HOUSEHOLD STRUCTURE AND HOUSEHOLD INCOME IN LAHORE

Wenonah Lyon and Michael Fischer

Family type and household structure not only respond to economic conditions, but can also have economic consequences of their own. Households are structured within a particular set of economic and social conditions, and increasing the ratio of income earners to non-income earners within a household is one strategy used to deal with poverty. A socio-economic survey of 457 households in a lower income district in Lahore revealed that forty-six per cent of the households surveyed reported more than one earner in the household. The per capita income of households with multiple earners improves most particularly between one and two earner households. The increase in per capita income in households with three or more working members is slight. The levelling out of per capita income despite the number of earners within the household could be the result of simple economic pressure: individuals who earn little are forced to share a household, while members of better-off households divide earlier. Further examination suggests a more complex explanation. It is also in some cases the result of choice. Having satisfied basic income requirements, multi-earner households may choose not to maximize household income. Instead, occupants prefer other goals, such as job satisfaction, extra leisure time, further training in an apprenticeship, or semi-retirement for the head of the household. When basic needs are met, people find better things to do than work.

Increasing household income by increasing the number of earners within the household has been widely reported. This may be done by taking in an elderly relative or non-relative to care for children, thus freeing adult women for paid employment (Laslett 1972), by female kin receiving welfare benefits and sharing a home (Stack 1974), by delaying marriage of children, and thus retaining their wages within the family and, conversely, by reducing the time of childhood dependency by limiting education and beginning working life at an early age (Hackenberg, Murphy and Selby 1984). The 8extended workforce9, dependent upon more than one earning member, is widespread in the industrialized world as well as in less developed countries as a solution to low income and unstable employment, and serves as a response to poverty associated with both old and young age, and with having finished and just begun work.

In Greentown, a lower income district of Lahore, residents are wage earners, with low salaries but steady employment. They own their own homes and this security of housing allows household size to be exploited to improve the general standard of living within the household. Economies of scale, in addition to the elimination of housing costs for wage earning adult sons and daughters, provide greater security and a larger disposable income for the household in general. Household structure thus subsidizes low-earning members of the household, but it also provides a limited independence from wage-earning for individual members.

GREENTOWN

Greentown is a planned community, a project of the Lahore Development Authority (LDA), approximately half a square mile in area. The residential area is divided into plots each twenty by forty feet in size. In 1970, the LDA began planning a housing development for low level government employees. It also wanted to clear katchi abadi (squatter) settlements from government owned land. Katchi abadi residents were given a free house for the head of the household and free plots for each male in the household over the age of sixteen.

In 1974, the government began housing refugees from Bangladesh in the area, selling houses at low rates. Many of the residents of Greentown came to Lahore as a result of major political upheavals, with all the stress and danger and economic disability this involves. The birth of Pakistan in 1947 and its dismemberment in 1971 resulted in an influx of refugees to Lahore. Some of these have settled in Greentown. In a survey in 1982, we found that 50 per cent of the households in the sample came from areas which are not now a part of Pakistan. This figure is similar to that for Lahore as a whole, which Qadeer (1983) puts at 47 per cent.

As housing prices in Lahore increased, immigration to the area has increased and more plots have been sold to individuals. Rural immigrants, people from the villages of Punjab, the North-West Frontier, and the Tribal areas, have also bought plots in the area. There are also a large number of residents from India, primarily from East Punjab and Uttar Pradesh. In addition, a local Christian missionary group bought plots in Greentown and built housing for 106 Christian families with low incomes. Greentown is thus an area of ethnic, linguistic and cultural diversity.

These factors, peculiar to Greentown, influence the structure of both family and household. It is an immigrant community; however a comparison of the population of Greentown with that of Pakistan9s urban areas shows no significant difference in age and sex structure. Where possible, families rather than individuals migrated. The single young adult male migrating for work is not a typical resident. They are young single males who have come to Lahore to find jobs, but when they live in Greentown, they live with their uncles or brothers-in-law. Although Greentown is a 8new town9, it feels settled and complete. Old men sit gossiping in the market while young boys play cricked in the fields. This heterogeneity contributes to the pleasure of living in the community.

