The economist and Nobel Prize winner Oliver Hart has rightly said in a Spanish newspaper that a good referendum must be preceded by a contract. In the short space of a newspaper interview, he only says that such contract should clarify the time before another similar referendum should take place. The objective would be to avoid the instability created by the fact that the voters may realize after the vote that the terms of what was being voted were uncertain. Oliver Hart, like almost all economists, was against Brexit and probably he was thinking of this referendum, after which the United Kingdom has entered one of the most uncertain periods of its history. But of course surely Hart would agree that many more issues should be covered in a referendum contract, like the exact terms of the policies and institutions that would be implemented in case of each possible referendum outcome. This is of course very difficult, although there are precedents in history, especially when referendums were preceded by wide-ranging agreements among all relevant stakeholders. This happened for example in the referendum after the Good Friday Agreement in Ireland. Or it happened in the Constitution referendum in Spain in 1978. I could not vote in those years, but I remember that all voters received the long text of the agreed Constitution at home (in my family, the copy we received could be read in Spanish from one side and in Catalan from the other side). However, when the referendum precedes a negotiation, the terms of the vote are totally unclear, which facilitates the work of demagogues and opportunists, as it happened with the Brexit referendum. It would have been interesting to dig deeper into the thoughts of Oliver Hart in this case, because this economist has promoted the idea of incomplete contracts, which preside those relationships where it is impossible (or too costly) to foresee all possible contingencies. Clearly, in many cases, it is not at all clear what will happen. In the economics of the firm, which is where Hart and others have applied the idea of incomplete contracts, the allocation of property rights (understood basically as control rights, because property gives the right to decide in all those cases that are not covered by a contract) is crucial to determine outcomes when contracts are incomplete. That is why citizens should be vigilant when opportunistic governments (democratic or autocratic, national or regional), political parties or international powers promote referendums. They will try (especially if the referendum has no legal framework, like in Crimea) to manipulate the agenda, the date, the question and everything they can because they know that the relationship between them and the voters is presided by a very incomplete contract. The control rights are crucial to try to influence the outcome of the vote, and the management of the events before and after it. Perhaps I should add this insight if I (hopefully with the help of someone else) ever come to write an English expansion of something I wrote in Spanish about social choice after the Brexit referendum. I infer from his work on firms that Oliver Hart knows that, in the presence of incomplete contracts, the moral attittudes of political leaders are as important as the vote of citizens.