Governance in the Wild: a Theory of State Vs. Private Firms Under Weak Institutions

Abstract : We study a model in which an agent may sell essential inputs to a private or a state firm. We show that the two firms rely on different “social contracts” with the government to sustain production. In the state firm the government must promise not to expropriate the seller and to pay for his inputs. For the private firm to be sustainable, the government must promise not to expropriate the seller and the firm’s owner, and to enforce their contracts. We show that it is harder (easier) for the government to credibly make the promises needed to sustain the private firm, relative to those needed to sustain the state firm, when political constraints on expropriation and contract enforcement institutions are weak (strong). Our model explains why privatizations in developing and transition countries with stronger pre-existing institutions have succeeded, while those in countries with weaker institutions have failed. More broadly, our model provides a theoretical framework to understand the relationship between institutions and the organization of production.

---

Infrastructure and Institutions: Lessons from History

Abstract : Many economic studies quantify the effects of infrastructures on income and welfare in the past, with the aim of informing policy makers today. A different group of historical studies examines how infrastructures came about, how they were owned and regulated, and how they performed. While diverse in their message, the fundamental role of institutions is a common theme. This paper explores how democratic institutions affected the railway sector in more than 20 countries between 1870 and 1912. The results suggest that formalization of democracy (de jure institutions) tended to encourage network expansion, increased company participation, lowered construction costs, and increased rates of return. Greater democratic participation (de facto institutions) generally had the opposite effect and led to lower average passenger fares with higher freight rates. More generally the analysis suggests a focus on institutions and its role in the infrastructure sector offers a different perspective and lessons for current development.

---

A Model of Censorship and Propaganda

Abstract : A fundamental constraint of information design is that propaganda must be believed to be effective. The presence of outside information tightens this constraint. We extend the canonical model of Bayesian persuasion to endogenize the presence of such information. The sender (government) chooses both the precision of an outside signal—it chooses a level of censorship—and a probability distribution over propaganda messages, where the distribution can be conditioned on the outside signal. The government bears an opportunity cost of censorship, in that the government relies on outside information to decide whether to repress rather than persuade. In equilibrium, the government employs both censorship and propaganda, but only if the citizen is ex ante close to indifferent between taking and not taking the action desired by the government. In this region of the parameter space, censorship and propaganda are perfect substitutes.

---