Efficient breach of contract - the role of damage measures

Table of Content

1. Introduction

2. Efficient Breach of Contract and Damage Measures
2.1 Incomplete Contingent Contracts and Inefficiency
2.2 Damage Measures as Substitutes of Completely Specified Contracts
2.3 Evaluation of Damage Measures
2.4 Summary

3. Critical Issues for Damage Measures
3.1 Transaction Costs, Damage Measures and Alternative Contractual Designs
3.2 The Problem of Full Compensation
3.3 Summary

4. Conclusion

References

1. Introduction

On the level of economic transactions, contractual relations have emerged over time to govern behavior of people involved in doing business in order to promote the efficient allocation of scarce ressources.^[1] Thereby, contracts create order, reduce uncertainty or transform uncertainty into risk and thus are basic premises allowing for the exchange of goods or services.^[2] However, depending on the nature of the economic transaction, parties involved in a contractual agreement may prefer to grant breach of contract if it proves to be efficient compared to performing the contract throughout duration. Hence, in order to guarantee the mutual benefit and thus, in fact, pareto efficiency of breach of contract, contractual settings have to be designed sophisticatedly to account for situations, where a contractual party may want to default and breach a contract.

Based on a paper by Steven Shavell,^[3] in the following, damage measures shall be critically discussed as efficient coordination mechanisms of interests in the event of breach of contract.

First of all, the need for contractual settings in order to promote efficient breach of contract given incomplete contingent contracting will be outlined. Then, a light shall be shed on the model of damage measures as used and introduced by Shavell. Thereafter, Shavell’s approach shall be critically discussed and some further implications will be given.

2. Efficient Breach of Contract and Damage Measures

2.1 Incomplete Contingent Contracts and Inefficiency

Since the famous article „The Problem of Social Cost“ by Ronald Coase,^[4] the fundamental role of transaction costs for the contractual governance of economic relations has received a growing attention in the interdisciplinary domain of Law and Economics. Contracts do not only involve production costs thus the costs of executing a contract, but also transaction costs, which consist of “the costs of arranging a contract ex ante and monitoring and enforcing it ex post”^[5].

Under a regime of transaction costs, complete contingent contracting, providing for each contingency that could arise during contract duration thus being pareto efficient ex ante, proves to be impossible, especially if a contractual agreement is complex.^[6] Complete contingent contracting would involve excessive contract negotiation and high information costs to verify and provide for the probability of every contigency perceivable and to ensure information symmetry between the players involved.^[7] In fact, as players are subject to bounded rationality, information may even be exclusive to one party or generally not available.^[8] As a result, contracts will not provide for any contingency and are therefore necessarily incomplete.

Based on the assumption that contracts are necessarily incomplete, an unexpected contingency may arise after a contract has been agreed on, which alters the expected value of a party involved in the contractual agreement. As a result, unexpected contingencies may induce one party to defect and breach a contract at the other party’s expense.^[9] In this particular case, pareto efficiency with respect to breach of contract is violated, as one player will be worse off as compared to the preceeding allocation.

Instancing, a simplified model shall be considered. It is supposed that seller S and buyer B join a contractual agreement. Under this contract, S is obligated to deliver a book to B, while B is obligated to pay €4. Furthermore, S has to spend €2 as transportation costs if performing. The contract price has already been transferred from B to S.

Before the book is delivered, an alternative buyer C offers seller A 8€, if S would deliver the book to C. It is supposed that this contingency was not perceived when designing the contract.

In this case, S will breach the contract as the potential gain of defaulting (€8 - €2) clearly exceeds the gain of performing the contract (€4 - €2). Contrariwise, B will be worse off as B already spent €4 as contract price and is not able to realize the full value of if the contract would have been performed, not anticipating C as an unexpected contingency. Hence, breach of contract is pareto inefficient.

