Voter participation in one election is a good predictor of turnout in later elections. It is therefore often said that voter participation is a habit. However, previous studies have not been able to completely isolate the direct effect of previous participation on voting. I use data from a trial in which the voting age was lowered from 18 to 16 in a group of Norwegian municipalities in the 2011 local elections. 16- and 17-year-olds in these municipalities where given the opportunity to take part in an election, while others at the same age could not. If voting is habit-forming, one would expect the 16- and 17-year olds who were given voting rights in 2011 have higher levels of turnout in the 2013 parliamentary election, than other voters at the same age. Panel data on turnout in both elections (from the electoral roll) is used to test this proposition.
