# commentsmankiwcampbell - Some remarks on Mankiw and...

This preview shows pages 1–2. Sign up to view the full content.

Some remarks on Mankiw and Campbell's paper. I only highlight some important elements that were missing in your commentaries. 1) Since, by definition, the change in permanent income is unforecastable, adding a white noise to the budget constraint will not change anything. 2) The permanent income is only a theoretical notion, so it is not observable. Furthermore, it cannot be estimated because its variation is unforecastable. Consequently it is impossible to test directly the theory (equation 9 or 10). The authors are not testing the Keynesian consumption function against the permanent income hypothesis (PIH). Indeed, if λ >0, we know that the PIH is not completely verified but we are not able to interpret what it means for the form of the consumption function. λ>0 could be consistent with keynesian theory (there is a constant part in the keynesian function) if interest rates variations explain (1-λ) for example. The authors are not testing directly the random walk hypothesis ( if they do, it would be sufficient to regress Ct on Ct-1 as in Hall's paper). On page 729, they explain why, under certain assumptions (non correlation between changes in current and permanent income, and weak predictability of the change in current income), a positive λ can be consistent with the random walk hypothesis. Indeed consumption will follow a random walk with a drift (constant). This is very paradoxical : theoretically the random walk result is less general than PIH (consumption smoothing is compatible with non linear utility function while the random walk is not. ..), but Mankiw and Campbell's test allows for the random walk to be more generally accepted than the PIH. .. 3) There are many reasons that can justify the endogeneity between ΔCt and ΔYt , that is why the former cause the latter and vice versa (reverse causality) : Keynesian theory (a rise in consumption affects aggregate demand and income at the same period), statistical construction (since the data are often quarterly averaged, consumption in January can have automatically an impact on income in

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
This is the end of the preview. Sign up to access the rest of the document.

## This note was uploaded on 01/12/2011 for the course ECO 010023 taught by Professor Mrraggillpol during the Fall '09 term at Paris Tech.

### Page1 / 3

commentsmankiwcampbell - Some remarks on Mankiw and...

This preview shows document pages 1 - 2. Sign up to view the full document.

View Full Document
Ask a homework question - tutors are online