The study exploits a natural experiment, and so the random selection of the villages into treatment groups was done by the Indian government prior to any researcher involvement. Prior to an election, village council positions in a district were randomly assigned across three lists to ensure some positions are reserved for disadvantaged minorities: Reserved for scheduled castes (SC), Reserved for scheduled tribes (ST), and Unreserved. Within a list, village councils are ordered by serial number. In 1998, every third council starting with number one on each list was required to be reserved for a woman. In 2003 every third council starting with number two on each list.