Incumbency and Information∗

A long tradition in political science is normatively troubled by the incumbency advantage. High reelection rates raise the concern that incumbents use the perquisites of office to insulate themselves from electoral threat. We uncover a previously unrecognized mechanism contributing to the incumbency advantage. Incumbents govern, and challengers do not. This generates information about incumbents that is not available about challengers. As a result, incumbents systematically win reelection at a different rate than they would were they in an open-seat election. This is so even absent any partisanship, electoral selection or challenger scare off. In particular, it holds even if incumbency and challenger status are randomly assigned. The information-based incumbency advantage co-varies positively with voter welfare. So, even with an ideal empirical research design, a positive incumbency advantage does not have the straightforward normative implications hypothesized by the literature. ∗This paper was inspired by the discussion at the 2016 CHEAP conference. We thank Anthony Fowler, Navin Kartik, Jim Snyder, and seminar audiences at Columbia, Harvard, Stanford, Utah, the Berkeley Political Economy and Governance Conference, the LA Theory Day, the Warwick/Princeton/Utah Political Economy Conference 2018, and Washington PECO 2018. †Harris School of Public Policy, University of Chicago, email: sashwort@uchicago.edu ‡Harris School of Public Policy, University of Chicago, email: bdm@uchicago.edu. §Eller College of Management, University of Arizona, email: amanda@amandafriedenberg.org At least since Erikson’s (1971) agenda-setting paper, political scientists have been normatively concerned about the high rate at which incumbents win reelection. Do high reelection rates imply that incumbents use the perquisites of office—e.g., greater access to campaign resources, gerrymandering, challenger scare off, etc.—to insulate themselves from electoral threat? (See, e.g., Fiorina (1989).) If so, then electoral accountability may be ineffective at generating good governance outcomes. Cox and Katz (2002, p. 7) summarize the conventional wisdom: Whenever the resources of public office are used to insulate individual politicians from electoral risk, their accountability to their constituents is weakened.. . . Thus, insulation from electoral risk of the kind suspected would, at a single stroke, debilitate the two fundamental accountability relationships of a democratic system of government. Of course, high reelection rates need not reflect insulation against electoral threat. Those rates may instead reflect the reasons incumbents get elected in the first place. One reason is party match—partisan voters may systematically elect politicians from the party they prefer. A second reason is electoral selection (Samuelson, 1984; Zaller, 1998; Ashworth and Bueno de Mesquita, 2008; Fowler, 2018). During electoral campaigns, voters may learn important information about candidates’ characteristics, e.g. competence, ideology, or character. Voters use this information to select candidates who appeal to them. As a result, incumbents have characteristics that are, on average, more appealing to voters than do challengers. Importantly, to the extent that party match and electoral selection are major explanations of high incumbent reelection rates, the normative implications are reversed: High incumbent reelection rates might just mean voters do a particularly good job of identifying the politicians they want to have in office. Open Seat Election Party Match Electoral Selection Governance Period Office Holding Effect Engage in Public Policy Exercise Perquisites of Office Closed Seat Election Challenger Scare Off High Reelection Rates Figure 1: Illustration of the Literature Figure 1 summarizes the literature’s understanding of incumbent retention rates. It points to four mechanisms. The first two mechanisms arise from the act of selecting an incumbent. In an open seat election, voters choose politicians based on both partisanship and information about politician characteristics that they learn in the course of the campaign. As a consequence, on average, voters prefer incumbents over challengers in the later closed seat election. Both of these mechanisms benefit voters. The third mechanism arises during the course of holding office. While governing, the incumbent works on public policy and exercises the perquisites of office. These perquisites directly advantage her in the closed seat election, by insulating her from electoral threat. The literature views this office holding effect as detrimental for voters. It reduces both