CC BY
43NEET-молодежь Таджикистана: роль образования и семья
о сч о
СЧ СП
о ш m
X
<
m о х
X
Миров Лоикджон Лутфулоевич,
аспирант, кафедра «Экономика и управление», Технологический университет Таджикистана, Mirov_loiq@mail.ru
Научная статья посвящена исследованию основных факторов, которые влияют на молодежь Таджикистана в период их перехода от образования к труду, и станут причиной роста объема молодежи категории NEET. К группе NEET (NEET — англ. Not in Employment, Education or Training) относятся молодые люди в возрасте 18-29 лет, которые являются безработными или экономически неактивными и при этом не учатся и не охвачены профессиональной подготовкой/переподготовкой. В данной статье оценивается вероятность попаданию в категорию NEET в зависимости от семейных и личных характеристик молодежи. Эмпирической основой исследования выступают данные исследования Перехода от школы к труду в регионах Кавказа и Центральной Азии (TEW-CCA) за 2017 г., которые охватили 2000 молодежь Таджикистана. При анализе данных использованы методы регрессионного анализа и оценка Каплана - Мей-ера. Результаты анализов показывают важную роль образования на начальном этапе карьеры, особенно для девушек. Семейные факторы, такие как образование и занятость родителей, а также финансовое положение семьи снижают вероятность попаданию в категорию NEET, т.е. повышают вероятность успеха на рынке труда.
Ключевые слова: NEET-молодежь, не работает и не учится, молодежная безработица, молодежь Таджикистана, переход от образования к труду
Introduction
Annually 150 thousand of youths enter the labor market of Tajikistan, but only 30-40 thousand of them are able to find work in Tajikistan, the rest part goes abroad or ends up in so-called economically inactive group.
The lack of suitable work, the big competition in labor market in the homeland create additional difficulties. Failure on the home labor market and/or abroad leads youths to alienation, frustration and some of them - to radicalization. But even if young people keep composure after long and unsuccessful job search, they leave the fighting for a job and join a number of economically inactive population. Each year the youth unemployment inactiveness grows in Tajikistan. Officially registered unemployment level is 2,4%, but last LFS2016 (Labor Force Survey) shows 6,9% of overall unemployment rate and 10,6% for youth 15-29 y.o.. [1, p.78-79. Traditional methods for estimating of unemployment are weak in developing countries with young population, where labor market is not able to create necessary number of productive jobs. For the last two-three decades other indicators were developed to better understand the situation with youths. One of indicators is NEET (Not in Education, Employment, or Training) that shows both number of unemployed and number of economically inactive people who are out of school. According to the last researches the level of NEET-youth is 38-41% in Tajikistan which is the highest level in Post-Soviet Union countries [2, p.4-5]. More and more youth people with tertiary education and VET are lost in transition after school.
Beside of education, as in any other traditional communities of the World, in Tajikistan the impact of family values - family background is significant on transition period. Researchers of Tajikistan show high impact of family members [3, p.16] on main life decisions like marriage, education and employment.
All the above-mentioned testify that there is a big mismatch between youth people's background and labor market which leads to high level of NEET.
NEET is a new term in science which researchers started to use about three decades ago in England. According to the first delineation of the phenomenon, the term NEET indicated only subjects aged 16-18 who had completed the compulsory school path and who decided not to continue the path of education, nor to take courses of professional qualification and as a result they cannot find a job [4, p.350].
From the early 2000th, the topic of NEET has begun to be addressed systematically at the European level, thanks to the statistics compiled by supranational institutions like the European Commission or the OECD. These, especially in recent times, have begun to pick up, in different countries, a significant amount of data about the phenomenon, each country adopting its own definition of the category. Initially, an important distinction is that the OECD refers to the term NEET as the age group between 15 and 24 years, whereas the European Commission extends the limits of the phenomenon up to 29 years [5, p.3]. In the American context, the NEEt category was "addressed" long before in the UK.
NEET-youth has next criteria:
- Age 15-29,
- Not in educational or training system, not attaining course to get a job, and do not doing apprenticeship, is not waiting to start a course
- Not working and not attained any job related training for the last 4 weeks, not on vocation and not waiting to start a job.
In our countries all papers on NEET started being published in 2015.
