When Elites Pick Winners: Candidate Selection Boosted Local Development in Postindependence Tanzania

Why Candidate Selection Matters

Jeremy Bowles asks how the process by which parties pick candidates shapes what politicians actually deliver in office. The question matters because candidate selection can alter who wins—even when voters have strong local preferences—and thereby influence the distribution and effectiveness of public goods, especially in regimes where parties dominate political life.

The Tanzanian Case

The study examines Tanzania’s early single-party legislative elections after independence, a period when the ruling party heavily influenced who ran for office. The paper focuses on rural development—a policy priority at the time—and asks whether candidates preferred by party elites produced different local outcomes than those preferred by voters.

Research Strategy and Data

Bowles exploits a distinctive feature of these elections: a deterministic assignment of ballot symbols that was unrelated to candidate attributes but had large effects on vote outcomes. This quasi-experimental variation creates plausibly exogenous differences in electoral success. Using newly assembled candidate-level data, the analysis links which candidates won (elite-preferred versus voter-preferred) to subsequent local provision of salient public goods.

Key Findings

• Electing candidates favored by party elites led to substantial increases in the supply of salient local public goods tied to rural development.
• Party elites tended to prioritize candidates’ national prominence, while voters prioritized local ties and connections.
• Rather than reflecting simple elite capture or misaligned incentives, the pattern is consistent with elites in the emerging regime having better information about which candidate characteristics would translate into effective policy implementation.

What This Implies for Politics and Policy

The results show that candidate selection matters not only for representation but also for on-the-ground policy delivery—even in nondemocratic or single-party settings. Where parties can identify traits linked to performance, elite-directed selection can improve responsiveness and alter the distributive allocation of goods. For scholars of comparative politics and development, the paper highlights a mechanism through which institutional choices about nominations shape governance outcomes.