AN ECONOMIC ANALYSIS OF OCCUPATIONAL DIVERSIFICATION AMONG HOUSEHOLDS IN ANDHRA PRADESH

Transcription

1 Indian Agricultural Research Institute From the SelectedWorks of A Amarender Reddy 2003 AN ECONOMIC ANALYSIS OF OCCUPATIONAL DIVERSIFICATION AMONG HOUSEHOLDS IN ANDHRA PRADESH A Amarender Reddy Available at:

2 AN ECONOMIC ANALYSIS OF OCCUPATIONAL DIVERSIFICATION AMONG HOUSEHOLDS IN ANDHRA PRADESH A. AMARENDER REDDY DIVISION OF AGRICULTURAL ECONOMICS INDIAN AGRICULTURAL RESEARCH INSTITUTE NEW DELHI

3 AN ECONOMIC ANALYSIS OF OCCUPATIONAL DIVERSIFICATION AMONG HOUSEHOLDS IN ANDHRA PRADESH by AMARENDER REDDY A thesis submitted to the Post Graduate School, Indian Agricultural Research Institute, New Delhi In partial fulfillment of requirements for the award of degree of DOCTOR OF PHILOSOPHY IN AGRICULTURAL ECONOMICS 2003 Approved by 1. Dr. Praduman Kumar (Chairman) 2. Dr. R.P. Singh (Co-Chairman) 3. Dr. B.R.Atteri (Member) 3. Dr. V.K. Gupta (Member) 4. Shri. H. Sikarwar (Member)

4 Dr. Praduman Kumar Professor Division of Agricultural Economics Indian Agricultural Research Institute NEW DELHI: (INDIA) CERTIFICATE This is to certify that the thesis entitled An Economic Analysis Of Occupational Diversification Among Households In Andhra Pradesh, submitted to the faculty of the Post Graduate School, Indian Agricultural Research Institute, New Delhi, in partial fulfillment of the requirements for the Doctor of Philosophy in Agricultural Economics, embodies the result of bona fide research work carried out by Mr. A. Amarender Reddy, under my guidance and supervision. No part of the thesis has been submitted for any other degree or diploma. He has duly acknowledged all the assistance and help received during the course of the investigation. July 19, 2003 New Delhi-12 (Praduman Kumar) Chairman Advisory Committee.

5 Acknowledgement Dr. Praduman Kumar, Professor, Division of Agricultural Economics, IARI and Chairman of the advisory committee for his inspiring and erudite guidance, motivation and encouragement provided during my doctoral programme and the preparation of this thesis. He has provided valuable insights into this study, I have benefited immensely from my association with him for which I shall ever remain grateful to him. Dr. R.P.Singh, Head, Division of Agricultural Economics, Co-Chairman of my advisory committee for his constant encouragement and efforts to provide all support as head of division and co-chairman of advisory committee. He has evinced keen interest in the progress of my study at IARI and has contributed heavily to shape this thesis in its present form for which I deeply indebted to him. I express my sincere thanks to Dr. B.R. Atteri Principal Scientist, Division of Agricultural Economics, IARI for offering valuable suggestions and constructive criticism as member of my advisory committee. Dr. V.K.Gupta and Shri.H.Sikarwar, Division of Agricultural Statistics and Division of Computer Applications respectively for offering valuable suggestions and constructive criticism as members of my advisory committee. I express my sincere gratitude to them for helpful guidance, which I received from them at various stages of my academic pursuit. I am thankful to Dr Pramod Kumar and Dr. N. P. Singh for very cordial and helping hand in thesis preparation and for the valuable suggestions.

6 I express my sincere thanks to Dr. Masood Ali, Director, Indian Institute of Pulses Research, Kanpur for his kind-heartedness in grating leave to complete my research work at IARI. I express my sincere thanks to Dr.I.P.S.Yadav, Head, Agricultural Economics, IIPR, Kanpur for his help in thesis preparation. I sincerely acknowledge and thanks to officials of National Sample Survey Organisation (NSSO), New Delhi for their kind help in providing literature and valuable data for the study. I am happy to acknowledge the help offered by Maneesh, Sujit, Arya, Veketachalam and Govindraj during my stay at IARI. I wish to acknowledge my parents, brothers, sisters and my wife for their affection and encouragement. I wish to acknowledge Director, IARI and Dean IARI for their kind-heartedness in permitting me to complete my Ph. D work at IARI, New Delhi. The SRF awarded by the IARI New Delhi made this doctoral programme possible. Date: Place: New Delhi ( A. Amarender Reddy)

7 S.No. Contents Page. No. 1 Introduction Objectives Organization of thesis Review of literature occupational structures of households/workers employment and unemployment factors influencing employment structure gaps in existing literature Methodology the data Analytical Approach Description of the Study area Location Historical background of the region Agro-climatic conditions Socio-economic profile Land utilization Non-agricultural sector Results and discussions Employment pattern among households and workers Distribution of persons by employment status Determinants of employment structure Summary and concussions Bibliography i-vii

