My research lies at the intersection of political and institutional economics, while also reflecting a broader interest in authoritarian politics and history. More specifically, I study how economic and political decision-making interact, with particular attention to the strategic behavior of politicians as they pursue electoral and political goals.
Methodologically, I mainly use formal modelling, though more recently I have also developed a strong interest in empirical research, especially in building original datasets from the ground up. For more information, you can find my CV here. If you work on related questions or share similar interests, I would be very happy to hear from you.
A common expectation is that weak incumbents are the ones most willing to rely on patronage, trading state resources for the support of elites, since that support can help them survive electoral competition. We show that this intuition changes once patronage is durable, developing an infinite-horizon model of electoral competition in which an incumbent can adopt such an exchange with elites once and for all. Because the channel attaches to office rather than to its creator, it can later benefit a rival who wins office after turnover. This creates a tradeoff: patronage raises the chance of staying in office today but builds an asset that may strengthen an opponent tomorrow. We characterize an adoption threshold and show that strong incumbents are more willing to adopt than weak ones, expecting a longer tenure in which to use the channel and a smaller risk of handing it to a rival. Secondary forces work through the same logic: larger elite contributions raise adoption, electoral stability raises it except for a weak and patient incumbent, and patience raises it only for a strong incumbent, with each effect single-peaked outside those cases.
Although repression generally reduces political participation and thereby strengthens an incumbent’s hold on power, incumbents are often observed to refrain from employing maximal levels of repression even when doing so appears feasible. This paper examines how repression shapes opposition interaction in hybrid regimes and, in turn, incumbents’ prospects for remaining in power. We develop a dynamic model in which opposition factions decide whether to remain politically active under repression and, conditional on participation, whether to cooperate or pursue independent strategies. Repression has two effects. First, it raises the cost of political engagement, reducing the number of active opposition factions. Second, repression strengthens incentives for cooperation among those factions that remain active by intensifying a shared threat and reducing the number of actors who must coordinate. The two effects generate a non-monotonic relationship between repression and the incumbent's chance to stay in power. Intermediate levels of repression limit opposition activity while keeping cooperation unstable. The model thus explains why incumbents may rationally refrain from maximal repression.
This paper examines Political Budget Cycles in federal systems, focusing on how a central incumbent allocates discretionary transfers across states in response to electoral incentives. We develop a theoretical model predicting that average discretionary transfers increase during federal election periods. While swing states consistently receive higher transfers due to their electoral competitiveness, the election-period increase is larger for non-swing states. This occurs because non-swing states are targeted primarily during federal elections: allocating transfers to them in state elections is not advantageous for the federal incumbent, as it has little effect on the probability of winning those state elections. To test these predictions, we compile a panel dataset of Indian states from 2006 to 2022. Using fixed effects specifications, we find evidence consistent with the theoretical model: discretionary transfers are significantly higher in federal election periods, swing states receive more discretionary transfers in non-election periods, and the election-period increase in discretionary transfers is more pronounced for non-swing states.