And where better to deliver such a speech on the one-year anniversary of Lehman Brothers' collapse than Wall Street?

Yesterday's address will no doubt build on the administration's report card, issued last week, on the US$787 billion American Recovery and Reinvestment Act. The president's Council of Economic Advisers determined that the ARRA added about 2.3 percentage points to real gross domestic product growth in the second quarter and created or saved about 1 million jobs as of August.

These estimates "are made by comparing actual economic performance" to the baseline forecast, the CEA said, which is like comparing real life to a Hollywood movie.

While the CEA included the caveat that any definitive assessment of the stimulus's impact is impossible because of the lack of a control study - no one knows what would have happened without the intervention - the Council concludes the intervention had "a substantial positive impact on real GDP growth and on employment in the second and third quarters of 2009."

Nonsense, according to Stanford University economics professor John Taylor. "I can't see any evidence the stimulus is working," Taylor said at a Sept. 10 dinner in New York sponsored by the Hoover Institution, where he is a senior fellow.

Real GDP fell 1 percent in the second quarter compared with declines of 5.4 percent and 6.4 percent in the previous two quarters, respectively. It sure looks as if some kind of dam acted to hold the waters back in the April-to-June quarter.

The only component of final domestic demand to show an increase was government spending. Real consumer spending fell 1 percent, residential investment was down 22.8 percent and business fixed investment slumped 13.5 percent, all at a seasonally adjusted annualized rate.

If the stimulus were working as advertised, consumption would have increased, Taylor said.

Economists have been arguing about the merits of fiscal stimulus for as long as it's been a fixture of crisis management. All governments are guilty when it comes to intervention in the economy, tweaking the name and spinning the effect - supply side, demand side - to suit their political purposes.

The question of whether it works is never settled to the satisfaction of either the proponents or opponents. How can we expect politicians to enact legislation that's good for the economy if the economics profession can't agree on what works and what doesn't?

There is something intuitively appealing about the idea of fiscal stimulus, although it probably has something to do with the name. When animal spirits are depressed, some entity has to step in to get the economy moving, spending money that generates income for others, which begets more spending.

Think about it this way. The federal government is just like you and me (well, sort of). It gets a "paycheck" in the form of taxes (no, you can't withhold the paycheck for non- performance of services). And it uses the money for "household expenditures," including basic necessities (food and shelter for our armed forces), support for the unemployed, sick and elderly, and various pork-barrel projects.

And just like many of us, the government spends more than it "earns." So it has to borrow or raise taxes, in which case you and I have less money to spend.

In other words, government spending is a wash in terms of the U.S. dollars spent now or in the future, except that you and I are more discerning about what we buy and how much we pay.

Sure, when you throw enough money at the economy, there will be some GDP response, now versus later. But there's no free lunch.

There is, however, a third possibility, akin to a magician's hat. You and I can set up a high-tech counterfeiting operation in the basement, forging crisp hundred-dollar bills. The central bank does pretty much the same thing, creating money out of thin air, in which case the federal government has money to spend and it doesn't come from you and me. The only difference is that we go to jail if we're caught.

If it sounds too good to be true, it is. Printing money increases aggregate demand in the short run but leads to higher inflation down the road.

Any discussion of fiscal stimulus - whether it works, how big it should be, where it should be targeted - must therefore start and end with monetary policy. That's what provides the stimulus.

It would go a long way toward alleviating the confusion if we stopped referring to bridge building and farm-price supports as stimulus and called them by their proper name: government spending. By that name, it doesn't smell as sweet.