Causes of youth unemployment in Pakistan.
JEL classification: C25, E24, J64
Keywords: Youth Unemployment+ Pakistan, Educational Levels, Structural Mismatch of Skills, Discrimination
Over the years, many less developed nations and the developing nations have tried to wipe-out the intensity of unemployment, which seems quite prevalent and widespread among these countries. The reasons and causes for this have remained subject to different interpretations depending upon the specifications, demographics, and regional profiles of different countries, thus the perceptions regarding having a plausible solution lacks concurrency. Not surprisingly among these unemployed a vast majority and victims belongs to young age group. According to ILO, there are 160 million unemployed people in the world and 40 percent of those out of work are young people (World youth report 2003). Pakistan is no exception to this, not only unemployment rates have been beyond reasonable limits but also a vast majority who fall prey to this belongs to youth category (Labour Force Survey 2003-04, 2005-06).
As observed in various countries the unemployment graph accounted for age specification reveals a U shaped pattern high unemployment-in the initial phases or among youth category, a moderate trend for the middle age group people and again highly intensive among old. Pakistan's trend is not far apart from this general trend. Moreover distribution of unemployed according to gender figures out high unemployment rates on the part of females. The recent government programmes have contributed to a high degree in confiscating the child labour by launching the educational campaigns rigorously and by increasing the enrolments at primary levels especially in urban areas. This policy of promoting the education has confined the variance of unemployment rate to reasonable extent on the part of males but on part of females the conventional and traditional norms are still strong enough to impede the way of females from coming out of this hall. Table 1 has fortified these distributional impacts for the economy of Pakistan which clearly testifies that among the young age group unemployment rate has been higher and also among old, furthermore the data also figures out the intensity of unemployment among females.
High unemployment rates as recorded by 8.3 percent in FY02, 7.7 percent in FY04, 6.2 percent in FY06 and 7.8 percent in FY07 reveals some serious weaknesses in making policies in line with the requirements of combating unemployment appropriately and renders this problem subject to serious monitoring and check. The adverse fact is that not only unemployment rates have been high but also youth unemployment transcended the adult unemployment rates. The causes of this high youth unemployment are manifold: lack of education, lack of skills, structural mismatch, divergence between the demographics of urban and rural areas, lack of experience, regional or province wise discrimination in the provision of job opportunities, sectored imbalance etc. The aim of this research is to scrutinise all those elements which cause unemployment with specifically focusing on youth unemployment because the fate of a nation is ultimately laid down by youth. If initially talents of this delicate group of society are sabotaged because of unfavourable job opportunities, then there would be a great chaos in the economy. The objective of the research is to sort out the reasons of unemployment and thus examining these reasons to make the prevailing situation in Pakistan in compliance with the ground facts and to trace out divergence (if any) to adopt the policies in simulation with the other nations who have successfully combated it. For a comprehensive and detailed perspective of the issue an attempt has been rendered to capture all those variables which could play a momentous role in determining the levels of unemployment and with respect to which we can make a variant analysis effectively.
2. LITERATURE REVIEW
Considerable amount of literature is available in this respect which helps us to scrutinise the different aspects of unemployment as observed throughout the world. The literature helps us to understand its determinants, causes, degree of its spread in different historical phases, the reasons of failures in different societies to outwit this curse and the policy implication as promulgated by different economies to chalk it out. One perception is that the young people are more probable of quitting their jobs voluntarily. The young people are more likely to quit off their jobs in the initial phases of their professional carriers because of their aspirations to seek the best, Flower and Freemen (1996) tested it empirically for the US economy. But this image of the young workers of deliberately not accepting jobs and quitting off is not supported by the study conducted by Lynch (1983) for the United kingdom he found in a survey that only 15 percent young men were those who had multitude job offers at their disposal the rest were scarcely able to be offered only once. The intensity of youth unemployment in comparison to adult has been proved empirically Higgin's (1997), same was proved for the South Asian countries Shehzad (2004). A research conducted by Khan and Ali (1986) explored a bulk of educated unemployed falling below the age of 30 in Pakistan. A research conducted by Gayur (1989) on the repercussions of the unemployment of educated youth attributed the unemployment to the malfunctioning of the educational and training system, which merely neglects demands of the labour markets. Chaudhry and Hamid (1998) divulged the low quality of human resource culpable for unemployment. A study by Lynch (1983) on the spells of youth unemployment declared ethnicity as a significant variable for the long spells of unemployment in Great Britain. A study by Cartmel and Furlong (2000) on the urban and rural youth unemployment revealed long term unemployment to be less common in rural areas than urban areas. A study by Andrews and Nickel (1982) on the after world war phase translated the increment in real wages in lengthening of unemployment spells, a 1 percent rise in real wage resulted in 2 to 5 percent increase in duration of unemployment.
