Ezra Klein writes :
[N]ot all “temporary stimulus spending” is the same. The theory behind a temporary payroll tax cut, for instance, is that it gives Americans a bit more money to spend. But that only gets the economy moving in a significant way if Americans are sufficiently interested in spending that money. The argument against a temporary tax cut is that Americans know it’s temporary and they know that they will eventually have to pay for this sort of spending and so they save the tax cut rather than spend it.
In contrast, Klein believes a new proposal by Senator Reid would lead to more net spending. That’s because instead of channeling the money to taxpayers who might save it, the Reid proposal would channel it to state and local governments who will surely spend it.
The problem is that John Taylor and John Cogan have convincingly shown  that in the last stimulus, despite Keynesian hopes, the state and local governments didn’t use the extra money to increase their net spending. Instead, they used it to decrease borrowing.
In other words, the federal government borrowed and transferred hundreds of billions of dollars to the states which used the money to reduce their own borrowing, largely offsetting the federal increase in borrowing. In the end, little if any new purchasing power was created or advanced through time.
It isn’t clear to me why we should expect anything different this time around.