3.4 The incomplete-contracts approach: An empirical example

To illustrate the approach of Grossman and Hart, consider the following elaboration of the numerical example provided by Aghion and Holden (2011:183 -184) in which there is a B(uyer) and S(eller) of an intermediate good that is used to produce a final good that B sells to consumers. Both parties can make private, non-contractible investments[4] relevant to this production process:

• S can make an investment costing $5 that will make the intermediate good cheaper to produce. This investment will reduce the cost of the intermediate good from $16 to $10.
• B can make an investment costing $5 that will make the final good more valuable for consumers. This investment will increase the value of the final good to consumers from $32 to $40.

The results of different allocations of residual control rights for this production problem are summarised in Table 1. If contracts were complete, B and S would write a contract that specified both types of investment ex ante, with the result that the total surplus generated would be $20 (sale price of $40, less investment cost of $5 for S and $5 for B, less the cost of the intermediate good of $10). However, where the investments are non-contractible there are three possibilities:

• S and B continue as separate firms. Contractual bargaining about the price that B pays to S results in an agreement to split the surplus 50:50. In this case, if B invests but S does not, B will bear a private cost of $5 but obtains half of the increase in surplus generated = ($40 - $32)/2 = $8/2 = $4. B therefore will not invest since $4 is less than the investment cost. Similarly, if S invests but B does not, it will bear a private cost of $5 and obtain half of the increase in surplus of ($16 - $10)/2 = $6/2 = $3. S will therefore not invest, since $3 is less than the investment cost. When neither party invests, we have the base case for incomplete contracts where total surplus is $16.
• S and B are integrated with S acquiring residual control rights (ownership) over the entire project. In this case, the investment by S will be made, but the investment by B will not, resulting in a total surplus of $17 (Table 1) which is superior to separation of S and B.
• S and B are integrated with B acquiring residual control rights (ownership) over the entire project. In this case, the investment by B will be made, but the investment by S will not, resulting in a total surplus of $19 (Table 1) which is both superior to separation of S and B and superior to residual control rights being allocated to S.

In this example, incomplete contracts mean that production will most efficiently take place in an integrated firm owned by B.

The incomplete contracting perspective embodied in this example represents a sharp break with the earlier transaction cost-based literature on the firm. Whereas incomplete contracts imply that inefficiencies arise because it was hard to foresee and contract about the uncertain future, earlier literature tended to take a “complete contracts” perspective in which imperfections arise as a result of moral hazard and asymmetric information. The shortcoming in the latter approach is in its lack of attention to the impact that ownership and other aspects of organisational form have on the efficiency of the allocation of residual decision rights: these aspects of ownership were subsumed to the focus on the moral hazard and asymmetric information problems faced by all owners employing professional managers. Relying on transaction costs alone to generate differences in organisational form under this complete contracts approach appears highly unsatisfactory. In contrast, ownership matters where contracts are incomplete, because the rights of the owner provide the power to take decisions in relation to those matters not explicitly covered in contracts covering the use of assets, supply of inputs or service to customers (Hart, 2003:296).

In subsequent chapters of this paper, we will explore the wide range of issues to which this framework is applicable. Appreciation of the breadth of the potential applications flows from the fact that it may be applicable whenever relationship specific investments are important in the determination of economic efficiency. In its most general form, it implies that the party whose non-contractible effort, knowledge, or investment makes the greatest contribution to total surplus should have residual control rights. Further, it provides a framework for thinking about both the boundaries of state-sector ownership of assets and service delivery, but also about the potential to redefine the boundaries of state ownership of activities to allocate residual control rights for particular tasks to public or private entities depending on the value added by different ownership structures for the activity. Each of these lines of inquiry is explored below.