Centre for Development Economics
and
Department of Economics, Delhi School of Economics

ANNOUNCE A SEMINAR

Mechanism Design for Land Acquisition

by

Soumendu Sarkar

ISI Delhi

Thursday, 5th March 2015 at 3:00 PM

Venue : Seminar Room (First Floor)
Department of Economics, Delhi School of Economics

All are cordially invited
Abstract

A model of land acquisition consists of a buyer and n sellers. The buyer wants k<n plots of land and sellers own a plot each. We have two variants of the model, the LA problem where the buyer does not require the plots to be contiguous and the LAC problem is one where he has such a restriction.  LA model is the tuple: n, k and a prior mu representing independently distributed valuations. LAC problem is the tuple: G, k and mu where G is a graph such that Sellers are nodes on the graph and they share an edge if their plots are adjacent. In LAC, the buyer wants a path of length k. Our first set of results show that both LA and LAC are problems where the Myerson-Satterthwaite impossibility does not hold. We offer necessary and sufficient conditions on mu for possibility. In our second set of results, we find necessary and sufficient conditions for both models on the prior such that the VCG mechanism almost always results in a surplus when number of sellers become large. The third set of results characterize the optimal mechanism for both problems. We prove a general result that optimal mechanism converges to efficiency, if and only if the VCG mechanism results in an expected surplus in the limit. The implications of the second set of results for the the optimal mechanism are then straightforward. For LAC, we offer a notion of "critical sellers", who are present in every path of length k. We show that the conditions for possibility and convergence become stricter in the presence of critical sellers.

back to seminars...