First, we examine the impact of economic integration on female labor force participation in Vietnam. Of its export industries, the textile and apparel sector is the largest employer of female labor.1,2 The Multifiber Arrangement (MFA and Vietnam's accession to the WTO in 2007 are the relevant trade agreements for female labor force participation in Vietnam. following from the work of Narayan and Nguyen (2016) , which shows the importance of the economic status of trading partners in determining trade in Vietnam.3 Besides, as mentioned above, Sauré and Zoabi (2014) suggested that the impact of economic integration on female labor force participation depends partly on the trading partners.
Third, we check the significance or not of the economic integration variable in the female labor force participation model. The model includes the variables of economic integration (MFA, WTO and trade) and the traditional drivers of female labor force participation.
LITERATURE REVIEW
We note that there were minor changes in average female labor force participation during the WTO (non-MFA) and non-WTO (MFA) periods. In the MFA (non-WTO) period, average female labor force participation was 80% compared to 78% in the non-MFA (WTO) period. This reciprocal relationship is also supported by Cheng (1996a) for African American women's labor force participation. 1997) found unidirectional causality running from fertility to FLFP in the USA and Japan.
They confirmed a unidirectional causality from fertility to FLFP in the long run for women aged 25-34. In the USA, Salamaliki et al. 2013) found evidence to support a bidirectional causality between fertility and FLFP for two groups of women, those aged 25-34 and over 16. Women in good health and with a better educational background (or higher level of schooling) may have a better chance of participating in the labor market. 2016) found a strong association between FLFP and health status in Colombia.
8 While women's labor force participation may depend on worker age, number of children, or marital status, as mentioned in the previous section, we did not find consistent data on these variables for Vietnam for our study. In theory, the improvement in living standards encourages higher education, improved health conditions, better training and support systems for working women, and the replacement of work with technology for household chores, which should improve women's participation in the workplace. For India, Hasan et al. 2012) found evidence of a negative relationship between unemployment and lower tariffs in the short run, but a positive association in the long run.
Sauré and Zoabi (2014) developed an extensive theoretical model where they show that trade between a capital-poor and a capital-rich economy widens the gender gap in the capital-rich country, and under the equilibrium factor allocation it reduces total female labor. force participation in the capital-rich country.
EMPIRICAL MODEL
Rashid and Akram (2017) examined the impact of international trade competitiveness on the UK manufacturing sector, focusing on both trade liberalization and the exchange rate. The authors found that contributions to job creation and job destruction depend on firm size. Gender-specific research on the impact of trade/trade liberalization on the labor force is currently thin, likely due to a lack of consistent data.
The authors find support for their theory on the example of the USA and Mexico during the period that also coincides with the NAFTA agreement between the two countries. Here, cross-sectional differences are identified with Vietnam's trade with trading partners captured in bilateral export data (Tradeij). Based on these trading partners, four panels are developed to estimate equation (1): full sample; and three income groups.
Consistent with Narayan and Nguyen (2016), income-based panels cover Vietnam's bilateral trade from high-income (SMHI) trading partners; upper middle income countries (UMICs); and low- and middle-income countries (LMCI). All other FLFP determinants relate only to Vietnam and are therefore repeated for each cross-section (i) in the panel. In the full sample there are 54 cross-sections, while the SMHI, SMUMI and SMLMI income groups consist of 30, 16 and 9 cross-sections, respectively. 11.
Given the small sample period and large number of determinants of FLFP, only a one-year delay could be allowed in the estimates.
DATA AND PRELIMINARY ANALYSIS A. Data
To capture the dynamics of bilateral trade relations between Vietnam and each of her trading partners while providing average coefficients, we run equation (1) under a panel construction. Expenditure on education in relation to total public expenditure: Total general (local, regional and central) public expenditure on education (current, capital and transfers). In addition, public expenditure on education in relation to GDP, fertility rate and female youth without school attendance as a percentage of females in secondary school was also stronger in the 1990s than in the period 2000-2014.
In contrast, female life expectancy at birth and the standard of living were stronger in the 2000-2014 period compared to the 1990s. It is followed by upper middle income (SMUMI) and low and lower middle income (SMLMI) trading partners.13. As part of our preliminary analysis, we performed tests to examine the stationarity of the variables.