People living in Greentown thoroughly like their community. People point out that it has good shopping, a good bus service and good neighbours. When Greentown was built, the government supplied sewage, water, and electricity. Houses are built of brick covered with concrete. They are rectangular boxes with flat roofs. Those supplied by the government have one room, a small storage room and a latrine at the corner of the lot next to the street. A number of people have added an extra room, and almost all the Muslims in Greentown have built a wall around their plots. Most of the people moving into the area moved into better housing than they left. In the twelve years we have been familiar with the community, people have become better-off financially. The original monotonous block dwellings have been individualized. Some families have added a second storey, growing vertically. People in Greentown are optimistic. Most think they have done well, and expect to do better.

In 1982, we conducted a census and socio-economic survey in Greentown. [1]This socio-economic survey supports Greentown residents9 description of themselves as 8good hardworking people9. In the survey of 457 households, 86 per cent (394) owned their own homes. They have, for the most part, permanent long term jobs. Of the major earners in the household survey, 85 per cent worked 52 weeks in 1981. They are not rich: 85 per cent of the major earners in a household earned less than Rs. 1000 a month. [2]

FAMILY STRUCTURE

People in Greentown describe the ideal family as one consisting of a man and wife and their married sons living in a single household, in harmony. Brothers should remain together, people say. United brothers are like the fingers of a single hand, strong and powerful. The 8extended family9, with married sons remaining in the household of their fathers, has obvious advantages in an agricultural community: fathers with land live with sons who labour. Access to the results of the labour of more than one adult member of a household, whether in the form of agricultural produce or wage labour, improves the living standard of the household. This is true in an urban as well as a rural setting.

At the same time, people are very aware that this causes problems: people fight. Even in a rural setting, land is almost invariably divided on the death of the father. In an urban setting, land is almost invariably divided on the death of the father. In an urban setting, without land, and possibly with disparities in salary among individual sons, there is even less economic impetus for families to remain in a single household. Individuals in Greentown, describing what they want for their own families, say that they would like their married sons, and daughters as well, to live nearby, a few houses away. They remain committed, certainly, to the united family but not the single household.

The physical provisions made by the LDA presume nuclear families, defined as consisting of one marital unit. 8One room and kitchen supplied9 would seem to discourage multi-marriage occupancy. Although people quite commonly add additional rooms to their houses, there is no statistically significant relationship between the number of rooms and the number of marital units within a household.

Large families are the rule. The average number of people in a household in Greentown is 6.5. In the socio-economic survey of 457 households [3], 170 households (37.2%) have five or fewer members; 199 households (43.5%) have six to nine members; and 87 have 10 or more members.

Despite the absence of adequate household facilities and despite the expressed preference for nuclear families, a large number of households consist of more than a man, his wife and their children. In the socio-economic survey 11 per cent of households are composed of patrilineal extended families, with a married father and at least one married son. Another 2 per cent of households are composed of married brothers. Multiple married couples within a household are, despite ideology, relatively rare. However, 46 per cent of households report more than one earner in the household.

In examining the 457 households of the 1982 survey, 256 (54.8%) consisted of a married couple living with unmarried sons and/or daughters. The other 45 per cent vary from this pattern. 52 households (11%) contained married parents and at least one married son. Another 44 households had a widowed parent of the household head as a member: more commonly, a mother (34 households) than a father. Thirty households had no married couple as members. Fourteen households were composed of married brothers. In the remaining households, additional members are primarily other relatives. Household members may be the wife of a son of the household, a nephew, an uncle or a cousin, as well as members of the household head9s nuclear family. It is more common for these relatives to be members of the head of the household9s patrilineage than that of his wife, just as it is more common for brothers (95 people) to live with the household head than sisters (37 people). But parents, siblings, aunts and uncles of the wife do join the household as well. Approximately 15 per cent of the 3008 household members surveyed are related to the wife of the household head. These households are large sprawling affairs, composed almost entirely of relatives. (In the household sample, we found nine boarders, three servants and four friends.)

These are also households at particular point in time. They divide, or grow larger. Daughters marry and leave. Sons bring wives to their parents9 house. Young men are supposed to marry when they have a good, secure job. It is quite common for a couple to begin married life with the young man9s parents, without intending to establish an extended household. The young wife can be introduced to household tasks by her mother-in-law and the couple can save money to establish their own household later. The couple leave the household, most commonly, when a younger brother of the husband marries and brings his own wife to live in the house of his father. Other family members join the household: a nephew or wife9s brother from the village comes and stays for several years, working or going to school. Elderly widowed parents join their children. Daughters (or daughters-in-law) and their children will join the household while husbands work outside Lahore. Household membership is flexible, responding to the needs of individual family members.

