A good friend, Rucheeta Kulkarni, sent me this interesting piece by Ben Stein. My sense is that it takes basic (I mean, VERY basic) concepts from neoclassical theory to generate some insights on how to approach relationships.
One thing I'd like to see is parallel piece from the standpoint of behavioral economics. Many people systematically DO NOT follow Stein's advice and it would be worth understanding (the patterns underlying) why. Hmmm....now THAT would make for a good column in Cosmo, Maxim, etc.
3 comments:
Forget Cosmo or Maxim. I think this type of thing could make it into a pretty good academic journal -- Journal of Personality and Social Psychology, or some social policy journal about marriage and families.
Theres definitely some behavioral theories that one might test empirically about this stuff. As an example, one could do research on loss aversion and how people stay in bad marriages and relationships because they fear being alone.
Similarly, the endowment effect and the underweighting of opportunity costs. People probably underestimate how much they could potentially gain from going on the singles market and finding a new partner. So they stay in a bad relationship or marriage because they underestimate potential gains that could be achieved through leaving the relationship.
Thanks for the comment!
I suppose you could add the (irrational?) salience of sunk costs to your list as well?
I'm curious about how you'd test the behavioral theories empirically, while being able to rule out other non-behavioral explanations. For example, "underestimation" of gains in the singles market could simply reflect incomplete information...
Thoughts?
(Also, point noted on your comment on the corruption post...certainly this stuff is context specific. Your point gets at external validity, and some sort of replication is required, both across countries and across transactions to really understand who bears the differential burden of corruption.)
For the bribe thing, I think its interesting that there was a heterogeneous treatment effect by class in your experiment, but not in that other one. Particularly, in your study the authorities could *see* class whereas in the other experiment, they could only infer class from what they read in the applications. I wonder if there is something about seeing poor people that impacts the bribe.
Post a Comment