E.Ya. Varshavskaya in her work (Russian NEET-youth: characteristics and geography) estimated the level of NEET in Russia and some characteristics of this group of young people. It noted that in 2014 18% of the youth of Russia was NEET-youth. This group includes young people aged 18-24. The level of education in this group is lower than that of working group. 71.7% of the members of this group have no work experience, and of those who used to work almost half of them worked before at the last job they worked for less than half a year. Also, the author noted that the probability of NEETs to be involved into informal job is higher than the worker group. The author divided the NEET youth into subgroups: NEET- unemployment and NEET - economic inactivity. The work shows that the gender of the respondent and the type of settlement where the respondent lives play significant role as well. The work covers the data of 20122014 [6, p.124] .
The last paper of Anna Zudina "The Pathways That Lead Youth in NEET: The Case of Russia" analyses causes and consequences of the NEET in Russia. As the research finds, "Despite the heterogeneous nature of the Russian NEET, the risks of falling into this state are mainly associated with education - either with its insufficient level (in the case of inactive NEETs) or with its low quality (in the case of unemployed NEETs)". Another factor which can lead to the NEET is "changes in the marital status". Economic activity as well can play significant role in job finding process, "The probability of finding a job next year reaches 50% for unemployed NEETs, and about 30-40% for inactive NEETs". [7, p.21]. Another paper which was published in August 2018 is about NEETs in Kazakhstan. Author Dinara Alimkhanova in her paper called "Understanding the Rising NEET Phenomenon in Southern Kazakhstan" showed factors affecting NEET level in Southern Kazakhstan. The paper using descriptive analyses shows significant correlation between being NEET and holding lower education level. The paper is weak of literature review and beeper analyses.
As for NEETS in Tajikistan it was mentioned in 2017 for the first time in the World Bank report. The paper analyzed results of a survey which was conducted 2013. The scope of the survey were migration, skills and employment. According to the report "NEET rates among youth have proven to be very persistent, despite a decade of strong growth in Tajikistan. Between 2003 and 2013, the share of NEET among youth increased from 37 to 41 percent despite relatively favorable economic conditions". 41% is one of the highest level in the region. Particular, in case of female youth, report shows that the level of NEET is twice higher than share of NEETs between males. NEET rates of female youth rose from 47 percent to 60 percent between 2003 and 2011 [2, p.4-5]. Another paper on NEETs of Tajikistan is an analytical report LFS2016 Tajikistan (Labor Force Survey by International Labor Organization). The paper shows some frequently tables on NEETs. According to the report 30% of youth are NEETs, which is about 690 thousand from 2.3
million youth population aged 15-29 . 90,5% of NEETs in Tajikistan are girls [1, p.78-79]. From 2013 to 2016 NEET level dropped from 41% to 30%. Analyzing last report make any researcher to think twice before accepting of 30%. If one check NEETs 15-24y.o. the level of NEET is 29,3%, but for age cohort 25-29 which the level of NEET much more higher (less people in education), the level of NEET for 15-29 should be not less than 34-38%.
Literature review showed that there are very limited number of papers on factors affecting NEET on youth of Tajikistan, and there are only a few researches on NEETs in the region. It is necessary to conduct independent research on NEETs of Tajikistan and using modern methodology to fill the gap in literature.
Research questions
This paper investigates main factors affecting NEET in Tajikistan. Mainly it will try to estimate the impact of education and family background on NEET-youth.
The net effect of higher levels of education on labor outcomes for youth should be positive according to the basic human capital theory. However, the opposite has been observed for a number of developing countries [8, p.95-105]. For example, for girls with higher level of education typically takes longer to find a job that is "safer and more regular." Some of them choose to prolong their education instead of staying home and doing house chores [9, p.8].
As for family background effect we will estimate the impact of parental migration history, parental education, parental working status and financial situation in the family.
Some empirical researches investigated the impact of migration of the parents to the educational attainment of the children in Tajikistan. Labor migration may have negative impact on education through higher absenteeism of school children, whose parents are abroad and cannot control school attendance. Nevertheless, a number of empirical studies show that money remittances have positive impact on education in migrant household as opposed to household without migrants [3, p.16]. Literature shows that in households with migrants the mobility of girls much more lower than in household without migrant.
