The liquidity trap really isn't something to be feared, as long as we approach it the right way. And it's an important inflection point for human civilization.
Not that he agrees with my general spin on the circumstances, of course, but Scott Sumner writes:
"Matt Yglesias seems to agree that interest rate targeting won’t work in the future:
'Interest rate targeting had a good run because everyone understood the convention, but the combination of low inflation and population aging means that negative demand shocks are now going to regularly put us at-or-near the zero bound. That means we’d be well-served to find some other monetary routine to appeal to.'
As is often the case, I think he’s ahead of most economists in spotting this problem. I don’t sense that most macroeconomists understand that rates will frequently fall to zero in future recessions, despite not having hit zero once in the half century before the current crisis. We need a different monetary policy instrument/short term target."
I don't know... I feel like I've heard this point a lot, but maybe. But I think Scott misdiagnoses what the problem really is. Sure interest rate targeting is going to be a stranger way to think about monetary policy. It never really was the right way to think about it - it was just nice because it described the monetary policy reaction function in a way that reflected the way central bankers talked in the last half of the twentieth century.
What would the alternative be? What Sumner Rule would replace the Taylor Rule?
It would pretty much look like the Taylor Rule.
On the LHS you'd have "NGDP". On the RHS you'd just reparameterize everything, probably using the exact same data that Taylor used. Bingo: a Sumner Rule.
I would still complain (as I have with the Taylor Rule) that NGDP is no more a policy lever than the interest rate and that you really need to put OMOs on the LHS and then reparameterize everything. But nobody really cares about my complaints on that and the central bankers will know how to operationalize it all, and the Sumner Rule will suit the purposes of macroeconomists just like the Taylor Rule suited their purposes for the last twenty years or so.
And nothing really will change at all about how we think or talk about monetary policy and the macroeconomy. You'll still have an IS curve. You'll still have a Phillip's Curve. You'll still have a monetary policy reaction function (in this case a Sumner Rule), and everything will run about the same as it did before the crisis.
The interesting part is the Keynesian part. With the likelihood that interest rates will stay near zero for a while we'll have to talk more seriously about fiscal policy. And that's actually a good thing. The surplus value produced by the market economy has, in the past, allowed us to collectively provide things that met a cost benefit test like education and infrastructure, and some that didn't pass a monetary cost benefit test but that were consistent with our values - like a safety net. Our Keynesian future gives us even more leg room to socialize investment. And there is a lot of good that can be done.
That's not to say I wouldn't want private investment demand to be sufficient to meet all the forces alligned against it. That would be wonderful too (and that could be the case... I've expressed my doubts about predictions about private demand in Keynes's Economic Possibilities for Our Grandchildren in the past). But in the absence of that, it doesn't have to be doom and gloom - if we play our cards right.