Female labor force participation and its influencing factors in the urban areas of Afghanistan - a case study of Herat City

Transcription

1 Abdul Qayoom Qaeem Female labor force participation and its influencing factors in the urban areas of Afghanistan - a case study of Herat City Volume 031 Bochum/Kabul

2 Female Labor Force Participation and its Influencing Factors in the Urban Areas of Afghanistan - A Case Study of Herat City Abdul Qayoom Qaeem Keyword list: Labor Force Participation, Female Labor, Employment, Urban Areas Abstract This study is an effort to find the influencing factors of female labor force participation in Herat City, Afghanistan. The objective of the present study is to identify the influencing factors that can affect negatively or positively on females employment in Herat City. In the present study, data was obtained from two urban areas of Herat city through interviews with 320 qualified females from age of 15 through 64. The logic model was used for analysis and Logit regression model was used to estimate the determinants of the females employment. The results indicate that education, skill, age, internal and external migration, and numbers of employed females have significantly positive effects on employment of females in Herat City. Conversely, the study found that family assets and number of educated persons in the family have a significantly negative effect on the employment of women in this city. The study also shows that household size, gender of family head, number of employed males, number of children (under 7), marital status, number of participants contributing in the family expenditure provision, and location have no significant effect on female labor force participation. Description of Data The study is mostly based on primary data collected by well-designed questionnaires during empirical research in Herat City. The data was obtained from two urban areas of Herat city (district 5 as the wealthiest and district 9 as the poorest district) through interviews with 320 qualified females (160 employed and 160 unemployed females) from age of 15 through 64. Considering the female s employment status, and the influencing factors including socio-economic characteristics of household, characteristics of respondent females, and immigration two separate questionnaires were developed for employed and unemployed females. The primary data related to 1) - socioeconomic characteristics such as household size, number of employed males, number of employed females, and number of educated persons 2) - characteristics of respondent females such as education, skill, and marital status and 3) - their migration background was collected from the target group. In addition, secondary data was used to compare female labor force participation in the 1

3 country with that of females in neighboring countries, South Asia, and the world in order to have an overall picture of female labor force participation in Afghanistan. The data was collected from two out of 12 districts of Herat city. One district is the wealthiest and the other one is the poorest. For the selection of the wealthiest and poorest districts, the data was obtained from Herat Strategic Master plan. It used the spatial distribution of households according to per capita income, showing that district 5 is the wealthiest. This is because concentration of households with a monthly per capita income more than 2000 Afghani is high there. Districts 1, 4, and 8 also demonstrate a high concentration of wealthy families yet there are poor families in these areas. While in the eight other districts, wealthy families are scarce. So the district 5 has chosen as the wealthiest area. In addition to see the difference between poor and rich families, one poor district was also randomly selected. Hence, districts 1, 4, and 8 have both poor and rich families. These three districts were excluded and from the remaining districts (2, 3, 6, 7, 9, 10, 11, and 12) with high concentration of poor families, district 9 was randomly selected for data collection. In the second stage, the researcher randomly chose two sub-districts from each of the two selected districts. Systematic random sampling procedure was used in this study in order to determine the females for interview. The number of households in each sub-district was obtained from the municipality of Herat city. Based on the number of households in each sub-district, the relative numbers of questionnaires were systematically allocated to each sub-district. Using systematic random sampling, first, the researcher calculated the sampling interval through dividing the total number of households in each sub-district by the number of females that researcher wanted in the sample and then, the researcher selected a number start between 1 and the sampling interval randomly. The selected number was considered as the first household for interview. Considering the sampling interval, selecting the other households was easy. Overall, 320 qualified females, 0.13 percent of the total female population in the urban areas of Herat city were randomly selected for the study. Research Question/Theoretical contextualization Women as a part of every community can play a very important role in the economic development of a country. The experiences of developed countries also provide us with a reason to emphasize that women can have a large share in the growth and development of all countries. But, this part of labor force is ever faced with various barriers for entrance into the labor market. 2

