The gender gap’s persistence and globalisation
The power of observable characteristics in explaining the gender wage gap has decreased over time. More and more women get higher education and combine family life with having a career.1 Yet, there is still a gender wage differential, even in gender equal societies such as the Nordic countries. As argued by Goldin (2014), the gender wage gap persists partially because firms have an incentive to disproportionately reward individuals who work long and particular hours. Differences in pay between genders may also be due to differences in flexibility and commitment. Moreover, if women are less committed and flexible, globalisation may exacerbate the gender wage gap, because it increases the value of worker flexibility to firms.
By the virtue of being exposed to higher competition, exporters require greater commitment to work and greater flexibility of their employees. For instance, working for an exporting firm may require taking late night phone calls to communicate with customers in different time zones and international travel arranged at a short notice. The employees may be expected to be available around the clock seven days a week in case of unexpected problems arising on the production line or shipments being delayed. If commitment is not easily observable and women are perceived to be less committed workers than men, exporters will statistically discriminate against female employees and will exhibit a higher gender wage gap than non-exporters. This is likely to be true particularly among highly skilled employees.
New evidence on the gender wage gap and firms’ exports
Given this background, we have set out to investigate the link between the gender wage gap and firms’ exports (see Bøler et al. 2015). We ask two questions:
- Do exporters pay women relatively less than non-exporters?
- How did a shock to the perceived commitment of women affect the gender wage gap differential between exporters and non-exporters?
We employ a matched employer-employee data set covering close to all firms in the Norwegian manufacturing sector and their full-time employees between 1996 and 2010. We estimate a Mincerian wage regression controlling for a host of worker and firm characteristics. We find evidence of a substantial overall gender wage gap of around 24%. The gender wage gap appears to be smaller in exporting firms where it is equal to 19%.
Thus, working for an exporting firm is associated with closing almost a fifth of the observed gender wage gap. These results are in line with previous research on the relationship between trade and gender.2
This analysis, however, does not take into account unobservable worker characteristics that are likely to be correlated with the choice of working for an exporter. Interestingly, and in line with our hypothesis on gender differences in flexibility and commitment, controlling for unobservable worker characteristics (or unobservable worker-firm heterogeneity) reverses the results.
- The findings indicate that exporting firms exhibit a higher gender wage gap than non-exporters with the difference reaching about 3 percentage points.
As expected, this effect is present only among college graduates. In other words, college educated women earn higher wages at exporting firms than at non-exporters, but they are underpaid given their unobservable characteristics.
Exploiting shocks in commitment and flexibility
Next we move on to exploit an exogenous shock that is widely believed to have changed the perceived commitment gap between the genders. The shock consisted of a series of legislative changes that gradually increased the number of weeks of parental leave that are available only to the father, the so called paternal quota. A quota of 4 weeks was first introduced in 1993. It was extended to 5 and 6 weeks in 2005 and 2006, respectively. In 2009, the number of weeks was further extended to 10. This nudge on the part of the Norwegian government has resulted in a huge increase in the fraction of fathers taking a substantial period of paternity leave.
As a proxy for the change in social attitudes, and thus the perceived commitment gap between men and women, we use the fraction of fathers taking at least 8 weeks of leave in a given year. The 8 week cutoff is chosen because a 2 month long absence from work is substantial and disruptive enough to affect the employers’ perceptions. The share of fathers taking at least 8 weeks of leave went up from 8.4% in 1996 (the first year of the sample), to 10.8% in 2003, and 40.6% in 2010 (the last year of the sample).
- While the gender wage gap was found to be higher in exporting firms than in non-exporters, the difference between the two narrows down with the changes in social attitudes.
In line with our priors, the results are stronger for workers in their reproductive years (i.e., those under 45 years of age). They are also stronger for college graduates. The magnitude of the estimated effect is economically meaningful. The change in the fraction of fathers taking at least 8 weeks of paternity leave between 1996 and 2008 corresponds to the gender wage gap in exporters dropping from being on average 4.5% higher than in other firms to being only 3.3% higher.
We also exploit an alternative shock to commitment and flexibility. In 2003, the Norwegian government initiated an ambitious expansion of kindergarten coverage (see the figure below). The goal of this policy was to offer all children a high quality low price place in a public or private kindergarten. But it was also seen as a measure to increase gender equality through the provision of reliable and formal child care. The initiative resulted in a massive increase in the share of children attending kindergarten. Nine of ten children between the ages of 1 and 5 attended kindergarten in 2009, as compared to 54% in 1996, and 63% in 2001. The expansion of kindergarten coverage proceeded at different speeds in different municipalities, providing geographical variation in coverage rates.
Figure 1. Share of children (age 1-5) in kindergarten (national average)
We examine whether the gender wage gap differential between exporters and non-exporters was affected by the increase in availability of kindergarten places in a given municipality. The logic behind this approach is that greater availability of reliable child care is associated with fewer disruptions to work hours and thus allows women to exhibit greater commitment to their jobs. The results are in line with those found for the paternity leave case. The gender wage gap differential between exporters and non-exporters decreased as a larger share of children was enrolled in kindergarten. Again the observed effect is stronger for workers under 45 years of age.
In summary, our findings are consistent with exporters being able to attract more able, more ambitious or more committed females, whom they subject to a larger degree of statistical discrimination than do other firms. This may seem at odds with previous research on the relationship between trade and gender which tends to find that gender wage gaps decrease with trade liberalisation. However, none of the previous studies have been able to control for unobservable worker heterogeneity. We show that doing so is crucial and reverses the sign of the effect of trade on the gender wage gap.
Understanding how globalisation affects inequality is crucial for policymakers and researchers alike, as is explaining what lies behind the remaining gender wage gap. Our results are consistent with exporters attaching greater importance to worker commitment due to operating in more competitive and fast changing environments than other firms. Thus, globalisation appears to be exacerbating the gender inequality, at least through this particular channel.
Beaudry, P, and E Lewis (2014), "Do Male-Female Wage Differentials Reflect Differences in the Return to Skill? Cross-City Evidence from 1980-2000", American Economic Journal: Applied Economics, 6(2): 178-94.
Bøler, E A, B Javorcik, and K-H Ulltveit-Moe (2015), “Globalization: A Woman’s Best Friend? Exporters and the Gender Wage Gap”, CEPR Discussion Paper 10475.
Goldin, C, L F Katz, and I Kuziemko (2006), "The Homecoming of American College Women: The Reversal of the College Gender Gap", Journal of Economic Perspectives, 20(4): 133-156.
Goldin, C (2014.), “A Grand Gender Convergence: Its Last Chapter,” American Economic Review, 104(4):1091-1119.
Juhn, C, G Ujhelyi, and C Villegas-Sanchez (2013), "Trade Liberalization and Gender Inequality", The American Economic Review, 103(3): 269-73.
Oostendorp, R H (2009), “Globalization and the Gender Wage Gap”, World Bank Economic Review, 23 (1): 141-161.
1 See, for instance, Goldin et al. (2006) and Beaudry and Lewis (2014).
2 Juhn et al. (2013) argue that exporting firms are more technologically advanced and that the marginal productivity of women increases in the presence of new, less physically demanding, technology. They find support for their predictions using Mexican firm-level data. Oostendorp (2009) also finds evidence of smaller gender wage gaps in more open countries in a cross-country study.