274
Views
2
CrossRef citations to date
0
Altmetric
Original Articles

The links between international parity conditions and Granger causality: a study of exchange rates and prices

, &
Pages 3491-3501 | Published online: 08 May 2008
 

Abstract

This article investigates Granger causality between exchange rates and prices for the US and four of its trading partners: Canada, Germany, Japan and the UK. We emphasize the distinction between direct and indirect Granger causality: exchange rates directly cause prices if movements in exchange rates lead movement in prices, and exchange rates indirectly cause prices if deviations from the Purchasing Power Parity (PPP) condition can help forecast movement in prices. But only by including the interest-rate differential in our error correction model do we obtain results that align with economic theory. The economic theory of PPP suggests that exchange rates and prices are cointegrated, with exchange rates moving proportionally to prices in the long run. In general, we find either direct or indirect feedback mechanisms between exchange rates and prices.

Acknowledgements

We are grateful to Judy McDonald and Youngsoo Bae for helpful comments. Any shortcomings of the article, however, are our own.

Notes

1 Frenkel (Citation1981) discounts the belief that PPP binds in the short run. Indeed, the observed high variance and/or persistence of real exchange rates make it unlikely that PPP binds in the short run, and certainly not instantaneously.

2 Consider the International Fisher Effect (IFE): E(%ΔS) = i f i d . Levich (Citation1998, p. 144) shows that if PPP binds instantaneously, an implicit assumption behind the IFE is that real interest rates are equal across countries. If so, the IFE reduces to E(%ΔS) = E(%ΔP f ) − E(%ΔP d ), and thus, we expect exchange rates and interest-rate differentials to be positively correlated. But PPP is expected to bind only in the long run and the correlation between the IRD and inflation differential is much less than unity for any of our four countries. As such, the observed negative correlation between the exchange rate and the interest differential is quite plausible; see also Krugman and Obstfeld (Citation2003, p. 421).

3 Our sample consists of monthly observations from March 1974 to February 2005; but for Germany, we ended our sample in December 2001 due to the end of using the Mark for all businesses in the Eurozone. Our data are primarily from the IMF's International Financial Statistics.

4 Due to space considerations, we do not report numerical results for routine tests. Results are available from the authors upon request.

5 Stock and Watson (Citation1993), Ball (Citation2001) and Hu and Phillips (Citation2004) believe that nominal interest rates are integrated processes. The belief in nonstationary interest rates, however, does not contradict the belief in stationary IRDs.

6 Cheng et al. (2006) do not consider including the IRD in their models. Rather, they emphasize a model of exchange rates and prices that does not impose PPP.

7 To attach orders of magnitudes to our variables, consider that the average Canadian money market rate from 1974 to 2005 was 0.079 with a SD of 0.038. Relative to the US, the average Canadian WPI in logs was 0.186 with a SD of 0.061, and the average Candian CPI in logs was 0.256 with a SD of 0.039.

8 Of course, policy makers may force interest rates to rise in an effort to curb inflation. But, estimates of the adjustment lag of the economy with respect to monetary policy vary from 5 to 24 months. For example, Friedman (Citation1961) suggests that the economy requires about 15 months to adjust fully to exogenous shifts in monetary policy. However, in the short run, a rise in relative interest rates should foretell of an increase in the relative inflation rate between two countries.

9 A good discussion of consistency is found in Diebold (Citation2001).

Reprints and Corporate Permissions

Please note: Selecting permissions does not provide access to the full text of the article, please see our help page How do I view content?

To request a reprint or corporate permissions for this article, please click on the relevant link below:

Academic Permissions

Please note: Selecting permissions does not provide access to the full text of the article, please see our help page How do I view content?

Obtain permissions instantly via Rightslink by clicking on the button below:

If you are unable to obtain permissions via Rightslink, please complete and submit this Permissions form. For more information, please visit our Permissions help page.