Cullen Roche cites John Carney in saying QE is akin to a tax increase:

A tax takes dollars out of the private sector, leaving households and businesses with fewer dollars and the government with more dollars. When the government buys something for $10 and sells it back to the private sector for $12, the net effect is the same as if the government had taxed away those $2.
Bernanke doesn’t come out and call quantitative easing a tax. But he comes close.
"The Fed’s asset purchases are not government spending, because the assets the Fed acquired will ultimately be sold back into the market. Indeed, the Fed has made money on its purchases so far, transferring about $200 billion to the Treasury from 2009 through 2011, money that benefited taxpayers by reducing the federal deficit," he explains in one of the prepared slides.
Here’s a good rule of thumb. If something reduces the federal deficit, it is either the equivalent of a spending cut or a tax hike.

So in some ways this makes sense. But in one really really big way it totally fails to make sense.

If the Fed buys something for $10 and sells it for $11 tomorrow, this would be exactly right. But if it buys something in 2009 for $10 and doesn’t sell it for $12 until 2011 that is very, very different – the dollars they placed into the economy have been circulating for two years! A lot has happened between "buy" and "sell" and in the meantime changing the dollars-to-assets ratio in the economy in favor of dollars may very well have had the stimulative effect the Fed was seeking. So when they are selling for more than they bought a) there’s been some inflation since then, enough to make a non-zero dent in the nominal difference in the prices, and b) they are selling into an economy that may be healthier than the one they bought out of. It is true that now that they are selling and replacing dollars with assets that could have an anti-stimulative effect, but they could always just respond to that with more stimulus until the economy is healthy enough that the Fed can relieve it’s "unconventional" balance sheet without much worry.

This reminds me a little of both arguments in favor of "dynamic scoring" for tax cuts as well as arguments in favor of debt-financed stimulus that take projected growth into account in order to mitigate the impact of the new debt issued to finance the stimulus. The reason I am sympathetic to the latter as opposed to the former is that it’s situational – the latter is only invoked when the Fed’s weapons are uniquely insufficient to the task of putting the economy back on trend, while "dynamic scoring" assumes tax cuts have magic powers at just about all times. The logic of "Keynesian" stimulus only applies to ZLB situations; the "logic" of dynamic scoring implies that we should practically eliminate taxes and solicit donations from the far larger economy that would result.