Stock markets and female participation in the labor force

https://doi.org/10.1016/j.intfin.2021.101297Get rights and content

Highlights

  • We study the relation between female participation in the labor force and stock markets.

  • We focus on the first teaching day of the year.

  • This day has lower trading volume, on average.

  • Female workforce participation is negatively associated with the trading volume.

  • If nondiscrimination laws that govern hiring exist, this negative effect is even more pronounced.

  • We document a significant societal phenomenon with a profound impact on stock markets.

Abstract

This paper examines the relation between female participation in the labor force and stock market trading volume. We focus on the first teaching day of the year at kindergartens and primary schools, treating it as an event with characteristics that resemble an exogenous event, and verify this using a web-based survey. We find that the first day of school has lower trading volume than the average daily trading volume during the rest of the year. Moreover, we document that female workforce participation is negatively associated with the trading volume on this day and that, in societies that have legislated gender nondiscrimination laws that govern hiring, this negative effect is even more pronounced. Our main findings document a significant societal phenomenon with a profound effect on stock markets, and specifically on the quality of the information environment, and therefore merit further attention.

Introduction

The increase in female participation in the labor force has been one of the most significant developments in the global economy over the last few decades (Eckstein and Lifshitz, 2011, OECD, 2012). While this increase has been neither linear nor uniform across countries and depends on the social norms and traditions of each society (OECD, 2014), its effects across the globe have attracted interest from academics, politicians, policymakers, and social organizations. The literature has documented both economic and non-economic activities that have been affected by the change in female labor force participation (e.g., Hausman, 1979, Tam, 2011). Rogers and Youssef (1988) even present a relationship between female involvement in economic activities and improvement in child nutrition and health. In this paper, we focus on an unexplored influence of female labor force participation—its effect on the stock markets. Our findings indicate that changes in female labor force participation have a significant effect on capital markets, both cross-country and over time.

The motivation for this research is rooted in the growing interest in the association between social considerations and stock markets. This interest is amply documented in the literature (see Section 2). Prominent examples include Meir et al. (2016), who find that social factors influence individuals’ decisions to participate in the stock market, and Ke (2018), who finds that traditional gender norms are negatively associated with direct household stock market participation. These findings prompt us to explore whether changes in social norms affect additional attributes of the stock market.

In this paper, we conjecture that a change in female participation in the labor force is a result of a change in social norms. In other words, we assume that a change in the traditional duties of males and females within the nuclear family—which is a result of a change in the social norms of the society—will lead to an increase in the female labor force participation (for simplicity, we refer to a society with higher gender equality as a society with “modern” norms). It should be noted that the outcome of the change in gender-related social norms is reflected in the higher participation of both parents in the economic and non-economic activities of the nuclear family, as both are performing the same duties interchangeably (relative to a more traditional family in which each parent has her/his own duties). That is, a change in the level of female labor force participation implies that, relative to traditional social norms, females (males) are less (more) engaged in households’ activities and more (less) engaged in outside employment activities.

Based on this assumption, we hypothesize that a higher gender equality—which is demonstrated in a higher participation of females in the labor force—will cause higher sensitivity of financial markets to cultural- and social-related events and activities (for example, large gatherings of families, such as open houses, festivals, and competitions). That is, in a traditional family, events that relate to the family will affect one of the parents, while, in a modern family, the same event will affect both parents because the dichotomy between the duties of male and females in the latter case is less clear. This will result in a greater effect on capital markets relative to societies in which the traditional male and female duties are preserved. The implication of this hypothesis is that social factors should be considered when studying capital markets.

To examine this research question, we focus on an event that, on average, affects females (both genders) in a society with traditional (modern) social norms: the first teaching day in kindergartens and primary schools (hereafter: “first day of school”). We view the first day of school as a proxy for an exogenous event in which changes in female participation in the labor force are reflected within and across economies. Specifically, higher female participation will increase the involvement and participation of both parents in the new beginning for their children; therefore, trading activity in capital markets is expected to decrease. Hence, we expect to find a negative correlation between female participation in the labor force and trading volume on the first day of school, on average.

