We exploit the time-series properties of charitable giving to provide additional insights into the relationship between charitable contributions and government spending. Cointegration tests reveal a significant long-run relationship between several categories of charitable giving and government spending. Granger causality tests provide evidence on the short-run giving and spending relationship. Evidence suggests that charitable contributions to education respond quite differently to state and local government education expenditures versus federal government expenditures. We argue that the government spending and charitable giving relationship depends on the source of government revenue, how this revenue is used, and the rational ignorance of private donors.