Evidence on the Functional Relationship between Relative Price Variability and Inflation with Implications for Monetary Policy

Abstract

This paper considers the functional relationship between relative price variability (RPV) and inflation using quarterly personal consumer expenditure data in the US over 1967(I)-2003(I), deriving some implications for monetary policy. Non-parametric regression techniques are used to explore the shape of the RPV function, and this shape is related to the theory outlined by J. J. Rotemberg. The empirical RPV function yields a plausible 'optimal' value of inflation in the region of five percentage points, which is consistent with this theory and lies within the range of values cited by recent evaluations of 'sand' and 'grease' effects. Copyright (c) The London School of Economics and Political Science 2007.

Relative price variability and inflation: New evidence

“…This evidence indicates that the inflation-RPV nexus exhibits quite different dynamics depending on the inflationary environment, consistently with Choi et al (2011), where the inflation-RPV relation is found to be linear in high inflation regimes, while nonlinear and U-shaped in more stable environments. Although this hump-shaped relation seems inconsistent with the U-shaped relation found in the empirical literature (e.g., Choi, 2010, Choi and Kim, 2010, and Fielding and Mizen, 2008, because this study covers periods with much higher levels of inflation (ranging between 0 % and 90 %), this result may be considered a generalization of the results in earlier studies, suggesting that the U-shaped relation can be confined to periods with relatively low levels of inflation but not long-lasting high inflation.…”

“…Results for the Euro area presented by Nautz and Scharff (2012), indicate that inflation significantly increases RPV only if inflation is either very low or very high in the range of their sample values. 3 More recently, conformable with recent monetary search and Calvo-type model predictions (see Head andKumar, 2005 andChoi, 2010), evidence has been provided of a U-shaped relationship between inflation and RPV by Choi (2010) for the US and Japan, Choi and Kim (2010) for the US, Canada and Japan, Becker (2011) for a panel of European countries, and Fielding and Mizen (2008) for the US. 4 Moreover, in a more recent study of the effect of inflation targeting (IT) on the inflation-RPV nexus, Choi et al (2011) analyzed a data set of twenty industrial and developing countries consisting of 12 targeters and eight non-targeters, including Turkey, during the so-called great moderation period.…”

“…5 In this paper, we contribute to this body of literature by estimating the relationship between RPV and inflation using a semi-parametric method that allows us to estimate varying coefficients capturing changing effects of inflation, if they exist, on RPV at different levels of inflation. In this respect, we use an estimation method similar to those of Choi (2010), Choi and Kim (2010), Choi et al (2011) andMizen (2008) in a panel data context by introducing further regional dimensions in addition to goods levels. This unique data set covers quite a large range of (an-3 In their studies, a very low level of inflation refers to a per annum rate below 0.95 % and a very high level is 4.96 %.…”

Evidence on the effects of inflation on price dispersion under indexation

“…They explicitly state that they use lagged squared inflation to avoid the endogeneity problem, but the endogeneity status of the conditional inflation variance remains dubious. There is also a recent literature where the marginal impact of inflation depends on the inflation level in a different way than a second-order polynomial of inflation : Nautz and Scharff (2007) and Bick and Nautz (2008) use threshold models with low, interme-diate, and high inflation regimes (but use a static relation), whereas Fielding and Mizen (2007) fit a non-parametric smooth-transition function.…”

“…In contrast, most theoretical models posit that positive as well as negative monetary shocks should increase RPV. In terms of dynamics the specification of Fielding and Mizen (2000) is related to ours, but they do not test the possible endogeneity of the inflation rate. Grier and Perry (1996) use a model where lagged RPV, lagged squared inflation and the conditional variance of inflation all affect RPV.…”