To provide observational support for the assumption regarding the participation of both parents in activities on the first day of school, we conduct a cross-country survey. The survey is conducted on Monday, September 2, 2019: the first day of the 2019 school year in most countries in the northern hemisphere. In most countries, the first day of school falls somewhere in early September (in the northern hemisphere) or early March (in the southern hemisphere), although there is considerable national and regional variation. The survey includes 607 anonymous participants from 32 different nations. Overall, more than two-thirds of the participants report that they take some time off work on the first day of the school year. The participants also indicate that 87.7% of their close circle accompany their children to school on that day. There are no gender differences across the participants’ answers: 67.33% (202 of 300) of the female participants accompany their children on the first day of school, compared with 67.75% of males (208 of 307 participants). Overall, the survey’s findings are consistent with our assumption that both parents are involved in the activities occurring on the first day of school.

To examine the relation between female participation in the labor force and trading activity on the first day of school, we employ a dataset of trading volumes and labor statistics of 74 countries. The sample period spans 1967 to 2018.1 In the analysis, we normalize the trading volume into two measures in a manner that, regardless of the currency and number of trading days, we are able to compare results across countries (see Section 3 for details).

First, we examine the relation between trading volume on the first day of school compared with the other trading days of the year. As per our hypothesis, we expect to find lower trading volume on the first day of school compared with the rest of the year, because, while it is a regular working day, its importance to family members distracts market participants. Indeed, our analysis shows that, after controlling for country and year fixed effects, trading volume on the first day of school is significantly lower than on the other trading days of the year, on average. This effect disappears in the trading days that follow the first day of school. Moreover, the economic magnitude of this decrease is meaningful: the results demonstrate an average decrease of approximately 12% in trading volume on the first day of school relative to other trading days.

Second, we examine the relation between female participation in the labor force and trading volume on the first day of school, and find a significant negative association This analysis is conducted using annual data, as macro data are measured and reported annually (see Section 3 for details).2 We find that higher female participation in the labor force in a given society causes lower trading volume on the local stock market on the first day of school. This relation holds for both national and international modeled estimates of female participation rates. To examine the robustness of this negative relation, we control for economic variables that may affect the results, such as gross domestic product (GDP) per capita,3 the proportion of females with account ownership at a financial institution (or with a mobile money service provider), and the gender ratio at birth, which previous literature suggests is negatively correlated with female labor force participation (Angrist, 2002).4 Consistent with our hypothesis, the negative correlation between trading volume and female labor force participation on the first day of school is preserved.

Third, we consider an additional aspect of a change in social norms that affects female integration in the labor force—laws that mandate gender nondiscrimination in hiring. Laws enacted by legislative bodies reflect social considerations accepted by the majority of the population. Therefore, the existence of gender nondiscrimination laws indicates a profound shift in social norms compared with “old-style” values. Note, however, that this assumption is stated with caution, as the causality in this case is less direct than the causality we attribute to female workforce participation, as it is based on the notion that politicians should act according to the agenda of their senders. Nevertheless, we employ a dummy variable that indicates the existence of such nondiscrimination law in society and include it in the analysis. Similar to the effect of the female participation rate, we find a significant and negative association between trading volume and the existence of a gender nondiscrimination law. More importantly, the negative effect of the gender nondiscrimination law is above and beyond the effect of female workforce participation. That is, in countries with legislated gender nondiscrimination laws, a higher female participation rate further decreases trading volume on the first day of school.

Finally, we conduct a battery of tests to ascertain the robustness of our findings. First, we control for the possibility that the first day of school is also the first trading day of the week. The literature documents that the first trading day of the week—known as the Monday effect—exhibits a lower trading volume, on average (e.g., Fishe et al., 1993, Lakonishok and Maberly, 1990). Therefore, our findings may (partially) arise from this effect. However, estimation results show otherwise. We begin by confirming that, while the first trading day of the week exhibits lower trading volume, it does not harm the finding that trading days on the first day of school have even lower trading volume. We then verify that, even after controlling for the first trading day of the week, there is a negative and significant relation between both female workforce participation and the existence of nondiscrimination legislation and trading volume during the first day of school. Second, we test the dependence of our findings to geographical location by splitting the sample into the southern and northern hemispheres (in these regions, the first day of school begins in different months). We find that the results still hold in both regions: the first day of school exhibits lower trading volume, and this trading volume is negatively related to female participation in the labor force. Finally, we estimate the effect of female workforce participation on trading volume on the first day of school in each country separately and find that 72% of the countries in our sample have a negative relation between these variables.

