SOCIAL DIVISIONS: India’s caste system as an exemplar of why members of different groups refuse to interact
01 Nov 2018
Social divisions – such as the divisions between ‘tribes’ or ‘clans’ in developing countries, or the divisions between racial or religious groups in developed countries – are a major problem in many societies.
The defining feature of these divisions is that members of different groups refuse to interact with each other in certain ways. For example, there may not be intermarriage across groups or employers may not hire members of different groups.
What causes these divisions? Research by James Choy, published in the November 2018 issue of the Economic Journal, suggests a new answer.
According to the new study, people may refuse to interact with members of a different social group in order to preserve their reputations with members of their own group.
To see how this works, consider the caste system in rural India, the example that inspires the new research. Suppose that in this system, a member of an upper caste is trying to decide whether to enter into a business partnership with another member of the same caste.
The partnership requires trust – for example, perhaps the first person is trying to decide whether to lend money to the second person so that the second person can make improvements on his farm. The lender may worry that the borrower will take the money and not repay the loan.
One factor that the lender may consider when thinking about the borrower’s trustworthiness is whether the borrower has alternative relationship partners from whom he can borrow. A borrower who has fewer alternative options is more trustworthy, because the cost to him of destroying a business relationship by failing to repay a loan is likely to be larger.
Now suppose that the borrower refuses to interact with some class of people, such as members of lower castes. By artificially limiting his alternative options, the borrower makes himself more trustworthy to members of his own caste. This effect helps to explain why people do not interact with members of different groups, even if there are no fundamental differences between members of different groups.
An interesting feature of this theory is that it explains the hierarchical nature of the Indian caste system and other systems of social division. Consider again the example of the upper and lower castes. If the upper-caste borrower has interacted with members of lower castes in the past, the lender believes that the borrower will be able to receive loans from members of lower castes in the future.
But this belief is reasonable only if members of lower castes are generally willing to interact with members of upper castes. Thus, the reputation effect implies a hierarchical system of social division, in which members of upper castes refuse to interact with members of lower castes, but in which members of lower castes are willing to interact with members of upper castes.
Again, this result appears even though there are no fundamental differences between members of upper castes and members of lower castes.
‘Social Division with Endogenous Hierarchy’ by James P. Choy is published in the November 2018 issue of the Economic Journal.