Job differentiation and mobility amongst manual workers in Lima, Peru

JOB DIFFERENTIATION AND MOBILITY AMONGST MANUAL WORKERS IN LIMA, PERU By Alison MacEwen Scott Series: Studies in Employment and Rural Development No. 71 Division: Employment and Rural Development Department: Development Economics Development Policy Staff International Bank for Reconstruction and Development This paper was prepared by the author in her capacity as a consultant. The views expressed are those of the author and the report may not be published or quoted as representing the views of the World Bank or its affiliated organization. Washington, D. C. August 1981 1. PREFACE This report is based on researcn carried out in Peru between 1972-74, financed by the Social Science Research Council (U.K.). It was written up mainly at the Employment and Rural Development Division of the Development Economics Department at the World Bank during the first half of 1980. I benefitted greatly from discussions with many individuals during my time in Washington, and would particularly like to thank Mark Leiserson, Richard Webb, Joan Nelson, Nancy Birdsall, Cynthia McClintock and Dipak Mazumdar for their helpful comments on-an earlier version of this report. It is necessary to mention that this research was carried out at a time when the effects of many reforms introduced by the Milita&y Government of post 1968 had not yet worked their way through to manual workers, and the economy as a whole was still enjoying the benefits of the relatively sustained growth of the fifties and sixties. However, since the mid-seventies, Peru has experienced one of the longest and most severe recessions in its history. Care should therefore be taken before applying any of the results of" this report to present circumstances in Peru. 2. JOB DIFFRENTIATION AND MOBILITY AMONGST MANUAL WdORERS IN LIMA, PERU C ON TENTS Preface ............................................. Page 1 Chapter 1 INTRODUCTION AND SUM1ARY ................... 3 Chapter 2 METHODOLOGY ................................ 7 Chapter 3 GRCWTE OF TIE LIMA LABOUR WAREET ............ s Chapter 4 DIFEPETMTIATION AYONGST MANUAL OCCUPATIONS . " 47 Chapter 5 SCCIAL FRATERORK OF JOB MOVMENT ........... 76 Chapter 6 PROFILES OF MOVE1,NT ....................... 89 Chapter 7 CONCLUSIONS ................................ 107 Bibliography ..................o........................ 3. Chapter 1 INTRODUCTION AND SUMMARY The research upon which this report is based was formulated originally as an analysis of the relationship between economic development and working class formation. The main questions had to do with the transformations in the structure of work resulting from changes in industrial organization, the patterns of movement between different types of jobs, and the effects of accumulated work experiences on attitudes, expectations and views of the world. The fieldwork was carried out in Peru betwejn 1972 and 1974 and consisted primarily of case studies of 192 manual workers,(l) supplemented with secondary analysis of survey and census material. This report will present an analysis of job differentiation and patterns of movement between jobs amongst manual workers in Lima, Peru. The main aspects of differentiation relate to enterprise size, employment status and skill. A job change is defined as an alteration in employment situation arising from a change in any of these variables. Job changes can result from movement between and within enterprises, and occur within as well as between occupations. The report has two main objectives: first to portray the degree to which job changes in the manual labour market result in growth in personal incomes and upward mobility in terms of occupation status. Second, to assess the significance of differences in types of enterprise (large or small scale) and employment status (wage labour or self-employment) for levels of earnings and opportunities of upward mobility. Some questions include: does job changing amongst manual workers involve shifting between unskilled jo'bs, or does it involve 'career' -type trajectories of movement towards more highly skilled and better paid jobs? Do workers have a better change of embarking on the latter type of trajectory if they are employed by large firms as opposed to small workshops? Are workers in small enterprises doomed to low incomes and high job instability? Does movement into self-employment involve an increase or a decrease in incomes? This research is relevant to and seriously questions a body of literature on LDC urban labour markets which places emphasis on labour market segmentation (see, e.g., M1iller, 1971). According to this model the urban economy is divided into two sectors, the formal or modern and the informal or traditional. Workers in the formal sector are characterized as having higher productivity, skill levels, wages, job stability and greater fringe benefits than those in the informal sector. The division between the two sectors is based on a number of ()manual work was used as a proxy for working class status. See Chapter 2 for definitions. assumptions: first, institutional interventions (by union action, labour legislation) play a large part in creating privilege in the final sector; employment opportunities are limited by capital intensive technology; and mobility of labour into the sector is restricted by barriers to entry (e.g., screening for high educational levels). Second, high wages in the sector relative to rural incomes attracts 'excessive' migration of workers to the city, who, being unable to find jobs in the formal sector, have to seek or create employment in the informal sector. This is possible because of the ease of entry into informal sector jobs. The influx of labour-there creates a downward pressure on incomes, accentuating the contrast with modern sector wages. This model relies on the concept of a 'surplus of labour' which moves from the countryside into the urban informal sector and is identified in terms of increased numbers.of under- or unemployed workers. Job changes are taken to be a reflection of competition for unskilled jobs, and the scope for raising personal incomes is assumed to be minimal. The emphasis of our research has been to examine the structure of employment at a far greater level of disaggregation than is usual in many of these studies. Detailed data show evidence of considerable variations in income, skill and job duration amongst both groups of workers. In particular, upward movement through different skill levels provides an important way of raising incomes in both formal and informal sectors. More detailed findings are as follows: Findin)s (1) 'orkers in both large and small scale sectors are highly differentiated in terms of skill. Incomes are dispersed in both sectors and skill is a major rea.zon for this. (2) The incomes of the self-employed are not noticeably below those of wage labourers, and have a similar distribution. The self-employed are a heterogeneous group, especially with respect to skill. (3) There is scope for personal income growth and upward mobility within both sectors, and this takes the form of movement through different skill levels. (4) Movement into self-employment occurs at all skill levels, and, depending on the industry, often involves departure from large enterprises. (5) There is more movement between large and small scale sectors than is generally supposed, especially at the lower skill levels. 5. (6) On the whole workers remain in their jobs for long periods, and this -pplies to the small scale sector as well as to larger firms. As far as workers in small enterprises are concerned the level and instability of incomes is more relevant than that of employment. (7) Insofar as there is segmentation in the manual labour market, the most significant dichotomy is not between small and large scale enterprises, but between men's and women's jobs. Within both sectors, men tend to have greater access to skill and capital, and hence to better income levels and mobility chances than women. - The stereotyped image of employment in the informal sector is actually a more accurate description of women's jobs. Imolica.tions Our general conclusion is that the labour market expansion which occurred in Lima did not produce an urban labour surplus of the proportions suggested by the segmented labour markot model, in spite-of .the fact that Peru has a large impoverished peasantry from which such a fund of labour could easily have come. Because of a number of linkages between large and small scale sectors, and the particular structure of the market in Lima, the economic expansion of post-War Peru up until 1974 provided possibilities for growth in both small and large scale sectors. The influx of labour from the interior was rather well attuned to these markets, and does not appear to have concentrated in low paid 'refuge' employment. One possible explanation of this could be that the use of personal networks in the recruitment process has a regulatory effect on the supply of labour, constraining migration until job opportunities actually come up. Another is that workers' preferences in the labour market are not exclusively oriented towards formal sector employment, but also consist of a desire for self-employment. If their information networks are to be efficient in this respect, the discrepancy between expected and actual earnings is likely to be smaller. The fact that the typical 'refuge jobs' turn out to be women's jobs has implications as far as policy is concerned. According to our data, a very small proportion of these women were primary income earners. Insofar as most of these women perform a crucial role in supplementing the household income, the alleviation of their situation might well be best served by stabilizing the incomes of their spouses and gaining better access for women to more highly skilled jobs. 6. This report is organized in the following sections: Chapter 2 outlines some of the major definitions of terms used in this analysis (especially job differentiation and job change). Chapter 3 provides an analysis of changes in the Lima labour market between 1940 and 1972, with special reference to finding evidence of the development or not of an 'urban surplus', income growth and- occupational mobility. Chapter 4 outlines the major aspects of job different- iation with special reference to enterprise size and employment status. Chapter 5 considers some social institutions which affect job choices, and Chapter 6 outlines the main profiles of movement. Finally, the conclusion provides a review of the major findings and suggests some wider implications arising from them. 7. Chaoter 2 METHODOLOGY This research is based on three sources of data: National Censuses, the Ministry of Labour employment survey of 1973, and my own case studies conducted between 1973-74. Before outlining the general characteristics of these data sets, I shall discuss the definitions of terms and concepts which are central to the study, referring where necessary to the different sources. I DEFINITIONS The analysis of job differentiation. requires careful definitions of terms such as 'work', 'job', 'occupation' and 'mobility'. UnfortuMately such definitions are problematic because of the variety of meanings attributed to them by social scientists and by actors within different cultural contexts. Brown (1978) writes: 'There is no agreement amongst social scientists as to the definition and significance of work" and the International Encycloepedia of Social Sciences (1968) warns that "..There is no universally accepted taxonomy of occupations and industries". The papers presented to the 1979 ASA Conference on the 'Anthropology of Work' published in Wallman (1979) amply demonstrate the varieties of meaning attached to folk concepts of work in different societies. Generally speaking economic activities are more easily defined and measured where there is greater influence of the market: their purpose is identifiably 'economic', the relationships are more prone to contractual agreement (written or verbal), and remuneration is clearly specified. However, even in today's ADCs there-are still areas of imprecision (as the iecent feminist literature has pointed out), and the problem is far greater in Third World countries such is Peru, where an inestimable amount of economic activity still takes place outside formal labour and commodity markets. In these circumstances ambiguity surrounds such questions as what constitutes work, which activities are considered as 'jobs' and how such jobs should be classified into occupations. Without particular knowledge of local economies and cultures it is easy for discrepancies to arise between technical and folk definitions, with the risk that data may be distorted and interpretations be inappropriate. Work and Jobs For most social science analysis, the concept of'work is taken to refer to 'employment' in a 'job' where members of a 'labour force' engage in the 'production of goods and services'. However, the problems of measuring participation rates and employment levels in the LDCs are well known. The difficulty arises from the fact that some activities are not perceived as 'work' and some 'work' is not considered to constitute a 'job'. The concept of 'work' has several elements: the physical expenditure of energy on specific tasks, a set of goals, an involvement in relationships with other individuals or groups, and certain effects such as economic returns, social prestige etc. Although it is possible to think of work purely in terms of the task involved, the categorization of an activity as 'work' will depend largely on the wider social context within which it takes place. If the context of a particular set of activities is defined in terms of kinship (eg child labour), religion (eg cultivating sacred fields, faith healing etc) or community (eg community work parties) rather than in terms of personal subsistence, they may not be defined as 'work,'. This ambiguity applies particularly to the activities of women and children, primarily in rural areas, but also to some extent in towns and cities. This problem was particularly relevant to our research:because of the decisions that had to be made about wh.i,gh activities should be considered to form part of the occupational career and which ignored and classified as unemployment or inactivity. It was particularly crucial for the identification of 'first jobs' and the establishment of first entry into the labour market. For many workers this involved working at a very young age on peasant holdings, in informal apprenticeships, or simply accompanying a parent to work. A second problem area was the categorization of women's work in the home, which ranged from identifiable jobs such as dressmaking or running a small shop, through helping a spouse with his job (eg cooking food for food sellers, ironing clothes for tailors), and performing small sporadic services for neighbours (eg laundry and childminding). As far as our own research was concerned, activities were classed as 'work' if they contributed to the pursuit of livelihood of the individual or group of which the individual was a member. But this definition did not produce a clear cut dichotomy between labour force participation or inactivity, for often informants would not declare certain forms of 'work' as 'jobs'. These were activities referred to as'ayuda'and'cachuelos'. The category of work referred to as 'auda' (help) is one of a number of reciprocal labour services found amongst Peruvian peasants (see Mayer 1974 for a detailed classification). It is the most regular of these services and the one which is most frequently found in urban areas. It includes the kind of work contributed by members of the nuclear family to the household and also to some services exchanged between extended kinsmen. The regulation of behaviour in this situation is customary rather than contractual and remuneration is not normally involved. Activities such as child labour and collaboration of spouses, and house construction for kinsmen fall under this category. Cachuelos are best translated as 'odd jobs' or 'sidelines'. Unlike avuda they are essentially remunerated activities, but differ from 'proper jobs' in that they are occasional in.nature. There is a definite sense in which cachuelos are seen to be supplementary, implying that they take up little time, or yield little income, or both. Typical cachuelos involve services performed for the neighbours, eg house construction, repairs, dressmaking, laundry etc. Such activities are considered cachuelos whether or not the individual has a main job. Unfortunately the spanish term for a job is the same as that for work (trabajo) and can also refer to a.trade skill. Workers will talk of having a job (tengo un trabajo) in a contractual sense, referring to a specific arrangement whereby their labour is purchased (if a wage-earner) or to a contract with a client (if self-employed). In the latter case, they will also talk of having work, in the more general sense of there being a demand for their products or services. Finally they will refer to their job in a technical sense referring to their trade. "He hecho el mismo trabajo toda mi vida" (I've done the same job all my life) This usage is particularly common in the building trade where workers pass through a rapid succession of contracts. 10. Amongst these nuances of meaning, there is a concept akin to that of a 'proper job' which refers to remunerated actiities in either wage labour or self-employment, which are regular, routinized and legal, This is illustrated in the way that workers differentiate avuda and cachuelos from proper jobs: "Esto no es trabajo Se' orita, es un cachuelo nomas" (This isn't a job Miss, it's only a sideline). "No tenia trabajo todavia, sino ayudaba a mis padres en la chacra" (I didn't have a job yet, but helped my parents on the farm). Most agencies which have been involved with the production of labour force data in Peru have been concerned to capture some of these ambiguities relating to work. Both the National Censuses aid the Ministry of Labour employment surveys enquire about unpaid family work, and classify it as a 'main job'. However, in both cases a minimum limit of hours worked is imposed (fifteen hours per week) which probably means that some but not all forms of 0i) 'ayuda' are included but many cachuelos would be excluded. Only the Ministry of Labour enquires directly about cachuelos but unfortunately does not use the specific term, referring somewhat clumsily to 'other work'. The inadequacy of the questions on this.topic may explain the extremely low figures reported in their data. In our own case studies, as long as the work was reasonably regular and not referred to as a cachuelo (these being enquired about separately), it was classified as a job. Our occupational careers therefore include unpaid child labour and apprenticeships, but not the more sporadic activities in which women were involved. Jobs and Occupations In both folk and scientific usage, the terms 'job' and 'occupation' tend to be interchangeable since they have in common a reference to the technical content of work activities. Occupations are merely classifications of similar types of jobs with deffering degrees of aggregation. Thus the International Standard Classification of Occupations (1969) defines the The introduction in the 1961 Census of a minimum hourly limit for the inclusion of work in labour force data coincided with a dramatic fall in the participation rate. A debate about whether this was cause or coincidence still rages. 11. respective terms as follows: "Workers whose principal tasks are identical may be considered as having the same type of jobs... An occupation covers jobs or positions held by individual workers". Similarly the British Classification of Occupations and Directory of Occupational Titles (CODOT, 1972) defines a job as "all the tasks carried out by a particular worker to complete his duties", and an occupation as "a collection of jobs sufficiently similar in their main tasks to be grouped under a common title for classification purposes. However, a distinction has to be made between jobs as specific -instances of employment, and occupations as trade skills possessed by workers whether or not they'have a job at a particular moment in time. This distinction may get lost when workers are referring to their occupational histories, unless it is explicitly mentioned that job change rather than trade or occupational change is referred to. (1) This raises the question of how mobility is defined. Sociological mobility studies tend to rely on occupational classifications, which are highly aggregated, hierarchical and based on broad groupings of skills. Because of the way they are constructed, the movement which is identified tends to be unidimensional (skill) and vertical. The high degree of aggregation also means that a large proportion of the movement is lost, some of which might have been just as important in improving material and social status as that which occurs across the boundaries of the broad groups. Economists tend to analyse mobility in terms of changes in jobs and locations. Job change is usually identified as a change in employment contracts or a change in employers. However, this criterion loses much of the movement within enterprises,tend cannot deal with non contractual situations such as that of the self7 employed. In this research, the criterion of job change is more detailed, since the definition of 'job' is more complex, extending beyond the technical function of a wor:ker within an employment contract. For our purposes, a job refers to a particular employment context involvina attachment to a particular enterprise, employment status (wage labour, self-employment), a position within (1) This is a particular problem with conAtruction workers who tend not to rementer each contract they have had 12. the enterprise determined by the internal division of labour and a particular form and level of remuneration. Job differentiation refers to variations in any of these variables. A job change is considered to have occurred when there is any change in either the technical definition of the work task, the position within the firm, the employment status or the employer. Under this definition the discrepancy between job change and occupational change is very large, and increases with the level of aggregation in the occupational classification. As a result the rates of mobility tend to be raised and the patterns of movement are more complex. Manual work It is common in Sociology to use the distinction between manual and non-manual work as a proxy for the division between the working class and the middle class. Although the distinction is blurred at the dividing line, it is considered to be useful since it alludes to differences between types and levels of skills and tends to correlate with variations in education 'and income. It also corresponds to differences in employment contracts, especially regarding the stability and permanence of employment, the form of remuneration, and the provision of fringe benefits.2) Finally, there is some evidence that the types of authority and control over the,two groups of workers differ qualitatively. This division, which in essence refers to the relative proportion of physical rather than intellectual effort at work, therefore underwrites a broad distinction in the position in the labour market and social status of the two categories of workers. In ADCs there are two ways of identifying manual labour, first through the occupational classification which ranks manual labourers in terms of their skill level. Second, through employment status, specifically through the distinction between wages and salaries. In the case of Peru there is no such easy identification. First, the occupational classification which is commonly used in census and survey data does not classify manual workers in terms of their skill but in terms of the sector with which they are typically (1) Mobility = a icb change tather than an occupational change. (2) In Peru, until 1974, there were separate social security systems for manual and non-manual workers. associated. Thus below the level of office worker, the groups are: salesmen, farmers, drivers, production workers, and service workers. Many of these groups contain workers such as policemen, sailors, athletes, divers etc. who would be difficult to classify as manual. The most difficult group to locate is that of salesmen, which places street pedlars and insurance brokers in one group. Secondly, the categories of wage-earner .and salaried worker only apply to persons employed in a formal labour market, and leaves nearly a quarter of the labour force (the self-employed and family labourers) unclassified. Self-employed workers present particular problems since they straddle the lowest and the highest occupations and nearly half of them are employed in the ambiguous occupation of salesman. The following table gives an indication of the relationship between occupational groups and employment status. As can be seen the category wage earner is heavily concentrated in production and service work, while that of salaried worker predominates in the white collar occupations (professional managers and office workers). In this research the manual labour sample was identified by a detailed examination of occupations at three digit level, together with information about the employment status and access to different social security systems. As far as sales workers were concerned, street pedlars and low income retailers and shop assistants were included as part of the manual labour force. 14. TABLE 1. OCCUPATIONAL GROUPINGS AND EMPLOYMENT STATUS, IN LIMA 1967. Emoloyment Status Occupational Wage Salaried Self Family Grouo earner & Employ- Employed Labour Total domestics ment Professionals 1.2% 21.4 % 5.0% - 7.9% Managers - 13.3 4.7 - 4.9 Office Workers 2.8 40.0 0.5 6.8 12.9 Salesmen 3.5 14.9 45.9 48.6 17.8 Farmers 1.1 - 2.8 3.4 * Drivers 5.4 1.4 6.9 - 4.9 Production workers 55.1 4.5 25.1 28.9 32.7 Service workers 30.8 5.0 9.1 11.8 16.8 100.0 100.0 100.0 100.0 :100.0 N. 739 491 362 76 1,668 from Gianella, 1970, page 52, Table 218, * below 0.5% II DATA SOURCES This research relies on three sources of employment data, whose main features are set out below: (i) National Ponulation Censuses. Population Censuses were carried out in 1940, 1961 and 1972. All three gave coverage of data on occupations, economic sector, employment status (wage labour, self-employment etc) and personal characteristics of the labour force such as age, education and migration. There are slight changes between the Censuses regarding definitions, forms of classification, and levels of aggregation of the data. The 1940 Census considered the unemployed to be part of the inactive population, and did not impose a minimum limit of hours worked as -a criterich 'for labour force participation. Its definitions of self- employment and owners of businesses also differs from the other two censuses. The 1972 Census incorporated a new classification of economic activities recommended by the United Nations which places a large number of repair jobs formerly classified as part of manufacturing into the services sector. Also, it provides some data only for the population of 15 years or over, a practice which was not followed 'in the previous Censuses. The Census data are used in this research mainly for the analysis of changes in the labour force over time. Censuses are the only source of such data since employment surveys were only started in the late 60's and vary greatly in their coverage. The Cdnsuses are also appropriate sources of information since they span the major period of economic growth in Peru's recent history. 