As detailed above, we state that the first day of school serves as an exogenous event to examine the relation between social norms and financial markets. While the first day of school is predicted, families rarely outsource their participation in the activities of this day. To rephrase this in financial terms, while the date and the activity of this event are anticipated, the exposure to this activity cannot be mitigated. Moreover, the results of the web-based survey detailed above also support the notion that parents participate in the activities of the first day of school. Such behavior enables us to examine the link between female labor force participation and financial markets both across countries and across time. We should mention, however, that the use of the first day of school has its drawbacks. First, the importance of the first day of school varies across societies. Second, the first day of school is a significant event in households with small children, yet there are many households without small children. Therefore, an insignificant relation between trading volume on the first day of school and female labor force participation does not necessarily imply that the examined relation does not exist. Therefore, our results can be interpreted as a lower bound of the effect of female workforce participation on capital markets.

Our findings contribute to the growing body of evidence that social factors affect capital markets. Traditional asset-pricing models focus on the characteristics of stocks and firms, such as volatility and firm size. We provide additional evidence supporting the notion that factors representing the real economy cannot be ignored when examining a society’s capital market, even if such factors are not related to the particulars of a given firm. Moreover, it should be mentioned that trading volume is used as an important indicator of capital markets, as it provides insight into information disclosure (Beaver, 1968), investment decisions, and investor disagreement (Bamber et al., 2011; see Section 2 for details), and may also have implications for a firm’s cost of capital (Barron et al., 2005, Verrecchia, 2001).

The remainder of this study is organized as follows. Section 2 reviews the related literature. Section 3 presents the research design and data. Section 4 presents our primary results, and Section 5 tests their robustness. Section 6 presents the results of the web-based survey, and Section 7 concludes.

Section snippets

Related literature and research question

There is growing evidence that social factors influence capital markets. One prominent example is stock market participation, which relates to trading volume, as higher market participation implies a higher trading volume. In this respect, Ke (2018) finds that the prevailing gender norms of a society are an important cultural factor. In a cross-country analysis, he finds that traditional gender norms are negatively associated with direct household stock market participation. Further, Ke (2020)

Data and methodology

We collect data on trading volumes at 79 stock exchanges, as well as statistics on the labor force in 74 countries. Our sample period starts in 1967 and ends in 2018. Trading volume data are obtained from Datastream: we collect the monetary trading volume of each exchange in the currency in which it trades. Statistics on labor are obtained from the World Bank and include annual data on female labor force participation from two sources: the World Bank’s estimates and data reported by the

Results

Panel A of Table 1 presents descriptive statistics for the main variables of interest. The first two rows in Panel A report the trading volume measures used in the analysis (as presented in Section 3). Given that the volume distributions contain extreme observations, this variable is winsorized at 5% (i.e., 2.5% on each side). The winsorized variable is reported in Panel A and used in our analysis. As a robustness check, we also winsorize trading volume at 2% (1% on each side) and at 10% (5% on

Robustness tests

In more than half of the observations, the school year starts at the beginning of the trading week. Therefore, as a robustness test, we add a control variable for the first trading day of the week and perform an additional analysis. Formally, Eq. (5) is an extension of Eq. (2), and appears as follows:RDVi,t=α0+β1DSi,FDS+β2DSi,FDS-1+d=36βdDSi,FDS+d-2+β7FDWi,t+δiDcountryi+θtDyeart+εi,tFDWi,t is a dummy variable that equals 1 if the trading day is the first trading day of the week, and 0

Qualitative evidence from web survey

Our main hypothesis relies on the assumption that a change in the female workforce participation rate reflects a change in a country’s societal norms. The traditional duties of males and females in a nuclear family are evolving, and this would affect financial markets. Moreover, we assume that this effect should be reflected in the economic indicators recorded on the first day of school, with both parents taking part in activities related to the school day.

