We report experimental results from three Colombian villages concerning the impact of a voting mechanism on interpersonal trust and trustworthiness. The vote is purely consultative in that participants are asked to declare in a secret ballot the most “appropriate” plan of action for individuals involved in a “Trust Game”. The plan of action that is most voted is then publicly announced. The mechanism is unbinding, as only the aggregate result of the voting is disclosed and it has no bearing on individual decisions. In spite of the strategic irrelevance of the announcement, we observe an increase in both trust and trustworthiness after the announcement is carried out, in comparison to the baseline case where no voting takes place.