<bold>Problem, research strategy, and findings:</bold> Underpriced and overcrowded curb parking creates problems for everyone except a few lucky drivers who find a cheap space; all the other drivers who cruise to find an open space waste time and fuel, congest traffic, and pollute the air. Overpriced and underoccupied parking also creates problems; when curb spaces remain empty, merchants lose potential customers, workers lose jobs, and cities lose tax revenue. To address these problems, San Francisco has established SF<italic>park</italic>, a program that adjusts prices to achieve availability of one or two open spaces per block. To measure how prices affected on-street occupancy, we calculated the price elasticity of demand revealed by over 5,000 price and occupancy changes during the program's first year. Price elasticity has an average value of -0.4, but varies greatly by time of day, location, and several other factors. The average meter price fell 1% during the first year, so SF<italic>park</italic> adjusted prices without increasing them overall. This study is the first to use measured occupancy to estimate the elasticity of demand for on-street parking. It also offers the first evaluation of pricing that varies by time of day and location to manage curb parking. <bold>Takeaway for practice:</bold> San Francisco can improve its program by making drivers more aware of the variable prices, reducing the disabled placard abuse, and introducing seasonal price adjustments. Other cities can incorporate performance parking as a form of congestion pricing. <bold>Research support:</bold> University of California Transportation Center.