4 In economics, the common factors of production are technology, physical capital, and human capital, which these factors have an obvious role in the growth process of any country. Yet developing as compared to developed countries are faced with limited resources in technology and physical capital, therefore the abundance of the labor force as a factor in developing countries can have an important role in accelerating the economic growth process. Human capital includes the numbers of both sexes with their skills. This means in human capital that the quantity and the quality of the labor force are important. Yet developing countries do not use the full capacity of this factor, especially for females as one part of the labor force (Hussain, 2012). The optimum utilization of the factors of production and reach to economic growth and development regardless of gender seem impossible and nowadays, the issue of female participation in the labor force over the last century is one of the most important social phenomena. The rate has changed over the years. This change is in addition to the labor market effect on family life, men s lives, children and also the lives of women themselves. At the beginning of the twentieth century, the participation of women in the labor force was less than it is now fewer than 20% of married women worked outside the home. This rate, at the end of the century, in the United States and most developed countries, had reached at least 60% or more. Overall, the participation rate increased around the world but it is different in developed and developing countries; in developed countries, it is higher. For example, in the United States in 2002 about 92% of men and 76% of women of prime working age (25 to 45) participated in the labor force (Hoffman & Averett, 2005). In 1991, half of all women in Italy worked (Borjas 2000). Yet in developing countries this rate was lower. For example, the lowest rate in the world was in the Middle East and North African countries at only 20% of women of the age of 15 and above, while the highest rate in this region was in Lebanon, Morocco, Turkey and Yemen at about 25% (Roudi-Fahimi and Moghadam 2003). Female labor force participation in Afghanistan is lower than the average participation of women in the world and the average of South Asia. According to World Development Data, in 2011 the average FLFP of the world was 51.3%, and in South Asia it was 31.8%. The rate in Afghanistan was 15.7% (World Bank, 2011). Even if one compares the participation of women in Afghanistan with its neighboring countries in 2011, it is less (See Table 1). Table 1: LFPR of Afghanistan and its Neighboring Countries from the World Development Indicators Country Female labor force Male labor force Total labor force participation % participation % participation % Afghanistan

5 Iran Pakistan Turkmenistan Uzbekistan Tajikistan China Average Source: Own Compilation The figures in the above table show that Afghanistan s male labor force participation rate is above those of its neighboring countries (except Pakistan) and also that of the average total rate ( ). Conversely, Afghanistan s FLFP rate is lowest among its neighboring countries with the average FLFP rate ( ). In case of total LFPR, Afghanistan is behind the neighboring countries. This percentage is 49.2 in Afghanistan which is even lower than the average of total LFP rate in the region (except Iran). This comparison shows that Afghanistan does not use its total capacity of labor force as a main factor of production. Moreover, according to an NRVA report from 2007/2008, female labor force participation in urban areas was 19%, which is lower than in rural areas or among the Kuchi 1 (54% and 64% respectively). It seems that there are various barriers to the entry of women into the labor market which prevent countries from using their full labor potential. Several studies have been done to recognize the influencing factors on participation of female labor force in the market. Oppenheimer (1973) in his paper examined the influences of demographic variables on the employment of females in American society. He showed how socioeconomic and demographic factors have interacted over time to produce changing patterns of female labor force participation. He also showed that marriage and the presence of children of preschool age have a negative effect on women s labor force participation, and that females participate more in the labor force before marriage and fertility while after children are grown and go to school, their participation goes up. Of course, besides marriage, age is also an 1 Kuchi is a group of people who have a nomadic way of life. According to an NRVA report 2007/2008, 1.5 million of the population of Afghanistan is Kuchi. 4

