The expansion of retail recreational marijuana laws (RMLs) has increased access to cannabis among older adults, yet evidence on potential cognitive and mental health implications remains limited.  This study examines changes in cannabis use and self-reported cognitive and mental health outcomes among adults aged 45 and older following RML implementation, with particular attention to subpopulations aged 55+ and 65+.  Using Behavioral Risk Factor Surveillance System (BRFSS) data from 2015–2023, we implement a staggered treatment difference-in-differences design that exploits variation in the timing of state RML adoption.

We find that RML enactment is associated with a 1.3 percentage point increase in past-30-day marijuana use among adults aged 45 and older, with similar magnitudes across older age subgroups.  Despite increased reported use, we find no evidence of changes in self-reported confusion or memory loss.  However, RML adoption is associated with a 1.7 percentage point reduction in the probability of reporting any poor mental health day in the past 30 days.  These findings suggest that increased cannabis use following legalization among older adults is not accompanied by short-run increases in self-reported cognitive difficulties and may coincide with improvements in mental health indicators.

To address limitations of self-reported cognitive measures in BRFSS and to examine longer-term outcomes, we plan to extend this analysis using restricted-access Health and Retirement Study (HRS) data with detailed geographic identifiers.  This ongoing work will allow for a more comprehensive examination of the prevalence, progression, and severity of Alzheimer’s disease and related dementias (ADRD) in relation to cannabis exposure following legalization.  Together, these results provide early evidence on behavioral and mental health responses to cannabis policy among older adults and lay the groundwork for future causal analyses of cannabis use and ADRD risk.

This study examines how university enrollment affects local cannabis retail sales, using administrative sales data from the Washington State Liquor and Cannabis Board covering monthly dispensary sales from 2014 through 2017.  We focus on cannabis retailers located in proximity to four-year colleges and universities and exploit within-year variation in student presence driven by the academic calendar.  To do so, we implement a stacked event-study design in which monthly sales from each year are aligned around the start of the academic year in August, allowing us to compare sales dynamics before and after student arrival while controlling for seasonality.

Our empirical framework links monthly dispensary-level sales to university enrollment using institution-level data from the Integrated Postsecondary Education Data System (IPEDS).  We examine heterogeneity by student characteristics, including full-time status and indicators of remote or non-traditional enrollment, and account for local market structure by controlling for the density and proximity of competing cannabis retailers.  Additional robustness analyses assess the sensitivity of results to cross-border competition by excluding or isolating dispensaries near Idaho, Oregon, and the Canadian border.

We exploit the differences in age restrictions policies for recreational and medical retail as we explore differences in sales responses across cannabis type, leveraging variation in tax treatment between the two.  Previous studies assess policy impact through surveys, subject to reporting bias.  Our approach provides new evidence on consumption changes on how higher-education demand shocks interact with legalized cannabis markets.