مالکیت محلی به عنوان اطلاعات خصوصی: شواهد در نظارت نقدینگی تجارت کردن

We investigate how ownership patterns affect the way the firm is monitored, the liquidity of its shares, and its stock price. We show that informed ownership improves governance and induces value-enhancing decisions (less over-investment and fewer but better acquisitions). At the same time, it increases the adverse selection discount required by less informed investors to trade, reducing the firm's liquidity. Both effects are impounded in the stock price. This explains why ownership seems to be unrelated to performance. Informed investors affect prices in opposite directions: monitoring would raise prices, but the lower liquidity induced by their presence would reduce them.

Recent corporate events have reignited the debate on both the role of corporate governance and the best way to enhance it. Central to this debate is the modern corporation's separation between ownership and control. The idea is that at least some monitoring by informed shareholders is necessary to prevent self-interested managers from undertaking suboptimal decisions, which implies that there should be a relation between a firm's shareholding composition (i.e., ownership structure) and its performance. This implication has been extensively investigated in the literature (e.g., Morck, Shleifer, and Vishny, 1988; McConnell and Servaes, 1990; Holderness, Kroszner, and Sheenan, 1999; Himmelberg, Hubbard, and Palia, 1999). Unfortunately, the endogenous nature of ownership structure (Demsetz, 1983; Demsetz and Lehn, 1985) complicates the analysis of this issue. Equilibrium ownership patterns depend on their relative costs and benefits. Given the lack of a suitable identifying restriction, however, the literature has been unable to produce conclusive evidence on this issue. Thus, while the current view argues that no relation exists between ownership and performance (Demsetz and Villalonga, 2001), it falls short of analyzing the underlying economic story that describes the costs and benefits of ownership. Central to any such story is the role of information. In an important contribution, Kahn and Winton (1998) show that there is a trade-off for an informed institution between profitably trading on its private information (“speculation”) and using that same information to monitor the firm (“intervention”). The choice between these two actions depends on the size of the informed shareholder's stake. For instance, high stakes provide an incentive to monitor the firm; however, they make uninformed shareholders require an adverse selection premium to trade in the firm's shares. The presence of such stakes therefore has observable implications for both the way the firm is run and the liquidity of the firm's shares. Our paper uses a proxy of informed investment to study the relations among the existence of informed shareholders, corporate governance, and stock liquidity, as well as their implications for stock prices. Specifically, we measure the degree of informed investment in a particular stock as the fraction of a mutual fund's investors that are headquartered nearby. This metric comes from Coval and Moskowitz (2001), who find that US mutual funds earn substantial abnormal returns in their nearby equity holdings and that the amount of local investment in a stock is positively correlated with the stock's expected return. Indeed, Coval and Moskowitz argue that local mutual fund ownership may “offer a unique method of identifying... perhaps the first set of seemingly informed investors.”1 More recent research is consistent with this conclusion,2 and shows that local fund managers, who are both informed and more proximate to the company headquarters, are in a better position to influence the firm's management in governance-related issues. The important feature of the above measure is that the main determinant of proximity, namely the place where the investor is located, is reasonably exogenous. We can therefore generate a set of instruments to address any potential residual endogeneity. In summary, by using local ownership as an explanatory variable we obtain an identifying restriction that enables us to study the economic implications of informed investment. Using local ownership as a proxy for the amount of private information, we empirically test the mechanism at work in Kahn and Winton's (1998) model. Our main testable hypothesis is that, as the stake of local (privately informed) ownership increases, we should observe a higher amount of monitoring, that is locally held firms should exhibit a higher quality of corporate governance (Hellwig, 2000). Simultaneously, the presence of informed investors should increase the adverse selection discount associated with the trading of the stock, which implies a decrease in the liquidity of the firm's shares. To test this intuition, we construct data on the amount of local ownership (defined as holdings of mutual funds located in a 100 km radius of the firm's headquarters) for a broad panel of US firms for the 1990–2002 period. We measure the quality of the firm's governance using the Governance Index of shareholder rights compiled by Gompers, Ishii, and Metrick (2003), and the liquidity of the stock using the illiquidity measure proposed by Amihud (2002). We find that locally held firms are associated with stronger shareholder rights (higher quality of corporate governance) but more illiquid shares. This effect is economically and statistically significant, particularly on the impact on illiquidity: a one-standard deviation in increase local ownership roughly doubles the observed price impact of a trade. In addition, we find that the impact of informed investors on governance is stronger in companies for which there is more public information available (large firms, firms with greater analyst following, and firms with higher idiosyncratic volatility), which provides additional evidence in favor of the information story. The theoretical framework we use allows us to offer an economic interpretation to the ownership-performance debate. An increase in informed holdings has two opposite effects: on the one hand, valuable monitoring activities increase the value of the firm and boost prices; on the other hand, uninformed investors require a premium to trade locally held firms, making them less liquid and thereby decreasing prices (Amihud and Mendelson, 1986; Amihud, 2002). Once these two effects are properly taken into account, ownership should have no effect on prices. These predictions are exactly what we observe in the data. Moreover, our econometric approach provides strong evidence of causality from