A fall in cannabis arrests is meaningful. It does not, by itself, tell us whether enforcement has become equitable. To answer that second question, we need to look at who is still arrested, which offenses are counted, and how rates change within each group.
Fewer arrests can coexist with a larger disparity
Firth and colleagues examined Washington arrest records from 2012–2015. After legalization of possession, arrests fell substantially among adults aged 21 and older, including Black adults. The absolute difference between Black and White arrest rates narrowed. Yet the ratio between those rates increased. Opening retail stores did not produce another statistically significant overall reduction. Those are findings about a particular state and period, not a current national estimate. Read the Washington study.
The distinction between an absolute gap and a ratio matters. As a hypothetical example, rates of 100 and 50 arrests per 100,000 people have a gap of 50 and a ratio of two. Rates of 30 and 10 have a smaller gap of 20 but a larger ratio of three. Both groups experienced fewer arrests; the relative disparity grew. Neither measure should be reported as if it were the other.
Age and policy design change the picture
A study of 38 states covering 2000–2016 found that both legalization and decriminalization were associated with lower adult possession-arrest rates. Youth arrests declined after decriminalization, but the analysis did not find a significant youth reduction after adult-use legalization. A policy benefiting adults therefore cannot automatically be assumed to benefit younger people in the same way. Read Plunk and colleagues’ analysis.
A separate analysis of 43 states, using data through 2019, also found that racial disparities persisted across the observation period. The timing of changes differed across policies and populations, and disparities increased in states without a policy change. Read the 43-state study.
What to check before accepting a headline
When someone says a policy “ended disparities” or “made them worse,” ask:
- Does the statistic describe arrest counts, population-adjusted rates, a rate difference, or a rate ratio?
- Does it separate adults from young people?
- Does it measure possession alone or other cannabis offenses too?
- Which jurisdictions and years are included?
- Does the study describe an association, or does its design support a stronger causal conclusion?
These studies evaluate historical enforcement patterns. They do not describe every community’s experience or determine anyone’s present legal rights. Their practical lesson is to evaluate reform with more than a single overall arrest number—and to keep evaluating who remains exposed to enforcement after the law changes.