To provide empirical support for our

Conclusion

This paper explores whether social norms affect capital markets. The specific social norm we examine is the level of gender equality among males and females in a given society. As a proxy for this norm, we choose to refer to female participation in the labor force. We assume that higher (lower) female participation in the workforce implies a larger (smaller) change in the traditional duties of male and females in a nuclear family. An additional proxy for such a change is the presence of

CRediT authorship contribution statement

Menachem (Meni) Abudy: Writing - original draft, Validation. Yevgeny Mugerman: Data curation, Conceptualization, Methodology, Writing - review & editing. Zvi Wiener: Supervision, Funding acquisition.

References (36)

  • J. Angrist

    How do sex ratios affect marriage and labor markets? Evidence from America’s second generation

    Quart. J. Econ.

    (2002)
  • L.S. Bamber et al.

    Trading volume around earnings announcements and other financial reports: theory, research design, empirical evidence, and directions for future research

    Contemp. Account. Res.

    (2011)
  • O. Barron et al.

    Evidence that investors trade on private event-period information around earnings announcements

    Account. Rev.

    (2005)
  • W. Beaver

    The information content of annual earnings announcements

    J. Account. Res.

    (1968)
  • Be’eri, O., Kedar-Levy, H., & Amar, M. (2019). The effect of gender differences on the emergence of the disposition...
  • A. Bernasek et al.

    Predictors of women’s involvement in household financial decision-making

    J. Assoc. Financ. Counsel. Plan. Educat.

    (2002)
  • J.R. Brown et al.

    Neighbors matter: causal community effects and stock market participation

    J. Finance

    (2008)
  • F.K. Changwony et al.

    Social engagement and stock market participation

    Rev. Financ.

    (2015)
  • Cited by (7)

    • National pride, investor sentiment, and stock markets

      2023, Journal of International Financial Markets, Institutions and Money
    • The role of education in capital Markets’ liquidity

      2023, Journal of International Financial Markets, Institutions and Money
    • Can gender Pay-Gap disclosures make a difference?

      2023, Finance Research Letters
      Citation Excerpt :

      Female labour force participation contributes to the economy in other aspects. For example, Abudy et al. (2021) show that female participation in the labour force can contribute to market liquidity, a key factor in facilitating and promoting economic growth via capital markets. Tsani et al. (2013) show that higher female labour force participation rates have a positive impact on growth.

    • Do computerized traders follow social norms? Evidence from the holocaust remembrance moment of silence

      2022, Finance Research Letters
      Citation Excerpt :

      First, computerized traders (CTs) minimize false or incorrect decisions that stem from irrational and biased human behavior. Common biases such as loss aversion (Odean, 1998), overconfidence (Odean, 1999), reluctance to realize losses (Grinblatt and Keloharju, 2001), various types of investor sentiment (Mugerman et al., 2020; Abudy et al., 2022), limited attention spans, information overload (Mugerman et al., 2022), and even social considerations (Abudy et al., 2021) are not expected to affect the decision-making processes of automated machines. Second, CTs offer improved asset pricing (D'Acunto et al., 2019; Chaboud et al., 2014).

    • The economic consequences of violence against civilians: Developing economic resilience to violence

      2022, Journal of International Financial Markets, Institutions and Money
      Citation Excerpt :

      According to Zemishlany (2012), “…a preventive approach towards terror-exposed communities should focus on helping the vulnerable groups such as disadvantaged city populations who lack economic and social support” (p. 308). Indeed, contemporary economics and finance studies, including ours, highlight the inter-relations between societal phenomena and economic fundamentals (e.g., GDP growth and stock market returns; see, for example, Abudy et al., 2021). Economic theory suggests that violence against civilians manifests itself in the economic decisions of individuals and governments.

    View all citing articles on Scopus
    View full text