آیا سهامداران کوچک به حساب می آیند؟

We hypothesize that age similarity among small shareholders acts as an implicit coordinating device for their actions and, thus, could represent an indirect source of corporate governance in firms with dispersed ownership. We test this hypothesis on a sample of Swedish firms during the 1995–2000 period. Consistent with our hypothesis, we find that compared with shareholders of differing ages, same-age noncontrolling shareholders sell more aggressively following negative firm news; firms with more age-similar small shareholders are more profitable and command higher valuation; and an increase (decline) in a firm's small shareholder age similarity brings a significantly large increase (decline) in its stock price. The last effects are more pronounced in the absence of a controlling shareholder.

he corporate finance literature suggests that dispersed shareholders leave the company at the mercy of the managers who can expropriate from the firm's owners at will. This literature stresses the role of controlling shareholders as the main monitors of managers and, therefore, as key determinants of firm value (Holmstrom and Tirole, 1993, Bolton and von Thadden, 1998a, Bolton and von Thadden, 1998b, Kahn and Winton, 1998, Noe, 2002 and Faure-Grimaud and Gromb, 2004). In general, small noncontrolling and nonstrategic shareholders are assumed not to monitor, as each small shareholder has little power and no incentive to engage in monitoring. At the same time, standard asset pricing models predict that the actions of every shareholder, including those of the small shareholders, could affect the price of the stock (Hong, Kubik and Stein, 2004). While we are not aware of any study that explicitly shows how small shareholders can directly condition the behavior of the manager and the firm's corporate policies, small shareholders as a group could play a critical role. Suppose that a large number of small shareholders coordinate the timing of their stock sales. Such action undoubtedly has a strong effect on the stock price, an effect that could last for a while and, hence, would certainly get the attention of the manager, especially when his compensation is strongly linked to the stock price. Given that the cost of coordination among many small shareholders is prohibitively high, this channel of corporate governance has been left largely unexplored by the literature. We argue that such coordination could arise unintentionally if small shareholders have common features that drive their information processing and stock sales. Thus, a high degree of similarity among the firm's current small shareholders could translate into correlated selling behavior that resembles the behavior of a large shareholder, even without deliberate coordination. That is, small shareholder similarity could lead small shareholders to also have a role in disciplining managers, even if these shareholders are neither more strategic nor more capable of monitoring management than usually assumed. For instance, these shareholders could be noise traders who happen to react to news in the same way without any underlying model (behavioral setting), or they could be rational traders who happen to interpret news in the same way (differences of opinions setting). This paper tests whether such unintentionally coordinated actions constitute a channel through which small shareholders as a group can discipline managers. The basic threat for managers is that if they disappoint a large group of similar small shareholders, those shareholders would take a “Wall Street walk” by selling the stock at the same time (effectively selling like a large shareholder would), bringing about a sharp drop in the stock price. The implicit threat of a shareholder sell-off makes it more expensive for equity-incentivized managers to engage in value-reducing activities. The crucial parameter here is the degree of coordination, which we conjecture to depend on the degree of similarity among noncontrolling shareholders. Our argument hinges on managers' awareness of the degree of similarity among the firm's shareholders. In most countries this information is not directly accessible to the manager. In Sweden, however, Central Security Registry (Värdepapperscentralen AB, or VPC) collects data on holdings of Swedish companies. SIS Ägarservice AB uses semiannual snapshots of these data (as well as their own proprietary data on voting pacts, family connections, trusts, strategic shareholdings via foreign holding firms, etc.), and sells these data to firms and other interested parties.1 This dataset contains information on practically all shareholders of all listed firms, including their age, location, and other demographic characteristics. We use this unique dataset of Swedish firms to test our hypotheses. More specifically, using the Swedish dataset for the 1995–2000 period, we create measures of shareholders' similarity based on age, wealth, and location. While we do not know a priori which traits are more important, age seems a likely candidate, as different cohorts are exposed to different fads and investment climates. For example, shareholders who have lived through a long bear market may react differently to information than shareholders who have experienced only a stock market boom.2 Age similarity is also related to the formation of social networks, which