Here is a chart compiled by Ezra Klein about a month ago, based on data from the Congressional Budget Office, showing the effects of the Recovery Act of 2009 on unemployment. (There are similar charts in the Klein article and elsewhere showing the effects of the stimulus on GDP, and on overall employment.)
It seems incontrovertible that if many thousands of people are hired to build roads and bridges and other projects, if taxes are cut, and if the federal government increases aid to the states, all these measures must create economic activity. It's not like all these stimulus projects could have taken jobs from the private sector, since the private sector has been busy laying off millions of workers since 2008. Not only can and has all this activity been measured, overall economic statistics have also improved since the stimulus was enacted. Yet those who opposed the stimulus when it passed Congress in 2009 continue to repeat, without evidence, that the stimulus is a failure, as if repetition of this mantra will make it true.
I try to understand the arguments of these opponents, but almost all of them merely illustrate a lot of well-known logical fallacies. For example, the fallacy of inconsistency, i.e., the stimulus was a failure because it consisted of too much government pork, instead of tax cuts which stimulus opponents would have supported. This argument overlooks the fact that the stimulus actually included almost $300 billion in tax cuts. If the government should have cut taxes instead, then a stimulus whose largest component was in fact tax cuts, could not have been a complete failure, by the opponents' own criteria.
Then there is the non sequitur: Christina Romer forecast that unemployment would rise to 9%, and unemployment actually rose to 10%, therefore the stimulus was a failure. All this argument proves is that the forecast was wrong. It says nothing about the effects of the stimulus, because it does not even address the question of what the unemployment rate would have been without the stimulus. That question is addressed in the data summarized in the chart above.
Another logical fallacy is to confuse association with causation, which in this situation can be briefly summarized as follows: the economy still sucks, therefore the stimulus was a failure. Again, the relevant question, admittedly not one to brag too much about, should be, how much worse would the economy have sucked without the stimulus, not how much the economy still sucks. Theoretically, the stimulus could still have helped even if the economy continued to get worse, if data showed that the economy would have fallen even faster without the stimulus. Instead the data show that GDP and employment have steadily risen since the stimulus, albeit at a slower rate than almost everyone would have liked.
Which leads to the next logical fallacy, the unfair argument that the stimulus might have helped, but it didn't help all that much, therefore it was a failure. That is like saying blaming someone for their lack of progress in digging a ditch, when you didn't give him a big enough shovel to dig it with. Many economists thought the size of the stimulus package should have been much larger. It is manifestly unfair to blame the limited success of the stimulus on its limited size. Especially when it was the opponents of the stimulus who caused its size to be so limited.
And if you want to see Rachel Maddow get just as frustrated as I do trying to explain all this stuff, check out this video: