Oil, Laws, and female labor force participation.
Keywords Female labor force participation * Oil and gas rents * Islamic family law * Middle East and North Africa
JEL Classification J71 * J21 * P48 * Q39
Women in the Middle East and North Africa (MENA) have had significant achievements over the past three decades. As shown in Table 1, education for girls has increased continuously in such a way that in some countries of the region, the average years of schooling for women ages 20 to 30 between 1970 and 2010 rose by more than eight years. For example, in Saudi Arabia, one of the most conservative societies in the region, average years of schooling increased by 9.1 years. As a result, in some countries, more than half of the college students are now women. At the same time, as depicted in Table 2, fertility rates have fallen. For example, in Iran, they have fallen universally in one of the largest and fastest transitions, from about 6.5 in 1984 to almost two children per woman in 2002.
Despite these achievements, female labor force participation (FLFP) rates are low and have remained at low levels for two decades. Table 3 reports FLFP rates in 1990 and 2010 for all regions of the world. Among them, MENA has the lowest rates by a large margin. It has remained a puzzle that despite improvements in education and fertility, these low rates persisted. (1)
There are many competing hypotheses that try to explain this puzzle. Discrimination against women on the supply as well as the demand side of the labor market is among the prominent ones. Traditional institutions, including Islam, are considered to play a major role (for example, World Bank 2004), though it has been difficult to provide scientific evidence for it. Using cross-country regressions, however, Ross (2008) argues that it is oil and gas rents that lower FLFP rates. He also shows that, controlling for country level time invariant factors, oil and gas rents have a significant and negative relationship with FLFP rates. But, Groh and Rothschild (2012) challenge these results and argue that Islam and geography may play a role. Nevertheless, they do not identify how.
This paper complements and advances the arguments in Groh and Rothschild (2012) and Ross (2008, 2012). It shows that the impact of oil and gas rents on FLFP increases when discriminatory labor laws are present. Thereby, it adds an additional mechanism to the impact of rents. It argues that by raising the ratio of male to female wages, oil and gas rents increase bargaining power of men inside the household which means they would have more say in household decisions, from how budget should be spent to who can participate in the labor market. But, this increase in the male bargaining power is more likely to affect female labor supply decisions when labor market institutions (for example, laws) make it acceptable for men to make decisions about female work in the household. On the contrary, when labor market institutions are egalitarian, rents which still increase male bargaining power in the household are less likely to affect female labor supply. The higher bargaining power, in this case, may affect consumption or investment decisions instead. Therefore, when labor market institutions, such as laws, make it "legitimate" and less costly for men to veto women's decision to work, it is more likely that the rise in the male bargaining power attributed to rents, reduces FLFP. For reasons explained in "The Impact of Oil and Gas Rents" section, Islamic family laws create discriminatory labor market institutions, although they were progressive at the time of their introduction. For example, these laws require women who want to work to ask for permission from their male guardian (husband, father, or eldest son) in advance. In effect, male guardians have veto power over women's working decisions. Controlling for unobserved heterogeneity across countries, this study illustrates that oil and gas rents reduce FLFP more where Islamic family laws are enforced.
The results for the Islamic family law are not only robust to specification and econometric modelling, but also to controlling for the joint impact of rents and all of the time-constant factors that are common across the MENA region, such as culture, social norms, history, and political and economic institutions. This means that even if we control for the joint impact of rent and all common regional unobservable factors that could be correlated with Islamic family law, we still get a negative and statistically significant joint impact of rents and these laws on FLFP. In other words, we can identify the effect of Islamic family law free of any potential bias.
To further strengthen the results, this study employs a strong predictor of gender discrimination in place of Islamic family law: historical plough use. This measure is introduced in the influential study of Alesina et al. (2013). It is used, because, as discussed in "The Impact of Oil and Gas Rents" section, the origin of gender roles is attributed to the introduction of the plough in agriculture that increased male productivity on the field and led to the allocation of male labor outside the home and female labor inside. But, this study finds no impact of this measure and rents on FLFP. In other words, the results for the Islamic family law cannot be replicated with historical plough use. The underlying reason for this result is discussed.
The paper does not argue that it is Islam that creates such an effect. It shows empirically that it is only the family law based on Sharia (a very specific aspect of Islam) that derives the result. Oil and gas rents have no impact in Muslim countries that do not enforce the Islamic family laws. Moreover, Islam is not a unique entity and one may argue that there are as many interpretations of it as there are Muslims in the world. But the Islamic family laws, particularly those pertaining to women's work, are consistent across various schools of law and would apply to all Muslims regardless of their personal interpretations or religiosity. This study is a step towards explaining the puzzle of female labor supply in the MENA.
The Impact of Oil and Gas Rents
Oil and gas resources create large windfalls of rents (profits), since their extraction costs are considerably smaller than their world prices. These rents may affect the economy and, some argue, the politics of countries rich in oil and gas, in both positive and negative ways. (2) This study only considers the impact of these rents on FLFP rates.
