سیاست سرمایه گذاری، تامین مالی داخلی و مالکیت تمرکز در انگلستان

This paper investigates whether investment spending of firms is sensitive to the availability of internal funds. Imperfect capital markets create a hierarchy for the different sources of funds such that investment and financial decisions are not independent. The relation between corporate investment and free cash flow is investigated using the Bond and Meghir [Review of Economic Studies, 61 (1994a) 197] Euler-equation model for a panel of 240 companies listed on the London Stock Exchange over a 6-year period. This method allows for a direct test of the first-order condition of an intertemporal maximisation problem. It does not require the use of Tobin's q, which is subject to mismeasurement problems. Apart from past investment levels and generated cash flow, the model also includes a leverage factor which captures potential bankruptcy costs and the tax advantages of debt. More importantly, we investigate whether ownership concentration by class of shareholder creates or mitigates liquidity constraints. When industrial companies control large shareholdings, there is evidence of increased overinvestment. This relation is strong when the relative voting power (measured by the Shapley values) of the combined equity stakes of families and industrial companies and the Herfindahl index of industrial ownership are high. This suggests that a small coalition of industrial companies is able to influence investment spending. In contrast, large institutional holdings reduce the positive link between investment spending and cash flow relation and, hence, suboptimal investing. Whereas there is no evidence of over- or underinvesting at low levels of insider shareholding, a high concentration of control in the hands of executive directors reduces the underinvestment problem.

In perfect capital markets, investment decisions are independent of financing decisions and, hence, investment policy only depends upon the availability of investment opportunities with a positive net present value (NPV) (Modigliani and Miller, 1958). In the standard neo-classical model of investment, firms have unlimited access to sources of finance and invest as long as the marginal dollar of the capital expenditure generates at least one dollar of a present value of cash flows (Tobin, 1969). Consequently, firms with profitable investment opportunities exceeding available cash flow are not expected to invest any less than firms with similar opportunities but larger internal cash flows. However, the empirical literature supports the model about the hierarchy of financing which predicts that the investment expenditure of some firms may be constrained by a lack of internally generated funds. For many firms, the cost of external capital does indeed seem to exceed the cost of internal funds. As profits are highly cyclical, the existence of liquidity constraints makes investment spending more sensitive to fluctuations in economic activity. Differing views on the riskiness of investment projects between shareholders and management and, hence, on the relevant discount rate may result in good investment projects being rejected. Underinvestment due to asymmetric information (Greenwald et al., 1984) results from the fact that the market requires—even for high quality firms/projects—a premium equal to the one required for investing in the average firm. Consequently, due to adverse selection, it may be the (relatively) lower quality projects which may seek external financing and some positive NPV projects are not undertaken at all.2Myers and Majluf (1984) have labelled the hierarchy of financing—driven by asymmetric information and/or the real direct and indirect costs of different sources of financing—the pecking order theory. Firms finance positive NPV projects in the first instance with internal financing, subsequently with debt (as the least risky form of external financing) followed by all kinds of hybrid debt with equity components and finally with external equity as a last resort. Consequently, a positive relation between investment and liquidity may result from asymmetric information because the lack of internal capital and the ‘high cost’ of external capital create an underinvestment problem. However, this positive relation may also be the consequence of an abundance of retained earnings which makes internal funds too inexpensive (from the management's point of view). In firms with insufficient monitoring mechanisms—e.g. in firms without performance-related managerial remuneration schemes, with diffuse ownership, with anti-takeover devices or with CEO dominated boards of directors—high managerial discretion may lead to considerable agency costs. In such cases, managers' interests are not perfectly aligned with those of the shareholders Jensen, 1986 and Bernanke and Gertler, 1989: managerial decision-making may be motivated by ‘empire building’ and lead to overinvestment. In this setting, managers may place a discount on internal funds and overspend by undertaking even negative NPV projects as long as there is excessive liquidity3 in the firm because managers may derive more private benefits by increasing their firm's size (Hart and Moore, 1995). The question whether or not the level of investment depends on corporate liquidity has drawn substantial attention over the last decade since the seminal paper by Fazzari et al. (1988) has rekindled the interest in the determinants of investments. However, relatively few papers test the investment–liquidity relation within a specific corporate governance framework. Notable recent exceptions are, e.g. Kathuria and Mueller (1995), Kaplan and Zingales (1997), Hadlock (1998), Gugler et al. (1999), Vogt (1994) and Cho (1998) for the US, Degryse and De Jong (2000) for the Netherlands, Haid and Weigand (1998) for Germany, and Gugler (1999) for Austria. This paper focuses on the impact of relative voting power and liquidity on investment spending in UK firms. The empirical version of the Bond and Meghir (1994a) Euler-equation model is extended by including variables capturing ownership concentration and shareholder coalition to answer the following questions. Are UK firms liquidity constrained? Does the presence of specific classes of shareholders influence the relation between investment spending and internally generated funds? Do shareholder coalitions influence the investment–cash flow relation in firms with dispersed ownership? Are companies with high leverage more liquidity constrained? The remainder of the paper is structured as follows. Section 2 presents the hypotheses and embeds them in the literature. Section 3 describes the data and explains the methodology. Section 4 discusses the results and Section 5 concludes.

The empirical literature documents that the level of internally generated funds significantly influences investment spending. The positive cash flow sensitivity of investments can result from excess cash flow which management perceives to be too inexpensive and therefore squanders in negative NPV projects. In contrast to such agency problems, the positive relation may also be the consequence of liquidity constraints which cause the company to pass up valuable investment projects if the premium paid for external financing is perceived to be too high. This paper has investigated this relation for a random sample of companies listed on the London Stock Exchange and has analysed whether the cash flow sensitivities differ for companies with financing needs and for companies with varying degrees of ownership control. To this end, the Bond and Meghir (1994a) model, which overcomes some of the drawbacks of the neo-classical and Tobin's q investment models, was extended. In addition, the model was estimated using the GMM in systems technique which avoids the estimation biases of the usual methods (like weighted least squares, and GMM in differences). For the whole sample, there was no evidence of a positive relation between the levels of internally generated funds and subsequent investment spending, or no evidence of consistent over- or underinvesting. However, companies with financing constraints seem to underinvest since their investment spending is strongly and positively related to the amount of internally generated funds. For companies in which institutions own a large amount of the voting rights, the relation between investment spending and cash flow is reduced. Whereas for companies without large share stakes controlled by industrial companies investment is not cash flow dependent, the presence of voting control by industrial companies induces a positive relation between cash flow and investment spending. This may result in either overinvestment—perhaps stimulated by industrial companies desiring to reap private benefits of control by tunnelling—or underinvestment if these large shareholders reduce the company's intention to attract external funding. Given that in the absence of concentrated control by industrial companies, investment spending does not depend on cash flow levels, the first interpretation seems the most plausible one. For the models with industrial ownership, a cash flow sensitivity is only observed for models with the Shapley values or Herfindahl indices. It seems that coalitions of a few industrial shareholders have an impact on investment policy rather than individual shareholders. Finally, there is some evidence that firms with low levels of managerial ownership suffer from underinvestment, a problem which is not present in firms where managers hold relative high stakes.