1940 was a period of relatively slow growth and little industrialization, while 1961 marks the middle of a period of massive foreign investment, buoyant exports and substantial industrial expansion, and 1972 marks the beginning of a period of significant political change. The discrepancies between the Censuses make for some difficulties in the establishment of comparability. However this has been overcome partly by reclassifying data (which was possible because of the highly disaggregated form in which the 1972 Census was published), and by working with higher levels of aggregation. In particular this has meant basing the tables for Lima on the Departmental rather than Provincial figures, which results in the inclusion of a small rural population. 16. (ii) The Ministry of Labour Urban Employment Survey (1973) The Ministry of Labour has been carrying out regular employment surveys since 1965. It has a national sample frame based on a multi-stage stratified area sample, which as far as Lima is concerned contains some 3,OC households. The survey of 1973 covered many aspects of labour market structure and movement relevant to this research. VariableG included population characteristics (sex, age, household composition etc), migration, formal education and other forms of training, characteristics of principal job (enterprise size, economic sector, occupation)employment status, recruitment, income, unionization, social security), second jobs and other sources of income, unemployment and occupational history (first job and other jobs during the last ten years). The Lima part of thi_s national survey covered 2,379 households and a population of working aqe (i.e. of fourteen years and over) of 7,812. The employed part of this population consisted of 3,588 persons, of which 2,377 were working in occupations defined by the criteria outlined above (1) lbu oc ewe to be manual . The distribution of this manual labour force between economic sector, employment status and sex is given in chapter 4 (tables la -c). In general the quality of data in the Ministry of Labour survey are good. There is sensitivity to the complexities of employment data in the construction of the questionnaires, and the interviewers are well trained and experienced. Possible weaknesses which should be mentioned here are firstly, a tendency toward under-reporting of incomes due to the failure to distinguish gross from net earnings and the lack of data on overtime, and secondly, an underestimation of job movement due to the confusion between 'jobs' and %occupations'. (1) This subset drawn from the survey population excludes unemployed persons. 17. (iii) Case Studies. More detailed data on job differentiation, occupational mobility and attitudes were collected in 192 case studies selected from the Ministry of Labour survey for re-interview. The emphasis in the selection of these cases was on adequate, rather than representative , coverage of key jobs. Therefore an equal number of cases (12) is represented in each of the cells in the matrix (1) above. This permits a typological analysis, but caution has to be taken in making generalizations about the frequency of such patterns in the general population. For the latter purpose, the Ministry of Labour survey data is used. The case study data was collected with the use of formal interview schedules organized into two parts. The first covered the detailed aspects of jobs (enterprise characteristics, technical and social relations of production, mechanisms of recruitment, terms and conditions of employment, remuneration, attitudes, preferences, and 'views of the world'). The second part consisted of the 'Life History' schedule, a format which permits chronological coverage of job movement, entering the major characteristics of eac,,L job (i.e. enterprise, employment status, occupation, recruitment, income, employment contract etc). This format considerably expands the amount of detail covered in the biography and encourages maximum memory recall. It is worth meritioning that only two major discrepancies were discovered between the data in these case histories and the originals in the Ministry of Labour Survey. The methodological style of this research has been to generate ideal typical profiles of jobs and job change from the case study data, and then estimate their frequencies inferentially from the more representative sample data. National Census data are used to provide some indication of trends over time. In this report the main patterns of differentiation and mobility will be discussed ideal-typically on the basis of case study data. The estimation of frequencies in the sample population will be produced in subsequent analysis. (1) The sample of case studies was stratified by economic sector, employment status and sex in order to provide comparison groups. Recause of the small number of cases in some cells, a representative selection would have given too few cases to provide any basis for analysis. 18. Chanter 3 GROWTH OF THE LIMA LAEOURb MARKET Economic background Peru can be considered one of the intermediate economies of Latin America in that it still has a relativelv large peasantry and a level of industrialisation which produces most durable consumer goods but is still heavily reliant on imr>orted inputs for industrial production . Table 1 gives some crude indicators of its level of development compared with other countries. TaLe 1. C<UDE U 1CATO~S OF DE'lELOPMEI!T, PERU /;D OH1IEPR SELECTED C3Z7IES .Io3.l C'? per V u e :. Z . 2 lanu,.r r i Hcuschold incc-' Pope acian --pic si auc- force in Co-ve in by by C- .: (=Il.2n') (.:$ 1977) curir.. (1975 primary sc:ardary 0owCsC hgigese ad pute i i iu tuc u.wtur 2G • 10 5.2 630 209 51 23 n.a. n.a. ¿J.ujLr 7. 790 420 47 24 n.a. å. ite 1o.4 840 40 20 1.9 42.9 Cclc bla 24.6 720 1,636 31 23 n.a, n.a. Cnile 10.6 1,160 1,86Ö 21. 27 4.4 34.8 26.0 1,730 8,635 14 29 4.4 35.2 a.ico 63.3 1,123 11,633 34 25 2.9 36.7 cc:il 116.1 1,3c0 17,312 42 20 2.0 50.6 Uniced Kingdom 55.9 4,420 35,004 2 43 6.3 23.5 Japua 113.2 5,670 91,770 14 37 7.9 27.2 Unzd s2ace 220.0 å,520 264,195 7 33 4,5 26.6 Source: Wird Devlnrn Indicacors, The World Eank, Juna 1979. DaeL- vury: Paru 1972, hile 1968, Argena 1970, Mexico 1977, Braz-il 1972, United Kingdom 1973, Japan 1969, Uniced Scacaj 1972. (1) 40% of the labour force in 1972 was emploved in agriculture; approximately 70% of rurdi lies were working on small peasant farms (less than 20 heetåz_) and 40% had less than 2 hectares of farinland (World Bank 1979 -304) (2 In 1969 durable consumer goods represented 3% of imports, while capital and intermediate goods for industry represented 72% (National Accounts 1973). 19. Partly as a result of an unusually diversified export package, Peru has had a sustained pattern of growth for most of this century, although short run fluctuations have affected particular products and regions. Exocrt growth has been especially marked since the 1950s and has been accompanied by high rates of industrial growth due to export processing and immort-substitution. Until the change in political regime in 1968, foreign capital was dominant in most sectors of the economy. There has been a substantial imbalance in the regional impact of growth. Broadly speaking levels of development have corresponded to the three major physical and demographic divisions of the country. Until 1940 the majority of the population (65%) lived in the mountainous region (sierra) on peasant farms or estates. The development potential of this area has been severely hindered by systems of land tenure(2) and a lack of natural soil fertility, and growth has mainly been centred around mines. Export agriculture (sugar, cotton and rice) developed on the coast, a desert strip which depended on irrigation for development. Expansion of economic opportunities due to agriculture, fishing, some mining and trade has resulted in a shift of population from che sierra to the coast, and its share of total population rose from 28% in 1940 to 461 in 1972. The Amazonian jungle has as yet little economic pctential (with tne exception of some possibilities of oil) and presently has about 10% of the population. . Peru has a long history of interregional migratiocn, much of which was of a temporary or seasonal nature and was directed towards plantations and mines as well as towns and cities. Since 1940 however, the volume of migration has increased substantially 3 , has become more permanent and is primarily directed towards urban centres. The proportion of the population living in centres of 2,00 people or more has risen from 27% in 1940 to 53% in 1972, and the proportion of urban population in towns of over 20,000 has risen from 55% to 76%. (1) Manufacturing output rose at approximately 7% between 1951-76 and its share of GNP rose from 14A in 1950 to 22% in 1972. During this period real GDP averaged about 5.5% per annum and per capita GDP doubled in spite of an increase in population growth (World Bank 1979 p 163A). (2) Land tenure was changed dramatically after the Agrarian Reform of 1968, althouc. this affected patterns of ownership rather than the distribution of land between larce and small holdings. According to a survey in 1970, 27% of the population were migrant, and a further 27% had lived Asewhere than their birthplace at some previous time. Taking these two figures together, over half the population have had some excerience of migration. (artinez 1973, p 18 & 65) 20. The rates of urban growth have been high in various marts of the country averaging 4.2% per year between 1940-61 and 5.6% between 1961-72, but they have been even higher in Lima where a disproportionate share of growth has been concentrated. In 1972 Lima accounted for about three quarters of industrial employment and output, 78% of commercial bank credit and a high proportion of public sector expenditure. Its share of the national population grew from lot in 1940 to 25% in 1972, and its share of the total urban population rose from 36A to 44%. Average annual growth rate of the Lima population was 5.1% between 1940-61 and 5.4% betweeh 1961-72. Real incomes in almost all sectors of the economy rose between 1950-66, but relative inecualitv widened both between social groups and regions. According to Webb's data for 1961, the average income of rural peasants in the sierra was just under half that or urban wace-earners and less than a fifth or urban white collar workers; rural incomes were about a third that of Lima incomes (Webb 1977 pp 10-11). Between 1950-66 the growth in real per (1) capita income of urban modern sector wage-earners was nearly double that of workers in the urban traditional sector and six times that of the poorer peasants. (P. 39). More recent Census data (1972) shows that the average blue. collar wage in Lima was 2.4 times the average wage in agriculture. Expansion of the Lima labour market Between 1940-1972, the population of Lima rose from 645,172 to 3,302,523, much of this increase being fed by immigration. The poportion of immigrants in (2' Lima rose from 33.4% to 41.4% and the net immigrant population ( which was barely 200,000 in 1940 rose to over a million in 1972. Although Lima has lower mortality and fertility levels than the rest of (3) the country , the hilh proportion of migrants who arrive at the beginning of their reproiuctive cycle has been pushing up the natural rate of increase in Lima's (1) Modern sector = all establishments employing five or more workers. (2) Net migration (immigrants minus emigrants) from the coast and sierra regions to Lima were 190.8 in 1940; 683.7 in 1961; and 1,194.6 in 1972 (thousands) (Calculated from World Bank 1979 o. 87) (3) The crude death rate per thousand in the 1969-72 period was 10 for Lima comparec with 23.9 for villages with a population of less than 2,500, and 18.7 for the whole country (Bagan, cited by World Bank 1979, p 65). The total fertility rate (average no. of children per women at the end of her reproductive age) in Lima in 1961 was 4.6 compared with 6.1 in the sierra and 6.0 for the nation (World Bank 1979, p 63) (4) 36% of the migrant population in 1970 was aged 15-29 compared with 24% of Lima born. (calculated from Martinez 1973, p. 35). 21. population and changing its demographic structure in favour of younger age groups. An indication of this is the rise in the total dependencv ratio from 36.O to 41.2t over the oeriod. The economically active pooulation in Lima ) has grown at an average rate of 4.5% per year in the 1940-1972, reachtna 1,077,335 by the latter date. The rate of growth in the labour force is lower than that of demographic increase largely because of a fall in the participation rate from 40.5 to 32.6(2) The age and sex specific participation rates set out in Table 2 show that these changes have been concentrated in the younger and older age groups, and explanations that have been offer.d include increased educational enrolments and better financial arrangements permitting early retirement (World Bank, 1979, pp 75-77). It is worth noting that Lima has a more highly educated population than- other areas of the country, and a greater concentration of educational establishments, especially at higher education level.(3) Table 2. AG a SEX SPSCIIC PARTICIPATIO-N RATES, METMOPOL.I"IAVI LI.MA', 19-10, 1961, 1962. 1940 1961. 1972 AGE I Total . F Total M F Total Pq 6 - 14 5.6 5.8 5.4 2.a 2.5 3.2 1.6 1.5 1.6 15 - 19 46.3 58.2 34.0 43.4 50.6 36.2 27,4 32.4 22.7 2 - 29 64.8 88.3 39.2 65.7 91.0 39.1 60.1 82.5 37.0 30 - 44 69.2 '95.3 3.5 65.1 98.7 29.5 64.8 97.4 30.7 45 - 64 64.0 92.2 33.9 58.3 93.4 22.8 56.1 89.6 22.3 65 * 49.2 80.7 27.7 32.7 60.7 10.2 24.8 44.9 7.7 TTAL 48.1 66.3 28.7 44.5 64.6 24.1 32.3 46.9 17.6 T ETAL 3AP 272,037 264,566 107,471 817,481 597,351 220,130 1,225,371 892,600 332,771 Sex 1 dis1ribution 00.0 71.1 28.9 100.0 73.1 26.9 100.0 72.8 27.2 'The age distribution of the labour force 1s not available at the provincial level, therefore this table qLves data for the Oepartment of Lima, plus Calao. The age categories are established from the 1940 Census, Cata includes unemployed. Source: National Censuses, 1940, 1961, 1972. (1) Population of 6+ years in work or actively seeking work in Lima (provicce of Lima and Constitutional Province of Callao) (2) Note that this downward trend in narticipation rates has occured throughout the period, and cannot thus be attributed to the definitional. changes between the 1940 and 1961 Censuses. (3) School enrolmont does not always result in a withdrawal from work. In 1972 almost half the 14-19 age group in the labour force were continuing their education (World Bank, 1979, p 74) 22. Because of the different age structure of the migrant population (61% of migrants in 1970 were of working age, compared with 39% of the Lima born, Martinez 1973 p. 155), migrants form a much larger proportion of the labour rorce than ,they do of the population as a whole (69.3% compared with 41.4%), and their predominance in the manual labour force is even greater (73% according to Martinez(1973 p. 178) 72% according to the 1973 Ministry of Labour survey). There appear to be two main streams of migration to Lima. The first, which contains roughly two thirds of the cases involves migration off young people from rural areas and agricultural employment in direct moves and accompanied by relatives from the nu,Iear or extended family. The educational level in this stream is generally low, although it is higher than (1) that of those who stay behind. The second stream involves migration from larger towns and cities in the interior,a slightly older age group and much higher levels of education (only 3-4% illiteracy, over 20% with completed secondary education and 12-15%-with some higher educationY. 10% of these migrants had- experience of professional or white collar work prior to migration (Martinez 1973, p. 200). One implication of this difference in the origin and skill level of migrants to Lima, is that their location in the Labour market on first arrival and their opportunities for upward mobility differ widely, although these variations appear to even out with time (see below). For a very large proportion of male migrants in the first and biggest stream, the conditions for economic improvement(i.e. continued education, apprenticeships, accumulation of savings etc) are all provided in the Capital: The situation for young girls however is rather different, most of them are destined for domestic service and their opportunities for upward mobility through the labour market are very low (see below). (1) In 1970 almost half the migrants to Lima were under 15 years of age, and nearly two thirds were under 25. 59% of the total migrant population and 61% of the migrant labour force were from small villages and settlements, and of those who had worked prior to migration, 54% were in agriculture, fishing and mining. As regards education, 10% were illiterata and 30-40% had incomplete primary schooling. Only 8-12% had secondary or higher education.. (Martinez 1973). 23. The structure o: the Lima labour force. Between 1940-72 there were 815,859 net additions to the Lima labour force which by the latter date stood at just over a million. Without any doubt, the certiar, sector (utilities, transport, commerce and services) absorbed most of this increase (61%) , and this is a common phenomenon in Third World cities. Superficially, this is often taken to mean that employment expansion consists mainly of street pedlars and shoe shine boys who contrihute nothing to productivity and create a social nuisance. In this section, I shall present a number of arguments which are intended to modify this view. Briefly they are that tertiary sector empLoyment has been as much due to an expansion of white collar as to manual work, that the rate of increase in manual tertiary occupations has been declining rather than increasing, and that the capacity of the secondary sector (manufacturing and construction) to exoand employment has been much underestimated. Finally, the general structure of the Labour force in terms of the sectoral distributions and the emoloyment categories within sectors has changed remarkably little, in spite of the fact that the Labour force has quadrupled in the period under consideration. Table 3 ECONOMICALLY ACTIVE POPULATION BY ECONOMIC SECTOR, METROPOLITAN LIMA (1) Sector 1940 1961 1972 Agriculture & fishing 24,793 9.5% 35,975 5.3% 36,932 3.4% Mining 863 0.3 3,839 0.6 4,208 0.4 Manufacturing 50,980 19.5 156,720 23.0 251,356 23.3 Construction 16,750 6.4 45,791 6.7 77,428 7.2 Utilities' 1,177 0.5 4,144 0.6 3,083 0.3 Transport 17,115 6.5 40,970 18.1 83,754 15.5 Commerce 34,222 13.1 123,082 6.0 167,116 7.8 Services 85,437 32.7 219,554 32.3 379,024 35.2 Unspecified 8,331 3.2 50,047 7.4 74,434 6.9 (Unemployed)(3) (21,808 8.3) TOTAL 261,476 100.0 680,122 100.0 1,077,335 100.0 Source: National Censuses (1) Province of Lima & Constitutional Province of Callao. (2) Sub categories in manufacturing, commerce & services have been relocated for comparability between Censuses. (3) Unemployed persons are not included in the 1940 labour force. In 1961 and 1972 the unemployed are included and classified according to the characteristics of their last job. Aspirants are located in 'unspecified'. 24. Table 4 ACCUMULATED RATES OF GROWTH OF LIMA LASOUR FORCE SY ECONOMIC SECTOR 1940-1961 1961 - 1972 Sector accumulated Net accumulated Net increase rate of growth increase rate of growth Agriculture & fishing 1.8 11,182 0.2 957 Mining 7.4 2,976 0.8 - 369 Manufacturing 5.5 105,740 4.4 94,636 Construction 4.9 29,041 4.9 31,637 Utilities 6.2 2,967 -2.7 -1,061 Transport 4.2 23,855 6.7 44,034 Commerce 6.3 88,860 2.8 42,784 Services 4.6 134,117 5.5 159,470 TOTAL 4.6 418,646 4.3 397,213 Source: table 3 Taking the last point first, Table 5 shows that any changes in the relative distribution of the different sectors has been small: the services sector has increased slightly from 32% to 35% but this has been offset by a decline in the share going into.commerce. There has,been a'rise in the> share of manufacturing employment but only by 4%. The sectors whose shares have changed most s:tgnificantly are relatively small; agriculture has declined from 9% to 3%, and transport has increased from 6% to 15%. Let us now examine the structure of the tertiary sector. Taken as a whole (utilities, transport, commerce and services) the sector absorbs a large proportion of the total labour force (over half) and has provided 61% of the net additions between 1940-72. However, the overall increase in this share has been relatively small (rising fron 53% to 59%) and has been largely due to an increase in service jobs rather than in commerce. In fact, the share of employment in the commercial sector has almost halved, and its rate of growth, especially in the second period has been one of the lowest (see Table 4). Within the tertiary sector, services occupies around 60% of the labour for=e, a proportion which has not changed between 1940-72. It alone has accounted for more net additions to the total labour force than any other sector, just over 36%, This change has been primarily due to an increase in white collar work, especially in the public sector. Table 7 shows that salaried 25. work now accounts for half of employment in the sector, having risen from 39% in 1940. On the other hand, the proportions of wage-workers and domestic servants have fallen from 51% to 36% in this sectsr. Indeed, domestic se_vice, which traditionally has been considered to form the bulk of service employment has dropped from 12% to 81 of the total labour force, from 38% to 23% of service employment and from 34% to 26% of total female employment. On the other hand, while the secondary sector (manufacturing and construction) did not increase in absolute terms as much as the tertiary sector, it did expand both absolutely and proportionately. Between 1940-72, the manufacturing sector was the next largest sector to grow after services, and its share of the total labour force grew from 19.50% to 23.3%, adding about a quarter of the total net increase. The growth rates of both the construction and manufacturing sectors have been between 4.4% to 5.5% over the period, about the same as the rate of growth of the services sector (Table 4). It should be noted, however, that a good part of the increase in secondary sector employment was also due to the expansion of white collar employment. (Salaried labour rose from 10% to 28% of manufacturing employment, and from 5% to 14% in the construction sector, see Table 7) Some of this increase may reflect the upgrading of manual occupations to white collar status in order to provide access to a superior social security system, but most of it is due to the expansion in clerical and technical occupations, associated with the growth of large scale firms. The main argument here is that changes in the relative sectoral distributions of the labour force have been affected more by changes in the rate of expansion of white collar than manual jobs. Table 5 gives a summary of those changes, in which it.can be seen that the share of &alaried.emoloyment has risen from 19% to 361 of the labour force; it has been by far the fastest growing category of employment and has accounted for 44% of the total net additions to the labour force. In comparison, although wage work has increased absolutely, its share of total employment has fallen from 50.9% to 39.2%, and the proportion of the self employed has remained at 22% throughout the period. 26. Table 5 DISTRIBUTION OF LA.BOR FORCE ( OCCU?ATIONAL CATEGORY IN METROPOLITAN LIMA, (2) 1940-72 Net Increase Increase 1940 1972 1940-72 1940-72 Self employed (3) 79,396 22.7 251,852 22.S 172,456 2177 Family Labour 16,772 4.8 13,957 1.2 -2,815 -17% Wage Work and Domescics 177,630 50.9 438,550 39.2 260,920 147% Salaried E=ployces 67,761 19.4 406,731 36.3 338,970 500% Unspecified 7,516 2.2 9,200 0.8 1,684 220% TOTAL 349,075 100.0 1,120,290 100.0 771,215 221 (1) Economically active populacicu of six years and over, excluding unemployed. (2) Departmenc OE Li. plus Coasticucional Province of Callao. 3 Independents and patrons. are grouped together because changes in the respective definitions between 1940 and 1972, Another way of appreciating the growth of non-manual occunations is to compare the distribution of employment between occupational groups, which unfortunately is not available for 1940. TablelO shows that with the exception of managers and executives which declined between 1961-72, the top occupational groups have had the fastest rates of growth. Even excluding the problematic category of salesmen, these top white collar jobs have accounted for about one third of the net increase in employment. Many of these jobs are located in the services sector (84% of professionals and technicians and 39% of office workers are employed there). A second point is that although the absolute growth in manual occupations far outstrips that of non-manual employment, the kinds of changes in the structure of the manual labour force which would have resulted from a disproportionate increase in 'marginal', 'refuge' or informal sector employment have not taken place. As can be seen from Tables, and 3 the proortions of the self-empLoyed have only expanded dramatically in small sectors of employment, which have a small share of total self employment (construction and transport). The relative share in commerce which was already large Prior to the onset of urban and industrial expansion has remained stable and in the manufacturing sector has declined. 27. 6 .* ..., .......... a~~n • :n' :.or. ::ic ie n cn c . -!,,c .1__'._______'::o r'r . l:, i-o':; n e ieT r -o- ei u . :: -· cur..i7 1':, 129 1,972 1,3 503 1r 51 n2- S..i2 171 10 2,124 541 1- 2 .u!ee i 1 7:5i 40,7 6,176 20 Cor.ructio ::18 16,937 983 -, ':n 3 22 70 1 4,912 4 20 . Co:-:erc2 17 1,21 4, 17' 16 ,6046 157 39, 59 arv1vi 3e6 4 7,661 35,901 2 033 92,25 c.c. c":11ifi à 916 2 2, ?94 1,279 4 437 - 79 1,772 177,050 67,761 7,516 309,-75 Source: National Census 1940 .e:gi-..c o, t:. .. ;:; s:conOlic ec sr and il( ceu at2c.n:n O eor (4) Fef Far:iily -a'e Work alaried Uisc- -1 Total up oyei Lanour J)or.,cstic, ,mv.ioyecs ci fem (- JI A,:-," culture -4,9- - ~ 5,277 .