The Dynamic Properties of New Keynesian Models with Learning

[This one is, as they say, wonkisk.] As James Bullard noted in the previous post, we "have one of the world's experts on the question of the dynamics" of dynamic stochastic general equilibrium (DSGE) models here at the University of Oregon, particularly models that involve learning.

There has been considerable controversy recently about the dynamic properties of DSGE models near the zero bound, in particular whether raising the federal funds rate target can help to avoid falling into a deflationary trap. So I asked George to explain this on video, and he graciously agreed to do so. The bottom line is that in these models, the type of policy discussed by Minnesota Fed President Narayana Kocherlakota increases rather than decreases the chance of a deflationary spiral. Here's George Evans

Please note: As soon as we were done, George realized there was a typo on the whiteboard. The horizontal axis on both graphs should have the low inflation steady state inflation value labeled β instead of β^-1. The value of .99 on the right-hand graph is correct. Here are corrected versions of the graphs in the video: Figure 1 (on the left in the video), Figure 2 (on the right).

Papers mentioned in the video:

• P. Howitt, Interest rate control and nonconvergence, Journal of Political Economy (1992), vol. 100, pp. 776-800.
• G. Evans, E. Guse and S. Honkapohja, Liqudity traps, learning and stagnation, European Economic Review (2008), vol. 52, pp. 1438-1463.
• J. Benhabib, S. Schmitt-Grohe and M. Uribe, The Perils of Taylor rules, Journal of Economic Theory (2001), vol. 96, pp. 40-69.