The economic theory explains that the decision to participate in the labor force for female (and male) members of a household depends on four main economic factors: 1) female member's wages, 2) male members' wages, 3) other prices in the economy, including the opportunity costs, and 4) non-labor income which contains direct transfers to households as well as returns to assets. Figure 1 depicts mechanisms by which oil and gas rents affect women's decisions to work. Each of these mechanisms changes one of the above factors: (3)
1. One mechanism by which rent may reduce FLFP is through affecting female wages. It is shown by the top branch in Fig. 1. In developing countries, two of the major sectors for female employment are agriculture and export-oriented manufacturing. (4) Manufacturing is a tradable sector and because of the Dutch disease, (5) it usually becomes weaker in economies that are dominated by natural resource rent. Hence, the demand of the tradable sector for female labor in such economies decreases.(6) This means fewer opportunities and lower wages for women. Lower wages would induce women to participate less only if the substitution effect is larger than the income effect of lower wages, which is usually the case according to empirical evidence.
2. Dutch disease enlarges the non-tradable sector, which is male-dominated. (7) This means that there will be more opportunities for men which, in turn, increase their wages. Theoretically, higher wages for male household members reduce female members' tendency to participate in the labor force. (8)
3. Another mechanism by which these rents affect FLFP is through direct transfers. When oil and gas rent is a major source of revenue for the government, the state may be tempted to transfer part of it to households for various reasons, such as gaining legitimacy or popularity. In these economies, the state can transfer income to households without dealing with the political consequences of redistribution from the rich to the poor. (9) The economic theory predicts that labor force participation (LFP) tends to fall if non-labor income, such as government transfers, increases. Since government transfers rise as oil and gas rents increase, one can conclude that larger rents would reduce LFP, especially for women. This mechanism is shown at the mid-bottom section of Fig. 1.
4. In addition to larger transfers, the influx of oil and gas rent into government coffers increases the state's tendency to make large investments in social welfare - from free education and health care programs to highly subsidized energy and food staples (bottom branch of Fig. 1). These investments reduce prices households face, that, in turn, have two impacts on hours worked: one is the substitution effect by which people substitute leisure with consumption of goods that are now cheaper, and the second one is the income effect which encourages people to consume more leisure as their purchasing power (real income) increases. People work less if the income effect dominates the substitution effect.
This paper, however, argues that, in addition to these, oil and gas rents may affect FLFP through another mechanism: by strengthening the discriminatory institutions inside the household wherever they exist. This mechanism is drawn with light gray color in Fig. 1. Imagine two countries: one which has institutions, such as laws, that give male members of the household the privilege to make or overrule decisions about female labor supply and another in which women have freedom to make their own labor supply decisions. By increasing the ratio of male to female wages, rents increase bargaining power of men inside the household in both of these countries. But, in the former, it is more likely that this higher bargaining power translates to lower FLFP. This is because, in the first country, it is legitimate and less costly for male members to exercise their higher bargaining power to limit FLFP. But, in the second country, it is not acceptable to use this higher bargaining power to affect female labor supply decisions (instead, the higher bargaining power may affect other household decisions, such as consumption and investment, i.e. how household budget should be spent.) Therefore, the negative impact of oil and gas rents on FLFP can be larger in countries where men have such privileges over women in the household. This is the hypothesis of this study and is evaluated empirically.
One institution that gives specific privileges to men with respect to female labor supply is the Islamic family law. Islam is a legalistic religion which sets laws for private as well as public spheres of individuals, breaching of which could be considered illegal. (10) For instance, unlike most religions, there are specific rules for doing business, making contracts, and forming partnerships (Kuran 2010). It also has a relatively extensive family law that lays out the rules pertaining to the rights, responsibilities, and roles of spouses in their families and the society. For example, Islamic law, at the time of its introduction, 14 centuries ago, gave women mandatory inheritance shares that were absent in many schools of law at the time. It gave them the right to own assets (and keep them even after their marriage), form partnerships, make contracts, and engage in business activities. Since this legal system was introduced in a society in which there was no limit to polygyny, it could have been considered as progressive at the time. But, today, Islamic family laws or at least their interpretations that are implemented in many Muslim societies may not be considered egalitarian. For example, these laws suggest that in marriage a husband's decisions is the final rule and a wife should be obedient of him. Moreover, women have to specifically ask for permission from their husbands (or male guardians, such as their father or the eldest son, in case they were not married) when they want to work. This gives full veto power to men on women's, decisions to work a power that the justice system supports and enforces. In addition, according to these laws, it is significantly more difficult for women to divorce and they have limited rights for child custody afterwards. Therefore, women have limited opportunities to use divorce as leverage in order to gain the right to work. Hence, these family laws not only give rights but also make it "politically" inexpensive (within the household) and socially acceptable for men to veto women's decision to work (for a comprehensive review of Islamic family law see Esposito and DeLong-Bas (2001) and Sonbol (2003)). It is, therefore, not a surprise that women's movements in Muslim countries, such as Morocco, Tunisia, Egypt, and Iran, have been almost entirely focused on changing family laws.
