Most government subsidy programs are badly designed, and the Minnesota College Savings Plan Matching Grant Program is no exception. The use of a means-tested matching grant means that, depending on your salary, you end up paying to work.
Example: suppose that during 2009 you contributed $4,000 to your child's College Savings Plan and had a family adjusted gross income of $80,000. You'd receive a matching grant from the State of Minnesota of $400. Now, if you were unlucky enough to have had a salary raise of just $1 during 2009, your government subsidy would have fallen from $400 to zero. In other words, an extra dollar of revenue would have led to an increase in your net tax liabilities of a little bit less than $400!
This is not an uncommon problem. At the federal level one can find many examples of poorly designed government programs. More interestingly, governments could eliminate these distortions by simply discarding stepwise income criteria for benefit concession. For example, since the Minnesota Matching Grant Program is capped at a relatively low value, it automatically becomes irrelevant as the income of a family increases. Furthermore, simplifying the program would reduce bureaucratic costs.
Given that the solution is so simple, why is it that governments insist in creating these public finance monstrosities? The explanation is given by public choice theory: the politician goal is not to maximize social welfare, but to maximize political power. Higher levels of control can be achieved by creating rules that increase the size of government bureaucracies and their effects on people's lives. The real solution to the problem, even if an uphill battle, would be to universally restrict the power of politicians to create programs that imply high marginal tax rates.