Other factor which will be tested in this research is an employment status of the parents. As Celik's (2008) study suggested, if a mother and father are unemployed or not in the labor force, their child, too, may be NEET, as the resources, skills or the social capital required to get a job might be missing, and the opposite may be true for those whose parents are employed [9, p.16].
Parent education is expected to impact youth labor market outcomes in various and distinctive ways. As parent education goes up, it is likely that parent income goes up and this might create more opportunities to stay in education longer or it might prolong unemployment or inhibit labor force participation and vice versa. People with higher education will look for a better job longer. So, parental education will impact to employment status via education, especially education of the mother. But father with higher education had higher chance to be employed. Father using his business network can increase the chance of the young person to be employed. [10, p.124].
Family financial situation can lead to higher education level followed by better chance to be employed. But, literature review shows that "often parents allocate family resources to sons because in many societies after marriage the daughter leaves the household while the son remains.
X X
o 00 A c.
X
00 m
o
ho o ho o
o cs o cs
an
O HI
m
X
3
<
m o x
X
Lower investment in daughters leads to lower educational attainment, reducing earnings [11, p.30.
Based on the previous evidences, we have a list of hypothesis:
H1: The net effect of higher levels of education on labor outcomes for youth is positive.
H1a: Effect of education is stronger for women than for men.
H1b: Women with tertiary education spent more time to find job, but after a period of the time has higher chance to be employed.
H2: parental education has no significant effect to the employment status of the youth, but they impact through education attainment of the children.
Method
To answer the questions of the research - the impact of an education and family background to NEETs, we predicted our dependent variables using independent variables. The type of the main dependent variable - NEET is binary. To predict the likelihood to be in we used logistic regression. Unlike descriptive statistics, logistic regression concurrently controls for multiple factors and generates the estimated probability of being NEET if all other indicators were held constant [12, p.40-50].
The base logistic regression equation (Logit Function):
Using logarithm we change the equation (1): ( Pr(NEET) \ L0<i-Pr(NEET)]=^+ ^ <2>
Equation (2) is a Logistic Model (Log (odds)). Both equations help us to estimate a negative or positive and significant association between our independent variables and the likelihood of NEET status. The intercept p0 estimates the probability of being NEET when all explanatory variables equal zero.
The type of our second dependent variable NEETlength is continuous. We approach this from an event history perspective. We estimate failure functions using the Kaplan-Meier product- limit method.
Modeling was done on statistical software package stata v14.
3.Results
Using dataset of "Youth of Central Asia: Tajikistan" we will show overall description of youth aged 15-29 based on their employment status. Drawing graph (see pic.1) approve the methodology that the picture of employment status for the age categories 15-19, 20-24, 25-29 are different both for men and for women. For ages 15-18 the proportions of NEETs is similar for both men and women. For ages 20-24 share of NEETs for women is 70,6% which is two times higher than for men. In this age group only 12,6% of women are employed, while share of employment for men is 29,9%. For third age category, youth ages 25-29 share of NEET is increases for both men and women. If 78,5% of women of ages 25-29 are in NEETs, for men share of the NEETs in this age group is 41,2%.
Graph on pic.1 shows that for men share of NEETs is about the same after the ages 17-18, but after the breaking point ages 17-18 share of NEETs grows for women repeatedly.
In order to determine the factors influencing NEETs - to answer the main questions of the research we will draw
some model, which can lead us to deeper understanding of the NEET.
Fig.1 Employment status of boys (left graph) and girls (right graph) over ages.
Source: Youth of Central Asia 2015; own calculations.
Before modeling we will observe our dataset. In this part we will explore data from TEW-CCA Survey. If in the first dataset we had data for youth ages 15-29, but second dataset contains data for people 18-29. As showed the pic.1 share of NEETs in ages 15-17 is very low. So we can use our second dataset for the purpose of the study. As well, second dataset - TEW-CCA data has more data than the first one and its methodology is better meet the purpose of our research.
256 out of 2000 observation were people of ages 30-35, whose data was removes from the dataset. For this study, individuals aged between 18 and 29 years are grouped into the following three states regarding labor market behavior:
1. Employment: employed individuals, excluding the students.
2. Unemployment: non-students who report having searched for a job in the reference period, and who are available to work in two weeks.