As shown in the gray part of Fig. 1, oil and gas rents increase wages for men and reduce them for women. This rise in male wages relative to female wages shifts the bargaining power inside the households towards men. Men can use this newly gained bargaining power to affect various household decisions, from consumption and investment decisions to labor force participation of household members. Islamic family laws make it "politically" cheap (within the household) and socially acceptable for men to veto women's decision to work and hence, where these laws exist, men can use this newly gained bargaining power to reduce women's participation. Where Islamic family laws are not present, it would be more "politically" costly (within the household) and less acceptable socially for men to use this gained bargaining power on vetoing women's decision to work. They may use it more often to affect household consumption and investment decisions. Therefore, oil and gas rents should reduce FLFP more in countries that enforce Islamic family laws.
Islamic family laws are enforced in many Muslim countries even those that abandoned Islamic criminal law. There are, however, Muslim majority countries that Islamic law plays very little or no role in their judicial system, such as Uzbekistan, Azerbaijan, Turkmenistan, and Tunisia. (11) In addition, family law on paper is not necessarily enforced in practice. For example, although some family laws are based on Sharia in Turkey, they are largely ignored in practice (ilkkaracan 2012). In this study, only countries in which Islamic family law exists and is enforced in practice are considered to have Islamic family law. (12)
One may argue that religion can be interpreted differently by individuals and it is not a unique entity. But, this study does not look at the impact of religion, rather it focuses on the institution of family laws. These laws apply a set of standard rules uniformly to every Muslim in a country regardless of her or his interpretation of Islam or degree of religiosity. It is the interpretation of the judicial system and not the individual that defines the law. Therefore, one can say that these laws, particularly those pertaining to women's work, are a relatively uniform entity across Muslim countries.
One may argue that discriminatory cultural norms, although possibly weaker than discriminatory laws, could be another institution that makes it less costly and more socially acceptable for men to veto women's decision to work. To test this hypothesis, this study uses an important exogenous measure of gender discrimination norms: historical plough use. Ester Boserup (1970) argued that unequal gender roles were formed or strengthened after the introduction of plough in agriculture. The plough, unlike the hoe or digging stick, requires strong upper-body muscles, grip strength, and the ability to burst power quickly, either when the plough is pulled manually or by an animal. As a result, whenever the type of soil and the environment required the plough agriculture, men had an advantage in tending the field. This advantage led to clear distinctions in gender roles. Men were specialized in working on the field, and home became the realm of women. This division of labor limited participation of women in the main economic activities outside the home and paved the way for discriminatory norms to flourish. These norms persisted up to this date. In an influential study, Alesina et al. (2013) provide evidence for the Boserup theory. Using cross-country, within-country, and instrumental variable estimates, they demonstrated that the use of the plough, centuries ago, induced discriminatory attitudes today, and reduced FLFP, female firm ownership, and political participation. Following this evidence, this study employs historical plough use as the measure of discriminatory institutions but finds no statistically significant joint impact of oil and gas rents and this measure.
As another strong robustness check (discussed in "The Economic Model" section), this study shows that the joint impact of oil and gas rents and Islamic family law remains statistically significant, even after controlling for the joint impact of rents and all factors common across the MENA region (such as norms, culture, and institutions).
The Econometric Model
The econometric model can be written as a simple regression of country-level variables. The FLFP rate is the dependent variable and the interaction of rents and discrimination is included as one of the independent variables to identify the joint role of oil and gas rents and discriminatory institutions on FLFP. The estimates of this cross-country regression, however, are suffering from omitted variable bias. Oil and gas rents are not allocated randomly across the globe. Countries that are endowed with oil and gas are different from the rest of the world in other aspects as well, some of which are not observable. For example, they tend to be in certain geographic regions, which led them to have interrelated history, culture, or institutions. If those unobserved characteristics that are correlated with oil and gas rents affect FLFP, the estimated coefficients will be biased. Moreover, the estimates are sensitive to model specification and choice of variables in the equation. Including or excluding variables may easily change the estimated coefficients or their significance levels. In addition, FLFP rates are measured differently in every country and they are not comparable. Therefore, we may not learn much from these regressions.
