Minimum Wages and the Gender GAP in Pay: New Evidence from the United Kingdom and Ireland

• Published date: 01 September 2019
• Author: Olivier Bargain,Karina Doorley,Philippe Van Kerm
• Date: 01 September 2019
• DOI: http://doi.org/10.1111/roiw.12384

© 2018 Internationa l Association for Res earch in Incom e and Wealth

514

MINIMUM WAGES AND THE GENDER GAP IN PAY: NEW EVIDENCE

FROM THE UNITED KINGDOM AND IRELAND

by Olivier bargain

University of Bo rdeaux, Institut U niversitaire de Fran ce, and Institute of L abor Economics

Karina DOOrley*

Economic a nd Social Resea rch Institute,Institu te of Labor Economi cs, and Trinity College D ublin

anD

PhiliPPe van Kerm

University of Lu xembourg and Lu xembourg Institut e of Socio-Econo mic Research

Women are disproportionately in low-paid work compared to men so, in the absence of rationing effects

on their employment, they should benefit the most from minimum wage policies. This study examines

the change in the gender wage gap around the introduction of minimum wages in Ireland and the

United Kingdom (U.K.). Using survey data for the two countries, we develop a decomposition of the

change in the gender differences in wage distributions around the date of introduction of minimum

wages. We separate out “price” effects attributed to minimum wages from “employment composition”

effects. A significant reduction of the gender gap at low wages is observed after the introduction of the

minimum wage in Ireland, while there is hardly any change in the U.K. Counterfactual simulations

show that the difference between countries may be attributed to gender differences in non-compliance

with the minimum wage legislation in the U.K.

JEL Codes: C14, I2, J16

Keywords: gender wage gap, minimum wage, distribution regression

1. intrODuctiOn

Recent research into the gender wage gap has increasingly focused on more

global methods than the evaluation of gender wage differences at the mean. Gender

gaps are often concentrated either at the bottom of the distribution (“sticky floors”)

or at the top (“glass ceilings”). This literature has benefited from the surge of meth-

ods extending Oaxaca–Blinder type decompositions to the whole wage distribution

(see the surveys in Melly, 2006, Fortin et al., 2011, Chernozhukov et al., 2013).

Most directly relevant for policymakers, distributional analyses provide some

insights into the intended or unintended effects of labor market policies on wage

Note: Doorley is grateful to the Fonds national de la Recherche Luxembourg (grant number

BFR08/041) for financial support.

*Correspondence to: Karina Doorley, The Economic and Social Research Institute, Whitaker

Square, Sir John Rogerson’s Quay, Dublin 2, Ireland (karina.doorley@esri.ie).

Review of Inc ome and Wealth

Series 65, Numb er 3, September 2019

DOI : 10.1111 /roi w.123 84

bs_bs_banner

Review of Income and Wealth, Series 65, Number 3, September 2019

515

© 2018 Internationa l Association for Res earch in Incom e and Wealth

inequality and, in particular, gender wage gaps. This is particularly the case for

policies such as the national minimum wage (henceforth NMW), which, by design,

affect workers at different positions of the wage distribution differently. NMW

policies tend to compress the bottom of the wage distribution, where women are

disproportionately represented. As a result, women should benefit the most from

NMWs, at least in the absence of changes to their employment status. A (possibly

unintended) consequence of the NMW is therefore a reduction of the gender wage

gap.

The testing of this prediction is usually complicated. At the macro level, it is

difficult to control for all sources of cross-country differences beyond wage dis-

tributions and NMW policies. A successful attempt to do so is Blau and Kahn

(2003), who check for a negative correlation between the gender gap and the “bite”

of NMWs (the NMW level as a proportion of the average wage). For Ireland,

McGuinness et al. (2008) use the proportion of NMW workers in a firm to iden-

tify the wage disadvantage to men and women who are employed in low-paying

firms. With micro data, time variations in NMWs are often too small to provide

detectable effects. Studies close to ours have used changes in NMW legislation in

the United States (U.S.) (Blau and Kahn, 1997), in Ukraine (Ganguli and Terrell,

2006, 2009), and in Indonesia (Hallward-Driemeier et al., 2017) to check how gen-

der gaps vary with NMW levels. In this study, we examine an even more radical

policy event, namely the introduction of NMW legislation.

We focus on the introduction of a NMW in the United Kingdom (U.K.)

in 1999 and in Ireland in 2000. Using the Living in Ireland survey (LII) and the

British Household Panel Survey (BHPS), we employ a flexible model of wage

distributions to construct counterfactual distributions of wages based on a fixed

distribution of covariates for women in each country. We estimate gender differ-

ences in wage distributions before and after the introduction of the NMW, sepa-

rating out workers’ characteristics (“explained/composition”) effects from residual

(“unexplained/discriminatory”) differentials. We can thus show how the gender

wage gap at the bottom of the distribution evolved after the introduction of the

NMW in each country, as well as measure possible “spillover” effects further up

in the distribution. It is noteworthy that we focus on two neighboring countries

that share a common past history, with highly centralized systems of collective

wage bargaining and a similar high level of “sticky floor” before the policy reform.

Beyond these common initial conditions, the almost simultaneous introduction of

a NMW in Ireland and the U.K. allows us to assess how much the impact may dif-

fer according to the level at which minimum wages are set (the “bite” of the NMW)

and to the degree of compliance.

Our results are as follows. A large reduction in the gender wage gap at the

bottom of the distribution is found after the introduction of the NMW in Ireland,

while there is hardly any change in the U.K. We perform several robustness checks

that include holding employment composition constant using panel data, detrend-

ing the effects (a triple difference approach), checking the sensitivity of our results

to the inclusion of occupation and industry variables, changing the reference group

and accounting for selection into work. Our conclusions are stable. To explain

the contrasted results between Ireland and the U.K., we suggest an extrapolation

exercise that examines the counterfactual effect of introducing the same NMW