Abstract
We know that ex-post Pareto efficiency is possible with strategy-proofness and weak envy-freeness while ordinally efficiency is incompatible with them. We strengthen this impossibility for random assignments and show that it prevails if ordinal efficiency is weakened to robust ex-post Pareto efficiency, an intermediate notion of efficiency weaker than ordinal efficiency and stronger than ex-post Pareto efficiency. An assignment is robust ex-post Pareto efficient whenever for all of its lottery decomposition, each deterministic assignment in its support is Pareto efficient.
Full Text 
Sign-in/Register to access full text options
Sign-in/Register to access full text options 
Published version (
Free)
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
