We find consistent evidence of negative autocorrelation in decision-making that is unrelated to the merits of the cases considered in three separate high-stakes field settings: refugee asylum court decisions, loan application reviews, and major league baseball umpire pitch calls. The evidence is...

We find consistent evidence of negative autocorrelation in decision-making that is unrelated to the merits of the cases considered in three separate high-stakes field settings: refugee asylum court decisions, loan application reviews, and major league baseball umpire pitch calls. The evidence is...

We document judicial leniency on defendant birthdays across 5 million decisions. Our results are consistent with reference-dependent social preferences. First, French sentences are 1% fewer and around 5% shorter. Second, U.S. federal judges also round down sentences except when rounding up makes...

What modes of moral reasoning do judges employ? We construct a linear SVM classifier for moral reasoning mode trained on applied ethics articles written by consequentialists and deontologists. The model can classify a paragraph of text in held out data with over 90 percent accuracy. We then...

We collect data on the record of every action in hundreds of derivative cases and merger class actions involving public companies filed in the Delaware Court of Chancery from 2004 to 2011. We use these data to analyze how markets respond to litigation in the most important court for corporate...

We document motivated reasoning among U.S. Circuit Court judges. We employ a supervised learning approach to measure partisan influences on prose (writing style), precedent (citations to previous cases), and policy (dissenting votes). We find persistent but low partisanship of language overall,...