-50,762 7, 476 ·481 1337 108 ,234 1. n .77 1 , G, 27 33 a:uractatrin- 36 -1 1,542 120,791 62,907 012 95 223 ,619 `cnbruc i ion 1~,198 184 49,968 10,222 191 290 73,053 ranLsport 24,374- 283 36,24,3 24,493 351 361 6,103 co1..i::ercC 100,376 4,146 34,226 90,161 048 1577 2 , crvices 2,848 41,140 126, 1,9 173,916 4996 631 346,162 Si n.cci~ i.c 1, (8 C;02 7,494i 3? ,8~1 1276 68 41 j,. -14 1,259 9 7 7,741 1,653 . 208 1173~ S L251, 82 U 97 4 38 , 50 40' 731 92001,10,2 Source: National Census 1972 .cw.o:lijc]y 1. Liv2 an~ Lrtion of six yoers Jr.d over, exci- cingc nx l-o'-. Uh: 1Y,0 censu5 doe :.CL - ive (Uoe itributio:. -t L'c level.n r oro tr ; fi c'::t.f rr!e tn the ähol.o '.1c nrt.entt of Li;-.a Ž.l :r':.v ncr: o` j:kiLAo, rtthe th u just i,eLo:o,itr:; .mi - r e(' ~l' E i :.0.; ion of Le fi: ro i . r i cul tu r a 1 1 1 (4) The 1972 census does not give data on employment status by economic sec-cor for workers between 6-14 yrs, 28. Table7 , Jini.rik"L..on c: . , .ri Zc caor .-u:r Cicsemtionn1 Cnter;ory Der't. LiN:, pnrd - Callino 1940 (1) :.ily aUe iork Salaried Unspecified Total yloycd Laour oamestic Ez2:lo-en, A-ricult"re 35.5 13.4 49.2 1.3 0.5 100.0 vi.ninz £.o 0.5 74.1 12.9 0.5 100.0 a:Uzacturin2 21.2 1.3 66,8 10.2 - 0.5 1c0.0 Corstruction 4.J C.1 29.9 5.2 0.2 100.0 TrOnSPOrt 17.1 C.4 57.8 24.5 0.2 100.0 -o' rce 43.. 3.1 10.6 42.1 0.4 100.0 Lervices 6.5 C.4 51.7 38.9 2.2 100.0 Lnspecified 9.4 2.1 29.7 13.1 45.6 100.0 j:-'. 22.7 4.2 50.9 19.4 2.2 100.0 Source: Table 6 Aistributicn of t:-:c ''.. :~co.ic .'Mector an.d ccuatic.1 Cate'ory Deut Li:a nd Callao 1972'' 15÷ * Hel£i 2naAe ':ork Salaried Unspe- 6-14 Total E=ployed Ladour Dom.estics X:ployees cifiad A;riculture 39.6 4.9 46.9 6.9 0.5 1.2 100.0 Vining 2.9 0,2 '59.0 37.1 0.4 0.4 100.0 anufacturir.j 16.4 0.7 54.0 28.1 0.4 0.4 100.0 Construction 16.7 0.2 68.4 14.0 0.3 0.4 IO.O Tr:nsport 22.3 0.3 42.1 28.4 0.4 0.4 100.0 Co:.:rcrce 43.4 1.8 14.8 39.0 0.4· 0.7 100.0 Services 9.5 0.3 36.5 50.2 1.4 2.0 100.0 f'nu.ccified 2.4 1.2 17.2 75.4 2.9 0.2 100.0 C-i, 10.7 7. 66.0 C 14.1 1.8 •00.0 WQ ý'a22.5 1.2 39.2 36.3 0.3 1CC. C Source: Table 6 (1) :.ice ::o t s 7- t:ablen; 3 :. d 29. Table 8 Percentage distribution of the EAP by Economic Sector and Occupational Category, Dept. Lima and Callao, 1940 Self Family Wage Work Salaried Employed Labour Domestics' Employees Total Agriculture 47.4 84.2 29.3 1.9 30.2 Mining 0.2 0.1 1.2 0.8 0.8 manufacturing 16.1 4.5 22.7 9.1 17.3 Construction 1.1 0.1 9.5 1.5 5.4 Transport 4.3 0.4 6.5 7.2 5.7 Commerce 21.8 7.3 2.4 24.6 11.3 Services 7.9 2.1 26.8 53.0 26.4 Unspecified 1.2 1.2 1.6 1.9 2.8 Total 100.0 100.0 100.0 100.0 100.O, Percentage Distribution of the EAP by Economic Sector and Occupational Category, Dept. of Lima and Callao, 1972. Self Family Wage Work Salaried Employed Labour Domestics Employees Total Agriculture 17.0 37.8 11.6 1.8 9.7 Mining 0.1 0.1 1.1 0.7 0.7 Manufacturing 14.5 11.0 27.5 15.5 20.0 Construction 4.8 1.3 11.4 2.5 6.5 Transoort 9.7 2.0 8.3 6.0 7.7 Commerce 39.9 29.7 7.8 22.2 20.6 Services 13.0 8.2 28.8 42.8 30.9 Unspecified 0.4 3.6 1.7 8.1 3.9 6-14 0.5 6.2 1.8 0.4 Total 100.0 100.0 100.0 100.0 100.0 Source: Table 6 30. 1ble ga Net increases in emoloyment 1940-1972 Dept. Lima and Callao Self amily ':la'e 'ork alaried .ris7e- Tot a z ,oyed La'cur )ozestics 2:eloyees cified culture 5,255 -8,852 -1,210 6,153 -22 2,691 xin2 66 - 2,752 2,522 23 5,396 .anufac turing 23,855 734 80,518 56,731 524 163,366 constructior 11,339 166 33,031 9,239 155 54,220 ransport 20,952 213 24,649 19,501 303 66,059 Cor:crrce c,034 2,915 30,051 73,315 691 191,783 Services 26,3 4 794 78,73 138,015 2,963 253,907 Us:ecified 132 298 4,600 31,556 -3,161 33,793 16-14 (1,259) (7) (7,741) (1,65) (208) '~'.T 172, 456 -2S15 20, 920 33, 970 1 ,64 771,215 Source: Table 6 Table 9i. Percentage increases in em*ployment in EAP 1940-72, Lima-Callao Self Fand.ly Wage work SaIaried erployed La'oour Domestic- Employees Total Agriculture 16 -63; -2% 465%3 Q3 130Q-' 466% 188% M rif acturing 187% 103% 200, 919 271 Construction 1320". 922' 195) 9L0. 288 Tr2nsport 612' 30Le 213- 399% 330 Commerce 479" 237% 72Q% 442 485 Ser7ices h205 229% 165- 384% 275 Total* 217% -17% 1U7 500: 221 * includes unspecificd and 6-14 ye,r olds, Source: Table 6 31. Table 1o Occupational Grouns Metropolitan Lima 196L, 1972 -- Absolute % Acc. 1961 1972 Increase Rate of Growth Prof. Technicians 42,199 124,981 82,782 LO.37 Managers, Admin. 24,801 11,318 -13,483 - 7.39 Office Workers 81,346 147,786 66,440 5.57 Salesmen & Pedlars 83,085 150,245 67,160 5.53 Farmers, Fishermen 33,753 35,405 1,652 0.44 Miners 777 701 -76 - 0.94 Drivers 33,031 40,199 7,168 1.80 Artisans & Workers 202,586 328,406 125,820 4.48 Service Workers 121,370 172.062 50,692 3.22 Unspecified 57,174 66.232 9,058 1.34 Total 680,122 1,077,335 397,213 4.27 1961 Sub- 19.72 Sub- M F Total M F Total Prof. Technicians 4.6 10.4 6.2 10.3 14.8 11.6 Managers, Admin. 4.5 1.4 3.6 1.4 0.2 1.0 Office Workers 10.6 15.4 12.0 11.6 19.0 13.7 Salesmen, Pedlars 12.7 11.1 12.2 14.3 13.1 13.9 Farmers, Fishermen 5.9 2.5 5.0 4.2 0.9 3.3 Miners 0.2 - 0.1 0.1 - 0.1 Drivers 6.7 0.1 4.9 5.2 . 0.1 3.7 Artisans & Workers 35.2 16.0 29.8 37.7 12.3 30.5 Service Workers 10.1 37.5 17.8 9.2 33.0 16.0 Unspecified 9.5 5.6 8.4 6.0 6.6 6.2 Total 100.0 100.0 100.0 100.0 100.0 100.0 Source: National Censuses 32. Another indication of the slow growth of' refuge'occupations is manifested in the absolute decline in family labour, which is often seen as the counterpart to self-employment, necessary to the latter's survival in the city. While doubts concerning the reliability of data in this area must be admitted, the fall in numbers is registered in all three censuses, and is consistent with the fall in participation rates and the rise in school enrolment ratios, mentioned above. In summary, the changes in the sectoral composition of the labour force which occurred in Lima between 1940-72 have been affected more by changes in the proportions between manual and non-manual labour than by an increase in so-called'refuge'employment within the manual workforce. The sectoral distributions within the wage labour force show marginal increases in manufacturing employment but otherwise remain broadly stable (Table8 ). While it is true that self-employment has expanded largely in the sphere of commerce, this has not affected its total share of commercial employment. This suggests that the size of the so-called 'urban labour surplus', and its accompanying assumption of income stagnation and occupational immobility, has been greatly exaggerated,. By implication, it would appear that the scope for upward social mobility and income growth may have been larger than previously supposed. The rest of this chapter will attempt to assess the evidence for these latter processes. Income distribution in Lima There are various sources of data on income distribution in Lima which vary slightly in their methodologies and coverage. . (See Webb 1976 for a comparison and critique of different sources). The figures given below are . derived from the Kinistry of Labour urban survey (1973) and refer to individual earnings from the main job. Income z::rm other sources such as odd jobs, property income,transfers are excluded. rhe income declared by the self-employed is supposed to be net of expenditure on production, although in practice this calculation is not precise. 33. The occupational classification which is used to present the income data is slightly different from the one normally used by Peruvian agencies. This was revised in 1973 in preparation for the analysis of occupational mobility, with the objective of making both the structure within groups and the overall basis of classification more consistent. The most important changes were the creation of a new white collar group of 'technicians' who had formerly been grouped with 'nrofessionals', a refinement of the division between skilled and unskilled production workers, and the removal of white collar salesmen to office workers. Table Ilgives the distributions of the labour force between these new occupational groups and the median monthly income of each group. These figures suggest a high degree of overall income inequality, which is reflected in the Gini coefficient of 4.7, calculated from the data. It may be noted, however, that the marginal differences between the median incomes in the middle of the hierarchy are relatively small, while at the opposite ends there are more marked differences.(1) This indicates considerable variation in income within the respective manual and non-manual groups, and a relatively small gap between them. Table 12 gives further evidence of the variance in incomes within groups, and of overlapping between them. This is particularly the case for office workers, drivers and skilled and unskilled production workers, between 30-40% of which groups all earn between 3,000-4,999 soles. Without access to the original data for regression analysis it is difficult to suggest reasons for this variance. Other research suggests that education, age, sex and productivity may be amongst the most significant variables. One implication of.this would be that incomes could rise considerably through an individual's lifetime without vertical occupational mobility. Vertical mobility might increase incomes in the long run, provided an individual worker had the right education and worked The incomes of service workers are particularly low because of the high proportion of domestic servants in the group, whose income includes a non-cash element (food and accommodation). Farmers and fishermen are a very heterogeneous group, and therefore are excluded from these observations. 34. Table 11 Occuoational GroursMetropolitan Lima 1973 No. 1 Men Women Median income (soles per nohth) Professionals 155 3.2 3.9 2.2 8,737 Managers 176 3.7 5.2 1.4 8,395 Technicians 327 6.8 6.6 7.2 5,606 Office Workers 709 14.7 13.9 16.0 3,885 Drivers 222 4.7 7.7 0.2 3,646 Skilled Production Workers 1,269 26.6 33.9 15.7 3,203 Unskilled Production Workers 99 2.1 3.4 0.1 3,101 Salesmen 733 15.2 12.2 19.7 2,483 Service Workers 1,004 20.7 10.1 36.4 1,791 Farmers & fishermen 108 2.3 3.1 1.1 3,290 Total 4,802 100.0 100.0 100.0 * includes unemployed Table 12 OccupationalGrouos By Income Level Metropolitan Lima 1973 1,000- 3,000- 5,000- 7,500- 10,000- 15,000- 0 2,999 4,999 7,499 9,999 14,999 + Total Professionals 1 5 16 21 14 11 32 1007. Managers 3 7 13 24 13 16 24 100% Technicians 1 12 28 29 15 9 6 100% Office Workers 1 20 34 24 10 7 *4 100% Drivers - 21 45 17 8 6 3 100% Skilled Workers 3 42 33 15 3 3 1 100% Unskilled Workers 3 44 39 4 4 3 3 100% Salesmen 6 60 21 6 2 3 2 100% Service Workers 2 73 16 6 2 1 0 100% Farmers & fishermen 8 41 18 6 11 10 6 100% Source: Ministry of Labour 1973 urban survey 35. in the right firm. In the short run, and without the right personal attributes, occupational mobility might provide little gain in economic situation. Individual incomes can of course rise because of a general increase in demand, and this appears to have been the case in Lima. The evidence of growth in incomes in Peru since the Second World War is fragmentary. The one comprehensive national study conducted by Webb (1977) only covers the period up to 1966 and does not specifically deal with the Lima Population, Other evidence for Lima comes from studies on partial populations, e.g. shanty towns. The conclusions from Webb's study was that (a) there has been some growth in all urban incomes over the period 1950-1966, but that (b) the incomes of modern sector workers had grown faster than those of the traditional sector resulting in an overall deterioration in relative income inequality. The figures are given below. Table 13 Trends in per caoita real income 1950-1966 Average income Annual income growth per capita (US $) 1950-1966 Modern Sector (1) 1,126 4.1 Wage earners 4.9 Government employees 3.6 White collar 3,3 Urban traditional(2) 407 2.0 Wage earners 2.5 Self Employed 1.9 Salaried workers 1.8 Domestic Servants 1.6 All Peru 433 2.1 (1) establishments with 5+ employees (2) establishments with less than 5 employees Source: Webb 1977 pp 10 and 39. 36. Alden Lewis (1973) has shown that the average income of households in shanty towns studied between 1956 and 1967 increased in real terms by about 33.5%. The extent of house construction in the shanty towns and the gradual upgrading of these areas in terms of their physical amenities bears witness to this income growth, and many writers have given illustrations of the process of upward mobility experienced by squatters (Mangin 1970, Turner 1968), It is possible that Lewis's findings cannot be generalized to the entire manual labour force. He does point out that the growth in incomes has been differential and that incomes in the tenth percentile of his sample were stagnant in real terms. Furthermore, it could be argued that some of the extreme poor live in the inner city slums rather than in the barriadas and therefore would not have been represented in Lewis's data. However, it is clear that the numbers which might have experienced stagnant or declining real incomes have been a fraction of that assumed in the conventional literature. According to Lewis, almost two thirds of the growth in incomes has been due to a general increase in the wage rate; only 12% was due to a shift into white collar work and 26% due to an increase in the average number of persons employed per household. This suggests that the increase in demand, while expanding employment and raising incomes, has not promoted widespread mobility between occupations l) We shall now examine some of the evidence on this aspect. Education and occuoational mobility Since one of the major constraints on upward occupational mobility is education, particularly because of the use of screening practices in job recruitment, I shall briefly review the educational profile of the Lima labour force. This is presented in Table 14. (1) Lewis refers only to movement into white collar work; he does not deal with occupational mobility within manual occupations. 37. -able 14 Occupational Grouos Ev Educational Level Metropolitan Lima 1973 Illiterate Primary Secondary Higher Total Professionals - 2 7 91 100% Managers - 14 49 37 100% Technicians - 7 38 55 100% Office Workers - 12 66 22 100% Drivers 52 43 5 100% Skilled Workers 1 50 46 2 100% Unskilled Workers 4 52 43 1 100% Salesmen 8 51 37 3 100% Service Workers 6 62 29 2 100% Farmers 10 60 26 4 100% * Less than 0.5% Source: Urban Survey, Ministry of Labour 1973. The education profile shows more bunching than the income distribution, which indicates that education is likely to be a better predictor of occupational status than income, although there is also some variation in educational levels within occupations. The educational levels in general are high compared with the rest of the country, and compared with some other Third World capitals. The level of illiteracy is extremely low and the proportions with incomplete primary education are also relatively small. Farthermore, around 40% of manual workers have some secondary education, and around 15% have completed secondary education. Amongst non-manual occupations the proportions with higher education rises with occupational status reaching its highest point amongst the professionals, 91% of whom have had some further education. 38. Table 15 Occuoational Groups by Educational Level, Lima 1973 I Incom- Comolete Incom- Comnlete- Incom- Complete Illit-10lete primary., plete second- plete higher Total erate :-- -!I e primary second ary higher Professionals - - 2.0 2.2 5.1 17.1 73.6 100.0 Managers 1.2 2.5 11.0 16.5 31.8 14.2 22.8 100.0 Te,hnicians - 1.8 5.0 12.2 26.0 39.7 15.3 100.0 Office workers - 2.8 9.1 17.9 48.0 16.1 6.1 100.0 Drivers 1.0 12.6 39.7 28.7 13.5 4.0 0.5 100.0 Skilled workers 1.2 18.0 32.3 30.8 14.9 2.5 0.3 100.0 Unskilled workers 4.1 26.3 24.8 28.4 15.1 1.3 - 100.0 Salesmen 9.0 30.6 21.3 21.4 15.4 1.8 0.5 100.0 Service workers 6.8 36.3 25.5 21.3 8.3 1.7 0.1 100.0 Farmers 10.5 1 24.8 35.6 17.0 9.0 1.5 1.6 100.0 Source: Suarez, 1975, p.24 Table 16 Exclusion effects of screening for a comnleted cycle of education Occupational % without % without % without % without Group literacy comDleted completed completed primary secondary higher Professionals 0.0 0.0 4.2 26.4 Managers 1.2 3.7 31.2 77.2 Technicians 0.0 1.8 19.0 *84.7 Office workers 0.0 2.8 29.8 93.9 Drivers 1.0 13.6 82.0 99.5 Skilled workers 1.2 19.2 82.3 99.7 Unskilled workers 4.1 30.3 83.6 100.0 Sales workers 9.0 39.6 82.3 99.5 Service workers 6.8 43.1 89.9 99.9 Farmers 10.8 35.3 87.3 98.4 Source: Table 15 39. The broad division between manual and non-manual work is evident in this table: a markedly higher proportion of manual workers do not go into secondary school compared with non-manual workers (rough percentages being 50-60% compared with 7-14%), and only a negligible proportion go into higher education. A more disaggregated analysis shows that while many workers do not complete the cycle of education on which they had embarked, numbers concentrate in the year of completion rather than in any previous year. This suggests that there is a premium on completion of educational cycles, and that employers are screening for this. Since the completion rates are lower in the unskilled occupations, presumably this tendency is less important there (See Table 15). If one examines the proportion of any occupational group which would be excluded from employment by the requirement of a completed cycle of education then the broad skill divisions of the population emerge clearly (See Table 16). Screening for completed secondary schooling would create a major barrier to entry into the professions, and into white collar work in general, excluding four fifths .of the workforce. Screening for completed primary education would create a barrier into skilled manual jobs but would exclude a smaller proportion of the labour force. The fact that some workers manage to enter occupations without these qualifications indicates that there is some return to incompleted cycles of education, and that the impact of education in occupati=nal recruitment varies in different parts of the labour market. Earlier in this chapter we have observed that (a) there was a massive expansion in employment at all occupational levels, (b) there wa' also an expansion in the educational system which led to declining child labour and increased enrolment ratios; (c) a large part of the increase in the labour force was due to immigration and most migrants had low educational levels. The questions which we now pos are the following: how far did the expansion of the labour market generate a process of upward occupational mobility? In which occupational groups was mobility greatest? What are the principal mechanisms or obstacles to mobility, and finally how far did immigrants move up the hierarchy? 40. Uo until the present, there has been no comprehensive analysis of occupational mobility in Peru, and the only source.which exists is a study of intra-generational mobility in Lima, based on the Ministry of Labour urban survey of 1973. This analysis consists of comparisons between the first and up to four previous jobs compared with the current one. It contains no information on the occupation of informants' parents nor their children, and therefore nothing can be said about intergenerational mobility. The data presented in the Ministry of Labour study exclude those who are still in their present job or who have ceased working since their first job. The patterns of movement thus refer to only those currently employed, who have experienced job changes, which is less than half the labour force. Care should therefore be taken before extrapolating genral mobility rates on the basis of these data. The includion of both groups would sharply reduce the amount of overall mobility. Our general concern in presenting this information is to indicate the directions of movement achieved through job changing and to discuss some of the variables which seem to be important in affecting this. For the main part of the analysis the occupations have been aggregated into four broad groups: professional and managerial workers (1), technical and clerical workers (2); drivers and-skilled production workers (3), and unskilled workers, sales and service workers and farmers and fishermen (4). Table 17 compares the distribution of first and second jobs in the four broad occupational strata. It indicates a general movement of population into higher-skilled occupations, resulting in a decline in the proportions in the lowest stratum. Tabl 17 Chanqes in the occupational structure between first and present job Occupational Strata First job Present Job Outflow Inflow Professional & 3.7 7.0 2.5 13.8 Managerial Technical & Clerical 19.0 21.4 15.9 24.2 Skilled manual 26.7 31.9 25.4 43.2 Unskilled manual 50.6 39.7 56.2 18.8 Total 100.0 100.0 100.0 100.0 Source: Suarez 1975 p. 3172 Tables 10 & 11 41. Table 18 is one of a number in the study which Dortravs the movement between first and present jobs in terms of projection, i.e. the rate and direction of outflow from the -first job. For example, while around 70-73% of those who were in groups 2 and 3 in their first job are still there in their present one, 11.6% of those originally in group 2 moved up into the professional group, while 16.71 of those in group 3 had moved dow-n into the unskillqd group. The broad trends indicated in Table 18 and others in Suarez's study show that the rate of movement out of most broad occupational strata varies between 20-40%, being higher in the lower occupational groups. Most mobility out of particular strata is destined for adjacent groups and this is both upwards and downwards. One effect of this type of 'short span' movement is that the overall flow between manual and non-manual occupations is low and typically takes place between lower white collar and skilled manual occupations. Only 8% of former professionals and managers and 15% of former technicians and clerical workers are now in manual occupations, while 12% of skilled and unskilled workers are now in non-manual jobs. Another effect is that considerable upward and downward mobility exists within both manual and non-manual groups. This is particularly striking amongst manual workers. There is a marked difference in the mobility patterns of men and women: women are more stable than men, have less experience of upward mobility and more of downward mobility. Upward mobility is positively associated with higher levels of education (Suarez, p 70). However, amongst manual, workers vocational training clearly plays an additional role offsetting a lower level of formal education and expanding an already high one.(1) The effect of age is somewhat unexpected. Amongst most groups, but especially skilled manual workers, it appears to increase the chances of downward rather than upward mobility. Only amongst technicians and white collar workers is there a tendency for upward movement amongst the older workers, (1) For these workers vocational training takes the form of apprenticeships and on-the-job training rather than in formal institutions. Note however the relatively small proportion with vocational training: 38% of the total labour force and 29% of the manual workers. 42. but there is also an equal tendency in the ooposite direction. For unskilled workers and women, age affects mobility chances neither postively nor negatively. Table 18 Mobility between first and present job, by sex First Job Present Total S E X Job Men Women Professional & 83,3 83.6 82.3 Managers (1) (2) 7.6 6.4 11.8 (3) 4.2 5.4 - (4) 4.9 4.6 5.9 Technicians (1) 11.6 16.5 3.3 Office Workers (2) (2) 73.5 63.3 91.0 (3) 9.5 13.1 3.3 (4) 5.4 7.1 2.4 Skilled workers (1) 2.9 3.1 1.8 Transport (3) (2) 9.7 10.3 6.6 (3) 70.7 71.0 69.5 (4) 16.7 15.6 22.1 Unskilled Workers (1) 3.5 4.6 1.6 Salesmen, Service 9.2 12.0 4.7 Workers (4) 24.3 35.6 5. (4) 63.0 47.8 88.0 Source: Suarez 1975, Table 32, o.61 * Vertical columns within each row sum to 100% 43. As far as migrants are conce.rned, it seems that white collar and skilled manual workers experience a drop in occupational status on first arrival to Lima. However, this recovers quickly and within five years the mobility patterns ace not dissimilar from those of the Lima born. (Suarez o 68). This means that for the vast prooo.:tion of migrants who are relatively unskilled on arrival, there is a 40.50% chance of movement into higher skilled occupations. Martinez' data shows more clearly the gradual increase in the proportions of migrants in the upper occupational groups, although the indication is that the process takes longer. Table 19 Occuoational Distribution of migrants, by years in Lima, 1970 Prof, Office Sales Prod- Service Other Total Years in Tech. workers workers uction workers Lima Managers workers O - 1 2.5 7.8 7.8 20.0 59.3 2.6 100 2 - 9 9.4 12.0 14.2 34.1 29.4 0.9 100 0 - 19 10.1 13.6 17.3 41.3 16.3 1.4 100 20 - 29 I18.31 12.5 20.7 35.1 .13.0 0.4 100 30 + 23.9 9.6 15.3 33.2 15.0 3.0 100 born 19.7 19.1 14.1 32.9 11.9 2.1 100 born Source: Martinez 1973, p. 198 and 178 The followihg matrix gives a more disaggregated picture of occupational movement. The data arranged in terms of extraction, which shows the rate and origin of inflow into current jobs. It indicates the degree to which particular occupations have recruited by'direct entry'(i.e. from persons entering the labour market for the first time), or from other occupations. Low figures along the diagonal indicate a high rate of inflow from other jobs, and high figures, the reverse. All the numbers to the right of the diagonal concist of downward movement, and those on the left upward movement. The lowest figures are to be found amongst managers and drivers, both of which are located at the top of the respective non-manual and manual groups. 