Corporate Conglomerates
Costs, Benefits and Internal Capital Markets

Corporate conglomerates have received an increasingly bad press both in the business world and in academic research.  They are deemed to be wasteful and inefficient in allocating resources. In the wake of a merger, conglomerates’ shares trade at an average discount of 13–15%. However, merger and acquisition activity proceeds relentlessly, with around 40% succeeding. This variation in performance has attracted the attention of researchers, who are now trying to understand its reasons. What are the factors that lead 40% of mergers to succeed? Is the recent merger-mania an attempt to imitate the ‘winners’?  What are the consequences of this merger wave for internal and external capital markets?

These were the questions discussed at a joint CEPR/CSEF Conference on ‘Core Competencies, Diversification and the Role of Internal Capital Markets’, which took place at the Istituto Italiano degli Studi Filosofici in Naples on 28/30 January 1999.

The existing literature largely focuses on corporate governance (how external investors can discipline both controlling shareholders and management), and on the boundaries of the firm (what should be managed inside a firm and what should be left to the markets). The papers presented in the conference mark a shift in focus from external to internal capital markets. Instead of posing the question of the external control mechanisms and boundaries of the firm, the conference focused on the mechanisms of the capital budgeting decision within organizations. The analysis concentrated mainly on the case of conglomerate corporations, where the allocation of capital across divisions is the main activity of headquarters. Other cases were also discussed, such as leverage buy out associations and venture capital.

The conference revolved around two main (related) themes: how the incentive problems that arise in organizations affect the rules used to allocate capital internally, and what are the cost and benefits of conglomerate corporations.

An immediate implication of the incentive constraints under which capital must be allocated internally is that a simple rule, such as net present value (NPV) maximization, may fail to be the efficient criterion to use for corporate headquarters. The paper by Elazar Berkovitch (Tel Aviv University) and Ronen Israel (University of Michigan), entitled ‘Why the NPV Criterion Does not Maximize the NPV’, shows how NPV criterion can fail, if divisional managers have more information than headquarters about the relative merits of alternative projects and their incentives are not perfectly aligned with shareholders’ value maximization. In these situations, alternatives, such as the Internal Rate of Return, the Profitability Index or other statistics, may produce a better allocation criterion than NPV maximization.

The tension between these conflicting objectives – eliciting information from various units within the company and providing them with incentives to exert effort – was also illustrated by the paper by Robert Gertner (University of Chicago) on ‘Coordination, Dispute Resolution, and the Scope of the Firm’. He studied how the conflict is modified by changing the allocation of control rights within the organization. In his model, two units need to coordinate on a decision such as the design of a new product. Each unit has private information about its own input. The question is, under which structure are the two units willing to share information efficiently in order to implement the value maximizing design? Gertner’s results suggest that organizations where both units report to a manager entrusted with the allocation of capital achieve more information sharing than those where there is a direct contractual relationship between the units.

In certain circumstances, however, contracts may turn out to be a preferable arrangement. In his paper on ‘Financing Mechanism and R&D Investment’, co-authored by Haizhou Huang (IMF), Chenggang Xu (London School of Economics and HIID, Harvard University) suggested that the efficiency of the venture capital mechanism in financing R&D investments may lie in its contractual foundation for soft budget constraints.

A second group of papers shed light on the costs and benefits of conglomerates and other diversified organizations such as universal banks. Understanding these costs and benefits is crucial to determining the motivations behind the current wave of mergers.

First among the benefits stands the conglomerates’ superior ability in taking advantage of deregulation and technological advances: being well-diversified across markets, this mode of organization possesses the option value of being in a market ‘before the others’. This option may be more valuable at times of great technological progress, and this can help explain the current merger-mania. This argument was offered by Arnoud Boot (Universiteit van Amsterdam and CEPR) with special reference to banking and financial conglomerates in his paper on ‘Expansion of Banking Scale and Scope: Don’t Banks Know the Value of Focus?’ co-authored with Todd Milbourn (London Business School) and Anjan Thakor. In the discussion it was pointed out that the model might be more suited to non-financial companies than to the financial sector, where technical innovation has been comparatively limited.