One way to mitigate these problems is to use panel data and control for any unobserved heterogeneity that is fixed over time, such as history and culture, in each country. This is possible by using country-fixed effects, which we do in this paper. Another approach is to use first-differencing that produces qualitatively similar results as those reported in this paper. The following shows the regression that employs country fixed effects:
[H.sub.kt] = [[beta].sub.1][Rents.sub.kt] + [[beta].sub.2][Rents.sub.kt] x [IL.sub.k] + [gamma][Z.sub.kt] +[c.sub.k] + [[delta].sub.t]+ [e.sub.kt] (1)
in which [H.sub.kt], [Rents.sub.kt], [Z.sub.kt], are respectively the FLFP rate, rents from oil and gas, and observed characteristics of the population in country k at time t. [IL.sub.k] is the Islamic family law index in country k and [c.sub.k] is the country fixed effects. We control for all unobserved time-constant factors specific to a country such as culture, social norms, and institutions, since [c.sub.k] contains these factors at the country level. Meanwhile, the year dummies, [[delta].sub.t], capture all unobserved factors that are changing over time but are the same across countries in the same year, [e.sub.kt] is the error term. [[beta].sub.2] captures the joint impact of oil and gas rents and Islamic family law on FLFP. The results of these regressions are reported in the first two columns of Table 5.
For further robustness check, the interaction of [Rents.sub.kt] and [MENA.sub.k], which is a dummy variable equal to one if country k belongs to the MENA region and zero otherwise, is added to Eq. (1). This interaction captures the joint impact of rents and any other common characteristic across the region, such as culture, social norms, history, and political and economic institutions. Since these common characteristics could be correlated with Islamic family law, this specification removes the potential bias in the estimates.
To strengthen the results, historical plough use, a strong predictor of gender discrimination, is employed instead of the Islamic family law index ([IL.sub.k]) in Eq. (1). This specification is used to make sure that the results for the Islamic family law index cannot be replicated with this strong measure of gender discrimination. The last two columns in Table 5 report the results.
In order to estimate these econometric models, one needs a longitudinal data set at the country level that has FLFP rates, a measure of oil and gas rents, country characteristics, represented by Z in Eqs. (1) and (2), as well as the measures of Islamic family law and historical plough use. All variables except the measures for Islamic family law and historical plough use come from the dataset in Ross (2008).(13) This dataset contains a robust measure of oil and gas rents which is equal to the value of total production of oil and gas (not just their exports), subtracted by the extraction costs, and then divided by population (not GDP). This measure shows the value that the natural resource creates per person in the economy.(14)
This study uses the log of oil and gas rents instead of their actual values. This is because, like GDP and GDP per capita, rents vary greatly across countries. One can control for this large variance in rents by taking the log of oil and gas rents, especially if the relationship between rents, its interaction, and FLFP is non-linear. (15) Other variables are FLFP rates, (16) GDP per capita, and percent of population aged 15 to 64.
The measure for the presence and enforcement of Islamic law is an index that is constructed by multiplying the share of Muslims in the population of a country by a dummy that is equal to one if the country applies Islamic law to family issues and zero otherwise. (17) There are some Muslim countries that do not have Islamic family law, such as Tunisia and Azerbaijan, and there are some non-Muslim countries, such as India, that accept Islamic family law as the jurisprudence for their Muslim citizens. For the former, the index is zero and for the latter it is a number between zero and one, equal to the share of Muslims in the country. Unlike Muslims, non-Muslims in a Muslim country are exempted from the Islamic law when they interact with non-Muslims, except on criminal matters. Therefore, the Islamic family law only applies to Muslims. This is why the share of Muslim population is multiplied by the legal system dummy. Hence, this is a measure of the intensity by which Islamic family law is applied in a country. Similar results would have been found if a dummy variable was used. The share of Muslim population was obtained from the 2009 Pew Report on the global Muslim population (Miller 2009). As this variable changes little over time, it is fair to use its values in one year. Figure 2 illustrates the intensity of the Islamic family law index globally.
The measure for historical plough use at the country level is obtained from Alesina et al. (2013). This index is the country level population weighted average of historical plough use for all ethnicities residing in a country. It measures the degree to which ancestors of ethnic groups in a country used the plough historically. The summary statistics of all these variables are reported in Table 4.
As discussed in "The Economic Model" section, to mitigate the problems of cross-country regressions, one may use country and year fixed effects, as in Equation (1). Table 5 depicts the results. All regressions correct for AR(1) process in the error terms. The FLFP rate, the dependent variable, is measured in percent. Following Goldin (1995), who argued a U-shaped relationship between economic development and FLFP, log of GDP and the square of itare included as independent variables. The other independent variables are share of the population aged 15-64, log of oil and gas rents, and the Islamic family law index. The complete longitudinal data from 1960 to 2003 are used to estimate the regression. The Islamic family law index is constant for every country in this period.
The first column shows that the coefficient of the log of rents is insignificant but the coefficient of its interaction with the Islamic family law index is negative and significant. Estimating the marginal effects, one finds that as rents rise (fall) by hundred per cent (that is, they double), in a fully Muslim country with Islamic family law, FLFP decreases (increases) by about 0.07 percentage points, and in a fully non-Muslim one, it does not change.