叫 45. As far as managers are concerned, a large part of the recruitment comes from other non-manual occupations, but 42% also were originally manual Workers. In fact, recruitment into this occupation constitutes the main access to non-manual work for manual workers. As far as drivers are concerned, oart oZ the inflow comes through a natural link between Primary production and wholesale transport, and the other major source is from skilled work. It is significant that self-employment Plays an important role in both of these occupations which suggests that access may be determined as much by canital and entrepreneurship as by education The second two occupations which have a relatively high rate of inflow are those of unskilled workers and salesmen, in both of these cases ( ownward mobility from. skilled work is a significant source of labour, which re-in-fcrces the point made earlier about the circularity of movement amongst (2) manual workers. on average, job changes result in occuoational mobility for around a third of the labour force. if the population still in firs" jobs or retired from them were included in these figures, the rate of a"Ctual mobility at any one time -would fall by.at least 50%, and the rates. within particular occupations might fall more if they had larger than average shares of such workers. However the Potential mobility would reamin the same if it could be assured that those currently in their first job might embark an similar patterns of movement as those who had already changed jobs. If downward mobility were made to cancel out vertical mobilit y, the percentage of total actual upward movement would be further reduced but not eliminated. However, bearing in mind the dispersion of incomes within occupations, one should not assurne that downward occunational mobility necessarily involves a deterioration in economic circumstances or a diminution in social status. The move into self-empLoyed commerce for example, which attracts many skilled manual. workers, is usually seen to mark the final stage of successful career consolidation. (1) Note that many drivers set themselves up in business with indemnity payments from skilled factor-y, work. (2) There is.the possibility that some of this movement is the result of ambiguities in coding and classification. This is particularly applicable to managers and farmers which are both very heterogeneous groubs. 46. As regards the direction of movement, three different patterns of mobility seem to emerge from these data. First, there is a trend towards net uoward mobility based on short span moves, which draws in the migrants (as well as the Lima born) from the bottom of the occupational structure and distributes them into higher skilled and better paying jobs. Second, there is a circulatory pattern within the manual workforce involveing both upward and downward mobility. Age appears to pily an important part in this process. Third, there is a major tendency for stability within occupations. In the upper occupational groups with the exception of managers, this suggests that labour is being recruited directly from educational establishments. Since education is usually aquired prior to entry into the labour market, mobility opportunities here are likely to be confined to the next generation. Although there is more mobility out of the lower occupational groups, the existence of a small 'traped' group suggests the presence of barriers to entry or supply constraints. The concentration of women in these groups suggests some possible lines of enquiry. The general conclusion of this chapter is that economic growth in Lima and Peru between 1940-72 massively expanded the labour market, resulting in income growth for most of the'labour force and upward occupational mobility for some. The opportunities for economic and occupational advancement amongst migrants are considerable, although the 'span' of such, processes is short and few of them will move out of manual occupations. The process of income growth has been differential and the mobility selective, leaving at the bottom of the labour market a group in trapped occupations with stagnant incomes. The size of this group however is considerably smaller than that suggested in the conventional literature. The rest of this report will investigate these processes of differentiation and selectivity. 47. Chapter 4 DIFFERENTIATION AMONGST MANUAL OCCUPATIONS In the last chapter we have tried to show that the amount of income growth and occupational mobility in the Lima labour force up until 1973 had been much greater than that portrayed in the literature on LDC urban labour markets. This was so amongst manual as well as non-manual workers, although a small group at the bottom appeared to be trapped in low income occupations. However, the true extent and pattern of these processes is largely hidden by the highly aggregated and clumsy classification of occupations which is used in the presentation of these data. This is particularly unsatisfactory as far as manual workers are concerned since it fails to deal adequately with skill differentials, being based instead on the economic sector within which jobs are located. Although there is an element of skill implicit in the sectoral classification (e.g. service jobs tend to be less.skilled than manufacturing jobs), this is far too vague for any precise measurement of mobility. The purpose of this report is to show that a more highly differgntiated analysis of jobs enables one to detect a variety of patterns of movement, each with a different logic of purpose and constraint. -Three general patterns of mobilitv are referred to in chapter 6 as 'escalators', 'oscillators' and 'dead ends'. The most general of these is the first which consists of a process of deliberate upward movement into better jobs. Such movement is on the whole gradual and slow, and involves a progression towards higher levels of skill and income. However, other criteria are also relevant. Case study data indicates that while workers clearly strive to maximize skill and income in making decisions about job changes, they also feel strongly about the degree of control they exercise in their work and value self-employment as a desirable (1) condition of emplovment. A move into self-employment may therefore be seen as a 'better job' even where it does not result in an increase in income. In this research, we have given emphasis to the subjective evaluations used by workers in obtaining employment, and therefore the criteria indicated (1) Given a choice between wage labour and self employment, 72% of the Case study informants preferred self-employment. After 'a better income', 'greater independence' was the second biggest reason given for changing jobs. 48. by them have played an important role in establishing the basis for analysing job differentiation. The main ones which we shall draw on in this report are enterprise size, ..,ill, Pm-ployment status and income. 'irst however, we shall qive a brief overview of the aeneral structure of the manual labour sampDle using the conventional categories. General features of occunational and income distributions amongst the manual labour sample (a) occupational structure Tables la - Ic show the distributions of the sample of manual workers by economic sector, employment status and sex. Compared with the aggregate distributions for the entire Lima labour force, this subset shows some interesting differences. First, regarding the sectoral distributions, there are much higher proportions of manual workers in manufacturing and commerce, but less in the services sector (compare Table 3, Chapter 3) . Second, there is a higher proportion of self-employed workers: 33% as compared with 22% (see Table 5 Chapter 3) . Finally, there is a sliqhtly higher proportion of women (33% comoared with 26%). Table la Samole distributions of manual workers by economic sector, employment status and sex. _ Men _Women_ Sector ~ ~_~ITotal Wage Self Sub Wage Self Sub ILabour Employed' Total Labour Employed Tota Manufact-uring1 505 12637 100 42 i42 779 Construction 1117 52 1619 1 170 ITransport115 7 9 Tr 72187 - - 187 1Commerce 1281 197 1325 53 177 230 555 Services 240 19 I259~ 335 84 I 419~ 6789 Total 1105 472 1577 489 303 792 2369 Source: Ministry of Labour urban survey 1973 49. Table lb Relative distribution of manual workers by economic sector, emnloyment Status and sex. Sector men J Women Total Wage Self Sub Wage Self Sub Labour Employed Total Labour Employed Total Manufacturing: 21.3 5.6 26.9 4.Z 1.8 6.0 32.9 Construction 4.9 2.2 7.1 * - * 7.1 Transport 4.9 3.0 7.9 - - 7.9 Commerce 5.4 8.3 13.7 2.2 7.5 9.7 23.4 Services 10.1 0.8 10.9 14.1 3.6 1 17.7 28.6 Total 46.6 19.9 66.5 20.5 12.9 33.4 100.0 * less than 0.05% Source: Table la Table lc Relative distribution of male and female manual workers by economic sector and employment status. Sector Men Total Women Total Wage Self Wage Self Labour Employed Labour Employed Manufacturing 32.0 8.4 40.4 12.6 5.3 17.9 Construction 7.4 3.3 10.7 0.1 - 0.1 Transport 7.3 4.6 11.9 - - - Commerce 8.1 12.5 20.6 6.7 22.3 29.0 Services 15.2 1.2 16.4 42,3 10.6 52.9 Total 70.1 29.9 100.0 61.7 38.3 100.0 Source: Table la 50. The distributions of the different sexes between sectors are markedly different: 40% of the men are in the manufacturing sector, and a further 11% in construction, bring the total in secondary sector employment to over half. Only a fifth are in commerce and less still in services. On the other hand, women concentrate overwhelmingly in the teitiary sector, (82%) with over half in the services sector alone. Less than a fifth are in manufacturing, and there are virtually none in construction and transport. This concentration of women in a particular area of the labour market is not restricted to economic sector, but is even more apparent at a more disaggregated level. Women participate in far fewer occupations than men overall, and are a minority in most of them. In 18 out of 32 manual occupations listed, the porportion of female workers did not even reach ten per cent. Women concentrate in three single occupations (dressmaking, retail sales and domestic service) and in two of these, tend of form the majority. In the garment trade, women typically make women's and children's clothing while men's clothing is the preserve of tailors. Within the retail trade, women tend to concentrate on certain lines and locations (Moser 1977, 1980), and within personal services there are also important differences in the work considered appropriate to the female sex (e.g hairdressers, boarding house keepers, laundry etc.). As the next chapter will show, this segregation in the labour market has profound implications for the mobility chances of women and tends to confine them to low income jobs. Part of the explanation can be sought in market forces which separate women out as a separate category of labour. But there are also cultural factors such as the 'sex labelling' of jobs which underwrite demand and supply in the market, (b) Income distribution (1) The figures on principal job earnings reported by the Ministry of Labour show a high degree of overall inequality as already mentioned in the orevious ,chapter. The top 1% of working people had earnings that were 25 times that of the lowest 10% and ten .times the minimum wage. However, the earnings distribution (1) All data on incomes refer to net earnings from the main job in the week or month preceeding the survey. There is a possible tendency towards under-reporting of earnings in these data due to the fact that wage-workers seem to report their take home pay net of deductions but without overtime bonuses, and the self-employed are vague and conservative in their estimations. (However, this may be coipensated by inadequate allowances for depreciation). The income figures for the entire EAP havc been adjusted for comparability with the sample population, which involves excluding unem- ployed persons classified with zero incomes. 51. of manual workers is also highly unequal, as is indicated in the fact that the Gini coefficient only falls to 4.1 from 4.7 when non-manual workers are excluded. In fact, since the former group represent two thirds of the total, it is likely that inequality amongst manual workers contributes significantly to overall inequality (see Figure 1). The distribution of manual earnings is highly skewed (see figure 2). A consequence of this is that the majority earn very low incomes, with 43% below the minimum wage. The distribution of female earnings is even more skewed and the prooortion below the minimum wage is almost double that of men. The discrepancy between male and female earnings is also idicated in the median income figures of both (Table 2); note also that the female median is well below the minimum wage of 2,400 soles. One possible explanation of the generally low level of female earnings would be the large proportion of domestic servants who receive part of their income in kind. However, this is only part of the story, for two thirds of the women are in other occupations and, as we shall see below, the earnings of women are lower than those of men in every occupation and sector of employment. Table 2 presents another interesting result: there is very little difference between the median incomes of wage labourers and the self-employed.(2) Figure 3 shows that the earnings profiles of the two groups are also remarkably similar. Furthermore, as regards the sectbral distributions, it is noticeable that in some cases the self-employed actually earn more than wage labourers, (in the case of men, in manufacturing, transport and commerce, and in that of women, in services). In general average incomes are higher in manufacturing, construction and transport than in commerce and services. This is especially true for men, regardless of their employment status, and almost two thirds of them are to be found in these sectors. Amongst women, the highest median is to be found in manufacturing, (wage labour), but this affects a bare 13% of them and the mean is well below the equivalent for men. On balance therefore, men tend to concentrate in the higher income sectors of the labour market, while the reverse is true for women. (2) Strictly speaking this comparison may not be legitimate, since the incom7s of the self-employed are not solely a return to labour but also to capital and entrepreneurship. However this distinction is academic in the case of many self-employed who use little capital. j 合 54. Table 2 MEDIAN MONTHLY E.IRNINGS, LIMA 1973 (in Soles) N 2,326 Male Female Total 9.3,010 0.1,289 0.2,556 Wage labour Self employed %,2,579 $.2,499 MALE FEMALE Economic Total Sector Sector W. Labour S. Employed W. Labour S. Employed Manufacturing N = 498 N = 131 N = 100 N = 40 N = 769 0.3,010 0.3,438 %.2,408 Z.1,287 %.3,009 Constiluction N = 1.16 N = 51 N = 167 .4,128 .3,010 %.3,869 Transport N 115 N 70 N = 185 S. ,930 S.4,502' Z.4,002 Commerce N = 126 N = 192 N 52 N 164 N = 534 Z.2,407 Z.2:999 Z.2,013 S.1,291 %.2,150 Services N = 23 - 6 N = 19 N z 330 IN = 80 N = 665 .2,800 ,2,000 %.1,000 ".1,506 0.1,499 TOTAL N = -1,091 N = 463 N = 482 N 284 N = 9,320 .3,010 %.3,010 9.1,200 .1,290 %.2,556 Note: Minimum wage level at this period %.2,40,0 55. * . ,L~TDÆ MiST Pulou 67 flLe N & ~J ~ ___ 0 ur, 0,,j-.. --.t v 4 2 4 wd >Z 1 - - -[ † *- -- - - - ...-- ... . • .. . - - - - - - - - - -- - --. ---. - - - - - * - • - - - .-kT 6 17z4 d- PrM--Y -k9y' .- - - m-.- ka-----7. - . .- ~~- -.--. . -- -- .... - -- - ... -. -- - - ----.- - - - ---- - - ------,- - - -iA-. ty' - -sj -At- 56. Table 3 MEASURES OF CONCENTRATION Ai:D DISPERSION OF MONTHLY EARNINGS' LIMA 1973* = arith-metic mean = co-efficient of variation (st-andard deviation) (arithmetic mean ) en Womer Wage Labour Self employed Total = 3609 := 1703 x = 2908 x = 3116 x = 2975 = 0.64 v = 0.88 v = 0.7. v = 0.83 v = 0.72 ~ ~ Men Women Economic sector -- Wme W. Labour S Employed W. Labour S. Employed Total Sector Manufacturing x = 3663 x= 3967 x = 2502 x = 1759 x = 3463 v = 0.59 v = 0.67 v = 0.45 v = 1.00 v = 0.64 Construction x = 3690 x = 4044 - - x = 3806 v = 0.39 v = 0.71 v = 0.52 Transport x = 4568. x = 5051 ..x = 4749 v = 0.55 v = 0.60 v = 0.57 Commerce x = 2378 x = 3458 x= 1824 x= 1878 x = 2558 v = 0.65 v = 0.71 v = 0.94 v = 0.96 v = 0.83 Services x = 3180 x = 2149 = 1229 x = 2166 % = 2060 v = 0.67 v = 0,60 v = 0.74 v = 0.95 v = 0.88 Total x = 3509 x= 3846 x= 1563 x = 1942 x = 2975 v = 0.61 v = 0.70 v = 0.76 v = 0.96 v = 0.71 * Excludes earnings over 14,999 (1% of sample). 57. The distributions of earnings within all categories are highly dispersed, and this should caution us against placing too much emphasis on measures of concentration. Coefficients of variation rarely fall below 0.5 and in many cases they approach unity. The highest co-efficients are to be found amongst women and the self-employed, and in sectors such as commerce and services which have large concentrations of both (see Table 3). The dispersion of incomes in these data indicates the need for greater disaggregation, particularly between high and low wage jobs and between successful and unsuccessful self-employed. The variables detailed below provide the basis for such disaggregation, and where possible their impact on earnings data will be discussed. In some cases however, this will not be possible because of unavailability of data, and illustrations from the case studies will have to be relied upon. Although income forms an important part of job differentiation, other elements which affect mobility chances in a more general way, are included in the'discussion which follows. I. Enterprise Size. Throughout all sectors of the Peruvian economy enterprises are highly polarized both in terms of their workforce size and in terms of a number of other variables such as labour productivity, gross value of production ,profitability and market concentration. Although the degree of correlation between these variables differs in specific industries, the dualities are sufficiently consistent for it to be possible to talk of a large and a small scale sector, crudely indicated by workforce size. The following tables are illustrative of this dualism: (1) (1) Unfortunately there has not been a comprehensive economic census in Peru since 1963 (the 1974 one referred onle tsmns bAe imrin), ho 1t ou. not easy to obtain data on establishments based in Lima without recourse to unpublished material. 58. Table Nn. 4 Manufacturing Sector in Peru - distribution of establishments labour force and qross value of production hy workforce size No. of Labour Gross value of production Workforce size establishments force (million soles) Less than 5 persons 21,126 83.4% 30,607 15.8% 575 1.8% 5 - 19 2,734 10.8 25,596 13.2 1,916 6.1 20+ 1,461 5.8 137,224 71.0 29,001 92.1 Total 25,321 100.0% 193,427 100.0% 31,492 100.0% Source: First Economic Census 1963 Preliminary Results. Table No. 5. Retail Commerce in Peru - distributions of establishment, labour force and value of sales by workforce size No. of Labour Value of sales Workforce size establishments force (million soles) Less than 5 bersons 56,076 96.0% 14,128 34.3% 4,966 39.0% 5+ 2,363 4.0 27,043 65.7 7,762 61.0 Total 58,439 100.0% 41,171 100.0% 12,728 100.0% Source: First Economic Census 1963 The polarization of workers between small and large enterprises shows up clearly in the Ministry of Labour data: 60% are in the group with 1-4 workers, and 19% in the group with 100 and more. Table No. 6. Distribution of the manual workforce by size of establishment Workforce size Men Women Total 1-4 711 47.8 649 83.7 1360 60.1 5-9 133 8.9 40 5.2 173 7.6 10-19 99 6.6 14 1.8 113 5.0 20-49 97 6.5 13 1.7 110 4.9 50-99 70 4.7 15 1.9 85 3.8 100+ 379 25.5 44 5.7 423 18.7 Total 1,489 100.0% 775 100.0% 2,264 100.0% Source: Ministry of Labour urban survey 1973 (missing data = 113) 59. Note that a very high peoportion of women are concentrated in the very small enter-rises (84s) and that barely 6% are in the largest ones. On the other hand, one quarter of the men are employed in the latter group and a much smaller orocortion are in the small enterprises. Figure 4 gives the profile within each sector, showing a repeated pattern of nolarization. The concentration of the workforce -in small enterprises is highest in commerce and services, particularly in the former where the proportion reaches 84%. In manufacturing, transport and construction, there is a higher proportion in large scale enterprises. For further analysis, the enterprise size categories have been collapsed into a two sector classification. The decision as to where to make the cut .off point is fairly arbitrary. Conventional practice in Peru has been to take establishments with five or more employees as the dividing line between 'modern' and 'traditional' sectors (e.g. We0b, L977, Fitzgerald 1976). Reichmuth (1978) places it at 10 employees, but has an additional classification which divides each sector into two subcategories. In this study we have placed the division at 20 workers on the grounds that it is at this size that an enterprise would be more likely to participate in formal economic institutions (1) and possibly to have a union . Whether or not the limit is set at 10 or 20 workers, the numbers involved are not very large. Table 7 shows the distribution of occupations between large and small scale sectors, and by sex. Note that very few occupations belong exclusively to one sector or another with the exception of retail and ambulatory sales, hairdressing and domestic service, which are overwhelmingly concentrated in the small scale sector.(2) Most occupations straddle both sectors, although some predominate in one sector rather than another.(3) This applies equally to skilled occupations such as textile workers, electricians and mechanics, and to unskilled work such as cleaners, porters and caretakers. (1) 20 is the minimum number of workers required in law before official recognition of union is permitted.- (2) Households employing domestic servants are classified on the basis of the number of servants employed. The self-employed are classified as one-man enterprises or, if applicable) on the basis of the number of workers employed. (3) Occupations with over 60% of workers in the small scale sector include drivers, garment workers, shoe makers, carpenters, bricklayers, painters, jewellers, metal workers, food processors, retail sales and personal service occupations. There are far fewer occupations with over 60% of workers in the large,scale sector: these include postman and messengers, rextile workers, radio and television engineers, packers and labellers, and stevedores. 引 Table 7: DISTRIBUTION OF OCCUPATIONS BY ENTFRPRISE SIZE (SECTOR) AND SEX Male Female Total SSS LSS SSS LSS -SS S LSS SSS Female Xerox operators - 3 - 3 Postman, messengers 6 6 15 28.6 11.8 Drivers 114 79 1 - 11-5 79 59.3 0.5 Textile workers 12 110 12 6 24 46 34.3 25.7 Garmen L workers 53 5 78 28 131 33 79.9 64.6 Shoemakers 36 19 2 3 38 22 63.3 8.3 Carpenters 53 14 1 54 14 79.4 1.5 Brick.1 ayers 811 35 - 2 84 37 69.14 1. 6 Painters 21 9 - - 21, 9 70.0 Welders, plumbers 29 24 - 9 24 51s.7 Electricians 29 26 29 26 52.7 - Mechanics 69 59 1. 1 70 60 53.8 1.5 Radio engineers 6 1-3 - - 6 13 31.6 - Jewellers 19 3 - 1 19 4 82.6 4.3 Metal workers 1.1 7 - - 11 7 61.1 - 011 Printers 9 12 2 1 11 13 45.8 12.5 Potters, brickmakers 8 12 1 1 9 13 40.9 .9.1 Food processors 34 12 10 2 44 14 75.9 20.7 Orher operatives 19 17 2 3 21 20 51.2 12.2 Packers labellers 11 20 3 10 14 30 31.8 30.0 Apprenticas, mates 20 17 2 - 22 17 56.1, 5.1 Stevedores, loaders 9 19 - - 9 19 32.1 - Retail salesmen 152 11 154 5 306 16 95.0 49.4 Street pedlars 102 3 77 - 179 3 98.4 112.3 Barbers, hairdressers 12 - 14 26 - 100.0 53.8 Barmen, waitresses 28 10 6 - 34 10 77.3 13.6- Cooks 11 7 8 2 19 9 67.9 35.7 Boardinghouse keepers 1. - 18 4 19 4 82.6 95.7 Launderers 1 3 41 2 42 5 89.4 91.5 Domestic servants 32 4 289 -7 321 4 98.8 88.9 Cleaners, shoeshiners 11 1.4 - - 11 V, 114.0 - Porters, caretakers 37 40 2 37 42 4 .8 2.5 Source: Ministry of Labour Urban Survey 1973, Manual Worker Sample. SSS = Small scale sector LSS = Large scale sector 62. Doubtless the nature of an occupation may be transformed by different olant technologies, and these broad occupational grouoings may conceal smaller more sectorally specific occupations, but nevertheless, in terms of broad categories there does not appear to be much evidence of occupational segregation between the two sectors, aoart from the exceptions already noted. One implication of this is that workers within a particular trade or occupation ought in theory to be able to move between small And large scale enterprises, considerably improving their incomes in so doing. In reality the extent of inter-sectoral movement depends on a number of factors, amongst them the skill levels of workers and the recruitment practices of firms. The fact that women tend to concentrate in the few occupations which are primarily located in the small scale sector naturally restricts their possibilities of such movement. Table 8 presents the average earnings of workers in establishments of different size. The first point of note is the relatively high mean in the smallest category, which is due to the inclusion of the self-employed there. When they are separated from wage and family workers, the mean drops considerably (see Table 9). With this exception, there is a progressive rise in income levels as the workforce size increases. However, the rise is only gradual until the 100+ category is reached, after which the rise is very pronounced. This association between enterprise size and level of average earnings applies to both men and women, although it is noticable that within each size group the income level of women is-substantially below that of men. Table 8 Mean earnings of manual workers by enterprise and sex* (coefficients of variation in parenthesis) Workforce size Men- Women Total 1-4 workers 3252 1494 2421 (0.75) (0.95) (0.91) 5-9 3237 1878 - 2923 (0.61) (0.76) (0.67) 10 -19 3331 2386 3209 (0.57) (0.24) (0.56) 20-49 3498 2361 3362 (0.44) (0.49) (0.46) 50-99 3587 2811 3447 (0.46) (0.24) (0.45) 100+ 4570 3465 4451 (0.51) (0.50) (0.52) Total 3609 1703 2975 (0.64) (0.88) (0.71) *excludes earnings over 14,999 soles. Source: Ministry of Labour urban survey. 63. The coefficients of variation in this Table are still sufficiently high to indicate considerable dispersion within both small and large enterprises. The small scale sector presents particular problems because of the different types of income represented there. However, Table 9 shows that the difference between the two types of wage income is relatively marginal and affects small numbers, compared with the distinction between self-employment and wage work in general. Table 9 Distribution of manual emoloyment and averaqe monthly earninqs in the small scale sector, according to tyne of remuneration Small scale sector Male Female Total (1-19 workers) N X N % X N % X Self-employed 457 (28.4) 4,141 284 (17.6) 2,080 741 (46.0) 3,351 Wage workers, cash only 449 (27.9) 2,825 219 (13.6) 1,439 668 (41.5) 2,371 Wage workers, cash and kind 25 (1.6) 1,148 175 (10.9) 1,091 200 (12.4) 1,098 Total 931 (57.9) 678 (42.1) 4609 0o0.0) 2,664 mean monthly earnings Source: Ministry of Labour Urban Survey (missing data 37) Amongst the most important of the factors which can affect income variance within both sectors is skill - a variable which also profoundly affects the mobility of particular workers. II Skill Skill distinctions are recognized within most manual trades or occupations and are considered to be crucial for the aquisition of high incomes, successful self-employment, pride in work and social esteem. The following statements from the case studies are illustrative: "conocer el oficio es basico" (you've got to know the trade skill) "se gana mas especializandose" (you earn more if you specialize) "esta es mi profesi'n, es lo 6nico que se hacer" (This is my trade, its the only thing I know how to do). "prefiero trabajar en una empresa grande para explotar mi saber" (I like to work in a big firm so I can exploit my skills). 