In 1994, Michigan legislators drafted and passed Proposal A: an education finance plan that resulted in drastic changes in the way public schools was funded. One of the main goals of Proposal A was to reduce inter-district spending disparities and eliminate educational inequality. Proposal A increased state aid to the lowest spending school districts and limited the future increases in spending in the districts that were already high spenders, thus abolishing local discretion over school funding. This was accommodated by a rise in the state sales tax, increasing it from four to six percent and reducing the reliance of school revenues on property taxes.. To determine the success of reforms such as this, Joydeep Roy from Columbia University conducted a study on the effects of Proposal A. He intended to analyze its impact on the distribution of resources and its educational outcomes. Roy used data from the state's 524 K-12 school districts over a period of approximately eleven years (1990-2001), to highlight the differences in the pre-reform and post-reform trends across those districts.
Evaluating Proposal A's effects on school funding, Roy used four different means of measurement: the gini coefficient, coefficient of variation, the Theil index, and a ratio of percentiles to ensure accuracy. The study separated the K-12 school districts into five equal groups based on the level of spending in that district. The study controlled for racial composition, location of district and size. In order to prove inequality he investigated the strength of the correlation between median household income and per pupil spending for the years 1990, 1994, 1998, and 2001. He found that the pre-reform gap between the highest spending and the lowest spending districts had been increasing. A distinct hierarchy became apparent; the higher the groups wealth (revenues and expenditures), the higher it's growth rate of spending on schools. The post-reform results indicate a significant decline in differences in per pupil spending between districts. In fact, the reform completely reversed the hierarchy.
Roy also studied both the pre-reform and post-reform effects of Proposal A on academic performance, using scores from standardized test such as the MEAP and the ACT. The study shows the distribution of the changes in fourth grade proficiency results in reading and mathematics in the highest spending groups and the lowest spending groups between 1995 and 2001. The results in either subject show a narrowing of the achievement gap between the highest and lowest spending districts post-reform. There is some evidence for academic improvement in low spending districts and a decline in the academic performance of students in higher spending districts in the post-reform period, however there is no clear trend in student performance across the years. When Roy used the ACT scores to determine if Proposal A had any effects on performance he didn't find any evidence to suggest that there was a positive effect of the reform on the lower spending districts.
Evidence from these investigations suggest an improved performance by the lowest spending districts later on in the decade, however the restraint on increases in funding in the highest spending districts seemed to have had a negative effect on the academic performance of it's students even though they were already spending at high levels. Overall, the improvements in the low spending districts roughly parallel the increase in per pupil spending suggesting that increases in spending and academic performance are causally related.