When Are Local Incumbents Held Accountable for Government Performance? Evidence from US School Districts

• DOI: http://doi.org/10.1111/lsq.12159
• Author: Julia A. Payson
• Published date: 01 August 2017
• Date: 01 August 2017

JULIA A. PAYSON

Stanford University

When Are Local Incumbents Held

Accountable for Government

Performance? Evidence from US

School Districts

Do voters hold local officials accountable for government performance? Using

over a decade of panel data on school district elections and academic achievement in

California, I causally identify the effect of test score changes on school board incumbent

re-election rates and show that incumbents are more likely to win re-election when test

scores improve in their districts—but only in presidential election years. This effect dis-

appears in midterm and off-years, indicating that election timing might facilitate local

government accountability.

Determining the conditions under which voters hold elected off‌i-

cials accountable for government performance is central to the study

of politics. We know that at the national level voters evaluate political

leaders on the basis of economic trends (Fiorina 1981; Key 1966;

Lewis-Beck and Stegmaier 2000). However, the vast majority of elected

off‌icials in the United States serve at the local level and have little control

over the national economy. Instead, over 90,000 local governments in

the United States employ 11 million workers and deliver public services

that directly impact the day-to-day life of citizens, including education,

police protection, and water and sanitation services (US Census Bureau

2012). An important open question is the extent to which voters hold

local politicians responsible for delivering these services. Only a handful

of studies have examined whether voters punish and reward local

government off‌icials on the basis of job performance (Arnold and Carnes

2012; Berry and Howell 2007; Hopkins and Pettingill 2015), and as a

result, we are only beginning to learn about how accountability works in

local government.

Do voters hold local off‌icials accountable for the quality of the pub-

lic services they provide? If so, under what conditions? The majority of

local governments in the United States are single-issue districts that

LEGISLATIVE STUDIES QUARTERLY, 42, 3, August 2017 421

DOI: 10.1111/lsq.12159

V

C2016 Washington University in St. Louis

oversee specif‌ic policy areas. Within this context, voters should have a

relatively easy time attributing responsibility to the incumbents who set

district policies. If residents are satisf‌ied with a service, we would expect

to see incumbents re-elected at high rates. If quality deteriorates, or if

voters are unhappy for any reason, one of the easiest ways to express

this dissatisfaction is at the ballot box. School board elections present a

unique opportunity to test this theory. Not only are school districts the

most common type of local government in the United States, but perfor-

mance data are readily available in the form of standardized test scores.

I use panel data on school board elections and school district perfor-

mance in California to show that district achievement inf‌luences election

results, but the strength and direction of this relationship varies based on

when an election is held. In presidential election years, when turnout is at

its highest, there is a strong, positive correlation between incumbent

performance and district achievement on standardized tests in the year

leading up to the election. However, this relationship is less pronounced in

midterm years and completely f‌lat in off-years. I argue that a likely expla-

nation for this pattern is the fact that the people who vote in low-turnout

elections are often a unique subset of voters with vested interests in the

election results (Anzia 2011; Berry and Gersen 2011). While ordinary dis-

trict residents may rely on district performance to guide their vote choice,

the voters who are active in nonpresidential school board elections—like

teachers and their unions, district staff, and other education-savvy

voters—likely evaluate incumbents on a much broader range of criteria.

I also investigate whether changes in test scores lead incumbents to

strategically exit school board races. I f‌ind no evidence that incumbents

are less likely to run or that challengers are more likely to enter school

board races when test scores fall, which suggests that my estimates of

incumbent performance are not biased as a result of strategic candidate

behavior. Finally, I assess whether low turnout in nonpresidential elec-

tions is a plausible mechanism driving the observed f‌indings. I show that

turnout in school board elections drops from roughly 32% in presidential

years to 25% in midterm years and to under 15% in off-year elections.

This f‌inding holds across districts but is also true for districts that switch

from off-year to even-year elections. When districts hold their elections

in nonpresidential years, turnout drops, and academic performance

matters less for incumbent re-election.

Voting and Accountability in the Local Context

Literature on government accountability typically claims that

elected off‌icials can be considered accountable if voters are able to

422 Julia A. Payson