Research

Job Market Paper

Matching with Property Rights: an Application to Korean Teacher Transfer Program [Current version: PDF]

I model a teacher transfer program in South Korea as a matching-with-contracts problem. The current (non-centralized) system allows a teacher wishing to make an inter-state transfer to submit at most one application and will approve only a one-on-one exchange. Consequently, not only is this current system unstable and wasteful, but it also does not adequately reflect their preferences.
In contrast, the centralized mechanism I propose fully reflect teachers’ preferences and is less wasteful. One key requirement of a teacher transfer program is that it must protect the rights of currently existing teachers who wish to transfer from their current positions for a short period of time only, possibly to satisfy familial needs. However, as I demonstrate, no stable mechanism exists to always protect the rights of such teachers to the positions they currently occupy. 
To solve this problem, I introduce a novel concept called a claim contract. A claim contract, in a matching-with-contract environment, protects existing teachers by removing contracts that do not match the specified duration set by these teachers from the set of potential matches. In this paper, I propose a new mechanism based on the claim contract that is strategy-proof and approximates stable matching.
 

Working paper

Parking Space Assignment Problem: a Matching Mechanism Design Approach

I extended the claim contract concept to a dynamic setting by applying it to a parking space assignment problem. I study a down town parking problem in a many-to-one matching frame and used the claim contract idea to protect residential permit parking holders. I model parking market as a strategic game and show that the set of Nash equilibrium outcomes of the cruising game is equivalent to the set of stable allocations. However, it is not reasonable to expect the drivers to reach a Nash equilibrium in the decentralized system due to the lack of information and coordination failure. I suggest a centralized mechanism with which a parking authority can assign available spaces to drivers in a stable way. Since the natural of the parking problem required dynamic considerations, I am investigating this problem with an algorithmic approach in order to find an approximation solution. 

  • Parking Space Assignment Problem: presented at the ITEA Annual Conference on Transportation Economics, Barcelona, Spain (June 2017) 
  • Poster session at the International Workshop on the Economics of Parking, Barcelona, Spain (November 2016)

Housing Market with Contract: Korean Teacher Exchange

I study a housing market problem with contract, where resources can be offered in two contractual terms, with a leading example of teacher exchange in Korea. I impose a restriction to this market, namely equal-term constraint, which requires exchanges be made in the same contractual terms. With this restriction, I show that there is no mechanism that is Pareto efficient, individually rational, equal-term, and strategy proof. I further show that strategy proofness is not compatible with Pareto improving from initial endowment. Therefore, I suggest a mechanism called Top Trading Cycle (TTC) with Counter Offer, which is individually rational and efficient under equal-term restriction.  TTC with Counter O er works as a normal TTC, except that it allows a counter offer which demands a contract with different terms, thus allowing it to find a potential cycle with equal terms.

  • Teacher Exchange in Korea: presented at the Conference on Economic Design, Budapest, Hungary (June 2019) 
  • Housing Market with Contract: presented at the 2018 SAET Conference, Taipei, Taiwan (June 2018)

Research in Progress

A Modification to Korean College Admissions
 
Time Inconsistency in Willingness to Wait (with Sera Linardi)
  • Presented at the Experimental Brown Bag Seminar, University of Pittsburgh (Oct 2019)
Market Design Approach to Congested Transportation System (with Sera Linardi)
  • Presented at the NSF PittSmartLiving Internal Workshop, University of Pittsburgh (Jan 2019)