We use data from the Veterans Administration to examine the efficacy of primary care providers (PCPs). Leveraging quasi-random assignment of veterans to PCPs, we measure effectiveness using ambulatory care sensitive conditions (ACSC) and hospitalizations/emergency department (ED) visits for mental health or circulatory conditions. PCPs variation along these dimensions predicts future outcomes. For example, a one standard deviation improvement in mental health effectiveness reduces patient risk of death by 3.8% and lowers costs by 4.4% over the next three years. More effective PCPs do more with less: their patients have fewer primary care visits, specialist referrals, lab panels, or imaging tests.