This paper introduces and applies synthetic control methods to comparative case studies, focusing on the effects of Proposition 99, a large-scale tobacco control program implemented in California in 1988. The authors discuss the advantages of synthetic control methods and demonstrate their application to estimate the impact of Proposition 99 on tobacco consumption in California. They find that following the passage of Proposition 99, tobacco consumption in California fell significantly compared to a synthetic control region. By 2000, annual per-capita cigarette sales in California were about 26 packs lower than they would have been without Proposition 99. The paper also addresses the limitations of traditional regression methods in comparative case studies and proposes new inferential techniques that are valid regardless of the number of comparison units, time periods, or data type (individual or aggregate). The synthetic control method is shown to be effective in producing informative inference and is applied to cross-country data to estimate the impact of German reunification on the West German economy.This paper introduces and applies synthetic control methods to comparative case studies, focusing on the effects of Proposition 99, a large-scale tobacco control program implemented in California in 1988. The authors discuss the advantages of synthetic control methods and demonstrate their application to estimate the impact of Proposition 99 on tobacco consumption in California. They find that following the passage of Proposition 99, tobacco consumption in California fell significantly compared to a synthetic control region. By 2000, annual per-capita cigarette sales in California were about 26 packs lower than they would have been without Proposition 99. The paper also addresses the limitations of traditional regression methods in comparative case studies and proposes new inferential techniques that are valid regardless of the number of comparison units, time periods, or data type (individual or aggregate). The synthetic control method is shown to be effective in producing informative inference and is applied to cross-country data to estimate the impact of German reunification on the West German economy.