The theory of internal capital markets offers diverging views on the efficiency of resource distribution across a firm’s operating units. The efficient hypothesis posits that internal capital markets benefit from stronger control rights and superior information quality, which enable the CEO to make better allocation decisions.
The alternative hypothesis states that internal capital markets suffer from multi-layered agency motives of divisional managers and the CEO, who pursue their private interests via rent-seeking, empire building, or entrenching. Despite the key role of divisional executives in the theory of internal capital markets, there is little empirical evidence on what influence, if any, they have on capital budgeting and firm value of conglomerates.
We construct the most comprehensive dataset of divisional executives at the S&P 500 firms from 2000 to 2008. The rich set of managers’ characteristics enables us to evaluate their involvement in the firm via various channels, ranging from formal, such as board membership and seniority, to informal, such as ties to top management via prior employment, educational institutions, and nonprofit organizations. The broad selection of firms in our sample enables us to investigate the cross-sectional implications of divisional managers for firm value and operating efficiency. We construct a detailed network of internal connections between divisional executives and top management, applying, for the first time, social networks techniques to the study of internal capital markets. This setting allows us to provide evidence on the different hypotheses about internal capital allocation proposed in the theoretical literature.
One view, which we refer to as the agency hypothesis, is that divisional managers attempt to attract more capital to their divisions in order to extract private rents, such as personal benefits, higher status, and better job security (Scharfstein and Stein 2000; Rajan, Servaes, and Zingales, 2000). If internal connections allow managers to have greater influence on capital budgeting, this view predicts higher capital allocations to divisions of well-connected executives and a negative effect on investment efficiency and firm value. An alternative agency hypothesis is offered by Xuan (2009). Under the “bridge building” hypothesis, the CEO entrenches himself by allocating capital to unconnected divisional managers in an effort to win their support. This hypothesis predicts higher capital allocation to unconnected managers and a negative effect on investment efficiency.
Another conjecture, which we label the information hypothesis, posits that divisional executives possess valuable information about their segments. Yet the CEO cannot take advantage of this private information, because managers tend to overstate their investment opportunities due to overconfidence (Heaton 2002; Malmendier and Tate, 2002) and preferences for empire building (Jensen 1986, 1993; Stulz 1990; Hart and Moore, 1995). In equilibrium, the CEO adjusts divisional capital allocation downward to account for possible overstatement of investment prospects. Therefore, if social connections facilitate information transfer (Cohen, Frazzini, and Malloy, 2008, 2010; Engelberg, Gao, and Parsons, 2010) and reduce information asymmetry, they are likely to result in higher capital allocation to connected divisions and a positive impact on investment efficiency and firm value.
Download
PDF Ebook Divisional Managers and Internal Capital Markets
