Should broadband subsidies make service affordable, or build it where it does not exist? The United States committed roughly $65 billion to both margins at once, and in concentrated markets the two interact: the incidence of a demand subsidy depends on firms’ pricing and product responses, while a deployment grant can reshape competition through entry. I evaluate the $14 billion Affordable Connectivity Program (ACP) and the $42 billion Broadband Equity, Access, and Deployment (BEAD) program in one welfare framework, pairing continuous-treatment event studies with a demand-and-entry model spanning more than 70,000 census tracts. I bound fixed costs with moment inequalities, making nationwide entry estimation feasible without solving the full equilibrium game. ACP exposure cut posted prices by 7–10 percent at peak and shifted demand toward faster plans; BEAD’s effects emerge only after the construction lag implied by its rollout, consistent with entry as the operative channel. A 75 percent BEAD subsidy returns about $7 per dollar over five years, while ACP roughly breaks even. The gains barely overlap: BEAD’s gains concentrate in monopoly tracts, exceeding duopoly gains tenfold, while ACP’s gains concentrate among low-income households. The programs are therefore complements, not rivals: build where deployment is unprofitable, then subsidize affordability where price binds.