Research

Work-in-progress

‘Shuffling to co-opt: Subnational governance, patronage, and political careers in Kazakhstan’. PDF [last update: 25 March 2025].

Abstract: Why do autocrats ‘shuffle’ elites around positions? Existing work suggests this practice aims to boost performance, with underperforming officials more frequently rotated. Yet I show that in Kazakhstan there is no association between performance and rotation. Instead, I explain shuffling as a strategy of co-optation. Shuffling prevents some of the potential downsides of co-opting elites through state office by disrupting network formation and freeing up positions for junior cadre. At the same time, it keeps co-optation credible by reassuring most elites of their long-term seniority. To test this argument, I present a detailed biographic dataset of regional governors (akims) in Kazakhstan between 1997 and 2022. Consistent with my argument, elites holding these posts are frequently shuffled to and from other senior positions. By contrast, there is a robust lack of association between regional socioeconomic measures and when a governor is rotated or dismissed. Sometimes, shuffling aims more at enhancing elites’ loyalty than their performance.

‘Personnel appointments and loyalty in bureaucracies: Evidence from Kazakhstan’.

Abstract: When do bureaucrats with managerial powers leverage personal ties to build their careers or engage in corruption? How do central governments respond? Existing work on bureaucratic managers’ use of delegation powers largely does not go beyond established democracies or test the role of social ties. I focus on non-democracies, where loose hiring rules allow managers to engage in favouritism. Addressing this gap, the paper analyses new data from a most-likely case, Kazakhstan, with a corpus of 6,000 district bureaucrat biographies, time-series data on thousands of appointments by district managers (rayon akims), and performance measures. Using fixed-effects and natural language processing approaches, I examine appointments, the role of social ties, and government responses. I find that managers use appointment powers to strategically build ‘teams’ of personally loyal subordinates who—based on interpersonal ties—they trust to follow orders or hide corruption. Managers engaging in excessive appointments are more likely to be dismissed, and I test the role of performance as a moderating factor. The findings provide new evidence on bureaucracies outside established democracies, the behaviour of subnational officials, and state solutions to bureaucratic disloyalty. They highlight the trade-offs faced by governments balancing bureaucratic autonomy with the risk of network-based corruption.

‘Local Elections and Elite Management in Authoritarian Regimes: Evidence from Kazakhstan’, with Kirill Melnikov and Eleonora Minaeva.

Abstract: This study examines the introduction of local executive elections in Kazakhstan, where rural executives (akims) were previously appointed through a centralized system. We analyze the impact of these elections on elite turnover and recruitment, leveraging a complete dataset of all subdistrict elections and a pilot study of akim biographies. Elections resulted in frequent leadership replacement. Nevertheless, we show that they did not result in inclusion of a broader group of local or the selection of more locally-embedded officials. The results suggest that elections may serve more to formalize the renewal of the same bureaucratic elite than to co-opt new groups or widen access to power.