could facilitate the sharing of information and opinions and thus lead to coordinated actions. For each firm, we measure the degree of similarity across all its individual noncontrolling shareholders. To show that our argument has support in the data, we need to perform the following sequence of tests. First, we must show that more similar individual shareholders do act in a coordinated fashion in their selling decisions, especially in response to bad news, which is essential to attract the attention of managers. Next, we must show that managers react to this information. That is, we must provide evidence that firms with more similar small shareholders exhibit higher profitability. Finally, we must show that the markets value the similarity of small shareholders, all else equal. We perform each of these tests. The results suggest that small shareholder similarity acts as a (hitherto unexplored) driver of corporate value creation.3 We start by showing that more age-similar small shareholders tend to sell less under normal conditions. This suggests that they tend to hold on to a stock longer. This makes them an identifiable and persistent group. At the same time, we show that firms with a higher percentage of age-similar small shareholders experience larger sell-outs following bad news. A negative surprise shock translates into 11% higher sales for stocks with above-median homogeneity of common shareholders. As predicted, this behavior is asymmetrical: Shareholder similarity does not play any role in shareholders' response to positive news, and buying behavior is not affected either. This is because share ownership itself introduces the asymmetry; that is, shareholders are more likely to sell than to buy. We do not find much effect of wealth similarity. In robustness tests we show that individual small shareholders tend to show more herding in their investment decisions with investors in their own age cohort than with investors from other cohorts. Overall, these findings suggest that age-similar small shareholders tend to react more aggressively to bad news about the firm and that this threat of unintentionally coordinated action is persistent. Next, we relate small shareholder similarity to firm profitability and value. Here we must address the potential endogeneity of ownership structure (Demsetz, 1983 and Demsetz and Lehn, 1985). Relying on the findings on home investment bias in Coval and Moskowitz, 1999 and Coval and Moskowitz, 2001, Huberman (2001), and Hong, Kubik and Stein (2008), we exploit the location of shareholders as an identifying restriction. The availability of complete information on the location of all shareholders for each firm allows us to devise a unique identifying strategy that pins down the exogenous component of the shareholder structure. We find that firms with small shareholder similarity that is one standard deviation greater than the average display 14.5% higher than average Tobin’s q which is about 10% of the total standard deviation of Q. Moreover, their profit margins are 9.5% higher than average [4.4% and 16%, respectively, for return on assets (ROA) and return on equity (ROE)]. When we repeat this exercise on the subsample of investors born in the same country in which they reside, we obtain similar results. Finally, we examine the effect of annual changes in shareholder similarity on subsequent stock returns. The way the similarity measures are constructed allows us to say something about the source of the change, especially for the large changes. A large decline in similarity is unlikely to come from a random event, as this would imply that a large number of dissimilar shareholders randomly chose to sell together, which is unlikely. Thus, the larger the decline, the more likely it is that the decline was driven by a coordinated sale of the similar shareholders. That is, similar shareholders (e.g., shareholders belonging to the same age group) interpreted the news in a similar way and sold the shares en block. A significant increase in similarity is likely due to coordinated buying, i.e., an event that appealed to many similar shareholders. For example, it could be driven by an article in a publication that appeals to this age group. We show that firms that experience an increase in shareholder similarity observe positive returns. More specifically, a one standard deviation increase in similarity is associated with a 2.18% increase in annual stock return. Also, when we form five portfolios of firms based on the change in the degree of shareholder similarity and compare the performance of these portfolios over the subsequent year, we show that a value-weighted portfolio that is short the most-declining similarity quintile and long the most-increasing similarity quintile generates a 1.6% monthly return. The α of the four-factor model run on this portfolio is 1.6–2% per month, and it is highly statistically significant. This result is not an anomaly as information on the change in similarity is not available to the market at the time of portfolio formation, but it illustrates the magnitude of the effect. Taken together, our findings provide evidence that small shareholders could, unintentionally, have a significant governance role if they are sufficiently similar. We can interpret these findings in light of two non-mutually exclusive hypotheses. The first posits that local