8 LIST OF TABLES S. No Table Title Page No A statistical account of different regions in Andhra Pradesh Percent distribution of households by principal occupation in and in rural Andhra Pradesh Poverty head count ratio by occupation group in and in rural Andhra Pradesh Percent distribution of households according to principal occupation by income class in rural Andhra Pradesh ( ) Percent distribution of households according to principal occupation by land holding group in rural Andhra Pradesh ( ) Percent distribution of households according to principal occupation by educational of head of household in rural Andhra Pradesh ( ) Percent distribution of households according to principal occupation by age of head of household in rural Andhra Pradesh ( ) Percent distribution of households according to principal occupation by social group and religion of Household in rural Andhra Pradesh ( ) Percent distribution of households among region and household occupation in rural Andhra Pradesh ( ) Percent distribution of households among region by income group Percent distribution of workers among activity status/sector of employment in to (age above 15 years)

9 S. No Table Title Page No a Table 5.10a: Percent distribution of workers among activity status and sector of employment in to in rural Andhra Pradesh Percent distribution of workers among activity status by household occupation in rural Andhra Pradesh ( ) Percent distribution of workers among activity status of employment by income group in rural Andhra Pradesh ( ) Percent distribution of persons among employment category by landholding group in rural Andhra Pradesh ( ) Percent distribution of workers among activity status of employment by social group and religion in rural Andhra Pradesh ( ) Percent distribution of workers among activity status of employment by educational level in rural Andhra Pradesh ( ) Percent distribution of persons among activity status of employment by age group in rural Andhra Pradesh ( ) Percent distribution of workers among activity status of employment by region in rural Andhra Pradesh ( ) Percent distribution of workers among sector of employment by household occupation in rural Andhra Pradesh ( ) Percent distribution of workers among sector of employment by income group in rural Andhra Pradesh ( ) Percent distribution of workers among sector of employment by landholding class in rural Andhra Pradesh ( ) Percent distribution of workers among sector of employment by social group in rural Andhra Pradesh ( ) Percent distribution of workers among sector of employment by educational level in rural Andhra Pradesh ( )

10 S. No Table Title Page No Percent distribution of workers among sector of employment by age group in rural Andhra Pradesh ( ) Percent distribution of workers among sector of employment by region in rural Andhra Pradesh ( ) Percent distribution of workers by sector/activity status of employment by household occupation in rural Andhra Pradesh ( ) Percent distribution of workers by sector/activity status of employment by income group in rural Andhra Pradesh ( ) Percent distribution of workers among sector/status of employment by landholding group in rural Andhra Pradesh ( ) Percent distribution of workers among sector/activity status of employment by social group and religion in rural Andhra Pradesh ( ) Percent distribution of workers among sector/activity status of employment by general education level in rural Andhra Pradesh ( ) Percent distribution of persons among sector/activity status of employment by age group in rural Andhra Pradesh ( ) Percent distribution of workers among sector/activity status of employment by region in rural Andhra Pradesh ( ) Herfindahl Index of occupation of workers by sector of employment and region (NCO 1968) in rural Andhra Pradesh ( and ) Employment status in rural Andhra Pradesh (age above 15 years) Percent distribution of persons among household occupation and employment status in rural Andhra Pradesh ( )

11 S. No Table Title Page No a Participation of women (percent of respondent women) in different household activities in rural Andhra Pradesh ( ) Percent distribution of persons among income group and employment status in rural Andhra Pradesh ( ) Percent distribution of persons among landholding group and employment status ( ) Percent distribution of persons among social group and religion by employment status in rural Andhra Pradesh ( ) Percent distribution of persons among educational level and employment status in rural Andhra Pradesh ( ) Percent distribution of persons among age group and employment status in rural Andhra Pradesh ( ) Percent distribution of persons among region and employment status in rural Andhra Pradesh ( ) Percent distribution of persons among sub-round (season) and employment status in rural Andhra Pradesh ( ) Workdays per week by different socio-economic groups in rural Andhra Pradesh ( and ) Parameter estimates of household principal occupation (against self-employed in agriculture as the control) choice model Parameter estimates of activity status of employment among persons of above 15 years age Parameter estimates of sector of employment against agricultural workers as the control among workers of above 15 years of age Wage income per week among different socio-economic groups in rural Andhra Pradesh in and at prices

12 S. No Table Title Page No a Wage income per week among different sector of employment in rural Andhra Pradesh in and at prices Factors influencing wage income per week among wage labourer in rural Andhra Pradesh a Factors influencing wage income per week among wage labourer in rural Andhra Pradesh

