This story is not subject to our Creative Commons license.

This story is co-published by The New York Times Sunday Review. The author, ProPublica reporter Michael Grabell, has written a new book, Money Well Spent? The Truth Behind the Trillion-Dollar Stimulus, the Biggest Economic Recovery Plan in History.

The polarized rhetoric of the 2012 election cycle presents voters with a false choice of whether the government can create jobs or should just get out of the way. The real debate should be about which policies work and which don’t.

I spent three years reporting on the $840 billion stimulus plan that the Obama administration pushed through Congress in 2009. My conclusion: government can create jobs — it just doesn’t often do it well.

The stimulus — a historic package of tax cuts, safety-net spending, infrastructure projects and green-energy investments — certainly did a lot of good. As the economists Alan S. Blinder and Mark Zandi have noted, it’s one of the key reasons the unemployment rate isn’t in double digits now.

But the stimulus ultimately failed to bring about a strong, sustainable recovery. Money was spread far and wide rather than dedicated to programs with the most bang for the buck. “Shovel-ready” projects, those that would put people to work right away, took too long to break ground. Investments in worthwhile long-term projects, on the other hand, were often rushed to meet arbitrary deadlines, and the resulting shoddy outcomes tarnished the projects’ image.

After trumpeting shovel-ready projects as the bedrock of his stimulus plan, President Obama admitted famously that “there’s no such thing.” But there were, and are, shovel-ready projects. The administration just needed to find them. Case in point: the nuclear cleanup at the Savannah River Site in Aiken, S.C., which received $1.6 billion in stimulus money. As soon as the money arrived in the summer of 2009, the retired cold war nuclear plant hired thousands of workers to decommission reactors, install pumps in the liquid waste tanks and ship barrels of solid waste to a salt formation in the Chihuahuan Desert. Workers from out of town filled up nearly all of the area’s apartments, hotels and restaurants. The county’s unemployment dropped to 8.5 percent from 10.2 percent in a matter of months.

Why did it work? Because the government could immediately send billions of dollars to contractors who were already in place for a project that had well-established plans.

The problem with most of the projects was that the Obama administration and Congress had defined “shovel ready” too broadly. The original plan called for putting “shovels in the ground” within 90 days. But when the rules were written, states ended up with 120 days to have their road projects “approved.” It often took six more months to a year before most of the projects were under construction.

Weatherization, for example, was billed as the low-hanging fruit of the clean-energy movement. But states are still sitting on roughly a billion dollars in unused grant money because of a tortured bureaucracy, in which the federal government paid the states, which paid local nonprofits, which then hired the contractors.

Neither states nor nonprofit groups were prepared to handle 20 to 30 times more money than usual. And federal officials brought ready projects to a standstill in the first year by applying new rules regarding prevailing wages.

As a result, the stimulus didn’t provide enough oomph in the first year to overcome the effects of the European debt crisis and rising gas prices in 2010.

The stimulus effort should have contained more programs like Cash for Clunkers, which pulled car sales forward, emptied dealership lots and prompted auto plants to bring back thousands of employees.

“We’re trying to figure out, ‘Man, how did that thing just blow up the way it did?’” President Obama later said. “Essentially, all the auto companies did the marketing. They did the advertising in a way the government just can’t do it and, frankly, even if we did it, people wouldn’t listen.”

The stimulus also could have been more powerful if the administration had pursued a temporary jobs program similar to the Works Progress Administration, which directly employed eight million people during the Depression.

As it was, states could create temporary jobs programs through a $5 billion emergency welfare fund. Not enough states took advantage of it, but those that did saw real results. Fresno County, Calif., where unemployment was 18 percent, found jobs for 2,000 people who were out of work or underemployed.

It also helps to avoid losing jobs in the first place. The promise of $50 billion in state fiscal relief prompted school districts to forgo layoffs. By early 2010, the stimulus money had saved the equivalent of nearly 300,000 full-time teachers and support staff.

Even $50 billion, though, wasn’t enough to plug the budget gaps. The administration should have shifted more money there, and it could have tried to prime a similar effort in the private sector.

Germany’s “work-sharing” program — in which companies reduce hours rather than lay people off, with the government providing partial unemployment benefits to make up for lost wages — has helped keep its unemployment rate below 8 percent since 2008. It also will let companies ramp up quickly when the economy recovers.

Looking ahead

No matter who wins in November, the appetite for a big fiscal stimulus package won’t be there. So what can be done about the 5.5 million Americans who’ve been unemployed for six months or more — a group that includes older workers whom Rutgers labor experts have called “the involuntarily retired”?

A temporary jobs program similar to the one tried in the stimulus, but aimed at the long-term unemployed, could help these people get the skills they need to return to work.

Shovel-ready isn’t as important as it was in early 2009 because we’re not scrambling to stanch economic bleeding. But the lingering malaise gives us an opportunity to make smart decisions about our infrastructure.

The American Society of Civil Engineers has given the nation’s infrastructure an overall grade of D. Fixing deficient bridges, tunnels, dams and sewage-treatment plants, not to mention expanding high-speed Internet and modernizing the electricity grid, should be clear priorities.

Typically, the government spreads money like peanut butter, so that no one can do anything significant and every program is starved. Separate agencies oversee highways, aviation, transit and railroads. The 2009 stimulus introduced a better model: a competitive, $1.5 billion grant program for transportation, called Tiger, that forced local leaders to think regionally about strategies that combined multiple modes of transportation. The money untangled freight rail lines in Chicago, financed streetcars in Dallas and rapid buses in the Washington area, and helped Philadelphia build a 128-mile network of bike and walking trails. It should be a model for future transportation grant programs.

Investments in solar and wind energy, electric cars and high-speed rail make sense, but to have an impact there must be certainty around them. The fluctuations in America’s energy policy, the absence of a trust fund for high-speed rail as there is for highways and aviation, and the clear lack of a plan to tackle the deficit hinder the recovery instead of helping it.

In short, there are areas where the government should get out of the away, by clearing bureaucratic hurdles. But it’s equally important for politics to get out of the way of smart government policies that can help the private sector create jobs.