This paper studies the degree of downward rigidity in nominal wages in the United Kingdom using micro-data. Around 9 % of employees who remain in the same job from one year to the next have zero pay growth. But on investigating the causes of rigidity we ®nd that up to ninetenths can be attributed to `symmetric ' causes (such as contracts and menu costs) or to error. Thus only 1 % of workers have pay that may be downwardly rigid. This suggests asymmetric, downward rigidity is not large enough to have serious macroeconomic consequences. The labour market provides almost no evidence to support a positive in¯ation target. The idea that nominal wages might be rigid has a long history, stretching back at least as far as Keynes (1936). In addition to any intrinsic interest in the behaviour of the price of labour, the issue has important implications for the macroeconomy. Downwardly rigid nominal wages could provide a rationale for a positive optimal rate of in¯ation and hence for the pursuance of a positive in¯ation target. As Tobin (1972) argued, if wages are rigid downwards, negative demand or positive supply shocks leave real wages higher than marginal product, resulting in unemployment. Higher in¯ation could alleviate this

The paper look for evidence of grease and sand effects in Europe, in particular the possibility that the natural rate of unemployment is affected in the long run by the inflation rate. LookPP at four countries, France, Germany, the Netherlands and Switzerland, the paper reports some preliminary evidence that the long-run rate of unemployment is a nonlinear function of inflation. The particular shape of the empirical relationship supports the viewthat a moderate level of inflation provides some "grease" to the price and wage setting process. In particular, the long-run rate of unemployment is found to reach a maximum between 0.5% and 1%, and toquickU decline for higher rates of inflation. For the range of inflation rates observed in the sample countries, there is no evidence of sand effects, that uncertainty associated with inflation adversely affect the longrun rate of unemployment.

ecogni?:: of the need togiq central banks iksr:?:J:rI and clear mandates to pursue theobjectir of pri'GGrIq'A?Gr . As a result, most developed countrid are currently experilyr?: rates ofirq`MAqr i consumerprium that are low, at least by recent hi?Aq - ir? standards. Therefore I would argue that the realquestiM now iw `Why notpri` stabiG?rIq Gita that any transi'rIqA costsi movir topri' stabiMMr have already beenie curred, can any respectable case be madeagai:: the currentsirentr` of very lowiwr`M tiw and id perpetuatiqr What would be thebenefi' ofallowi: ilowi:r tori? agaiA Ithi? that the answers to thesequesti:A are clear and that there are no compelli ? arguments that would lead us to return to thesir:OGJG thatexir" i the 1970s and 1980s wheninrqqJAM di so much damage to many ofoureconomiM . 4 WhyPric Stability? OtmarIssi' * I would lid to thank Juli' Morgan for hi valuablecontrir:G"G 1 Indeed, accordiG to the survey results f

1 Comments and discussions with V. Gaspar, K. Masuch, S. Nicoletti-Altimari and H.-J. Klöckers are gratefully acknowledged. Comments by G. Kenny, F. Mongelli, R. Motto, P. Moutot,

411 144 ecb d All rights reserved. Reproduction for educational and non-commercial purposes is permitted provided that the source is acknowledged. ISBN 92-9181-373-7 (print)

Using the accurate and extensive data available in the UK New Earnings Survey, this paper investigates the extent to which nominal wages are downwardly rigid and whether such rigidity interferes with necessary real wage adjustments when inflation is low. Despite the substantial numbers of individuals whose nominal wages fall from one year to the next, we find that if long-run inflation is one percent higher, the number of individuals with negative real pay growth increases by around 1.4 percent. This is controlling for the median and dispersion of the real wage change distribution.

3 Business cycles asymmetry and monetary policy: 4 a further investigation using MRSTAR models a b c,

new survey evidence on price stickiness

time-varying price rigidity