جایگزین مدیریتی و سیاست های مالی شرکت ها با ارزش مدیریتی خاص درون زا

This paper studies financial policy, investment decisions and the threat of dismissal when managers value control and investments generate manager-specific value. A high probability of investigation focuses the manager on the profitability of replacement and therefore manager-specific value. The probability of an investigation increases when the firm enters bankruptcy. Thus, high debt levels focus the manager on investments that dissuade replacement during bankruptcy procedures. Dividends relax the manager's focus on manager-specific value since there is a lower probability of an investigation following a missed dividend. The ability to make dividend payments, however, is related to ex-post performance and can improve replacement decisions. When managerial quality is commonly known, the expected value of the firm is maximized with a combination of debt and dividend commitments. When managerial quality is hidden information, it is optimal for the combination of debt and dividend commitments to signal quality.

Since the pioneering work of Modigliani and Miller (1958) and Miller and Modigliani (1961), there has been considerable progress in developing a robust theory of corporate financial policy. However, there remain many unexplained aspects of debt and dividend policies (especially the links between the two). This paper develops some potential determinants of debt and dividend policies arising from interactions between firm-specific value, managerial benefits of control and information asymmetries. The importance of firm-specific value has long been recognized Coase, 1937, Alchian and Demsetz, 1972 and Williamson, 1985, and has increased significantly with the onset of the “knowledge economy” (in which the value of intangible assets now outweighs the value of plant, property and equipment on corporate balance sheets). An interesting dimension of firm-specific value is its relation to the study of potential agency problems that can arise from the separation between ownership of and control over corporate resources. A potentially costly agency problem can arise when managers become concerned with protecting their positions. In addition to poison pills (see Jensen and Ruback, 1983), managers may undertake investments fostering firm-specific value that is tied directly to themselves, i.e. manager-specific value (Shleifer and Vishny, 1989). There is considerable evidence suggesting that managerial decisions are related to concerns with manager-specific value. Mitchell and Lehn (1990) find that managers alter the set of assets under their control (i.e. restructure) in response to a takeover threat. Dennis et al. (1997) and Lang et al. (1995) provide similar evidence suggesting that managers become concerned with their comparative advantage during periods of financial distress or takeover threats. Weisbach (1995) finds that divestitures are related to managerial replacement and that incoming CEOs often reverse the strategies of previous CEOs, and Dennis and Dennis (1995) find that competition for control increases significantly when incumbents are forced out. The influence of the threat of dismissal on managerial decisions has been related to disciplinary actions by the board of directors (Weisbach, 1988), debt commitments (Gilson, 1988), takeovers and voting rights Linn and McConnell, 1983, Roe, 1990, Stulz, 1988 and Pound, 1991. More recently, researchers have studied the potential to improve managerial incentives by reforming bankruptcy laws Jensen, 1991 and Bradley and Rosenzweig, 1992, takeover laws (Comment and Schwert, 1995), proxy rules (Karpoff et al., 1996), and board requirements (Brickley et al., 1995) to increase the threat of dismissal. Our paper contributes to this literature by extending the analysis back one period, and considering the incentives for ex-ante investment strategies that develop manager-specific value. We develop a model in which the ex-ante incentive to pursue manager-specific value is influenced by the ex-post threat of dismissal. Ex-post replacement decisions must be dynamically consistent, and are therefore based on ex-post estimates of the profitability of replacement. These estimates depend on the information that is available ex-post, which depends on the intensity with which ex-post value is investigated and therefore the firm's ability to meet its financial commitments. The link between financial commitments and managerial incentives is related to Jensen's (1986) free cash flow theory and Harris and Raviv's (1990) informational theory of debt, in which default causes an investigation that improves ex-post decisions. In our paper, default also causes an investigation that improves ex-post decisions. However, bankruptcy investigations are also linked to the manager's ex-ante investment choice, since investments that develop manager-specific value decrease the ex-post profitability of managerial replacement and “hold-up” replacement. This limits the optimal level of debt, since high debt levels focus the manager on surviving bankruptcy and exacerbate the hold-up problem (the incentive to choose investments with excessive levels of manager-specific value). Interestingly, dividends can substitute for higher debt commitments without the adverse investment incentive. This is because the probability of an investigation is lower if a dividend payment is missed. In addition, the ability to make dividend payments is related to the ex-post performance of the firm, which also affects the ex-post profitability of replacement. Thus, augmenting lower debt commitments with a dividend commitment improves ex-post decisions without affecting ex-ante investment, increasing the expected value of investors' claims. The model therefore produces the commonly observed mix of debt and dividend payments that is sub-optimal in the theories of Jensen (1986) and Harris and Raviv (1990). The advantage of dividends relative to debt contributes to the literature studying the persistence of dividends despite lower taxed substitutes (debt payments are taxed lower due to the deductibility of interest; see the “capital structure puzzle” (Myers, 1984)). The literature studying the “dividend puzzle” (Black, 1976) typically focuses on share repurchases as a lower-tax substitute for dividends. The advantage of dividends in our model can be lost if repurchases are used since disproportionate repurchases increase the incentive to investigate value, just as in Brennan and Thakor (1990). Our model contributes by illustrating potential investment distortions in addition to the purely redistributive effects illustrated by Brennan and Thakor. The equilibrium financial policy in our model is determined jointly by the manager (who proposes financial policy) and the board of directors (who oversees the manager and must approve his proposal), similar to practice. The financial policy implemented depends on whether managerial quality is asymmetric information. When managerial quality is commonly known, all firms adopt a combination of debt and dividend commitments since the board can clearly determine that this financial policy maximizes value.1 With asymmetric knowledge of quality, however, expected value is maximized when financial policy is used to signal quality, since knowledge of managerial quality improves replacement decisions. In this case the board allows the manager greater discretion over the financial policy choice, and the good manager signals his quality by augmenting his debt commitment with a higher dividend commitment. This is costly to the manager since, ceteris paribus, he obtains higher control benefits when free cash flow is retained in the firm. However, signaling is beneficial to the manager since it reduces the probability of replacement. For a good manager, signaling also increases monetary compensation (his share of firm value) since the profitability of replacement is lower for a good manager. In contrast, signaling reduces the monetary compensation of a bad manager since replacement decisions are then based on incorrect information. The higher cost for the bad manager to signal implies that a signaling equilibrium can obtain. Interestingly, the signaling equilibrium also requires the combination of debt and dividends to be “hard” enough to deter mimicking, yet “soft” enough to dissuade the managers from pursuing investments that develop excessive manager-specific value. Ravid and Sarig (1991) also consider the ability of debt and dividend commitments to serve as signals. In Ravid and Sarig, debt and dividend commitments are perfect substitutes as signals, and the optimal mix is that which minimizes the cost of signaling. Our analysis differs from theirs in that we link financial commitments to ex-ante investment choices and we incorporate the differing hardness of debt and dividends commitments. This leads to a benefit of dividends (discussed above), which is not present in Ravid and Sarig. This benefit affects the mix of (hard) debt commitments and (soft) dividend commitments that is required to obtain a signaling equilibrium, as above. This also implies that it can be optimal to maintain dividends at a strictly higher marginal cost, as found in the dividend puzzle literature (in contrast to Ravid and Sarig where the optimal mix equates marginal costs). The ex-post information structure in our paper is similar to the security design literature. Townsend (1979) and Gale and Hellwig (1985) provide pioneering work in this area, focusing on optimal verification (investigation) strategies when ex-post earnings are asymmetrically observed and can be appropriated by manager-owners. Our paper focuses more on the separation of management and ownership, such that the manager's manipulation incentives are caused by his desire to maintain private control benefits (rather than appropriating funds). In addition, we focus on dynamically consistent investigation decisions when the cost of investigating the managerial position is related to the firm's ability to meet its financial commitments. In contrast to the papers above, firm value is not maximized with costless investigation. This is due to the managerial hold-up (entrenching investment) problem considered here. Our paper suggests that securities can optimally combine soft and hard commitments to produce an ex-post probability of investigation that induces efficient ex-ante investment choices. Our paper also illustrates how this combination can be designed to mitigate asymmetric knowledge of ex-ante managerial quality. The remainder of the paper is organized as follows. The next section presents the model and develops the links between investment decisions, financial policy, and managerial quality. Section 3 extends the analysis to incorporate asymmetric knowledge of managerial quality. Section 4 discusses the robustness of the results to extensions. Section 5 discusses empirical implications and Section 6 concludes.