13 CHAPTER 1 Introduction The second five year plan, formulated in 1956, had envisaged that the development of the economy at the expected pace would change pattern of employment by lowering the proportion of the labour force engaged in agricultural occupation from 70 percent to about 60 percent by 1976 ( 20 years hence ), that absorb growing labour force thereby limit unemployment to tolerable levels (Government of India 1956: 317). Consequently every five-year plan reiterated this point and introduced many non-farm employment generation programmes in rural areas. However, the above said target has been reached only recently in the year (that is 23 years later than the target year of 1976). There was significant difference between the male and female workforce engaged in agriculture. About 53.1 percent of the male workers were engaged in agriculture in the year , whereas among female workers, about 74.8 percent were active in agriculture at national level (Sundaram 2001a). The slower change in employment structure in terms of share of non-agricultural sector, however, was not enough to provide employment opportunities for increasing population, resulting of high unemployment, underemployment and low work participation rate. Recent studies have also shown that during years when non-agricultural rural employment increases, rural poverty declines, and that the converse also

14 holds. In India, the unemployment rate is 5 percent but poverty is more than 25 percent( Dev 2000). In other words, much of the employment is not adequately productive or remunerative. It indicates that access to different occupations can determine another 20 % of poverty levels in India. Even though the structural change in work force was sluggish in rural areas of the country, there has been some clear indication of a decline in the share of agriculture in the workforce. The share of agricultural workers in total rural workforce decreased from 86.1 percent in to 76.2 percent in The share of agricultural workers in the male work force had dropped from around 84 percent in 1961 to about 70.4 percent in The rural female work force however, showed a slower decline from about 90 to 86.3 percent between and (Chadha and Sahu 2002). In addition to gender differences, the pattern of change was not uniform among different states of rural India. For example share of employment in the non-agricultural sector is quite high in southern states like Tamil Nadu, Kerala and Karnataka while on the other hand, share of non-agricultural employment is quite low in northern states namely Bihar, Uttar Pradesh, Rajasthan and Madhya Pradesh. More over shift from agriculture to non-agricultural employment is rapid in southern states compared to northern states of India. The slow structural change did have its negative impact on work participation rate, among rural men work participation rate reduced from 553 in to 531 in A similar drop in work participation rate was recorded among rural females from 328 in to 299 in (Sundaram 2001b). In addition to sector of employment, activity statuses of employment (self-employed, regular employed 2

15 and casual labourer) play an important role in increasing livelihood of rural workers. For example, Sundaram (2001a) concluded that the average wage incomes of regular wage/salaried workers would be higher than that received by the casual labourers. In recent years, these changes in structure of employment, reduction in work participation rates and increase in unemployment rates in rural India have generated a lot of interest among researchers to understand the causes for these changes. There were a good number of studies that focused on the structure of employment and unemployment. But most of them were concentrated in analysing macro-level data i.e., at state and national level. At micro level very few studies were available that explore association of household and personal characteristics with occupational structure and findings vary with location and not conclusive. For example, Livingstone (1999) in his study found that in urban areas, higher educated persons were mostly employed in non-agricultural sector thereby earning higher incomes than less educated since 1970, but the same was not true in case of rural areas. Most of the higher educated and high resource endowed persons either remained in low productive agricultural sector or underemployed in rural areas. While studying Pakistan rural labour market, Fafchamps and Agnes (1998) concluded that households with better educated and resource endowed males divert labour resources away from agricultural activities toward non-agricultural work, as there were no significant returns to higher education in agricultural sector. However the same was not true in case of women in rural Pakistan. Analyses of Ghanaian rural households 3

16 by Jolliffe (1996) and rural China by Yang (1997) gave similar results. Contrary to above arguments, a widely cited survey by Lockheed, Jamison and Lau (1980) summarizes from 18 different studies in 13 countries, concluded that education has a positive effect on incomes of self-employed in agriculture in rural areas. However Phillips (1987) argues that these results vary substantially by geographic region. However in the context of rural India in addition to educational levels, many researchers found other socio-economic and traditional factors were observed to influence employment diversification to non-agricultural sector. In a study, Ghose (1999) found that even where the education and skill levels are similar, gender, kinship, caste, tribe etc, remain important determinants in rural areas in India. In conformity with above, Dreze and Srinivasan (1997) concluded that female-headed households concentration in agriculture was more compared to male-headed households. In another study, Ray (2000) concluded that the SC/ST households generally experience significantly lower standards of living than others as they mostly depend on agricultural labour or non-agricultural labour in rural India. In Indian context, the divergence in opinion in relative importance of education and other socio-economic factors in influencing occupational pattern/employment status (employed, unemployed and underemployed) necessitated the need for a micro-level study to analyse why are people in a particular part of the sector/activity status of employment? How a person to be employed or unemployed and what personal and household characteristics 4

