Essays on Elite Persistence, Accountability, and Representation

Abstract

This dissertation consists of three empirical chapters on elite

persistence, accountability, and representation.

The first chapter examines the determinants of dynastic politicians

in democracies. In the chapter I introduce a theoretical framework

where parties face a trade-off between nominating strong but

undisciplined dynastic politicians, or loyal but weak non-dynastic

candidates. Under this framework, I predict that

parties rely on dynasts only in districts where their organizations

are weak. I test this prediction in the context of Victorian Britain,

and show results broadly consistent with the theory. The main findings

in the chapter suggest that party strength is key to explain the

variation in the incidence of dynasties across time and countries.

In the second chapter, I examine how partisan alignment between

politicians (co-partisanship) affects bureaucratic performance and

policy outcomes. The chapter introduces a theory where only

co-partisan legislators can credibly threaten to punish bureaucrats. I

predict that co-partisan legislators are more likely to sponsor, and

bureaucrats to approve, projects associated with higher rents. I also

predict that legislators, anticipating a favorable disposition from

bureaucrats, use more resources during periods of partisan

alignment. I provide evidence supporting these predictions based on a

unique dataset of works implemented under India's Member of Parliament

Local Development Scheme.

Finally, in the third chapter, co-authored with Scott Abramson, we

estimate the impact of personal power on stability and institutional

development in autocracies. Following the literature of dynasties in

democracies, we propose a leader's tenure as proxy for his political

capital. We then exploit the random timing of natural deaths for a set

of European monarchs, and show their successors were deposed less

frequently and less likely to face parliamentary constraints. We also

show that the effect of tenure on successor deposal is at least as

large as the one associated with succession orders -- an institution

that has received recent attention in the literature. Our results are

consistent with a theoretical account we develop wherein leaders

accumulate political power the longer they are in office which then

determines patterns of succession, stability, and institutional

development in autocracies.