Research

What does society gain and lose when states reduce criminal penalties for drug possession? Using FBI arrest and crime records and CDC death records, I compare Oregon's and Washington's 2021 reforms with weighted comparison states that tracked them before 2021. Drug-user arrests fall 68 percent with no shift to other charges, while overdose deaths rise 34 percent and property crime 17 percent. Deaths rise in every comparison group, including groups matched on COVID, fentanyl's spread and state politics. Restoring penalties in Washington brings arrests back but not deaths down. Health and crime costs outweigh the gains from less enforcement: a net social cost of $26.6 billion by mid-2023, mostly from 1,971 excess deaths. The results show that reducing penalties for drug possession can cut arrests sharply and still leave society worse off.

Large language models (LLMs) are increasingly used as synthetic personas representing survey respondents. Their validity as substitutes for particular respondents depends on whether they reproduce individuals' decisions. We examine what information helps synthetic respondents predict each individual's later choices, using five conditions that add progressively richer information: no personal information, demographics, personality traits, cognitive scores, and finally the respondent's earlier survey choices as behavioral history. We use a two-wave panel of 845 US adults who completed measures of 14 behavioral biases (spanning risk, time preferences, overconfidence, and reasoning), so each respondent's earlier answers provide a human test–retest benchmark; in the behavioral-history condition, all items that score the target bias are withheld. At the population level, the average number of biases per respondent in every condition is close to the human average (7.1–8.1 biases, against 7.1 for humans). This aggregate similarity masks differences in variance: persona descriptions recover only 53–67% of human between-person variation, whereas adding behavioral history restores it to approximately the human level. At the individual level, description-based personas achieve only 7–12% of the informedness observed in human test–retest responses, while adding behavioral history raises this to 28%. The condition including behavioral history has the highest estimated informedness in all 17 demographic groups, whereas description-based conditions provide little or no information for some groups. Synthetic responses also exhibit stronger education- and income-related differences than human responses. For LLM synthetic personas, a respondent's past answers add more to individual-level prediction than a description of who they are.