Skip to main content

Bayesian markets for unverifiable truths

Periodic Reporting for period 3 - BayesianMarkets (Bayesian markets for unverifiable truths)

Reporting period: 2019-01-01 to 2020-06-30

Economic analysis and policy design must often rely on expectations (about future events) and other subjective assessments (self-assessed health, life satisfaction, happiness). For instance, Stiglitz, Sen & Fitoussi (2009) recommended using subjective measurements of life satisfaction in welfare analysis. Similarly, environmental policies are based on expert opinions about climate change. But how can we make sure experts report their true estimates? How can we know whether people are truly satisfied with their life? GDP is an objective measure that is compatible with the revealed-preference paradigm prevailing in economics. Existing measures of life satisfaction are not and they have been criticized for this.
If the judgment or opinion we try to collect is related to an observable event, solutions exist. For instance, prediction markets (markets on which agents buy and sell bets that give a money amount if a defined event occurs) offer a way to elicit beliefs of agents and to aggregate them to an average belief (the market price). However, prediction markets require that either the event on which the bet is defined or its complement actually occurs. In horse race betting markets, we know, at the end, which horse wins the race and who should be paid. Prediction markets cannot be applied to subjective judgments, or beliefs about unverifiable events.
The project aims to develop methods that will incentivize (i.e. reward) truth-telling even for completely unverifiable truths.
"The work performed in the first half of the project is organized in 4 parts:
A. The first part is theoretical. We developed a new form of markets, on which people bet on what others think. Their bets reveal what they themselves think.
B. We worked on implementing the new markets in practice. We introduce new, simple bets, that we call ""top-flop bets"". We can use these bets to reveal what people like. It can be used for instance in marketing. Imagine you watch a movie before everyone else (or you can test a product before it is officially launched). The movie producer would typically ask you what you thought of the movie, whether you liked it. But would you always tell the truth? You may want to please the movie producer and say it was nice. What we propose is to make you bet on the performance of the movie, e.g. whether it will get better review than another random movie. We showed how to organize this betting to reveal the most important information: whether you really liked it.
C. The methods we are working on also help us identify experts. In some cases, we do not have experts' track records, so we do not know how good they are (and therefore whether they are real experts). So we do not know whom we should follow. Consider a question such as “Is Proposition X true?” On a Bayesian market (developed in part A), people bet on what others believe the answer to that question is. We demonstrated how studying the earnings of the various agents on the market can tell us whether the proposition is actually true or not. We call it a “follow the money” algorithm.
D. Finally, we want to apply our new methods to socially and economically relevant topics. We identified situations in which what people do differ from what they say. We recently discovered such a situation and called it the ""cybersecurity paradox"": people do not do what they say they should do in terms of cybersecurity (e.g. falling for phishing emails).

We introduce a new type of markets, called Bayesian markets, to reveal what people think or like by making them bets on what others think or like.
Our approach is simpler than what has been proposed in the literature so far (more complicated scoring methods).
We expect the project to show how our new markets and, in general some simple bets, can be used to know what people truly think.

One may wonder why this is interesting: maybe people mostly tell the truth. We are therefore also studying when people lie and when they don't do what they say. Our research on this topic also contributes to the literature on deception (new ways to detect when people lie), but also in cybersecurity (where people don't do what they say they should do). We expect new results in these domains by the end of the project.

Finally, we expect to develop new ways to identify who knows the truth. The majority is sometimes mistaken and for some problems, it is not clear which expert to trust. We are developing a new algorithm, in which studying earnings of people on our new Bayesian markets inform us about the quality of the information / knowledge people have. Our results will contribute to the literature on expert identification and wisdom of crowds.