17 influence employment pattern among households and workers? In light of the above background, the current study was undertaken in rural Andhra Pradesh as the poverty reduction rate was highest (5.9 percent per annum) in rural Andhra Pradesh during , that reflecting rapid change in rural Andhra Pradesh (Reddy et al. 2003). That necessitated the current study with the following specific objectives. 1.1 Objectives The specific objectives of the study are as follows: 1. To examine the occupational structure of households in different regions of rural Andhra Pradesh. 2. To assess the level of unemployment, underemployment and poverty in different categories of households. 3. To find out various socio-economic factors affecting employment pattern in different regions and among households, and 4. To suggest policy measures to reduce unemployment and underemployment. 1.2 Organisation of the thesis The rest of the thesis is organised as follows. Chapter II presents a brief review of the studies related to the present investigation. Chapter III depicts methodology used in this study. A brief description of the general profile of the study area is described in chapter IV. Chapter V examines the general profile of the households in each region. While chapter VI is devoted to the analysis of employment and unemployment details of the persons for each region and for 5

18 different socio-economic groups. Chapter VII examines the occupational structure among members of households in each region. Finally chapter VIII summarises findings, outlines the limitations and highlights policy implications of the study. 6

19 CHAPTER 2 Review of Literature The employment is a crucial factor in determining the overall development of households and persons. The structure and composition of employment and unemployment is the result of a complex phenomenon, which encompass socio-economic and household characteristics of persons and the location. This chapter reviews the various issues related to the employment structure and characteristics, the causes of differences in employment structure and composition among households and individuals. There is a large pool of literature that studies the employment structure and composition with special emphasis in agricultural vs. non-agricultural employment growth. However, most of the studies concentrated at macro level analysis of employment growth and its composition with respect to structure of employment. Very little attention is paid on the structure and composition of employment, unemployment and underemployment at micro level due to lack of reliable data at micro level as most reliable National Sample Survey Organisation (NSSO) employment and unemployment survey results were available at national and state level until recently. There were two major national level data collection agencies, which collect information on population and labour force in India. The census data 7

20 provide information on the population and workforce by age and gender disaggregated to the level of states, districts and blocks. The most widely used censes are of 1961, 1971, 1981 and The NSSO is the other national agency that collects data on employment in its quinquennial employment and underemployment surveys of , , 1983, , and These data provide information on employment in much greater detail and are also considered more reliable and stable than censes. Most of the studies have concluded that the employment structure is shifting in favour of non-agricultural sector and this trend is more conspicuous among males and in urban areas compared to females and rural areas respectively. This trend is especially conspicuous from the late eighties and it is continuing in the nineties also. The Fiftieth round of National Sample Survey Organisation (NSSO), results indicate that the trend of a shift in the structure of the work force away from the primary sector continues at least for the male workers. The tide, however, appears to have slower for female workers in rural areas (Sundaram 2001a). 2.1 Occupational structure of households/workers In India, the unemployment rate is 5 percent but poverty is more than 25 percent. In other words, much of the employment is not adequately productive or remunerative (Dev 2000). It indicates that access to different income/employment sources can determine another 25% of poverty levels in India. The recent NSSO household employment data also shows that there is greater variation in incomes from different sources of employment. In 8

21 highlighting earning differences Sundaram (2001a) concluded that the average wage incomes of all regular wage / salaried workers (other household type) would be higher than that received by the casual labourer (agricultural and non-agricultural labourer households) and that, the growth in labour productivity and hence in labour incomes of those (self-employed with asset-base) who hire the casual wage labourer Employment and household characteristics A complex web of household characteristics determines access to different sources of income/employment. The importance of incorporating household size and composition in welfare analysis has long been recognised (Buhmann et al.1988). Household type and household structure not only respond to economic conditions, but can also determine employment sector/status of persons (Laslett 1972). There is an evidence that education, skill and assets determine most part of the variation in income/employment sources of households. Even where the educational and skill levels are similar gender, kinship, caste, tribe etc., remain important determinants of access to different sources of employment/income of households (Ghose 1999). Empirical work on Indian data has been relatively scarce one exception is the study by Dreze and Srinivasan (1997) who utilized disaggregated data on household size and composition to analyse the economic position of female headed households in India and concluded that female headed households (hhs) mostly engaged in agricultural sector. In another study, Ray (2000) concluded that the SC/ST households, i.e., the backward castes, generally experience significantly lower standards of living than others whose sole 9