WORK IN GREENTOWN

Informants were asked which members of the household were employed in the week immediately prior to the survey. Table 1 lists those working by age and sex.

The majority of male unemployment is simply due to age: these people are children or students. If males under the age of fifteen are discounted, 82 per cent of males were employed in the preceding week. Far fewer women are employed; among women 15 years or older, 8 per cent worked in the preceding week.

Between the ages of 21 and 60, almost all men work, either full-time or part-time. (These figures include part-time, intermittent work as well as full-time work.) Between the same ages, few women work.

TABLE 1 : PEOPLE EMPLOYED BY SEX AND AGE

Age	0-20	21-40	41-60	61-80	80+	Total
Male
Employed	104	404	157	32	3	700
Unemployed	781	43	16	21	10	871
% employed	11.7	90.3	90.7	60.3	3
Female
Employed	13	32	14	5	0	64
Unemployed	836	358	129	34	3	1360
% employed	1.5	8.9	9.7	12.8	0

In considering the actual types of work done, 46 per cent (358 men) were categorized as working in service occupations. Little advancement occurs in such jobs. The next largest category, craftsmen, was assigned to 189 men (24.4%). Electricians, bricklayers, carpet makers and goldsmiths are all considered 8craftsmen9 in the survey, and job security and wages vary tremendously among them. Bricklayers do casual work, influenced by the weather, and can find employment for only part of the year. Carpet makers who hire other men and contract to make carpets make a good wage and have steady work. Their employees make less. Goldsmiths generally work on commission, and have high incomes and very steady work. (As long as there are prostitutes and weddings, a goldsmith will do well. Prostitutes are probably more important to the trade; a good woman marries once, but prostitutes are constantly having their jewellery re-set.) In Greentown, 10.6 per cent of the working men are shopkeepers or vendors. This is a high risk occupation, and there are more ex-shopkeepers than current ones. While one can do well, one is more likely to just get by. Very young and very old men often work as vendors, selling baked sweetcorn, fruit, soft drinks, tea, cold water, a few vegetables (or even just one) from a cart in the bazaar of hawked through the streets. It requires a small initial investment, and has equally small returns. 6.7 per cent (52 men) work in factories. This is most unpopular work. Salaries are low, the work is boring, there are very few holidays and few ways of regularly making money on the side. Office work is done by slightly more men361 (7.8%). It has many of the disadvantages of factor work. The status is higher, but the salaries are no better. Corruption and bribery improve salaries with some office work, but such jobs are rare.
In Greentown, there is no statistically significant relationship between age of earner and salary. (A comparison of mean salaries at different ages does show a slight increase3about ten rupees a year in the age groups between 20 and 60. However, the amount of variation within each group makes any sort of correlation unreliable and the small difference irrelevant.) There is no 8career structure93a man of 20 has much the same sort of work as his father of 50, and makes no more money. Since there is no relationship between age of earner and salary, more expensive points in the family cycle, when children marry and are in full time education, are not matched by the work cycle, when a middle-class Pakistani or North American might expect, due to promotions and seniority, to have a larger salary. People want to marry their children respectably, to educate at least some of their sons, and to have a secure old age. Adult sons remaining in the household and contributing their salaries to general household funds is the major way in which this is done.

The combination of more workers in the household and economies of scale does allow for substantial savings. Per capita cost decreases significantly with the number of earners in the household. This allows more disposable income for the household, and much of what is saved is used for marriages and for establishing one9s children in general. There is, however, an interesting contrast between the wages of the major household earner and those of the second worker in the household. [4] Table 2 compares the salaries of major and secondary earners by occupation.

In the socio-economic survey, informants were asked about the monthly income of the major earner in the family. They were also asked about other sources of income, including the earnings each month by all other household members. A major earner will, by definition, have the largest salary in the household. The range of salaries earned by major earners within the household shows no difference between multi-earner and single earner households. The average income of the major earner of the family is Rs. 812 a month. The average income of all other working members of a family, in contrast, is Rs. 431 a month. This discrepancy needs to be examined more closely, looking at age and occupation of secondary workers.