6 important factor affecting the participation of women in economic activity. Kulik (2000) suggests that the employment of married women in the upper age ranges is more than in the lower ones, because their responsibilities are fewer than those of younger women and they can look for employment opportunities. Khan and Khan (2009) using data from 3,911 observations, showed that factors such as the age and education of women, size of family, position of women as head of a household, the number of girls in a household between 5 and 15 years, the number of daughters over 15 years, poverty and rural locality all have positive influences on the participation of married women in the labor force and that these effects are significant, whereas the education of husbands, household per capita income, being a nuclear family, the number of infants, the number of sons over 15 years, the ownership of assets by a household, and husband s unemployment all have a negative effect on the participation of women in the labor force. Haffman and Averett (2005) using a simple model of women s participation, explains that the existence of children can have a negative effect because in this model a woman can spend her whole time engaged in three kinds of activity: working for pay in the labor market, working in the household, and leisure. Because a woman who has children will spend one part of her time working in the household, the presence of children will have a negative effect on female labor force participation. Naqvi and Shahnaz (2002) using data from the Pakistan Integrated Household Survey , examined the effects of various socio-economic, human capital and demographic factors on the participation of women in economic activities. They used multi-logit and probit models to estimate the effects of these factors and found that women who are older, better educated, the head of family or come from smaller, better-off urban families were more empowered to make decisions about whether to work or not than women who were younger, poorly educated and from larger families. As mentioned previously, the participation of females in Afghanistan as a developing country is significantly lower than the average of the world, South Asian region, and its neighboring countries. Females in Afghanistan are faced with various barriers for entrance into the labor market and the objective of the present study is the recognition of the influencing factors that can affect negatively or positively on their employment. This study will consider the following research question. What are the determinants of female labor force participation in the urban areas of Afghanistan? Field research design/ Methods of data gathering 5

7 This study used the data collected from field research. The dependent variable in the present study is a binary (dichotomous) variable generated from responses to the question: Why do some females work and some of them don t? Independent variables include binary and continuous variables. As the dependent variable is a binary variable, it takes the value of 1 if the respondent female is in the workforce and 0 if she is not in the workforce. The OLS procedure for estimating qualitative as response regression models is not suitable. If the dependent variable is a binary variable, it suggests the probit and logit model (Gujarati, 2004). These models prepare a prediction for the probability of female labor force participation (employment) with a given set of characteristics. Hence, considering the characteristic of the dependent variable, the logit model is suitable and used in this study. Logit regression model is as follow: When there is more than one independent variable, according to Gujarati (2002), the model can be written as follows Where Employment status = e z (1 + e z ) z = α + β 1 HSIZ i + β 2 NMEMP i + β 3 NFEMP i + β 4 NEDUCP i + β 5 NPEXP i +β 6 LOCTN i + β 7 ASSET i + β 8 GFH i + β 9 EDUC i + β 10 AGE i + β 11 AGESQ i +β 12 MARTS i + β 13 CHLDU7 i + β 14 SKILL i + β 15 INMIG i + β 16 EXMIG i + ε i The independent variables include three categories of socio-economic characteristics of household, characteristics of respondent females, and immigration. The following table describes the variables and the Logit regression model. Table 2: Description of variables in Logit Model Abbreviation Description α Constant β 1, β 2,, β k Coefficient s of Independent Variables X 1, X 2,, X k Explanatory Variables HSIZ Number of household members NMEMP Number of employed males in household NFEMP Number of employed females in household NEDUCP Number of educated persons in households NPEXP LOCTN Number of participants contributing in the expenses of family Location (Binary Variable) 6

