Pay Growth, Fairness, and Job Satisfaction: Implications for Nominal and Real Wage Rigidity

• DOI: http://doi.org/10.1111/sjoe.12091
• Published date: 01 July 2015
• Date: 01 July 2015
• Author: Jennifer C. Smith

Scand. J. of Economics 117(3), 852–877, 2015

DOI: 10.1111/sjoe.12091

Pay Growth, Fairness, and Job

Satisfaction: Implications for Nominal and

Real Wage Rigidity∗

Jennifer C. Smith

University of Warwick, Coventry, CV4 7AL, UK

jennifer.smith@warwick.ac.uk

Abstract

Theories of wage rigidity often rely on a positive relationship between pay changes and

utility, arising from concern for fairness or gift exchange. Supportive evidence has emerged

from laboratory experiments, but the link has not yet been established with field data. This

paper contributes a first step, using representative British data. Workers care about the level

and the growth of earnings. Below-median wage increases lead to an insult effect, except

when similar workers have real wage reductions or when firm production is falling. Nominal

pay cuts appear to be insulting even when the firm is doing badly.

Keywords: Gift exchange; pay cuts; social comparisons

JEL classification:E24; J28; J33; M52

I. Introduction

Wage dynamics are complex: micro data reveal notable heterogeneity, with

significant nominal or real rigidity for some, coexisting with apparent

flexibility for others (e.g., Smith, 2000; Dickens et al., 2007). At present,

the foundations for wage rigidity are poorly understood. Here, I examine

the extent to which wage dynamics are consistent using a model of wage-

setting based on fairness, where workers’ utility depends on how their own

income compares with some reference income.

There is strong evidence from laboratory experiments that fairness con-

siderations drive participants’ reactions to wages and wage changes, and

surveys of managers consistently report concern for workers’ morale when

setting wages (Blinder and Choi, 1990; Fehr et al., 1993, 2009; Agell and

Lundborg, 1995, 2003; Campbell and Kamlani, 1997; Bewley, 1999). How-

ever, the link between wage rigidity and potential fairness foundations has

not yet been established with field data. Using British field data from 1991

to 2007, I examine the effects of compensation and, in particular, changes

∗I am grateful to Chris Doyle, Barry Smith, Jeremy Smith, two anonymous referees, and

participants at the Royal Economic Society Conference 2013 and EALE-SOLE 2010. I would

also like to thank Truman Bewley and Peter Howlett for comments on related issues.

CThe editors of The Scandinavian Journal of Economics 2014.

J. C. Smith 853

in compensation on job satisfaction. Three questions are addressed. Does

job satisfaction depend on the change in pay in addition to the level of

pay? If so, does the effect of a pay change differ between increases and de-

creases (either real or nominal)? Does satisfaction with pay changes depend

on what is happening elsewhere?

Few researchers have investigated the effect of pay changes on job sat-

isfaction. British, German, and Japanese field data have been analysed by

Clark (1999), Grund and Sliwka (2007), and Kawaguchi and Ohtake (2007),

respectively, but these studies are limited in scope and do not allow robust

detailed conclusions to be drawn about the effect of pay cuts. Performance

declines were found after a pay change loss by Mas (2006) and, to some

extent, by Lee and Rupp (2007), using data on specific employee groups

(police and pilots, respectively).

A small but growing number of field experiments have also examined

the relationship between wage, fairness, and effort. The detrimental effect

of downward deviation from a reference point is a reasonably consistent

result. Findings are more mixed concerning the effect of a rise above the

reference point (e.g., Cohn et al., 2013; Kube et al., 2013).

Recent experimental investigations have also unearthed sources of het-

erogeneity in reactions to pay changes stemming from comparisons with

other workers (Cohn et al., 2011), firm profitability (Hennig-Schmidt et al.,

2010), perceived pay fairness (Cohn et al., 2013), and expectations (Abeler

et al., 2011). I examine whether these findings carry over to representative

natural field data.

This study shows that pay growth has a positive effect on job satisfaction,

consistent with explanations for downward wage rigidity based on fairness,

such as the gift exchange model of Akerlof (1982). If earnings are cut, job

satisfaction suffers even more than would be expected, given the general

effect of pay growth; indeed, there is a step reduction in job satisfaction

(i.e., the insult effect of Bewley, 1999). This reduction in satisfaction occurs

once earnings growth falls below the median, implying that employers need

to keep earnings growth at or above average in order to avoid damaging

workforce morale.

Comparisons with the firm and comparisons with salient others affect

reaction to a given pay change by determining whether an insult is felt. If

comparison pay falls, a firm could get away with a real – but not nominal

– pay cut without a step reduction in morale, even if its output is rising. If

the pay of similar other workers increases, such a pay cut would have an

insult effect if the firm’s output is rising, but not if it is falling. No matter

how badly the firm is doing, nominal cuts cause a step decline in morale,

unless other workers are also experiencing pay cuts. Thus, results provide

particular motivation for downward nominal rigidity.

CThe editors of The Scandinavian Journal of Economics 2014.