authorDiscover the Best Books Written by Bruce de Mesquita
Bruce Bueno de Mesquita (/məˈskiːtə/; born November 24, 1946) is a political scientist, professor at New York University, and senior fellow at Stanford University's Hoover Institution. Bueno de Mesquita graduated from Stuyvesant High School in 1963 (along with Richard Axel and Alexander Rosenberg), earned his BA degree from Queens College, New York, in 1967, and then his MA and Ph.D. from the University of Michigan. He specializes in international relations, foreign policy, and nation-building. He is one of the originators of selectorate theory and was also the director of New York University's Alexander Hamilton Center for Political Economy from 2006 to 2016.
He was a founding partner at Mesquita & Roundell until that company merged with his other company, Selectors, LLC, which used the selectorate model for macro-level policy analysis. Now, the company is called Selectors, LLC, and uses both the forecasting model and the selectorate approach in consulting.
Bueno de Mesquita is discussed on an August 16, 2009, Sunday New York Times Magazine article entitled "Can Game Theory Predict When Iran Will Get the Bomb?" In December 2008, he was also the subject of a two-hour history channel special entitled "The Next Nostradamus" and featured on the 2021 Netflix series How to Become a Tyrant. He is the author of many books, including The Dictator's Handbook, co-authored with Alastair Smith, and the forthcoming book The Invention of Power (January 2022).
Into the early 2000s, Bueno de Mesquita was known for his development of an expected utility model (EUM) capable of predicting the outcome of policy events over a unidimensional policy space. His EUM used Duncan Black's median voter theorem to calculate the median voter position of an N-player bargaining game. He solved for the median voter position as the outcome of several bargaining rounds using other ad-hoc components in the process.
The first implementation of the EUM was used to successfully predict the successor of Indian Prime Minister Y. B. Chavan after his government collapsed (this was additionally the first known time the model was tested). Bueno de Mesquita's model not only correctly predicted that Charan Singh would become prime minister (a prediction that few experts in Indian politics at the time predicted) but also that Y. B. Chavan would be in Singh's cabinet, that Indira Gandhi would briefly support Chavan's government, and that the government would soon collapse (all events that did occur).
From the early success of his model, Bueno de Mesquita began a long and continuing career of consulting using refined implementations of his forecasting model. A declassified assessment by the Central Intelligence Agency rated his model as being 90 percent accurate.
Since 2005 or so, Bueno de Mesquita developed a superior model, now known as the Predictioneer's Game or PG, that forecasts in a multi-dimensional space, use the Schofield mean voter theorem, and solves for Perfect Bayesian Equilibrium in an N-player bargaining game that includes the possibility of coercion, essentially a greatly generalized version of the 2-player game in War and Reason. This model predicts significantly more accurately and does a substantially better job of identifying opportunities that players have to improve the outcome by exploiting uncertainties. This model is documented in A New Model for Predicting Policy Choices: Preliminary Tests and discussed and applied to examples in The Predictioneer's Game.
Bueno de Mesquita's forecasting model has greatly contributed to the study of political events using forecasting methods, especially through his numerous papers that document elements of his models and predictions. Bueno de Mesquita has published dozens of forecasts in academic journals. The entirety of his models has never been released to the general public.
