I'm an optimist because I can imagine as a very realistic prospect at least two other worlds:
1. A world where ARRA failed and QE was never tried - a really awful, awful world that could have happened but didn't, andOf course, we got the world we got, which carries frustrations with it but also could be worse.
2. A world where ARRA was twice as big and spent more wisely and where QE was bigger, had better forward guidance, and IOR was scrapped - a better world.
Market monetarists tend to be very dismal either because they define success as adopting their particular policy instrument (rather than a particular monetary stance), or because they define the monetary stance according to macroeconomic outcomes and not relative to a plausible counter-factual policy. I have some sympathy with the latter - we're clearly "tight" now. But I have little sympathy with defining the policy stance itself by whether or not the desired outcome is achieved - this "do or do not, there is no try" crap. That makes for good movies but it's dumb practically speaking. We can definitely try things that fail and you know what - we've still tried that thing.
Keynesians are somewhat better on this count, so it's unfortunate that Krugman is such a downer about it. Keynesians have output gaps, Okun's law, and multiplier estimates that make it easier to avoid the Yoda garbage and just say "you know what - it could have been worse and at least we got a couple hundred billion in". Kind of like the New Deal in fact. It didn't do much, as we all know, but Americans still radiate with pride decades later over the work of the WPA and the CCC.
Interestingly enough the roles are switched here - Krugman is talking pessimistic and Sumner is talking optimistic. But generally speaking I find that market monetarists are more downers than the Keynesians because of the difference in how they define their goals (Keynesians also think market monetarists have it half right whereas market monetarists seem hell-bent on refusing to take "yes" for an answer from Keynesians on monetary policy).
This is also why I simply don't get the "public choice" argument (I have to put that in quotes for the sake of the dignity of public choice theory) against Keynesianism that says that if politicians are going to botch it up you are somehow obligated to abandon it. I think the actual world is better than the first scenario I described above even though I'd prefer to see the second scenario to both the actual world and the first scenario. It makes exactly zero sense to say that because there are policy failures out there and I can't trust politicians to implement Keynesianism perfectly that it's worthless enterprise. That's like saying because I won't be a completely perfect dad the most sensible thing to do is to never have kids in the first place. Nonsense.
It's very easy to lose perspective amidst the short-run disappointments, but in the big picture there's good reason to stay optimistic and at the very least recognize that it could have been worse and that it's a good thing it wasn't.