In light of the literature, we conducted pairwise Granger causality tests to examine whether each of the factors identified above Granger causes FLFP in Vietnam and vice versa. Consistent with some evidence in the literature, our results suggest that in Vietnam there is a bidirectional relationship between FLFP and fertility, education, health, and income growth (see Section III). 13 For statistical analyzes and discussion of trends and patterns in relation to Vietnam's exports of goods with these 54 trading partners, see Narayan and Nguyen (2016).
At the same time, the availability of relatively cheap female labor in Vietnam, which is one of the drivers of FDI in export industries, is driving exports (see discussion in Part II).
EMPIRICAL RESULTS
The explanatory variables are the health of the female labor force in Vietnam, measured by female life expectancy at birth (FLE);. So far in the discussion we have observed that exports and the agreement of the MFA have positively influenced the participation of women in the workforce, while the WTO has negatively influenced the participation of women in the workforce. First, the effect of economic integration variables has been lower in magnitude than all the traditional determinants of female labor force participation, with the exception of government spending on education.
At the same time, economic integration variables moderated the impact on female labor force participation of other traditional factors, namely fertility rate (FERTI) and female life expectancy (FLE). This is determined by: the health of Vietnam's female workforce, as measured by female life expectancy at birth (FLE); education and skills of the female labor force in Vietnam, measured as adolescent girls who have not attended school (as the ratio of women in lower secondary school) (FEOUT); government expenditure on education as a ratio of total government expenditure in Vietnam (EDUEXP); Vietnam's total fertility rate (births per woman) (FERTI); and Vietnam's GDP per capita (current US dollar) (GDPP) captures the level of development or standard of living. First, contrary to general theory on female labor force participation, we find that lower fertility rates discouraged FLFP.
More recent studies also tend to find a positive relationship between the fertility rate and female labor force participation (see Kinoshita and Guo (2015) for Japan and Da-Rocha and Fuster (2006) for OECD countries). Da-Rocha and Fuster (2006) emphasized labor market frictions to explain the positive relationship between fertility rates and female labor force participation in OECD countries. The association between fertility and female labor force participation was only slightly weakened when the economic integration variables were introduced.
Fifth, increased government spending on education (EDUEXP) in Vietnam has increased women's labor force participation, and the role of government-funded education programs in this labor force participation probably became more relevant after economic integration.
CONCLUSION
Thus, although economic integration – through the use of technology – has helped to reduce some pressure on the health of workers, leading to a reduced correlation between health and female labor force participation, health, represented by life expectancy, is still the most important determinant of female labor force participation in Vietnam. Second, developments during the MFA period, namely the EU preferential quota, the lifting of the US embargo, and improved trade relations between the US and Vietnam through a reduction in tariffs faced by Vietnamese exports to the US, have increased female labor (average) improves participation in Vietnam. Third, despite the growth in exports from Vietnam since their WTO membership, the WTO period had a negative impact on female labor force participation.
The significant trade-female labor force participation relationship portrayed in this paper implies that trade policymakers should not discount their ability to influence women's economic well-being. The characteristics of Vietnam's female labor force are more similar to female labor forces in low- and middle-income groups than to those of high-income countries, so our results may be depicting that Vietnam has a greater comparative advantage. strong in the employment of women for export in those with high incomes. trading partners than in the case of middle or low income trading partners. The negative impact of the WTO (or non-MFA period) on the FLFP suggests that Vietnam had lost some of its comparative advantage in what the female labor force produced, especially in terms of textiles and clothing.
An examination of the traditional determinants of female labor force participation concluded that health status not only has a positive effect, it is also an important determinant, more important than economic integration variables or the other variables examined here. We also found evidence that economic integration variables likely played a role in enhancing the labor force impact of out-of-school female adolescents, and that government spending on education affected female labor force participation, both of which imply that economic integration increases employment opportunities for women in Vietnam in the studied period. The U-Shaped Female Labor Force Function in Economic Development and Economic History (NBER Working Paper No. 4707).
On the relationship between female labor force participation and fertility in G7 countries: Evidence from panel cointegration and Granger causality. Female labor force participation and total fertility rates in the OECD: New evidence from panel cointegration and Granger causality testing. Female labor force participation, fertility and infant mortality in Australia: Some empirical evidence from Granger causality tests.