It might be thought that second and third workers in a household are all young, with the lower mean income the result of beginning work. This is not supported by an examination of the ages of second and third workers: out of 207 worker, 39 are less than twenty years of age, 98 are between the ages of twenty and thirty-nine, and 70 are forty or more years old. 43 per cent of second workers are between twenty and thirty-nine, and 24 are between twenty-five and thirty-four, when we might expect a man to be in permanent full-time work. With the exception of industrial workers, the mean salaries of second workers in a household are much less, and show much greater variation within the occupation, than salaries of major earners. The size of the standard deviation among second workers suggests part-time work or apprenticeship, which would pay very little. The exception of industrial workers supports this: factory work involves no apprenticeship, and one is not allowed to work part-time.

TABLE 2 : COMPARISON OF SALARY OF MAJOR AND SECONDARY EARNERS BY OCCUPATION

Occupation Mean No. of Std. Dev Mean Salary No. of Std. Dev Salary Cases 2nd Worker Cases Maj. Worker

Industrial 605.65 31 208.54 590.00 32 444.31 Service 665.67 193 272.21 382.49 68 273.57 Office 753.30 44 242.56 397.67 15 264.93 Crafts 856.10 86 328.63 277.27 22 272.87 Shop/Vend. 735.11 47 352.62 253.57 14 217.03 Police/Mil. 655.00 10 118.91 430.00 5 284.17 Prof. 735.50 8 244.06 350.00 2 70.71 Other 569.79 24 336.83 387.50 8 425.73

HOUSEHOLD STRUCTURE AND HOUSEHOLD INCOME

A comparison of families with multiple earners and single earners has somewhat surprising results. Per capita income in multi-earner households remains relatively flat, despite number of earners (see Table 3).

Multi-earner households can provide extra opportunities and savings for individuals, or they can be a response to a low income earned by the household head. There seems to be a basic minimum per capita income, and if the major earner9s income falls below this, his children (or his wife) work. It is easier for an eight or nine year old boy to earn small sums of money than for his mother to do so. In addition to working for a craftsman learning his trade, a child of this age runs errands in the market, and delivers tea to shopkeepers. An uneducated woman in Greentown can work as a part time servant in other households or make handicrafts for sale in the bazaars of Lahore. There are not many demands for servants in Greentown, and it is not a respectable thing to do. Women in Greentown, as elsewhere, differ in their sewing and embroidery skills. A willing eight year old, on the other hand, can always pick up a few rupees. Three or four rupees will buy sufficient vegetables and beans for a family of six. (It will not buy meat or staples: the flour, the oil, the spices.) Quite young children can contribute significant, although small, amounts to the household income. In the sample census, thirty boys under the age of 15 work in Greentown. In a larger family, where the major earner makes a low salary, children begin work earlier. One common strategy in such a household is for elder sons to begin work early, and concentrate education within the family on younger sons. Either the earning of a very low income by the head of the household or choosing alternatives to full time employment can lead to multi-earner households. Case studies suggest that both occur. In the three case studies that follow, earning strategies within the households, and their income structure, differ radically.

TABLE 3 : MONTHLY PER CAPITA INCOME BY NUMBER OF EARNERS IN HOUSEHOLD

Number of households % of sample Number working Average per capita income

6 1.3 0 81.66 254 54.4 1 165.19 139 29.8 2 191.46 46 9.9 3 182.23 12 2.6 4 204.12 6 1.3 5 155.55 4 0.9 6 218.05

Case Study 1

The household consisted of the father, mother and eight children. The father was an electrician, and irregularly employed. The eldest son was apprenticed to a carpet maker at the age of six. The next two sons began work at ten and eleven, apprenticed to a relative who was also an electrician. The eldest two daughters began work at fourteen, making cosmetics in a factory where only women work. (Such factories are very respectable, but the wages are low. In 1982, this factory aid Rs. 300 a month, compared to the average wage of Rs. 700 a month in factories employing male workers.) The youngest son and two younger daughters remained in school. The father of the family hoped that his youngest son would get a good education, at least a B.A. (He wanted him to be a computer programmer.) None of the children in the family were married, although marriage with a relative had been arranged for the eldest daughter. (This is a 8cheap marriage9, since relatives traditionally demand less dowry.) The family was one of the few in Greentown not to own their own house. The father intended, first, to buy a house, then to arrange marriages for his other daughters, and then marriages for his eldest sons. He thought that this could be accomplished within five years. None of his children objected to this plan.

