Keefer

Incumbent Advantage, Voter Information
and Vote Buying ∗

Cesi Cruz† Philip Keefer‡ Julien Labonne§
November 2015
Abstract
Results from a new experiment shed light on the effects of voter informa-
tion on vote buying and incumbent advantage. The treatment provided voters
with information about a major spending program and the proposed alloca-
tions and promises of mayoral candidates just prior to municipal elections. It
left voters more knowledgeable about candidates’ proposed policies and in-
creased the salience of spending, but did not affect vote shares and turnout.
Treated voters were more likely to be targeted for vote buying. We develop a
model of vote buying that accounts for these results. The information raised
voter expectations regarding incumbent performance, especially in incumbent
strongholds. Incumbents increased vote buying in response. Consistent with
this explanation, both knowledge and vote buying impacts were higher in
incumbent-dominated municipalities. Our findings show that, in a political
environment where vote buying is the currency of electoral mobilization, in-
cumbent efforts to increase voter welfare may take the form of greater vote
buying.
Keywords: Political Economy, Vote Buying, Information, Elections
∗ This project would not have been possible without the support and cooperation of PPCRV vol-
unteers in Ilocos Norte and Ilocos Sur. We are grateful to Michael Davidson for excellent research
assistance and to Prudenciano Gordoncillo and the UPLB team for collecting the data. We thank
Marcel Fafchamps, Clement Imbert, Pablo Querubin, Simon Quinn and two anonymous review-
ers for comments on the pre-analysis plan. Pablo Querubin graciously shared his precinct-level
data from the 2010 elections with us. We thank Michael Davidson, Andrew Foster, James Fenske,
Gyung Ho Jeong, Chris Kam, Pablo Querubin and Dean Yang, as well as conference and seminar
participants at George Mason University, MPSA 2015, University of British Columbia, University of
Copenhagen, University of Gothenburg, University of Michigan University of Oxford and Univer-
sity of Washington for comments. We are grateful for funding from the Research Support Budget
of the World Bank. The project received ethics approval from the University of Oxford Economics
Department (Econ DREC Ref. No. 1213/0014). The opinions and conclusions expressed here are
those of the authors and not those of the World Bank or the Inter-American Development Bank.
† University of British Columbia; email: cesi.cruz@ubc.ca
‡ Inter-American Development Bank; email: pkeefer@iadb.org
§ Yale-NUS College; email: julien.labonne@yale-nus.edu.sg
1 Introduction
The results of a novel experiment in the Philippines extend research on voter in-
formation, vote buying, and incumbent advantage. First, we find that voters’ in-
formation about what politicians can do for them increases politician incentives
to engage in vote buying. Second, it reduces incumbent advantage, the degree to
which some incumbents can secure a higher vote share than others without exert-
ing greater effort on behalf of constituents.
These conclusions emerge from an experiment in which, just before the munic-
ipal elections in the Philippines, we gave voters information about a large fund,
provided by the central government to every municipality and intended to fi-
nance municipal development projects. The intervention was implemented in 142
randomly selected villages in 12 municipalities and significantly increased voter
knowledge about the funding program and the concrete policies that candidates
promised to fund under the program. It also increased the salience of spending:
voters were more likely to report that an important determinant of their vote choice
was whether candidates proposed to spend the municipal budget on things that
were important to their household. Finally, voters who received the information
were more likely to report vote buying.
We develop a framework to understand how voters and incumbents should
react to an information shock that makes voters aware of the potential resources
incumbents have to provide public goods. This reaction depends on the size of
the pre-shock information asymmetry regarding this potential. Consistent with
the model, and information-based theories of incumbent advantage more gener-
ally, the effects of the intervention on knowledge and vote buying are larger in
incumbent strongholds (municipalities where incumbent margins of victory were
largest in the previous election). Since incumbents are able to respond to the infor-
mation shock by increasing vote buying, we do not expect a change in turnout or
candidate vote shares; the treatment has no effect on either.
Voter knowledge of the services that incumbents could provide them matters
most in political settings where candidates cannot make credible pre-electoral com-
mitments regarding the policies they plan to implement after the elections. In these
settings, voters rely more on retrospective voting rules based on the observed per-
formance of the incumbent, ignoring challenger characteristics. Incumbents with
1
an information advantage are able to persuade voters that there was little they
could do for voters, leading voters to hold them to a lower performance challenge
than they hold incumbents who do not enjoy this informational advantage.
The Philippines has several characteristics that make it ideal for this study. Po-
litical parties are weak and evanescent, such that politicians cannot make credible
pre-electoral commitments; and vote buying is widespread. We do not seek to
explain systemic incumbent advantage (e.g., why U.S. Congressmen are difficult
to dislodge) or disadvantage (why Indian MPs experience notably high turnover).
However, the Philippines allows us to explore how information affects dominant
relative to other incumbents: dominant incumbents are common in the Philippines
and in our study area.
Our results have several important implications. We provide the first direct
evidence that vote-buying can be a response to higher voter expectations for in-
cumbent performance. Information interventions that attenuate the advantages
of dominant incumbents, making elections more competitive, do increase politi-
cal incentives to deliver benefits to voters. However, as Khemani (2013) notes, the
equilibrium outcome of increased political competition in clientelist settings tends
to take the form of greater vote buying and worse public service delivery. As a re-
sult, despite increasing the leverage of voters vis-à-vis incumbents, interventions
to make public spending decisions more salient in the week before an election are
still not enough to shift the political equilibrium towards less vote buying. How-
ever, the results of the intervention offer a strong motivation to examine the ef-
fects of a similar intervention earlier in the electoral cycle, when incumbents have
greater opportunity to react by increasing the provision of public goods.
The remainder of the paper is organized as follows. Section 2 nests our analysis
in a wide range of earlier contributions to the literature. Section 3 describes munic-
ipal elections in the Philippines and the experiment and data. Section 4 details the
direct effects of the information intervention. In Section 5, to interpret our results,
we develop a retrospective voting model of political competition where candidates
cannot make credible pre-electoral promises. The model yields a number of ancil-
lary predictions that we are able to test; these predictions are examined in Section
6. Section 7 concludes.
2
2 Literature Review
Although there is a considerable literature on vote buying, information and in-
cumbency advantage, ours is the first paper to examine these phenomena jointly,
and to show that the provision of information about what governments can do
for citizens can have significant effects on vote-buying and incumbency advan-
tage. The sections below explain in more detail our contributions to research on
vote buying; information; and on the relationship between voter information and
incumbent advantage.
2.1 Vote Buying

In many countries, vote-buying (gifts of goods or money before the election, aimed
at persuading recipients to vote in a particular way, or to vote at all) is pervasive
and pre-electoral transfers to voters are substantial. Across the 17 countries sur-
veyed in the 2005-06 wave of the Afrobarometer survey, 19 percent of more than
20,000 respondents reported that they had been offered a gift in the last election.
Brusco et al. (2009) surveyed nearly 2,000 respondents in three Argentine provinces
three months after the October 2001 elections. Forty-four percent of respondents
said that parties had distributed food, clothing and other items to homes in their
neighborhoods. In our data, 14 percent of households (in the control group) report
vote buying.1
Ours is the first analysis that links vote buying to voter information and in-
cumbent dominance. In contrast, a large literature focuses on enforcement, such
as party machines and social networks (Brusco et al., 2009; Cruz, 2013) or norms of
reciprocity (Finan and Schechter, 2012), and on the extent of leakage (voters who
do not necessarily vote for the politicians who pay them, as in Schaffer (2007)).
Hicken et al. (2014) show that exhortations not to buy votes interact with voter
incentives to pre-commit to resist the temptation to sell their votes: voters who
promise not to sell their vote are less likely to sell their vote than are voters who
promise to vote their conscience even if they do sell their vote. Vicente (2014) looks
at the effects of normative information on the acceptability of vote buying. Leaflets
distributed to voters ahead of the 2006 elections in Sao Tome and Principe, discour-
1 Nichter (2008) observes that transactions that are commonly labelled vote buying may, instead,
be turnout buying. Our analysis is consistent with either.
3
aging them from allowing vote buying to influence their vote, significant reduced
turnout and challenger and incumbent vote shares.
A substantial literature emphasizes the potential tradeoffs between vote buying
and other narrowly targeted transfers, on the one hand, and the provision of broad
public services, on the other. For example, models developed in Keefer and Vlaicu
(2008) and Hanusch and Keefer (2012) link preferences for targeted transfers and
vote buying, respectively, to the inability of politicians to make credible commit-
ments to citizens. Kitschelt (2000) concludes that vote-buying is more common in
countries with non-programmatic political parties, such as the Philippines. These
analyses do not address, as we do here, the effects on vote buying of voter un-
certainty about the benefits that incumbents provide them, nor the heterogeneous
incentives of dominant and non-dominant incumbents to engage in vote buying.
2.2 Information Provision

We examine a unique treatment, informing voters about what government could
do for them. A large literature on information and electoral behavior examines
other types of information, particularly on voting procedures and voting irregular-
ities (Vicente, 2014; Aker et al., 2011); valence issues such as candidate corruption,
criminal records, education and attendance at parliamentary sessions (Banerjee
et al., 2011; Chong et al., 2011; Humphreys and Weinstein, 2013) and radio broad-
casts that increase the demand for public services (Keefer and Khemani, 2014).
Aker et al. (2011) conducted three information interventions before the 2009
elections in Mozambique, providing treatment areas with brochures about how to
vote, SMS messages about electoral problems observed in senders’ local areas, or
free copies of the largest newspaper in Mozambique, which included information
about how to vote and how to send messages about electoral problems. All of the
treatments increased voter turnout. Information about how to vote (though not
about electoral irregularities) favored the incumbent party in Mozambique.
Chong et al. (2011) provided voters with information on incumbent corruption
prior to local elections in Mexico. The treatment reduced turnout and the vote
shares of both incumbents and challengers. Banerjee et al. (2011) distributed in-
formation to voters in India about the effort exerted by the incumbent legislator
(the legislator’s committee attendance, legislative activity, and spending of discre-
4
tionary jurisdiction development funds), as well as about the wealth, education
and criminal record of the incumbent and two main challengers. They find no
effects on gift giving, which is similar to vote buying in the Philippines.
As in our intervention, Fujiwara and Wantchekon (2013) also provide voters
with more neutral policy information, but in the context of deliberative town hall
meetings in Benin, where candidate positions on policies may or may not have
been revealed. Their intervention did not have a significant effect on survey re-
sponses regarding vote buying, though it did significantly affect a broader index
of survey responses that they jointly characterized as reflecting voter attitudes to-
wards clientelist forms of electoral mobilization.
Our information intervention emphasizes broadly-targeted spending programs
in a context where targeted pre-electoral transfers (vote buying) are common. In
empirical research on Benin, Keefer and Khemani (2014) find that radio broadcasts
of information touting the benefits of health and education services shift house-
hold preferences towards candidates who promise jobs for a few rather than more
health and education benefits for all. However, consistent with the results reported
below, they also find that exposure to this programming has no effect on prefer-
ences for candidates who give electoral gifts, when it is unclear whether those gifts
come at the expense of health and education services. In addition, this information
does not seem to affect actual government behavior: villages with better access to
radio do not have better schools nor access to more, free, anti-malaria bed nets.
2.3 Incomplete Information and Dominant Incumbents

A growing literature focuses on incumbent advantage in developing countries or
new democracies. Most of the research documents a tendency for incumbency to
reduce prospects of re-election (Uppal, 2009; Klasnja, 2014). We are not concerned
with the extent of systematic incumbent (dis)advantage in the Philippines. Instead,
our analysis offers the first direct test of the effects of information on the behavior
of dominant incumbents – incumbents elected by a large margin and who are more
likely to possess an incumbency advantage.
Information plays a prominent role in research on the sources of incumbent
advantage in developing countries.2 Boas and Hidalgo (2011) find that control of
2A large literature considers the sources of incumbent advantage in the United States, focusing
5
community radio stations in Brazil increases incumbent vote share. Though their
evidence does not allow them to unpack the radio programming that gives rise to
this advantage, it is most plausible that it simply restricts the airtime of challengers
and exaggerates the contributions of incumbents to listener welfare. MacDonald
(2014), investigating a pattern of incumbent disadvantage in Zambia, argues that
incumbents enjoy a smaller incumbency advantage in Zambia in districts where
citizens have greater access to radio. These contributions highlight the importance
of control over or access to sources of information, but do not investigate the par-
ticular types of information that might matter and the effects of access on vote
buying. One contribution of this paper is to highlight the effects of one particular
type of information, about the capacity of the incumbent to deliver public goods,
on incumbent advantage.3
The results of the Banerjee et al. (2011) experiment also support the notion of
an information-linked incumbent advantage. They inform voters about incum-
bent legislative effort and the criminal records of incumbents and challengers. The
strongest effects on turnout and vote-shares emerge in jurisdictions where the in-
cumbent’s performance rating (e.g., attendance at committee meetings) was worse
and where the challengers were better qualified–that is, where incumbent perfor-
mance was worse than uninformed voters might have anticipated. Our experiment
instead examines the effects of information that signals what benefits incumbents
could have provided voters. Ansolabehere et al. (2006) trace the electoral advan-
tages of US congressional incumbents to the greater coverage they receive in print
media. Using data from the United States, Kasnja (2011) concludes that an increase
in political awareness–knowledge of basic political facts–systematically reduces
support for incumbents accused of corruption.
An implication of these analyses is that incumbents should increase their per-
on issues such as name recognition, resource and staffing advantages, and better connections with
campaign financiers. These are all present in the context of developing country democracies, as
well. Our analysis and review of the literature are concerned with other features of developing
country democracies that contribute to incumbent advantage and are more pronounced in these
countries, such as imperfect information about the existence of government programs and incum-
bent control of the media.
3 Klasnja (forthcoming) argues that incumbent advantage arises when corruption increases with
incumbent tenure (because of learning or fixed costs in the establishment of corruption networks).
Voters then have an incentive to replace incumbents sooner. Klasnja and Titiunik (2013) attribute
incumbent disadvantage in Brazilian municipalities to term limits and the weak attachment of in-
cumbent mayors to their parties.
6
formance on behalf of voters when their information advantage shrinks. The in-
formation intervention we examine has this effect: it tells voters that incumbents
have greater capacity to deliver public goods than voters thought.
The theoretical literature also describes circumstances under which citizen in-
formation undermines incumbents’ advantages, though because of strategic reac-
tions by incumbents to voter information, the effects are subtle and often indirect.
Gordon et al. (2007) endogenize both the quality of challengers who enter a race
and voters’ decision to acquire information about challengers. Incumbency ad-
vantage arises from the presence of incomplete voter information about incum-
bent quality.4 Hodler et al. (2010) argue that incumbents’ advantage emerges from
their better knowledge of the state of the world. Incumbents use this knowledge
to strategically choose inefficient policies in areas where they are more competent
than the challenger. Their actions produce outcomes that are just adverse enough
to encourage voters to re-elect them, since they are the candidates best placed to
address the policy problem.
3 The Experiment and Data

