Sunday, June 23, 2013

Two questions for Sunday: in re Fed policy...

- by New Deal democrat

I have two questions that I would really like to see academic economists answer, based on past and intended Fed policies:

1. Professor Brad DeLong this morning faults Ben Bernanke for failing to create credible inflation expectations significantly over 2%.

My question is, What would an average wage-earner, who has seen his/her wages go up on average 1.5% or so for the last few years, do if they actually believed that 3%, 4%, or 5% price inflation was around the corner? I believe they would actually spend less and save more to guard against the coming diminution of their earning power. Does anybody have any historical examples of what has happened when wages stagnated in the face of increasing inflation? I seem to recall reading that this has happened a number of times in Latin America, and the outcomes were not pleasant. If so, why do you believe the outcome would be different here?

2. There's a reason Mish is on the sidebar, even though I disagree with a large majority of his opinions. And that is because, especially when he digs into the data, he always makes me think. Three weeks ago he posed a challenge specifically to Prof. Paul Krugman: "When does it [quantitative easing] stop Paul? When?"

I think this is actually an excellent question. Is there an academic model for when quantitative easing should end? Empirically the Fed can ease off, and if the economy falters, start back in, but that is rather like feeling around in an unfamiliar room in the dark.

Krugman and others have quantiative academic models for when quantitative easing is necessary (the zero lower bound) and the forms it might take. But at what level or trend in employment, unemployment, wages, interest rates, or other relevant parameters should it be scaled back? How should it be scaled back? When can it be completely ended? I'm not aware of any such mathematical model, and if one doesn't exist, now would be an excellent time for academic economists to consider one.