22 employment source was agricultural labour or non-agricultural labour in rural India Trends in occupational diversification Most of the studies on sectoral and regional trends in employment are based on NSSO and census data sources. Summarising the results of a number of earlier studies based on NSSO and census data Visaria and Basant (1994) noted three broad trends in rural employment pattern in India. (1) During the last three decades ( ), the share of the rural non-agricultural sector in the total rural labour force has increased. The trend is more clearly evident among male workers than among female workers; (2) Within the rural nonagricultural sector, the increase in the share of tertiary sector exceeds that in the secondary sector; (3) Bulk of increase in the rural non-agricultural sector is explained by the increase in the proportion of casual workers. As noted by Ghosh (1995), the expansion of non-agricultural sector was partly due to government employment generation programmes. The demand for non-agricultural goods was maintained in years of lower agricultural incomes by stepping up government s expenditure in rural areas by expanding employment programmes and rural development schemes (Sen 1997). In a series of papers, Bhalla (1997) analysed changes in the workforce structure using the censuses of 1981 and In the decade , there was a sharp contraction in the absolute number of non-farm workers in rural areas. However, in the subsequent two decades, the absolute number of nonfarm workers as well as share of rural areas in the growth of non-farm workers 10

23 grew. Forty percent of all new non-farm jobs during the 1980 s were created in rural areas. Further, in rural areas, non-farm jobs ( ) account for close to a third of all new jobs for male workers (Bhalla, 1993b). A number of authors have analysed the workforce change and sectoral patterns using NSS data (Basant and Kumar, 1984; Visaria, 1989; Unni, 1991; Chadha, 1993, 1997). There was a steady increase in the share of rural male and female workforce engaged in non-agricultural employment by usual status during the period to In , however, while the proportion of male workforce in the non-agricultural sector continued to record a rise, that of the female workforce declined (Unni, 1996b). Hence, though the share of male workforce in non-agriculture in continued to show a rising trend, its growth was relatively lower compared to the earlier period Except for community and personal services among the non-farm activities, the employment growth slackened in every branch of economic activity for rural male workers. Analysis of the NSS data shows that the states where non-farm employment for male workers expanded fast during the 1970 s and 1980 s were Punjab, Haryana, Gujarat, Himachal Pradesh, Rajasthan and Tamil Nadu. A modest expansion was witnessed in Karnataka, Maharastra, Orissa and West Bengal (Chadha, 1997). Besides the share of non-agricultural sector in the rural workforce, Bhalla (1993a and 1997) has computed a diversification index at the state level for 1961, 1971 and This is a more sensitive measure, which takes in to 11

24 account the changing composition of the non-agricultural segment as well as agricultural and non-agricultural shares in the total workforce. According to her, rural India has experienced, over the two decades up to 1981, a deepening of workforce diversification, an increasing complexity within the non-cultivating segments-rather than a significant widening of rural workforce diversification. Of the three broad sectors of the rural economy the tertiary sector has diversified fastest, the secondary sector next, while the primary sector has scarcely diversified at all. Indices of diversification computed at the district level show more complex spatial patterns. Analysis of district level data reveals that existence of three kinds of regions: The first, concentrated in Bihar, display symptoms of agricultural involution; the second, high farm productivity districts of Punjab and Coastal Andhra Pradesh, shows increasing concentration in agriculture; The third is now a large block of highly diversified districts clustered around industrial towns, or forming long geographical corridors, linking large urban conglomerations (Bhalla, 1997). Most conspicuous of this extends from Delhi through Haryana, Rajasthan and Gujarat all the way to the coast, viz., Ahmadabad and Surat (Bhalla, 1993b). As many studies have been concentrated in studying at macro level by relating public expenditure, agricultural gross domestic product, inflation and prices, unemployment and poverty, there is very little empirical evidence on employment status and structure at household level Gender differences in employment structure Women were mostly concentrated in agricultural sector with little 12

25 diversification to non-agricultural sector. The trend of an increasing participation in the non-agricultural sector was observed for women as well in the 1980 s. The geographical variation in the pattern of employment of women compared to men is also very large as indicated by much higher coefficient of variations across states (Mitra, 1993). This lower participation is also observed when sex ratios (number of female workers per thousand male workers, in the rural non-farm sector and farm sectors) are computed (Mitra, 1993). The sex ratio is obviously lower in the non-farm sector, using both census and NSS data. Across states, the sex ratio in the non-farm sector increased or remained unchanged in Andhra Pradesh, Haryana, Karnataka, Kerala, Maharastra, Tamil Nadu and West Bengal. In nonhousehold manufacturing it shot up in a majority of states, except Bihar, Gujarat, Haryana, Orissa, Punjab and Rajasthan. Household manufacturing recorded a sharp increase in the sex ratio in all states except Karnataka (Mitra, 1993). Most of the differences in employment pattern between men and women were due to differences in educational levels. India has one of the highest female-male literacy gaps. According to Human Development Report (1998), only 7 countries have higher gap than India. The economic incentives for investing in girls' education are weaker than the incentives for investing in boys' education, for various reasons. Gender division of labour combined with partially property rights makes the employment possibilities for educated women much more limited than the possibilities for educated men. As a consequence, it becomes less probable for women to study. In addition after marriage, contrary to boys, 13