We focus on the electoral incentives of mayoral candidates in the Philippines. Since
the passage of the 1991 Local Government Code (LGC), municipalities have had
an important role in the delivery of basic services. The code devolved a number
of responsibilities, such as responsibility for nutrition programs (Khemani, 2013),
and transferred a large number of civil servants to municipalities (Llanto, 2012).
Despite the presence of one vice-mayor and eight municipal councilors, mayors ex-
ert significant control over how municipal resources are spent (Hutchcroft, 2012).
However, local officials exert little control over the size of the municipal budget.
The typical municipality relies on fixed fiscal transfers from the central govern-
ment, with the average municipality taking 85 percent of its revenues from those
fixed transfers (Troland, 2014). Laws governing transfers to municipalities encour-
4 Animprovement in voter estimates of incumbent quality, which they infer from incumbent
performance, either increases the probability that the uninformed voter retains the incumbent, or
the probability that the uninformed voter exerts effort to become informed; either outcome boosts
the unconditional probability of incumbent victory. An information shock that reduces voter per-
ceptions of incumbent quality relative to the challenger correspondingly reduces incumbent ad-
vantage.
7
age municipalities to allocate 20 percent of transfers to development projects.5
Second, municipal electoral campaigns tend to be centered around personali-
ties and family alliances rather than around policies and party platforms (Hutchcroft
and Rocamora, 2003; Kerkvliet, 2002). Filipino mayors are often viewed as local
bosses (Capuno, 2012; Sidel, 1999) with substantial control over municipal budgets
and spending decisions. As a result, party affiliations do not provide information
about proposed policies and programs, but rather about alliances between local
and national politicians. Consistent with the lack of programmatic parties, vote
buying is prevalent and tends to take place a few days before the elections (Cruz,
2013). In addition, prior research has documented that Filipino voters use retro-
spective voting rules when deciding whether to re-elect the incumbent (Cruz and
Schneider, 2013; Labonne, 2013).
3.1 The Experiment

Our intervention is unique in eliciting and disseminating the public spending promises
of candidates. It was designed with and implemented by the Parish Pastoral Coun-
cil for Responsible Voting (PPCRV). We collected data from candidates on their
proposed policies and platforms and distributed that information in brochures to
voters in randomly selected villages ahead of the May 13, 2013 mayoral elections.6
These brochures and the household visits by PPCRV staff to distribute and explain
the brochures increased voter information about the types of public good provi-
sion that mayors could provide and to inform them of the availability of a large
fund to finance provision of municipal public goods. A detailed timeline of the
experiment is available in Table A.1. The pre-analysis plan (PAP) was registered
with J-PAL’s hypotheses registry on May 12, 2013.7
5 Section 287 of the Local Government Code states that Each local government unit shall appropriate
in its annual budget no less than twenty percent (20%) of its annual internal revenue allotment for develop-
ment projects. Copies of the development plans of local government units shall be furnished the Department
of Interior and Local Government. In its Memorandum Circular 2010-138, the Department of Inte-
rior and Local Government further clarified that development means the realization of desirable, social,
economic and environmental outcomes essential in the attainment of the constitutional objective of a desired
quality of life for all. Those guidelines were further refined in the Joint Memorandum Circular 2011-1,
issued on April 13, 2011 by the Department of Interior and Local Government and the Department
of Budget and Management.
6 A copy of a flyer is included as Figure A.1.
7 The submitted documents are available at: http://www.povertyactionlab.org/Hypothesis-
Registry
8
In April 2013, we interviewed every candidate for mayor in twelve municipal-
ities in the provinces of Ilocos Norte and Ilocos Sur, in the northern reaches of the
Philippines. Candidate names were taken from the official list of the Commission
on Elections (COMELEC).8 In the course of the interview, we gave each candidate
a worksheet with a list of sectors.9 Candidates were told the average amount that
they would have to spend from their local development fund (LDF) and asked to
allocate money across sectors.
To facilitate this decision, candidates received 20 tokens to place on the work-
sheet and were told that each token represented five percent of the total LDF. They
were instructed that they could put the tokens on any combination of sectors that
they wished. Once they indicated they were satisfied with the sector allocation, the
enumerators would record it. This information would then be given to randomly-
selected barangays in their municipality before the elections.
If candidates did not give careful consideration to the exercise, the sectoral al-
locations described in the brochures could represent only a weak signal of what
they would do if elected. Fortunately, the opposite was the case. Candidates were
eager to participate in the program. Once they learned of it, they called PPCRV to
make sure that they would be included.10 They also took the process of allocating
tokens across sectors seriously, typically spending several minutes to arrange and
rearrange the tokens after reconsidering their allocation.
Further evidence of candidate engagement comes from comparisons of what
incumbents said they would do with the money and what they actually did with
it. For incumbents we can compare how they allocated the budget while in office
with what they proposed. We use budgetary data for the last full fiscal year before
the election (2012) and compute the correlation between the share of the budget
8 Ilocos Norte is the home province of the Marcos family and, both Imelda Marcos and her
daughter Imee Marcos are still politically active there. Imee has been governor of Ilocos Norte
since 2010 and Imelda has been Representative for Ilocos Norte’s Second District since 2010.
9 The sectors: public health services; public education services; cash or in-kind transfers (such
as loans or job assistance); water and sanitation services; road construction and rehabilitation; con-
struction of community facilities (such as multipurpose halls or basketball courts); business loans
and other private economic development programs; agricultural assistance and irrigation systems;
peace and security; community events and festivals.
10 Only one candidate refused to participate. One might think that more incumbents would have
refused to participate, since the brochures would increase voter knowledge of incumbent capacity
to provide them with public goods. In fact, because the brochures were going to be disseminated
even if incumbents chose not to have their promises included in them, it was in the incumbents’
interest to participate.
9
spent on each sector with the share of the budget that the incumbent proposes to
spend on the sector. We find that the correlation is 0.55 which we interpret as large
particularly since changes in priorities and errors in budget data make a perfect
correlation unlikely. In addition, this is indicative that candidates conscientiously
allocated their tokens.
Finally, we asked candidates to list projects and programs that they would im-
plement if elected. Most of the proposals were quite specific, reducing concerns
that candidates allocated tokens in an arbitrary and ad hoc manner. In addition,
the sectoral allocations of candidates were in line with their listed projects and pro-
grams. Candidates consistently allocated a greater share of their proposed budget
to those sectors that matched their promises (see Figure A.2).
Within each target municipality, villages were allocated to the treatment and
control groups using a pairwise matching algorithm. First, for all potential pairs,
the Mahalanobis distance was computed using village-level data on population,
number of registered voters, the number of precincts, a rural dummy, turnout in
the 2010 municipal election and incumbent vote share in the 2010 elections. Sec-
ond, among 5,000 randomly selected partitions, the partition that minimized the
total sum of Mahalanobis distance between villages in the same pairs was selected.
Third, within each pair, a village was randomly selected to be allocated to treat-
ment; the other one serving as control. The final sample includes 142 treatment
and 142 control villages in twelve municipalities (cf. Table A.2).11
PPCRV put together flyers comparing the proposed allocations of all candi-
dates in each municipality.12 Then, in the week leading up to the election, PPCRV
volunteers, trained to use a detailed script, distributed the flyers through door-
to-door visits in the target villages. On average, 44.5 percent of voters resided in
treated villages and 55.5 in control villages. Importantly, the flyer and the script
did not mention vote buying, reducing any concerns related to social desirability
bias.
In our interpretation of the findings, we emphasize that the flyer informed vot-
11 The list differs slightly from the one included in the Pre-Analysis Plan as volunteers could not
distribute the flyers in Banayoyo, Pagudpud and Tagudin. In addition, we had to drop one pair
in Pasuquin as we found out during the endline survey that the control village in that pair was a
military camp. This is consistent with the protocols listed in Section 5.4 "Potential Adjustments for
non compliance" of the Pre-Analysis Plan.
12 There were only two or three candidates in each municipality. The figures are available in Table
A.2.
10
ers of the existence of a key government program to provide infrastructure, along
with evidence of the significance of the program (the flyer, the participation of a
prominent NGO, and the participation of the candidates). All of the effects that
we identify are consistent with this interpretation. However, the flyer also pro-
vided voters with information about the candidates, especially who the candidates
were and the allocation preferences of candidates. Neither of these can account for
the results we present below. Knowledge of candidates was unaffected by the in-
formation treatment. Furthermore, all of our results are robust to controlling for
differences between the allocation preferences of candidates and those of house-
holds. Indeed, consistent with our assumption, and the consensus of observers,
that politicians cannot easily make credible pre-electoral promises in the Philip-
pines, we do not observe that the treatment leads voters to vote for candidates
closer to their own preferences.
Evidence that the treatment and control groups were similar prior to the in-
tervention indicates that the randomization strategy was successful. The results
in Table 1 indicate that the village-level variables used to carry out the pairwise
matching exercises are well-balanced. We also use data from the survey to test if
the treatment and control are balanced along sectoral preferences, alignment with
the candidates (incumbent and challengers), household composition, households
assets, etc.13 Out of the 42 village- and household-level variables for which we
test balance, only 4 exhibit differences that are significant at the 10 percent level.
Controlling for those variables does not affect results reported below.
3.2 Data
We use two main data sources in the paper. First, we gathered precinct-level elec-
tion results from the COMELEC.14 The data included information on the number
of votes obtained by all candidates in the mayoral elections. We then used data
from the Project of Precincts to match precincts to villages. Every village contains
at least one precinct. Second, we implemented a household survey in 284 villages
in twelve municipalities in June 2013. Twelve households were interviewed in
each village for a total sample size of 3,408 households. These interviews yielded
13 This set of results is available in Table A.3-A.5.
14 The data were available at: http://2013electionresults.comelec.gov.ph/res_reg0.html
11
Table 1: Balance Tests
Treatment Control T-test K-Smirnov test OLS
(1) (2) (3) (4) (5)
# Precincts 1.092 1.099 0.181 0.021 -0.007
(0.289) (0.363) [0.857] [1.000] [0.828]
Registered Voters 526.261 544.937 0.484 0.070 -18.676
(306.785) (342.678) [0.629] [0.842] [0.526]
Population 842.197 895.923 0.826 0.056 -53.725
(492.927) (598.277) [0.410] [0.969] [0.305]
Turnout 0.785 0.785 0.010 0.056 0.000
(0.082) (0.081) [0.992] [0.969] [0.982]
Incumbent vote share 0.728 0.720 -0.314 0.063 0.008
(0.212) (0.230) [0.754] [0.919] [0.520]
Incumbent vote share [adj.] 0.559 0.552 -0.452 0.056 0.007
(0.130) (0.145) [0.652] [0.969] [0.526]
Rural 0.880 0.873 -0.180 0.007 0.007
(0.326) (0.334) [0.857] [1.000] [0.836]
Notes: n=284. The standard deviations are in (parentheses) (Columns 1-2). In Columns 3-4, the test
statistics are reported along with the p-values [bracket]. Each cell in Column 5 is either the coef-
ficient on the dummy variable indicating whether the campaign was implemented in the village
from a different OLS regression with pair fixed-effects or the associated p-value in [bracket].
the key variables that we use in the analysis. Descriptive statistics are reported in
Table 2.
Political Knowledge In order to examine whether the intervention was effective,

we compare the knowledge of budget allocations between treated and untreated
households. For each of the ten sectors about which respondents received infor-
mation, respondents were asked to name the candidate with the highest proposed
allocation. Following Kling et al. (2007), we create an index aggregating the various
indicators of knowledge of the campaign promises by taking the simple average of
the demeaned indicators (divided by the control group standard deviation). So if
Kis is individual i’s knowledge about sector s promises (i.e., whether they correctly
identified the candidate who proposed to spend the largest share of the LDF on
sector s), then the knowledge index is:15
15 The formula included in the PAP contained a typo.
12
1 Kis − K̄s
Ki = ∑
10 s σs
where K̄s and σs are respectively the control group mean and control group stan-
dard deviation.
Salience Another test of effectiveness of the intervention is whether treated house-

holds cared more about local development spending than untreated households.
To establish the salience of local development spending in household voting de-
cisions, we asked respondents about six possible influences on their decision to
vote.16 One of these was whether candidates spend the municipal budget on
things that are important to the household. The other five were the preferences
of friends and family; gift or money from the candidates before the elections; the
candidates’ ability to use political connections to get money and projects for the
municipality; fear of reprisal from candidates; and the approachability or help-
fulness of candidates.17 They rated how important each of these was on a 0 - 4
scale, from “not important” to “very important”. Respondents took flashcards,
each with a reason for voting, and laid it on a worksheet with the numbers 0 -
4, to indicate the importance of that factor. We use both the raw responses and
responses adjusted for the average answers in the other five categories.
Preferences over candidates and spending allocations We analyze the treat-

