As we discussed Tuesday, most Tennessean's aren't willing to raise taxes to fund education. But they are willing to fund education through the purchase of lottery tickets (which some characterize as a tax on the mathematically challenged). According to the Tennessee Lottery Corporation's latest annual report (2005), ticket sales topped $800 million, netting $227 million for education after prize payouts and operating costs.

Launched in 2004, Tennessee's lottery funded scholarship program is by all accounts a success. At U.T. Knoxville, Tennessee's flagship university, there were more than 12,000 applications for 4200 freshmen openings in the Fall of 2006. The incoming class has an average ACT score of 25.8 and a GPA of 3.6, up from an average ACT score of 24.3 and an average GPA of 3.3 in 2001, suggesting there were qualified students who couldn't afford college before the lottery scholarships.

According to the Tennessee Higher Education Commission, there are more than 57,000 students in Tennessee's universities, colleges, and technical schools receiving some form of lottery scholarship.

But the lottery funds pouring into state coffers are creating new problems, the most recent of which is what to do with all that money. The state legislature is now considering how to allocate more than $300 million in surplus that has accumulated.

Some suggest expanding the lottery program by lowering the GPA requirements or by targeting low-income students who otherwise don't qualify. Others want to keep the money in reserve, possibly in an endowment. Others suggest targeting "back to school adults", i.e. adults who attended college but never completed their degrees. The Governor would like to use some of the funds for capital building projects.

All of that sounds great, but there are other problems that aren't being widely reported in media accounts of the lottery program's successes.

Back in the fall, a Tennessee Education Lottery Scholarship Annual Report found some troubling disparities in success rates based on income and race.

The report found that 55 percent of students from families with income less than $36,000 failed to maintain a sufficient GPA to retain their scholarships, while only 42 percent of those from families with income of more than $75,000 failed to retain their scholarships.

The report also found that 63% of African-American students lost their scholarships, as compared to 47% of white students.

There was also an analysis last year suggesting that upper-income families benefit more from the scholarship program than families in lower income brackets:

In Hamilton County, for example, the average student with a lottery scholarship comes from a family that makes $71,980 a year, compared to the $38,930 median household income the U.S. Census Bureau reports for the county.

[..]

In Hancock County, where the median household income is $19,760 a year _ the lowest in the state _ lottery scholarship recipients on average come from families that make $34,146 annually.

Households in Grundy County, the third-poorest county, pull in a median income of $22,959 a year, but lottery recipients from that county on average come from households that make $40,540 annually.

Then there are the findings of a 2005 study by the Tennessee Independent Colleges and Universities Association. In their recommendations to the Tennessee Education Lottery Scholarship program, they cite "lessons learned" from other state lottery programs and some of the unintended consequences.

Regarding enrollment, the study found that in most states lottery scholarships increase enrollment by about 7% overall, except in New Mexico where they shifted enrollment from two-year to four-year institutions.

On outcomes of students staying in school and completing their degrees, the study found mixed results. According to the report, lottery scholarship students in New Mexico were more likely to drop out, while in Georgia lottery scholarship students had higher GPAs and were less likely to drop out.

The report suggest that Georgia's scholarship program resulted in some grade inflation at the high school level to help borderline students qualify for the scholarships. There is also evidence that private colleges in Georgia increased their tuition to get more of the state funding. There was also a finding that the scholarships allowed money from family budgets to be used for other purposes such as purchasing new cars.

According to the report, increased funding for scholarships has not closed the gap between enrollment among lower income families and enrollment among higher income families. While enrollments have increased, the gap remains constant at around 30% more enrollment from higher income families.

The report also found that 93 percent of the Georgia lottery scholarships went to students that would have gone to college anyway, meaning that Georgia spent $9 to get $1 worth of additional college access. The report says "This situation raises questions about the efficiency and effectiveness of the use of state money in Georgia."

In it's summary of "lessons learned", the report says:

- Lottery scholarships have increased higher education participation in states where they have been implemented.

- Lottery scholarships do seem to decrease the out-migration of students.

- There is mixed evidence on the effectiveness of lottery scholarships in increasing student retention and completion.

- There is no evidence one way or the other on the effectiveness of these programs in increasing the educational attainment of the population.

- These programs have some unintended consequences, including grade inflation and alternative consumption.

- The first years of a lottery scholarship program are not representative of its future states, either in terms of revenues or expenditure obligations.

In other words, while no doubt helpful, throwing lottery money at a state's higher education system is not a magic bullet. Which reminds us this quote from last week's discussion of the Education Week "Quality Counts" survey:

...education does not exist in a vacuum. Rather, broader social policies may be needed to address issues of changing demographics, health care, concentrated poverty, and an economy increasingly stratified by wealth.

"Equal opportunity," Richard Rothstein, a research associate at the Washington-based Economic Policy Institute, argues, "requires a full menu of social, economic, and educational reforms: in employment policy, health care, housing, and civil rights enforcement, as well as in schools."