Health insurance copayments and deductibles can distort the healthcare consumption of those with no cash. We test for that possibility by studying the distribution of monthly Social Security checks among Medicare recipients. On the day that Medicare recipients receive their Social Security checks, they become 6-12 percent more likely to fill a prescription. That ``payday effect'' is concentrated in low-income areas, and is of a similar magnitude for a set of drugs where imperfect adherence has important health consequences: blood thinners, insulins, anti-psychotics, and anti-seizure medications. We observe no effect of Social Security checks among Medicare recipients who pay no copays, but we see strong effects for very low-income recipients who face subsidized copayments of only $2-$5. Payday effects disappear when recipients are re-categorized to face no copayments. By contrast, the effects diminish, but do not disappear, when recipients are re-categorized to face lower but non-zero copayments, indicating that only full insurance can eliminate this distortion.