Direct Election, Bureaucratic Appointment, and Responsiveness in Taiwan

• In Event: People in Politics

Abstract

Does local democracy induce responsiveness to citizens? While elected officials can be punished at the ballot box if they fail to address citizens’ needs, appointed bureaucrats may have greater policy knowledge that enables them to serve citizens well, and may be less vulnerable to vote-buying and other forms of corruption. However, determining whether democracy causes greater responsiveness or simply arises in places where officials are more likely to be responsive is difficult, because the types of places that hold elections are fundamentally different from those that do not.

This paper exploits spatial and temporal variation in local political institutions in Taiwan to assess whether direct election of local officials induces greater responsiveness to citizens. Taiwan’s district magistrates--officials responsible for overseeing “street-level bureaucracy” and local public service provision--are directly elected in some districts and appointed in others; in some cities, elected district positions have been turned into appointed ones. Following Nielsen’s (2016) approach to combining nonparametric matching methods with case studies, I select matched pairs of districts that are broadly similar with respect to relevant background characteristics but differ in terms of the method of (s)election of their district chief. I combine these case studies with process tracing of a “natural experiment” in which a 2014 policy change replaced elected district chiefs with appointed ones across one large city. Finally, I use data from a contacting experiment to compare responsiveness across appointed and elected district leaders; while a given leader’s status as elected or appointed cannot be randomly assigned, the data point to substantial descriptive differences across the two groups.

Taken together, these data suggest that local officials in Taiwan are, by international standards, exceptionally responsive to citizens’ requests for help. However, responsiveness by elected and appointed officials takes different forms. Appointed bureaucrats display high levels of policy responsiveness. These district magistrates serve as an extension of the city government and have relatively elite educational and professional backgrounds; they are knowledgeable about public policy and are able to connect citizens to city-run public welfare programs. They are also motivated to undertake innovative district-level initiatives by a desire to impress their superiors, who are elected at the city level. However, their insulation from direct electoral pressures leads to the perception that they are irrelevant and do not care about voters. By contrast, elected magistrates display greater personal responsiveness: they have greater name recognition among citizens and are perceived by voters as more responsive to their needs. However, they are also widely perceived by elites as corrupt and personalistic, and experimental data suggests that they are less responsive to citizens’ demands for help than appointed officials are.

My findings suggest two mechanisms by which bureaucratic appointment of local officials may improve local governance. First, while some citizens perceive appointed magistrates as less accountable, accountability typically relies on participation in factional and family networks. Those who are excluded from these networks may find bureaucratic appointees relatively responsive. Second, the shift to bureaucratic appointment encourages new types of people—typically those with more elite professional and educational backgrounds than elected magistrates—to enter local government, improving policy knowledge and innovation at the local level. These findings point to some of the limitations to democratic consolidation in a context where elections are increasingly institutionalized but the influence of informal politics remains strong.

Author

• Sara A. Newland, Smith College