Further evidence on money and inflation in the long run

Abstract

The long-run money–inflation relation is re-examined in the context of De Grauwe and Polan's (2005) recent claim that the monetary aggregates are of no use for monetary analysis at low rates of inflation, and Nelson's (2002) counterclaim that in cross-country analysis money should be defined as currency and allowance should be made for lags. I report cross-section results that support a role for the monetary aggregates at low inflation when currency is the monetary aggregate, and panel data results that find no such role, notwithstanding the use of currency and allowing for lags.

Notes

¹De Grauwe and Polan (2005) presented their paper as an attack on Friedman's famous statement that ‘inflation is always and everywhere a monetary phenomenon’ (Friedman, 1963, p. 17). Other economists (e.g. Begg et al., 2002; Svennson, 2002) have cited the DP study as providing strong evidence for de-emphasizing the role of monetary aggregates in monetary policy analysis even in the long run. The timeline of the papers reflects the fact that Begg et al. (2002), Svennson (2002) and Nelson (2002) referred to an earlier draft of the De Grauwe and Polan (2005) paper.

²The variables are the annual average growth of currency in circulation; inflation, as measured by the average annual per cent change in the consumer price index; and output growth, as measured by the annual average change in real Gross Domestic Product (GDP). The data source is the IMF's International Financial Statistics database.

³The outlier inflation countries are Argentina (208.3%), Bolivia (301.7%), Brazil (432.7%), Democratic Republic of the Congo (850.7%), Peru (282.2%) and Zimbabwe (634.7%).

⁴De Grauwe and Polan (2005) argue that grouping observation using the level of inflation (the left-hand-side variable) would create a potential bias, and group them according to money growth (a right-hand-side variable).