3. DATA AND METHODOLOGY
For the purpose of this research the data has been taken from secondary source of labour force survey (2003-2004). Although according to United Nation's definition, the youth comprises of the age limit 15-24 but to focus primarily on the young people residing in Pakistan the age constraint of 15-29 has been inculcated. Some exceptions of excluding the category of widowers and the data of FATA and Azad Jammu and Kashmir have been made to avoid any discrepancy. The labour force survey renders an opportunity to comprehensively study the important characteristics of households by encompassing the demographic, cultural, educational features; this momentous feature distinctly plays a pivotal role for research prospects and provides a healthy opportunity for a meticulous and detailed analysis of the different household categories. A sample of 14515 households has been taken for this research, out of these 1151 are found to be unemployed and 13364 are employed.
Certain variables are included by distributing them in different categories or profiles. Demographic profile [including four provinces, region (rural and urbane), household size and migration], personal profile [including age, sex] and educational profiles (including primary, matric, college and higher levels of education and training). The youth unemployment itself is taken as dependent variable and all other categories in different profiles are taken as independent variables. The dependent variable youth unemployment is dichotomous because of its quantitative and qualitative nature. All the independent variables are discrete in nature except age and household size which are continuous variables. The provinces are included to figure out how each province is contributing towards generating youth unemployment. Educational categories are incorporated to translate the variation in youth unemployment for every level of education. Age groups specification scrutinises the age limits in which the young people are most likely to be out of the work force. Region elucidates the share of urban and rural areas in the variance of youth unemployment by encompassing structural flaws of urban areas and skill promoting deficiencies of rural areas. The training variable could be helpful in defining how the training and skill development techniques consequent upon the confiscation of variance of youth unemployment.
In order to deal with the dichotomous dependent variable we have used the normal probability model that emerges from a normal cumulative distribution function. Suppose [yi.sup.*] is the ability to be a part of labour force and be potentially active in the work field it actually depends upon some actively provoking elements or factors denoted by "Xi". The designed equation is as follows:
[yi.sup.*] = [beta]Xi + [epsilon]i Where [beta] represents a row vector of parameters, and Xi is the column vector of the variables that affect [yi.sup.*] and [epsilon]i is normally distributed with 0 mean. The observable binary variable is depicted as [yi.sup.*] in a following sense.
Y = 1 if [yi.sup.*] > 0 (if a person is unemployed)
= 0 otherwise
Given the normality assumption, the probability that [yi.sup.*] is less than or equal to Yi can be obtained from standardised normal cumulative distribution function as
[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]
[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]
Where f(z) represents the density function, Z is the standard normal variable with 0 mean and unit variance and its value is
Z= Xi - [mu]/[sigma]
[P.sub.i] is the probability of person being unemployed.
4. ESTIMATION, RESULTS, AND INTERPRETATION
To estimate the various categories of variables, we have used the probit model to deal with the dummy variables. For interpretation of the results marginal effects for each variable have been calculated to see how the assignment of an additional unit from each variable consequent upon unemployment for three different profiles personal, demographic and educational. The results estimated are given in Table 2. The results shown in the Table 2 depicts the marginal effects of each of the independent variable associated with casting its impact on unemployment. Within demographic boundaries the results show that if an additional individual is from Punjab this will have an impact of decreasing the probability of being unemployed by 1.07 percent, if additional individual is from NWFP this will likely to increase the unemployment by 4.7 percent and if he belongs to Sindh, this will caste a slight incremental effect on the likely outcome of unemployment by a factor of 0.1 percent. Thus the data suggests that there is well provided evidence of difference in unemployment between Balochistan (taken as base) and NWFP. The unemployment status of Sindh and Balochistan is more or less the same or there is only a scant difference. The results are highly significant for NWFP and insignificant for Punjab and Sindh. The intuition behind this trend implies that unemployment is accounted more, or it holds more weight age for NWFP in the sample. Thus if the individual is selected from NWFP a province suffering from high unemployment he will be more probable of being unemployed than any one selected from the rest of provinces.
