How many times have you heard someone say that the Federal Reserve's asset-purchase program known as quantitative easing was ineffective?

At least, that’s what I keep hearing from the usual pundits arguing their case on television and in print.

I have my own suspicions that monetary stimulus works well enough to thaw frozen credit markets, as it did amid the 2008 financial crisis. But if we want to stimulate the economy, then a fiscal, rather than monetary policy, is the way to go (nevermind that lots of pundits also claim that the federal government's 2009 economic stimulus package didn't work either).

But proving the efficacy of monetary or fiscal policy can't easily be accomplished. This is one of the biggest challenges for economics, and why critics sometimes say that economists suffer from physics envy (I'm ignoring the Freudian implications here, but would note that the phenomenon has its own Wikipedia entry).

The problem with any line of economic policy argument is that there often isn't much of a way to test any one thesis in order to prove or disprove the underlying claim. There is, for example, no laboratory where we can test side-by-side how QE performed for the U.S. economy versus an alternative scenario without QE. If we had that ability, we could get a better idea of whether QE, zero interest rates, fiscal stimulus or other programs work. Alternatively, we could run experiments about what would happen if there was no central bank or federal government economic intervention, letting the system cleanse itself of excesses.

If only. Instead, many pundits fall back on partial arguments without recognizing what the lack of a counterfactual means to their analysis. As we noted two years ago:

This flawed analytical paradigm has many manifestations, and not just in the investing world. They all rely on the same equation: If you do X, and there is no measurable change, X is therefore ineffective.

The problem with this non-result result is what would have occurred otherwise. Might "no change" be an improvement from what otherwise would have happened? No change, last time I checked, is better than a free-fall.

This, or course, has consequences not just for economic-policy choices, but also for  corporate decision-makers, investors and -- not least of all -- voters in the U.S. presidential election.

There is some good news on this issue: the states, those so-called laboratories of democracy, have been engaging in a variety of different policy experiments. These are obviously not perfect scientifically rigorous experiments; it's impossible in complex social and economic systems to create two identical test groups, with the variable to be verified versus a control group. However, the various states can give us some ideas about how successful one form of public policy may be versus its peers and the national average.