Abstract
This paper contributes to the literature on the drivers of export diversification by tackling an issue that has received scant attention in this literature, that is, the effect of women empowerment on export product diversification in developing countries. The analysis focuses on a sample of 106 developing countries covering the period 2002–2018. The results suggest that women’s political empowerment and gender parity in secondary school enrolment have a positive effect on export diversification. On the contrary, gender vulnerable employment inequality and gender wage inequality have a negative effect on diversification. These findings suggest that policies targeting SDG 5 – gender parity and women empowerment – can potentially improve trade performance.
Disclosure statement
No potential conflict of interest was reported by the author(s).
Notes
1 Empowering women in the economy and closing gender gaps in the world of work are key to achieving the 2030 Agenda for Sustainable Development (SDG 5). Women’s empowerment is defined, according to Kabeer (Citation1999), as the process through which women acquire the ability to make strategic life choices in a context where this ability was previously denied to them (Kabeer Citation1999). Recently, Duflo (Citation2012) defined women’s empowerment as a process through which the ability of women to access the constituents of development in particular health, education, earning opportunities, rights and political participation is improved. When turning to the political area, Sundström et al. (Citation2017) define women’s political empowerment as a process of increasing women’s capacity, through greater choice, agencies and participation, in societal decision-making.
2 In this article, we use women’s empowerment and gender equality interchangeably, assuming that by empowering women, people compete on a level playing field.
3 See Stanistreet et al. (Citation2007) for more details.
4 This point has been made notably by Aghion et al. (Citation2010).
5 On the link between wage inequality and lower effort, see Akerlof and Yellen (Citation1990) and Levine (Citation1991).
6 Since the seminal work of Melitz (Citation2003), recent trade literature has shown that exporters are usually the more productive than non-exporters. Thus by lowering work efforts and the overall productivity, gender wage inequality can indeed reduce export diversification.
7 As explained in section 2, vulnerable employment is characterized by low earnings, low productivity and difficult working conditions. The low earnings - low productivity nexus discussed above explains why this may lead to fewer goods being exported.
8 Data are taken from World Bank Business Law (WBL)
9 Relevant means that the instruments are strongly related to the endogenous variables conditional on the other exogenous variables. Exogenous means that the instruments themselves must not be correlated with the error term. The instrument also needs to be redundant.
10 We use codes from Osakwe and Kilolo (Citation2018) to compute several measures of export diversification based on the BACI2002 database: HHI, Theil and the number of export products.
11 CEPII stands for Centre d'Études Prospectives et d'Informations Internationales; It compiles Trade data avalable online at http://www.cepii.fr/CEPII/en/bdd_modele/bdd_modele.asp.
12 This index is also referred to in the literature as the Herfindahl index.
13 Persson and Wilhelmsson (Citation2016) compute a diversification index, IHHI, that correspond to the inverse of HHI (IHHI = 1/HHI).
14 Centre d’Etudes Prospectives et d’Informations Internationales.
15 BACI is the World trade database built by the CEPII from the UN Comtrade database. It is developed through an original procedure that reconciles the declarations of the exporter and the importer.
16 As shown in Table A.3, missing values concern 3 variables: gpi, av year schooling and xr.
17 Persson and Wilhelmsson (Citation2016) note a caveat in using concentration indices since price changes between consecutive periods may induce variation when trace volumes do not change.
18 Using GDP-weighted mean.
19 See https://ilostat.ilo.org/resources/concepts-and-definitions/description-employment-by-status/ accessed on 05/07/2022
20 We used WPEI rather than the other indexes of women’s political empowerment. This V-Dem WPEI are far more extensive both spatially and temporarily (Sundström et al. Citation2017). The aggregated WPEI and its three sub-indexes ranges from 0–1 (high empowerment).
21 This term captures the curvature of GDP per capita. This variable captures the threshold effect.
22 Data are taken from World Bank Business Law (WBL)
23 Relevant means that the instruments are strongly related to the endogenous variables conditional on the other exogenous variables. Exogenous means that the instruments themselves must not be correlated with the error term. The instrument also needs to be redundant.