local ownership to quality of corporate governance and stock illiquidity, and no evidence of reverse causality in the other direction. We complement the findings above along two further dimensions. First, we investigate whether shareholders’ information affects corporate investment policies. We find that locally held firms exhibit lower investment, and they make fewer but better acquisitions. Furthermore, local investors demonstrate a preference for geographically proximate acquisitions, and these acquisitions generate superior returns. These results confirm that indeed the increase in governance brought by local ownership is associated with value-increasing decisions. Second, we focus on gathering evidence on the possible channels through which local owners make themselves heard. We show that local shareholders are more effective in improving governance in cases in which they have a stronger bargaining position vis-à-vis the firm's managers (firms that are more equity dependent), or other forms of shareholder control are weaker (firms with a smaller fraction of block holders, or firms in which executive equity incentives are lower). Our paper makes four main contributions. First, methodologically, we focus on a measure of informed shareholding that is less subject to the usual criticisms of endogeneity and we control for residual endogeneity. Building on existing theory, we use local ownership as an identifying restriction to obtain an empirical decomposition of the effects of ownership structure that is absent in the extant literature. In doing so, we bring to the data the mechanism that determines the trade-off between the two components of the impact of insider behavior: gains from monitoring and gains from trading. Second, to our knowledge, our paper is one of the first attempts to understand the corporate finance consequences of the geography of investment, an area that has recently been the focus of increased attention by the finance literature. Our findings are consistent with the idea that (at least among the type of sophisticated investors we analyze here) investing locally is an inexpensive way to acquire information. However, if local ownership is valuable, why don’t we observe (even) more of it in equilibrium, as Coval and Moskowitz (2001) suggest? The answer is that the information the local investors obtain comes at a cost, specifically, the cost of making their stakes more illiquid! Our paper suggests that, in addition to generating potentially adverse diversification effects, investing locally may have liquidity consequences that counteract the informational benefits. Third, we contribute to the understanding of liquidity. O’Hara (2003) points out that both the existence and the source of common liquidity factors have not yet been satisfactorily identified. We show how the quality of corporate governance may change jointly with liquidity, with both affected by the same driving factor — private information. Indeed, if private information increases both the quality of a stock's corporate governance and its degree of illiquidity, the quality of corporate governance is an additional element that alters any direct estimate of the relation between stock prices and liquidity that does not properly account for it. Finally, our paper provides useful normative insights to the debate on the trade-off between liquidity and control. Regulatory measures (such as Regulation Fair Disclosure introduced in late 2000) designed to level the playing field between investors reduce the informational advantage of some investor groups, thereby increasing liquidity. However, these same measures might reduce the quality of corporate governance because in equilibrium informed traders will lower their holdings to adjust to the new trade-off between informational benefits and liquidity costs, which in turn will reduce the amount of socially valuable monitoring. Our results therefore suggest that efforts to increase the market's liquidity must simultaneously consider measures to provide incentives for economic agents to engage in monitoring activities. The extent to which this objective can be achieved via legal constraints (such as the Sarbanes-Oxley Act of 2002) remains to be seen. The impact of investor location on the issues of liquidity and governance has received very little attention. An exception is Dahlquist Pinkowitz, Stulz, and Williamson, (2002), who, in an international context, link the geography of investment to corporate governance measures across different countries. They suggest that the control premium associated with holding shares in countries with poor investor protection can partly explain the home-bias phenomenon of overweighting domestic shares (see Lewis (1999) for a survey of the home-bias literature). The remainder of the paper is developed as follows. Section 2 lays out our main testable hypotheses. Section 3 describes the sample and the variables we use. Section 4 analyzes the impact of local ownership on a stock's governance characteristics and on the illiquidity of the firm's shares. Section 5 lays out the implications of these findings for firm valuation, in the context of a simultaneous equation model. Section 6 investigates the relation between local ownership, corporate investment policies, and the merger behavior of firms with local shareholders. Section 7 focuses on evidence concerning the channels through which local owners can influence firm governance. Section 8 discusses our results. A brief conclusion follows.

Our paper uses local ownership as a measure of informed shareholding that is less subject to the usual criticisms of endogeneity. This measure provides an identifying restriction that allows us to empirically disentangle the effects of ownership structure. The evidence presented in this paper is consistent with the idea that geographical proximity is an inexpensive way to obtain information about a firm. Local investors can gather valuable information about the firm, which they can use to influence the firm's course of action. However, outside investors will recognize this informational ability and, as a consequence, the liquidity of the stock will suffer. In their paper, Coval and Moskowitz (2001) conclude that it is puzzling that so few funds actually “go local” by investing in stocks in which they have informational advantages. Our paper suggests that, in addition to the underdiversification motive that they mention, the liquidity costs of such strategy will weigh on the valuation of the shares. In equilibrium, the observed ownership level of local investment adjusts to reflect the trade-off between these respective costs and benefits.