64. The concept of skill is complex and cannot be measured by any crude indicators such as the level of formal education or the 'sector of employment. It is likely that skills can be differentiated qualitatively as well as quantitatively, and that a particular skill level or type will encompass a range of qualities. Furthermore, the precise categorization of skill made by employers may be a function of a particular system of incentives and control mechanisms as well as the contribution of the workers to the production process. There are therefore enormous difficulties in comparing the skills of different groups of workers. In general however, most workers and employers recognize four broad levels of skill which refer to the degree of technical knowhow and responsibility involved in the trade. These correspond to the following: unskilled work, usually involving heavy labour (neon), trainees, apprentices and mates (ayudantes), journeymen (operarios)and masters (maestros). These distinctions apply to almost all occupations except commerce and domestic service (1) and are to be found in all types of enterprises, even the smallest. Within the small scale sector, it is' common for young children to work in workshops or on construction sites as ayudantes, although they are often used as a source of cheap(2) unskilled labour, fetching and carrying, acting as guardians and so on. After a while they are introduced to the skills of the trade, and gradually reach the stage where they can carry out certain tasks without supervision, at which point they have become journeymen (operarios)and are paid a stable cash wage. Some crucial expertise has still to be acquired before they can pass to the status of master(3) Within these small enterprises passage from one status to another does not appear to be formalized by certification and the length of the training period is indeterminate. Promotion between skill levels is largely dependent. on the personal whim of the 'boss' (patron) and external recognition of skill (1) within domestic service there are status distinctions but they are based more on responsibility than skill) (2) They usually do not receive a cash wage at this point, though they may be given food and/or accommodation and pocket money. (3) Example from the case studies suggest that this consi-sts of non-repetitive skills involving design (e.g. cutting out cloth amongst tailors and seam- stresses or interpretation of plans and installation of equipment amongst construction workers), 65. status is dependent on the boss's recommendation or a test (pruebal This fourfold classification seems to be widely used within large scale enterprises also, though the distinction between unskilled labouring and traineeship is clearer and the training periods in apprenticeshio both shorter and more soecific. Table 10 shows that manufacturing firms vary greatly in the skill composition of their workforce, and employment of highly skilled labour does not necessarily correlate with size of firm. The branches of industry which have the highest prooortions of skilled labour are the metal working industries, printing, wood products and furniture. These are also the industries which have the lowest proportions of unskilled labour. Industries such as food products, footwear and clothing, paper, leather and rubber and chemical products appear to have a greater concentration of workers in the unskilled and semi-skilled levels, poysibly because of less technological sophistication and A greater emphasis on assembly and packaging. Within the construction industry the skill distinctions are shared by large and small firms alike, especially since many of them are directly linked through the practice of subcontracting. The existence of an internal hierarchy of skills within most enterprises produces a corresponding stratification of income levels there. Unfortunately it is difficult to demonstrate this from the case studies since the cross sectional data do not have a good distribution of persons in the lower skill levels , i n particular there are no children or family labourers in the category of apprentices.(1) Furthermore, there are not sufficient number of cases in large firms which could demonstrate the impact of skill distinctions on incomes in such enterprises. However, the longitudinal data from the occunational histories show clearly that most upward moves in skill status were accompanied by an increase in income. Table 11 shows the considerable variation in incomes which corresponds to the different skill levels in all three sectors.(2) Note that the range of variation between the skilled and unskilled means is widest in the small enterprises and amongst the self-employed, and that the means for skilled workers are remarkably similar in the three groups. Note also the lower coefficients of variation in this table. (1) For logistical reasons, only heads of families were interviewed, which Produced a bias against younger age groups, except in the case of domestic servants who were classified by the Ministry of Labour as heads of their own families. (2) The income data in the case studies refers to daily and not monthly earnings. 拓 . 져■뱐■큐■스뻐으．..5.8므1...．ㅉ겐p엔츠뜨뜨흐느뜨뜨坐뜨뜨쓰느느쓰뜨二쑈ㄴ뜨프그느스쁘르 Branch ㅇf ：그dusㄷry 51:e ㅇ： 媤αㅗㄷ己 cㅇ1ㅗar Skㅗ11ㄹ쑈 se＝ㅗ－S노ㅗlled Uusk：ㅗlad Traㅗ仕eㄹ 』＿__ ∼－－즈뜨르므드느느－.-----－·―--,--------－■－ 功ad ㅈoduc觸（聊d. &aahㅂ삐 34.8 2. 7 9·3 41·0 12,2 10-49 날醱kars 50+ 27 . 3 3 . 2 10 .0 3쇽 .0 25 · 5 Bakarles 10-49 14 . 6 0 .6 50. 4 24 · 9 9 · 5 SG논 10 .7 1 . 7 16 . 1 53 . 5 17 · 9 3ave:ag탭5 10－쇽9 51.3 2.1 7·6 30·7 8 ·3 50+ 38 ,3 10 . 6 5 , 9 26 ，ㅗ 19 ,1 T蠟“1단5 1Oee쇽9 29 ,4 5 . 7 今1 . 7 7 · 7 15 · 5 5아 . ㅗㅗ．4 4 . 2 52, 2 ㅗ5 ·3 ㅗ7 ·9 Fㅇ戚＝긋ear & LOee〈ㅑ9 31.5 5 .3 36 . 5 24 ·8 9 ·9 clㅇ댜노1ㅍg 5Q+ 16.7 5·ㅇ 29．ㅗ 42.1 · 7.1 vㅇ얘 pr。ducㄷ5 1ㅇ－49 28.6 16·7 14.3 17 ·ㅗ 23.3 50+ 29 .4 ㅗ9 . 6 5 .4 21 . 7 23 , 9 FLlr:Lㅗ뎨：e 10·｀9 22. 7 28 . 3 16 .3 14 ·6 ㅗ8 ,1 50ㅓ· . 22，ㅇ 21 .8 21 .0 11 .3 23 · 9 ?ap슨r Prㅇduc& ㅗo-^9 貂，6 4.9 3 ·ㅇ 슘0.5 12．ㅇ 50+ 22 . 9 4 .4 4 . 9 47 .9 19 . 9 。ㅑr土ncㅗng , 10ㅢ‘9 57 .3. & 5·3 1p·8 . · ？·겯 · 믄·구 50+ 45 . 3 18 , 2 : 8 . 9 ㅗ」 · 。 ㅗ슥 ·ㅗ Leaㄷher 10-&9 24 .3 4 . 3 8.8 48 ,0 14 ·6 P＝ㅃucㄷ5 5Q+ 18. 8 3 . 6 2 , 2 57 ·8 17 ·6 肋bb·r ㅗㅇ－49 - 28.2 5 .5 22, 7 2ㅇ·9 22·7 . Pr。ducㄷ5 5盼 44 ．쇽 6. 7 24 ．ㅇ 11 ·1 13 ·8 山e祗caㅗ5 10ㅢ卜9 44 ．ㅇ 1.4 14.9 29 .1 10·6 SG, 39 .3 3 ,0 12. 2 31 . 7 ㅗ3 . 8 Nㅇn ㅁeㄷallㅗc lo-49 23, 9 8 . 6 16 . 7 & 33 ·ㅇ , 17 ·8 吻eru prㅇducㄷ5 5盼 . 18．ㅗ 9.1 17·9 44 ·｀ 10.5 贓eㄷaㅗ Prㅇducㄷ＄ 10-49 , 33.9 1ㅗ． 9 18 .5 6 ·2 29 ·5 excㅗ． 醜Chㅗㅍe=2 5α， 17.9 13.8 3쇽 ·4 17 ·7 16.2 NQn-eㅗacㄷr1Cal 10－쇽9 23 . 9 30.9 18 .8 4 ·4 22·ㅇ 凹chiㅍery s아· , 洲·ㅇ 30·3 12.5 4.5 26·‘ uacㄷricaㅗ 10-49 29 ．ㅇ 25, 3 14 , 9 9 ·5 20 ·3 飇。hlㅍar7 5아 .20.6 20．ㅇ 35.2 13·0 11.2 Tr狐spㅇr: 10－쇼9 29．ㅇ 24 . 5 17 . 7 . 3 ·7 25 ·l 飇ch소ery 33·5 22·6 23 ·2 6·2 14.5 Dlv딘rse IOee49 27 . 9 8 . 7 9 . 5 37 , 1 16 . 8 SG논 24 . 3 6 , 6 9 . ㅇ 쇼0 . 3 二9 . 8 Sㅇurce : x스，15ㄷry ㅇf Labour (SE묘E) 馳Pㅗㄻ y Rㅍaraclarl딥S &n la l=duscr스a X革｀u：·acㄷurara, u그a, 1970 67. Table 11 Mean Dailv incomes of case studies bv work'oece size and skill (coe_ 'ficients of variation in carenthesis) Skill level Self-emzloyed wage work 01 1-19 workers 20+ workers Skilled 207 (0.6) 204 (0.4) 225 (0.5) 74 Semi-skilled 90 (0.5) 109 (o.3) 152 (0.3) 47 Unskilled 86 (0.6) 55 (0.6) 136 (0.5) 49 Total 149 (0.7) 115 (0.7) L78 (0.5) 170 88 51 31 Source: Case studies (missing data = 23) The generalized use of skill distinctions within trades and enterprises implies both vertical and lateral mobility. In small enterprises where workers are involved in the same trade, an upward movement from acorenticeshim into mastership status and from wage labour into self- employment is common. within larger enterprises internal careers are also theoretically possible. To the extent that enterprises rely on these definitions of.skill in their recruitment practices, mobilit- i between firms at a given level of skill is also frequent. Examples are to be found in the-construction and transport industries, and in.branches of, manufacturing such as the garment trade. One factor which can hinder this movement however, is the development of firm-specific skills which derive from the use of rare technologies.. At almost all levels of skill with the excention of this last exa-mole (firm-snecific skills) workers have the action of moving into self- employment. Many do so voluntarily because they actually prefer more indeDendence in their work situations; others(such as women) do so because there is no other ontion. We shall now examine the variety of situatidns which can be found under this rubric. III Emoloyment Status ,The most commonly used categories of employment status are those of wage labour and sel-f-employment, but this dichotomy is too crude to express the variety of relationships between producers and workers within the manual labour force in Lima, The widespread prevalence of subcontracting, outwork, 68. commissioned sales and work sharing has created a number of ncermediace categories between the extremes of autonomous production and dependence suggested by the conventional cerminology (MacEwen Scott, 1979). In order to appreciate this complexity it is necessary to examine the concepts which are embodied in the self-employed/wage-labour duality. Particularly, one must distinguish the basis of the economic relationships from the form of remuneration. Precise definitions of self-employment are often ambiguous. What is generally alluded to is some concept of independence in the process of production, a normal condition of which is direct participation in the product market, control over the decision-making processes involved in production, and ownership of the means of production. The autonomous self- employed producer owns his own plant and equipment, purchases inputs, hires labour or works himself, as necessary, and determines the price of his final product (subject to competition), as it goes on to the market. His income is thus a return to all his factors of production, and*is appropriated as net profit. The wage labourer on the other hand, is the owner only of his own labour, which is purchased in the labour market by entrepreneurs who control the other factors of production and who determine the conditions of work and intensity of effort of the worker. The income of the wage worker thus represents a return to labour only, and is appropriated in the form of a time or piece wage. Intermediate forms of employment represent a combination of these two states. Outworkers may own part of their equipment, but are reliant on the provision of crucial raw materials for their work. Their output is not destined for the final product market but is retained in the sphere of production, often for further elaboration. Outworkers have the freedom to determine the location and pace of their work, but the price is determined by the contracting firm. The form of remuneration is mostly but not totally a return to labour, and consists of a piece rate. The degree of real or apparent independence of these workers therefore varies enormously according to the criteria used. The subjective categorization defined by workers - the usual basis for collection of statistics - may not be a reliable expression of these complxities. The structure of these intermediate production and market relations varies between different sectors. On the whole they are most prevalent in the manufacturing sector (particularly the clothing industry), and in construction and transport. Unfortunately, there are no statistical sources which permit reliable quantification as yet. Within the manufacturing sector, there is a range of different situations including truly autonomous producers participating directly in the final product market; debt-bound artisans whose reliance on credit from merchants results in a loss of control over their decision making process (as regards the volume and price of their output); outworkers who, as outlined above, are reliant upon che provision of raw materials although they work with some of their own instruments of production and receive their income as a piece rate determined by the industrial or commercial subcontractors; and workshops produciq under contract for medium and large firms, many of which could be regarded as expanded forms of outwork. The effect of this comphx set of relationships is that workers within a similar occupation may be involved in vastly different contractual relationships. This complexity is illustrated in figure S- The symbols in this figure signify enterprises, and the shaded sections represent workers within a similar occupation in different employment situations, 70. mnanuul workers Large facter,es desinano~n of output SmaIt workshCQ . .subcontracting Arisns Outworke ,Wgelabourers , S M.reSants Store% Distrbutors Cients Mass market Figurc 5 Relutions or dependence und independence in dhc manuf icturinu. sector man.ai workers Large and medium Consortia destination of service companies - ---+- subcontracting Labouf - Smali contractori . builders lndependent specialists W y age ~ Laour s Clients Mass market Fi.ure 0 Rclafions o( derndence and erndence in the constructi sector 71. arranged from left to right in terms of their independence or dependence as producers. The variations in the shape of the symbols indicate differences in the scale and internal division of labour of the enterprise. The continuous lines represent destination of output, and reflect.the fact that much of the output of workers in the intermediate area goes to other producing units as well as into the sphere of commerce. The discontinuous line represent ties of dependence such as outwork and subcontracting. Note that these ties control the destination of output and that they proliferate in the centre of the figure. This complexity is further increased by the fact that workers may involve themselves with different contractual arrangements simultaneously. The construction sector is organized on similar lines, see figure 6. At one end clients approach builders and specialized tradesmen dire.ctly, advance them the money for materials and pay a fee when the work is completed. At the other,concracts are drawn up with construction companies who employ labourers in the full range of specialist occupations and at all levels of skill on a wage labour basis. In the intermediate area, small businesses and gangs are formed by master builders who employ a small number of journeymen, apprentices and labourers on a wage labour basis, and who obtain work both directly from clients and through subcontracting from larger companies. In most of these cases the employment contract lasts only for the dtration of the building work, and the form of remuneration varies between a regular time wage and piece rates. Specialist tradesmen almost always operate on a 'self-employed' basis, however, their fee can be seen as a direct return to labour, similar to a piece rate. 72, The entire construction industry relies on the process of contracting and subcontracting, which fosters the development of inter- mediaries (master builders, architects and engineers) whose function is to provide labour for companies and members of the public. In this sector it is much less easy to make distinctions between the product market and the labour market and beteen returns to labour and returns to entrepreneurship. 4 The situation is further complicated by the fact that workers will move between a variety of contractual rela- tionships, and unless employed on the permanent payroll of the largest companies, tend to see themselves as self-employed. The transport sector is largely polarized between self-employed owners of vehicles such as taxis, lorries and microbuses and wage earners employed by the large bus companies. However, in the intermediate area there ,are workers who borrow or hire vehicles belonging to other workers in return for a percentage of their fare takings. This practice is most common amongst owners of microbuses and provides them with a means of financing the large capital outlay involved in the purchase of such vehicles. The contracts between substitute drivers (palancas) and the owners of the vehicles vary: in some cases it consists of a verbal agreements as to the size of the 'rent' or commission (20%-50% depending on the profitability of the bus route), in others the contract consists of written agreement about the hours worked, the number of times the route is driven, etc. At one ex- treme of this continuum, the palanca system could be seen as a form of work or capital sharing amongst self-employed producers, but at the other, it is barely disguised wage work, in which the income constitutes a form of piece wage. 73. Within the commercial sector, the extremes of self-employ- ment and wage work are represented on the one hand by traders who bear all the costs of purchases of merchandise, freight and other fixed costs (such as stalls, tills, storage, etc.) , and on the other by commercial companies and scores who employ workers on the permanent payroll, and who are paid a wage or salary in return for .heir work. In the second case, the remuneration consists of a combination of time and piece wages, the latter consisting of a -percentage commission ono sales designed to increase productivity. In the intermediate area, there are workers, usually ambulatory salesmen whose income consists entirely of commissions. Although many of these workers would consider themselves self-employed because of their freedom to work when and where they wish, the fact that they do not own the merchandise, decide the price to consumers nor the size of the com- mission, suggests that their contract in fact consists. of a form of de- pendent wage work. Within the service sector the demarcation is between workers who are employed directly by service establishments such as government agencies, restaurants hotels etc., or individuals who sell services di- rectly to the public such as street photographers, and shoeshine boys. The interdeterminate area consists of personal services such as domestic service and laundry work. While domestic service hired under stable contract in private homes in usually classified as wage work, women who perform some of this work in their own homes are considered to be self-employed. How- ever, some women perform this service in other peoples' homes on an hourly basis: are they to be considered self-employed or wage-earners? 74. In conclusion, the distinction between wage labour and self-employment breaks down under examination into a wide variety of contractual situations. However, since workers themselves tend to emphasize independence in the actual conduct of their work, rather than other formal aspects of their economic relationships, and since most surveys record information on this aspect of employment in terms of actors' subjective declarations, most of the variety becomes submerged into the single category of self-employment. As a result it is high differentiated internally, not only because of the variety of contractual situations, but because of differences in skill and income. As far as skill is concerned, the self-employed include skilled maestros running their own businesses with a fairly high degree of autonomy, semi-skilled operarios doing outwork for other small enterprises or for the bigger firms, and unskilled service workers undertaking work such as laundry and food preparation in their own homes. As far as income is concerned, variation is partly a product of-skill differentials, but also it is related to enormous differences in the amount of capital used in production (being highest amongst driver-proprietors of minibuses), and entrepreneurship (ability to create a stable clientele). The self-employed may be seen as a segment of the manual labour market which runs parallel to that of wage labour in terms of skill and income, and therefore presents alternative paths in the trajectories being mapped out by workers. (see figure 7). For most workers, self employment is a final destination in a trajectory, in that, except in the case of the most successful who could conceivably expand into managers of small factories, it does not lead on or up. The aquisition of skills which would be necessary to pass from relatively low skilled into higher skilled self-employment would involve a return to wage labour. For this reason, the age profiles of the self-employed are older 75. than those of manual labourers, and the least skilled groups of self- employed workers consist of those who have little prospect of further training: the old and women. As destinations therefore, the different types of self-employment represent the ends of vastly different trajectories. For those at the top it will have represented the culmination of a gradual process of skill aquisition and income growth, while for those below it will signify both rejection by the wage labour market and failure in self-employment. Figure 7 shows the interaction of the three aspects of job differentiation in stylized form. The figure presents the range of different types of jobs which could be found within a particular trade or occupation (e.g. the building trade, the garment trade). The symbols again represent enterprises, and variations in their shape indicate differences in workforce size and whether they consist of wage labourers or self-employed persons. The horizontal divisions in the figure represent different levels of skill. Finally, the arrows represent possible trajectories of movement through the various jobs. The major purpose of this figure is to show how different jobs within an occupation have different consequences for workers both in terms of their.earnings level and their mobility chances. It also indicates that the routes towards better jobs are many and various. self employed - - direction of movement wage labourers, skilled 4 semiskilled unskilled ' / trainee - small scale large scale enterprises enterprises 71,zure 7 Job differentiation accordin9 to enterrise size, C-Muloyment status and skill, with.-aatterns of mobility. 76. Chapter 5 THE SOCIAL FRAMEWORK OF JOB MOVEMENT Before analysing in detail the particular profiles of movement between jobs it is necessary to describe certain aspects of the labour market which affect decisions about jobs and access to them. This concerns, first, the institutional framework within which decisions about jobs are made (which particularly affect first entry into the labour market and the patterns of employment of women), second, the role of personal networks in affecting recruitment to particular jobs and, third, the personalistic nature of skill acquisition . These factors affect not only the supply of labour into particular jobs but the shape of particular profiles of movement and the allocation of workers to specific trajectories. Institutional factors affecting first entry into the labour market Within traditional areas of production in Peru there are several institutionalized systems of labour allocation, which involve the employment of certain workers (mainly women and children) being decided by third parties, with varying degrees of coercion. These systems are relevant for the present Study, firstly, because many workers in Lima initiated their working life -within them and indeed may have migrated to the Capital because of them, and, secondly, because these systems exert a normative if not actual influence within the urban labour market today, especially regarding female participation patterns and the employment of young workers within the small scale sector. The first system for allocation of labour is rooted in the peasant household, in which patriarchal authority provides the basis for distributing work to each oE the members. Women perform in a number of economic activities along with their domestic and reproductive roles (Deere 1977, Bourque 197b). Children are entrusted with such tasks as lookin4 after sheep, accompanying parents to market and so on, from the age of five or six. A few years later they may assist with cottage industry or work alongside their parents in various forms of wage work (e.g. harvest work, casual work on building sites, etc.). Their contribution to family income is considered a normal and correct part of family life, and is in most cases unremunerated. The important features of this system of labour allocation are first, that economic participation is based on the needs of the household rather than the desires of the individual (although they 77. may not necessarily be incompatible) and, secondly, that the decision making structure is rooted in the kinship system, which includes such elements as patriarchal authority (ability of the husband and father to determine the activities of wives and children) and filial loyalty. This system exerts a Powerful influence over the activities of women and children in the urban labour market, as we shall see below. The second system of labour allocation involves reciprocal exchanges of labour and services between households connected by neighbourhood or kinship ties. In the peasant communities of the Andes, this sytem enables individual households to cope with fluctuations in the labour requirements of the farm and also provides a security system for those hit by temporary misfortunes. As migration introduced differentiation into these communities, so the exchange of services began to straddle town and countryside, with the relatives of departed kinsmen looking after lands and livestock in the formers' absence and urban migrants helping to find work for the offspring of those who stayed behind (Alderson Smith 1975). This system of exchange was by no means regular, was often unequal and served different purposes at different times. However, it provided a source of cheap labour for urban traditional producers and a means of. offloading excess dependents from peasant'households experiencing economic stress. This exchange of services was frequently matched by a similar flow of subsistence goods; wages flowing from the city and rural produce going to the town.. A grat many children are brought to Lima by this means, particularly young boys for apprenticeships and young girls for informal domestic service. Statements such as the following are common: "mi tio me trajo de Huacho para trabajaren su panaderla (my uncle brought me from Huacho to work in his baker's shop) mi tio me puso de aprendiz en su taller" (my uncle put me to work as his apprentice) "vine con mi t1a para cuidar sus nenes" (I came with my aunt to look after her children) "nos mandaron llamar desde Lima" (they sent word from Lima for us to come). The flow of goods appears to be very sporadic and may, in the case of goods going to Lima, merely serve a ritual function, designed to maintain the system as such. The flow of wages, however, is crucial to the survival of many peasants. (Conversation with Adolfo Figueroa 1980) 78. In most of these cases the decision is made between the uncle in Lima and the father of the child, as is indicated by the widespread use of the passive tense by informants when describing these situations. The system relies both on the parental authority and the system of service exchange between relatives. Doubtless in many cases the children involved would have been happy with the decision, especially if it gave them a chance of visiting the big city and acquiring crucial skills. In some cases, however, there were expressions of discontent and mention of escape. It is clear that considerable coercion could be involved and many abuses were committed: Juan was brought down from the mountains by an aunt, who promised to look after him while he continued with his schooling, but instead set him to work as a cleaner in her boarding house. When he was eighteen she got him a job in a chalk factory opposite her house, where he worked day and night, sleeping in the factory. The aunt appropriated his wages, leaving himonly with a little pocket money. A third system of labour allocation took place within the agricultural estates (haciendas)1. Landlords often had rights over the family members of their.hired workers, which extended well beyond formal wage contracts (Deere 1977, Hobsbawn 1965). Debts could be paid Off in labour, particularly in the form of domestic service. (2) Landowners were able to move their workforce round the country between estates and often took their personal servants to work in their urban residences.4) In some cases this developed into a thinly disguised trade in domestic servants, where landowners' wives acquire maids from their estates for their friends in Lima.. There were many examples in the case studies of domestic servants who came to Lima in this way. (1) This system has probably declined since the Agrarian Reform. (2) This practice was apparently recognisable in law according to Chaplin (1967,64). (3) Miller (1967) gives examples of sugar estate owners on the coast who purchased Andean haciendas in order to have access to labour. (4) Skeldon (1974) shows that outmigration through domestic service opened the way for the rural exodus in Cuzco villages. 