Kin Investment in Wage-Labor Economies Effects on Child and Marriage Market Outcomes Mary K. Shenk University o f Washington Various human groups, from food foragers to inner-city urban Americans, have used widespread sharing of resources through kin networks as a means of buffering themselves against fluctuations in resource availability in their environments. This paper addresses the effects of progressive incorporation into a wage-labor economy on the benefits of traditional kin networks for two social classes in urban South India. Predictions regarding the effects of kin network wealth, education, and size on child and spouse characteristics and methods of financing marriages are tested using various regression techniques. Despite the rapid growth of participation in a wage-labor economy, it is found that kin network characteristics still have an important impact on investment behavior among families in Bangalore in both social classes. Network wealth is found to have a positive effect on child and spouse characteristics, and large networks are found to act as significant drains on family resources. However, the results for education are broadly consistent with an interpretation of increasing family autonomy as parents' education has a far stronger influence on child and spouse characteristics across categories than network education does. Finally, professional-class parents are found to prefer financing marriages using formal mechanisms such as savings and bank loans while working-class parents preferentially finance marriages using credit from relatives and friends. KEYWORDS:Human capital; India; Kin investment; Marriage costs; Social networks his p a p e r explores the resource-sharing practices o f groups who participate to varying degrees in a wage-labor economy. Embodied capital theory and theory from studies o f human food sharing are used to generate predic-
T
Received March 18, 2004; accepted August 4, 2004; final version received October 5, 2004.
Address all correspondence to Mary K. Shenk, Centerfor Studies in Demography and Ecology, Box 353412, University of Washington, Seattle, WA 98195. Email: mshenk@u,washington.edu Human Nature, Spring 2005, Vol. 16, No. 1, pp. 8 1 - 1 1 4 .
1045-6767/98/$6.00 = .15
82
Human Nature [ Spring 2005
tions about how the use of kin networks should alter with progressive integration into a market economy. These predictions are then tested by analyzing the differential effects of kin networks on child's and child's spouse's characteristics and the means of financing weddings using data from members of two social classes in urban South India. I will begin with an overview of relevant theory from studies of food sharing among foragers and resource sharing among agricultural and urban communities. I will then give brief summaries of embodied capital theory and the basic tenets of parental investment theory, following which I will describe the urban Indian social context as it relates to both parental investment and investment by kin. I will then integrate these disparate topics to develop hypotheses and predictions regarding the effects of kin networks on child and spouse characteristics and mechanisms of marriage financing. After summarizing the dataset and the methods of analysis, I will present my results and discuss the relevance of these findings. I will finish by summarizing my findings and commenting on their importance in a broader theoretical context. THEORETICAL
BACKGROUND
Kin Networks and Resource Sharing A large body of literature in evolutionary anthropology addresses the question of the causes of food sharing among foragers. One typical interpretation of this phenomenon is that it is a form of risk-sharing which compensates over time for unpredictability in the rate of return to foraging (particularly hunting) and temporal fluctuations in the availability of resources (e.g., Gurven 2004; Kaplan and Hill 1985; Winterhalder 1997). In this way, the resource--which in many cases is not subject to storage--is converted from a tangible but perishable physical resource into an intangible but presumably more durable resource, variously either (a) a forager's stake in expected reciprocity (i.e., the right to be shared with by others in return) or (b) some form of social status which may never be directly repaid in calories but which may have other effects on the survival or reproductive success of a forager or his/her children (Bliege Bird and Bird 1997; Kaplan and Hill 1985). Among others, Kaplan and Hill (1985) argue that the unpredictability of returns to foraging as well as temporal fluctuations in food availability are important conditions leading to the adoption of food-sharing practices. Specifically, when unpredictability in foraging returns or food availability is high, food sharing becomes a beneficial strategy that will buffer against the effects of hunger at any particular time. Conversely, when unpredictability in foraging returns or food availability is low, food sharing will not benefit those who are more efficient foragers or better at storing food. In these circumstances we
Kin Investment in Wage-Labor Economies
83
