INCENTIVE PAY: Helping to equalise access to information in rural India
07 Jan 2019
New research in rural India suggests that the detrimental effects of entrenched social barriers can be overcome by paying piece rates to local people to spread information.
The study by Erlend Berg and colleagues, which is published in the January 2019 edition of The Economic Journal, finds that the agents whom they commissioned to spread information about a public welfare scheme were generally more effective at transmitting knowledge to people who were similar to themselves in terms of caste, education and poverty status.
But how the agents were paid was key: those whose pay included an incentive payment that depended on how well the recipient households scored in a knowledge test were more effective at transmitting information than agents whose pay was fixed. What’s more, social differences between the agents and recipients no longer seemed to matter. Indeed, agents appeared to spend less time on recipients similar to themselves and more time on those different from themselves.
These findings suggest that financial incentives can help to overcome our natural inclination for dealing with those who are similar to us, which can be a kind of passive discrimination against groups that are more socially distant.
There are many jobs that involve communicating important information. Think, for example, of teachers, health providers or welfare officers.
Even when the person responsible for the task is not someone who shirks her duties in the sense of putting in less effort, they may still not allocate their efforst efficiently across service recipients. For example, we may suspect that teachers sometimes prioritise some pupils over others. This could be due to a conscious or unconscious bias on the part of the teacher – they may simply like some pupils more. But it could also be that some pupils are easier to communicate with.
This research looks at the role of social barriers in hindering the flow of information. The context is rural India, where social markers such as caste are particularly entrenched. The researchers hired and paid local women as ‘agents’ to spread information about a public welfare scheme – the National Health Insurance Scheme – to the local eligible population. Months later, they returned and tested the knowledge of villagers about the scheme.
The researchers find that social differences do matter for the flow of information. Overall, the agents were more effective at transmitting knowledge to people who were similar to themselves in terms of caste, education and poverty status. The detrimental effect of social barriers could be due to a bias on the part of the agents – perhaps they prefer to spend time with people who are similar to themselves, or care more about them. It is also possible that they found it harder to communicate with those who are different from themselves.
But how the agents were paid also mattered. Overall, agents whose pay included a ‘piece rate’ – a payment that depended on how well the recipient households scored in the knowledge test – were more effective at transmitting information than agents whose pay was fixed.
Moreover, agents receiving a piece rate appeared to overcome these social differences. In other words, when the researchers tested the level of knowledge among recipients in villages where the agents received a piece rate, differences in social markers between the agent and the recipients no longer seemed to matter.
By asking the recipients how much time the agent had spent with them, the researchers are also able look at how agents divided their time. It turns out that agents with incentives – those receiving piece rates – did not spend more time on the job overall. But they appeared to be spending less time on recipients who are similar, and more time on those who are different from themselves.
This research adds to our understanding of the connection between incentives and performance. It is true that there is some evidence suggesting that the use of incentives can crowd out intrinsic motivation or attract undesirable types of people to a job.
But we can think of social distance as a form of ‘negative’ intrinsic motivation – preferring to deal with those who are similar to us is a form of passive discrimination against groups that are socially distant from us. This research shows that financial incentives can, at least in some circumstances, crowd out this form of ‘negative’ intrinsic motivation.
‘Motivating Knowledge Agents: Can Incentive Pay Overcome Social Distance?’ by Erlend Berg, Maitreesh Ghatak, R Manjula, D Rajasekhar and Sanchari Roy is published in the January 2019 edition of The Economic Journal.
University of Bristol | firstname.lastname@example.org
London School of Economics | email@example.com
Institute for Social and Economic Change, Bangalore | firstname.lastname@example.org
Institute for Social and Economic Change, Bangalore. | email@example.com
King’s College, London | firstname.lastname@example.org