One may argue that there are other unobservable factors that are common across the MENA region and correlated with the Islamic family law index (such as culture, tradition, social norms, and political and economic institutions). Interaction of these factors with oil and gas rents is not included in the regression and it may bias the estimates if this interaction is correlated with the interaction of Islamic family law and rents. To address this problem, we control for such unobservable factors with the interaction of rents and the MENA dummy ([Rents.sub.kt] x [MENA.sub.k]). Controlling for this changes the coefficient of interaction between Islamic family law index and rents from -0.074 to -0.047, but it remains strongly significant. The coefficient of the interaction of MENA and rents is also significant with almost the same size. (18) This means Islamic family laws have the same impact as all those factors such as social norms and culture combined.
These results support the hypothesis that oil and gas rents have a larger impact when the Islamic family laws that limit women's participation in the labor market are enforced. Rents raise the bargaining power of men in the household. As a result, men would be able to affect the household decisions more. As discussed in "The Impact of Oil and Gas Rents" section, Islamic family laws make it legitimate, socially acceptable, and substantially less "politically" costly for men to make decisions about female labor supply in their household. When Islamic family laws are not present, it is socially less acceptable and more "politically" costly within the household to veto female labor supply. In those environments, men may make decisions about household consumption and investment instead.
When one uses the historical plough use instead of the Islamic family law index (the last two columns of Table 5), she finds no statistically significant difference in the impact of oil and gas rents on FLFP across countries with various degrees of historical plough use.(19) This result is interesting as it shows that historical plough use, a strong source and predictor of gender discrimination, is not able to replicate the same results as Islamic family law index. In other words, Islamic family law is special and its impact may not be reproduced by cultural factors that exogenously measure gender discrimination across countries. This result is consistent with expectations. Discriminatory cultural and social norms exist in many places in the world but they should be very strong to make it cheap and acceptable for men to veto the female labor supply decision. The strongest form of discrimination happens when it becomes the law of the land and is enforced by the judicial system. Therefore, the interaction of Islamic family law and rent can affect FLFP, but historical plough use does not have such power.
Policy Implications and Conclusion
This paper illustrates that oil and gas rents reduce FLFP more in countries that enforce the Islamic family law. It adds new mechanisms that were overlooked: oil and gas rents may reduce FLFP more where Islamic family laws are present, by making it more likely that those laws are utilized in practice. The rise in rents increases the ratio of male to female wages and therefore raises male bargaining power inside the household. When Islamic family laws are enforced, this rise in men's bargaining power makes it more likely that they take advantage of the privileges given to them by these laws and limit FLFP.
Like social and cultural institutions, religion can have as many interpretations as the number of believers and it is not a unique entity. But, when it comes to law, the judicial system has only a few interpretations that it imposes on individuals, regardless of their-personal beliefs, interpretations, and degree of religiosity. This study is not about the impact of Islam which is not a unique entity, rather it is about specific laws and their joint impact with oil and gas rents.
Results also showed that there are still other unobservable factors that are common across the MENA region and have joint impact with oil and gas rents on FLFP. But Islamic family law has the same impact as all of those factors combined. Further research is necessary to identify and understand which factor(s) among these unob-servables is (are) at play.
One productive policy for the governments is to limit the amount of rent that enters the economy to a minimum and invest all the rents in diversified assets outside the economy. This is known as the Hartwick rule and has been suggested as the best long-run policy for countries with natural resources. (20) Besides the long-term economic benefits from following the Hartwick rule, the reduction in the amount of rent entering the economy will reduce the impact of rents on labor markets for women, particularly in countries with the Islamic family law. Following the Hartwick rule, however, is politically costly, as governments need to sacrifice today's income for prosperity in the distant future.
Although it is relatively easier to implement policies to mitigate the potential negative impacts of oil and gas rents on the economy, (21) it is harder to mitigate the joint impact of Islamic family laws and rents in reducing FLFP. For example, changing the family laws may not be feasible at least in the short run. But, economic and social globalization has a positive influence on the social values, increases tolerance, and reduces gender inequality (Berggren and Nilsson 2015). Therefore, globalization which is gaining pace with the fast and better modes of communication could be of help in the future.
Another more practical policy recommendation can be that the government makes it easier for female workers to be hired in the public sector. For example, the government of Qatar offers many positions for women, as a result of which it has one of the highest FLFP rates among countries with Islamic family laws.
According to World Bank Economic Indicators, about 51 % of females 15 and older are participating in the labor force which is comparable to this rate in Germany (53.5 %). Further research is necessary to understand the ways to improve FLFP rates in the MENA region.
Acknowledgment The author wishes to sincerely thank the editor and the anonymous reviewer for their helpful comments. Special thanks goes to John Marthinsen and Jeffery B. Nugent for reviewing an earlier version of this paper. Funding for this study was generously provided by the Glavin Council of Chairs at Babson College. All errors remain mine.
Alesina, A., Giuliano, P., & Nunn, N. (2013). On the origins of gender roles: women and the plough. Quarterly Journal of Economics, 128(2), 469-530.
Barro, R. J., & Lee, J.-W. (2013). A new data set of educational attainment in the world. 1950-2010. Journal of Development Economics, 104(C), 184-198.
