Working Paper: NBER ID: w2223
Authors: John V. Campbell; Richard H. Clarida
Abstract: Recent theoretical research in open-economy macroeconomics has emphasized the connection between a country's current account and the intertemporal savings and investment choices of its households, firms, and governments. In this paper, we assess the empirical relevance of the permanent income theory of household saving, a key building block of recent theoretical models of the current account. Using the econometric approach of Campbell (1987), we are able to reject the theory on quarterly aggregate data in Canada and the United Kingdom. However, we also assess the economic significance of these statistical rejections by comparing the behavior of saving with that of an unrestricted vector autoregressive (VAR) forecast of future changes in disposable labor income. If the theory is true, saving should be the best available predictor of future changes in disposable labor income. We find the correlation between saving and the unrestricted VAR forecast to be extremely high in both countries. The results suggest that the theory provides a useful description of the dynamic behavior of household saving in Canada and Britain.
Keywords: household saving; permanent income; Canada; United Kingdom; current account
JEL Codes: E21; F32
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| saving (E21) | changes in disposable labor income (J39) |
| changes in disposable labor income (J39) | saving (E21) |