evidence that mailing checks to people doesn't work

Early in 2008, Bush and his Congresses tried their second Keynesian tax cut (mail checks to people and hope for more consumption). And with Obama in office and the recession dribbling along, we may see more of the same.

But here's a great graph, from an article by John Taylor in the WSJ...

Here's some of Taylor's analysis...

These results may seem surprising, but they are not. They correspond very closely to what basic economic theory tells us. According to the permanent-income theory of Milton Friedman, or the life-cycle theory of Franco Modigliani, temporary increases in income will not lead to significant increases in consumption. However, if increases are longer-term, as in the case of permanent tax cut, then consumption is increased, and by a significant amount.

After years of study and debate, theories based on the permanent-income model led many economists to conclude that discretionary fiscal policy actions, such as temporary rebates, are not a good policy tool. Rather, fiscal policy should focus on the "automatic stabilizers" (the tendency for tax revenues to decline in a recession and transfer payments such as unemployment compensation to increase in a recession), which are built into the tax-and-transfer system, and on more permanent fiscal changes that will positively affect the long-term growth of the economy.

Why did that consensus seem to break down during the public debates about the fiscal stimulus early this year? One reason may have been the apparent success of the rebate payments in 2001. However, those rebate payments were the first installment of more permanent, multiyear tax cuts passed that same year. Hence, they were not temporary.

What are the implications for a second stimulus early next year? The mantra often heard during debates about the first stimulus was that it should be temporary, targeted and timely. Clearly, that mantra must be replaced. In testimony before the Senate Budget Committee on Nov. 19, I recommended alternative principles: permanent, pervasive and predictable.