to attract the services of individuals. Thus, the labor market value of human capital is reflected in the compensation amounts received by human capital investors. What remains to be determined is how human capital is reflected in stock valuation. Human capital is an important parameter of firm performance. Theory and empirical evidence suggest that maximizing firm value involves optimally mixing people, technology, and physical assets in order to produce strategic goals (see, among others, Myers, 1999, Becker and Huselid, 1998, Boudreau and Ramstad, 1997, DiFrancesco and Berman, 2000, Huselid, 1995, Lev, 2001 and Provo, 2000).1 Even though the importance of human capital and its valuation in the labor market have been studied extensively, human capital intangibles are either difficult to incorporate in financial valuation or simply missing in traditional accounting valuation methods. This is because the input of human capital cannot be measured clearly and the product of human capital is often masked by other factors. Thus, it is difficult to gauge the stock market’s ability to recognize the quality of human capital, i.e., whether firms’ equity valuation is commensurate with their employees’ collective skills and effectiveness relative to that of their industry peers and competitors. In theory, the link between firm valuation and human capital can be understood in the context of a simple model wherein the firm’s total market value, MV(F), is decomposed into the present values of profits attributable to tangible assets (MV(T)) and intangible assets (MV(I)), respectively. If human capital-related intangibles contribute a sizeable component of MV(I), then they should also have an impact on MV(F). However, since under current US accounting standards intangible assets are not reported in financial statements, there is considerable doubt about the market’s ability to correctly assess MV(I) (see Lev and Zarowin (1999)). Past studies examining the impact of intangibles on valuation have concentrated on R&D (and advertising) expenditures as a proxy of a firm’s intangibles (see for example, Chan et al. (2001)). However, these measures suffer from two limitations: first, they are available only for a small subset of public firms, and second, they do not account for a possibly sizable portion of the firm’s intangible assets that is attributed to the firm’s employees.2 In this paper we use a more general and representative measure of the market’s valuation of firm intangibles than those based on R&D and advertising expenses. Ideally we would like to have employed a measure that is based on employee wages. Unfortunately, this information is mostly absent in data extracted from COMPUSTAT,3 and, in addition, it is not clear whether salaries correctly reflect investors’ assessments of how much human capital intangibles contribute to firm performance.4 We therefore devise a relative stock market valuation measure of human capital intangibles (EVHC) based on the ratio of the market value of equity per employee to the industry’s median market value of equity per employee and examine whether it is related to future return performance. In an efficient market EVHC would correctly impound all relevant information about the value of the firm’s intangibles and therefore there should be no systematic relationship between EVHC and future returns. Conversely, if reliance on human capital intangibles constitutes a risk not accounted for in conventional asset pricing models, then EVHC will be related to future returns. Moreover, EVHC will be related to future returns if extreme values of EVHC reflect the market’s inability to correctly assess the value impact of intangibles due to either limits to arbitrage, noise trading, or behavioral biases. For example, Hirshleifer (2001) points out that people are more prone to biases, and thus misvaluation is more likely, when they try to value securities for which information is sparse. Alternatively, if the value of human capital is hard to pin down, then stock prices may be noisy, making mispricing more likely. Thus, since information about the quality of human capital and the intangibles’ advantages derived from it is not readily available, it is reasonable to expect that investors may misinterpret the value impact of human capital for intangibles-intensive firms. We find that portfolios of low EVHC firms systematically outperform portfolios of high EVHC firms by an average 1.34% per month. This is primarily a small firm effect. Our asset pricing tests show that the in the case of small firms the difference in the return performance of the low- and high EVHC portfolios cannot be explained away by conventional risk factors, whereas in the case of large firms the excess returns of the arbitrage portfolio the is long low EVHC firms and short high EVHC firms is zero. Our results also indicate that EVHC is highly correlated with several mispricing measures. We conduct a number of additional tests aimed at clarifying the nature of the link of between EVHC and future returns. Overall, our results are consistent with two alternative views of EVHC: as a proxy for risk associated with firm’s reliance on human capital intangibles, or as a mispricing proxy that reflects the market’s inability to correctly assess the true value of human capital, especially for small firms. The rest of the paper is organized as follows: Section 2 describes the data selection process and provides detailed definitions of the variables used in the study with special focus on EVHC. Section 3 presents and describes the empirical results. Section 4 provides a summary and some concluding remarks.
نتیجه گیری انگلیسی
Even though intangibles are recognized as an important parameter of firm value, finance literature has not provided an answer to the question of whether, and to what degree, the stock market is able to recognize the quality of human capital, i.e., whether firms’ equity valuation is commensurate with their employees’ collective skills and effectiveness relative to that of their industry peers and competitors. In order to overcome the problem that arises from the lack of availability of human capital intangibles information in accounting statements, we devise a relative stock market valuation measure of human capital intangibles based on the ratio of the market value of equity per employee to the industry’s median market value of equity per employee. We examine whether this “excess value of human capital”, EVHC, reflects the value of human capital-related intangibles by examining whether is related to future return performance. We find that portfolios of low EVHC firms systematically outperform portfolios of high EVHC firms by an average 1.34% per month. Our asset pricing tests show that this is primarily a small firm effect. Our results also indicate that EVHC is highly correlated with other commonly used mispricing measures. Overall, our tests allow room for two alternative interpretations of EVHC as a measure of the value investors assign to human capital-related intangibles. On one hand EVHC can be thought of as a measure of risk, associated with heavy reliance on human capital. On the other hand, especially for small firms, one can think of EVHC as a measure of mispricing that reflects the market’s inability to correctly assess the true value of human capital. The average monthly return from an arbitrage portfolio that buys low EVHC stocks and sells high EVHC stocks seems sizeable, especially for small firms. However, additional tests (not tabulated here for the sake of brevity, but available upon request) suggest that after accounting for the costs of arbitrage (using the stock’s idiosyncratic volatility to measure arbitrage risk (see Doukas, Kim and Pantzalis among many others) investors may not be able to derive substantial profits from this strategy. There is also another, more general, and perhaps equally important, implication of the findings presented in this study. Namely, this paper is another example of the inadequacy of the four-factor model to capture all systematic risks. This has been demonstrated in a different context by several studies in the past.23 Our results provide support for those who question the ability of the traditional four-factor model that includes the market, distress, size and momentum factors to capture different types of risk. In this paper, EVHC, which is constructed as another price-to-fundamentals ratio, adjusted at the median industry level, appears to also behave as a persistent mispricing variable. Even though more work would be required to show that using these four factors to capture all systematic risk is impossible, our findings are consistent with the notion that the four-factor model is not explaining much beyond what it was designed around.