This paper examines how minimum wage increases affect firm employment and production decisions across firm size in the manufacturing and retail sectors. Using state- and county-level variation and restricted-use firm- and establishment-level data from the US Census, we estimate the average treatment effect on the treated (ATT) of minimum wage increases on employment with a synthetic difference-in-differences approach. We estimate that minimum wage increases in retail shift employment toward firms with 250-2,499 employees and away from the smallest and largest firms, while manufacturing employment declines across all firm sizes. We use these estimates to discipline a dynamic model of monopsonistic competition with endogenous firm entry and exit. The calibrated model implies wage markdowns as approximately 91-92% of the marginal product of labor. We then use the model to quantify the short- and long-run macroeconomic effects of minimum wage increases of varying magnitudes. A permanently indexed ``$15" minimum wage results in long-run employment declines ranging from 6% to 21%.