Carolinians have come to doubt that public colleges are using their tax dollars wisely,... and this has made lawmakers reluctant to spend more on higher (Austin Gilbert of South Carolina's Commission on Higher Education, quoted in Schmidt, 1997, p. A26). Introduction South Carolina is one of many states to question whether higher education institutions operate efficiently. Government officials suggest that using performance measures and limiting state appropriations for higher education will control higher education costs and force institutions to provide an education more efficiently. For example, the 1999 Almanac issue of The Chronicle of Higher Education reported that 22 states provide appropriations to individual public colleges partly based on their meeting specific performance measures. Although state appropriations for higher education have generally increased in recent years due to a strong economy, it is a matter of time before the economy slows. In some cases, even a strong economy is not enough to increase state appropriations. For example, Governor Pataki's 1999 budget proposal sought reductions in state appropriations for the State University of New York and City University of New York systems. Appropriation reductions were sought to encourage what Governor Pataki called continued efforts to improve cost-efficiency and productivity (Chronicle of Higher Education's Daily Report, January 28, 1999). Why some universities are more efficient at providing public higher education to students and the relationship between efficiency and an institution's revenue structure are not adequately addressed in the literature. This article attempts to partly fill this gap with a three-step analysis. First, the changed structure of public university revenues between 1991 and 1995 is documented. Reductions (or limited growth) in state appropriations combined with increases in tuition revenue resulted in universities receiving a smaller proportion of revenues from state appropriations and a larger proportion from tuition. [1] The second step analyzes how this changed revenue structure influenced cost efficiency defined as the difference between actual costs and an estimated minimum cost. Attention is focused on whether public sector universities that received a larger proportion of funds from state appropriations were relatively more or less efficient than schools that received a larger share of funds from other sources such as tuition. [2] The third step examines how universities' cost efficiency and revenue structure changed between 1991 and 1995. In particular, this paper determines whether schools with a greater decline in the share of revenues from state appropriations improved efficiency relative to schools with a smaller decline in the share from state appropriations. Literature Review Before getting to these questions, we discuss some of the literature examining higher education costs (e.g., Adams, Hankins, & Schroeder, 1978; Brinkman, 1981; Brinkman & Leslie, 1986; Brovender, 1974; Hoenack, Weiler, Goodman, & Pierro, 1986; Koshal & Koshal, 1995; Tierney, 1980; Verry & Layard, 1975). Early studies concentrate on a single output in higher education, namely undergraduate students measured by full-time equivalent students, number of credit hours, or number of students. This is problematic, for colleges and universities have various different outputs. James (1978) noted that focusing only on undergraduate students and ignoring other higher education outputs, such as graduate education and research, overstates the cost of providing an undergraduate education and creates biases across types of institutions. Recent studies overcome this problem by considering universities to be multi-product firms producing undergraduate education graduate education, and research. Cohn, Rhine, and Santos (1989) use Higher Education General Information Survey data on more than 1,000 institutions to estimate cost functions for public and private schools. …