Considerable attention has been given to the managerial incentives associated with debt commitments. Many have advocated efficiency gains that can result from improved ex-ante incentives (e.g. higher effort or fewer perquisites) and improved ex-post decisions (e.g. discontinuance of sub-optimal strategies). This paper illustrates how managerial discretion over ex-ante investment strategies may limit the efficiency gains associated with hard financial commitments. Although high levels of debt lead to more efficient continuation decisions (through investigations in the event of default), they may also distort the manager's incentive to pursue ex-ante investments that better suit an investigation. In particular, it is more efficient to continue with managers that have more firm-specific value. Thus, a high probability of bankruptcy increases a control-oriented manager's interest in investments that develop high levels of manager-specific value. This possibility limits the ability of debt commitments to improve exit decisions and foster creative destruction as argued by Jensen (1993). Indeed, high debt may lead to excessive manager-specific value and, ultimately, excessive destruction. This possibility is consistent with the recent empirical investigations of Hotchkiss (1995), who finds that firms retaining their management through chapter 11 continue to perform poorly and Dennis and Dennis (1995), who find similar results for firms that avoid bankruptcy. In contrast to most agency models of financial policy, our model considers the optimal mix of debt and dividend commitments. The substitution of dividend for debt commitments can mitigate incentives to pursue entrenching investments and provide information that can improve ex-post replacement decisions. Moreover, the substitution of share repurchases does not provide the same advantage as dividends. Thus, our paper also provides a new contribution to the literature studying dividends. We find that the optimal combination of financial commitments should be designed to maximize the expected profits from replacement while accounting for managerial incentives to pursue entrenching investments. When managerial quality is common knowledge, this involves each firm setting a relatively low debt commitment augmented with a dividend commitment. When managerial quality is asymmetric information, allowing managers greater discretion over financial policy can lead to a signaling equilibrium where better managers signal with higher financial commitments. Signaling increases the expected value of claims since knowledge of managerial quality again improves ex-post replacement decisions.