When you are unhappy about public policy where you are you can either influence it by voting, or you can vote with your feet and move to an area that follows your ideals. But what if you could do both? It is not obvious that this would be welfare improving as the fact that one can move away makes that poorer choices could be offered. And moving implies costs that can be substantial.
Alessandro Innocenti and Chiara Rapallini take the experimental approach. They compare two games, one where dictators impose local payouts and people can move, and one where people vote on payouts and then can decide to move. It appears voting with the opportunity of moving is better than only moving, which makes sense if there are no moving costs: it gives more degrees of freedom and better outside options. And this is what Tiebout's theory was all about. However, I would have loved to see results of experiments with moving costs. This makes people captive, and strategies become more complex as arbitrage by moving is more difficult.