School suspension rates have been rising since the early 1970s, especially for children of color. One body of research has demonstrated that suspension from school is harmful to students, as it increases the risk of retention and school dropout. Another has demonstrated that school dropouts impose huge social costs on their states and localities, due to lost wages and taxes, increased crime, higher welfare costs, and poorer health. Although it is estimated that reducing school suspension rates in Texas would save the state up to $1 billion in social costs, only one study to date has linked these two bodies of research. The current study addresses some of the limitations of that study by (1) estimating a stronger causal model of the effects suspension has on dropping out of school, (2) calculating a more comprehensive set of the social costs associated with dropping out, and (3) estimating the cost of school suspensions in Florida and California, and for the U.S. as a whole. The results show that suspensions in 10th grade alone produced more than 67,000 dropouts in the U.S. and generated social costs to the nation of more than $35 billion. These results are undoubtedly conservative, since the California and U.S. estimates were limited to 10th-grade students, while the Florida estimates were limited to 9th-grade students. Thus, they did not capture the effects of suspensions in earlier grades. The report contains recommendations for policymakers, an appendix detailing the Educational Longitudinal Study of 2002 analysis, and endnotes.