The papal conclave is often portrayed as a closed, ritualized event. But behind the secrecy lies a voting procedure with clear strategic logic. In this interview, CERGE-EI faculty Jan Zápal introduces the study Electing the pope: Elections by repeated ballots which he coauthors with Clara Ponsati. It explains how the pope is elected through repeated rounds requiring a two-thirds majority, why this system can in principle last for months or even years, and how economists model such elections to understand what kinds of winners the rules tend to produce. Along the way, they connect the conclave to other institutions that use similar repeated voting, and argue that long-standing rules can also confer legitimacy on the final choice.
Can you describe the mechanism for selecting a new pope?
The election of the pope proceeds in successive rounds: in each round, the cardinals vote, each casting a vote for a single candidate. If any candidate obtains two thirds of the votes, he becomes pope. If no one reaches this threshold, the election continues to another round and the entire procedure is repeated.
In this respect it differs, for example, from the election of the president in the Czech Republic: in the first round the votes are counted and either someone is elected by an absolute majority, or—if no one obtains an absolute majority—only two candidates advance to the second round, so that in the second round a president must be elected. For the pope it does not work like that: candidates are not formally eliminated between rounds, and the two-thirds rule does not have to be satisfied “by a certain round.” In theory, therefore, the election can last a very long time. Historically there have been cases where a conclave lasted, say, three months, and even several years.
How does the election mechanism translate into the qualities of the winner?
Every electoral rule affects, to some extent, who will ultimately be selected under it. In our research we focus on two properties of the candidates who become popes.
Each cardinal might have candidates that he finds completely unacceptable in the sense that he would rather remain in the conclave for a very long time than voting for such a candidate. For someone to become pope, he must, therefore, be acceptable to a sufficient number of cardinals—specifically, to two thirds of them.
“We write down a ‘conclave game’ that formally describes who the players are, what they can do, and how they evaluate outcomes.”
One important property of a candidate who becomes pope is what we call stability: there is no other candidate whom two thirds (or more) of the cardinals would clearly prefer more than him. In our research we ask what must hold for there to exist a candidate who is both acceptable to a sufficient number of cardinals and stable—because there is no other candidate who could, in some way, defeat him in the election.
Game theory meets papal elections
How do economists study such a problem? How do they approach it?
Depending on the research question, economists choose different methodologies: sometimes they work with data (for example, tracking how prices change or how many people go to work), sometimes they conduct experiments in an artificial environment and draw conclusions from them, and they also often build theoretical models—for instance using game theory.
We study the papal election precisely in this way: we write down a “conclave game” that formally describes who the players are, what they can do, and how they evaluate outcomes.
In the model, the conclave rules are as follows: the cardinals convene; in each round, each writes one name on a ballot. If someone gets two thirds of the votes, he becomes pope. If not, the ballots are burned—this is why we see black smoke rising from the chimney installed on the roof of the Sistine Chapel—and voting proceeds to another round.
In the model we assume the players are the cardinals, and each has preferences: there are candidates he would prefer to see becoming a pope and candidates he would rather not see becoming a pope. At the same time, remaining in the conclave is costly in some way—more costly for some (they want to go home, they have other duties), less for others (they are willing to stay longer). In each round they vote, and if their preferred outcome is elected, they obtain a certain “utility” from it.
What is your main conclusion?
We examined what the voting rule must be like for there to exist a candidate who is both stable and acceptable to a sufficient number of cardinals. The main result of our study is that, under certain additional conditions, this holds precisely when the threshold for election is a two-thirds majority—that is, the rule that has been used in the conclave essentially from the beginning, for roughly a thousand years.
Similar elections: Italy, party conventions
In what other settings or organizations is a similar method of election used?
Repeated voting in successive rounds without formal elimination of candidates is surprisingly common in the selection of various representatives.
Under certain conditions, for example, candidates for U.S. president are chosen in this way at party conventions. And there is also a less well-known variant in the presidential election process itself: normally the president is chosen by the Electoral College on the basis of the popular vote, but if the Electoral College fails to elect a president (because an absolute majority is required and, with multiple candidates, that may be problematic), the election moves to the House of Representatives. The House then votes in successive rounds without eliminating candidates—similar to a conclave, with the important difference that instead of two thirds, an absolute majority is sufficient.
A similar method is also used to elect the president of Italy, or for example the chief conductor of the Berlin Philharmonic.
Based on your study, would you recommend any changes to the rules of the conclave?
Probably not. One thing economists do not usually study directly as a property of electoral rules is the legitimacy of the chosen representative. It is often the case that if a rule is widely perceived as correct and fair, it also confers legitimacy on whoever is elected under it.
An important aspect of legitimacy is also that the rules are long-term stable and unchanging, which largely holds for the papal election, because these rules have changed only minimally for almost a thousand years. Precisely for that reason they are generally seen as stable and “correct.” For this reason, I do not think I would change them.
Did you have the opportunity to work with any data or information besides models, given that the conclave is officially secret?
Our research is theoretical, so in principle it is enough for us to know the rules used to elect the pope. Nevertheless, we also consulted historical sources describing individual conclaves.
A conclave is supposed to be secret, and information should not be taken out either during the election or afterward. In practice, however, this was not the case—and even today it is not entirely so—so there are historical books documenting what happened in particular conclaves. Information often found its way to the public in the past as well.
For example, in Rome it was historically common to bet on the outcome of the papal election; bookmakers with better sources had an advantage because they could offer more favorable odds. And in the Middle Ages it also happened that the Roman poor looted the palace of the man who was believed to have become pope—on the assumption that, once he became pope, he no longer needed worldly possessions.
Sometimes it turned out that the report of his election was false: the person’s house had been looted, but he ultimately did not become pope.