ment’s effects on voter preferences for candidates as a function of the proximity
of candidates’ allocations and voters’ preferred allocations. We therefore collected
data on respondents’ candidate preferences and vote choice and asked them to ex-
press their preferences over the ten different spending categories that were given
to the mayoral candidates. Each respondent rated all mayoral candidates on a 0 -
4 scale (strongly disagree to strongly agree). In addition, in order to reduce over-
reporting votes in support of the winner, we used a secret ballot.18
16 Although the ordering of the six alternatives may have affected the response, the ordering was
the same across treatment and control groups.
17 To ensure that the six possible influences were all salient to respondents, the lists were exten-
sively field-tested by one of the authors ahead of a similar survey carried out in the nearby province
of Isabela.
18 Respondents were given ballots with only ID codes corresponding to their survey instrument.
The ballots contained the names and parties of the mayoral candidates in the municipality, in the
13
With respect to preferences over spending allocations, a similar procedure was
used as with candidates. Respondents were given 20 tokens and asked to allocate
the tokens in any manner they wished across the ten categories. We then calcu-
lated how close the preferences of the candidates were to those of the household
by comparing the share S that voter v allocated to sector s with the share that can-
didate c allocated to the sector. We then construct an agreement index, defined
as Avc = ∑s min (Svs , Scs ): the total spending over which the candidate and voter
agree.
A potential concern with this variable is that, since it represents a choice that
respondents aren’t used to making, the quality of the data collected might suffer.
To check this, we regress preferences on a number of household characteristics that
we expect to be correlated with preferences for a given sector. For example, we ex-
pect families with children to favor spending on education and farmers to favor
spending in agriculture. Results presented in Table A.6 suggest that stated prefer-
ences over spending priorities match up relatively well with observable household
characteristics.
It is also possible that, since household preferences were collected after the in-
formation about candidates’ promises had been distributed to voters in the treat-
ment group, voters might have adjusted their preferences to match their preferred
candidate’s promises. Two pieces of evidence suggest that this is not the case. First,
we are unable to reject the null hypothesis that the alignment between respondents
and their preferred candidate is the same between the treatment and control group.
This holds whether we define the preferred candidate as the top ranked candidate
on the 0-4 scale or as the candidate whom respondents indicated voting for in the
secret ballot exercise. Second, the correlation between alignment and support for
given candidates is essentially the same across the treatment and control groups
(Results in Table A.7).
Occurrence of vote buying Observers assured us that vote buying occurs in the
days before the election and that candidates and their brokers can re-target vote
same order and spelling as they appeared on the actual ballot. The respondents were instructed
to select the candidate that they voted for, place the ballot in the envelope, and seal the envelope.
Enumerators could not see the contents of these envelopes at any point and respondents were told
that the envelopes remained sealed until they were brought to the survey firm to be encoded with
the rest of the survey.
14
buying quickly.19 Hence, we expected that our information intervention, only a
week before the election, might still affect vote buying. We measure vote buying
through a series of questions asking whether respondents were aware of any case
of vote buying in their village and if, during the recent election, someone offered
them money for their vote. In the Philippines, social desirability bias associated
with vote buying is low and responses to direct questions provide credible esti-
mates of vote buying incidence. In Isabela, a province near our study area, Khe-
mani (2013) uses vote buying estimates from direct questions. Further, Cruz (2013)
reports that, in the same province, the vote buying estimates obtained through
direct questions are similar to the ones obtained with the unmatched count tech-
nique.20
4 Results: Effects on Knowledge and Vote Buying

Two direct effects of the experiment are of particular interest: did the information
treatment in fact increase relevant political knowledge? And did it influence vote
buying? The results in this Section first verify the essential assumption that treated
voters are indeed more informed about candidates and in the issues that they re-
gard as electorally salient. We then report results showing that the intervention
increased vote buying. To explain the increase in vote buying, in Section 5 we
develop a model that yields a number of additional predictions related to hetero-
geneity in treatment effects and voter choice. We test these predictions in Section
6.
19 Stokes
et al. (2013) argue that brokers, who are local, are given resources by the candidate to
ensure a certain level of support for the candidate. They seek to retain some of these resources as
rents for themselves, but rapidly disburse when they observe an information shock that reduces
support for their candidate.
20 In the pre-analysis plan, we indicated that we intended to also use an unmatched count tech-
nique to asses the extent of social desirability bias. The question was included in the household
questionnaire but it referred to vote buying on ’election day’. Specifically, the question was Here are
some things that can happen to people during election day. How many of these things happened to you? You
don’t have to tell us which things happened, just how many. Given that in Ilocos, vote buying tends to
take place over a longer period of time, those estimates are unreliable and we do not use them in the
analysis. Consistent with this, the unmatched estimates are much lower than the direct questions.
Note that this is the opposite of what we would expect if the direct questions elicited significant
social desirability bias.
15
4.1 Did the Treatment Increase Knowledge?
The descriptive statistics reported in Table 2 suggest that voters tend to be poorly
informed about candidates promises. Voters in the control group make an average
of seven mistakes over the ten sectors. In this Section, we report results concerning
the information’s treatment effect on voter knowledge.
Table 2: Descriptive Statistics

Treatment Control
(1) (2)
Know promises 0.051 0.000
(0.501) (0.470)
Know politicians 0.034 0.009
(0.544) (0.573)
Salience sectors 2.458 2.300
(1.470) (1.513)
Salience sectors (adjusted) 0.884 0.776
(1.188) (1.207)
Error incumbent 4.763 4.974
(3.080) (3.110)
Error challenger 2.059 2.084
(3.017) (3.111)
Relative preference 12.597 12.529
(12.101) (11.663)
Turnout (self-reported) 0.965 0.965
(0.183) (0.184)
Vote Buying 0.165 0.138
(0.371) (0.345)
Notes: n= 3,408. The standard deviations are in (parentheses) (Columns 1-2)
To test the validity of the assumption that the treatment affected relevant po-
litical knowledge, we estimate the intervention’s impacts on knowledge of candi-
dates’ promises using regressions of the form:
Yijk = αTj + vk + uijk (1)
where Yijk is the knowledge index for individual i in village j in pair k, Tj is a

dummy equal to one if the campaign was implemented in village j, vk is a pair-
specific unobservable and, uijk is the usual idiosyncratic error term. To account for
16
the way the randomization was carried out, standard errors are clustered at the
village level. As indicated in the Pre-Analysis Plan, we estimate equation (1) with-
out fixed-effects, with municipal fixed effects and with pair fixed effects (Brunh
and McKenzie, 2009). Our preferred specification is the one with pair fixed effects
and without additional controls. We also test if results are robust to the inclusion
of the four variables that are not balanced between treatment and control.
As expected and outlined in the PAP, voters in treatment villages are more
likely to know which candidates is promising to spend the largest share of the LDF
on any given sector. Results are available in Table 3.21 Supporting the strength of
our randomization strategy and the balance between treated and control groups,
the point estimates are essentially constant across the four different specifications
though the standard errors get smaller as we include more fixed-effects and addi-
tional control variables. As is the case with a number of other outcome variables,
the fixed effects explain a large share of the variation in voter knowledge.
Table 3: Effects of Treatment on: Knowledge of Promises

(1) (2) (3) (4)
Treat 0.051 0.051** 0.051*** 0.052***
(0.036) (0.022) (0.015) (0.015)
Municipal Fixed-Effects No Yes No No

Pair Fixed-Effects No No Yes Yes
Additional Controls No No No Yes
Observations 3,408 3,408 3,408 3,408

R-squared 0.003 0.245 0.326 0.326
Notes: Results from individual-level regressions. The dependent variable is an index capturing the
respondent’s knowledge of candidate promises. In Column 4, the regression includes a dummy
equal to one if someone in the household is a member of any group, a dummy equal to one if
someone in the household participated in any collective action activity in the village in the past six
months, the share of the local budget the respondent would like to spend on water and the share of
the local budget the respondent would like to spend on roads. The standard errors (in parentheses)
account for potential correlation within village. * denotes significance at the 10%, ** at the 5% and,
*** at the 1% level.
The fact that we asked voters precisely about the information provided during
21 Thetreatment had no effect on dimensions of political knowledge not included in the flyers
(Table A.8).
17
the intervention, and that we can confirm that the intervention indeed increased
voter knowledge, is an improvement over the existing literature, where these tests
have not been possible. Among the exceptions, such as Banerjee et al. (2011), re-
searchers have tested treatment effects on voter knowledge, but the knowledge
tested differed from the information provided to voters through the intervention.
We further explore whether the treatment affected how respondents decide
which candidate to elect. In particular, we are interested in the treatment’s effect
on the salience of local development spending on vote decisions. We estimate:
Yijk = αTj + vk + uijk (2)
where Yijk captures how salient sectoral allocations are when individual i in village
j in pair k decides which candidate to vote for. The set-up is equivalent to the one
used for equation (1).
As expected and outlined in the PAP, voters are more likely to report that candi-
dates’ proposals for local development spending are important when they decide
which candidate to vote for. Results are available in Table 4. This holds whether
raw or adjusted ratings are used.
4.2 Information Effects on Vote Buying

Survey results indicated high levels of vote buying - 13 percent of voters in the
control group indicated being offered money for their votes. We can show that
vote buying increased in the treatment villages, estimating equations of the form:
Yjk = αTj + vk + u jk (3)
where Yjk is the prevalence of vote buying in village j in pair k during the May
2013 elections.22 The set-up is equivalent to the one used for equation (1).
Importantly, as described above, vote buying tends to take place a few days
before the elections and so, even though our intervention was rolled out shortly
before the elections, candidates had sufficient time to adjust their campaigning
strategies. Results presented in Panels A and B of Table 5 indicate that vote buying
22 Recallthat, as indicated in the PAP, we run those regressions at the village-level. We obtain
similar results if we run those regressions at the individual-level instead (Table A.9).
18
Table 4: Effects of Treatment on: Salience of Budgetary Allocations
(1) (2) (3) (4)
Panel A: Salience
Treat 0.159* 0.158* 0.161** 0.168**
(0.096) (0.094) (0.070) (0.068)

Observations 3,346 3,346 3,346 3,346

R-squared 0.003 0.014 0.146 0.150
Panel B: Salience (adjusted)
Treat 0.107* 0.107* 0.109** 0.113**
(0.060) (0.058) (0.044) (0.044)

Observations 3,346 3,346 3,346 3,346

R-squared 0.002 0.013 0.089 0.091
Notes: Results from individual-level regressions. In Panel A, the dependent variable is rating given
to "Whether candidates will spend the municipal budget on things that are important to me and
my family" when the respondent was asked about ’voting influences’. In Panel B, the variable is
adjusted to account for the average rating given to the other categories. In Column 4, the regression
includes a dummy equal to one if someone in the household is a member of any group, a dummy
equal to one if someone in the household participated in any collective action activity in the village
in the past six months, the share of the local budget the respondent would like to spend on water
and the share of the local budget the respondent would like to spend on roads. The standard errors
(in parentheses) account for potential correlation within village. * denotes significance at the 10%,
** at the 5% and, *** at the 1% level.
intensified in treated villages.23

23 The specifications we examined in Table 5 were anticipated in our PAP. However, in the PAP we
predicted that information would reduce vote buying. The theory underlying the PAP anticipated
that the flyers would increase the salience and credibility of candidate promises regarding public
good provision, leading them to substitute away from vote buying in treated areas. However,
although an important part of our treatment was to argue that the PPCRV would monitor whether
new mayors would adhere to their spending promises, nothing about our intervention increased
the potential sanctions that voters could impose on candidates who reneged. For example, the
19
In Panel A, the outcome variable is the share of respondents who were aware
of instances of vote buying in their village. This is an imprecise measure of vote
buying, since voters have incomplete information about whether their neighbours
have been targeted for vote buying. In Panel B, the outcome variable is the share of
respondents who were directly offered money for their votes. The point estimates
are very close in both specifications. However, consistent with the fact that the
variable used in Panel A is a noisier measure of vote buying than the one use in
Panel B, we can only reject the null of no effect in Panel B. The intervention led
to a 3.4 percentage points increase in vote buying (31 percent of the control group
mean).24
5 Accounting for the Results

The fact that an apparently desirable information intervention should have in-
creased vote buying is surprising. This Section offers an explanation for it, and
Section 6 provides a wealth of evidence supporting the explanation. The model
developed here embeds two key characteristics that are central to mayoral elec-
tions in the Philippines: political competition does not center on policy promises,
which are not credible; the mayoral office is a strong one and mayors are often
dominant; and the provision of public goods out of funds provided by the central
government is a key policy for mayors.
Previous research on targeted transfers and vote buying assumes that politi-
cians can make credible commitments to some voters, but not all, and use a proba-
bilistic voting model in which challenger promises influence voter decisions. Ret-
rospective voting models emphasize the key insight that vote buying is linked to
the absence of credible commitment, but assume that politicians cannot make cred-
ible commitments to any voters. Unlike probabilistic voting models, though, ret-
rospective models are a natural vehicle for understanding voter uncertainty about
intervention did not affect households’ capacity to engage in collective action. On the other hand,
the PAP did not anticipate that the flyers would give households new information that would lead
them to update incumbent performance thresholds. It is this effect that we model and test. It is also
important to reiterate that, while the intervention increased vote buying, we argue that this was a
result of an intervention that incumbent incentives to improve voter welfare.
24 A common issue with is how to deal with the limited number of respondents who refuse to
answer the vote buying questions. In the main regression we code ’refuse to answer’ as missing.
We obtain similar results if we code ’refuse to answer’ as yes (Table A.10).
20
Table 5: Effects of Treatment on: vote buying
(1) (2) (3) (4)
Panel A: Are you Aware of Instances of Vote Buying in your Village?
Treat 0.033 0.033 0.033 0.030
(0.029) (0.023) (0.023) (0.025)

Observations 284 284 284 284

R-squared 0.005 0.370 0.688 0.695
Panel B: Did Someone Offered you Money for your Vote?
Treat 0.034 0.034** 0.034** 0.043***
(0.021) (0.016) (0.016) (0.016)


