I just read an outstanding essay in this month’s Atlantic Magazine titled “Can Government Play Moneyball.” It asks whether the federal government can analyze the effectiveness of programs in the same way that baseball now analyzes the effectiveness of ballplayers. This means they are evaluated based on statistical data and verifiable criteria to judge effectiveness, and not hunches and wish full thinking. The authors, both former senior high level budgetary officials, one in the Bush administration and the other in the Obama administration (Orszag), describe their frustrations in trying to evaluate the actual effectiveness of programs, and disappointment that ineffective programs could not be killed because of political support.
The article is available on line here: Can Government Play Moneyball
The article notes that administrations since Clinton have tried to evaluate, on some level, the effectiveness of programs. Remember Clinton’s Reinventing Government? The Bush Administration established the Program Assessment Rating Tool, which the Obama Administration was continued. All of these efforts have identified government programs that simply don’t work, but in many cases, when the administration tries to defund the program, it runs into opposition for Congress. The article mentions two programs, one supporting individualized payments under Medicare, rather than bundling, and the other an after school education program, that were both shown, irrefutably, to be ineffective, but neither administration was able to kill the program due to opposition from Congress.
The article notes that the Bush program evaluated roughly 1000 programs, and 19% were rated effective, 32% were moderately effective, 29% adequate, 3% ineffective, and another 17% un-ratable due to insufficient data. I was surprised by these numbers. I would have thought that 10 to 15% of the programs would be ineffective, so I’m surprised that such a high percentage were at least somewhat effective.
The point is that government programs should be evaluated for effectiveness, and those that are not effective should be jettisoned. The problem is that there is never any provision in the bill creating a program to evaluate its effectiveness. The article suggests a number of proposals to address this defect. One proposal suggests that 1% of the budget of a program should be set aside for evaluation. The results of the evaluation can be used to improve the program if it needs it, or justify killing the program if it is proving ineffective. Another proposal, which New York Mayor Michael Bloomberg is using, is to make many new programs temporary, and they only become permanent if they prove their effectiveness.
I think that these are excellent ideas. Ineffective government programs should be eliminated, but their effectiveness should be evaluated based on valid criteria and good data, and not political beliefs and desires. I particularly like the idea of subjecting all new programs to a “probationary” period for them to prove their effectiveness. Government should not be wasting money (the taxpayers money) on programs that don’t work.