Balancing Risk and Benefit
The increasing use of randomized evaluations in economics has brought an increase in discussion about ethical issues. We argue that while there are ethical issues specific to randomization, most important ethical challenges are not unique to this methodology. The rise in direct researcher involvement with antipoverty programs that has accompanied the rise in randomized evaluations has made ethics issues more salient and raised complex regulatory questions. Though the principles of respect for persons, justice, and beneficence outlined by the 1978 Belmont Report continue to provide a useful ethical framework, we note a number of challenging tradeoffs in applying them including those around data confidentiality, informed consent, and misleading research subjects. We conclude by discussing how ethical guidelines are applied in practice, noting a number of gaps, ambiguities, and areas where we believe practice is diverging from the underlying principles. These issues apply with equal force to all empirical methodologies.