In order to avoid inefficient breach of contract and ensure pareto efficiency, several contractual countermeasures may be set, such as intrinsic rewards or the possibility of renegotiation.^[10] In a paper by Steven Shavell, the vital role of an appropriate design of damage measures for breach performance is highlighted, which will be discussed in the following chapter.

2.2 Damage Measures as Substitutes of Completely Specified Contracts

If liquidated damages, that is, damage measures are appropriately designed, they will create incentives for contractual parties to behave in a way “that approximates what they would have explicitly agreed upon under a fully specified contract”^[11]. Damage measures may hence serve as substitutes for completely specified contracts and thus provide for unexpected contingencies, which would imply efficient breach of contract.

This is central to the program of Shavell’s paper, as it is the underlying proposition of his analysis. In his model of damage measures, it is assumed that a risk neutral buyer and a risk neutral seller agree on a contract for the delivery and production of a certain good or for the performance of a service. Furthermore, it is supposed that both parties have agreed on a certain type of damage measure for breach of contract. The possibility of renegotiation or partial performance is not considered in the model.^[12]

Ex post the contract is ratified, both parties will decide simultaneously about whether they want to perform or whether they wish to default. Hence, both parties will try to maximize their expected value associated with the contract.

If a party decides to perform, he will determine his level of reliance, which, according to Shavell, is defined as “actions taken before and in anticipation of contract performance”^[13]. This, for example, may include a seller’s expenditures such as production costs or a buyer’s specific investment into employee training. To determine his level of reliance, a party has to consider the probability of becoming a victim of breach as, in this case, the party may not realize the full benefits of reliance. Based on this calculation, the party will determine his level of reliance.^[14]

Now it is assumed that a party may exclusively observe a certain unexpected contingency and, based upon that, will decide to default. In this case, the party will default whenever his position, given breach of contract, will have more positive implications than if he was to stick to the contract.^[15]

Both the decision about reliance and the decision about breach are influenced by the damage measure that was agreed upon.

On the one hand, the defaulting party will only default if the difference between his position, given breach and given a certain damage the defaulting party has to pay to the other party, and his position, given performance, is non-negative.^[16] This breach set is rather straightforward. The defaulting party has to renounce the benefits of performance and pay damage depending on a certain damage measure, while receiving the benefits of breach under a certain contingency. Based on a non-negative cost benefit analysis, the party will default. Given this breach set, the damage measure hence sets disincentives for breach of contract by reducing the benefits of defaulting.

On the other hand, the relying party’s decision about his level of reliance is also influenced by the damage measure. To begin with, the relying party may not be able to observe the unexpected contingency, but he is certainly aware of the nature of the damage measure. If the relying party understands the logic of the other party’s breach set, which was outlined before, the damage measure helps to determine the probability of becoming victim of defaulting as the relying party is well aware of the amount the defaulting party has to pay to him.^[17]

[...]

^[1] See Macaulay (1963), pp. 56-62.

^[2] See North (1991), pp. 97-108; See Dequech (2006), pp. 112-120.

^[3] See Shavell (1980), pp. 466-490.

^[4] See Coase (1960), pp. 1-44.

^[5] Matthews (1986), p. 906.

^[6] See Macaulay (1963), pp. 55-58; See Shavell (1980), pp. 468-469; See Williamson (1999), p. 1089.

^[7] See Eisenberg (2000), p. 1781.

^[8] See Radner (1968), pp. 31-32; See Shavell (1980), pp. 468-469; See Simon (1985), pp. 294-295.

^[9] See Shavell (1980), p. 470.

^[10] See Murdock (2002), pp. 650-653; See Huberman/Kahn (1988), pp. 471-473.

^[11] Shavell (1980), p. 466.

^[12] See Shavell (1980), pp. 469-470.

^[13] Shavell (1980), p. 470.

^[14] See Shavell (1980), pp. 470-471.

^[15] See Shavell (1980), p. 475.

^[16] See Shavell (1980), p. 470.

^[17] See Shavell (1980), p. 470.