79. As suggested, the case study data shows that these traditional forms of labour allocation are still important in the urban labour market, particularly in the small scale sector. Martinez's data shows that in 1970 a very large proportion of migrants to Lima come at an age when they would normally be involved in such systems. Most of these children aged between 10 - 19 were accompanied rather than arriving alone and, although many came with parents, half arrived with 'significant others' (Martinez 1973). According to our data, 86% of migrants obtained their first jobs in the small scale sector, where such forms of labour use are prevalent. Case studies show that these first jobs often have crucial consequences for the subsequent mobility opportunities of workers. If children are pushed on to the labour market at an early age at the expense of continuing their education it restricts their choice in later years. Many children find difficulty in breaking out of the situations in which they fidd themselves, because of their dependence on the employer for food and accommodation, their lack of familiarity with the city and their isolation from people in similar situations. This limits their chances Of changing jobs in the early years and retains them in many an exploitative Situation. Institutional factors affecting female participation in the labour market The levels and types of participation by manual women in Lima are strongly influented by the factors discussed above, i.e. they are determined by household based considerations, very often decided by cettain.'significant others' and occasionally involve elements of coercion. It is useful to distinguish four separate variables affecting female participation: first, normative influences within the nuclear family determine whether she should or should not work, and what kind of work she should do; second, economic pressures determine whether or not she will work; and, finally, the availability of support systems may somewhat mitigate the effect of her domestic constraints. The impact of these'factors varies at different points in the domestic cycle, as is shown below. 80. Fig. 1 Variables affecting female participation amongst manual workers Stage in the Domestic Cycle Pre-marriage Marriage and Childbearing Post Childbearing or pre-union Normative Parental Spouse's influence Spouse's and influences pressure adult children's influence Economic Parental Inadequacy of spouses' Adult children's pressures needs earnings, abandonment reauirements by spouse, widowhood Domestic constraints t------------------ Childcare ------------------------- Support Availability of kin/neighbours systems Criche facilities at work All of these influences over a woman's economic participation are rooted in the kinship system. Fathers and spouses have the authority to determine whether women will workor not, while mothers, sisters and daughters provide the support systems which make it possible to work when necessary. Curiously, the normative influences on women may be contradictory; the higher the social status of men, the more susceptible they are to the dominant ideology that women should not (have to) work. Thus one group of women reported: "mis padres decian que yo no debo trabajar mientras que ellos esten vivos" (my parents said that I shouldn't work as long as they are alive - her father was a community leader in a small coastal town) "mi esooso no cueria cue trabajara" (my huband didn't want me to work) "mi esposo me sac6 del servicio domestico" (my husband got me out of domestic service) "actualmente no coso para la calle, mis hijos no quieren" (I've given up sewing for the public, my children don't want me to any more) while others said: "me gustaba trabajar, porque asi ayudaba un poaquito a mis padres" (I liked to work because that way I could help out a bit at home) 81. "mi esooso se podria aburrir de mantenerme a mi y a mi mama" (my husband could get bored with maintaining my mother and me) "estaba acostrumbrada a trabajar, parecia raro estar sin trabajo" (I've always worked, it seemed strange to be without work) While the normative influences on women's decisions to work may be contradictory, the economic pressures are extremely consistent. Girls and women are relied upon to relieve economic crises throughout their lives. Domestic service jobs provide a particularly attractive solution to economic pressures since they are easy to come by, provide an immediate wage (unlike apprenticeships, which pay very little at the beginning) and assume many of the subsistence costs of the girl. The kinds of crises which propel girls into domestic service are all too frequent in a society with widespread poverty. Well over a quarter of the case studies (men and women) had no father at the time when they entered the labour market, due to death or abandonment. Frequent stories were: "falJeci6 mi padre y me vela obligada a buscar trabajo en lo que encontrara" (my father died and I had to find a job, whatever it might be) "me puso mi mama en el servicio domestico porque el negocio no abastecia para comprarme ropa" (my mother had to put me into domestic service because the business could not even provide for my clothes) One little girl was offloaded on to an aunt in Lima at the age of five. The aunt placed her in domestic service a year later but the child escaped and found her way back to the aunt. She then looked after the aunt's children for four years until the latter was able to place her again in domestic service, this time successfully. - After setting up a home through marriage or consensual union, and with the birth of children, women who had withdrawn from the labour market are brought back in because of the increasing demands on the family budget and the inadequacy or instability of the spouse's income. Specific crises such as medical costs arising from illness, desertion by the spouse or widowhood are often critical factors in the return to work. 82. The childcare c.onstraint may.affect women through the greater oart of their lives. Although this apolies primarily to women who are in the process of rearing their own children it also affects elder daughters and/or grandmot-ers, who have.to look after children if the mother works. For most women in the childbearing phase of the domestic cycle this constraint severely narrows their occupational choices, forcing them to seek part time jobs, ones which might be carried out in their homes or one to which they can take their children (such as street selling). However, this problem is not unsurmountable: childcare facilities may be assumed by local family and neighbourhood networks or, in a few cases, by the provision of creche facilities in modern sector establishments. These factors affect women's mobility chances in a number of ways. The early placement of young girls into domestic service precludes the possibility of obtaining access into better paying jobs as shop assistants or in manufacturing, since both require higher educational levels. Subsequentl,,, the constraints of childcare result in their having to take jobs beneath their skill levels, mostly in self-employed activities with low profit margins. The role of Dersonal networks in labour recruitment There is everwhelming evidence that personal relationships play an imoortant role in the recruitment of labour, in both Large and small scale sectors, and affects access to self-employment as well as to wage earning jobs. It is useful to distinguish two aspects of labour recruitment: first, the processes of information diffusion through which workers come to hear about Abs and, second, the selecticroceuresu by'emploe when actually hiring labour, Similarly, as far as the self-employed are concerned, there are processes whereby workers hear of business opportunities and the actual system of resource mobilization required for the establishment of the business. In the manual labour market (including self-employed) it appears that neither jobs nor key resources necessary for the establishment of self-employed businesses are available on the open market. This places a premium on information tor obtaining access to a job. For these reasons 83. the response to the question 'how did you get your job?' usually produces a reference to the agents responsible for suoplying the key information. The Ministry of Labour data shows that a very small u:oportion of workers obtained their jobs on the open market and the majority got them through friends and relatives or through direct contacts with employers. Table I Mechanisms of recruitment N Employment agencies 68 2.8 Employers 431 18.2 Professional association, unions 14 0.6 Newspapers 89 3.8 Relatives and friends 1,231 51.9 Other 539 22.7 Total 2,372 100.0 Source: Ministry of Labour Urban Survey (missing data = 5) Many small scale employers seek out labourers directly when they need them; they take good workers with them when they change lines of business, they offer jobs to people seen to be working well in other people's employment and they send word to relatives when they need extra hands. This may be the result of a .reliance on cheap labour provided by the kinship system as outlined above but it may also be the product of a need for special skills or trustworthy workers. Amongst larger enterprises, employers may raly on the existing workforce to pass the word around about forthcoming vacancies and rely on the recommendation of trusted workers as the principle of selecti6n. Yet others use these methods in addition to initial selection tests and a three month trial period before finally committing themselves. The evidence from the case studies suggests that amongst many of the larger scale enterprises it is only the unskillled jobs which are screened in this way, higher skilled jobs are filled by a process of internal promotion in which Personal relationships are even more important, Passage into self-employment is similarly constrained by personal networks because of the importance of trust in business relationships, the restricted access to some crucial resources and reluctance or difficulties 84. in using the banking system. Within the construction industry, for examole, the widespread practice of subcontracting relies heavily on personal re:-ommendations. In the transport sector access to microbus routes is limited by professional associations, and the availability of places is passed by word of mouth to friends and relatives. In commerce there is a preference for trading relationships to be kept 'within the family' and access to the more prosperous jobs depends on possession of puestos (stalls) or franchises, which are sold amongst friends. Personal networks are also the means whereby clients are passed on and credit is raised; many crucial resources such as tools and machinery are acquired through inheritance. One result of this personalization of recruitment is that groups of kinsmen tend to ioncentrate in particular occupations. Skeldon (1974) describes the process whereby paperselling in Cuzco was dominated by workers from one particular village, whereas the streetsweepers came from another. "A survey of municipal workers showed that 75% of the street sweepers came from the provin-ce of Acomayo ... these jobs were never advertised and there were never any vacancies. The job is passed on from one worker to another and the length of employment and time of transfer are arranged by the workers themselves ... Once the position is in the possession of a group of campesinos there is little chance of breaking in from outside." (Skeldon 1974) Chaplin (1967) writes of similar processes in the textile industry in Peru; Laite (1977) describes t1he invasion, of particular departments of the Cerro -- ~ ~ ~~de asomining comanyer rm~ ~ T ~~hTf (1975) reports that strawberry sellers in Lima were all from a tiny community in the Mantaro valley. in the case study data we heard of the trade in tomatoes being controlled by extended kin networks from a particular area in Huanuco, while that of onions was in the hands of families from Arequipa. Another effect is that personal networks tend to concentrab-in occupations which are connected through geographic or economic linkages. Thus workers in rural areas link up easily with transport and commerce, both of which are connected with agriculture; people move from working the land to selling and distributing the produce and thence into urban commerce and transport. Some workers move through different occupations 85. within the hacienda system. One worker had learned his construction skills on an hacienda, another had acquired mechanical expertese working on harvest machinery, and yet another moved from agricultural work to become stable boy and ended up looking after the hacendadds racehorses in the Lima hippodrome. In urban areas, common residence provides another element in the formation of personal networks through which workers gain access to different jobs. The effect as far as particular workers are concerned is to constrain their job opportunities in particular directions, making it difficult to obtain entry into jobs where the right contacts do not exist. me gustar'a,:ser t.ecanico, "pero no he- thido 'relaciones con amigos mec nicos' (I'd like to be a mechanic, but I don't have any friends who are mechanics) es dificil cambiar de trabajo porque no somos conocidbs' ( its difficult to change jobs because we are not known) me gustarla ser joyero, pero por el momento no tengo apoyo' ( I'd like to be a jeweller, but up to now I don't have a patron). Curiously, this constraint on-access to jobs does not apply tb such an extent to women. The most open labour market, in the sense of relying on impersonal mechanisms of recruitment is that of domestic service. Jobs are commonly advertised through special agencies, in the newspapers or by notices placed in the windows of private homes, Although some domestic servants are recruited through personal networks, by far the majority obtain their jobs through the open market. Retait sales workers find it relatively easy to set up operations in their own homes, although access to market stalls may be more restricted. Finally, access to manufacturing jobs is relatively open provided the girls have the right skills. Most clothing workshops advertize their jobs openly and screen their applicants with a trial sewing test. Screening for permanent jobs takes place a few months later if at all. In contrast with men's jobs which have a combination of personalistic recruitment and some screening for promotion and training, women's jobs are characterised by relative 'ease of access' Per-or-1.3ation of the training process For the great majority of workers, formal education contributes little towards vocational training. Just over a third of manual workers had 86. been to the vocationally oriented types of secondary schools ( see tables 2 and 3) Table 2: LEVELS OF FORMAL EDUCATION AMONGST MANUAL WORERS, LIMA 1973 Men Women - Total None 29 1.8% 75 9.4% 104 4.4. Preschool 4 0.3 10 1.3 14 0.6 Primary 833 52.7 512 64.3 1352 56.6 Secondary 673 42.6 ..193 24.2 866 36.4 Higher 42 2.7 6- 0.8 48 2.0 Total 1581 100.0 796 100.0 2377 100.0 Source: Ministry of Labour Urban Survey Table 3 : TYPE OF SECONDARY EDUCATION (857 cases) Men Women Total General 507 76.8% 129 65.5% 636 74.0% Commercial 78 11.8 - 37 18.8 115 13.4 Technical 75 11.4 31 15.7 106 12.3 Total 660 100.0 197 100.0 857 100.0 --rc '. try 2 ~- .. Table 5 shows that workers with trade skills obtain them over- whelmingly on-the-job or in apprenticeships with master craftsmen or their equivalents (78%). Formal education, vocational training,O)and military service are almost negligible in importance. 14% acquire skills informally, ' on their own initiative, i.e. picking them up from relatives or neighbours, or through the domestic culture ( i.e. at home). (1) Only 20% of the sample had Taken vocational courses, and many of these were not relevant to their present occupation. 87. Table 4 HOW TRADE SyILLS WEERE ACQUIRED Men Women Total On the job 313 55.0% 27 22.9 340 49.5 Apprenticeships 163 28.6% 34 28.8 197 -28.7 Formal education 15 2.6 19 16.1 34 '4.9 Military Service 15 2.6 - - 15 2.2 Inform.al classes 9 1.6 7 5.9 16 2.3 own initiative * 53 9.3 31 26.3 84 12.2 Other 1 0.2 - - 1 0.1 Total 569 100.0 118 17.2 687 100.0 source: Ministry of Labour Urban Survey * Applies only to workers with trade skills (687 cases) * atching other people, asking around, etc. These tables indicate some major differences in the skill levels and - training processes of men and women. Amongst the latter there is a much higher illiteracy rate, and far fewer numbers with secondary education compared with men. The process of skill acquisition is also different: on-the-job training and apprenticeships play a -smaller-role and formal training and domestic culture are more important. These differences are in part a product of the processes described above ( i,e . early placement in the labour market) and have the esult of seriously limiting the mobility -UinJ. u UnhUguIU L11ir JJU r1sV57ies 1fe rmas eLow therefore refer to men. A high proportion of men obtain their skills in small scale enterprises where there is little standardisation of training procedures. According to the case studies Maestros can spin out the training time to as much as five or six years, and workers are dependent on their recommendation as the only proof of the skill acquired. This is another factor which ties apprentices to particular patons and slows down their rate of movement through this level of the occupation. It means that subsequent movement into other employment is often dependent on the networks of the maestro, particularly as far as recommendation for subcontracts are concerned ( e.g. 883. in the construction industry). The process of 7kill acouisition which is crucial for upward mobility and entry into self-employment is therefore highly variable and is crucially tied to the recruitment process. In summary, the salience of these institutionalised aspects of labour supply plays an important role in determining the direction and rates of movement between jobs. The household or family which determines when and where a child will be put to work will orient him or her towards a particular social network which already has a specific occupational and sectoral bias. The structure of authority in the kinship system and the employing enterprise (which may overlap), together with the highly idio- syncratic nature of the apprenticeship system, may retain the child in that job for a longer period than if the structures were more voluntaristic and impersonal. Thereafter the speed and direction of movement will, other things being equal (i.e., there being no changes in demand), depend on the way in which his personal network is set up. As we shall see in the next chapter, this personalisation of the labour market has the overall effect of constraining movement to within occupations and sectors. 89. Chapter 6 PROILS OF MOVEMENT General Patterns The general patterns of movement between jobs may be summarized as follows: (1) First entry into the labour market is predominantly in the small scale sector. This often consists of family or household enter- prises, particularly in the case of workers from peasant backgrounds. While household enterprises are less common in the urban areas, there are many cases of children and adolescents being initiated into work alongside their parents, especially when the latter are self-employed. (2) Small enterprises provide the major opportunities for skill acquisition, especially at the early stage in an occupational career. Subsequently workers may move into large scale enter- prises where their skills are adapted to particular technologies. (3) The ports of entry into large and small enterprises are at the unskilled level. Thereafter there is a process of upward movement between skill levels; in small enterprises this'is the result of - increased expertise on the part of the worker; in large ones, the product of internal promotion to different positions in the firm. (4) Most movement is confined to occupations; interioccupational or intersectoral movement occurs mainly at unskilled levels. (5) Movement into self-employment occurs at almost every skill level. (6) The trajectories of women differ in several ways from those of men, tending to show less upward movement and confinement to lower levels of skill and income. (7) Patterns of movement vary within different economic sectors because of differences in industrial structure and variations in the role of skill and capital in access to jobs. Processes of upward movement are most pronounced in manufacturing, construction and transport, although the speed of movement is slower in the first two than in the third. In commerce and services the scope for upward mobility relies less on s-ill than on capital and entrepreneurship, and is restricted to far fewer numbers. 90. Sectoral Patterns In this part of the analysis, occupations have been grouped together in - terms of the industry with which they are most frequently associated,;rather than the enterprise in which they work. Thus, drivers are grouped together in transport whether or not they work in factories, shops or domestic employment. This permits greater comparability between workers with similar skills and also reveals the different labour markets open to workers in a particular trade. (a) The Manufacturing Sector Four general points may be made about profiles of movement in this sector: first, there is a bias in recruitment towards workers with an urban background, prior connections with manufacturing trades, and higher levels of formal education. Although many workers were from peasant backgrounds, a significant number were from homes where at least one parent had been employed in a traditional craft. The group as a whole had higher education levels than those in some other sectors; 7C0 had completed primary schooling, several had secondary schooling and there were no illiterates. Secondly, entry into the labour market is usually directly into manufacturing and into small enterprises. This is the case even amongst workers with relatively high levels of formal education, and with previous employment in higher level occupations (e.g., minor clerical jobs). Except in the case of women in the dressmaking trade, who enter factories directly at a semi-sdilled level, most workers enter at the unskilled and apprenticeship levels. Thirdly, movement into self-employment appears to depend crucially on skill levels and access to clients. For autonomous producers, a good location for the workshop appears to be the main key to success, while for those who depend on subcontracts, good relations with firms are important. Access to means of production are less crucial since mechanisms exist for sharing and renting these out amongst workers. Fourthly, amongst women, major differences exist because $f the fact that the basic skill in the commonest occupation (dressmaking) can be acquired outside the workplace (in homes or on vocational courses), and their immobility subsequent to childbearing confines them to less than optimal locations from the point of view of clients. These patterns are portrayed in figures 1 and 2. Because of the differences between men and women, we shall deal with them in turn. 91. Firure 1. Profiles of movement of Ken in the Manufacturinz Sector 4oth &(&SJ a s e r s T c i raty p S P A I ?4LO LLL-. There are two general. patterns of movement amongst Men in manufacturing jobs. The first'and. most'general is a_process of upward movement.within 'both large and, small enterprises. The second is a circulatory pattern which moves between jobs in different enterprises at the unskilled level. Both patterns converge initially in the small enterprises, as may be seen in Figure 1. Even those who had first obtained unskilled jobs in large scale enterprises later left to join small workshops, and all the inform- ants who were currently employed in large factories had done apprentice- ships in small enterprises. A current mechanic working in a large garage had been recruited from a smaller one, a solderer working at the Electrolu factory had started. off as a blacksmith's assistant, and a shoemaker in the largest shoe factory in Lima had done his apprenticeship in a small workshop. All of these workers had nevertheless been taken on as trainees in the larger establishments, although the apprenticeships were considerably shorter than in the small scale sector (see below). They had then ascended to their present skill level through internal promotion within the firm. In one or two cases, workers had changed to another firm after reaching the highest skill level, twice on personal invitation from their next employer. In most cases, however, they stayed with the firm in which they trained, and at the time of interview had been there for between ten and twenty years. 