Over the past six years, America’s economy has received numerous doses of fiscal and monetary stimulus, beginning with the relief package (a mix of tax-rebate checks, temporary business-tax breaks, and temporary increases in federal mortgage limits) that President Bush signed into law in February 2008, all the way through the Federal Reserve’s current round of “quantitative easing” (in which the central bank has been purchasing Treasury and mortgage-backed securities).

Yet when you mention “the stimulus,” most people assume you’re talking about the 2009 American Recovery and Reinvestment Act (ARRA), which Congress passed less than a month after President Obama took office, with only three Republicans voting in favor. (One of those Republicans — the late Arlen Specter — would soon switch parties and become a Democrat.) Half a decade later, the ARRA remains as controversial and politically polarizing as ever.

Some analysts complain that it was insufficiently large. Some believe it was poorly designed. Some think it should’ve been complemented by a more aggressive monetary policy, and perhaps even a higher inflation target. Some contend we should’ve eschewed fiscal policy entirely and relied on the Fed to stabilize nominal-GDP growth. Some say our most critical mistake was failing to embrace a bold strategy for reducing post-bubble housing debt.

Writing in the Washington Post, columnist Robert Samuelson recently examined competing arguments about the ARRA’s relative success or failure. He noted that economists are divided on the Keynesian “multiplier” effect, with a Congressional Budget Office (CBO) survey finding that (in Samuelson’s words) “multiplier estimates for government investment spending ranged from 0.5 to 2.5. This means that a dollar of spending generates between 50 cents and $2.50 in added output (gross domestic product). Quite a range!” Samuelson then offered his “messy verdict” on what the ARRA accomplished:

When proposed, President Obama’s stimulus was desirable. (Disclosure: Though disliking details, I favored it.) Regardless of multipliers, it supported the economy. It also sent a message along with the auto-industry bailout, the Federal Reserve’s easy money and the Troubled Asset Relief Program: The government won’t let the economy collapse. This was crucial to restoring confidence. The stimulus was a justifiable emergency measure.

But the emergency has passed. The economy, though struggling, is not failing. The administration attributes its sluggishness to many causes (household debt, Europe’s problems, Washington’s political discord). Maybe. But more stimulus won’t cure underperformance and may perversely contribute to it. By highlighting the economy’s weakness, it may magnify consumer and business caution. Pessimism becomes self-fulfilling. An economy dependent on periodic shots of stimulus is an economy in eclipse.

The key question about the ARRA is not whether it had a temporary impact on the economy. (As Douglas Holtz-Eakin, the former CBO director who served as John McCain’s top economic advisor during the 2008 presidential campaign, acknowledged in a 2011 interview with journalist Ezra Klein, “If you throw a trillion dollars at the economy, it has an impact. I would have preferred to do it differently, but they needed to do something.”) Rather, the key question is whether it addressed America’s most urgent economic needs, whether it was properly structured, and whether its short-term benefits outweighed its long-term costs.

Two years ago, Political Science Quarterly published a fascinating study of the ARRA’s geographic distribution by scholars James Gimpel and Frances Lee of the University of Maryland and Rebecca Thorpe of the University of Washington. (Alas, the full article is hidden behind a pay wall.) “Contrary to the expressed intent of the legislation,” they concluded, “additional funds were not directed to areas where the recession’s downturn was most severe” (emphasis added). As the authors explained:

Counties that emerged with more ARRA funds were not those that suffered more from the recession, but were instead locations in which funds could advance other prominent policy goals that were important to members of Congress and the administration at the time. The areas that received greater federal support better advanced policy goals in promoting clean energy, fostering medical and scientific research, repairing existing infrastructure, and subsidizing state and local government.

If you’re still wondering why “the stimulus” proved a disappointment, that’s one big reason.