8 ASSET Asset (Binary Variable) GFH Gender of family head (Binary Variable) EDUC Education of females in years AGE Age of females in years MARTS Marital status of females (Binary Variable) CHLDU7 Number of children under the age of 7 SKILL Skill of females (Binary Variable) INMIG Internal migration (Binary Variable) EXMIG External migration (Binary Variable) Results In the present study employment status of females as the dependent variable is binary and takes two values 1 if female works and 0 otherwise. Independent variables include binary and continuous variables. The obtained results related to the values of Cox and Snell R-square shows that 57.4% of variation in the dependent variable is explained by the explanatory variables. A total of 76.6% is explained according the value of Naglkerke R-square. Table 3, shows the summary of descriptive statistics for this study which can be useful for a brief overview. Table 3: Definition and Summary Statistics of Important Variables Symbol Definitions EF 2 UF 3 Total N Number of observations Mean Mean Mean Socio-economic Characteristics HSIZE Household size NMEMP Number of employed males in family NFEMP Number of employed females in family NEDUCP Number of educated members in family NPEXP Number of participants in family expenses F-FH Female as family head (proportion) M-FH Male as family head (proportion) Characteristics of Respondent Females EDUC Education in years AGE Age in years MF Married females (proportion) UNF Unmarried females (proportion) CHLDU7 Children under 7 years of age (proportion) Employed Females 3 Unemployed Females 7

9 SKILL Skill of females (proportion) INCF Income of EF per month (in AFN) INCEF Income of educated EF per month (in AFN) INCUF Income of uneducated EF per month (in AFN) 2840 WHF Working hours per day for EF WHEF Working hours per day for educated EF WHEF Working hours per day for uneducated EF Migration (Proportional) INMIG Internal migration EXMIG External migration Sector of Employment (Proportional) PU Public PR Private NGOs Non-governmental organizations SE Self-employed Family Income Level in Afghani (Proportional) Low-Income Females in income level of 0-6, Middle- Income Females in income level of 6,001-30, High-Income Females in income level above 30, Source: Own Data The following table shows the Logit regression results for the present study. Table4: Logit Regression Results of the Field Research Data (Dependent variable: Probability of Female Employment) Explanatory Variables Coefficients S.E. Wald Sig. Odd ratios HSIZ NMEMP NFEMP *** NEDUC * NPEXP GFH LOCTN ASSET **

10 EDUC *** AGE *** AGESQ *** MARTS CHLDU SKILL *** INMIG ** EXMIG *** Constant Number of observations 320 * Significant at 1 percent, **Significant at 5 percent and *** Significant at 10 percent Source: Own Data Results and Discussion As seen in the table 4, the coefficient of household size is positive as expected, but it is not significant. The result indicates household size does not have any effects on employment of females. The coefficient of number of participants in family expenses is negative and insignificant. The number of employed males, contrary of the research expectation, is positive and insignificant. As expected, the number of employed females in the family has a positive relationship with the employment of females and significantly influences their participation in the labor market. The result shows that one member increase in employed females in family increases times the odds of the female employment. This may be due to the increased role of females for decision-making in the family and encourage their authority for working. The number of educated persons negatively and significantly influences female labor force participation. The result indicates that the increment of one educated member in the family decreases the odds of female participation in the labor market by 30.5%. Although, the coefficient associated to the location turns to be negative but statistically insignificant. It indicates that there is no significant difference in the female labor force participation across district 5 and 9. The result may be due to scattering families with different economic levels in both location of living. The presence of family assets as expected has a negative and significant effect on the employment of females. The result shows that if female s family has assets, this 9

11 ownership decreases the odds of her employment by a factor equal to In other words, presence of assets decreases the odds of employment for females by 62.3%. The existence of assets may be considered as a source of income. According to the theory of income effect, the increment of income causes the demand for leisure hours to be increased and desired hours of working decreased (Ehrenberg, 2008). Therefore, the existence of assets can decrease the probability of females participation in the labor market; the result is comparable with Faridi, et al. (2009). Gender of family head is as a factor that can influences on the decision-making related to employment or unemployment of females in the family. In the present study, the coefficient of this variable is positive but, contrary to the researchers expectation, is insignificant. Education has effects on participation of females in the labor market. In the present study, the coefficient of education variable is positive and significantly influences the employment of females. The result indicates that the increment of a year of education increases the odds of employment for females by a factor of In other words, the increment of a year of education increases the odds of females participation in the labor market by 26.2%. The obtained results are comparable with the theory: the productivity of female labor force will increase if they investment more time on education and therefore, they increase chances for employment with higher earnings in the labor market. It means increase in the level of education increases the productivity of individuals. Then higher education increases the efficiency of females for home production and they do their responsibilities with less time in the family. They can allocate more time to participate in the labor market. This result is supported by findings of Shaheen, et al. (2011), Khan and Khan (2009), Harandy and Jamshidi (2008), Naqvi and Shahnaz (2002), and Ackah et al. (2009). Age of females, as an influencing factor, significantly has a positive effect on their employment. The result shows one year increment in the age of females increases times the odds of their employment. One additional year of age increases the odds of female participation by 98%. The significant and negative coefficient of agesquare shows this increment is not monotonic and progressively reduces as age increases. The negative and significant coefficient of age-square shows age-participation profile. In early workingage, the additional year of age substantially increases the chance of females to participate in the labor market, while later, it progressively decreases the chance of females to participate in the labor market and become negative. The positive and significant effect of age on the employment of females is also supported by Khan and Khan (2009) in Pakistan and Mahmudian and Reshvandi (2012) in Iran. 10