26 girls leave their parents' homes, and that reduces parents' benefit from their daughters' education Employment structure among children Child labour is a colossal problem in India. India has the largest number of child labourers in the world. They are all around, little figures picking rags and toiling long hours in appalling conditions within repair shops, machine shops and roadside restaurants. Less visible and more exploited are the millions of others employed in vast, mushrooming, often hazardous, industries manufacturing beedis, carpets, locks, slates, glass products, brassware, garments and leather good (Mishra, 2000). Chaudhri (1997) stated that about 80 percent of India's child labour in rural areas are marginal workers engaged in such occupations as dry land farming. Household surveys in West Bengal, Bihar and Eastern Uttar Pradesh reveal that all girl children in rural households are in fact disguised child labourers, with the degree of exploitation varying with socio-economic backwardness (Mukhopadhyay 1994). Most of the children work at simple repetitive manual tasks that do not require long years of training or experience. The work is low paying, involves drudgery and is hazardous (Swaminathan, 1998). Bhatty (1996) concluded that child labour is less a phenomenon of poverty than of social attitudes and sensibilities. Learning skills through education is a sure way to break the cycle of child labour and low income. Child labour in India is still a rural phenomenon. Primary education should be brought to the forefront of political agenda to eradicate child labour. 14

27 2.1.5 Activity status of workers Activity status of workers generally classified as self-employed (working in owned enterprises), regular employed (salaried workers) and casual labourer (working as hired workers on daily basis) according to status of work. Visaria and Minhas (1991) argued that in view of the resource crisis and other structural rigidities, the organised sector would be unable to provide a high growth of employment in the coming years. Therefore, the large majority of nearly 80 million persons who would join the labour force during will have to find work as self-employed and casual workers. Evidence from West Bengal indicates that rural workforce diversification to non-farm activities appears to offer greater scope for self-employment and regular employment than casual wage employment. Self-employed activity constituted various kinds of primitive manufacturing like earthen utensils and bamboo work, catering to the local market (Chandhra Shekar, 1993). Similar evidence of a scope for self-employment and regular employment rather than casual work in non-agriculture was also observed in Gujarat (Unni, 1996c). Another study on rural labour markets by Kannan (1990) noted distress-induced self-employment. Wage employment (regular or casual) is unlikely to absorb the growing labour, so that domestic manufacture and petty commodity production coexist with various forms of wage/self employment. 15

28 2.2 Employment and Unemployment Unemployment rate is an important indicator, which shows overall health of the economy in one single measure. The unemployment rate to a greater extent depends on factors such as general level of economic development, business cycles, inflation, labour intensive vs. capital-intensive technology as well as micro socio-economic factors such as household characteristics and individual literacy rate etc. A general fear is that the employment scenario in the coming decade would deteriorate further under the impact of liberalization, privatisation and Structural Adjustment Programme (SAP) of Government of India. Access to employment by more than one adult member of a household, whether in the form of self-employment or wage labour, improves the living standard of the household. Increasing household income by increasing the number of earners within the household has been widely reported. This may be done by taking in an elderly relative or non-relative to care for children, thus freeing adult women for paid employment (Laslett 1972; Stack 1974), by delaying marriage of children, and thus retaining their wages within the family and conversely, by reducing the time of childhood dependency by limiting education and beginning working life at an early age (Hackenberg et al. 1984). The greater concern is high unemployment rate among young and high rate of not in labour force among women. It suggests that the recent decline in the labour force participation rate of youth can be explained by the rise in proportion of youth (particularly in the age group years) attending an educational institution. This was indicated by not only the percent of those who 16

29 reported studying to be their usual activity but also by the proportion of those who responded affirmatively to a direct question about their current attendance at a school or college (Visaria 1998). However the high level of unemployment rate among this group suggests presence of unexplained problems around high unemployment among youth. Low work participation rate among women is common in rural India. Most of the studies indicated that employment of women particularly had positive impact on child welfare and nutrition. Employment generation for poor women is being promoted as a crucial component of poverty alleviation strategies by the Indian government (World Bank 1991). Various indicators of women s status such as sex ratio, control of assets, access to communal resources and access to kin networks has also been associated with access to employment to women (Agarwal 1990). Women take care of most of the household duties which absorb most of the time of women thereby reduces time available for so called economic activities and that causes low work participation rate (Jain 1985). The important interrelationship between earnings and spell of unemployment has been explored by many researchers, in human capital theory perspective, Human capital theory indicates that job termination incurs the permanent loss of firm specific human capital, while an unemployment spell not only precludes the accumulation of work experience but may also bring the deterioration of general skills, future earnings in employment will therefore be lower than if unemployment had not occurred (Wiji et al. 2001). Policy initiatives to encourage individuals to stay on at school and acquire the necessary 17

