Academic journal article The European Journal of Comparative Economics

Properties of the Histogram Location Approach and the Extent and Change of Downward Nominal Wage Rigidity in the EU

Academic journal article The European Journal of Comparative Economics

Properties of the Histogram Location Approach and the Extent and Change of Downward Nominal Wage Rigidity in the EU

Article excerpt


The histogram location approach has been proposed by Kahn (1997) to estimate the fraction of wage cuts prevented by downward nominal wage rigidity. In this paper, we analyze the validity of the approach by means of a simulation study which yielded evidence of unbiasedness but also of potential underestimation of rigidity parameter uncertainty and therefore of potential anticonservative inference.

We apply the histogram location approach to estimate the extent of downward nominal wage rigidity across the EU for 1995-2001. Our data base is the User Data Base (UDB) of the European Community Household Panel (ECHP). The results show wide variation in the fraction of wage cuts prevented by nominal wage rigidity across the EU. The lowest rigidity parameters are found for the UK, Spain and Ireland, the largest for Portugal and Italy. Analyzing the change of rigidity between sub periods 1995-1997 and 1999-2001 even shows an widening of the differences in nominal wage rigidity. Due to the finding of large differences across the EU, the results imply that the costs of low inflation policies across the EU differ substantially.

JEL Classification: J30

Keywords: Wage rigidity, Inflation, Unemployment

1. Introduction

Of the existing approaches to testing for nominal wage rigidity, the histogram location approach seems promising. Contrarily to most approaches, the histogram location approach proposed by Kahn (1997) allows an explicit estimation of the share [rho] of persons potentially subject to nominal rigidity, referring to the highest effective share in the case of substantial deflation. The share [rho] multiplied by the share of persons who would face negative wage changes in the event of zero nominal wage rigidity yields the share of currently effected employees. Hence, the share of ultimately effected employees differs according to the location of the distribution of wage changes. The more the distribution of nominal wage changes is shifted to the left, the higher the share of ultimately effected employees, who according to the neoclassic paradigm in the following will face a strongly increased probability of becoming unemployed. Hence, if a significant and relevant share of employees is found to be subject to nominal wage rigidity, low inflation may cause unemployment, which, in this situation could be avoided with a policy leading to higher nominal wage changes at unchanged real wage changes due to higher inflation. Therefore, empirical evidence of relevant nominal wage rigidity would be a strong argument for allowing moderate inflation rates above zero inflation. (3)

The histogram location approach suggested by Kahn (1997) has been subsequently applied to different data sets by Beissinger and Knoppik (2001) and Knoppik and Beissinger (2003, 2005) who found strong evidence of downward nominal wage rigidity in Germany and the EU. Christofides and Leung (2003) applied the approach to Canadian contract data and they found strong nominal wage rigidity which might be due partly to the use of union contract data. Holden and Wulfsberg (2004, 2005) apply a non-parametric approach by means of bootstrap and simulation methods which allows the use of much less suitable data at the industry level. Despite the averaging effect of wage changes within industries, their analysis also yielded strong evidence of nominal wage rigidity in the EU. Their estimates are comparable to estimates obtained from micro data. Crawford and Harrison (1997) analyse macro data on wage union settlements 1952-1996 and find the share of constant or almost constant wages to decrease systematically with inflation rates. Using additional micro data sources they find wage cuts to be more common for employees of smaller firms and nonunionised employees.

Smith (2000) discusses such data issues as rounding and measurement error in some detail, based on the British Household Panel Study which could potentially lead to overestimation of the extent of rigidity measures based on survey data. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.