Berggren, Niclas and Therese Nilsson (2015) "Globalization and the transmission of social values: The case of tolerance." Journal of Comparative Economics, Available online 2 March 2015, ISSN 0147-5967, http://www.sciencedirect.com/science/article/pii/S0147596715000293.
Boserup, E. (1970). Woman's role in economic development. London: George Allen and Unwin Ltd..
Caselli, F., & Cunningham, T. (2009). Leader behaviour and the natural resource curse. Oxford Economic Papers, 61(4), 628.
Caselli, F., & Michaels, G. (2013). Do oil windfalls improve living standards? Evidence from Brazil. American Economic Journal: Applied Economics, 5(1), 208-238.
Esposito. J. L., & Delong-Bas, N. J. (2001). Women in Muslim family law. Syracuse, NY: Syracuse University Press.
Goldin, C. (1995). The U-shaped female labour force function in economic development and economic history. In T. P. Shultz (Ed.), Investment in Womens Human Capital and Economic Development (pp. 61-90). Chicago: University of Chicago Press.
Groh, M., & Rothschild, C. (2012). Oil, Islam, women, and geography: a comment on Ross (2008). Quarterly Journal of Political Science, 7, 69-87.
Ilkkaracan, I. (2012). Why so few women in the labor market in Turkey? Feminist Economics, 18(1), 1-37.
Killingsworth. M., & Heckman, J. (1987). Female labor supply: a survey. In O. Ashenfelter, & R. Layard (Eds.), Handbook of labor economics (vol. 1, ch. 02, p. 103-204). Amsterdam: Elsevier.
Kuran, T. (2004). Islamic Statecarft and the Middle East's delayed modernization. In P. Competition (Ed.), Innovation, and Growth in the History of Asian Civilizations, edited by Peter Bemholz and Ronaliz Vaubel (pp. 150-183). Cheltenham, UK: Edward Elgar.
Kuran, T. (2010). The long divergence: how Islamic law held back the Middle East. Princeton, NJ: Princeton University Press.
Majbouri, M. (2010). Against the wind: labor force participation of women in Iran. Ph.D: Dissertation, University of Southern California.
Majbouri, M. (2015). Calculating the income counterfactuals of oil exporting countries of the MENA region. Resources Policy, 44, 47-56.
Majbouri, M. (2016a). Against the wind: labor force participation of women and economic instability in Iran. Forthcoming in Feminist Economics, (2016), 22(4), 31-53.
Majbouri, M. (2016b). Fertility and the puzzle of female employment in the Middle East. Working paper.
Majbouri, M. (2016c). Oil and entrepreneurship. Energy Policy, 94, 10-15.
Miller, T. (Ed.) (2009). Mapping the global Muslim population: a report on the size and distribution of the World's Muslim population. Pew Research: Center accessed 2009-12-17.
Moghadam, V. (2005). Gender and social policy: family law and Women's economic citizenship in the Middle East. International Review of Public Administration, 10(\), 23-44.
Robinson, J. (2005). Female labor force participation in the Middle East and North Africa. Wharton Research Scholars Journal: University of Pennsylvania.
Ross, M. L. (2008). Oil, Islam, and women. American Political Science Review, 102(\), 107-123.
Ross, M. L. (2012). What's so special about the Arabian peninsula: a reply to Groh and Rothschild. Quarterly Journal of Political Science, 7, 89-103.
Sonbol, A. E.-A. (2003). Women of the Jordan: Islam, labor and the law. Syracuse, NY: Syracuse University Press.
Spierings, Neils, and Jeroen Smith (2007) Women's labour market participation in Egypt, Jordan, Morocco, Syria & Tunisia: a three-level analysis. Working paper.
van der Ploeg, F. (2011). Natural resources: curse or blessing? Journal of Economic Literature, 49(2). 366-420.
World Bank (2004). Gender and development in the Middle East and North Africa. MENA Development Report, Washington: World Bank.
Mahdi Majbouri (1)
Published online: 28 November 2016
[c] International Atlantic Economic Society 2016
[??] Mahdi Majbouri
(1) Economics Division, Babson College. 231 Forest St., Babson Park, MA 02457, USA
(1) Labor force participation of women in the MENA has been discussed in many studies such as Spierings and Smith (2007), Moghadam (2005), Robinson (2005), and Majbouri (2010, 2016a, 2016b).
(2) See van der Ploeg (2011) for a detailed discussion. For recent empirical evidence on the economic impacts see Caselli and Michaels (2013) and Majbouri (2016c).
(3) Ross (2008) describes these as well.
(4) This is because jobs in the export-oriented manufacturing sector, such as in textile and electronics factories, do not need physical strength or much education, but require precision and smaller hands.
