How is stimulus doing? Fair reckoning is due

Of all President Obama's efforts to jump-start the economy, his Cash for Clunkers program easily has been the most popular.

What's not to like about a government handout to buy a new car?

But a study released last week by Edmunds.com, the California-based auto industry analysts best known for their consumer guides to car prices, concludes that Cash for Clunkers was pretty much a lemon, shoveling most of its $3 billion into the hands of people who would have bought a car anyway.

Edmunds' analysis is intriguing, though the White House says otherwise. Edmunds fairly calls into question the real impact of the Cash for Clunkers program and, more broadly, reminds us that the impact of every multibillion-dollar program created by the White House and Congress to stimulate the economy must be dispassionately assessed.

Let's learn what works and what does not, without regard for the political fallout.

This page has supported bailouts and economic stimulus efforts by both President Bushs and Obama. A year ago, when the economy was tailspinning toward a second Great Depression, government intervention was urgently needed. Today, thanks in large part to that bold action, the economy has begun to stabilize, even as the recession lingers.

But with that bold action comes the need for a full and accurate reckoning of the costs and benefits, and on that score recent developments suggest the Obama administration is falling short. First the Edmunds study raised our doubts, then on Wednesday the Chicago Tribune reported that the White House may have overestimated the jobs-saving impact of $1.25 billion in stimulus money given to Illinois schools.

According to Edmunds, five out of every six of the 690,000 cars sold during the Cash for Clunker promotion last summer -- with rebates of up to $4,500 -- would have been sold anyway. Only 125,000 cars were sold to people who were not already in the market for a new vehicle.

Doing the math, Edmunds says, that means "Cash for Clunkers cost taxpayers $24,000 per vehicle sold."

The White House has snorted at the Edmunds study, saying it rests on false assumptions, and has accused the company of trying to "get on cable TV." But the administration's own study, which credits Cash for Clunkers for a whopping 500,000 additional car sales in July and August, relies on its own debatable assumptions.

Edmunds' approach was to look at sales of gas guzzlers, such as SUVs, during those months that were not eligible for the Cash for Clunkers program. Historically, Edmunds points out, there is a consistent relationship between how many of those vehicles are sold and how many more fuel-efficient vehicles are sold. For every 10 SUVs sold, for example, a certain number of Ford sedans typically find buyers.

From that, Edmunds calculated roughly how many cars they think would have been sold even if there had been no Cash for Clunkers program -- five out of six.

The Obama administration's own study, by the president's Council of Economic Advisers, offers a more complex, but not necessarily more convincing, analysis. It relies heavily on the car industry's estimates of how many new cars typically are bought in normal times by people unloading seriously old cars -- clunkers. From that, it extrapolates how many more cars were bought because of Cash for Clunkers.

Who's right -- Edmunds or the council? We can't say for sure. But the White House's efforts to shoot down the Edmunds study look pretty feeble so far.

Faced with an economy falling apart earlier this year, Obama took fast and sometimes risky action. The risk of doing nothing was so much greater.

Now let's learn what has worked best.