Do bureaucrats acculturate? Evidence from a long-running natural experiment in India

This project explores the extent cultural norms around corruption and inaction are adopted over time by bureaucrats. By utilising the random assignment of elite bureaucrats to states with varying cultures of corruption and inaction, the team studies how the bureaucrat’s environment influences their preferences, values, and behaviour.

Inaction and corruption of bureaucrats in developing countries have potentially significant implications for both resource allocation and economic activity. These features of bureaucratic behaviour have two broad classes of influences: incentives and cultural norms or preferences. While the role of incentives has been studied extensively, we know little about the role of norms and preferences in driving the behaviour of bureaucrats. Thus, the researchers aim to answer five primary research questions:

  1. Do bureaucrats assigned to corrupt states become: (i) more dishonest, (ii) less willing to punish dishonesty in others, (iii) less likely to emphasise integrity as an important bureaucratic value?
  2. What is the time path of acculturation? How do effects depend on the number of years of exposure?
  3. Are values explained more by home environment or work environment?
  4. Do acculturation effects persist even after bureaucrats face a common environment?
  5. To what extent are bureaucrats aware of the influence of their environments

Strong evidence of acculturation entail re-designing allocation rules to counteract the acculturation of officers assigned to corrupt states. By exploring the time-path of acculturation effects, we can inform whether or not and when officers should be transferred from more corrupt states to less corrupt states (and vice versa). Moreover, evidence of acculturation will confirm that ethics can be shifted in adulthood, and yield suggestions for other interventions to initiate shifts towards a non-corrupt equilibrium fostered by social norms.


  • Research in progress.

    Project last updated on: 9 Nov 2020.