Men are from Mars…
Posted by fazeer on 17 April, 2008
The standard economic model usually starts with something like this: “We consider an infinite horizon economy, with a continuum of individuals of measure 1, etc.” These individuals have are neither men nor women, because economists, in their quest for simplification, assume that men and women make choices in much the same way. While this assumption is perfectly fine for most stories that economists want to tell, some experiments show that gender matters. Here are two examples, taken from a survey paper on Field Experiments in Development Economics by Esther Duflo:
1. GENDER AND TIME INCONSISTENCY
…the existence of time-inconsistent preferences is well established in the lab … A key question when analyzing the consequences of these preferences is whether people are sophisticated or naïve in how they deal with their temporal inconsistency … Sophisticated [people] will therefore have a demand for commitment devices, whereas naïve [ones] will not.
Ashraf, Karlan, and Yin (2006) designed a commitment savings product for a small rural bank in the Philippines. Individuals could restrict access to the funds they deposited in the accounts until either a given maturity or a given amount of money was achieved. Relative to standard accounts, the accounts carried no advantage other than this feature. The product was offered to a randomly selected half of 1,700 former clients of the bank.
Prior to offering the products, the experimenters asked the standard hypothetical preference-reversal questions. They found that among women, those who had more tendency to exhibit preference-reversal were also the most likely to take up the product…the fact that the time-reversal questions predict take-up for women, but not for men, is not something that was predicted by the theory.
2. GENDER AND ASYMMETRIC INFORMATION
[In South-Africa], Karlan and Zinman (2005) set out to test the relative weight of ex-post repayment burden (including moral hazard) and ex-ante adverse selection in lending. In their setup, potential borrowers with the same observable risk were randomly offered a high or a low interest rate in an initial letter. Individuals then decided whether to borrow at the solicitation’s “offer” rate. Of those that responded to the high rate, half were randomly given a new lower “contract” interest rate when they actually applied for the loan, while the remaining half continued to receive the rate at which they were offered the loan. Individuals did not know beforehand that the contract rate might differ from the offer rate.
The researchers then compared repayment performance of the loans in all three groups. This design allows the researchers to separately identify adverse selection effects and ex-post repayment burden effects (which could be due to moral hazard or sheer financial distress ex post) … The study found that men and women behave differently: while women exhibited adverse selection, men exhibited moral hazard.
Obviously, marketing people have known this all along. In their same study, Karlan and Zinman find that “for male customers, having a photo of a woman on top of the offer letter increased take-up as much as a 1 percent reduction in the monthly interest rate.“
Microfinance has built its success on giving loans to women principally, as they are more likely to repay than men (97% of the borrowers at the Grameem Bank are women).
Gender differences affect public choice too. In “Women as Policy Makers: Evidence from a Randomized Policy Experiment in India“, Duflo shows how “women elected as leaders invest more in drinking water and roads in West Bengal and drinking water in Rajasthan. They invest less in education in West Bengal and roads in Rajasthan.“
Are gender differences caused by institutions and society or do they have deeper causes? In two recent NBER working papers, Muriel Niederle, Alexandra H. Yestrumskas, Carmit Segal, Lise Vesterlund conduct lab experiments to show that while men are eager to compete and women often shy away from competitive environments, these are attributable to institutions and can be remedied by affirmative action.
