This paper is published in the Brookings Papers on Economic Activity.
This is an excellent journal that publishes papers that are presented
at the Brookings Institution's semi-annual conference on macroeconomics.
The papers are of very high quality, but the journal does not require
them to be edited to normal journal length, so they tend to be very
long. You need not read every word on every page; one of the purposes of
the papers of the week is to help you learn how to read papers
selectively to extract the information you need.
Pages 1-19 are a summary of much of the new Keynesian literature we
have been studying. There shouldn't be much there that is new to you, so
read it quickly, or in more detail if you find the review of our
classroom models to be useful.
The section starting on page 19 introduces a model of price
stickiness from the same family as those we have studied. Do not fret
excessively over the specific form of the equations, but try to get the
basic idea that culminates in the theoretical predictions of Tables 1
and 2. You are reading this paper for its empirical tests, not for the
theory.
The "robustness" section starting on page 29 discusses extensions to
their basic model; skim this. The section on the neoclassical model
summarizes the Lucas model that you have studied. You can skim this
section, too.
For your purposes, the heart of the paper begins on page 33 with the
section on international evidence. Their empirical work takes off from
the Lucas paper you may have read a week ago. Read and analyze this part of
the paper carefully.
Questions for analysis
The Lucas model predicts that high variance of aggregate-demand changes should cause the short-run aggregate-supply curve to be inelastic, but that the expected level of
AD change should have no effect because people can plan any amount of
expected inflation into their expectations if they know about it. How
does introducing price stickiness change the prediction about the effect
of a high level of expected AD change on supply behavior? Interpret
briefly the results in Tables 1 and 2 and explain how the theoretical
implications of the Lucas model would be different.
What is τ (tau) in this paper? To what coefficient in the Lucas model of Romer's Section 6.9 does it most closely correspond?
Explain what the estimated coefficient -1.347 for mean inflation in
the first column of Table 5 means (no pun intended, for once).
Taken as a whole, what does the evidence in Tables 5 through 9 imply
about the applicability of new Keynesian model and the Lucas model?
Explain.