12 The researcher found a negative relationship between marital status and FLFP 4, as expected. The coefficient of this variable is insignificant, showing that employment of females is not influenced by this factor. That may be because the majority of respondent females in the sample are married and there is less variation in the data. The coefficient of number of children (less than 7 years of age) is negative, as expected, but not significant. The result indicates the number of children under 7 has not effect on the employment of females. Families in the urban areas have lower numbers of children compared to rural areas and, therefore, they can allocate some of their time by working in the labor market. It could be a reason for the result obtained in this study. Or perhaps some other females in the family are involved in taking care of the children. Also the descriptive data show more than half of respondent females have no children less than 7 years of age, may be a reason for the insignificant result in the present study. As mentioned in human capital theory, skill is an important factor influencing participation in the labor market. Existence of skills in females increases their chances for getting a job and working outside the home. The obtained results from the study are compatible with theory; the sign of skills is positive and statistically significant. The result shows that existence of skills in females increases the odds of their participation in the labor market by a factor of or increases the odds of a female s employment by 317% if skilled. The internal migration of females (alone or with family) significantly has a positive effect on their employment. The result indicates the migration of females increases the odds of female employment by a factor equal to The reason may be due to migration of females during their lives (alone or with family) creating awareness of customs, traditions, and culture of other regions and, therefore, making their opinions positive about the role of females in society. External migration is one of the important factors in the present study. During three decades of war, many citizens of Afghanistan migrated to other countries and, therefore, they were influenced by different cultures. Going back to their land home from other countries with new views and cultural influences of customs and traditions changed their views. This factor significantly has a positive effect on the participation of females in the labor market. The result shows that external migration increases the odds of females for employment by a factor equal to The result may be due to the changing in view of females and their families about the role of females in the society; also external migration may cause individuals to acquire new skills and become familiar with new jobs (Usmani et al. 2010). According to human capital theory, level of skills has an important role for getting job opportunities in the labor market for both male and females. 4 Female Labor Force Participation 11

13 Conclusion The present study examines the influential factors on females labor force participation in the urban areas of Afghanistan; the case study was in Herat city. The data was collected through a welldesigned questionnaire from 320 females (including 160 employed and 160 unemployed) from two districts of the city of Herat in the winter of Logit regression model is used to estimate the determinants of the females employment. The result of Logit regression indicates that human capital variables are the main determinants of female employment as well as demographic and socioeconomic factors. The results show that education has a significantly positive effect on employment of females. The estimates indicate that an additional year of education increases the likelihood of female labor force participation by 26.2%. Also the estimates of skill show that having skill increases the likelihood of employment for females by a factor of The estimates of the education and skill confirm the explanatory power of human capital theory, stating that higher contents of human capital increase the likelihood of female participation in the labor market. Results indicate that each additional year of age increases the likelihood of female labor force participation by a factor of However, the significant and negative coefficient of age-square indicates that increment progressively decreases as age increases. Results confirm that the participation pattern of women in the labor market changes during their life-cycle. In addition, results indicate that migration positively effects the participation of females in the labor force. The estimated result shows that internal migration has a significantly positive effect on female employment, showing that those females who experienced migration individually or with family are more likely to work as compared to their counterparts without migration. Also external migration has a significantly positive effect on the employment of females, showing that those females who have had the experience of external migration alone or with family are more likely to participate in the labor force. Results of internal and external migration show that immigration has had a positive influence on the social and economic perspective of the families and females. In addition, it increased the human capital investment of families on female members, enabling females to work beside their male counterparts in the labor market. The Logit estimate shows that the number of employed females in the family has a significantly positive effect on female labor force participation. One additional employed female in the family increases the odds of female employment by a factor of This significantly increase in the participation of females in the labor force may be due to the increase in their economic and social independence in the family. Yet the study found that family size, number of children, marital status, and gender of family head have no significant effect on female labor force participation. 12