Carpet-making pays well after an initial apprenticeship. By the age of ten, a lad can contribute to the family income. If he likes the work, and does it well, a young man can establish his own carpet frame and make a very good income by the time he is twenty-five. It is, however, not something everyone wants to do. The eldest son in this family disliked it a great deal. After his two younger brothers began work, he quit carpet making and worked as an apprentice jewel-setter. He is now trained, and working independently.

The young man9s work was an important, indeed necessary, contribution to the family income when he was quite young. (At times, he was the only person working in the family.) He could leave carpet-making only because his younger brothers9 salaries allowed him to do so: but, if his earnings supported the household, it is also true that the household supported him during his apprenticeship in a new trade.

The major earner9s income was intermittent, and his salary low. The boys of the household began work initially to pay for necessities. Now, as adult men and women of the household work, the household is more concerned with other expenses: first, the ritual expense of marriage; next, buying a house; then further marriages. When children work, the money is spent on physical necessities, but when more than one adult works the money can be spent on cultural necessities. At all but the lowest level of living, social expenses are as important as food and shelter. To be a respected member of Greentown, you must marry your children appropriately. A household of adult multi-earners is under strong social pressure to satisfy this kind of obligation.

The position of the family in the second case study is quite different from that of the carpet-maker. This family was, initially, among the better-off families in Greentown. Their financial position came, in part, from the contributions of adult sons and a daughter. As household divisions occurred, the financial position of the family changed.

Case Study 2

In January 1982, the household consisted of a man, his wife, three unmarried, employed sons (27, 25 and 18 years old) and three daughters (aged 22, 20 and 14). The two elder girls worked in a school which they had established in Greentown. The father owned two houses, the house the family lived in and the house used as a school run by his daughters, very near the household home. The man, head of the household, had retired several years earlier. He had also purchased a plot in another section of Greentown, to build a house for his eldest son on his marriage. The family had savings of more than Rs.40,000 invested in government prized bonds. The two older sons had also purchased prize bonds. The man also belonged to a kameti , a neighbourhood association for saving. Each member of the kameti paid a set amount into the fund each month, and kameti members in turn received the amount paid in. Talking about his savings, the man said that they were intended to establish his children, provide for their marriages, and take care of him and his wife in their old age.

In May 1982, the eldest son married. His wife came to live in the household, while construction began on the new house. In September 1982, the middle daughter married and moved to her husband9s house, about a mile from Greentown. (More usually, daughters marry in order of age. The eldest daughter wished to continue her education). In January 1983, the eldest married son went to Karachi to work for a year. His wife remained with her parents-in-law.

In 1984, the eldest son9s wife had her first child. Her husband returned from Karachi, and the family moved into the completed house in Greentown. The second son married, and brought his wife to live in his parents9 house.

In 1985, the father became ill, and died the following year.

In 1987, the youngest son, two daughters and mother moved into the second house, which continued to be used as a school by the two daughters of the family. The second son, his wife and children, remained in the original family house.

In October 1990, the youngest son went abroad to work. He continues to send money to his mother and sisters.

In December 1990, the youngest daughter married and moved into her husband9s household, a few doors from the school.

The original household now consists of a woman and unmarried daughter.

In eight years, household size and composition underwent dramatic fluctuations. This had resulted in economic fluctuations as well. Household composition had direct financial consequences. In 1982, the household had five employed members out of a household of eight, as well as extensive savings. The father of the family was determined to see his children educated. His eldest son had an M.A. in business studies, his second had a B.A. in business studies, and his two daughters had matriculated (completed grade 10). His eldest daughter was studying part-time for a B.A. In 1980, when the two eldest sons got good jobs, the man retired. After his retirement, he spent much of his time investigating possible marriages for his children, something he enjoyed immensely. Once his sons and daughters began working, the family invested a good part of their income in savings to cover the major expenses of marriages.