3. Inactivity: individuals who are not attending school at the time of the survey, who do not have a job, and who report not having searched for a job in the reference period.
The last two states constitute the NEET group and first group is Non-NEET one.
To make the data more appropriate for logistic regression, many variables were recoded to binary. Table 1 presents properties of the variables.
variable name type variable label variable values
IsFemale Binar y Gender 1 - female
age Float Age of the respondent 18 - 29 y.o.
Education
EdInSecAndLow er Binar y Initial Secondary and lower 1 - Initial Secondary and lower
EdSecond Binar y Secondary completed 1 - Secondary completed
EdInitProf Binar y Initially prof education 1 - Initially prof education
EdSecondProf Binar y Secondary professional education 1 - Secondary professional education
EdBachalor Binar y Bachalor 1 - Bachalor
EdMAAndHigher Binar y Master or higher 1 - Master or higher
brAndSist doubl e How many brothers and sisters fo you have?
mothEdVETorHi gher Binar y Education of mother 1.VET or higher 0. Secondary or lower
fathEdVETorHig her Binar y Education of father 1.VET or higher 0. Secondary or lower
mothIsEmp Binar y Employment status of mother 1.Eemployed O.Not employed
fathIsEmp Binar y Employment status of father 1.Eemployed O.Not employed
Regions of TJK
isRural Binar y Urban Rural 1- Rural
isNEET Binar y Is respondent NEET 1- NEET, 0 -employed.
more likely to be NEETs with general secondary education than young men with Secondary professional education or higher education. First model for female and male shows that young women with initially professional education are less likely to be NEETs young women with general secondary education than but in case of men vice versa.
Table 2. Results of Regression analyses. People with General secondary education defined as control group. Table 2
Logistic regression (odd ratio) estimating the likelihood of being in
According to the data 50% of individuals are girls. All individuals in the sample are between the ages of 18 and 29 and average age is 24.
About half of individuals in the sample have General Education. Analyses show that proportion of individuals by regions of the country is closer to total population of the regions. 74 per cent of individuals live in urban area. Individuals in the NEET category constitute 50 per cent of the sample.
Results of regression
Using the odd ration and varying significance by the p-value young women with incomplete secondary and lower educational attainment are more likely to be NEETs than young women with general secondary education. However young women are less likely to be NEETs with general secondary education than young women with VET education or higher education.
Young men are more likely to be NEETs with Incomplete general secondary education and Professional Technical education (Initial vocational education) than young men with general secondary education. However young men are
Dependent variable: isNEET Model 1 Model 2
Male Female Male Female
EdInSecAndLower 1.47 1.48* 1.43 1.60*
(0.32) (0.29) (0.32) (0.35)
EdInitProf 1.24 0.17* 1.07 0.09*
(0.42) (0.14) (0.38) (0.09)
EdSecondProf 0.56* 0.25*** 0.56 0.26***
(0.16) (0.07) (0.17) (0.07)
EdBachalor 0.44*** 0.40*** 0.46*** 0.37***
(0.09) (0.10) (0.10) (0.10)
EdMAAndHigher 0.25*** 0.20*** 0.24*** 0.19***
(0.06) (0.05) (0.06) (0.05)
mothEdVETorHigher 1.60* 1.12
(0.36) (0.25)
fathEdVETorHigher 1.01 1.08
(0.16) (0.17)
mothIsEmp 0.84 0.71
(0.14) (0.12)
fathIsEmp 0.88 0.67
(0.19) (0.16)
brAndSist 0.98 0.86***
(0.04) (0.03)
N 858 886 797 812
pseudo R2 0.052 0.077 0.052 0.101
Source: TEW-CCA 2017; p<0.10, **p<0.05, ***p<0.01;
own calculations. Note:
On second model we add family background variables, as Number of brothers and sisters, education of father, education of mother, employment status of father, and employment status of mother. Model shows that impact of last four variables is not significant, but only number of brothers and sisters. Observing odd rations of education level of youth show that family background has high impact on education attainment of the youth, but not on labor market outcome. Number of brothers and sisters has significant negative effect on being in NEET for youth, but significance level for women is higher.
To make stronger model we run model for urban and rural separately.