R-squared 0.009 0.478 0.730 0.740
Notes: Results from village-level regressions. In Panel A, the dependent variable is the share of
respondent who indicated being aware of instances of vote buying in their village. In Panel B, the
dependent variable is the share of respondent who indicated that someone attempted to buy their
votes [with ’refused to answer’ coded as ’missing’]. In Column 4, the regression includes the share
of respondents with an household member who belongs to a group, the share of respondent who
participated in any collective action activity in the village in the past six months, the village-average
share of the local budget that respondents would like to spend on water and the village-average
share of the local budget the respondent would like to spend on roads. The standard errors are (in
parentheses). * denotes significance at the 10%, ** at the 5% and, *** at the 1% level.
the benefits that incumbents provide them. Challenger characteristics are irrele-
vant (since challengers cannot credibly commit to pursue different policies than
the incumbent), but voter beliefs about incumbent performance are central.
Retrospective voting models provide a natural vehicle for illustrating the ef-
fects of our intervention, given that the treatment was intended to affect voter be-
liefs about what the incumbent could do for them. The treatment informs voters
about a public spending program with concrete outputs and benefits, about which
21
voters may be ignorant. Incumbents know about our intervention and know that
voters can observe whether or not they have received benefits from the spending
program to which their attention has been drawn. If the voters do not observe any
benefits, and if incumbents believe that the brochure has convinced voters that
they should have observed them, the incumbent will respond by buying votes. The
gap between voter expectations about incumbent performance before and after
seeing the handout should be largest where the incumbent was most successful in
suppressing expectations–where the incumbent was dominant. The effects of the
intervention, therefore, should be greatest in incumbent strongholds. This Section
presents a more formal analysis of this logic.
5.1 Basic Set-Up

We model incumbent advantage as emerging from imperfect voter information
about the public goods that incumbents can provide. In contrast to previous re-
search, in our analysis the information available to voters is a parameter. Prior
research is concerned with the emergence of incumbency advantage and therefore
models voter information as a choice variable. Our question, however, is how an
exogenous information shock affects public policy choices in the presence of an
incumbent advantage.
Under retrospective voting, voters establish a performance threshold for in-
cumbents and vote for or against the incumbent depending on whether the in-
cumbent has met the threshold. If the threshold is too high, incumbents make no
effort to deliver benefits to voters and, instead, maximize private rent-seeking. If
the threshold is too low, voters extract fewer benefits from the incumbent than they
could have. Assuming that voters can spontaneously coordinate on this threshold,
as in Ferejohn (1986) and Persson and Tabellini (2000), their main challenge in set-
ting the threshold is uncertainty about the welfare that the incumbent could have
potentially delivered. Voters’ incomplete information makes it difficult for them to
distinguish incumbent shirking from an unfavorable state of the world that would
keep any incumbent from improving welfare.
We begin with a standard setup (see, e.g., Persson and Tabellini 2000, pp 236
- 238). There are N arbitrarily small groups of voters indexed by i. Incumbent
mayors can spend money either on public goods such as infrastructure, g, or on
22
direct transfers to voters, f i . Since subnational governments in many countries,
including the Philippines, rely on transfers from the central government, the gov-
ernment budget is exogenous and given by M. Public goods deliver welfare H ( g)
to each voter, while transfers deliver welfare equal to the amount of transfers that
the voter receives. The cost of all transfers received by voters is given by ∑ f i .
Our field experiment gave voters information about the public goods that in-
cumbents could provide and about which they were previously unaware. Analyt-
ically, this information could be modeled as shifting voter beliefs either about the
ability of government to finance services of any kind - the government budget con-
straint - or about the relative costs to government of turning budgetary resources
into public goods versus transfers. Since the intervention informed voters only
about the existence of an important program to finance infrastructure, the anal-
ysis below adopts the second approach. However, a model that adopts the first
approach yields the same conclusions that we report below.25
The cost parameter governing public good provision is given by θ̄ and total
costs of providing public goods are given by θ̄g. The cost is higher when there
are restrictions on the type of public goods that can be purchased, when the costs
of inputs and construction are high, or when the bureaucracy is incompetent. As
long as the costs θ̄ are not too high, government decisions to spend more on local
public infrastructure delivers greater welfare to voters per peso of spending than
does spending on direct transfers.
Mayors therefore choose direct transfers and public good spending to maxi-
mize their pecuniary rents, r = M − ∑ N f i − θ̄g, and the non-pecuniary rents from
being re-elected, R:
M − ∑ f i − θ̄g + pR
N
where p is the probability of re-election. In the event that they do not expect to be
25 For beliefs about the size of the government budget to matter, voters must believe that the par-
ticipation constraint of the incumbent binds when they choose the performance threshold. Condi-
tional on this, the incumbent provides the public goods consistent with the performance threshold;
when voters are unexpectedly informed that the budget is larger than they thought, support for
the incumbent drops unless the incumbent makes targeted transfers to some voters. The identity of
those voters is established in Proposition 2, below-the conclusions of the proposition are indepen-
dent of whether the information shock affects voter beliefs about the budget or about the relative
costs of providing public goods versus transfers.
23
re-elected, they set g = f = 0 and take as pecuniary rents the entire budget.
The welfare of voters in (arbitrarily small) group i is given by ω = f i + H ( g).
Voters prefer that the mayor dedicates the municipal budget to public goods until
Hg ( g) = Nθ̄ , the Samuelsonian condition for public good provision, and then to
distribute any remaining budget in the form of transfers.
We add three features to this standard set-up. First, for most public goods,
such as infrastructure, spending takes time to implement before voters perceive a
change in their welfare. Mayors must therefore decide to spend money on public
goods early in their terms in order to ensure that it has an electoral impact (Robin-
son and Torvik, 2005) Transfers, however, can be implemented quickly, even at the
end of the mayor’s term, right before the next election. Mayors have two oppor-
tunities, then, to make budget decisions. Earlier in their tenure, they can decide
to supply public goods or transfers (though, for any expenditure amount, public
goods deliver greater welfare to voters). Late in their tenure, they can only deliver
transfers. This accurately reflects the limitations on incumbents’ ability to react to
information shocks in the weeks before an election.
Second, voters are uncertain about the costs to the incumbent of providing
them with public goods. Just before the election, each voter’s beliefs about about
the costs of producing public goods are drawn from a uniform distribution given
by θi ∼ [1, 2θc − 1], θc > 1. Incumbents know this distribution, but not the beliefs
of individual voters. The median belief about the incumbent’s costs of producing
public goods is given by the cost parameter θc . The ability to produce is never less
than one - it can never cost less than g to produce g.
Our intervention is equivalent to an unexpected shock that shifts this distribu-
tion for a randomly-selected fraction δ of all voters, δ ≤ 1 (approximately 44.5 per-
cent in the case of our treatment). Incumbents know which voters are subject to the
shock, but beyond that only know that the distribution of beliefs about the costs
of producing public goods follows θi0 ∼ [1, 2θc0 − 1], where θc0 = θc + k θ̄ − θc ,

k ∼ [−1, 1]. Recalling that citizens do not know θ̄, the true cost of producing pub-
lic goods, the effect of the information shock reflects the assumption that the more
accurate the beliefs θc of citizens regarding the costs of public good provision, the
less they change in the event of a shock. Where this gap between the true cost of
providing public goods and the cost perceived by the median voter is large, the in-
formation shock has a potentially larger effect on voter beliefs; where it is small, it
24
does not. This is consistent with our experimental intervention, since we provided
voters with the “true” ability of politicians to provide public goods; those voters
who knew this already were therefore unaffected by the intervention.
The information shock in our field experiment, and in the model here, is unan-
ticipated. Hence, incumbents do not take it into account when deciding on public
goods.26
As usual in retrospective voting models, citizens coordinate on a voting rule
that is conditional on their beliefs about the costs of public good production just
before the election, after the mayor has provided public goods. Here, the voting
rule that voters establish at the beginning of the mayor’s term is that, given their
individual draw from the distribution of potential pre-electoral beliefs, θi0 , about
the costs of public good production, they will support the incumbent who meets
2θ 0
the performance threshold ω̄i ≥ H ( gθ 0 ), where gθ 0 is given by Hg ( gθ 0 ) = Ni .27
The third feature of the set-up is that it provides an immediate link to the liter-
ature on incumbent advantage. Incumbents with a significant advantage are those
for whom voters’ beliefs about the costs they confront are above their true costs of
providing public goods. Recall that voters draw their pre-electoral beliefs from the
distribution θi ∼ [1, 2θc − 1]. Where incumbents have an information advantage,
θc > θ̄.
The stages of the game are therefore the following:
26 We abstract from anticipated information shocks. Their inclusion would complicate the anal-
ysis, but not change the key results. An anticipated shock would take the form of some ran-
dom variable z that would change the cost parameter in the distribution of beliefs according to
θc + z θ̄ − θc . As in standard retrospective voting models, voters and incumbents would be
aware of the distribution of z and anticipate the possibility of the shock in the construction of their
performance thresholds and decisions regarding public goods. Again, however, once equilibrium
public goods are provided and the unanticipated shock occurs, the dynamics of vote buying remain
the same: in the event of negative shocks (that reduce voter beliefs about the costs of providing pub-
lic goods), as long as they are not too large, the incumbent buys votes from the voters who were
subjected to the shocks.
27 In the usual retrospective voting model, both an economic shock and government policy affect
voter welfare; voters do not observe either, but take the distribution of the shock into account when
setting a performance threshold for the incumbent. The shock occurs, observed by the incumbent,
but not the voter, and then the incumbent makes policy. In the analysis here, neither politicians nor
voters observe the information that voters will have about politician ability before politicians must
make decisions about public good provision. Politicians can therefore not exploit an information
asymmetry between themselves and voters, as in the canonical model of retrospective voting: there
is no asymmetry at the time that they decide on public good spending. Voters, therefore, can do
no better than to require politicians to meet the performance threshold that is indicated by the
revelation of θ 0 , voters’ best information about the true efficiency of public good provision.
25
1. Incumbents and voters observe the distribution of beliefs about the costs of
public good provision, θi ∼ [1, 2θc − 1], that voters will have before the elec-
tion.
2. Voters coordinate on a voting rule ω̂ = ω ( gi ), where gi is given by Hg ( gi ) =

2θi
N.
3. Incumbents choose the level of public good provision g.28
4. A randomly-selected subset of all voters δ ≤ 1 are subject to an unanticipated

shock k to the distribution of their beliefs about the costs of producing public
goods, such that for these voters θi0 ∼ [1, 2θc0 − 1], where θc0 = θc + k θ̄ − θc ,

k ∼ [−1, 1].
5. Incumbents choose the level of spending on transfers to voters.
6. Voters’ individual beliefs about the costs of public good provision are re-
vealed to them.
7. The election takes place.

Proposition 1 establishes the equilibrium level of public good provision, and that
and then the conditions under which vote buying emerges. The remainder of the
analysis then describes the conditions under which vote buying takes place, the
amount of vote buying, and the voters targeted for it.
Proposition 1 Incumbents set public good provision to meet the expected performance
threshold given the voting rule, ω̄ = H ( gθc ), where public good provision is given by
Hg ( gθc ) = 2θNc .
Proof: See technical appendix.
Lemma 1 simply confirms the effects on voter support of unanticipated infor-

mation shocks that increase or reduce voter expectations about the costs of provid-
ing public goods, given the public goods that the incumbent chose to provide.
28 First,when voters observe public good spending g, from the participation constraint of the in-
cumbent they can infer an upper limit on the cost of providing public goods, θ ≤ Rg . The voters
who believed that the cost was higher than this immediately update their beliefs about costs. How-
ever, this updating does not change their voting behavior, since incumbent spending that satisfies
the performance threshold of voters who believe the costs were θ by necessity satisfies those who
believe the costs were higher, and who set a lower performance threshold.
26

Lemma 1 After a positive unanticipated information shock, k θ̄ − θc > 0, increasing
the beliefs among a fraction of voters δ about the costs of providing public goods, the public
goods provided by the incumbent meet the performance threshold of more than half of the
voters. After a negative unanticipated information shock affecting a fraction of voters δ,

k θ̄ − θc < 0, public good provision meets the threshold of less than half of the voters.
Proposition 2 describes the incumbent response to an information shock. In

particular, if the shock is adverse (it tells voters that it is less expensive to provide
public goods than the incumbent anticipated), incumbents have an incentive to
target those affected by the shock rather than others, targeting a larger fraction of
those voters when their information advantage is larger.

Proposition 2 After a positive unanticipated information shock, k θ̄ − θc > 0, there
is no change in public policy. In the event of a negative unanticipated information shock,

incumbents target transfers f k = H ggθc +k − H ( gθc ) to a fraction α of voters in δwho re-

∗ M−θ̄gθc + R+lδ f k 1 k (θ̄ −θc )
ceived the information shock, where α is given by α = δ fk ,l = 2
θc +k (θ̄ −θc )−1
The model accounts for the puzzling result from Section 4, that more informa-
tion can trigger greater vote buying: voters who receive more accurate information
about the public goods the incumbent could have provided raise their performance
threshold, in accordance with the voting rule. Since incumbents cannot adjust
the provision of public goods in time for the election, they respond to the higher
threshold with vote buying. This result is only possible if public good spending
begins substantially before the election, while transfers can be made right before
the elections. Evidence from the Philippines, discussed earlier, supports these as-
sumptions.
Section 6 offers evidence for other assumptions and ancilliary predictions of the
model. First, from our assumption that politicians cannot make credible commit-
ments, following the logic of retrospective voting models, voters care most about
incumbent characteristics and performance. We should therefore observe stronger
27
information effects with respect to incumbents than to challengers on voter knowl-
edge of candidates and on candidate vote buying. These stronger effects are found
in the data.
Second, we assume that the information advantage of incumbents emerges
from the degree to which voter beliefs about the costs of providing public goods
are higher than the actual costs. Especially in the Philippines, it is reasonable that
dominant incumbents might control the information reaching voters in their ju-
risdictions. For example, journalists in the Philippines are subject to significant
pressure relative to other countries. Freedom House (2014) report on press free-
dom rates the Philippines as only partly free, with journalists the target of libel
suits by local politicians, harassment and assassination attempts (many success-

ful). The term k θ̄ − θc embeds our assumption that when the difference between
voter beliefs and the actual costs of public good provision is greater, as with domi-
nant incumbents, information shocks should have a larger impact on voter beliefs.
This is consistent with the evidence below.
In addition, it is straightforward to show that the shocks should have larger
effect on vote buying. Differentiating α∗ with respect to incumbent advantage,
∗
θ̄ − θc , is equivalent to ∂α ∂l ∂l
∂l ∂(θ̄ −θc ) . The term ∂(θ̄ −θc ) describes the effects of in-
cumbent advantage on the fraction of voters whose beliefs are shifted by the infor-
k (θ̄ −θc )
mation shock. Since for k θ̄ − θc < 0 ,− 12

increases in the distance
θc +k (θ̄ −θc )−1
∂l
between θ̄ and θc ,
∂(θ̄ −θc )
> 0: the fraction increases when incumbent advantage
is greater–more voters are disappointed. Then differentiating α∗ with respect to l
equals one: conditional on meeting the participation constraint, the share of vot-
ers α∗ whose votes are bought rises one for one with the fraction of voters who are
dissuaded from supporting the incumbent by the information shock. The fraction
of voters whose vote is bought therefore rises with incumbent advantage. We find,
indeed, that vote buying is higher in incumbent strongholds.
6 Results: Mechanisms and Alternative Explanations

