I study security design with moral hazard and two-sided limited liability when an entrepreneur can choose any distribution of project returns subject to a cost. I first characterize which return distributions can be induced by a security and when the first-best can be financed. Whenever the first-best can be financed, it is optimal and can be implemented by debt; more generally, any security that implements it pays the investor a constant amount on the support of the first-best distribution. The agency costs that may prevent first-best financing are related to the convexity of the cost function. For a given return distribution, greater convexity of the cost function increases the rent that the entrepreneur has to retain and can therefore make first-best financing more difficult. When the first-best cannot be financed, I derive second-best distributions for cost functions that are monotone in risk or depend on generalized moments. Under suitable conditions, the second-best is first-order stochastically dominated by the first-best. The optimal return distributions I characterize have finite support. Since an optimal security is uniquely determined only on, but not off, the support of the optimal return distribution, optimal securities are not unique.
No takes yet. Share an insight, caveat, or question.
Daniel Krähmer (2026) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: