136 • Chapter 4 • Integration and Its Alternatives

GOVERNANCE

In our example above, we assumed that after firm 1 acquired firm 2, the managers of firm 1 could dictate what actions the managers of firm 2 should take. But integration does not turn managers into marionettes or eliminate self-interest. The same managers at firm 2 who sought to gain at the expense of firm 1 when transacting at arm’s length may also behave selfishly when working in the same firm. It is often taken for granted that contracting inefficiencies disappear when decision makers are joined in the same organization. But this is not necessarily so. Whether integration reduces or eliminates holdup and coordination problems depends on governance arrangements. If we think of contracts as delegating decision rights and control of assets between firms, then governance arrangements delegate decision rights and the control of assets within firms.

An analogy with contracts suggests a useful way to think about governance. When independent trading partners disagree about the proper course of action or division of rewards, they rely on contracts and the courts to sort things out. When individuals within a firm disagree, they rely on the guidance and authority of the central office to sort things out. The central office must be aware that employees, like independent firms, are tempted to act in their own selfish interests. Through the judicious use of carrots and sticks (i.e., bonuses, promotions, and job terminations) the central office can tame that self-interest and get employees to act in the firm’s best interests. The success of integration therefore depends on the central office’s ability to reward and punish workers in ways that contracts cannot.

Contracts are effective when rewards and punishments can be based on objective criteria. Thus, it is important for the central office to use excellent judgment in implementing subjective criteria. (Otherwise, there would be no benefit from integration.) Moreover, the central office should be sensitive to those aspects of production that are most likely to break down because of selfishness and contractual incompleteness. Thus, the central office should reward workers who sacrifice to assure coordination of design attributes; the central office should not exploit workers who make firm-specific investments (for example, by withholding salary increases relative to workers who have more general human capital); and the central office should punish workers who hold up coworkers (for example, by demanding resources) to extract quasi-rents.

Delegation

Another critical role of the central office is delegation—determining which decisions will be made by the central office and which will be left to workers. PRT helps clarify which decisions should be delegated. Consider two types of decisions that a worker can make—decisions about how to use physical assets (e.g., equipment) and decisions about how to use human capital (e.g., the time and effort devoted to work). The central office may gain control over physical assets, but it can never gain full control over human capital—it is up to each employee to decide how hard to work. As we previously discussed, PRT concludes that the central decision-making rights for an activity should be given to those managers whose decisions will have the greatest impact on the performance of that activity. Bearing this in mind, if a manager’s investment in human capital is essential to the productive use of physical assets, then control over the physical assets should be delegated to the manager. Through delegation, the manager is encouraged to make the necessary human capital investments.

Governance • 137

We conclude that when human and physical capital are highly complementary within a given application, then delegate authority. When physical capital is complementary across applications, then centralize. For example, consider a hospital with a surgery suite. If all surgeons in the hospital can benefit from upgrades to the surgery suite, then the hospital should maintain control over upgrade decisions. If the surgery suite has specialized equipment so that it is only suitable for heart surgery, then control of the suite could be delegated to the cardiac surgery team.

Recapping PRT

To summarize, PRT says that the central decision-making rights for an activity should be given to those managers whose decisions will have the greatest impact on the performance of that activity. This leads to several possible merger scenarios:

•If the success of a merger between firms A and B depends on the specialized knowledge of the managers of firm B, then decision authority should be given to the managers of firm B. This would typically mean that firm B should acquire firm A, giving B ownership and control of A’s assets. The merger could also succeed if A acquired B, provided that A delegated decision rights to B’s managers.

•If success depends on synergies associated with the combined assets of firms A and B, such as through the resolution of coordination or holdup problems between a buyer and a supplier, then A and B should merge and the decision-making authority should be centralized.

•If success depends equally on the specialized knowledge of both firms’ managers and there are no synergies from combining assets, then A and B should remain independent.

Path Dependence

Of course, governance arrangements are not always optimal. Often, the process by which governance develops exhibits path dependence. That is, past circumstances could exclude certain possible governance arrangements in the future. For example, if the period following a merger is marked by conflict, an efficient governance structure requiring cooperation between acquired and acquiring firm managers might not be feasible. These same considerations will also apply to disintegration. One might expect a business unit that was spun off to the market to act as an independent market firm. Initially, however, managers in that unit will not be used to making decisions as an autonomous market actor and may continue to rely on associations with managers in the former parent firm. This would make the relationship between the two firms after a spinoff not a market transaction, but rather a long-term informal association, which is somewhere between being part of an integrated firm and a specialized market actor.

The path-dependent nature of the processes by which firms develop can also affect the firm’s capacity to sell the products of a unit to other downstream buyers besides itself. In Chapter 3, we suggested that market specialists could gain economies of scale by selling to multiple downstream buyers. Firms manufacturing for internal use do not typically sell excess output to other firms because this would be both a distraction and an activity for which the firm lacked the requisite skills. If a firm acquired rather than built its supply capacity, however, the situation would be different. The acquired firm would know how to sell to multiple buyers. This marketing capacity would presumably be one of the resources acquired by the parent through the