This Section systematically explores the heterogeneous effects of information on
voter knowledge and vote buying, comparing incumbents and challengers and
dominant and non-dominant incumbents.
28
6.1 Heterogeneous Effects on Knowledge and Vote Buying across
Incumbents and Challengers
Our interpretation of the effects of the information shock, formalized in the model,
implies that incumbents should be more affected than challengers. First, incum-
bents, but not challengers, should exhibit significantly fewer information advan-
tages among informed than among uninformed voters. Second, incumbents, but
again, not challengers, should react to the information shock with greater vote
buying. The data are consistent with both of these predictions.
To see the contrasting effects on voter knowledge, we create a variable that cap-
tures whether respondents make a mistake in identifying which candidate promised
to spend the greatest share of the development budget on some sector s. We clas-
sify any error made by the respondent as favoring the incumbent when the respon-
dent claims that the incumbent promised to spend the greatest share, but actually
the challenger did; and favoring the challenger in the reverse case. Consistent
with an information-based theory of incumbency advantage, on average, five of
the seven errors made by the average respondent favor the incumbent.
We estimate equation (1) where Yijk is either the number of errors that favor the
incumbent or the number of errors made that favor the challenger. The results dis-
played in Table 6 indicate that we can reject the null hypothesis of no effect for the
incumbent-favoring errors but not for the challenger-favoring errors. The point es-
timate of the effect of the information shock on incumbent-favoring errors is more
than eight times larger than point estimate for challenger-favoring errors. Treated
households–more informed households–made significantly fewer errors favoring
the incumbent, consistent with our earlier arguments that incumbent advantage is
related to voter over-estimates of incumbent contributions to voter welfare (or, in
terms of model parameters, that voters in incumbent strongholds are particularly
likely to over-estimate the costs of providing public goods).29
By assumption, the model predicts that challengers do not react to the infor-
mation shock. The information effects on respondent knowledge about incum-
bents and challengers support this assumption. Nevertheless, it is possible that it
is not incumbents, but challengers, who increase their vote buying in incumbent
29 Importantly, as reported in Table A.11, incumbent and challenger promises do not seem to
differ systematically, which could explain those results.
29
Table 6: Effects of Treatment on : Errors
(1) (2) (3) (4)
Panel A: Incumbent-Favoring Errors
Treat -0.211 -0.211* -0.211** -0.218**
(0.281) (0.120) (0.089) (0.090)

Observations 3,408 3,408 3,408 3,408

R-squared 0.001 0.477 0.527 0.528
Panel B: Challenger-Favoring Errors
Treat -0.0250 -0.0250 -0.0250 -0.0260
(0.216) (0.177) (0.118) (0.118)

Observations 3,408 3,408 3,408 3,408

R-squared 0.001 0.117 0.253 0.254
Notes: Results from individual-level regressions. In Panel A, the dependent variable is the number
of errors made by the respondent about candidate promises that were favoring the incumbent. In
Panel B, the dependent variable is the number of errors made by the respondent about candidate
promises that were favoring the challenger. In Column 4, the regression includes a dummy equal
to one if someone in the household is a member of any group, a dummy equal to one if someone in
the household participated in any collective action activity in the village in the past six months, the
share of the local budget the respondent would like to spend on water and the share of the local
budget the respondent would like to spend on roads. The standard errors (in parentheses) account
for potential correlation within village. * denotes significance at the 10%, ** at the 5% and, *** at the
1% level.
strongholds. We did not ask respondents which candidates bought their votes be-
cause this would have been too sensitive.
However, we implemented a secret ballot to find out from respondents whom
they voted for. If the information treatment primarily affected incumbents, then
we would expect treated areas to express less support for the incumbent, but
those whose votes were bought should express more support for the incumbent.
These effects would not emerge if the treatment affected both incumbents and chal-
30
lengers, or only challengers. We therefore estimate equation (1), where Yijk is a
dummy equal to one if the respondent declared voting for the incumbent. We con-
trol for treatment status, whether the individual was targeted for vote buying, and
their interaction. The results, in Table 7, are consistent with our argument that the
information provided to voters unexpectedly raised their performance threshold
for incumbents and that incumbents reacted by buying more votes. First, the treat-
ment is associated with lower support for the incumbent. Second, the interaction
term is positive and we are unable to reject the null that the sum of the treatment
dummy and the interaction term is zero.
Table 7: Effects of Treatment on : Self-reported vote for the incumbent

(1) (2) (3) (4)
Treat -0.030 -0.041** -0.041** -0.041**
(0.036) (0.019) (0.019) (0.019)
Vote-buying -0.156*** -0.001 -0.001 0.004
(0.037) (0.034) (0.034) (0.034)
Interaction 0.043 0.039 0.039 0.032
(0.056) (0.049) (0.049) (0.048)

Observations 2,882 2,882 2,882 2,881

R-squared 0.012 0.314 0.314 0.319
Notes: Results from individual-level regressions. The dependent variable is a dummy equal to
one if the respondent indicated voting for the incumbent. In Column 4, the regression includes a
dummy equal to one if someone in the household is a member of any group, a dummy equal to one
if someone in the household participated in any collective action activity in the village in the past
six months, the share of the local budget the respondent would like to spend on water, the share of
the local budget the respondent would like to spend on roads, alignment between the respondent
and the incumbent, how long the respondent has lived in her current village of residence, family
size, respondent’s age, whether the respondent receive remittances from abroad and whether the
respondent benefit from a large-scale CCT program. The standard errors (in parentheses) account
for potential correlation within village. * denotes significance at the 10%, ** at the 5% and, *** at the
1% level.
31
6.2 Heterogeneous Effects on Knowledge and Vote Buying across
Dominant and Non-dominant Incumbents
The model further predicts that information shocks should matter most where in-
cumbents are dominant and where, as a consequence, their pre-intervention in-
formation advantage was greatest. To test this prediction, we examine the role of
incumbent dominance, proxied by incumbent vote share in the previous munici-
pal elections that took place in May 2010.30 We look at dominance measured in
terms of both barangay (village) and municipal electoral results. Vote share, in
turn, is measured as a percentage of the registered population.31 To facilitate inter-
pretation, the interacted variables are demeaned so the coefficient on the treatment
dummy still captures the average treatment effect.
The estimates again support the hypothesis that the information treatment has
the strongest in municipalities with dominant incumbents (Table 8). The posi-
tive impact of the treatment on knowledge of campaign promises is significantly
higher in incumbent-dominated municipalities and villages. Moreover, the effect
is driven, as we would expect, by a reduction in the number of incumbent-favoring
errors. There is no associated reduction in the number of challenger-favoring er-
rors.
The model also predicts that the intervention should have larger effects on vote
buying in incumbent-dominated villages and municipalities.32 The data provide
substantial evidence of the predicted effect. Proxying incumbent strongholds ac-
cording to incumbent vote share in the 2010 elections, a one standard deviation
in incumbents’ 2010 vote share increases the impact of the intervention on vote
buying by more than 10 percent.
These results differ from those in prior research in ways that are consistent with
30 Since the theory predicts that dominant incumbents provide fewer public goods than non-
dominant incumbents, a natural empirical exercise would be to compare effort across these two
incumbent classes with respect to spending out of the Local Development Fund. This would require
spending data across the different municipalities, but also data on the quality of the spending
(the actual projects that were implemented). We hope to undertake this significant data-gathering
exercise in the future.
31 Results are similar when using vote share defined as a percentage of the voting population
(Table A.12).
32 This effect would be muted if incumbent politicians are not able to adjust vote buying eas-
ily. In fact, vote buying in the Philippines is conducted by local brokers with close connections
to politicians, giving politicians significant logistical capacity to adapt vote buying to changing
circumstances in their strongholds (Cruz et al., 2014).
32
Table 8: Heterogeneity
Know Errors Salience Vote
Promises Incumbent Challenger Buying
(1) (2) (3) (4) (5)
Panel A: 2010 Incumbent Vote Share (Municipal)
Treat 0.051*** -0.211** -0.025 0.109** 0.034**
(0.015) (0.088) (0.118) (0.044) (0.015)
Interaction 0.007*** -0.024*** 0.002 0.006 0.004***
(0.001) (0.008) (0.012) (0.004) (0.001)
Observations 3,408 3,408 3,408 3,346 284

R-squared 0.330 0.528 0.253 0.090 0.742
Panel B: 2010 Incumbent Vote Share (Barangay)
Treat 0.051*** -0.212** -0.039 0.104** 0.035**
(0.015) (0.090) (0.117) (0.043) (0.015)
Interaction 0.006*** -0.019*** 0.000 0.009** 0.003**
(0.001) (0.006) (0.009) (0.004) (0.001)
Observations 3,408 3,408 3,408 3,346 284

R-squared 0.330 0.528 0.255 0.093 0.741
Notes: Results from individual-level (Columns 1-4) )and village-level (Column 5) regressions. All
regression include pair dummies. In Column 1, the dependent variable is an index capturing the
respondent’s knowledge of candidate promises. In Column 2, the dependent variable is the number
Column 3, the dependent variable is the number of errors made by the respondent about candidate
promises that were favoring the challenger. In Column 4, the dependent variable is rating given
my family" when the respondent was asked about ’voting influences’. The variable is adjusted to
account for the average rating given to the other categories. In Column 5, the dependent variable is
the share of respondent who indicated that someone attempted to buy their votes [with ’refused to
answer’ coded as ’missing’] . The standard errors (in parentheses) account for potential correlation
within village (Columns 1-4). * denotes significance at the 10%, ** at the 5% and, *** at the 1% level.
our explanation of the effects we identify. The information interventions of Chong

et al. (2011) in Mexico and of Vicente (2014) in Sao Tome and Principe both re-
duced turnout. In contrast, in our experiment, the information intervention led
dominant incumbents to increase their use of the main mobilization tactic at their
disposal–vote buying. Vote buying by dominant incumbents offset the effects of
the information shock on the performance threshold that voters establish for in-
33
cumbent re-election, leading to no net change in voter behavior (either with respect
to turnout or to vote shares).33
6.3 Robustness Checks and Alternative Explanations

We contend that our main results and heterogenous treatment effects are explained
by a reduction in the informational advantage that gives incumbents an electoral
advantage. The evidence is robust both for effects on political knowledge (the
intervention has an unfavorable effect on knowledge about incumbents, and the
effect is strongest in municipalities with dominant incumbents) and on politicians’
reactions (vote buying rises in treated areas, with the largest increases in munici-
palities with dominant incumbents). It is nevertheless possible that the interven-
tion affected voters and candidates through other channels and that dominant in-
cumbents were better able to react to the intervention.
One alternative explanation for our findings is that challenger promises regard-
ing infrastructure spending are actually credible and are more appealing to voters
than incumbent promises in incumbent strongholds. For example, dominant in-
cumbents may believe that they are more insulated from competition and better
able to spend resources according to their personal preferences and not those of
their constituents. When a challenger emerges who promises to spend in accor-
dance with constituent preferences, incumbents are forced to react by buying more
votes. If this were the case, however, our results would disappear after controlling
for the overlap between candidate promises. In Panel A of Table 9 we report results
of a specification that includes two additional terms compared to those of Panel A
in Table 8: the measure of the overlap between candidate promises (the share of
the budget on which the candidates agree), and the interaction of this variable with
the treatment dummy with a measure of overlap between the candidate promises.
33 These two hypotheses were also included in the PAP and are reported in Table A.13. The
differences are likely related to the nature of the information provided. Voters in Mexico were
provided with salient information about corruption cases, reducing voter expectations about the
effectiveness of electoral accountability in improving voter welfare and, therefore, their incentive to
vote. The information provided in Sao Tome and Principe explicitly discouraged vote buying, but
not other features of the electoral system, and therefore may have reduced the benefits that voters
expected to receive from electoral participation, again suppressing turnout. Our intervention was
neutral with respect to the desirability or feasibility of vote buying, but not neutral with respect to
the effects on voter expectations regarding incumbent performance. Its effects on vote buying and
voter turnout were correspondingly different.
34
The inclusion of these additional terms does not affect our results.
A second alternative explanation for our findings is that dominant incumbents
are of better quality and therefore better able to react to the intervention. Again,
this explanation does not account for the disproportionate effects of the interven-
tion on political knowledge in incumbent strongholds. In addition, if this were the
mechanism accounting for our results, the interaction of the treatment dummy and
incumbent stronghold would disappear after accounting for incumbent quality. To
test this, we further interact the treatment dummy with measures of incumbent’s
education levels and affiliations with national and provincial politicians. In the re-
sults reported in Panel B of Table 9, however, the interaction of treatment dummy
and incumbent stronghold remain significant.
7 Conclusion
We present results from a unique intervention that provided voters with informa-
tion about the existence and importance of a large public spending program, the
types of services the program could finance, and candidate priorities and promises
regarding the program just prior to the May 2013 municipal elections in the Philip-
pines. The intervention led to significant changes in voter knowledge about in-
cumbents and also led candidates to expend more resources on vote buying. We
account for these results with a new model of vote buying and incumbent infor-
mation advantage in an environment where candidates cannot make credible com-
mitments. Information shocks that raise voters’ thresholds for incumbent perfor-
mance shortly before an election oblige incumbents to do more to increase voter
welfare than they anticipated. With little time before the election to improve the
provision of public goods, incumbents turn to vote buying.
The theory predicts that the effects of the intervention on voter beliefs and
vote buying should be strongest in incumbent strongholds. In fact, voter beliefs
about incumbents change more than their beliefs about challengers and the effect
is strongest for dominant incumbents. In addition, vote buying significantly in-
creased in areas exposed to this unanticipated information shock, and that effect
was, again, greatest in incumbent strongholds.
The results raise questions for future research. Our intervention took place
shortly before the election, which we argue is the reason that it increased vote buy-
35
Table 9: Alternative Channels
(1) (2) (3) (4) (5)
Panel A: Controlling for Overlap between Candidate Promises
Treat 0.050*** -0.209** -0.025 0.109** 0.034**
(0.015) (0.088) (0.118) (0.044) (0.015)
Treat*Strongholds 0.007*** -0.024*** 0.003 0.006 0.004***
(0.001) (0.007) (0.012) (0.004) (0.001)
Observations 3,408 3,408 3,408 3,346 284