ownership increases shareholder similarity. Managers could be subject to greater peer pressure from firm shareholders that live close by (e.g., Kandel and Lazear, 1992). The higher the fraction of local shareholders, the more that pressure from the local community acts as a disciplining device. In addition, the more similar the local shareholders are, the less the manager (himself part of the local community) wants to disappoint them. Local connections, social interactions, and local culture impose costs on cheating one's neighbors. The alternative hypothesis posits that managers (or, to some extent, controlling shareholders) whose objective function is explicitly or implicitly linked to the equity price do not want to disappoint small shareholders for fear of the price impact of a massive sell-off. Several recent papers explore the theory of disciplining managers by “voice through exit”. Admati and Pfleiderer (2009) show that a large minority shareholder could wield a certain amount of power over the manager due to his ability to exit the company and depress its stock price. Edmans (2009) argues that block holders monitor the firm's fundamental value and sell their stakes upon negative information, thus making prices reflect fundamental value and affecting managers' incentives. Edmans and Manso (2011) argue that competition among smaller shareholders (who hold sizable positions) can affect managerial decision making by coordinating exit or entry in some circumstances. In both cases, shareholders behave strategically, choosing to become more informed and more active depending on the scenario. We propose a complementary argument. Our results suggest the existence of a novel channel through which shareholder characteristics affect the value of the firm even though the shareholders are small and act completely nonstrategically. In particular, small shareholder age similarity acts as a coordination device that replaces strategic consideration. Lack of strategic consideration acts as a commitment device to sell, thus strengthening the impact of age similarity. Miller (1977) argues that higher heterogeneity of opinions should increase the price of the stock, as short positions are more expensive. An extensive recent literature tests this hypothesis by using the dispersion of analysts' forecasts as a proxy for opinion heterogeneity (e.g., Diether, Malloy, and Scherbina, 2002). One could argue that higher shareholder similarity should be associated with lower heterogeneity of opinions, leading to lower stock prices for firms with more similar shareholders. The fact that we find the opposite result does not reject this hypothesis but instead indicates that the magnitude of this effect is lower than that of the disciplining effect. Our paper also contributes to the literature that relates shareholder composition to firm performance (e.g., Morck et al., 1988, McConnell and Servaes, 1990, Himmelberg et al., 1999, Holderness et al., 1999, Franks and Mayer, 2001 and Franks et al., 2001). While the extant literature focuses mostly on large/controlling shareholders, we show an important role played by small shareholders, which indicates that the entire ownership structure could affect firm performance. Moreover, our unique data yield an identifying strategy that pins down the exogenous component of the shareholder structure, which is not an easy task in this literature. Finally, our results contribute to the debate on the extent of disclosure of shareholders' information. If knowledge of shareholder characteristics positively affects a manager's incentives, especially in firms with dispersed ownership, then perhaps such information should be widely disseminated in the U.S. and the U.K., where such firms are a norm. This issue requires more study. The paper is structured as follows. In Section 2, we develop our testable hypotheses. In Section 3, we describe the data and the construction of our measures of similarity. Section 4 relates shareholder similarity to shareholders' reactions to news. In Section 5, we relate shareholder similarity to firm profitability, and in Section 6 we study how similarity affects stock price. A discussion of the results is provided in Section 7. A brief conclusion follows in Section 8.

In this paper, we study how a firm's shareholding structure affects its financial and operating performance. We argue that the degree of shareholder similarity affects firm value. In particular, similar shareholders are better able to informally coordinate their actions, especially sales of stocks following negative news about the firm, and thus similar shareholders can play a monitoring role that translates into higher profitability and higher stock prices. We test this hypothesis by using a novel dataset containing information on all the shareholders for each firm in Sweden over the past decade. We construct a proxy for similarity based on age. We show that similar shareholders react in the same way to firm news and that greater similarity among shareholders increases firm profitability and returns. This evidence is consistent with the existence of a channel through which the distribution of shareholders affects firm value. Our findings thus raise a new set of questions in asset pricing and corporate finance regarding the path dependency of shareholder composition and its effects on management and firm valuation.