What is Bayesian Efficiency?
Bayesian Efficiency is a formula to allocate resources across a population to optimize for a particular parameter, even if not all of the population’s details are unknown. This efficiency approach is a variation of Pareto efficiency that uses Bayes’ theory of probability to fill in the knowledge gaps about a population. The ultimate goal is still the same though: to allocate resources in such a way that one preference criterion is optimized and any further reallocation would make at least one individual in the population worse off.
What’s the Difference between Bayesian Efficiency and Pareto Efficiency?
While having the same goal, Bayesian efficiency addresses the unsolved problems from Pareto efficiency in three ways:
- Uses probability to account for incomplete information about individual members or a whole population’s parameters.
- Decides when to conduct the efficiency evaluation. The efficiency check can be made before the agent sees the population details (ex-ante stage), at the interim stage after the agent sees population details, or later when the agent has complete information about all variables (ex-post stage).
- Finally, Bayesian efficiency adds an incentive qualifier so that the allocation rule is actually followed.