The expenses for the eldest son9s marriage in 1982 amounted to Rs. 10,000. The father had bought the Greentown plot some time earlier (for Rs. 30,000), and this and the cost of building a house (in 1982, about Rs. 30,000-50,000 in Greentown) are not included in the marriage costs. The middle daughter was married four months later, and approximately the same amount (Rs. 10,000) was spent on her dowry and on the wedding celebrations.

A wife joined the family, a daughter left: per capita income remained the same. The eldest son9s employers wanted him to work in Karachi for a year, to gain further training. Much of the eldest son9s salary was then spent on living expenses in Karachi, travelling expenses, and materials to build his house in Greentown. The disposable income of the family declined, and less could be invested in savings.

The second son was married, the eldest left home. Savings were spent, but the family was still well off. Shortly after the second son9s marriage, the father of the family became ill. He dies of cancer, a long and expensive illness. This exhausted funds saved for the dowries of his two remaining daughters. After his death, his eldest son sold his house in Greentown and moved with his family to Karachi. He gave part of the money which he had received from the sale of the house to his mother, to help pay his father9s medical expenses. He refused to contribute to the maintenance of his mother and sisters. His mother and sisters said that, with four young children, he could not be expected to. (Neighbours said that he had requested a transfer to Karachi to avoid both family arguments and family financial obligations.)

The subsequent division of the household, when the youngest son with his two sisters and mother moved into the school, caused great financial hardship. The youngest son, with his father9s support, had intended to open a small shop in Lahore. Without his father9s financial backing, this was impossible. Instead, he became an apprentice, and the family lived on the earnings from the school, about 1200 rupees a month. [5]

The youngest son9s primary financial worry during this time concerned the dowries of his two sisters. It is very difficult for a family with a single earner to save enough money to provide a dowry. A girl needs a brother as well as a father for a proper marriage.

After finishing his apprenticeship, the youngest son was relatively successful financially. He saved enough money to go abroad to work, and has continued to send money to his sister and mother. He intends to return to Lahore and open his own business.

The elder sister arranged a marriage for her younger sister. The cost of the wedding arrangements and dowry was Rs. 70,000. Her elder brother in Karachi contributed Rs. 20,000, the husband of the middle sister contributed a further Rs. 20,000 with the remainder coming from the elder sister, and from the youngest brother who was working abroad.

In this case, it is clear that a traditional extended family was not seen as desirable, or possible. It was taken for granted that sons, after a time, establish independent households. But, equally, wage-earning adult sons and daughters were expected to remain in the household and contribute to family savings which were spent primarily on them. The possibilities open to individual members of the family were considerably enhanced by the presence in the household of more than one earner.

The children were able to marry only because the family, including an affinal relative in the case of the youngest girl, contributed to their marriage expenses. The eldest son had a house only because his father had supplied a plot, and because the other working members of the household had paid his household expenses while he built his house. His father, in addition, contributed to the actual cost of building the house. (The second son has a house as well, but this was achieved with some acrimony.)

The youngest son was able to learn a trade, rather than being forced to accept any available work, because his working sisters supported the household. He contributed to one sister9s marriage. He can work abroad because his unmarried sister has assumed responsibility for his mother. This sister is not sure that she wishes to marry, and there is no pressure on her to do so. With the house, which her mother owns, she can continue her school and support both herself and her mother. Her brother is quite willing for her to remain a member of his household after he marries, and this is what she thinks she may prefer to do.

In Case Study 1, the earnings of more than one member of the family were necessary for household survival. In Case Study 2, the excess of earnings over requirements were spent on increased economic opportunities for the members of the household. In Case Study 3, household income was manipulated to allow for other sorts of opportunities to family members.

Case Study 3

A neighbour did nothing but play cricket until he was 26 years old. He then joined his elder brother at work, and his father retired and did nothing but gossip in the market place. He was one of the best cricket players in Greentown, a cricket-mad place, with five teams. Two of the teams were quite good, three were not so good. He played cricket for other teams in Lahore as well, and was paid small sums of money. He said once that the only reason he minded being poor was that he would never know how good a cricket player he could have been. 8Indulged Youth9 or 8Youth Cheated9? Both.