30 qualifications and skills should be one of the priorities to eliminate spell of unemployment. The micro level studies on unemployment were relatively less. In this category, the most relevant examples are Ham et al. (1998), Ham et al. (1999), Finta and Terrel (1997), Abraham and Vodopivec (1993). The results vary in significance but all of them find that higher education have a positive effect on the probability of unemployment. Vocational training or more generally, a higher level of firm and industry specific skills is found to improve the chances of finding a job, while Earle and Pauna (1998) find a weak negative effect of specific skills on probability of unemployment. Another set of studies with large sample age is found to be weak positive effect of the probability of existing unemployment (Storm and Terrell 1997). Using categorical definition of age, Bellmann et al. (1995) and Earle (1997) find that older individuals have a lower probability of dropping out of the labour force. Jones and Kato (1997) find higher education and experience increase women s chances of finding jobs. When usual principal status alone considered Andhra Pradesh is one of the major states recording high rates of unemployment among both males and females (Parthasarathy 1999). Unemployment among rural educated is high and growing, resulting in the rise of a frustrated youth attracted to naxalism in rural Andhra Pradesh. Author own studies examined the interlinkages between agriculture and labour markets in Andhra Pradesh and India (Reddy and Kumar 2006; Reddy 2010; Reddy 2011; Reddy and Kumar 2011; Reddy and Bantilan (2013); Reddy, 2013; Reddy (2004); Reddy (2006); Reddy (2009a); Reddy (2009b); Reddy (2010b); Reddy (2011a); Reddy et al., (2011)) 18

31 2.3 Factors influencing employment structure These studies are of two types. The first is cross sectional econometric estimates of the relationship between the level of employment diversification and either level or growth of the explanatory variables. The second try to estimate the dynamic association between growth of employment diversification and changes in various macro indicators (Lanjouw and Lanjow, 1995). While analysing NSS regional data (Dev, 1990 and Unni, 1991) stated that there was no significant association between crop output per head of agricultural population and the level of non-agricultural employment. However a more equal distribution of land (Gini concentration ratios) was associated with greater diversification (Vaidyanathan, 1986; Dev, 1990 and Unni, 1991). Bhalla (1993a) while analysing district level data from the censuses for the country argues that the shift of consumer demand in favour of better quality products, together with a switch to urban produced inputs is the decisive factor in districts that have reached high level of agricultural productivity. Another prime mover, which operates from outside agriculture, is the growing influences of urban centers. Workforce diversification occurs in the surrounding countryside (districts) increased rapidly in 1971 and Such diversification was not seen in That proximity to urban centers or urbanisation has an important influence on the growth of rural non-farm sector (Unni, 1991; Jayaraj, 1994 and 19

32 Eapen, 1995). Papola (1992) highlighted the role of small towns in the rural hinterland in the employment of rural workers and in promoting diversification of employment in rural areas. These towns generate employment in rural areas through backward and forward linkages. He observed that productivity and earning levels are higher in rural non-farm enterprises in regions where urban settlements are more widely spread in the rural hinterland, than where there is a concentration of a few towns. These small towns also appear to increase the viability and sustainability of rural enterprises. During 1971 to 1981, small towns with a population size of 20 to 50 thousand experienced the fastest growth and new workers in towns were more likely to come from rural areas. Another factor, outside agriculture, that is observed to influence the nonfarm employment is the level of education of the population. This has been noted at the regional level mainly in studies on Kerala, the most highly literate state in India. The level of education appears to facilitate and enable shifts of the workforce from agriculture to non-agriculture (Eapen 1994). However, it was noted that the educated person in Kerala preferred salaried employment rather than self-employment in the non-agricultural sector (Mathew 1995). A positive relationship between literacy rate and the rural non-farm sector has been observed for Gujarat (Sant 1993), Tamil Nadu (Jayaraj 1994) and Orissa (Samal 1997). An increase of 1% literacy rates in the region would result in 21.9% increase in wages over and above the effect of household own education (Shira klien 2002). In a study about effect of education on productivity of farmers in 20

33 rural India, Surabhi and Kumar (2000) concluded that literacy has a positive and significant relation with crop productivity, hence high returns to investment on education are expected among self-employed in agriculture. Investment in formal education was associated with both higher individual earnings and growing societal wealth. These relationships have been extensively conceptualized and documented by human capital theory which stresses the value of peoples' learning capacities as a factor of economic productivity (Becker, 1964). We are now living in a knowledge society, in the sense that people, in general, are devoting more time to learning new knowledge and skills than ever before. But as this aggregate knowledge increases, the opportunities to apply it in paid workplaces have not kept pace. As a result, the underemployment of educated adults is a large and growing. The notion of " "knowledge-based economy" remains largely illusory for most of the rural population. The individual level relationship between educational attainment and income has remained strong in relative terms. In all advanced industrial market economies, university graduates have consistently earned significantly more than high school graduates since 1970 (see Livingstone, 1999, p. 163). Secondly, some human capital advocates have suggested that declining quality of schooling in terms of relevance of education for employment creation is now the central problem. Additional analyses indicate that underemployment is often greatest among young people and visible minorities among women; all these factors tend to heighten class differences in underemployment. Ultimately, 21