R-squared 0.330 0.528 0.253 0.090 0.743
Panel B: Controlling for Incumbent Quality
Treat 0.050*** -0.210** -0.023 0.109** 0.034**
(0.015) (0.088) (0.117) (0.044) (0.015)
Treat*Strongholds 0.007*** -0.028*** 0.002 0.006 0.003**
(0.002) (0.008) (0.013) (0.005) (0.002)
Observations 3,408 3,408 3,408 3,346 284

R-squared 0.331 0.529 0.255 0.090 0.743
Notes: Results from individual-level regressions. All regression include pair dummies. In Panel
A, regressions also control for the interaction between the treatment dummy and overlap between
candidate promises. In Panel B, regressions also control for the interaction between the treatment
dummy and incumbent education and affiliations with national and provincial politicians. In Col-
umn 1, the dependent variable is an index capturing the respondent’s knowledge of candidate
promises. In Column 2, the dependent variable is the number of errors made by the respondent
about candidate promises that were favoring the incumbent. In Column 3, the dependent variable
is the number of errors made by the respondent about candidate promises that were favoring the
challenger. In Column 4, the dependent variable is rating given to "Whether candidates will spend
the municipal budget on things that are important to me and my family" when the respondent was
asked about ’voting influences’. The variable is adjusted to account for the average rating given
to the other categories. In Column 5, the dependent variable is the share of respondent who indi-
cated that someone attempted to buy their votes [with ’refused to answer’ coded as ’missing’]. The
standard errors (in parentheses) account for potential correlation within village (Columns 1-4). *
denotes significance at the 10%, ** at the 5% and, *** at the 1% level.
ing. Additional research is needed to assess an important corollary of this argu-

ment, that if the intervention had occurred earlier in the electoral cycle (or at least
if incumbents knew earlier that the intervention would take place), it might have
prompted incumbents to provide more public goods, with no change, or even a
36
reduction, in vote buying. In addition, the information in the intervention related
primarily to local infrastructure. A further open question is whether information
about service delivery, such as the quality of health facilities or the effectiveness
of schools, would have elicited similar responses with respect to voter knowledge
and politician vote buying,
Our findings also have implications for improving the accountability effects of
elections in developing countries. It demonstrates that voters are poorly informed
about what politicians can do for them and that relatively simple information inter-
ventions have a significant effect on this information asymmetry. Moreover, since
the asymmetry tends to reduce the incentives of incumbents to improve citizen
welfare, such an intervention has potential welfare effects. Consistent with this,
incumbents in our treatment area made significant attempts to increase voter wel-
fare. In our setting, where their time for reaction was short and only vote buying
was feasible, they significantly increased vote buying in areas where voters were
better informed.
37
References
A KER , J., P. C OLLIER , AND P. V ICENTE (2011): “Is Information Power? Using Cell
Phones during an Election in Mozambique,” mimeo.
A NSOLABEHERE , S., E. C. S NOWBERG , AND J. M. S NYDER , J R . (2006): “Television

and the Incumbency Advantage of U.S. Elections,” Legislative Studies Quarterly,
31, 469–490.
B ANERJEE , A., S. K UMAR , R. PANDE , AND F. S U (2011): “Do Informed Voters

Make Better Choices? Experiment Evidence from Urban India,” MIT, mimeo.
B OAS , T. AND F. D. H IDALGO (2011): “Controlling the Airwaves: Incumbency

Advantage and Community Radio in Brazil,” American Journal of Political Science,
55, 869–885.
B RUNH , M. AND D. M C K ENZIE (2009): “In Pursuit of Balance: Randomization in

Practice in Development Field Experiments,” American Economic Journal: Applied
Economics, 1, 200–232.
B RUSCO , V., M. N AZARENO , AND S. S TOKES (2009): “Vote Buying in Argentina,”

Latin American Research Review, 39, 66–88.
C APUNO , J. (2012): “The PIPER Forum on 20 Years of Fiscal Decentralization: A

Synthesis,” Philippine Review of Economics, 49, 191–202.
C HONG , A., A. L. D E L A O, D. K ARLAN , AND L. WANTCHEKON (2011): “Looking

Beyond the Incumbent: The Effects of Exposing Corruption on Electoral Out-
comes,” NBER Working Paper 17679.
C RUZ , C. (2013): “Social Networks and the Targeting of Vote Buying,” Annual
Meeting of the American Political Science Association.
C RUZ , C., J. L ABONNE , AND P. Q UERUBIN (2014): “Politician Family Networks

and Electoral Outcomes: Evidence from the Philippines,” Annual Meeting of the
American Political Science Association.
C RUZ , C. AND C. S CHNEIDER (2013): “The (Unintended) Electoral Effects of Mul-

tilateral Aid Projects,” University of California - San Diego, mimeo.
F EREJOHN , J. (1986): “Incumbent performance and electoral control,” Public

Choice, 50, 5–26.
F INAN , F. AND L. S CHECHTER (2012): “Vote-Buying and Reciprocity,” Economet-

rica, 80, 863–882.
F REEDOM H OUSE (2014): Freedom of the Press 2014.
38
F UJIWARA , T. AND L. WANTCHEKON (2013): “Can Informed Public Deliberation
Overcome Clientelism? Experimental Evidence from Benin,” American Economic
Journal: Applied Economics, 5, 241–255.
G ORDON , S. C., G. A. H UBER , AND D. L ANDA (2007): “Challenger Entry and

Voter Learning,” American Political Science Review, 101, 303–320.
H ANUSCH , M. AND P. K EEFER (2012): “Promises, Promises: Political Budget Cy-

cles, Parties and Vote-buying.” World Bank Policy Research Working Paper 6653.
H ICKEN , A., S. L EIDER , N. R AVANILLA , AND D. YANG (2014): “Temptation in

Vote-Selling: Evidence from a Field Experiment in the Philippines,” University of
Michigan, mimeo.
H ODLER , R., S. L OERTSCHER , AND D. R OHNER (2010): “Inefficient Policies and

Incumbency Advantage,” Journal of Public Economics, 94, 761–767.
H UMPHREYS , M. AND J. W EINSTEIN (2013): “Policing Politicians: Citizen Empow-

erment and Political Accountability in Uganda - Preliminary Analysis,” Working
Paper, International Growth Center.
H UTCHCROFT, P. (2012): “Re-slicing the pie of patronage: the politics of internal

revenue allotment in the Philippines, 1991-2010,” Philippine Review of Economics,
49, 109–134.
H UTCHCROFT, P. AND J. R OCAMORA (2003): “Strong Demands and Weak Institu-

tions: The Origins and Evolution of the Democratic Deficit in the Philippines,”
Journal of East Asian Studies, 3, 259–292.
K ASNJA , M. (2011): “Why Do Malfeasant Politicians Maintain Political Support?

Testing the Uninformed Voter Argument,” New York University, mimeo.
K EEFER , P. AND S. K HEMANI (2014): “Radio’s Impact on the Support for Clien-
telism,” World Bank Research Group, mimeo.
K EEFER , P. AND R. V LAICU (2008): “Democracy, Credibility, and Clientelism,”

Journal of Law, Economics and Organization, 24, 371–406.
K ERKVLIET, B. J. (2002): Everyday Politics in the Philippines. Class and Status Relations
in a Central Luzon Village, Rowman and Littlefied Publishers.
K HEMANI , S. (2013): “Buying Votes vs. Supplying Public Services: Political Incen-
tives to Under-Invest in Pro-Poor Policies,” World Bank Policy Research Working
Paper 6339.
K ITSCHELT, H. (2000): “Linkages Between Citizens and Politicians in Democratic

Polities,” Comparative Political Studies, 33, 845–879.
39
K LASNJA , M. (2014): “Corruption and Incumbency Disadvantage: Theory and
Evidence from Romania,” Mimeo.
——— (forthcoming): “Increasing Rents and Incumbency Disadvantage,” Journal
of Theoretical Politics.
K LASNJA , M. AND R. T ITIUNIK (2013): “Incumbency Disadvantage in Weak Party
Systems: Evidence from Brazil,” mimeo.
K LING , J., J. L IEBMAN , AND L. K ATZ (2007): “Experimental Analysis of Neigh-
borhood Effects,” Econometrica, 75, 83–119.
L ABONNE , J. (2013): “The Local Electoral Impacts of Conditional Cash Transfers:
Evidence from the Philippines,” Journal of Development Economics, 104, 73–88.
L LANTO , G. M. (2012): “The Assignment of Functions and Intergovernmental Fis-
cal Relations in the Philippines 20 Years after Decentralization,” Philippine Review
of Economics, 49, 37–80.
M AC D ONALD , B. (2014): “Incumbency Disadvantages in African Politics? Re-
gression Discontinuity Evidence from Zambian Elections,” London School of Eco-
nomics - mimeo.
N ICHTER , S. (2008): “Vote Buying or Turnout Buying? Machine Politics and the
Secret Ballot,” American Political Science Review, 102, 19–31.
P ERSSON , T. AND G. TABELLINI (2000): Political Economics: Explaining Economic
Policy, Cambridge, MA: MIT Press.
R OBINSON , J. A. AND R. T ORVIK (2005): “White Elephants,” Journal of Public Eco-
nomics, 89, 197–210.
S CHAFFER , F. C. (2007): Elections For Sale. The Causes and Consequences of Vote Buy-
ing, Quezon City: Ateneo de Manila University Press.
S IDEL , J. (1999): Capital, Coercion, and Crime: Bossism in the Philippines, Contempo-
rary Issues in Asia and Pacific, Stanford, CA: Stanford University Press.
S TOKES , S., T. D UNNING , M. N AZARENO , AND V. B RUSCO (2013): Brokers, Voters,
and Clientelism. The Puzzle of Distributive Politics, Cambridge University Press.
T ROLAND , E. (2014): “Do Fiscal Transfers Increase Local Revenue Collection? Ev-
idence From The Philippines,” UCSD, mimeo.
U PPAL , Y. (2009): “The Disadvantaged Incumbents: Estimating Incumbency Ef-
fects in Indian State Legislatures,” Public Choice, 138, 9–27.
V ICENTE , P. (2014): “Is Vote Buying Effective? Evidence from a Field Experiment
in West Africa,” Economic Journal, 124.
40
Technical Appendix
In this technical appendix we report the proofs of the Lemma and Propositions
discussed in Section 5.
Proof of Proposition 1
The proof follows from the canonical model in Persson and Tabellini (2000). Re-
call that the information shock is unanticipated. Hence, voters coordinate on the
pre-electoral performance threshold according to their expected individual beliefs
about the costs of providing public goods, drawn from θi ∼ [1, 2θc − 1]: for all vot-
ers, the expected cost of providing public goods is given by θc . Voters would most
prefer to set the performance threshold to require that public goods be provided
at the Samuelsonian optimum or, given their expected beliefs, at Hg ( gθc ) = θNc .
The performance threshold of the median voter would then be ω̄ = H ( gθc ), where
Hg ( gθc ) = θNc . However, voters anticipate that incumbents can marginally reduce
public goods, saving incumbents θc in expectation, thereby reducing the utility
of each voter by θNc . Incumbents can then offset this welfare loss for N2 voters
by offering transfers to them that total N2 θNc = θ2c <θc . This tradeoff continues
to be feasible for the incumbent, in expectation, until public good provision falls to
Hg ( g∗ ) = 2θNc and the cost of using transfers to offset the welfare losses from ad-
ditional marginal reductions in public good provision exactly equals the reduced
cost of providing public goods, N2 2θNc = θc . The provision of g∗ is feasible as long as
it is less than gmax , defined by the incumbent’s participation constraint, including
the actual costs of providing public goods, M − θ̄gmax + R ≥ M. The performance
threshold sets transfers to zero since, as in Persson and Tabellini (2000), voters an-
ticipate that competition between voters to be part of the majority that receives
these transfers drives actual redistributive transfers to zero.
Proof of Lemma 1
Based on the performance threshold ω̄ = H ( gθc ) incumbents provided gθc . In
the event of an unanticipated shock, a fraction δ of voters have beliefs distributed
according to θi0 ∼ [1, 2θc0 − 1], where θc0 = θc + k θ̄ − θc , and the remaining (1 − δ)

voters have beliefs distributed as before, θi ∼ [1, 2θc − 1]. Therefore, one-half of the
41
voters who were not subjected to the information shock, given by 12 (1 − δ) < 21 ,
are expected to conclude that the incumbent met their performance threshold, as
before.

Case 1: The unanticipated shock is positive (k θ̄ − θc > 0). The shock shifts up
the median of the distribution of beliefs about the costs of providing public goods
among a fraction δ of voters. Consequently, among the δ fraction of voters exposed
to the shock, the incumbent’s performance will, in expectation, meet the threshold
for some voters for whom it previously did not. Recalling that their beliefs are now
distributed according to θi0 ∼ 1, 2 θc + k θ̄ − θc − 1 , the fraction of voters in δ

for whom the incumbent’s

performance is expected
to be sufficient,
but previously
0
θc −θc k ( c)
θ̄ − θ
was not, is given by
2(θc +k (θ̄ −θc ))−2
= 12 θ +k θ̄ −θ −1 > 0. The total fraction
c ( c)
of voters in δ for whom the incumbent’s performance is expected to be sufficient
k ( θ̄ − θ c )
is therefore 12 1 + > 21 . Incumbents have the support of one-half of
θc +k (θ̄ −θc )−1
the voters who were not exposed to the shock and more than one-half of the voters
who were, and are therefore re-elected with no additional effort. However, they
provided more public goods than they needed to in order to secure the support of
N/2 voters.

