Since the mid-1990s, the Indian state of Gujarat has had a fairly robust set of environmental regulations on its books. But pollution from industrial plants in Gujarat remains ubiquitous, and it’s related to another common problem: corruption.
Four US-based economists, in association with MIT’s Poverty Action Lab, collaborated with the Gujarat Pollution Control Board on a two-year study that explored those linked issues. Their goal, in particular, was to see if altering the current regulatory structure might affect the accuracy of reporting by the third-party auditors who verify the plants’ compliance with environmental laws.
The basic flaw in Gujarat’s system (and it’s a problem hardly restricted to India) is that monitored firms are responsible for hiring the auditor who will report on their activity to the government. “And the auditors are basically saying that every plant is compliant—which we know is not true, because the pollution problem is still quite severe,” says Nicholas Ryan, Prize Fellow in economics at Harvard University.
Ryan and his colleagues set up randomized controlled trials in which one set of plants received nonstandard regulatory treatment. Those plants, instead of selecting their own auditors, had auditors assigned to them randomly. The researchers also fixed the price that plants could pay for audit services. In addition, to verify the accuracy of the auditors’ results, they enlisted people from a local university to conduct “backcheck” reviews of conditions at plants both in the experimental group and in a control group. One year into the experiment, the researchers began using incentive pay to encourage auditors to report plant conditions accurately.
There were three main results of the study: First, in many instances, auditors were falsely reporting that plants were compliant with the law. Second, the auditors who were subject to the nonstandard regulatory treatment filed more truthful reports, and those reports showed that roughly three-quarters of the plants were non-compliant. And third, more accurate reporting led plants to reduce their emissions in order to comply with regulations.
Seen from one angle, those results are somewhat obvious. Yet because of their practical relevance, they are also compelling, says Kelsey Jack, assistant professor of economics at Tufts University, who now uses this study to teach one of her courses. “It shows that a very simple fix can make a huge difference in how well these regulations are implemented,” she says. This research reinforces a broader point, Jack adds: “Ensuring that there’s no easy way for people to cut corners or make a buck is as important as having a law on the books in the first place.”
Jack says that this approach is “extremely replicable” in other countries, and not just in developing nations such as India. “The auditing problems around the US financial crisis are another great example of people responding to incentives that are in front of them,” she says.
Firms that issue financial products, for instance, pay credit-rating agencies such as Standard & Poor’s and Moody’s to rate the quality of those very products. The solution in that case, Ryan notes, could be to have investors rather than securities issuers pay for credit ratings.
More generally, Ryan suggests, governments have a role to play in checking audits independently and in publicizing information about them. That way, companies would have a reputation-based incentive to be honest. “Short of wholly changing the incentive structure, as we did [in the Gujarat experiment],” he says, governments and other entities “could still do a lot to minimize conflicts of interest by releasing information about the quality of the audits coming out.”
Esther Duflo, Michael Greenstone, Rohini Pande, and Nicholas Ryan, “Truth-Telling by Third-Party Auditors and the Response of Polluting Firms: Experimental Evidence from India,” The Quarterly Journal of Economics, 129, February 2014.