should expect food-sharing practices to be less important. Furthermore, Kaplan and Hill discuss the ability or inability to safely store food as important indicators of whether or not these fluctuations will produce sharing behavior. If storage is difficult or impossible, foragers will be exposed to strong unpredictability risks and will be strongly motivated to share. On the other hand, if food can be stored, more-successful foragers may be able to buffer themselves against unpredictable returns to foraging and may thus have a reduced motivation to share. A potentially analogous situation has been described in numerous ethnographic accounts of the economic practices of the rural and urban poor, specifically those living on the margins of modern market economies (i.e., Dohan 2003; Lewis 1959, 1970; Scheper-Hughes 1992; Stack 1974). Authors describing environments as diverse as shantytowns in periurban Brazil, villages in rural Mexico, and modern Americans living in the inner city give details of remarkably similar systems of economic obligation and exchange. Perhaps most famously discussed by Oscar Lewis (1959, 1970) as part of the "culture of poverty," these individuals and households are described as existing at the center of wide-ranging kin networks, the members of which have strong reciprocal obligations to each other with regard to many sorts of goods and services. If one family or individual has an economic surplus of any sort, that surplus is expected to be shared with members of the family network who have more immediate need of it. Given the size of these kin networks and the general poverty of their members, there are, of course, almost always relatives in need. Moreover, in these settings families are interconnected in ways which make concealment of wealth flows difficult and the social consequences of refusing to share quite problematic. In fact, people may describe themselves as being forced to share to avoid negative social repercussions (e.g., Stack 1974), an idea reminiscent of Blurton Jones's (1984) notion of tolerated theft among food foragers--that sharing occurs because of the reduced marginal utility of large resource packets and the comparatively high costs of resource defense, rather than because people are individually motivated to share. Yet social attitudes toward sharing in these cultural settings tend to emphasize generosity, and there tends to be trust that others will share if one is in need (e.g., Farrer 1996; Stack 1974). The result is that individuals effectively store or save wealth in community goodwill and obligations rather than bank accounts, believing that it is their right, in return, to ask for money from relatives if and when they need it. In contrast, members of urban middle or upper classes in these cultural contexts are typically described as not relying as heavily on social networks as a means of gaining access to resources. Instead, their patterns of resource use are often more similar to the ideal of the independent household unit common in economic theory (see Becker 1991 for a sophisticated treatment).
84
Human Nature I Spring 2005
Embodied Capital and Parental Investment Kaplan (1996; also see Kaplan and Lancaster 2000; Kaplan et al. 1995, 2002), drawing on the work of Becker (1975, 1991) and basic tenets of behavioral ecology, argues that in allocating resources to having and raising children, parents should act to maximize the total lifetime allocations of their children to their children's children. To do this, parents must optimally allocate their resources among investments in the number (quantity) and the embodied capital (quality) of their children. Under Kaplan's model, embodied capital can be any physical capacity, skills, or knowledge that increases a child's lifetime income. Income can be measured in terms of calories, cash, or any other salient currency that humans use to fund their survival and reproduction. Embodied capital is Kaplan's term, and it implies a somewhat broader and more explicitly evolutionary version of Becker's concept of human capital; the two terms will be used interchangeably in this paper, as the outcome variables discussed clearly fall under both definitions. In behavioral ecology, parental or kin investment is typically understood to be any type of investment, whether in terms of resources or care, that increases one child's survival or ability to reproduce at the cost of the ability to invest in other offspring (Trivers 1972). Investment in a child can thus include investment in food, shelter, or health care, but in a developed or developing economy the largest and most variable aspect of investment is often in human capital (in most cases education)--the skills necessary to obtain employment in competitive labor markets. In such environments, Kaplan argues, the returns to investment in human capital do not begin to diminish until very high levels, resulting in an increase in optimal per-child investment and a consequent decrease in fertility among those who have access to education and market-linked employment. Furthermore, Kaplan argues that if those with more education are more efficient at investing in the educational achievement of their own children than those with less education, more-educated people should tend to have both lower fertility and higher levels of per-child investment than lesseducated people in the same economic and cultural environment.
Parental Investment in Urban India Investment in children in India can take two primary forms: investment in a child's labor market attributes and investment in a child's ability to compete in the marriage market. I define labor market investments as those that result in higher-quality child characteristics as measured in terms of either human capital endowment (measured here by amount of education) or earning power (measured here by occupational rank). In contrast, marriage market investments are those that result in a higher-quality spouse for the child in question; in this paper spouse quality is also measured in terms of both human capital endowment (education) and earning power (occupational rank).