Despite of the fact that the NWFP accounts for a small proportion in the data but this could not have taken away the culpability on government in its negligence towards this province. NWFP, on basis of low educational opportunities and flaws in the system of allocation of resources is subject to serious backwardness. For reaping potentially maximum output, the labour force within this province has been in wretched state which is still observable even today. This wretched state can also be attributed to the lack of investment in human capital which could have led this work force to cope up the new emerging technological advancement and update them. The fact that a lot of illegal businesses are carried out in NWFP, and insufficiency in provision of actual figures might have caused the figures to overstate unemployment and understate employment. On the other hand job creation, work opportunities, training and educational standards, multiple investment schemes are far better and mobile in Sindh and Punjab that is why their capacity to accommodate more of workforce is evident from the data.
If urban areas are to be compared at the margin of rural areas, it is clear that an additional individual or household worker from urban areas is responsible for increasing the probability of unemployment by 1.3 percent. There is considerable difference between the employment status of rural and urban areas. Unemployment is high in urban areas as compared to rural areas as the highly significant result implies. The high unemployment in urban areas as compared to rural areas is due to the structural mismatch of skills provided and demanded. The point to ponder over these findings is that the urban areas are usually thought to hold high employment levels than rural areas but Table 2 delivers opposite result for the economy of Pakistan. The reason could be that Pakistan is an agricultural country and almost 70 percent of its population lives in rural areas and due to disguised unemployment most of the people seem to be actively participating in economic activities but actually not. In short because of overwhelming dependence of Pakistan on agriculture (specific to rural areas) and presence of disguised unemployment might have caused the figures to overstate the rural employment and understate unemployment. Furthermore the situation would have been caused because of excessive trend of urbanisation; the rate at which the labour force is moving towards urban areas must have contributed to the high unemployment levels of urban areas. This is because even big cities in Pakistan are not efficient enough institutionally to facilitate the increasing number of upcoming labours from villages, furthermore the qualifications ranked by the employers of industries and manufacturers are not properly met by the village workforce. As rural employed people are more concentrated in small firms and small scale industries, the opportunity for young to undergo training and to enhance their skill is limited that is why these young people with low qualification cannot be absorbed in big industries in cities.
Considering the personal information, the results regarding age are also significant. For every additional increase in age responsibly reduces the probability of becoming unemployed by 0.2 percent, this also accompanies the theoretical intuition that the unemployment in adult ages is less as compared to young age group which has been substantiated and proven on solid grounds by many studies conducted all over the world. The results of the data are in harmony with the fact that youth unemployment is more severe as the significant results of the age variable suggest because the age limit taken for this research merely focuses the youth. Its causes are manifold lack of skills owed by the young labour class as they newly enter the market, incompetent youth because of absence of those institution which could offer proper counselling and training to make them compatible, lack of experience which at least in Pakistan surpasses every thing, reluctance of the employers to appoint young people on jobs because of their unawareness about the potentials of youth(initially), disregard of merit based selection, hiring of young cohorts on contract basis from where they can easily be fired as compared to the old workers (low opportunity cost faced by the firms for firing the young workers).
The state of significance is not very much different in relevance of sex, if an additional worker is male, the probability of being unemployed is found at a decline by a factor of 4.8 percent, not surprisingly more males are employed than females owing to their high labour force participation ratio and large share in the labour force statistics. Thus unemployment is less for males. Discrimination among sex is a dilemma which is generally practiced all over the world and manifold theories of labour force participation have fortified the truth of this reality. In conventional economies of Asia, females are restrained only to their typical household cores and tasks and the work they perform in the villages despite of being reflecting the image of employment is not taken into account and is seemed oblivious without any acknowledgment. Scrutinising the marital status implies that if an additional household is single, probability of unemployment increases by a factor of 4.2 percent. Thus unemployment among non married is considerably higher than among married because married people have more liabilities to meet and to cope up well with these liabilities they engage in more working activities. The reason for this trend might be the usual practice in the society which declares the choice of occupation and being an earning hand the first priority for a contended married life and as an obligation before a person is married. Employment for the married households is a constraint which they necessarily have to fulfil for leading a better standard of living. The inclusion of this variable of marital status is a provoking element for understanding the causes of unemployment in a sense that before marriage a sense of no compromise upon wages prevails among the young people. As they finally get into the bond of marriage they have to overlook their previous wage standards and sometimes revise it by accepting the even low paying jobs for the sake of meeting their liabilities.