Case 2: The unanticipated shock is negative (k θ̄ − θc < 0). When the unan-
ticipated shock reduces the beliefs of a fraction δ of voters regarding incumbent
costs, these voters expect higher performance, on average, than the incumbent
anticipated they would. Some of these voters would have believed that the incum-
bent met the performance threshold in the absence of the shock, ω̄i ≤ H ( gθc ), and
now do not believe this, ω̄i > H ( gθc +k ). Now, the fraction of the voters exposed to
the information
shock who are satisfied by the incumbent’s performance is given
k ( θ̄ − θ c )
by 21 1 + < 21 . Fewer than one-half of the voters subjected to the
θc +k (θ̄ −θc )−1
information shock, and therefore fewer than one-half of all voters, are satisfied by
incumbent performance. However, these incumbents can still be re-elected if they
use transfers to increase voter welfare.
42
Proof of Proposition 2
Recall from Lemma 1, Case 2, that public good provision meets the performance
threshold of fewer than half of the voters in δ. Incumbents cannot increase public
good provision to recapture the support of N2 voters, but they can use transfers.
It follows immediately that the transfers must be sufficient to meet the condition

that f k = H ggθc +k − H ( gθc ): they must be enough to bring voters’ evaluation of
incumbent performance up to the performance threshold for enough voters such
that the incumbent has the support of N2 voters. Note that inter-voter competition
for transfers does not drive transfers to zero because incumbents have no incentive
to initiate it. Voters have already coordinated on a voting rule. Consequently, in-
dividual voters cannot credibly commit their vote to the incumbent if they receive
transfers that are lower than needed to bring the incumbent’s performance up to
the threshold that is consistent with the voting rule.
If incumbents could, they would target these transfers to the most persuadable

voters, those for whom transfers f k = H ggθc +k − H ( gθc ) are just sufficient to shift
their support to the incumbent. However, incumbents know only the distribution
of voter beliefs and not the beliefs of each voter. Hence, they have to make transfers
to voters without knowing whether those voters already support them, even with-
out transfers, or whether those voters will not support them, even with transfers.
We first show, therefore, that incumbents prefer to target voters inδ with transfers
rather than other voters. We then establish the fraction of voters in δ whom they
target.
1 k (θ̄ −θc )
Recalling that k θ̄ − θc is less than zero, − 2 is the fraction of
θc +k (θ̄ −θc )−1
voters in the group δ that received the information shock and would be “per-
suaded” by a transfer f k . Other voters in δ either already support the incumbent
or are sufficiently hostile to the incumbent that they would not be persuaded by
the transfer. The fraction of voters in the group not exposed to the shockand that
k (θ̄ −θc )
would be equally persuadable by the transfer f k is given by - 21 θ c −1 . Since

(θc − 1) > θc + k θ̄ − θc − 1 for k θ̄ − θc < 0, the probability that a transfer
will reach a persuadable voter is greater if it is targeted to voters in the group δ.34
34 The intuition is straightforward. Incumbents would like to target transfers to voters for whom
the distribution of voter beliefs is most dense around the median: these are the most persuadable
voters. An information shock that tells voters that the maximum costs of providing public goods
43
Incumbents cannot identify these voters, however, since they know only the
distribution of preferences. Incumbents’ probability of re-election ρ is therefore de-
termined by the fraction α of the voters in δ to whom they provide the transfer
fk =
k ( c)
θ̄ − θ
H ggθc +k − H ( gθc ). The probability equals zero for α < 12

- if they
θc +k (θ̄ −θc )−1
provide transfers to fewer voters than those whose support they lost because of
the information shock, they cannot be re-elected, so they would prefer to provide
zero and forego re-election. The probability of re-election goes to one as all mem-
bers of δ receive the transfer, or as α goes to one. Hence, incumbentsif they choose
k (θ̄ −θc )
to seek re-election, incumbents will chooseα from 21 , 1 . Set l =
θc +k (θ̄ −θc )−1

1 k (θ̄ −θc )
2 . Since the distribution of voter beliefs about costs is uniform, the
θc +k (θ̄ −θc )−1

α−l 1 k (θ̄ −θc )
incumbent’s probability of re-election is therefore 1−l , for α ∈ 2 ,1 .
θc +k (θ̄ −θc )−1
−l

The incumbent chooses α to maximize expected rents, α1− l M − θ̄gθc − αδ f k + R ,
subject to non-pecuniary rents from seeking office continuing to be sufficiently
large that the incumbent still prefers to seek re-election, M − θ̄gθi − αδ f k + R ≥
M − θ̄gθi . Assuming the participation constraint does not bind, the incumbent
M−θ̄gθc + R+lδ f k
maximizes rents choosing α∗ = δ fk .
are lower than they thought reduces the upper limit, but has no effect on the lower limit, of the dis-
tribution of voter beliefs regarding the costs of producing public goods. Hence, the shock increases
the density of the uniform distribution at every point, including the median, making treated voters
more attractive targets for vote buying than untreated voters. This effect is not unique to a uniform
distribution, but occurs for any distribution for which the information shock increases the density
of voters at the median.
44
Appendix for Online Publication
A.1 Background on the Experiment
Table A.1: Timeline

Date Activity
April 17-29 Candidates Interview
April Randomization
May 5 Flyer printing
May 7-10 Flyer distribution
May 13 Elections
June Household survey
Table A.2: List of Intervention Municipalities

Province Municipality # Pairs # Candidates
ILOCOS NORTE BANGUI 7 2
BANNA (ESPIRITU) 10 3
DINGRAS 15 3
PAOAY 15 2
PASUQUIN 15 3
PINILI 10 2
SAN NICOLAS 11 2
ILOCOS SUR BURGOS 11 2
LIDLIDDA 5 3
MAGSINGAL 13 2
SAN JUAN (LAPOG) 13 2
SANTA LUCIA 17 2
Notes: The list differs slightly from the one included in the Pre-Analysis Plan as volunteers could
not distribute the flyers in Banayoyo (Ilocos Sur), Pagudpud (Ilocos Norte) and Tagudin (Ilocos
Sur). In addition, we had to drop one pair in Pasuquin (Ilocos Norte) as we found out during the
endline survey that the control village in that pair was a military camp.
A.1
A.2
Figure A.1: Flyer for the Municipality of San Nicolas, Ilocos Norte
A.2 Additional Results
Figure A.2: Candidate Proposed Projects and Budget Allocations
A.3
Table A.3: Balance Tests : Preferences Over Sectoral Allocations
(1) (2) (3) (4) (5)
Panel A: Sectoral Preferences
Health 18.915 18.714 -0.453 0.012 0.204
(12.834) (13.094) [0.651] [1.000] [0.641]
Education 17.378 18.048 1.530 0.021 -0.665
(12.527) (12.993) [0.126] [0.828] [0.124]
Help for Needy 8.086 8.162 0.228 0.012 -0.077
(9.610) (9.742) [0.820] [1.000] [0.814]
Water and Sanitation 7.822 8.500 2.363 0.036 -0.675
(7.989) (8.731) [0.018] [0.213] [0.017]
Roads 7.340 6.541 -2.792 0.050 0.791
(9.035) (7.587) [0.005] [0.028] [0.004]
Community Facilities 4.877 4.689 -0.891 0.021 0.186
(6.413) (5.872) [0.373] [0.851] [0.366]
Business Loans 5.802 5.637 -0.496 0.016 0.161
(9.917) (9.538) [0.620] [0.982] [0.624]
Agricultural Assistance 21.517 21.947 0.690 0.025 -0.422
(18.078) (18.249) [0.490] [0.644] [0.482]
Peace and Security 5.473 5.197 -1.238 0.018 0.272
(6.662) (6.363) [0.216] [0.934] [0.213]
Community Events 2.790 2.566 -1.354 0.023 0.223
(4.771) (4.859) [0.176] [0.759] [0.172]
Panel B: Alignment
Alignment 58.396 58.267 -0.285 0.011 0.128
(19.418) (19.674) [0.776] [0.984] [0.764]
Alignment (challenger) 57.168 57.063 -0.182 0.015 0.105
(18.562) (18.557) [0.856] [0.977] [0.849]
Alignment (incumbent) 59.883 59.727 -0.220 0.031 0.156
(20.314) (20.861) [0.826] [0.365] [0.811]
Notes: n=3,408 (Panel A) and 7,896 (Panel B). The standard deviations are in (parentheses)
(Columns 1-2). In Columns 3-4, the test statistics are reported along with the p-values [bracket].
Each cell in Column 5 is either the coefficient on the dummy variable indicating whether the cam-
paign was implemented in the village from a different OLS regression with pair fixed-effects or the
associated p-value in [bracket].
A.4
Table A.4: Balance Tests : Respondent Characteristics
(1) (2) (3) (4) (5)
Length of residence 40.693 40.390 -0.464 0.018 0.303
(19.069) (19.028) [0.643] [0.941] [0.639]
Family size 5.022 5.096 1.012 0.032 -0.074
(2.217) (2.045) [0.312] [0.321] [0.297]
Female 0.490 0.492 0.137 0.002 -0.002
(0.500) (0.500) [0.891] [1.000] [0.889]
Age 49.361 48.853 -1.103 0.034 0.505
(13.537) (13.374) [0.270] [0.262] [0.269]
Education (years) 9.462 9.353 -0.930 0.039 0.110
(3.468) (3.422) [0.353] [0.134] [0.345]
Remittances abroad 0.225 0.244 1.334 0.019 -0.019
(0.418) (0.430) [0.182] [0.899] [0.175]
CCT Beneficiary 0.154 0.159 0.377 0.005 -0.005
(0.361) (0.366) [0.706] [1.000] [0.700]
Group Member 0.671 0.644 -1.697 0.028 0.028
(0.470) (0.479) [0.090] [0.518] [0.077]
Barangay assembly 0.925 0.936 1.214 0.011 -0.011
(0.263) (0.245) [0.225] [1.000] [0.213]
Collective Action 0.736 0.767 2.102 0.031 -0.031
(0.441) (0.423) [0.036] [0.365] [0.026]
Religion: never 0.085 0.087 0.183 0.002 -0.002
(0.279) (0.282) [0.855] [1.000] [0.853]
Religion: weekly 0.373 0.363 -0.568 0.009 0.009
(0.484) (0.481) [0.570] [1.000] [0.561]
Notes: n=3,408. The standard deviations are in (parentheses) (Columns 1-2). In Columns 3-4,
the test statistics are reported along with the p-values [bracket]. Each cell in Column 5 is either
the coefficient on the dummy variable indicating whether the campaign was implemented in the
village from a different OLS regression with pair fixed-effects or the associated p-value in [bracket].
A.5
Table A.5: Balance Tests : Respondent Characteristics
(1) (2) (3) (4) (5)
Electricity 0.968 0.973 0.909 0.005 -0.005
(0.177) (0.162) [0.363] [1.000] [0.358]
Radio 0.735 0.742 0.507 0.008 -0.008
(0.442) (0.437) [0.612] [1.000] [0.605]
Television 0.877 0.876 -0.104 0.001 0.001
(0.328) (0.329) [0.917] [1.000] [0.915]
Phone 0.890 0.905 1.412 0.015 -0.015
(0.313) (0.293) [0.158] [0.992] [0.154]
wash mach 0.326 0.346 1.269 0.021 -0.021
(0.469) (0.476) [0.204] [0.855] [0.197]
Fridge 0.529 0.542 0.755 0.013 -0.013
(0.499) (0.498) [0.450] [0.999] [0.443]
Gas stove 0.603 0.614 0.632 0.011 -0.011
(0.489) (0.487) [0.528] [1.000] [0.515]
Bicycle 0.406 0.390 -0.945 0.016 0.016
(0.491) (0.488) [0.345] [0.981] [0.336]
Boat 0.023 0.026 0.549 0.003 -0.003
(0.151) (0.160) [0.583] [1.000] [0.558]
Motorcycle 0.535 0.549 0.825 0.014 -0.014
(0.499) (0.498) [0.409] [0.995] [0.402]
Notes: n=3,408. The standard deviations are in (parentheses) (Columns 1-2). In Columns 3-4,
the test statistics are reported along with the p-values [bracket]. Each cell in Column 5 is either
the coefficient on the dummy variable indicating whether the campaign was implemented in the
village from a different OLS regression with pair fixed-effects or the associated p-value in [bracket].
A.6
Table A.6: Correlates of Sectoral Preferences
(1) (2) (3) (4) (5) (6) (7) (8) (9) (10)
Health Education Emergencies Water Road ComFaci EconProg Agriculture Peace Festiva
Education (years) -0.106 0.347*** -0.195*** 0.064 0.128*** 0.052 -0.110** -0.262*** 0.125*** -0.045
(0.075) (0.069) (0.051) (0.039) (0.048) (0.033) (0.045) (0.087) (0.036) (0.025
Children below 6 0.022 0.843*** -0.034 -0.019 -0.116 -0.057 -0.469*** 0.124 -0.090 -0.204
(0.349) (0.296) (0.205) (0.170) (0.226) (0.132) (0.159) (0.331) (0.132) (0.102
Children 7-14 -0.605** 1.152*** -0.317** 0.005 -0.324** -0.106 0.128 0.366 -0.305*** 0.006
(0.242) (0.230) (0.161) (0.146) (0.154) (0.113) (0.235) (0.345) (0.109) (0.111
Farmer -1.582** -1.375*** -1.433*** -0.245 0.106 -1.285*** -1.337*** 8.810*** -0.993*** -0.666*
A.7
(0.666) (0.500) (0.445) (0.352) (0.381) (0.301) (0.429) (0.733) (0.281) (0.219
Business Owner -0.484 0.758 -0.413 0.033 -0.017 -0.187 0.943 -1.194 0.204 0.355
(0.808) (0.712) (0.582) (0.435) (0.478) (0.387) (0.643) (0.822) (0.377) (0.294
Female 0.931** 0.491 2.197*** -0.052 -1.488*** -0.575** 0.789** -2.199*** -0.070 -0.02
(0.469) (0.444) (0.368) (0.331) (0.326) (0.225) (0.362) (0.647) (0.246) (0.178
Observations 3,404 3,404 3,404 3,404 3,404 3,404 3,404 3,404 3,404 3,404
R-squared 0.007 0.028 0.022 0.001 0.012 0.013 0.012 0.077 0.015 0.008
Notes: Results from individual-level regressions. The dependent variables are the share of the LDF that the respondent would like to allocate
to Health (Column 1), Education (Column 2), Emergencies (Column 3), Water (Column 4), Roads (Column 5), Community Facilities (Column
6), Economic Programs (Column 7), Agriculture (Column 8), Peace and Order(Column 9) and Festivals (Column 10). The standard errors
(in parentheses) account for potential correlation within village. * denotes significance at the 10%, ** at the 5% and, *** at the 1% level.
Table A.7: Effects of Treatment on : self-reported support for candidate
(1) (2) (3) (4)
Panel A: Candidate Ratings
Treat 0.001 0.001 0.001
(0.009) (0.009) (0.006)
Alignment 0.009 0.009 -0.008 -0.018
(0.061) (0.061) (0.065) (0.454)
Interaction 0.013 0.013 0.020 -0.007
(0.091) (0.091) (0.097) (0.578)