A second benefit of merger activity is directly related to the efficiency of internal capital markets. This theme, which connects directly with the contributions of the first set of papers discussed above, was highlighted by Zsuzsanna Fluck (Stern School of Business, New York University) in her paper with Anthony W. Lynch on ‘Why Do Firms Merge and Then Divest: A Theory of Financial Synergy’. They develop a model where, in the presence of agency costs, mergers can increase efficiency in the allocation of capital across companies. In their view, a conglomerate merger is a technology that allows the financing of marginally profitable projects which, because of agency problems, would not otherwise be financed. Interestingly, the model predicts that merger activity would not proceed unchecked; external pressure by capital markets is predicted to lead conglomerates to dispose of projects which have already been financed and are able to proceed more profitably as standalone companies.

It is generally argued that a third benefit of mergers derives from the fact that diversification creates value. Synergies that arise from the merger of two or more firms can lead to cost reductions, to higher and more stable demand, to the ability to organize activities more efficiently, and to the exploitation of a greater range of opportunities. The value of diversification is at the heart of the paper by Arnoud W. Boot (Universiteit van Amsterdam and CEPR) and Anjolein Schmeits (Washington University) on ‘Market Discipline and Incentive Problems in Conglomerate Banks’. In their model, the cost of capital reflects the company’s risk, implying that a conglomerate – and in particular a conglomerate bank – benefits from a lower cost of capital because of diversification, other things being equal. However, Boot and Schmeits show that other things are not equal: because diversification also allows cross-subsidies and free riding between divisions in a conglomerate, the cost of capital ends up holding an ambiguous relationship with risk-taking.

The unpleasant side-effects of synergies were central to the paper by Raghuram Rajan (University of Chicago), Henri Servaes and Luigi Zingales on ‘The Cost of Diversity: The Diversification Discount and Inefficient Investment’, where the attempt to exploit these synergies may actually impose considerable costs. Rajan explained that it could be very expensive to exploit synergies when the incentives of different divisions are not aligned: the friction between divisions can lead them to forfeit value-enhancing investments that would potentially make them more dependent upon each other. To persuade the divisions to proceed with investments that require them to integrate their operations, corporate headquarters need to make transfers among divisions to align their incentives. As a result, the benefit of the synergies can only be reaped at the cost of distorting the allocation of resources in internal capital markets, allocating a disproportionate amount of funds to less profitable divisions in order to ‘bribe’ them into cooperating with the others.

Further implications of the emergence of conglomerates are the extent of market power they are able to exercise and their ability to span various interrelated markets. This point is emphasized in the paper by Oved Yosha (Berglas School of Economics, Tel Aviv University and Bank of Israel), Hedva Ber and Yishay Yafeh on ‘Conflict of Interest in Universal Banking: Bank Lending, Stock Underwriting and Fund Management’. Based on the analysis of a panel data set on Israeli banks, the authors find that the hypothesis that bank conglomerates use their market power to overprice the stocks they sell to their own mutual funds (thus inflicting a cost on investors in those funds) could not be rejected.

There are essentially two questions that the conference left for further research:

First, because of time constraints in processing information a lot of capital allocation is inevitably delegated to financial intermediaries and corporate headquarters within organizations.  What we need to understand better is the extent to which the market achieves an efficient level of decentralization and how incentive problems and corporate politics disrupt the process of capital allocation. More research is needed on the design of internal organizations for the effective allocation of capital within multidivisional firms.  The scope of investigation must be broadened beyond the allocation of capital to include the allocation of human resources and other inputs. The allocation of human resources is the least studied. This is partly because little systematic research has been done to identify empirical facts on human resource management. These issues clearly go beyond the narrow boundaries of economic analysis. What emerged from the discussions during the conference is that a theory that attempts to explain capital budgeting also has to take into account issues related to the sociology of organizations. Politics, power, authority and leadership, which influence behaviour at the top of an organization, all play an important role when trying to understand internal capital markets.

Second, the relationship between internal and external capital markets needs to be investigated further. The existing literature has shown how external capital markets may impose some discipline on insiders, thus shaping corporate governance. Contributions at this conference have instead shown how organizations determine the allocation of capital across alternative uses via internal capital markets, taking the external constraints on chief executive officers (CEOs) as given. In fact, external investors can influence the appointment and removal of CEOs, therefore affecting how they allocate capital within a conglomerate company. The power and authority of the CEO within the organization ultimately rests upon these external factors. Hence the next step must be to understand how the two decision mechanisms – internal and external capital markets – interact with each other.