Traveling Agents: Political Change and Bureaucratic Turnover in India
Executive Summary — Politicians and bureaucrats are two important pillars of governance, but while politicians are motivated by short-term electoral pressures, bureaucrats are driven by long-term career concerns. This difference in the nature of their incentives is, in most cases, deliberate and constitutionally provided for. Iyer and Mani address two key questions in this paper: How do politicians facing short-term electoral pressures control bureaucrats with low-powered incentives? In turn, how do bureaucrats respond to these incentives? The authors develop a simple framework and provide empirical evidence on both the politicians' and the bureaucrats' strategies, using a detailed data set on the entire career histories of officers in the Indian Administrative Service, the top layer of government bureaucracy in India. Key concepts include:
- The framework suggests that instituting limits to a politician's power to transfer bureaucrats across posts will favorably affect junior officers' incentives to invest in expertise.
- In India there is significant political influence on the bureaucracy through frequent transfers of bureaucrats across posts, despite the constitutional insulation provided to them against political pressures.
- Not all officers face the same odds of being transferred. High-skilled officers are much less likely to be transferred by an incoming politician and have more even career paths.
- Belonging to the same caste as the politician's party base was a factor that helped officers to secure more important posts.
We develop a framework to examine how politicians with short-term electoral pressures control bureaucrats with long-term career concerns. Empirical analysis using a unique data set on the career histories of Indian bureaucrats supports the key predictions of our framework. We find that politicians use frequent reassignments (transfers) across posts of varying importance as a means of control. High-skilled bureaucrats face less frequent transfers and a lower variability in the importance of their posts. There are alternative routes to career success: officers of higher initial ability are more likely to invest in developing expertise, but officers who belong to the same caste as the politician are also able to obtain important posts. Bureaucrats are less likely to be transferred if politicians have alternative means of control through subordinate politicians. Districts with higher rates of politically induced bureaucrat transfers are somewhat less successful in poverty reduction over the long run.