Back in October 2012, right before President Barack Obama won reelection, conservatives accused him of doctoring the unemployment rate for his own political gain. "Unbelievable jobs numbers..these Chicago guys will do anything..can't debate so change numbers," Jack Welch, the former CEO of General Electric, tweeted after the October jobs report came out.

It was a serious accusation with no basis in reality—and it's making a comeback.

Former Texas Governor Rick Perry, who spoke at Saturday's Iowa Freedom Summit hosted by Representative Steve King, revived the line Monday. “It’s been massaged, it’s been doctored,” he said of the unemployment rate, according to Bloomberg Politics’s Dave Weigel.

There's still no evidence of that. In fact, the conservative argument makes no sense. In 2012, the unemployment rate fell from 8.3 percent in January to 7.8 percent in October, a drop of five percentage points. It rose again in December 2012, and it fell from 8.0 percent in January 2013 to 7.2 percent that October, a drop of eight percentage points. If the White House is doctoring the unemployment rate for political purposes, why did it fall so much more in 2013, the year after Obama’s reelection, than in 2012?

You’ll often hear economists say that the unemployment rate doesn’t represent the actual state of the economy. That’s because millions of Americans dropped out of the labor force, too discouraged to look for work. Because they aren’t actively seeking work, those people are not counted in the official unemployment rate. As the economy improved, economists expected them to rejoin the labor market—and that has happened. But there are still people on the sidelines, causing the unemployment rate to misrepresent the state of the economy.

That’s an important caveat to the economic recovery, and it's a reason policymakers should consider a much wider array of statistics in their decisions. But it's much easier for Perry to accuse the president of doctoring statistics than it is to admit that the economy has improved substantially.