The Regulation of Moral Hazard in Retail Transactions
In transactions where costly efforts from both the seller and the buyer help prevent selling to the unfit buyer, the unique optimal law that restores the first-best requires the seller to refund the buyer and pay a punitive penalty of the buyer's loss. If the social planner can infer from a loss whether efficient efforts had been exerted, the law with contingent punitive penalty always restores the first-best with a balanced actual budget. The law is harsher for one-stop than for specialized purchase, as a loss is more likely due to shirking. Higher budget surplus cost can make the social optimum more implementable. We study constrained optima if the first-best is unattainable or if a balanced budget is enforced. With costless efforts or unilateral moral hazard or no effort choices, no punitive penalty is needed to always restore the first-best with a balanced budget. Bilateral moral hazard rationalizes punitive penalty and determinacy of law yet complicates regulation. Our model applies widely to many goods and services and yields novel predictions.
I also study efficient disclosure by contingent non-disclosure. I show that if the sender is not always active and receivers do not know if the sender is active, then compound information can be efficiently conveyed only if some information is withheld contingently. Our theory has wide applications in fields like political economy, public policy, and law. It is optimal for the benevolent social planner to purge dissident yet useful information so as to convey the more welfare-relevant information to its citizens. Even the most liberal and transparent government should implement undemocratic policies for the citizens' own good. Transparency or disclosure laws can hurt citizens. We contribute to the economic theory of information transmission, and identify a new situation where the sender's interest is perfectly aligned with receivers', more information always helps receivers, yet it is optimal to contingently block some information away from receivers. Our theory has novel predictions and policy implications.
This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 3.0 United States License.
Rights for Collection: Duke Dissertations