Discussion Paper

No. 2013-28 | April 22, 2013
Within-Establishment Wage Inequality and Satisfaction


The aim of this paper is to provide fresh empirical evidence of the mechanisms through which wage inequality affects worker satisfaction. Theoretically, wages of others may affect workers’ utility for two main reasons: Workers may derive well-being from their social status (comparison hypothesis) and/or they may use others wages to help predict their own future wage (information hypothesis). Both hypotheses are tested. To achieve her aims, the author models individual utility from pay as a function of a worker’s own wage and the earnings of all other workers within the same establishment, and she estimates the model using British employer-employee data. Incomplete information about others wages is assumed. The author finds that the comparison effects matter. Of most interest, she provides some first evidence about a positive relation between well-being and inequality. Her results are robust within the different specifications and different definitions of the reference group.

JEL Classification:

J28, J31, J33, C25


  • Downloads: 1297


Cite As

Ambra Poggi (2013). Within-Establishment Wage Inequality and Satisfaction. Economics Discussion Papers, No 2013-28, Kiel Institute for the World Economy. http://www.economics-ejournal.org/economics/discussionpapers/2013-28

Comments and Questions

Anonymous - Referee Report 1
June 19, 2013 - 08:47

see attached file

Ambra Poggi - answers to referee's comments
August 02, 2013 - 12:19

I’d like to thank the referee for the useful comments. I understand the concern of the referee about data limitations and the suggestion to be more cautions when describing the contribution of the paper. I modified consequently the paper pointing out that “the existence of a positive relationship between well-being ...[more]

... and inequality deserves to be further investigated to exclude that it depends on data specific characteristics.” (p 2, last paragraph). I also improved the discussion about data limitations in Section 4.

Please see below the answers to the referee’s specific comments.
1) The referee is worry about a possible high correlation between the main variables of interest (wage, upward and downwards comparisons). Indeed, these variable are correlated, but the correlation is not so high to be concerned about the estimations. See below the correlation matrix. I performed a formal detection-tolerance and the variance inflation factor (VIF) for multicollinearity. A tolerance of less than 0.20 or 0.10 and/or a VIF of 5 or 10 and above indicates a multicollinearity problem (O'Brien 2007). This is not our case, therefore there is no evidence of multicollinearity problems.

Correlation | w pride envy
wage | 1.0000
pride | 0.5346 1.0000
envy | -0.2183 -0.3148 1.0000

Collinearity Diagnostics
Variable VIF VIF Tolerance Squared
wage 1.41 1.19 0.7114 0.2886
pride 1.49 1.22 0.6730 0.3270
envy 1.11 1.06 0.8974 0.1026
Mean VIF 1.34

2) I followed the suggestion of the referee to be clear that the implementation of the assumption of incomplete information about others’ wages is imposed by data availability. See Section 4, pp. 8, first paragraph.
3) Figure 1 is indeed a histogram: I corrected the description of Figures 1. see Section 4, p.7, last paragraph.

Anonymous - Referee Report 2
July 16, 2013 - 12:10

see attached file

Anonymous - answers to referee's comments
August 02, 2013 - 12:23

see file attached