Pair Fixed-Effects No No Yes No
Individual Fixed-Effects No No No Yes
Observations 6,825 6,825 6,825 6,825

R-squared 0.411 0.411 0.414 0.437
Panel B: Self-reported vote choice
Treat 0 0 -0.001
(0.007) (0.007) (0.005)
Alignment 0.052* 0.052* 0.050* 0.048
(0.029) (0.029) (0.029) (0.174)
Interaction -0.006 -0.006 -0.031 -0.096
(0.048) (0.048) (0.048) (0.249)

Individual Fixed-Effects No No No Yes
Observations 6,793 6,793 6,793 6,793

R-squared 0.470 0.470 0.477 0.528
Notes: Results from candidate*individual-level regressions. All regressions include a full set of
candidate dummies. In Panel A, the dependent variable is the rating given to the candidate relative
to the average rating given to the other candidates. In Panel B, the dependent variable is a dummy
equal if the respondent indicated voting for the candidate in our secret ballot exercise. The standard
errors (in parentheses) account for potential correlation within village. * denotes significance at the
10%, ** at the 5% and, *** at the 1% level.
A.8
Table A.8: Effects of Treatment on : Knowledge of Politicians and Candidates
(1) (2) (3) (4)
Panel A: Knowledge of Local Politicians
Treat 0.025 0.027 0.029 0.029
(0.040) (0.030) (0.020) (0.020)

Observations 3,187 3,187 3,187 3,187

R-squared 0.001 0.160 0.277 0.278
Panel B: Knowledge of Candidates
Treat -0.022 -0.022 -0.022 -0.024
(0.040) (0.027) (0.020) (0.019)

Observations 3,408 3,408 3,408 3,408

R-squared 0.001 0.218 0.309 0.314
Notes: Results from individual-level regressions. In Panel A, the dependent variable is an index
capturing the respondent’s knowledge of politicians in their village, municipality and province. In
Panel B, the dependent variable is an index capturing the respondent’s knowledge of mayoral can-
didates’ political experience and education levels. In Column 4, the regression includes a dummy
equal to one if someone in the household is a member of any group, a dummy equal to one if
someone in the household participated in any collective action activity in the village in the past six
months, the share of the local budget the respondent would like to spend on water and the share of
the local budget the respondent would like to spend on roads. The standard errors (in parentheses)
account for potential correlation within village. * denotes significance at the 10%, ** at the 5% and,
*** at the 1% level.
A.9
Table A.9: Effects of Treatment on : vote buying (individual-level)
(1) (2) (3) (4)
Panel A: Are you Aware of Instances of Vote Buying in your Village?
Treat 0.029 0.033 0.035** 0.034**
(0.029) (0.023) (0.016) (0.016)

Observations 3,212 3,212 3,212 3,212

R-squared 0.001 0.105 0.193 0.193
Panel B: Did Someone Offered you Money for your Vote?
Treat 0.027 0.031** 0.032*** 0.033***
(0.021) (0.015) (0.011) (0.011)

Observations 3,181 3,181 3,181 3,181

R-squared 0.001 0.117 0.176 0.176
Notes: Results from individual-level regressions. In Panel A, the dependent variable is a dummy
equal to one if the respondent indicated being aware of instances of vote buying in her village. In
Panel B, the dependent variable is a dummy equal to one if the respondent indicated that some-
one attempted to buy their votes [with ’refused to answer’ coded as ’missing’]. In Column 4, the
regression includes a dummy equal to one if someone in the household is a member of any group,
a dummy equal to one if someone in the household participated in any collective action activity in
the village in the past six months, the share of the local budget the respondent would like to spend
on water and the share of the local budget the respondent would like to spend on roads. The stan-
dard errors (in parentheses) account for potential correlation within village. * denotes significance
at the 10%, ** at the 5% and, *** at the 1% level.
A.10
Table A.10: Effects of Treatment on : vote buying
(1) (2) (3) (4)
Panel A: Did Someone Offered you Money for your Vote ?
[Alternative Coding]
Treat 0.038* 0.038** 0.038** 0.047***
(0.021) (0.016) (0.016) (0.017)


R-squared 0.012 0.462 0.718 0.728
Notes: Results from village-level regressions. The dependent variable is the share of respondent
who indicated that someone attempted to buy their votes [with ’refused to answer’ coded as ’yes’].
In Column 4, the regression includes the share of respondents with an household member who
belongs to a group, the share of respondent who participated in any collective action activity in the
village in the past six months, the village-average share of the local budget that respondents would
like to spend on water and the village-average share of the local budget the respondent would like
to spend on roads. The standard errors are (in parentheses). * denotes significance at the 10%, ** at
the 5% and, *** at the 1% level.
A.11
Table A.11: Comparing Incumbent and Challenger Promises
Incumbent Challenger T-test K-Smirnov test OLS
(1) (2) (3) (4) (5)
Health 15.417 15.333 -0.033 0.117 0.256
(5.418) (7.188) [0.974] [1.000] [0.919]
Education 15.417 12.333 -1.393 0.283 3.333
(4.502) (6.510) [0.176] [0.544] [0.192]
Emergencies 6.250 4.333 -1.566 0.200 1.667
(3.108) (3.200) [0.130] [0.915] [0.195]
Water and Sanitation 8.750 5.333 -2.128 0.333 2.949
(4.330) (3.994) [0.043] [0.336] [0.122]
Road 10.417 11.667 0.587 0.300 -1.026
(5.418) (5.563) [0.563] [0.468] [0.610]
Community Facilities 6.667 5.333 -0.934 0.183 0.897
(3.257) (3.994) [0.359] [0.958] [0.465]
Business Loans 8.750 10.667 0.849 0.317 -2.308
(3.769) (7.037) [0.404] [0.398] [0.362]
Agricultural Assistance 14.167 24.667 1.809 0.267 -9.744
(7.017) (19.036) [0.082] [0.624] [0.077]
Peace and Security 10.000 7.000 -1.531 0.283 3.077
(5.222) (4.928) [0.138] [0.544] [0.100]
Community Events 4.167 3.333 -0.597 0.133 0.897
(3.589) (3.619) [0.556] [1.000] [0.349]
Notes: n=27). The standard deviations are in (parentheses) (Columns 1-2). In Columns 3-4, the test
statistics are reported along with the p-values [bracket]. Each cell in Column 5 is either the coef-
ficient on the dummy variable indicating whether the campaign was implemented in the village
from a different OLS regression with municipal fixed-effects or the associated p-value in [bracket].
A.12
Table A.12: Heterogeneity
(1) (2) (3) (4) (5)
Panel A: 2010 Incumbent Vote Share (Municipal, non-corrected)
Treat 0.051*** -0.211** -0.025 0.110** 0.034**
(0.015) (0.088) (0.118) (0.044) (0.015)
Interaction 0.003*** -0.014*** 0.001 0.003 0.002***
(0.001) (0.004) (0.006) (0.002) (0.001)
Observations 3,408 3,408 3,408 3,346 284

R-squared 0.330 0.528 0.253 0.090 0.745
Panel B: 2010 Incumbent Vote Share (Barangay, non-corrected)
Treat 0.051*** -0.210** -0.046 0.105** 0.035**
(0.014) (0.090) (0.115) (0.043) (0.015)
Interaction 0.003*** -0.012*** 0.001 0.005** 0.002**
(0.001) (0.004) (0.006) (0.002) (0.001)
Observations 3,408 3,408 3,408 3,346 284

R-squared 0.330 0.528 0.257 0.093 0.745
Notes: Results from individual-level (Columns 1-4) )and village-level (Column 5) regressions. All
regression include pair dummies. In Column 1, the dependent variable is an index capturing the
respondent’s knowledge of candidate promises. In Column 2, the dependent variable is the number
Column 3, the dependent variable is the number of errors made by the respondent about candidate
promises that were favoring the challenger. In Column 4, the dependent variable is rating given
my family" when the respondent was asked about ’voting influences’. The variable is adjusted to
account for the average rating given to the other categories. In Column 5, the dependent variable is
the share of respondent who indicated that someone attempted to buy their votes [with ’refused to
answer’ coded as ’missing’] . The standard errors (in parentheses) account for potential correlation
within village (Columns 1-4). * denotes significance at the 10%, ** at the 5% and, *** at the 1% level.
A.13
Table A.13: Further Results
(1) (2) (3) (4)
Panel A: Turnout
Treat 0.034 -0.109 -0.125 0.330
(0.729) (0.506) (0.510) (0.558)
Relative Preference 0.075 0.045 0.151 0.237*
(0.047) (0.075) (0.121) (0.126)
Interaction -0.053 -0.053 -0.072 -0.076
(0.071) (0.051) (0.053) (0.055)


R-squared 0.00500 0.513 0.722 0.732
Panel B: Candidate Vote Share
Treat 0.169 0.169 0.229
(0.312) (0.312) (0.304)
Alignment 0.003 0.003 -0.015 -0.087
(0.041) (0.041) (0.066) (0.288)
Interaction -0.056 -0.056 -0.080 -0.276
(0.066) (0.066) (0.098) (0.264)

Barangay Fixed-Effects No No No Yes

R-squared 0.860 0.860 0.864 0.873
Notes: Results from precinct-level regressions (Panel A) and candidate*precinct-level regressions
(Panel B). In Panel A, the dependent variable is turnout in the 2013 mayoral elections. In Panel B,
the dependent variable is the candidate vote share in the 2013 elections. All regressions include a
full set of candidate dummies. he standard errors (in parentheses) account for potential correlation
within village. * denotes significance at the 10%, ** at the 5% and, *** at the 1% level.
A.14

Keefer

Uploaded by

Copyright:

Available Formats

Keefer

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Keefer

Uploaded by

Copyright:

Available Formats

Incumbent Advantage, Voter Information

and Vote Buying ∗

Keywords: Political Economy, Vote Buying, Information, Elections

2.1 Vote Buying

2.2 Information Provision

2.3 Incomplete Information and Dominant Incumbents

3 The Experiment and Data

3.1 The Experiment

Political Knowledge In order to examine whether the intervention was effective,

Salience Another test of effectiveness of the intervention is whether treated house-

Preferences over candidates and spending allocations We analyze the treat-

4 Results: Effects on Knowledge and Vote Buying

Table 2: Descriptive Statistics

Yijk = αTj + vk + uijk (1)

where Yijk is the knowledge index for individual i in village j in pair k, Tj is a

Table 3: Effects of Treatment on: Knowledge of Promises

Municipal Fixed-Effects No Yes No No

Observations 3,408 3,408 3,408 3,408

Yijk = αTj + vk + uijk (2)

4.2 Information Effects on Vote Buying

Yjk = αTj + vk + u jk (3)

Municipal Fixed-Effects No Yes No No

Observations 3,346 3,346 3,346 3,346

Municipal Fixed-Effects No Yes No No

Observations 3,346 3,346 3,346 3,346

intensified in treated villages.23

5 Accounting for the Results

Municipal Fixed-Effects No Yes No No

Observations 284 284 284 284

Municipal Fixed-Effects No Yes No No

Observations 284 284 284 284

5.1 Basic Set-Up

2. Voters coordinate on a voting rule ω̂ = ω ( gi ), where gi is given by Hg ( gi ) =

3. Incumbents choose the level of public good provision g.28

4. A randomly-selected subset of all voters δ ≤ 1 are subject to an unanticipated

5. Incumbents choose the level of spending on transfers to voters.

7. The election takes place.

Proof: See technical appendix.

Lemma 1 simply confirms the effects on voter support of unanticipated infor-

Proof: See technical appendix.

Proposition 2 describes the incumbent response to an information shock. In

Proof: See technical appendix.

6 Results: Mechanisms and Alternative Explanations

Municipal Fixed-Effects No Yes No No

Observations 3,408 3,408 3,408 3,408

Municipal Fixed-Effects No Yes No No

Observations 3,408 3,408 3,408 3,408

Table 7: Effects of Treatment on : Self-reported vote for the incumbent

Municipal Fixed-Effects No Yes No No

Observations 2,882 2,882 2,882 2,881

Observations 3,408 3,408 3,408 3,346 284

Observations 3,408 3,408 3,408 3,346 284

our explanation of the effects we identify. The information interventions of Chong

6.3 Robustness Checks and Alternative Explanations

Observations 3,408 3,408 3,408 3,346 284

Observations 3,408 3,408 3,408 3,346 284

ing. Additional research is needed to assess an important corollary of this argu-

A NSOLABEHERE , S., E. C. S NOWBERG , AND J. M. S NYDER , J R . (2006): “Television

B ANERJEE , A., S. K UMAR , R. PANDE , AND F. S U (2011): “Do Informed Voters