Within educational profile the educational levels show a transitional increase in unemployment as compared to no formal education. If an additional individual gets primary education the probability of becoming unemployed is increased by 2.3 percent, for an additional individual passing out as matriculate, this probability is increased by 9.4 percent, similarly for an additional college pass out, the percentage is 18.3 percent and for tendency of an additional individual to become highly educated causes this probability to rise by 20.1 percent. For every additional level of education the difference in unemployment of non-formally educated individuals and individuals educated at different incremental levels is increasing, which suggests that individuals having no formal education accepts the jobs offering below subsistence level wages. As soon as an individual is getting indulged in more and more years of education his opportunity cost rises and his willingness to compromise with low wages diminishes. These results of high unemployment among educated and low among those with no formal education is not worth stupor for the economies of Pakistan where people start participating in labour oriented tasks abortively even before admitted in schools. This provides an evidence for the authenticity of manifold theories which posit unemployment a dilemma for the educated people or which proves unemployment to be more common in educated individuals. For training (technical/vocational) the results are significant, revealing if an additional individual holds training skills it will decrease the probable chances of becoming unemployed by 1.8 percent. Surely job givers or the employers, while giving jobs account skills quite highly. The lack of training facilities and institutions providing them do serve as an obstacle for the attainment of high employment levels and certainly a provoking factor for high unemployment among young cohorts. The results regarding migration, household size were found insignificant upon estimation.
If the results are to be dissected, even a synopsis depicts that most of the finding are totally in compliance with the idea based on former studies for different countries (Phenomenon of unemployment). Subject to formulated model various independent variables gave strong significant results for their impact on youth unemployment. As anticipated age was negatively related with unemployment. The regional aspect expounded that NWFP was inflamed with high unemployment. This haphazard scenario of NWFP is attributed to extreme failure on the part of administrative management which has been retaining the policy measures to combat this adverse situation into oblivion neglect throughout the history. The inverse relation of youth unemployment with training (technical\vocational) incites us to think over one provoking factor that training helps to diffuse the unemployment problem. The results also indicated that the "single" people are more unemployed than married because liabilities after marriage coerce them to condone their asking wages and accept even low paying jobs inevitably. Urban people are more unemployed than rural, which is attributable to the recent transitions in the manufacturing sectors and industries, the introduction of new technologies has brought a new consignment of machines and plants, which has reduced the demand for the manual work. The worse scenario among all is the high unemployment among the educated youth, with every level of education difference of educated unemployed from the people holding no formal education (taken as base) status increased. This testifies the existence of structural unemployment in Pakistan. Moreover the unemployment is lower for males than females, which is logically interpreted on grounds that in Pakistan in most families' males are the only breadwinners.
In nut shells youth unemployment in Pakistan is prevalent because there is improper counselling of future dimensions, there are no institutions which could guide the young students which field to adopt congenial to the requirements of the country, in foreign countries which have restricted the unemployment rate to a reasonable extent, young students are properly entertained, and are directed to those professions which are likely to open new avenues of success and prosperity for them in the upcoming future. This feature is seriously lacking in Pakistan. This structural unemployment is also the reason of high urban unemployment in Pakistan. Faulty administrative structure is no exception to this; it is also culpable because despite of encountering the proper figures of different provinces, emphasised policies are lagging far behind the actual levels which are desired to confiscate the spread of vicious cycles in the backward areas like NWFP. Pakistan seriously lack those institutions which could provide compatible training skills, this is perhaps the most neglected sector of economy. Unemployment is an ordeal for our youth; Okun's law suggested that a 2 percent decrease in GDP is responsible for a 1 percent increase in unemployment. If our economy is tested under his criterion, certainly Pakistan's GDP remained subject to sporadic fluctuation all throughout the history, which reasons that resources in our country are not utilised properly and economy has never been in state of full employment.
