The rule in Hadley v. Baxendale
The main conclusion of the previous section is that none of the three damage measures are able to achieve efficiency in both the level of care and the reliance investment. The results are summarised in Table 8.4.1.
Table 8.4.1 The effects of different damage measures for breach of contract
Damage measure |
Buyer |
Seller |
Expectation damages |
Over-reliance |
Efficient, given reliance, but since buyer overinvests in reliance, this means that care is inefficiently high. |
Reliance damages |
Over-reliance |
Inefficiently low. |
Restitution damages |
Under-reliance |
Inefficiently low |
As discussed earlier, the model that we have developed in this section is similar to the basic bilateral accident model developed in Chapter 5, and yet all of the damage measures we have examined so far fail the efficiency test. The reason is that none of them replicate or approximate the efficient rules we examined in the accident law setting. The expectation measure shows promise - it induces efficient behaviour by the seller - but ultimately fails because the buyer is fully insured against all risk and so there is a moral hazard problem with this measure.
Drawing on the lessons from Chapter 5, it is reasonable to expect that some kind of due standard or negligence rule might perform well in the contract law setting. Let us investigate such a rule, which is known in the literature as the rule in Hadley v. Baxendale, after the legal case in which it was (with suitable interpretation from economists!) first proposed.1
Suppose that courts award expectation damages to buyers, but only to the extent that such expectation damages are "reasonable". In particular, suppose that the court knows the efficient level of reliance, x*, and suppose that it implements a rule which awards the buyer V(x*) in the event of a breach, irrespective of the actual level of reliance investment that the buyer has made. Let us investigate the behaviour of the parties under this rule.