(5) Dutch disease refers to the negative impacts of a significant influx of foreign currency into an economy because of, for instance, discovery of natural resources or increase in their prices. The influx appreciates the domestic currency and this in turn makes the exports expensive and the imports cheap. Therefore, the domestic producers in the tradable sector lose competitiveness in the global and domestic markets and falter. The non-tradable sector, however, flourishes as the country became richer due to the influx and demand has increased. Since the industrial sector (manufacturing) is a tradable sector, the influx of foreign currency leads to de-industrialization. This phenomenon is called the Dutch disease after the Economist magazine coined the term in 1977 to refer to the economic maladies that happened following the discovery of vast natural gas reserves in the North Sea of the Netherlands in 1959.
(6) The non-tradable sector may not be able to compensate for this loss of female jobs in the tradable sector, although it becomes larger as a result of Dutch disease. This is because the non-tradable sector in a developing economy mostly consists of construction and services such as retail, which offer few jobs for women. Construction is a male-dominated sector as it requires physical strength and services, such as retail, are usually small shops with one employee who is the male owner.
(7) See the previous note.
(8) This is true if female and male labor are substitutes. For a detailed discussion of income and substitution effects of wages, and non-labor income, see Killingsworth and Heckman (1987).
(9) For example, in Iran, the government has recently transformed the large subsidies of energy and basic food into lump-sum monthly transfers to households.
(10) The legalistic nature of Islam is attributed to the fact that it had to govern soon after its introduction. Therefore, it needed to have laws to regulate the interactions of the pious in the society. Few religions, in the world, had such chance to be in government early in their formation.
(11) Sharia Courts are abolished in 1956 and inheritance law is among a few laws that remain based on Islamic law.
(12) Similar to the arguments presented here about family law affecting economic behavior, Kuran (2004) and Kuran (2010) explain that Islamic business and contract law though sophisticated and fair was one of the main causes of divergence of the Middle East economies from those of the west in the modem era.
(13) For further information about data see Ross (2008). The data is available for download at: http://dvn.iq.harvard.edu/dvn/dv/mlross/faces/study/StudyPage.xhtml?globalId=hdl: 1902.1/14307.
(14) It has several advantages over other measures. These advantages are discussed in Ross (2008).
(15) In fact, only non-linear results are significant.
(16) Collected by the International Labor Organization and released by World Development Indicators.
(17) Data on Islamic family law is from the project "Islamic Family Law: Possibilities of Reform Through Internal Initiative" at the Emory Law School, project director: Dr. Abdullahi A. An-Na'im (http://aannaim.law.emory.edu/ifl/index2.html).
(18) If, instead of the Islamic family law index, one uses the share of Muslim population in every country of the world in the regressions depicted in Table 5, she finds that the coefficient of the interaction of share of Muslims and rents becomes insignificant when one controls for the interaction of MENA dummy and rents. This provides further support for the special effect of Islamic family law. Another issue with using share of Muslim population in the regressions is that it assumes Islam as a unique entity that applies to every individual around the world similarly.
(19) Saudi Arabia may be an outlier and affect the results significantly. But, none of the results in this study change when one excludes Saudi Arabia from the sample.
(20) Majbouri (2015) calculates the benefits of following the Hartwick's rule as the right policy for countries in the MENA region in the long-run. See Majbouri (2016c) on the impact of oil on cntrepreneurship.
(21) In addition to the Hartwick's rule, a major and easier to implement policy recommendation is to regulate the amount of rent that enters the economy by setting up a stabilization fund. The goal is to isolate the economy from the shocks, created by the volatility in the oil and gas market, by allowing only a fixed amount of rent to go into the economy annually, regardless of the price of oil. Whenever the oil prices are high, the extra rent will be saved in the stabilization fund and when the oil markets crash, money is withdrawn from this fund. This policy immunes the economy from the many booms and busts in the oil markets and reduces uncertainty about the future.
