Friday, April 30, 2010

Learn from Child

Great talk, maybe that's lesson #1...

Wednesday, April 28, 2010

Who is OB Drive?

The other day, I looked up a published paper in Industrial and Corporate Change, which cited a recent publication by one professor in our strategy group, prof. Anne Marie Knott. The citation goes like this, "...Knott and Drive (2008)...", my immediate response is that who the heck is Drive? I am familiar with this paper and I am pretty sure that this is a single-author publication!!

Guess what? It turns out that this confusion is simply caused by the omnipotent search engine - Google Scholar! As you can see from here, the search result does look like that the paper is written by Knott and Drive. However, as I dig this a bit further (I simply typed in the keyword "OB Drive") , and from the search result, I could almost immediately explain what went wrong. Many academic papers whose authors are affiliated with Washington University suffer the same problem, because they all share the same critical attribute - their correspondence addresses are all listed as "One Brookings Drive"!!

If we could simply change this to "1 Brookings Drive", this problem might be resolved under the current algorithm of Google Scholar! See, here is at least "1" benefit of using numbers rather than words!!

Saturday, April 24, 2010

A Lecture from the New Clark Medalist

John Bates Clark Medal will be awarded annually from this year on, and this year's winner was announced by the American Economic Association yesterday -- the title goes to MIT development economist, Esther Duflo (also see here).
She will visit Wash U next week and give a lecture here.
Update: Here is a talk given by Esther Duflo on social experiements earlier this year.

Thursday, April 1, 2010

Insider Econometrics

Ichniowski and Shaw have a great review paper recently on how organizational economists address the question of the adoption of management practice and its potential effects on the productivity of workers, worker groups and firms. With the attractive title of "Insider Econometrics: Empirical Studies of How Management Matters", I am sure if you are interested in recent development in empirical personnel economics, this is a must read.

Insider Econometrics is termed to suggest "the use of rich mirco-level data on workers or work groups inside firms that share a common production function, and at the same time, also refers to the use of insights from insiders - from managers or employees - that inform almost every facet of the research." The latter point, according to the authors, is the defining characteristics of this empirical strategy.

Extrinsic versus Intrinsic Motivation

Can we use rational economic theory to explain voting, volunteering, giving to charity, helping strangers or even risking life? Behavioral economics often attribute these to altruistic preferences. However, some phenomena cannot be explained by sole presence of individuals with other-regarding preferences, such as the well-documented "crowding-out" effect. For example, when people start to get monetary benefits from donating blood, their actual donation will go down.

In a series of papers (see here and here), Roland Benabou and Jean Tirole attempt to explain this crowding out effect using rational economic models. Put it simply, they have built two different models to capture two such possible mechanisms that would eventually lead to crowd-out. One is the so-called informed principal problem. In this model, the principal wants to motivate the agent to make efforts. However, although the principal knows the critical factors of the task the agent is going to take (e.g., its chance of success, or the ability of the agent in doing the task), the agent doesn't know about it. It turns out that the principal would have more incentive to motivate the agent only if the agent is not that good. This would eventually become a bad signal for the agent to make such efforts.

The other mechanism, often documented in the psychology literature as the "overjustification effect", is called "multi-dimensional signaling" by Benabou and Tirole. They argue, that people don't care only about monetary payoffs, they also care about their self-fulfillment (e.g., intrinsic warm glow when helping others) and their reputation (e.g., what others would say about them when they do good deeds). If all people in the society could be characterized by parameters of intrinsic, extrinsic and reputational motives, and these parameters also follow some kind of distribution, then, as they have shown in their model, when we increase the monetary incentives, part of the population who are initial contributors would be crowded out by new contributors who are more greedy but less altruistic.