In rural women less likely to be in NEET group than in urban area, but for men vice versa. Because of less number of observations in urban area level of significance became lower. For rural, impact of education become stronger for both men and women. Last column of the table 3 shows that in rural area women with bachelor education about 7 times
X X
o
OD >
c.
X
OD m
o
io
2 O IO
o
*
o
CS
0
CS
en
01
o m m
X
3
<
m o x
X
and with master degree (including specialist, master or higher degrees) about 15 less likely to be in NEETs than young women with general secondary education.
Table 3
Logistic regression (odd ratio) estimating the likelihood of being in NEET category for different groups. Separate models for different
Source: TEW-CCA 2017; p<0.10, ** p<0.05, *** p<0.01;
own calculations. Note:
system if some is still NEET he or she has about 100% of chances to be in NEETs forever.
To examine the effect of education on the length of being in NEETs we will separately estimate failure functions for each educational group separately dividing the dataset to men and women.
Dependent variable: isNEET Model 3
Urban Rural
Male Female Male Female
EdInSecAndLower 0.29 0.72 2.05** 2.29**
(0.19) (0.30) (0.56) (0.64)
EdInitProf 0.76 0.29 1.25 1.00
(0.61) (0.42) (0.54) (.)
EdSecondProf 0.48 0.08** 0.56 0.22***
(0.36) (0.06) (0.19) (0.07)
EdBachalor 1.02 0.64 0.26*** 0.13***
(0.54) (0.36) (0.07) (0.06)
EdMAAndHigher 0.15** 0.24** 0.35*** 0.07***
(0.10) (0.11) (0.11) (0.03)
EdSecond (omitted) 1.00 1.00 1.00 1.00
(.) (.) (.) (.)
mothEdVETorHigher 0.93 0.92 1.49 0.98
(0.47) (0.36) (0.44) (0.29)
fathEdVETorHigher 0.79 0.72 1.08 1.64*
(0.31) (0.27) (0.21) (0.34)
mothIsEmp 1.57 0.89 0.65* 0.70
(0.62) (0.31) (0.13) (0.15)
fathIsEmp 0.90 0.52 0.97 0.64
(0.47) (0.26) (0.26) (0.20)
brAndSist 0.90 1.08 0.96 0.95
(0.11) (0.10) (0.05) (0.05)
N 189 221 608 586
pseudo R2 0.107 0.121 0.118 0.191
Duration from education to the first job - length of being in NEET group.
We describe the transition from education to first stable job (leaving NEETs) depending on the educational attainment. Transition from education to first job is our dependent variable NEETlength. We approach this from an event history perspective. We estimate failure functions using the Kaplan-Meier product- limit method [13, p.6].
The failure functions in Fig. 2 show the cumulative proportion of individuals that left NEETs at time t after leaving education, separated by Gender. The time at risk is defined as the period of time between leaving educational system and the month of first job. The figure shows that even after 120 month from of being in NEETs about 60% of women could not job, while for men it is only about 20%. Kaplan-Meier estimators show that as the length of being in NEETs become longer as the chance to leave the NEETs become lower. After five years of leaving the education
Figure 2.
Source: TEW-CCA 2017; own calculations
Regarding the educational groups, the results from the Kaplan-Meier estimators show that women with General basic or secondary education have the biggest problems when making their school-to-work transition. Even after 120 month 75% of women with General secondary education or lower have not leave NEETs. Vocational education holders between women spend more time in NEETs during first year, but have higher chances than any other groups to leave the NEETs.
In case of men, the effect of education on school to work transition is lower than for women. For the first 40 month as higher the level of education as higher the chance to leave NEETs, but after 60 months no any chance to leave the NEETs, no matter the education level. As longer men stay in NEETs as lower the effect of higher education.
Conclusions
Using two datasets we examined the effect of education and family background to NEET-youth of Tajikistan.
Going back to our hypotheses, we had these hypotheses:
H1: The net effect of higher levels of education on labor outcomes for youth is positive.
H1a: Effect of education is stronger for women than for men.
H1b: Women with tertiary education spent more time to find job, but after a period of the time has higher chance to be employed.