34 educational upgrading becomes a less and less viable means of coping with underemployment (Livingstone 2002). While studying village level data in rural India, Lanjouw and Shariff (2002) stated that education, wealth, caste, village level agricultural conditions, population densities and other regional effects influence the access to employment. Even where the educational and skill levels are similar gender, kinship, caste, tribe etc., remain important determinants of access to employment and the level of remuneration Ghose (1999). 2.4 Gaps in the existing literature While there were a good number of studies that studied the structure of employment and unemployment, but most of them were concentrated in analysing macro-level data i.e., at state and national level. At micro-level very little is known about the nature, extent, pattern and determinants of employment structure, unemployment and underemployment in rural areas. In order to faster non-agricultural activity at the micro level, it would be useful to know how and why workers choose various segments of employment. This would require a micro level analysis at the level of the household or individual. What is the interrelationship among work force structure and poverty and other socio-economic characteristics of households/persons? Why are people in a particular part of the sector/status of employment? How income levels vary in each sector/ status of employment? How a person to be employed or unemployed? These are some of the questions need to be understand for proper micro-level employment generation programmes in rural India. 22

35 CHAPTER -3 Methodology In this chapter, the sampling procedure adopted, techniques and tools used in the analysis of the data are discussed. 3.1 The Data The study used unit data relating to employment, unemployment and under-employment based on quinquennial survey on employment and unemployment conducted by the National Sample Survey Organisation, (NSSO) in its 50th and 55 th round survey on employment and unemployment. NSSO 50 th round data is used to compare temporal changes in employment structure to a limited extent. Schedule 10 was canvassed for employment and unemployment situation for the period and in NSSOs 50 th and 55 th rounds respectively. The geographical coverage in rural Andhra Pradesh is 496 villages, covering 4816 households in 50 th round and 432 villages covering 5184 households in 55 th round Sample Design The state was divided in to four NSSO regions by grouping contiguous districts, which are similar with respect to population density and cropping 23

36 pattern. They are Coastal consisting of 9 agriculturally prosperous districts, Inland Northern consisting 10 districts of Telangana region, which is relatively backward, but rich in natural resources, South Western region comprising only two districts of Rayalaseema region and Inland Southern consisting the remaining two districts of Rayalaseema. But for all analytical purposes, the last two regions of Rayalaseema are pooled together to overcome obstacles in interpreting small sample size and also due to historical, socio-economic and geographical similarities. So our final list of regions comprises Coastal (9 districts), Telangana (10 districts) and Rayalaseema (4 districts). Within each region, each district normally formed a separate stratum. However, if the census population of the district according to the sampling frame used exceeded 2.0 million, the district was split in to two or more strata by grouping contiguous blocks. A stratified two stage sampling design was adopted with census villages as First Stage Units (FSUs) and households formed the second stage units in rural sector. For the FSUs the sampling frame was provided by the 1991 census list of villages. Sample villages were selected by circular systematic sampling (with population as the size variable) from the appropriate sampling frame, in the form of two independent sub-samples. In sample villages, households in the second stage sampling frame were divided in to two strata as follows, households of any sign of affluence such as the possession of any of a specified list of assets including land in excess of a certain specified area, or a member holding a good salaried job or belonging to a paying profession, such as that of a doctor or 24

37 advocate, were designated as affluent households and the top 10 such households (subject to availability) placed in strata one. The remaining households formed second-stage strata two. Two households were selected circular systematically from second-stage strata 1. From second-stage strata 2, ten households were selected circular systematically after arranging the households, in the strata by means of livelihood ; this meant that households belong to Self-employed placed first, followed by rural labour households and then the other. Further, the households under other were arranged in five different land-possessed classes to ensure spread of the sample over households of different economic strata. However in , two households were selected circular systematically from second stage strata 1, from second stage strata 2 only eight households was selected circular systematically Conceptual framework The NSSO surveys on employment and unemployment aim to measure the extent of employment and unemployment in quantitative terms disaggregated by various household and population characteristics. The persons surveyed are classified in to various activity categories on the basis of the activity pursued by them during certain specified reference period. Three reference periods are used in these surveys these are i) one year, ii) one week and iii) each day of the week. Based on these three periods, three different measures namely usual status, current weekly status, and the current employment status respectively were measured. In this study, only current weekly status(cws) and current daily status(cds) were used to analyse status of employment so the 25