Mechanism design made simple

Posted by Marc Hodak on October 16, 2007 under Regulation without regulators | 5 Comments to Read

For my Canadian readers, you can check me out on B1 of your Globe and Mail. The author did a good job of representing me. She may have slightly overstated things to say that my “consulting business (is) based mainly on applying mechanism design theory,” (it’s more broadly about using incentives to minimize agency costs and improve decision making), but frankly I don’t know of anyone who incorporates mechanism design theory into actual, internal corporate practices to the extent I do. The author seemed thrilled, in fact, to find someone who actually applied this theory to real life business problems rather than another professor to provide theoretical solutions to hypothetical problems.

Alex Tabarrok did an excellent job illustrating the basic concept of mechanism design theory. The author mentioned two of my business applications (executive incentives and corporate investment decisions), but those are fairly complex, and reading what Alex wrote inspired me to think of a simpler example of an application I’ve implemented. So, here is one:

I some situations, one manager has to transfer an asset (say, real estate) to a particular other manager. There is no possibility of competing buyers or sellers. Manager A must sell it to Manager B. At what price? A simple negotiation would quickly break down. The seller would ask for too much; the buyer would offer for too little; neither side has an incentive to provide an honest valuation. Agreeing to split the difference wouldn’t help–in fact, it would simply polarize the bid and ask prices.

So I overlaid an additional rule to this negotiation. I said that if the two managers couldn’t agree on the price, we would set aside their respective best and final offers, and go to a third party appraiser with some experience with that asset (real estate, in this example) and provide an estimated value. The appraiser wouldn’t even see the offers. We would then open up the best and final offers of the two managers, and the offer closest to the appraised value got their price. Think about how the dynamics are changed with this rule, and why the managers, in fact, were generally able to quickly come to agreement, almost never needing an appraiser to actually get involved.

I didn’t mention to the reporter the fact that I was also a professor.