As state tax revenues plummet, some lawmakers and higher-education leaders are once again looking at loosening the bonds between state governments and public colleges to save money and give colleges the freedom to bolster their bottom lines in new ways. Over the past two decades, college officials have often lamented the growing need to secure money outside of appropriations. However, the continuing economic crisis has led to a new urgency on the part of some public colleges to shed more of their ties to states, despite the mixed results of previous such efforts. Operating more like private institutions not only would be a buffer from the recession and the volatility of state budgets, some college officials argue, but also may well be vital to the survival of many public colleges. Those that seek to thrive in the future must earn money from a variety of sources and continually cut costs in ways that don't harm the quality of instruction, says Philip J. Hanlon, vice provost for academic and budgetary affairs at the University of Michigan at Ann Arbor, which gets just 7%, or $320-million, of its revenue from the state. While no public college is likely to free itself entirely from fiscal ties to its state, many of the nation's largest public institutions, like Michigan, have evolved to operate nearly like private colleges. One of the surest signs of de facto privatization in public higher education is the increasing share of educational costs being covered by students through tuition and fees, says Jane V. Wellman, executive director of the Delta Project. The rapid growth of tuition has raised concerns that some colleges are more concerned about their profit margin than their public missions. Patrick M. Callan, president of the National Center for Public Policy and Higher Education, says privatization is basically a euphemism: "It's just a code word for "Let us raise tuition as much as we want."" While privatization has occurred in an ad hoc fashion in most of the country, a few states have moved purposely down that path, with mixed results.