While fiscal adjustment is commonly viewed as the cornerstone of macroeconomic stabilization, the effectiveness of alternative fiscal instruments in raising national saving is still poorly understood. This paper enters the debate by estimating a private consumption function that allows for two types of agents—finite horizons and liquidity constraints—and nests three different consumption hypotheses. Using a large-panel data set that includes both industrial and developing countries, we reject full Ricardian equivalence. We also find substantial differences between industrial and developing countries, regarding both the extent of Ricardian offsetting and the degree to which the government budget constraint is internalized.