Yusuke Narita and Kohei Yata
Algorithms produce a growing portion of decisions and recommendations both in policy and business. Such algorithmic decisions are natural experiments (conditionally quasi-randomly assigned instruments) since the algorithms make decisions based only on observable input variables. The researchers use this observation to develop a treatment-effect estimator for a class of stochastic and deterministic decision-making algorithms. The estimator is shown to be consistent and asymptotically normal for well-defined causal effects. A key special case of our estimator is a multidimensional regression discontinuity design. The estimator is applied to evaluate the effect of the Coronavirus Aid, Relief, and Economic Security (CARES) Act, where hundreds of billions of dollars worth of relief funding is allocated to hospitals via an algorithmic rule. The research estimates suggest that the relief funding has little effect on COVID- 19-related hospital activity levels. Naive OLS and IV estimates exhibit substantial selection bias.
Subscribe for Updates