The chances are very good that a young man in Greentown will work very hard in a dull job making little money. It is not surprising that these young men put off working full-time as long as possible. Young men say that they do not want to marry, because then they will have to work all the time. Generally speaking, a man who begins steady full-time employment at 20 is doing no better when he is 30 years old than a friend who waited until he was 25 to begin work. Brothers in their twenties are commonly found working part-time or intermittently while remaining in their parents9 household. Multi-earner households are required because of the low rewards of some occupations. But these same multi-earner households give individuals within them some independence from the labour market. They take advantage of it. They look for better work, and better work is usually defined in terms of what you do rather than what you earn. (You make about the same amount no matter what you do.) They postpone work. If people are given very small rewards, why should they work? If the same structure that subsidizes these small rewards3households3allows it, why not play instead? Some, for a time, do. Multi-earner households can be a response to individual household members9 low earnings. But these households also allow individuals (and families) to manipulate household structure to permit themselves greater economic freedom and flexibility. We might describe these two situations as: the 8Happy Lotus Eaters9 versus the 8Ground-in-the-Dust Poor9, perhaps. These are not competing, but complementary.

WORKING WOMEN

In two of the cases discussed, women working make fundamental contributions to the family income. Working women in Pakistan are accepted at the top and bottom of the social and economic scale. Very poor women work for wages, as very poor women have always worked. Women of the upper middle and upper classes also work, as doctors, lawyers, and in other professions. In the first case study, the daughters of the family worked in a factory because their salaries were required. In the second case study, the daughters of the family began teaching because their father thought it important for his daughters to be financially independent. (His attitude was and is unusual.) For eighteen months, the women of the family supported themselves and their younger brother with very little outside help. Both of these cases are unusual.

In general, in Greentown women do not work outside the house. There are too few women working in Greentown, and the positions of those who work are too different, to make generalizations. However, one interesting point is indicated by Table 4 which shows women working by the number of couples within their households. There is no significant statistical correlation between the two variables, but women in multi-couple households do seem more likely to work.

Table 4 does not include women making handicrafts in the home. Much of the intricate embroidery sold in the expensive bazaars in Lahore is made by women in Greentown. Most informants did not regard this as 8employment9, just as, in rural areas, women are officially unemployed despite the amount of agricultural labour they contribute. Eleven households reported receiving income from 8handicrafts9 in the previous month.

TABLE 4 : NUMBER OF WORKING WOMEN BY NUMBER OF MARRIED COUPLES IN HOUSEHOLD

Number of working

women in household 0 1 2 3 4 Total

1 324 34 4 1 1 364

No. of married 2 61 12 2 0 0 75

couples in household 3 17 3 1 0 0 21

4 2 0 0 0 0 2

Total 404 49 7 1 1 462

Thirteen per cent of one-family households contain working women, 18 per cent of two-family households, and 23 per cent of three-family households. One in eight women work in households with one married couple (40 of 324), while almost one in four (18 of 80) women work in households with more than one married couple. This might be thought to be simply a relationship between total household size and women working3there are more women to work in a larger household than in a small one. There is no significant correlation between the size of the household and women within it working, however. Inspection of a cross-tabulation of women working by number of household members does not suggest this sort of pattern. (In fact, households with very small numbers of people are more likely to have a woman employed 3 these are, for the most part, female-headed households, widows or divorcees.) The amount of work required of women in a household influences this, as well as cultural objections to women working in Pakistan.

Food has to be purchased and cooked daily, clothes must be washed, usually without a machine, and ironed. Somebody has to take care of the children. A housewife in Greentown has a lot to do, work that cannot simply be ignored. Women work very hard, and their work is necessary, even if unpaid. If more than one adult woman is present in the household, other adult women can consider working for a wage.

CONCLUSION

Ideologically, people in Greentown see the patrilineal extended family as typical of their society, and it is the family structure they prefer. Practically, they expect their own children to establish new households. Discussions of family types often focus on property rights: the control and transmission of property (Owens 1971; Yanagisako 1979). There is little property to control in Greentown. Families do remain together in Greentown after the marriage of sons of the household, either as a temporary condition or permanently. But the primary economic benefit of an extended household, whether patrilineal or with other relatives, does not involve the control of property but the sharing of salaries and responsibilities.

