Economic and Political Analysis of Institutions and Organizations

Paper Session

Saturday, Jan. 7, 2017 8:00 AM – 10:00 AM

Swissotel Chicago, Zurich C
Hosted By: Society for Institutional and Organizational Economics
  • Chair: Robert Gibbons, Massachusetts Institute of Technology

The Emergence of Weak, Despotic, and Inclusive States

Daron Acemoglu
,
Massachusetts Institute of Technology and NBER
James A. Robinson
,
University of Chicago

Abstract

This paper studies the evolution of state institutions and their relationship to civil society. The state is modeled as being controlled by a group which can invest resources to increase the power of the state, while civil society (the rest of society) can also invest to increase its defensive power. The powers of the state and civil society determine both the size of the national pie and its division, and the dynamic game between these two actors determines how state-civil society relations evolve over time. We show that, under the assumption that building the power of either the state or civil society is more difficult starting from very low levels, this dynamic game has three asymptotically stable steady-state equilibria. One corresponds to a despotic state which is relatively powerful against a very weak civil society; one to a weak state which is very weak being strongly constrained by a relatively powerful civil; and one, which we refer to as an inclusive state, where state and civil society have equal powers. In this last case, notably, the strength of the state is greater than in the despotic state equilibrium, but it is constrained by civil society. The dynamics of investments converging to this steady-state equilibrium have the flavor of a “red queen effect” whereby both sides are choosing high levels of investment to keep up with each other.

Policy Advice in a Complicated World

Steven Callander
,
Stanford University
Nicholas Lambert
,
Stanford University
Niko Matouschek
,
Northwestern University

Abstract

The internal organization of political and economic institutions is often predicated on enhancing the flow and use of expert information. In modeling expertise, however, the canonical representation of expertise takes a very simple form; namely the expert is presumed to know a single piece of information that the non-expert does not. In this paper we argue for a richer conception of expertise in which the expert's advantage is knowledge of a continuum of correlated variables. In a simple sender-receiver game with verifiable information, we show that this richer, more realistic, representation of expertise changes the structure of equilibrium communication. We identify a new form of equilibrium in which the expert conveys many pieces of information and that this strictly increases the expert's leverage over the non-expert. We explore the implications of this equilibrium for the nature of policy advice, policy choice, and organizational form.

Expressed Demands and the Emergence of Organizational Sub-Units: How Petitions Formed Congressional Committees, 1789-1865

Daniel Carpenter
,
Harvard University
Tobias Resch
,
Harvard University
Benjamin Schneer
,
Harvard University
Maggie McKinley
,
Harvard University

Abstract

Committee formation in the early state legislatures and the antebellum Congress coincided with the largest surge in petitioning in American history, yet the clear historical relationship remains unexamined in institutional political science. We develop a model of committee formation where the key input is the inflow of petitions (citizen complaints and requests) across a variety of topics. The floor of the legislature is uninformed about the petitions yet incurs penalties if it fails to respond to them. Given other methods of responding to petitions (select committees, processing them on the Floor, referring them to members as ombudsmen), the floor creates committees to process petitions only when asset specificity of the information combines with political generality. Under the Floor’s optimal policy, the probability of new committees forming is related positively to the frequency of petitions on a given topic (but not necessarily to petitions overall). We test these and other propositions by harvesting two original datasets -- one covering all legislative petitions to the Virginia House of Delegates from 1776 to 1789, the second covering over 100,000 petitions recorded in the House Journal from 1789 to 1875, combined with data on their introduction and referral to select and standing committees. Empirical analyses demonstrate that early American legislatures created committees when topic-specific petitions increased, and particularly when the entropy of petitions across constituencies was higher. Attention to an important, precedent-setting case -- the Virginia Committee on Religion of 1776 – demonstrates the role of topic specificity and geographic generality. Our theory and empirical analysis point to the manifold nature of asset specificity in policy information, give new meaning to the entropy of political agendas, and help reinterpret the origins of standing committees in American legislatures.

Measuring Corruption: Unmasking Strategic Data Manipulation in the Theft of Development Aid

Jean Ensminger
,
California Institute of Technology

Abstract

Corruption is a pervasive and destructive problem that undermines economic development, democracy, and the effectiveness of aid. Efficient identification and measurement are key to finding policy solutions. We present new tests to detect fraudulent patterns in data and apply these methods to a World Bank project in Africa. Digit analysis exploits the fact that humanly produced data follow different patterns than naturally occurring data. Building upon Benford’s Law of natural digit distributions, we develop tests to expose strategic and profitable data fabrication. Our new tests significantly expand the possible use of digit analysis to smaller sample sizes. A forensic audit of the same project we study was conducted by the World Bank and also found massive evidence consistent with fraud, providing external validity for our tests. Our methods improve upon existing anti-corruption techniques in two ways: their application requires little cooperation from potentially corrupt insiders, and they provide a scalable, real time, and cost-effective method to monitor suspected fraud in public and private expenditures.
JEL Classifications
  • H1 - Structure and Scope of Government
  • L2 - Firm Objectives, Organization, and Behavior