Reliance damages put the buyer in the same position as they would have been had the contract not been entered into. If we assume that the buyer had initial benefits of zero, this means that the reliance measure simply returns the initial price and the buyer's reliance investment wBxB. This means that:
Note that the level of damages again depends on the level of reliance taken by the buyer.
The seller's behaviour
Suppose that the parties know that reliance damages remedy will be awarded to the buyer in the event of a breach. The expected value of the contract to the seller is:
For any xB, the seller will choose precaution up to the point where private marginal benefit equals private marginal cost:
Since the buyer would never enter into the contract if the price paid plus the cost of reliance exceeded the value of the reliance, we must have P + wBxB < V(xB). This then means that:
For any xB. In particular, the inequality in (8.19) must hold for xB = x*. This means that:
Once again, since p'(xS) is negative, and since p"(xS) > 0, this means that xReHance < x*. The seller underinvests in precaution (alternatively, the seller breaches the contract too frequently, compared to the efficient breach frequency). Intuitively, since P + wBxB < V(xB), under reliance damages the seller is not faced with the full social costs of his breach decision, and he breaches too often.
The buyer's behaviour
The expected value of the contract to the buyer under reliance damages is:
Maximising this is equivalent to maximising:
As a result, the buyer again chooses an inefficiently high level of reliance. Because wBxB is returned to the buyer in the event of a breach, the buyer ignores the loss of wBxB in the event of non-performance. This results in overinvestment.