Peer-reviewed article · 2023
Poverty, Social Networks, and Clientelism
Nico Ravanilla and Allen Hicken · World Development 162(106128)
In brief
Why poor voters are disproportionately targeted with vote buying in the Philippines, and how membership in formal groups like workers' associations shields them from clientelist pressure.
Abstract
Why are the poor susceptible to clientelism, and what factors shield them from the influence of vote buying? We explore the role of both formal and informal social networks in shaping the likelihood of being targeted with private inducements. We argue that when the poor lack access to formal social networks, they become increasingly reliant on vote buying channelled through informal networks. To test our theory, we build the informal, family-based network linkages between voters and local politicians spanning a city in the Philippines. We then collect survey data on formal network connections, electoral handouts, and voting behaviour of 900 voters randomly drawn from these family networks. We show first that campaigns disproportionately target poorer voters. We then show that familial ties further influence targeting among poor voters. Finally, we show that access to formal networks such as workers' associations mitigate voter fears of punishment for failing to reciprocate.
Research annotationQuestion, variables, design, findings, and mechanism at a glance▾
- Research question
- Why are poor voters disproportionately targeted with vote buying, and what shields them from it?
- Independent variable (X)
- Voter poverty, measured as logged monthly household income; the voter's own centrality in the barangay family network, principally betweenness centrality; and membership in at least one formal organization such as a cooperative, sectoral association, professional group, or religious organization.
- Dependent variable (Y)
- Whether the voter was offered money, food, or household items before the election and the amount offered; beliefs about whether brokers, family members, or others could learn how the voter voted; and expectations about what would follow defection, namely nothing, shame, loss of future handouts, or reprimand.
- Identification strategy
- Barangay-level family networks for all 64 barangays of an anonymized city of roughly 150,000 to 170,000 in Southern Luzon, reconstructed from shared surnames on the 2016 Certified Voters List. These are paired with two rounds of an original voter survey, roughly three weeks before and three weeks after the 2016 election, on a random sample of 900 voters stratified by barangay, of whom 659 remain in the estimation sample after missingness. All models are OLS at the voter level, regressing targeting, monitoring beliefs, and enforcement expectations on logged household income, betweenness centrality, formal network membership, and the interactions of income with centrality and with formal membership, with barangay fixed effects and standard errors clustered at the barangay level.
- Main findings
- About 60 percent of surveyed voters reported being offered a handout, against a national figure of 22 percent. Poorer voters are more likely to be targeted: a one log point increase in household income lowers the probability of being offered a handout by 5.8 percentage points against a mean of 0.71, and lowers the amount offered by 0.74 US dollars against a mean of 8.12. Poverty and network position interact, so that among the poor the more central voters are the most heavily targeted while for wealthier voters network position has no bearing on targeting. The monitoring and enforcement hypotheses fail: neither income nor centrality nor their interaction predicts whether voters believe their vote can be observed, and none predicts fear of punishment. Formal network membership does not change the probability of being targeted, but poor voters in formal organizations are significantly more likely to believe brokers can do nothing if they defect and significantly less likely to fear reprimand, with both effects attenuating as income rises.
- Mechanism
- The paper rules out its own proposed mechanism rather than confirming one. Central poor voters are targeted without believing they are more observable or more punishable, so monitoring and enforcement cannot be what drives targeting here. The discussion offers only untested conjectures: collective rather than individual monitoring, handouts as brand-building rather than persuasion, and machine lists that recycle past patronage recipients.
- Why it matters
- Identifies a concrete, policy-relevant protective factor against clientelism in formal association membership, and separates the targeting of the poor from the monitoring story usually invoked to explain it, showing that brokers concentrate on centrally located poor voters even where the surveillance rationale finds no support.
- Speaks to
- clientelism and vote buyingsocial networks and political behaviorpoverty and developmentsocial capital
Cite
@article{ravanilla2023poverty,
title = {Poverty, Social Networks, and Clientelism},
author = {Nico Ravanilla and Allen Hicken},
journal = {World Development},
year = {2023},
doi = {10.1016/j.worlddev.2022.106128},
}