Our analyses show that in general education has positive effect of labor market outcome. It decreases number of youth in NEETs. We accept H1. Effect of education is stronger for women than for men, which means we accept H1a. Women with VET spent more time in NEETS, but after a period of the time has higher chance to leave the NEET. So, we reject H1b.
H2: parental education has no significant effect to the employment status of the youth, but they impact through education attainment of the children.
Analyses show that after adding family background variables to the model effect of education become stronger, but the effect of these variables was not significant. So, Family background has higher impact to education attainment, but not on being on labor market outcome. Family background can decrease the level of NEET by increasing the number of students, but not employed youth. Here we accept H2.
Concluding the results of the research, we can say that education can decrease the level of NEET in Tajikistan. Better family background can lead to better education and which will be followed by lower NEET level. These chain "family background -> education -> employment" is more important for women, that for men.
This paper is first paper on NEET youth of Tajikistan. More surveys with more observations are needed to find better models with significant estimators on micro and macro level to explain the reasons of high NEET level and to observe the dynamic of NEET-youth level in Tajikistan.
The impact of education and family background on NEET-
youth of Tajikistan Mirov L.L.
Technological University of Tajikistan
Article aims to investigate factors affecting youth during their transition from education to work. The focus of the paper is NEET-youth NEET (not in education, employment or training) of Tajikistan, and main factors that lead young people to this category. Particularly, this paper aims to investigate the impact of family background and individual characteristics on NEET in Tajikistan of age 18-29. Data from large-scale retrospective survey of 2000 youth, which was conducted in 2017, is used. Methods of regression analyses and Kaplan-Meier estimator are used to analyse data. Initial analyses showed that education has positive impact on labor market outcome and the effect is stronger for women. Family background has no significant impact on labor market outcome, but can increase level of education which decreases the NEET level. Educated parents have higher chances to have employed children. Key words: NEET-youth, not in education and employment, youth unemployment, Tajikistan youth, school-to-work
References
1. Tajstat (Statistical agency under the president of the republic of Tajikistan)." Situation on Labour market of Tajikistan. Report". Results of Labor Force Survey in 2016. 2017. Dushanbe, 220p.
2. Rokova V. St., Ajwad M. I. Tajikistan Jobs Diagnostic: Strategic Framework for Job. - Washington, DC: World Bank Group, 2017.-p. 19.
3. Saodat Olimova. The Impact of Labour Migration on Human Capital: The Case of Tajikistan, Revue européenne des migrations internationales [Online], vol. 26 - №3 | 2010, Online since 01 December 2013, connection on 30 September 2018. URL : http://remi.revues.org/5239 ; DOI : 10.4000/remi.5239
4. Fahmy, E. (2006). Youth, poverty and social exclusion. In C. Pantazis, et al. (Eds.), Poverty and social exclusion in Britain: The millennium survey. Bristol: The Policy Press. P.490,
5. Batini F. NEET: A Phenomenom Yet to Be Explored// Springer Science. 2016, October 18. -p. 19-37.
6. Varshavskaya E.Ya., Russian NEET-youth: characteristics and geography// HSE conference paper, 2017,p. 122-132
7. Zudina A.A. NEET-youth on labor market of Russia. HSE. WP3, Labor Market issues, 2017, p.21
8. O'Higgins, N. Youth unemployment and employment policy: A global perspective Geneva, International Labour Office, 2001, p.228
9. Çelik K. (2008) 'My state is my father': youth unemployment experiences under the weak state welfare provisions of Turkey, Journal of Youth Studies, 11:4, 429-444, DOI: 10.1080/13676260802106191
10. Kahraman, Berna, "Youth Employment and Unemployment in Developing Countries: Macro Challenges with Micro Perspectives" (2011). Graduate Doctoral Dissertations. p. 354
11. Becker, G. (1991) A Treatise on the Family. Enlarged ed. Cambridge, MA: Harvard University Press. p.441
12. Hosmer, D. W., and S. Lemeshow, and R. X. Sturdivant. 2013. Applied Logistic Regression, 3rd edition. New York: Wiley. p.522
13. Gebel, M., & Heyne, S. Delayed transitions in times of increasing uncertainty: School-to-work transition and the delay of first marriage in Jordan. Research in Social Stratification and Mobility (2016), p.12
X X
o
00 >
c.
X
00 m
o
ho o ho o