Societies have socially appropriate definitions of family and household3whom one should live with and how one should live. Kinship is the most important determinant of household membership: we live with relatives. 8Which relatives9 shows cultural variation, but it is commonly assumed that one will live with one9s culturally relevant family. As Bender (1967) points out, family and household are logically and analytically distinct. Family obligations extend beyond household boundaries and households contain members who are not kin. However, households are most commonly based on kin groupings, and the long-term reciprocity that distinguishes kinship allows the family to use household structure in ways not possible to a household of strangers. Household structure has economic consequences, just as it responds to economic conditions. Certain wages are possible because the earners of those wages have access to other income, whether in the form of state benefits, working spouse, sharing living expenses with a room-mate or lodger, or an extended family. Households are structured within particular set of economic and social conditions. The household provides for the practical implementation of the ideal of common fraternal responsibility, the rights and obligations of kinship. Common household membership is often the only way to satisfy these obligations and obtain these rights.

A household dependent on multiple earners is one response to a specific economic configuration, most particularly wage structure, within a society. Segments of the working population do not receive a 8family wage9, often not even a 8living wage9. This wage structure is dependent on household structure. Family and household structure allow individual survival on a low wage. Equally, wages and working conditions depend upon certain family and household structures being maintained. The wages of some workers, particularly in jobs traditionally done by women, do not allow for the survival of the worker, much less a family.

In Greentown, the low earning individual supports non-paid kin by common household membership. Most children, most women, and most of the elderly are maintained within one9s household. Such people are commonly described in English as 8dependents9. The household is used to support the cost of reproducing labour, non-paid work, and those the work force is not interested in using. Just as importantly, the household supports its individual members against the outside world. Multi-earner households allow their members greater choice: the possibility of retiring early, changing occupations, or simply enjoying additional free-time. The household provides for those who cannot earn, but also for those who decline to earn. Members are less concerned with maximizing income, than the maximizing choice within a system of wage labour. The structure of the household is used to provide individual members with independence from wage labour to increase such choice.

REFERENCES

Bender, D. R. (1967). A refinement of the concept of household: family, co-residence and domestic functions. American Anthropologist 69 : 493-504.

Hackenberg, R., Murphy, A. D. and Selby, H. A. (1984). The Urban Household in Dependent Development. In : R. McC. Netting, R. R. Wilk, and E. J. Arnold (eds.), Households: Comparative and Historical Studies of the Domestic Group. Berkeley: University of California Press.

Laslett, P. (1972). Introduction. In : P. Laslett and R. Wall (eds.), Household and Family in Past Time. Cambridge: Cambridge University Press.

Owens, R. (1971). Industrialization and the Indian joint family. Ethnology 10 : 233-50.

Quadeer, M. (1983). Urban Development in the Third World: Internal Dynamics of Lahore, Pakistan. New York: Praeger.

Stack, C. B. (1974). All Our Kin: Strategies for Survival in a Black Community. New York: Harper & Row.

Yanagisako, S. J. (1979). Family and household: The analysis of domestic groups. Annual Reviews of Anthropology 8 : 161-206.

---

[1] The research on which this chapter is based was funded initially by the Fulbright-Hayes Dissertation Abroad Award. Subsequent work was funded by Tri-Council HCI/Cog.Sci. Initiative grant SPG8920734, and by Economic and Social Research Council grants R000231113, R00231953, and R000233509.

In 1982, using field assistants at the Pakistan Medical Research Center, with whom we were affiliated, we conducted a census of the area. We reported a population of 15,794 people. At the same time, a socio-economic survey of 20 per cent of these households was conducted. The sample was chosen from numbered plot maps prepared by the Lahore Development Authority. We sent two field assistants out with maps before choosing the sample; empty lots and shops were eliminated from the pool from which the sample was drawn.

[2] Rs13 = US $1 [1982].

[3] The record in the Socioeconomic Survey was based on the household; in the Census, on the individual. The records of all individuals in households in the Socioeconomic Survey were taken from the general census to provide a census of the households in the survey.

[4] Major earners and secondary earners earning more than 2000 rupees a month have been eliminated from Table 2. A few cases can skew the means and variance to such an extent that the results are misleading. (For example, by including the three office workers who make more than 2000 rupees per month in the sample, the mean is increased to 1020.75 and the standard deviation to 1399.72. The very successful are very much more successful.)

[5] In 1982, Rs.13=$1. In 1987, Rs.20=$1. In 1982, a family with an income of 1200 rupees had slightly more than the average income of 1000 rupees. In 1987, 1200 rupees was less than the average income.