Why Keynesianism Works Better in Theory Than in Practice

This, by Tino Sanandaji, is an excellent piece about why Keynes was both right and wrong, and why the 2009 "stimulus" appears to have failed to stimulate...

...The Obama administration chose a strong dose of Keynesian medicine, confidently predicting that unemployment would be lowered to 6 percent by the end of 2011. As we now know, this promise was not fulfilled, with unemployment currently at 9 percent. This has, fairly or not, once again soured American voters on Keynes, with three-quarters of the public concluding that the stimulus failed.

Of course, the high unemployment rate alone does not prove that the stimulus failed. [LB: Absolutely true, and a point that far too many people miss, on this topic and many others. "Prove" requires far more evidence than we generally have when speaking of counterfactual possibilities. Events have, indeed, "proved" that the administration's predictions of the effects of the stimulus package were wrong, but they don't "prove" that the stimulus did more harm than good. I believe that to be the case, of course, and see the current situation as evidence in support of my position, but I don't claim that it's been proven.] It may be that unemployment would be 15 percent without it, as former House Speaker Nancy Pelosi has suggested. On the other hand, the poor health of the patient cannot be ignored when evaluating a Keynesian cure which in practice has a spotty track record. [LB: This, of course, is also relevant.]

Another argument, most notably made by Nobel Prize winner Paul Krugman, is that the high level of unemployment simply proves that the stimulus was too small....It is worth noting that Paul Krugman has used the same explanation for the failure of Japanese stabilization policy, i.e. that the failure of aggressive Keynesianism to achieve growth proves that deficits were too small.

In the decade following the 1991 crash, Japanese deficit spending was on average 5 percent of GDP per year, and 7 percent of GDP in the decade that followed. Japan’s debt increased from normal levels to a staggering 233 percent of GDP today, far higher than in every other advanced country, including Greece. None of this managed to stimulate growth.

In this vulgar form, Keynesianism is turned into a non-falsifiable theory. [LB: And, of course, falsifiability is essential for something to be a good scientific hypothesis. If the only two positions one allows for are a) the stimulus really worked or b) it really would have worked if it had only been bigger, then one has taken a position that's faith-based, not fact-based. (I recognize, of course, that the converse is true as well. Those of us who will only accept that a) the stimulus was a failure and b) a larger stimulus would have been a larger failure are also holding those positions as positions of faith, because, like the stimulus supporters, we can't prove it. We cannot go back in time and re-run the experiment. It does seem as if the evidence in support of our position is a little bit stronger and more fact-based, though...)] If borrowing and spending 150 percent of GDP fails to achieve growth, why, that merely proves Japan should have borrowed and spent 300 percent of GDP!

This part is good, too.

Still, a doctor whose remedy frequently fails can always say that things would have been even worse without it, but at some point he also has to be open to the possibility that his prescription is wrong.

Remember that the estimates of the number of jobs “created or saved” by the stimulus that we have seen used by the Congressional Budget Office, Mark Zandi, and others are not direct measurements. Rather, they are based on simulations where the effectiveness of the program is assumed a priori by applying a so-called multiplier to deficit spending. As Stanford University economist John Taylor points out: “You learn virtually nothing about the efficacy of a stimulus package if you use the same models to evaluate its impact ex post that you used to predict its impact ex ante.”

Read it all...