In the past year, several ideas have dominated the discourse on improving government. A couple of months ago, the President called for improving the management and results of government. A recent OMB Memorandum, M-13-17, specifically calls for “evidence based management” and calls for the use of “random trials” to explore and experiment with promising new ideas.
There is much to think about in the advent of so-called “Moneyball government.” Another term for what is purported to be “evidence-based government.” it all SOUNDS sexy and good. But… The notions of “evidence” in complex domain challenges don’t seem to be as claimed. In large part because the presumed “evidence” is derived from highly variable interactions leading to significant differences in data. Running government isn’t like doing a random clinical trial on a new drug. Government leaders can’t isolate a system to a single variable, or control their sample population characteristics and behaviors.