Do Brazilian mayors use empirical research to make policy choices?

Experimental Evidence from 2,150 Brazilian Municipalities

Decentralization can allow the public sector to be more inclusive and responsive, but only when local leaders make well-informed policy choices. This begs the question: what role can policy-effectiveness research play in spurring the spread of effective policies at the local government level and the abandonment of ineffective ones? For instance, do mayors consider evidence-based research when making local decisions to promote local economic development?

One possibility is that lack of (access to) research information is not a binding constraint on local policy choices, for example because political leaders are self-interested and electoral competitive pressures too weak to motivate the effort required to change policy, or because leaders have limited real power over the policies in use. Alternatively, frictions may constrain political leaders’ access to existing research.

A paper recently published in the American Economic Review investigates how informing political leaders about research findings affects policy beliefs and practice. Using experiments with the elected heads of Brazil’s local governments, mayors, the study finds that local political leaders value access to impact evaluations, and update their beliefs when informed of the research findings. Mayors (and other local policymakers in the study’s sample) appear to be fairly sophisticated consumers of accessible research, for example paying more for studies, such as those with a large sample size, that subsequently affect their beliefs more.

In addition, the analysis shows that providing mayors with research findings documenting positive impact of an inexpensive and easy-to-implement policy increases the probability that their municipality implements the policy by 10 percentage points. Making research information directly and easily available to policymakers therefore appears to influence policy. This suggests that information frictions may play an important role in explaining failures to adopt policies which have been proven to be effective.

Perhaps it is surprising that such information frictions persist. After all, even if political leaders themselves do not read academic journals, information frictions should generate incentives for actors interested in enhancing social welfare to access academic research and connect policy research with practice to eliminate these frictions.

Yet, the reach of think tanks and other organizations that institutionalize and scale up transmission of research findings to local political leaders still appears limited, especially in developing countries. While evidence-based research can be a helpful input in better local decision-making, there is a need to prevent local policymakers from facing the problem of information overload, with numerous motivated actors attempting to persuade them by providing them with selective pieces of evidence and information.

Access the study: Hjort, Jonas, Diana Moreira, Gautam Rao, and Juan Francisco Santini. 2021. “How Research Affects Policy: Experimental Evidence from 2,150 Brazilian Municipalities.” American Economic Review, 111 (5): 1442-80.