In Pakistan a diminutive portion of the total population has excessive resources to lead a luxurious life, the rest are subjugated badly as their income are depressed because either they lack the opportunity to be employed or if at all they get it, it is far below the subsistence level, among these victims a large proportion comprised of passionate youth who steps in the market with the devotion to thrive and ends up mostly in despair and devoid potentials. Thus low incomes and below subsistence wages are responsible for the perpetuation of the vicious circle of unemployment in Pakistan. The inflexible wage structure in Pakistan, minimum wage laws, poor monitoring of demand and supply of labour force all are contributing in aggravating the situation.
6. POLICY RECOMMENDATIONS
To safeguard and confiscate youth unemployment from soaring, it is necessary to take immediate corrective measures so that economy of Pakistan could withstand this reprehensible curse.
* Special focus should be placed on the issue of making students aware of the scope of education they acquire prior to the stage where they make themselves fully engaged towards their desired field of study.
* Special internship programme be launched so that students could acquaint themselves with experience side by side their education, this will help to alleviate the hesitancy on the part of employers in hiring the fresh students.
* To cope up with the trend of urbanisation, more and more industries be established in urban areas which can accommodate the increasing number of labour force.
* As it is obvious that large proportion of labours holding skills are more likely to get job, training facilities should be made accessible within the reach of a common man also by making the establishment of training institutes a necessary part of development programmes.
* Small scales industries should be established so that those labours who remain unemployed despite of owing skills in hand made items and handicrafts can play their active role in society as an employed member.
* Corrective measures must be taken by the government to achieve a smooth flow of productivity and growth persistently. Balance should be maintained in focusing on the objectives and performance of various sectors. No sector should be made to thrive at the cost of other. Moreover an attempt should be made to uplift all the sectors simultaneously.
* An increment in GDP be made in line with those standards which directly results in increasing the investment prospects and productivity. This aspect is particularly important because some factors sometimes over express GDP (such as excessive foreign remittances); such type of increase in GDP is not necessarily accompanied by productivity and employment opportunities.
The main aim of this research is to scrutinise all those elements which cause unemployment, with special focus on Youth Unemployment. This is very important issue for a country like Pakistan--a labour abundant country. The author choice of this vital subject is therefore commendable. However, I have few points which may help to understand the issue more clearly and make the study more invaluable.
(a) It is not clear to me from the paper that which definition of female labour force participation is used in the paper. Labour force survey provides two types of data for female participation.
(b) The study states that unemployment is high due to mal functioning of education and training system. At another page it states that unemployment increase by education level. This need to be analyse in more detail with mismatch analysis.
(c) It will be very informative if youth unemployment by education level will be calculated on the basis of demand and supply. Demand can be projected using sectoral growth and employment. Labour force survey provides information on employment by occupation and by industry also.
(d) Author points out that employment among age group of 50-59 have increased and voluntary unemployment has increased among youth. It can be concluded from this that voluntary unemployment among youth has forced older persons to work more.
(e) The study states that rural unemployment is lower than urban unemployment that may be due to under employment in rural areas--In majority of households all member of a family engaged in agriculture activities.
(f) The results show that singles are most likely to be unemployed. But I think this result may differ by gender as employers are most likely to higher single women.
(g) Wage have not been included in the analysis, therefore it cannot be concluded from the paper that wages are responsible for unemployment.
(h) The author concludes with a cyclical trend which has not been analysed in the paper. The author can delete the statements about it from the paper.
Pakistan Institute of Development Economics, Islamabad.
Andrew, M. and S. Nickell (1982) Unemployment in the United Kingdom since the War. Review of Economic Studies 49, 731-759.
Blanch Flower, D. G. and R. B. Freemen (1996) Growing in Work. OECD Employment Outlook, July.
Blomqvist, A. G. (1985) Unemployment of the Educated and Emigration of Post-secondary Graduates from the LDCs. The Pakistan Development Review 28:3&4, 643-654.
Cartmel and Furlong (2000) Youth Unemployment in Rural Areas. University of Glasgow. www.jrf.org.uk/knowledge/findings
Chaudry, M. A. and A. Hamid (1998) Unemployment in Pakistan. Pakistan Economic and Social Review 36: 2, 147-170.