Table 1 Average years of schooling for women ages 20 to 30 1970 2010 Change Algeria 1.3 10.2 8.9 Bahrain 2.9 11.8 8.9 Cyprus 7.8 13.0 5.2 Egypt 1.3 8.8 7.5 Iran 1.7 10.0 8.3 Iraq 0.9 5.4 4.5 Jordan 2.4 10.8 8.4 Kuwait 3.5 8.4 4.9 Libya 0.5 11.8 11.3 Morocco 0.9 5.4 4.5 Qatar 4.3 9.8 5.5 Saudi Arabia 1.5 10.6 9.1 Syrian Arab Republic 1.5 5.4 3.9 Tunisia 1.7 10.2 8.5 United Arab Emirates 3.9 10.5 6.6 Yemen 1.0 3.2 3.2 Data Source: Barro and Lee (2013) Table 2 Fertility transition in the Middle East and North Africa region Total fertility rates Change in fertility 1980-85 1995-2000 rates Algeria 6.4 3.3 3.1 Bahrain 4.6 2.6 2.0 Egypt 5.1 3.4 1.7 Iran 6.9 2.3 4.6 Jordan 6.8 4.7 2.1 Kuwait 43) 2.9 2.0 Lebanon 3.8 2.3 1.5 Morocco 5.4 3.4 2.0 Qatar 5.5 3.7 1.8 Sudan 6.0 4.9 1.1 Syrian Arab Republic 7.4 4.0 3.4 Tunisia 4.9 2.3 2.6 United Arab Emirates 5.2 3.2 2.0 Source: United Nations Population Division Estimates Table 3 Female labor force participation rates (% of female population ages 15+) 1990 2010 East Asia and Pacific 66.7 63.7 Europe and Central Asia 49.4 50.2 Latin America and Caribbean 40.2 53.1 Middle East and North Africa 18.3 20.9 North America 56.5 57.9 South Asia 35.8 31.7 Sub-Saharan Africa 58.5 63.0 World 52.1 51.3 Data Source: World Bank Development Indicators Table 4 Summary statistics (n = 5711) Variable Mean St dev. Min Max Female labor force participation rate 36.3 10.1 4.4 54.1 log(Oil and gas rents per capita + 1) 1.9 2.6 0.0 10.9 log(GDP per capita) 7.4 1.6 4.0 11.0 % of population aged 15 to 64 57.6 6.5 46.3 76.1 Islamic family law index 14.4 32.2 0 0.997 Historical plough use 0.5 0.5 0 1 Oil and gas rents per capita contain annual profits from oil and gas production (total value: total costs of production) in constant 2000 dollars divided by population of the country in the middle of each year. Islamic family law index is the result of multiplying the share of Muslim population in a country by a dummy that is equal to one if the Islamic family law applies to Muslims in that country and zero otherwise. Historical plough use is the population weighted average of ethnicities in a country who used plough historically Data Source: Female labor force participation rates, GDP per capita, % of population aged 15 to 64. and oil and gas rents per capita are from the dataset used in Ross (2008). The Islamic family law index is from the project "The Islamic Family Law: Possibilities of Reform through Internal Initiative" at Emory University as well as Miller (2009). Historical plough use is obtained from Alesina et al. (2013) Table 5 Panel regressions of Islamic family law and oil and gas rents on female labor force participation rate Islamic family law (1) (2) log(GDP per capita) 0.102 0.079 (0.194) (0.194) [log(GDP per capita)] (2) -0.017 -0.015 (0.014) (0.014) % of population aged 15 to 64 0.137 (***) 0.137 (***) (0.010) (0.010) log(Oil and gas rents per capita + 1) 0.005 0.005 (0.006) (0.006) x Islamic family law index -0.074 (***) -0.047 (**) (0.019) (0.023) xHistorical plough use xMENA -0.053 (*) (0.028) Observations 5,550 5,550 Number of countries 161 161 Adjusted R-squared 0.013 0.014 Historical plough use (3) (4) log(GDP per capita) 0.150 0.098 (0.194) (0.194) [log(GDP per capita)] (2) -0.021 -0.016 (0.014) (0.014) % of population aged 15 to 64 0.135 (***) 0.134 (***) (0.010) (0.010) log(Oil and gas rents per capita + 1) -0.007 -0.007 (0.010) (0.010) x Islamic family law index xHistorical plough use 0.004 0.014 (0.012) (0.012) xMENA -0.092 (***) (0.023) Observations 5,586 5,586 Number of countries 162 162 Adjusted R-squared 0.010 0.012 Female labor force participation (FLFP) rate is the dependent variable. The Islamic family law index is the result of multiplying the share of Muslim population in a country by a dummy that is equal to one if the Islamic family law applies to Muslims in that country and zero otherwise. The results are similar if one considers the share of Muslim population in a country as the Islamic family law index. Historical plough use is an index taken from Alesina et al. (2013) and is the population weighted average of the historical plough use in agriculture at the country level. MENA is a dummy variable that is equal to one if the country belongs to the MENA region and zero otherwise. All variables used in these regressions, except the Islamic family law index and MENA dummy that arc constant over time, are from an annual sample that covers 1960 to 2003. All regressions contain country and time fixed effects. All standard errors arc corrected for AR(1) process in the error term. First-differenced regressions produce the same results Data Source: Female labor force participation rates. GDP per capita. % of population aged 15 to 64. and oil and gas rents per capita are from the dataset used in Ross (2008). Islamic family law index is from the project "Islamic Family Law: Possibilities of Reform through Internal Initiative" at Emory University as well as Miller (2009). Historical plough use is obtained from Alesina et al. (2013).mmy that are constant over time, are from an annual sample that covers 1960 to 2003 (***) p < 0.01, (**) p< 0.05, (*) p < 0.10
|Printer friendly Cite/link Email Feedback|
|Comment:||Oil, Laws, and female labor force participation.|
|Publication:||International Advances in Economic Research|
|Date:||Feb 1, 2017|
|Previous Article:||Rolling regression analysis of the Pastor-Stambaugh Model: Evidence from Robust Instrumental Variables.|
|Next Article:||How much does the state of the economy influence the popularity of Austrian Parties? An empirical investigation.|