Representation and Accountability I

Carolina Torreblanca

University of Pennsylvania

Global Development: Intermediate Topics in Politics, Policy, and Data

PSCI 3200 - Spring 2025

Assignments

  • Two assignments due TODAY/Wed at midnight:

  1. Creating a Final Project Git Repo

  2. Idea for Final Research Project

  • Please send a Slack DM to both me and Jeremy

Agenda

  • Background (Accountability 101)
  • Ferraz and Finan, 2011

Democracy, Accountability, and Representation

Przeworski, Stokes, and Manin, 1999

What do we mean by representation?

  • Acting in the voter’s best interests

  • What is the claim connecting democracy and representation?

  • Under democracy, governments are representative because they are elected.

  • But why?

Mandate view:

Elections serve to select good policy

  • Fundamentally prospective
  • During campaign parties inform citizens what policies they want to follow and why
  • Voters select the best one
  • Politicians do what they propose (the platform is the mandate)

Accountability view

Elections serve to hold governments accountable for their past actions

  • Fundamentally retrospective

  • Voters retain politicians only when they acted in their best interest

  • Politicians anticipate this and serve!

Taking stock

  • Do we feel these mechanisms are plausible?

  • What might be some issues with these characterizations of representation?

Reconsidering: Some potential pitfalls

  • Citizens are not omniscient, for better and for worse

  • Imperfect evaluation of what politicians should do

  • Imperfect evaluation of whether they did what they ought to have done

  • Politicians have goals, interests, and values of their own, and monitoring is costly

Pitfalls in the Mandate View

  • Is it plausible to think governments will do what they propose?

  • Is it desirable?

  • Politicians are not legally compelled to abide by their platform in any democratic system! Why?

Pitfals in the Accountability View

  • Idea is reward or punish depending on their performance

  • Is it plausible to think citizens have enough information to evaluate politicians?

  • What if politicians do not value getting reelected

The Vote: Our One Blunt Tool

  • In the accountability view, voters are retrospective.

    • They use the vote to punish

  • In the mandate view, voters are prospective

    • They use the vote to select the best policy / best politicians

  • In reality, voters want to do both: select good policy and punish bad behavior

  • But we only have one vote. Can we achieve both goals?

Evidence from Brazil: Ferraz and Finan (2011)

Research question

  • Broadly: Do institutions affect accountability?

  • Specifically: Do elections work as a disciplining device?

  • Reelection Incentives \(\rightarrow\) corruption

Theory


Politicians who can get reelected have fewer incentives to steal funds

  • Why?

  • They want to get reelected! To get reelected, they need to serve the citizens.

  • Voting as a disciplining tool

Case, Data

  • Context: Municipalities in Brazil

    • Brazilian mayors allowed to run for reelection starting in 2000

  • Data: Misappropriated Funds

    • Random audits of municipalities since 2003

Hypotheses

  • Hypothesis 1: Mayors that can still get reelected will steal less

  • Hypothesis 2: Especially when the “theft” is very visible

Research Design

  • What is the treatment group?
  • What is the control group?
  • What are plausible threats to inference?

Research Design

Research Design

  Constrain the Comparison

  • Compare mayors who barely won and barely lost
  • Compare 1st term mayors that later won to 2nd term mayors
  • Compare mayors of equal experience
  • Compare similar places only

Findings

Findings

  • Mayors who can still run for reelection are less corrupt.

    • Around 1.9-4 pp less corrupt than 2nd term mayors

  • Because around 7.4% of all resources were stolen under 2nd term mayors, a 1.9-4 pp decrease is HUGE and A LOT of money

Robustness


“If my story is true, what else should I observe in the data?”

  • 1s term mayors procure more $ from the fed. gov (effort)
  • Effects smaller where opposition is weak (who else are you going to vote for??)

Implications for Representation

  • That 1st term mayors are less corrupt suggests votes are a good disciplining device

    • The accountability theory seems to have empirical support

  • But… is this evidence that voting does not help select good politicians?

  • Could both mechanisms be operational?

Policy Implications

  • Corruption is more pervasive in weakly institutionalized settings

    • How to make local governments in those settings more accountable?

  • Implications for term limits?

  • Implications for politicians’ salaries?