The evidence cited there is for government spending during World War II and the Korean conflict. This would seem to be grossly misleading. During WWII, in particular, the US was producing just about as much as it physically could, and a lot of what it was producing was war materials being shipped overseas. There couldn’t be a multiplier because there was excess demand in the US which simply could not be met – we could not produce more than we were. (Citizens were encouraged to save instead through war bonds.) That is one reason why after the war there was such a housing boom and the suburban US developed and the sales of cars and durables like refrigerators, washing machines, etc. exploded. Much of that spending was the multiplier of the WWII spending.
Also all tax decreases do not have the same multiplier. One reason why the SSI tax was cut recently was that this type of tax cut has a much bigger multiplier than cutting the capital gains tax, for example. The reason is simple – an increase in a low-paid worker’s take-home pay is quite often spent immediately on goods and services.
Plus, experiments in behavioral economics have shown that big individual checks go into a different mental bucket than small increases in pay. Bush gave tax cuts to the middle classes by single checks - this got saved. The Obama method is more likely to be spent. The downside was that the Bush one was more visible as a gift from the President while the Obama way was kind of invisible.
Cass Sunstein (co-author of “Nudge” with Richard Thaler) was working in Washington and might have inspired this stuff.
BTW, slightly off subject, the Times reported that a volunteer committee of behavioral scientists advised the Obama campaign, recommending the most effective ways of getting people to vote, for instance. Science not only is good for the world, it can help win elections.