14 The study found that family assets have a significantly negative effect on the employment of women. It indicates that those women in wealthier families are less likely to participate in the labor market. In addition, the result of location indicates no significant difference in the female labor force participation. References 13

15 Ackah, C., Ahiadeke, C., and Fenny, A. P. (2009). Determinants of female labor force participation in Ghana. Institute of Statistical, Social and Economic Research, University of Ghana. Borjas, G. J. (2000). Labor economics. 2 nd ed. Boston: McGraw-Hill/Irwin. Central Statistics Organization (CSO) & MRRD (2008). The national risk and vulnerability assessment (NRVA), Kabul. Jehoon Printing Press. Ehrenberg, R., & Smith, R. (2012). Modern labor economics: theory and public policy, 11 th ed. Boston: Pearson/Addison Wesley. Faridi, M. Z., Chaudhry, I. S., & Anwar, M. (2009). The socio-economic and demographic determinants of women work participation in Pakistan: evidence from Bahawalpur district. A Research Journal of South Asian Studies, 24(2), Gujarati, D. N. (2004). Basic econometrics. 4 th ed. Boston: Irwin/McGraw-Hill. Harandi, F., & Jamshidi, F. (2008). The effects of household and individual characteristics on the urban and rural women s employment probability. Knowledge and Development, (23) Herat Municipality (2010). General information of Herat city. Herat: Herat Municipality. Hussain, S. (2012). Growth effects and the determinants of female employment in Pakistan: a macro-and microeconomic analysis, UAMR studies on development and global governance, Vol. 61, Logos Verlag Berlin GmbH, pp.3-4. Khan, R. E. A., & Khan, T. (2009). Labor force participation of married women in Punjab (Pakistan). Journal of Economic and Social Research, 11(2), Loda, M. and Florentine team (2013). Herat strategic master plan: a vision for the future. EdizioniPolistampa via Livorno, 8/ Firenze. Mahmudian, H., & Reshvandi, M. (2012). Employment of higher graduate women and effect of socio-demographic factors (Case study - Tehran city). Journal of Women in Development & Politics, 10(1), Naqvi, Z. F., Shahaz, L., & Arif, G. M. (2002). How do women decide to work in Pakistan? The Pakistan Development Review, 41(2), Oppenheimer, V. K. (1973). Demographic influence on female employment and the status of women. American Journal of Sociology, 78(4),

16 Roudi-Fahimi, F., & Moghadam, V. M. (2003). Empowering women, developing society: female education in the Middle East and North Africa. Washington, D.C.: Population Reference Bureau, pp Shaheen, S., Sial, M. H., & Awan, M. S. (2011). Female labor force participation in Pakistan: a case of Punjab. Journal of Social and Development Sciences, 2(3), Usmani, A., Nooruddin, A., Jamaludin, S., & Mujtaba, A. (2010). Emigration of Afghans and their return to Afghanistan: factors and its social outcomes. Kabul: National Center for Policy and Research (NCPR), p.14. World Bank. Development Data Group (2011). World development indicators. World Bankfree PDF. 15