Ghayur, S. (1989) Educated Unemployed in Pakistan: Estimates of Imbalances in Current Flows. The Pakistan Development Review 28:4, 603-613.
Haque, I. (1994) The Significant Variables of Full Employment and their Explanatory Values. The Pakistan Development Review 33:4, 1345-1356.
Higgin, N. O. (1997) The Challenge of Youth Unemployment. (Employment Working Papers.) www.ilo.org Jacobson, T. and H. Ohlsson (2000) Working Time, Employment and Work Sharing:
Evidence from Sweden. Empirical Economics 25, 169-187. Khan, I. A. (2003) Impact of Privatisation on Employment and Output in Pakistan. The Pakistan Development Review 42: 4, 513-536.
Khan, S. R. and S. Z. Ali (1989) Some Findings about the Unemployed Highly Educated Persons in Pakistan. The Pakistan Development Review 25:4, 731-737.
Lynch, L. M. (1983) Job Search and Youth Unemployment. Oxford Economic Papers 35, 271-282.
Malik, M. A. A. (2004) A Country Report for Symposium on Globalisation and Future of Youth in Asia. Ministry of Labour, Manpower and Overseas Pakistanis. www.ilo.org/pub
Malinvaud, E. (1989) Lessons to be Learned from European Unemployment
of 80s. The Pakistan Development Review 28: 4, 311-333.
Robinson, W. C. and N. Abbasi (1979) Unemployment in Pakistan. The Pakistan Development Review 28: 4, 313-331.
Shezad, M. (2003) The Challenge of Unemployment. India's National Magazine 21:4.
Waqqas Qayyum <firstname.lastname@example.org> is a PhD student at the Federal Urdu University of Arts, Science, and Technology, lslamabad.
Table 1 Unemployment Rates Sex and Age (%) 2003-04 2005-06 Age Groups Total Male Female Total Male Female 15-19 13.2 12.8 14.9 9.98 10.02 9.82 20-24 10.3 9.3 15.0 7.37 6.87 9.43 25-29 7.1 6.1 12.5 4.88 4.30 7.34 30-34 4.5 3.8 7.4 2.85 2.45 4.33 35-39 2.9 2.0 7.2 2.37 1.67 4.87 40-44 2.9 2.5 4.8 2.68 2.00 5.45 45-49 3.5 2.3 9.5 2.87 2.08 6.04 50-54 5.1 3.5 12.2 6.32 4.52 14.48 55-59 7.1 4.5 20.7 8.35 5.89 19.50 60 and Above 12.8 8.9 36.1 11.62 7.18 34.12 Source: Labour Forty Survey 2003-04 and 2005-06. Table 2 Variable Coefficient dy/dx Z Personal Profile Age -0.0231 -0.0027 -4.49 Male * -0.3394 -0.0480 -7.64 Single * 0.3973 0.0422 8.58 Demographic Profile Punjab * -0.0908 -0.0107 -1.41 NWFP * 0.3380 0.0473 4.99 Sindh * 0.0111 0.0013 0.16 Urban * 0.1159 0.0138 3.36 Hsize -0.0030 -0.0003 -0.70 Mig * 0.0607 0.0074 1.11 Educational Profile Prim * 0.1825 0.0232 3.48 Matric * 0.6526 0.0946 14.37 College * 0.9282 0.1836 16.26 Higher * 0.9275 0.2013 7.61 Training -0.1585 -0.0187 -2.12 Cons -1.1219 -5.30 Probit Estimates Number of Obs = 14515 LR [Chi.sup.2] (14) = 842.06 Pseudo [R.sup.2] = 0.1047 Log Likelihood = -3600.3463 (*) dy/dx is for discrete change of dummy variable from 0 to l. Marginal effects after Probit y = Pr(unempl) (predict) = .05907343
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||GROWTH EMPLOYMENT / POVERTY ISSUES|
|Publication:||Pakistan Development Review|
|Date:||Dec 22, 2007|
|Previous Article:||An analysis of Pakistan's vulnerability to economic crisis.|
|Next Article:||Understanding poverty through the eyes of low-salaried government employees: a case study of the NED University of Engineering and Technology.|