92. Case No. 142 is a Solderer. His father was a blacksmith on one of the largest sugar plantations in Peru. From the age of thirteen he worked unofficially as his father's assistant there, and two years later he obtained an official position as cleaner in the forge. At the age of eighteen his uncle got him a job in a foundry in Lima as assistant boilermaker. He shortly rose to the status of master boilermaker. Six years later the same uncle helped him into an engineering plant as a trainee fitter. One year later he had ascended to be a master fitter. At the age of twenty-seven he was offered a job by the manager of the new Electrolux plant. Once again he reverted to the status of trainee, although his wage was higher than before. Ore year later he was a master solderer, mending articles which were being brought in from abroad. .Some years later the firm started assembling the same articles in Peru, and he had one of the three most highly skilled jobs in the plant. He has now been with this one firm for twenty-one years and earns two and a half times the minimum daily wage. Case No. 143 is a Yvechanic. Born of a peasant family in the sierra, he spent his adolescent years in casual agricultural work. He came to Lima to do his military service and afterwards stayed for a couple of years doing casual work there.. He then did two years apprentice- ship as a mechanic. He returned to the mountains and got married, but missed the bustle of Lima and was back within the year. Finding it difficult to get a job as a mechanic, he started another apprentice- ship as a carpenter and worked in this trade for eleven years. When he was forty years old a relative got him a job as a machinist's assistant in the State-owned cocaine factory. One year later he was a maestro machinist and he has stayed in this job ever since (ten years). Many of the workers who end up in skilled self-employment do so by entering taditional trades where they become apprenticed .to self-employed maestros, 'who are often kdnsmen. Most workers seem to do this from first entry into the labour market, especially when they come from families that already have a tradition of craft trades. Case No. 59 - Jeweller: This man came from a large provincial town near Lima where his father was a printing worker. After completing his primary school education, he went to visit some relatives in Lima, and became apprenticed to his uncle in the jewellry trade. His apprent- iceship lasted for three years, and after this he continued to work with his uncle on a wage labour / commission basis. After seventeen years working like this he acquired a locale next door (in the downtown area of Lima) and set up his own business. For him the cucial determinants of his success were the access to a central location and the network o-- clients he had established while working with his uncle. Case No. 35 - Tailor: This man came from a jeweller's family in Lima. His father died when he was thirteen and he went into an apprentice- ship with his brother-in-law who was a tailor. One year later he was employed on a regular basis in the same establishment as a pantalonero (trouser maker), one of the semi-skilled tasks in a tailor's shop, He retained ntS o u1!y-v -J - ... thirty-seven, he had learned how to make jackets also, and was able,to set up his own business. Five years later he had a flourishing business, built up on the basis of subcontracts with commercial shops. He employed six other workers and had a shop fronting on the street in the central business district. After four years of this success his store was robbed of all its machinery and he was left with debts which took years to pay off. Since then he has worked alone from his home, relying on the neighbourhood for work. 93. Case No. 75 - Baker: This man came from a peasant farm near Lima. He worked as a child on the farm, and then at the age of thirteen went into domestic service in a nearby town. At fifteen his uncle, who sold bread in Lima, took him to help on the rounds. He worked as an avudante with his uncle for five years, earning his food and keep and a little pocket money. At the age of twenty his uncle helped him to get his own round delivering bread for a bakery, and he worked in this job for fourteen years before setting up his own bakery. The business was established little by little, and was able to survive in the early years because he managed to retain his old clients. Now his major constraint is labour. He and his -wife work long hours, and as yet his children are too young to be able to help. Some workers enter this trajectory at a later period, after trying their hand at other jobs. However, whatever the prior erperience, the Dort of entry is always through the post of ayudante. Case No. 27 - Tailor - was the son of a salesman born in a valley near Lima. After completing primary school he worked for five years helping in his father's store. At !e age of twenty he moved to Lima and obtained a job in a bookshop where a neighbour worked. This .shop was-one -of -a chain in Lima,:and he-ascended to counter assistant and then to the Accounts Department, when the chain went bankrupt. At the age of twenty-seven, and with a wife and child, he decided he must learn a trade, and went into an apprenticeship in a tailor's workshop. He stayed in this place for ten years before going independent. Case No. 118 is a Brush ?aker. Originally of peasant background, he came to Lima at the age of twenty, and worked as a bread seller until he was called up for military service. While in the army, he worked. as a mechanic!s assistant .in the armed dar division.. Or leaving he obtained a job as a trainee machinist in a small factory which made paintbrushes. After he had worked there for a time (three years as a trainee aid four as a machinist) he decided to set up his own business. He went to work in a fishmeal factory to earn enough money to establish himself (the wage rate being more than double that of the brush factory), and a year later set up his own business. At present he employs two nephews and is earning almosttwice the minimum wage. The length of time which is taken before passing into the status of skilled self-employment is striking. Apparently the major obstacles are capital or a crucial skill. For some of these workers, rather than staying indefinitely with a maestro the solution is to take up outwork-contracts for shops or factories. In these cases the crucial design skill and costs of materials are assumed by the sub-contractor. For example: Case No. 38 - Tailor. This man was relatively highly educated; he completed his secondary schooling and studied for two years at the Topographical Institute in Arequipa, Peru's second largest city. He worked as a clerk on construction sites, and came to Lima to a similar job with one of the biggest construction companies there. H0 ever, when his contract finis-er EAd diffittli nci.ng aother job. Already married with one child, he joined his cousin in a tailor's workshop, and within a year was able to join him in completing outwork contracts. He has worked at this for the last six years. He works at home, with his own sewing machine and other tools, and makes up jackets and school uniforms for a shop down town. He says that the major obstacles to his setting up on his own are the large capital costs involved in renting an adeq.uate locale and purchasing a stock of materials. Also he does not know how to cut cloth t.o fit clients. 94. Subcontracting arrangements provide a means of risk evasion for skilled workers as well. This is shown in the case of an upholsterer -- who alternates between various combinations of wage labour and self- employment: Case No. 93 - Utholsterer - came into the trade at the age of sixteen after a number of casual jobs in agriculture and construction. He joined his cousin's workshop and worked for three years as an apprentice and a further three as an ocerario. He then set up his own workshop, where he worked alone for three years. However, the location of his workshop was not adequate and he agreed to manage a friend's business in a rented locale. He came to employ three assistants there, but the business could not flourish because of competition next door. He set up on his own again, and in a few years managed to land a big contract doing the upholstery of seats in a microbus plant recently set up. He works there to this day, employing a nephew as his assistant, and plying his trade on a contract basis. A number of workers remain at the unskilled level, either because they entered the trade too late or because they did not stay at it for long enough, but went off into other unskilled jobs such as street sales, construction, etc. For these men the options are to remain within wage employment as ooerarios working with a maestro (as in the case of No.133 below) or to set themselves up in unskilled self-employment doing small repairs. Case No. 133 is a Tinsmith. He spent most .of his life alternating between rural-base ottage industry (weaving) and very arduous work digging guano 1 on the offshore islands of Peru. When the latter source of employment declined and became too heavy for him, and thus ceased to provide the necessary supplement to his weaving, he was forced to bring his family to Lima. At the age of forty, and with no other skills, he had to rely upon his brother-in-law's offer of a job in his metal working workshop, and has worked there ever since as an onerario (eight years). Case No. 12 is a Shoe ReDairer. He was born in Ancash on a peasant farm. Between the ages of nine and fifteen he helped the family, looki.ng after animals; and then at fifteen joined the stream of labour from the area towards the sugar estates, getting work as a cane cutter. He stayed there until the age of twenty-si_, when he moved to Lima, getting work on a cotton plantation on The out- skirts of the city. Six years later he moved into the wholesale market of La Parada, where he worked in various unskilled jobs as a porter and street pedlar. Sulisequently he got work on construction sites where, owing to the boom in urban construction during the 1950s, there was much demand for labour. However, his age was against him, and this option began to close off as he was approaching the age of fifty. In between construction jobs he worked as a bottle seller and learned to repair shoes. At the age of fifty-seven, a friend told him that there were stalls available in one of the markets where he could work as a shoe repairer. At the time of interviow he had been working there for seven years. Guano is a manure derived from the excrement of Delicans. 95. Repair jobs also offer a source of employment to those who had fallen on hard times for other reasons, such as workers who were made redundant from factory work at an age when they were too old to obtain another job or were unable to set themselves up in a small workshop of their own. Case No. 34 is a Shoe Repairer. Born in Cajamarca, an area which traditionally provided labour for the sugar and rice estates, he spent most of his adolescent years working in casual agric- ultural employment. His father, who was a tailor, warned him of the need to learn a trade, but he was not inclined towards tailoring. At the age of sixteen he decided to become a shoemaker and entered apprenticeship with an uncle. At first he received no money while he was learning the trade, and then very little. When he married and his wife had' a child he set up a small shop, hoping to make more money, but he was barely able to make a living from it, Around the age of thirty he went to try his luck in Lima, leaving his wife with the shop. He spent the first three years in casual unskilled jobs - a waiter in a cafe, cleaner in a clinic, and labourer on a construction site. Finally, at the age of thirty-three, he found a job in the Singer sewing-machine factory which had recently been set up. He had worked there as a machine operator for five years, when the plant closed down. By now he was thirty-nine, and had great difficulty finding another job in a factory. He went back to his old trade and set up as a shoe repairer. He has worked in this job for the last six years, but makes very little money out of it,. He has no workshop and works in a tenement block, relying on his neighbours for trade. We will now consider the trajectories of women. Figure 2. Profiles of movement of Women in the anufacturing Sector // N 71,;7' 7 1 -0t 0 L:r Vy- 96. The trajectories of women in the manufacturing sector differ from those of the men in several important ways. First, since the majority of female employment in this sector has to do with clothing, the basic skills are often learned outside the labour market, either in people's homes (i.e., sewing, 1nitting and embroidery all form part of women's roles in the 'domestic culture'), or in vocational courses (dressmaking). This means that women are much less dependent on apprenticeships in small workshops for induction into the trade, and clothing factories can recruit directly into the semi-skilled levels. Only firms which have a specific quality component due to brand names insist on a specific training for their workers. As a result, there is a much greater convergence on large scale enterprises rather than small ones, as can be seen in Figure 2, and there is much more mobility at the semi-skilled levels, both in terms of direct entry and also movement between enterprises. Second, outwork-plays a far'greater role in the sector, largely as a result of the immobility of women after marriage and childbirth. These jobs often are continuations of previous full-time semi-skilled .jobs, but there are also some very unskilled operations such as assembly of eggboxes and folding of envelopes. Third, outside the garent trade, there is a higher proportion of direct recruitment of women into unskilled jobs in factories. . These jobs consist of packaging and labelling operations, do not involve any training processes and have few possibilities for upward promotion. Finally, the typical refuge jobs for women do not involve the heavy reliance on repairs which occurs amongst men, but involve basic manu- facturing operations at low skilled levels. Typical examples are knitting for the neighbours and outwork. Many of such jobs are sporadic and involve few hours in each day. Case No. 105 - Dressmaker - completed her primary schooling in Arequipa and at the age of thirteen joined the workshop of a friend of her grandmother. She stayed there for three years and then started to work on her own on the basis of outwork contracts from a nearby hospital. Five years later she came to Lima and got a job in a small dress shop, making clothes and attending to the public. Two years later she moved to another small workshop, and the next year to another. One year later she moved to a small factory where she stayed for about three years. At the time of interview, she had just left to set herself up independently. Case No. 98 is a Shirtmaker. Originally from a rural area, she came to Lima with her father at the age of fifteen and went into domestic service. She worked in a number of houses until the age of twenty- nine, when she obtained a job in a small factory ironing shirts. The next year she got a similar job in another factory, and yet another three years later. By this time a tailor friend had been teaching her to sew, and she was able to move to a machinist's job instead of ironing. After she married and had a child she left the factory, but continued to work for them at home on an outwork basis. 97. The change between the formality and turnover of employment of women prior to marriage, and the informality and stability afterwards is dramatic. Very few women continued in wage employment after marriage, although some did. Most entered self-employment on a skilled or semi- skilled / outworker basis. All complained about the unevenness of their businesses, due either to the sporadic nature of outwork demand or their deDendence on low income clients. In this latter respect the imobility of women is a crucial obstacle to the success of self-employment. In conclusion, trajectories within the manufacturing sector show a pattern of upward movement into skilled work for most men, involving relatively high incomes in large factories and relatively successful self-employment; and a pattern of-stagnation in-ow paid self-employment or unskilled work for most women and a few unskilled or redundant men. (b) The Construction Sector As already mentioned in the previous chapter, the widespread use of subcontracting in the construction sector makes many of the distinctions we have been relying upon ambiguous. This applies particularly to the wage labour / self-employment dichotomF and also to the question of the size and types of enterprises in which people work. Many workers employed on the large construction sites are in fact self-employed or work in small subcontracted gangs led by maestros. The fact that workers move between many contracts of different sizes and types also complicates .the picture. In general, however, most workers are recruited through small enterprises (in small gangs or with self-employed specialists) and have to acquire their skills within such groups. This is shown in Figure 3, where it can be seen that although large-scale enterprises may employ labour indirectly, there is no internal promotion jystem there because of the impermanency of the contracts. For most workers, hiring, training and promotion lies in the hands of the maestros. A broad distinction may be drawn between those in the general building trades (bricklaying, plastering) who work in the gang system and the specialists such as plumbers, electricians, carpenters, painters, etc. who usually work in very small groups consisting of a self-employed maestro and his mate(s). In a few cases, a small number of specialists are kept on the permanent payroll of the large companies. In all cases there is considerable movement between different types of employer and different sized contracts. However, skilled maestros who work on the biggest contracts for the large construction companies are usually wage employed, while the self-employed maestros tend to work on smaller jobs. The very smallest contracts are those done for 98. neighbours in shanty towns, by semi-skilled builders of oficial status. This type of job provides the main source of demand for th3 second level of self-employment in Figure 3. The entire industry runs on very personalistic relationships. Knowing the right person and having the right recommendation are crucial for finding jobs, gaining access to skills, obtaining recognition of skill levels and so on. There is also evidence of worksharing in this sector, which again is confined to friends and relatives. Entry into self-employment in this sector depends not on capital or workshop location, but on reputation and an efficient network of friends who will recommend clients and sub-contract work. Figure 3. Profiles of movement in the Construction Sector 601 It 4( (kKI-C t r)t - N1 There is considerable movement into this sector from unskilled work in other sectors, especially agriculture, commerce and services, and there is a generally low level of education. As a result, the ages at the time of entry into the trade vary enormously, and this affects the skill levels they can reach and the kind of contracts they can expect. However, once workers have entered this sector and have moved up into the semi-skilled levels, they rarely seek work again in other trades. Case No. 147 is an example of a Floor Specialist at the maximum skill level in the trade, and is on the permanent payroll of one of the biggest construction companies in Peru. He was. born in a northern coastal town and spent his early childhood helping his parents in agriculture. At the age of sixteen he joined a maestro and worked with him as a labourer for six years. At the age of twenty-two he was called up for military service, and when he came out a banker brought him to Lima to help build his house. Subseauently he obtained a job with a big construction company making special floors for bakeries. Two years later he was taken on by the largest company in the country, He has stayed with the company ever since (twenty-four years) and has a stable relationship with them even though there are sometimes short periods of unemployment in between contracts. Currently he earns the state regulated wage for skilled workers in construction, but he receives as much again in overtime payments and bonuses. 99. The following case provides an example of a maestro bricklayer who is now self-employed. Case No. 5 - Bricklayer - was born in the sierra of a peasant family. At the age of ten an aunt brought him to Lima, promising that he could continue his education with her. However, he was made to work in her pension (boarding house) for the next eight years, and then placed in a chalk factory nearby. He escaned from this place and found work on an hacienda (estate) outside Lima. While there he was moved out of agricultural work into construction, in order to help build some silos. He made some contacts with other builders while on this work, and soon was able to move into another job as a watchman on a construction site. From here he worked as a labourer on a number of sites, and at the age of thirty was taken on as oficial by a maestro for a series of contracts which were to last over the next five years. At the age of thirty-six, he set himself up as a self-employed maestro, and has been working ever since on small contracts. Currently he has two assistants, because he is working on a small job (a large one would need six to eight assistants). He is apparently quite successful since he matntains that he has only a few days without work in between each contract. This case provides an example of a worker who joined the building trade after a long history of unskilled work in other sectors. Case No. 163 - Bricklayer - was born in Ancash, another part of the rural sierra. At the age of eleven he travelled to the coast with his uncle to work as a labourer in construction. He worked there for seven years and then did three years on a farm, three in a charcoal factory, one in a chalk factory, and another two in agriculture. He then moved into transport, as a lorry-driver's mate,'and as a porter'.in a market. By this time he was thirty; he then returned to agricultural work for the next eleven years. At the age of forty-two he travelled to Lima to stay with his brothers and work with his brother-in-law in construction. At present he is working as an oficial in a small construction gang of eight people. In contrast to this case, all of the specialists who had reached the level of maestro, whether self-employed or wage-earning, had relatively thin occupational careers; they had entered apprenticeships early on in their lives and passed directly to the status of maestro. In one or two cases, entry into the trade had begun at a later date, but,this did not appear to have impeded passage into self-employment. In general the construction sector appears more open to entry at the unskilled levels for a longer period of time than in manufacturing, where age seems to close off such opportunities. However, it is misleading to conceive of this sector as primarily one of unskilled 'refuge' employment. Promotion into well paid, skilled jobs in widespread providing workers have the right social networks. 100. (c) The Transport Sector Jobs in this sector are markedly differentiated according to whether the mode of transport is vehicular or not. On the one hand there are drivers of buses, lorries, taxis and private cars, whose incomes are amongst the highst of any manual workers, and on the other there are loaders, porters and messengers whose earnings are amongst the lowest. The proportions in the latter group are, however, very small. As far as drivers are concerned, there are several features of this occupation which contrast markedly with the other trades considered so far. First there is no internal hierarchical skill structure through 'which it is necessary to pass before becoming a driver. Although many drivers have reached their current positions having been driver's mates or apprentice mechanics, these are not necessary requirements. The skills required for driving can be obtained outside the labour market. This means that recruitment is relatively open; workers come in to the occup9tion 'through a variety of` routes and do not have to spend time in lengthy apprenticeships. On the other hand, a large proportion of drivers are self-employed and have to purchase their own vehicles. This m.ans that recruitment tends to favour those in high income jobs which in turn means persons with higher than average levels of education and employment in large scale enterprises. In fact, only two workers in this sector had not completed .primary school, and almost half had some secondary schooling. Three quarters of the case studies in this sector had worked in large scale enterprises, and for many the indemnity payments obtained from some of these firms provided the major means of purchase of the vehicles. Secondly, while there is little vertical mobility through hierarchically differentiated skill positions in this occupation, there is a very high degree of lateral movement between small and large enterprises and between wage labour and self-employment. One reason for this lateral mobility is the high depreciation rate of the vehicles, encouraged no doubt by the intensive use to which they are put, which forces the self-employed back into wage employment of one kind or another. As a source of upward mobility, therefore, drivers' jobs represent a high income occupation which workers arrive at throuzh other occurations rather than through the trade itself. The rate of inflow from other sections is therefore high. However, since the long term returns in this occupation also are subject to substantial fluctuation, there is also some outflow to other occupations. This contrasts noticeably with other occupations which have remarkably little intersectoral movement once pre-migration experiences are discounted. 101. Access to driving jobs follows a variety of routes. First there are a number of workers who become driver's mates as young boys, either because they have a driver in the family, or know one in the local village. Second, there are workers who enter by moving into delivery services run by factories or shops. Third, there are some workers who enter the trade through domestic service. In one case a national service recruit became the private chauffer of a succession of army generals. Fourthly, many men got driving jobs after working as mechanics, and, finally, a few adventurers try their hand at driving because they can raise the money for an old second-hand car and are prepared to try their luck. The following cases provide some illustrations of typical trajectories in this sector. Case No. 174 - Microbus Driver. This man came to Lima at the age of eleven with the rest of the family. He and his brother worked as street pedlars while continuing their secondary education. When he was eighteen he worked as a labourer on construction sites, and at twenty-three he was called up for military service. On leaving this he found a job in a mining company as storeman. After three years there he left, and with his savings and indemnity pay bought a taxi in Lima. This worked well for a year or two, but the car was old and began to need a lot of repairs. He sold it and got a job as a driver on a new hydro-electric project in the mountains behind Lima, and he soon became one of the transport overseers. Seven years later, when the project was.finishing, he was made redundant, and with his redundancy pay returned to Lima and bought a microbus. Today he is still working the micro, but has declared that 'the job of a driver is not secure, there is no stability'. Case-No. 1 is a Mechanic-cum Tai .Driver. Born in a rural area, he worked on the family farm for the first ten years or so of his working life. When he was twenty-six, his sister, who was married to a policeman in Lima, sent word that a big import-export company was looking for staff. He immediately got a job there as a packer. While in this job he completed his primary schooling and did an eight month course in mechanical engineering. Soon he got another job in another similar company, with the added responsibility of organizing the stores. All the while.he was practising as a mechanic at the weekends, getting contracts from an engineering company. By the time he was forty-one he had moved into a job with a toy factory, repairing vehicles, and also acting as a travelling salesman. Soon he was able to buy a car and joined a line of colectivos. However, after some time his car was too old and was clearly beginning to eat into his profits. He now spends more of his time repairing vehicles on the line and only works his car occasionally. Case No. 128 - Microbus Driver - was from a-coastal town, where he entered the building trade at the age of sixteen. The next year he came to Lima as an apprentice mechanic, and the following year rose to the position of oficial. He then worked briefly in a paint factory before being called up for military service. During his time there he learned to drive and obtained his military driving licence. When he came out he bought a second-hand car and worked 102. as a taxi-driver for a couple of years. Eowever, his car broke a down a lot and he found a job in a textile factory as a driver, where he stayed for eight years. He left this job after a dispute with one of the supervisors and returned to taxi driving again. Three years later he was struck by ill-health and sought out wage employment again in order to get access to health insurance. Ee got a job as a driver in one of the Lima bus companies, but two years later he left again as the company went into liquidation. He now works as a Dalanca microbus driver and earns over three times the daily minimum wage. The workers in this sector who are involved with portering and loading are much lower paid. Their employment is more clearly polarized between large enterprises (e.g., the airport) where they are employed as wage labourers and one-man self-employed enterprises in such areas as markets and stations. Both of these types of jobs were entered at the end of a history of largely unskilled employment. The exception is to be found in the dockers, who form part of a closed shop cooperative. Entry into this coupation is extremely difficult, and depends largely on kinship links. The careers tend therefore to be concentrated within the one occupation, and permanent status in the cooperative is rewarded with very high wages. In summary, as far as drivers are concerned, self-employment has relatively easy access (as long as the capital can be raised), since it does not involve waiting time in apprenticeships. It is, however, a high risk occupation, and workers who fail have to return to wage employ- ment or go into other occupations. Neither of these solutions are to . - be found in other sectors, since self-employment iself constitutes a 'refuge' for unsuccessful workers. While it is true that porters and stevedores provide a certain source of refuge employment, it is for a very different kind of worker, one who is more commonly found in the next sector. (d) Commerce The major distinctions in this sector are between wage-earning shop assistants and self-employed retailers, and amongst the latter, between men and women. Taking the shop assistants first, as a whole they are a very selective and unrepresentative group, being young, relatively highly educated and having a high proportion born in Lima. Many of them could hardly be said to be manual workers, although in some cases they had arrived at their current positions through upward promotion from cleaning jobs. Because of the ambiguity about the manual or non-manual nature of these jobs, and also the fact that they form a very small proportion of the total sample, we shall not analyse this group in any further detail. 103. In contrast to the former group, the self-employed in commerce had much lower levels of education, were generally older and had a background of migration into Lima. vost had moved in from unskilled work in agriculture and services, and few had had experience of skilled jobs in manufacturing or construction, or of employment in the large scale sector. Almost all workers in this group had worked remarkably long periods in their current jobs. Within this group it is noticeable that men have businesses which involve some capital outlay and are in profitable lines such as meat, fish and groceries. In most of these jobs there is some barrier to etry, such as a licence, franchise, stall, or costly merchandise and equipment. In contrast, most of the women work in unprofitable lines, such as sales of perishable goods (e.g., vegetables), with very small amounts of capital, and where there are low, if any, entry costs. Case No. 8 is a Newsacer Seller. He started life, aged eigt, as a stevedore in the port of Huacho near Lima; then he became a fisherman for four yes,rs. At the ae of eighteen he became a -lorrydriver's mate, helpi:ng to deliver goods to the market in Lima. A year later he decided to stay there, working as a loader in the market. He remained in this job for the next four years, until he married a vegetable seller in the market. After his marriage he worked on the stall with her, and stayed in this job for the next twelve years. At the age of thirty-seven he inherited his mother-in-law's newspaper stall, and moved into that line of .business, where he has remained ever since (ten years). Case No. 190 is a Fish Seller. He was born in Arequipa, and started work on a temporary contract in an ice-cream factory, and then as a street pedlar. A year later, when he was nineteen, he got a job in a factory maki m rubber, products, and was doing well there when he had a quarrel with one of the supervisors and was dismissed. He then worked as a fisherman for a year, but decided to try his luck in Lima. Cn arrival he obtained a job as a bus conductor and he stayed in this job for ten years. At the age of thirty-two, his father sold him his stall in the fish section of a market and he moved into this job. He regards this job as better remunerated and less dangerous than his earlier one. Case No. 192 is a Butcher. He was born on the northern coast' of a peasant family and spent his early years working in agriculture. At the age of twenty-four he volunteered for the Civil Guard in Lima and worked there for four years. 'hen he fell ill, he was dismissed, and took up taxi-driving in a rented car. After four years of this he became an itinerant meat seller in one of the markets. This job proved to his liling since it was both profitable and less agitated than that of driving, and he soon rented a locale to expand his butchery business. Nine years afterwards, the owner of the locale sold one of his other butcher shops, and he now owns this business. In all these cases the monthly earnings were between 3,000 and 5,000 soles, but the amounts earned by women are lamentable in comparison. A very high proportion of women had worked in domestic service prior to entering sales, but some had not worked at all. Only one of the case studies had ever worked in manufacturing, and even that was only unskilled outwork in her home. None of the women had ever worked in large scale enterprises. 104. Case No. 102 - Vegetable Seller - was born on a peasant farm and was sent to Lima aged eight. She went into domestic service aged eleven and worked in various houses for eight years, finally leaving when she became pregnant. After the birth of her child she set up a small stall in La Parada and has worked there selling vegetables ever since. Case No. 67 started, selling vegetables at the age of forty-four when she came to Li,a from her village.in the northern sierra. She sells from her house, which is easier for her since she is illiterate. Case No. 90 is a Lottery Ticket Seller. She came to lima aged eight from Chile, and started selling tickets in the central square at the age of seventeen. She has been there for the last thirty-two years. Although some of these women show pride in their work, one gets the impression that these are dead-end jobs in stagnant trajectories. The fact is that choices are so severely constrained for women with no skills and many domestic commitments that their only solution lies in self- employed-commerce or services. The stability of employment here is but a reflection of this callejon sin salida (dead end). (e) Services Within the services sector, again there are differences between the trajectories of men and women. As far as men are concerned, there are two groups; the first contains occupations which had some element of skill involved, such as ph6tography or hairdressing, and jobs which involve somecapital, such as restaurants, canteens, studios, and barbers' shops. The second group includes jobs which are for the old and unskilled, such as janitors, shoe-shiners and cleaners. As far as women are concerned, the overwhelming proportion of wage labour consists of domestic service, and the majority of self-employment consists of jobs which are based around the home and have extended out of domestic service (e.g., laundry, food preparation). Only hairdressing presents an occupation with some career possibilities, but this requires a fair amount of canital investment. For some men, therefore, and a miniscule number of women hairdressers, there are possibilities of upward movement in this sector. For the rest, the majority of whom have been drawn from agricultural origins and have low levels of education, trajectories lead towards stagnation, i.e., low income self-employment. Conclusions This analysis of patterns of job change shows that almost all occupations have an internal career structure containing possibilities for upward mobility through skill acquisition or capital expansion, for movement between small and 105. large enterprises and between wage labour and self-employment. Each occupation also has a capacity to harbour 'refuge employment'. The existence of these internal career structures provides one reason why workers rarely change their occupations once they have begun to progress a little way along the trajectory, and also exclains the immense amount of variation within occupations which shows up, for example, in income dispersion. Almost all sectors have some recruitment of workers from peasant origin, although some more than others. Conditions for and mechanisms of entry vary greatly as to whether they rely on informal apprenticeships, etc., but the contingent factor is usually the availability of a kinsman who can provide information or employment at the right time. In general, the pattern seems to be that most workers enter trades through small-scale enterprises. It would appear that the extent of movement between large and small-scale enterprises has been underestimated,,as has the amount of internal promotion within both. There are major differences between the trajectories of women and men in all sectors where both are found. In conttast 'to the slow process of gradual upward movement in which men are involved, women tend to start off in forms of .wage labour that have rapid turnover (domestic service and dressmaking) with little prospect of skill consolidation, and then are dramatically channelled into forms of self-employment which cannot adequately make use of their skills even where they have them. The closing off of options which occurs at this point is reflected in the great stability in such occupations, In general it is possible to discern three patterns of movement. The slow process of upward movement we may refer to as the 'escalator' model. This is to be found in almost every sector; self-employed artisans, skilled factory workers, skilled builders, drivers, successful shopkeepers, barbers and restauranteurs all to a greater or lesser degree form part of this process. However, it appears that although successful workers may manage to acquire the necessary skills and/or capital to arrive at the top, their income is by no means secure. This is particularly the case with some self-employed artisans, builders and drivers. The second process involves lateral movement between jobs of similar skill levels. It is found particularly at the lowest levels of skill (e.g., workers who alternate agricultural, commercial and construction jobs), but also to some extent at the semi-skilled level (especially in the building trade and amongst dressmakers) and amongst drivers. What is unusual about this pattern of movement, as far as we can discern from the case studies, is that individual workers do not stay for very long in this particular pattern of movement. Possibly, the size of this group would depend on broader movements in the economy, shrinking in a period of expansion as workers moved out into an escalator model, and expanding in a slump as opportunities for upward mobility contracted. 106. The third pattern of movement is really one of no movement: stagnation in dead-end jobs. , As we have seen, this is the pattern to which many, if not most, manual women adhere after marriage and childrearing begins. The implic- ations of these three patterns as far as wider issues of employment and poverty are concerned will be raised in our general conclusions in the' next chapter. 10r7. Chapter 7 CONCLUSIONS This report has attempted to show that a number of common assum=tions concerning segmentation in urban labour markets do not hold in the case of Peru. In spite of the persistence of a large impoverished peasantry and a substantial rural-urban wage differential, a large reserve of 'surplus' labour does not seem to have developed in Lima. In Chapter III we gave some evidence from aggregate data to support this view. We pointed out, for example, that in two classic areas of informal sector work - self-employment in commerce and manufacturing - their relative share of the sectoral labour force had either remained stable or had declined during the period of major economic growth (1940-72). On the other hand, there was evidence that urban real incomes had risen during the period, and this had affected the informal sector as well as other groups, albeit at slightly lower rates. Mobility data showed a trend towards upward movement, especially from unskilled occupations. Taken together these data suggest that there was widespread, if unequal, growth in personal incomes and substantial, if selective, upward mobility, both of which are compatible with .the very considerable economic expansion that took place at this time. The assumption of stagnation in incomes and jobs for an overwhelming number of migrants simply was not proven. The analysis of job differentiation provided in Chapter 4 indicates that the assumed disparity between incomes in the formal and informal sectors was also less clear cut. Although average incomes tended to be hi- sr in larger enterprises than in smaller ones, there was considerable dispersion in each group. As far as the self-employed were concerned, there was little or no difference between their mean income and that of wage labourers, and the distribution of self-employed and wage incomes between different income classes was also very similar. Finally, the high degree of income variation within both groups indicated that neither could be considered homogeneous. A more disaggregated analysis showed that workers in all these categories were employed at different levels of skill, and it was suggested that this variable might account for some of the income variation. Almost all occupations had an internal hierarchy of skill levels which could be found in enterprises of different sizes. Amongst the self-emloyed, varying amounts of capital further differentiated workers. These two variables, skill and capital, formed the basis for 'career' formation which could result in both higher incomes and increased occupational prestige. Each occupation also had its 'refuge type jobs' which consisted of unskilled work. These processes were 6ccurring within occupations and were therefore undetected in analyses based on more aggregated data. ice. Actual trajectories of movement varied between those who embarked on an 'escalator' type which involved them in a gradual process of upward move- ment through different skill levels, and those who were engaged in lateral movement within a single skill level, often between enterprises of different sizes. This kind of movePnt occurred most frequently at the highest and lowest skill levels. A third type consisted of workers who were trapped in dead-end jobs with no possibilities of upward movement or of increased incomes. In this last group we found a large concentration of women and some older male workers. The factors which affect which kinds of workers get into particular jobs and on to certain trajectories include a number of social practices governing labour recruitment and participation. As we saw in Chapter 5, the practice of placing young children into domestic service and apprentice- ships has important consequences for them in their subsequent working life. While apprentices embark upon the crucial process of skill acquisition which might lead eventually into lucrative jobs, young girls forfeit their access to education, which is a more important determinant of their job prospects than apprenticeships.(1) The practice of moving labour around within the hacienda system also had the effect of feeding workers into an oscillating trajectory between unskilled jobs. The personalistic nature of the recruitment and training processes softens the possible discriminating effects of education. It opens up opportunities for those with the right contacts, even where they are lacking in the appropriate qualifications, and closes them off for other workers. The particular contacts in an individual worker's network may determine which occupation he entered and have the effect of slowing down or speeding up rates of movement through particular jobs in that occupation. At the aggregate level, it is possible that these networks affect the rate of migration to Lima and the level of oployment there. They provide important sources of information diffusion in the national labour market, helping migrants to assess more accurately their realistic probabilities of employment. These profiles of movement have been generated from a limited number of case studies, which were not selected on a representative basis. Unfortunately it is not possible to calculate the frequencies of the different profiles from the inistry of Labour survey, since they do not have data on skill levels or capital. Tables la and lb in Chapter 4 can, however, provide some broad hints. From the analysis we know that most career opportunities are concentrated in the manufacturing, construction and transport sectors. If we exclude women (who Some do continue their education while in domestic service, although a minority. 109, barely exist at high skill levels in these sectors) a proportion of the self- employed and of wage workers in large enterprises who might be unskilled (say 25%), we arrive at a figure of 384 workers who might, on the basis of inference, be enjoying the top positions in their occupational careers. This figure would represent 39O of all the men in these three sectors, 245 of all men and 16 of the sample as a whole. On the other hand, about 9C/ of all women and C, of the total manual labour force would be involved in dead-end trajectories. This would leave over half the sample somewhere in the middle, .half way up 'escalator' careers or 'oscillating' in between unskilled and semi-skilled jobs. The limited extensiveness of the upward movement should be kept in perspective. We should also remember that even at the top positions in the career, incomes may be unstable in the short run (as in the case of construction workers) or in the long run (as in the case of owners of fast depreciating vehicles). Nevertheless, the fact that around a quarter of the men experience relative 'success' in workers' terms may exert a powerful influence on the expectations and decisions of others lower down the hierarchy. The significance of the large proportion of women in dead-end jobs also has to be considered. Only a small. proportion of these women are primary income earners with dependents (175) and this group is only 5% of the total sample. Unfortunately we do not have data which permits us to analyse the degree of dependence of the household on these supplementary sources of finance. Since there is little chance of raising.the incomes of this group of workers, the best solution for them might be increased access to skilled occupations and a general improvement in the income levels of their spouses. This would suggest a priority for macro-economic policies aimed at raising incomes on a general level rather than projects aimed at particular target groups of women. It is pertinent to ask how stable is the overall pattern of different- iation and mobility in the manual labour market described in this report. It should be remembered that this research was conducted towards the end of a long period of economic expansion which had lasted in Peru for at least twenty years. The process of upward mobility amongst workers was underwritten by the fact that many large scale enterprises were being set up for the first time (and were thus more open to workers from other industries and lower skill levels) and that small enterprises were able to participate in the expansion either directly by producing for their own markets on the basis of some comparative advantage, or indirectly, through subcontracting arrangements. It is conceivable that the slump which affected Peru between 1975 and 1980 seriously reversed a number of these processes. In a contracting situation we would expect wage employment to fall as a result of redundancies, subcontracting 110. arrangements to be'withdrawn, and the incomes of the self-employed to be reduced, especially in sectors which are sensitive to changes in aggregate demand, such as construction. It is likely that many small producers wCuld shift their lines from manufacturing to repairs, and that the refuge occup- ations would swell in number. The numbers of women in dead-end jobs would substantially increase in spite of the low incomes there. One effect of the disaggregated approach pursued in this research is that it has been possible to identify the 'refuge' jobs within occupations. This permits one to forecast more accurately the kinds of labour market move- ment that would result from cyclical movements in the economy. 111. BIBLICGLAPHY Alden Lewis, R (1973) noloyment, Incore and the Growth of the Earriaas Ph. D. Dissertáion, Cornell University. Alderson Smith, C (1975) The Social 3asis of Peasant Political Activity Ph.D. Dissertáion, Sussex Universit-y Bouraue S and Warren (1976) (,jnpesinas y co::rii: su crnacica n la Sierra c ruanla" ras tud os an ii.ns ,5 Brown, R (1978) "ork" in P Abrams· (ed) Work, Urban.ism and Irecuality Vieidenfeld and !icolson Chaplin, D (1967) The Peruvian Industrial Labour Force Pr.r.ceton Urniv. Pres CODOT (1972) Classification of Occupations and Directory of Occuational Titles K -ISO Deere, C.D. (1977) .-Chan-in Social lelatiois of Production and Peruvian Peasants Woien's Iork' Latin American Persectives,Vol IV Fitzgerald, E.V.X (1976) The State and Economic Develor:ment Cambridge Univ. Press Gianella, J (1970) M,arginalidad en Lima Metrooolitana DESCO Hobsbawm, E.J. (1969) "A Case of Neo-Feudalism: La Convencion, Peru" Journal'of Latin A.merican Studies, I, pP 35-50 Laite, J (1977) The Migrant Worker: A Case Study of Industrialization and Social Stratification in Highland Per- Ph.i. Dissertation, University òf Manchester. Martines, H et. al (1973) 21 Exrodo Rural en el Peru. CEPD Mayer, E (1374) ieci·urocity, Self-Sufficiency and Market Relations in a Contemporary Coimunity in the Cerit-l ,ndes4l ?eru Ph. D. Dis:-ertation, Cornell University Mliller, Richard U (1971) "The Relevance of Surplus Labour Theory to the Urban Labour Markets of Latin America" Bulletin for the International Institute of Labour Studies Vel 5 Niller, Solomon (1967) "Hacienda to Plantation in Northern Peru: the Process of Proletarianization of a Tenant Farmer Socie tyl in Julian H Steward (ed) Contemocrary Chane in Traditional Societies Vol III Illinois Press oser, C (1977) "The Dual Economy and Marginality Debate =nd the Contribution of Micro Analysis: Market Eallers in Rogota" Develoument and Chanze Vol 8 (1980) "Women' s ',ork in a Peripheral nrconomy nimeo 112. Reichmuth, M r (978) Dualism in Peru: An Investi.lation into the Interrelation- shiDs between Lima's informal Clotnind- Industry and the Formal Sector 3 Litt Thesis, University f Oxford. Skeldon, R (1974) untitled draft Ph. D. Dissertation Suarez, F (1975) La Movilidad Ocunacional en Lima Metrouolitana Ministry of Labour Wallman, S (1979) Social Anthrorol·ogy of Work Academic Press Webb, R (1977) Government Policy nd "Oistribution of Income in Peru Harvard University Press World Eank (1979) Peru: Long Term Development Issues (restricted publilcatic