Kin Investment in Wage-Labor Economies
85
In urban India, the primary form of investment in sons is in education. Men are expected to be the primary (if not the exclusive) breadwinners for their families, and education is an increasingly important means of ensuring that a son has a good income. Furthermore, from the perspective of human capital theory, a son's education should not only increase his lifetime income but also make him more efficient at investing in the education of his own children. Finally, an educated son with a good income is in great demand in the marriage market, able to command an educated, beautiful, and/or wealthy bride, an expensive wedding, and potentially significant property transfers from his bride's family. Educating daughters has a somewhat different set of rationales. First, it may help the daughter's chances in the marriage market because better-educated grooms often demand better-educated brides (e.g., Caldwell et al. 1983). Second, education may enable a woman to get a job. If this is the case, her income will generally be under the control of her parents before she is married (and may be used by them to help defray the costs of her wedding) and under the control of her husband or her in-laws after marriage. Both of these reasons for educating daughters have evolved in the changing social and economic circumstances of the past several decades (Srinivas 1978, 1984), and even today working brides are an anomaly in some social classes. The second major form of parental investment in South Asia takes the form of marriage costs, a broad term that I use to apply to all types of expenses associated with weddings. In Indian society, it is virtually necessary to pay high marriage costs to get one's daughter married. This is especially true if she is to be married in a socially acceptable manner and to a socially acceptable groom. Most explanations for this "dowry marriage" system emphasize the economic dependence of women as key to the logic of the institution (e.g., Boserup 1970; Goody 1973); however, this social practice has become so normative that even women who are not in an economically dependant position are still expected to pay high marriage costs for grooms who match their status (Caldwell et al. 1983; Srinivas 1984). Although investment in marriage costs is not of direct economic benefit to a bride's parents or other kin, it is in keeping with a human capital framework to treat marriage costs as investment because they will affect a daughter's lifetime income (in the sense that she will share in the benefits of her husband's income and that of his family) and thus her ability to raise quality offspring. Parents also invest in the marriage costs of their sons, though usually significantly less than they invest in the marriage costs of daughters (Shenk 2004). These gifts and functions must be of a style that is appropriate to the status of the bride's family and the style in which the bride's family is celebrating the wedding. If a bride has wealthy parents or other relatives or a large income the wedding is likely to be lavish, and the groom's family has little choice but to match the style or lose face.
86
Human Nature ] Spring 2005
Kin Networks and Kin Investment in Urban India Traditionally in India kin are thought of as playing a significant role in the raising of any particular child. On the paternal side, these obligations are an important component of the joint family system, in which adult brothers continue to live with their parents even after their marriages. Consequently, parallel cousins often grow up in the same household, effectively as siblings (Harrell 1997). Although the joint family system is becoming less widespread, the social obligations of paternal kin toward each other's children are still significant. On the maternal side is the Indian system of dowry marriage, which has been widely practiced among higher castes in the country for the past several centuries (Srinivas 1984). This system entails not only significant economic contributions at the time of marriage but also significant, continuing economic and social obligations on the part of the bride's family to both the bride and the children that result from her marriage. While dowry (politically defined as a sum used to induce a groom into marriage) is technically illegal in modern India, gift-giving is not restricted and most families are happy to discuss details of property transfers (gold, saris, houses, furnishings, etc.) as long as undue stress is not placed on the topics of dowry "demands" or transfers of cash. Extensive ethnographic interviewing among respondents in my sample suggested that both maternal and paternal kin, especially aunts and uncles, make significant contributions of time, influence, and cash to aid both the labor market and the marriage prospects of their nieces and nephews. Common examples of labor market investment included: giving cash to the child's parents to help pay school fees, taking on a child as an apprentice in one's occupation, caring for a child while parents are working or ill, helping a child with his or her schoolwork, helping to pay a child's medical bills, and using one's influence to help a child obtain a job or school admission. Common examples of marriage market investment were: aiding in finding and/or vetting prospective spouses, aiding with marriage negotiations, giving gifts or loans to help with marriage expenses, and aiding with time-consuming and ritually intensive wedding preparations. In addition to these direct forms of aid, the characteristics and behavior of the members of one's kin network may indirectly influence a child's position in the marriage market. Wealthy or otherwise high-status kin may aid a child merely by their presence at marriage-related functions, while impoverished or low-status relatives may make a child unattractive to prospective spouses who fear loss of either status or resources by establishing a kin relationship with them through marriage. Social Classes, and Resource Sharing in an Emerging Wage-Labor Economy The research on which this paper is based was conducted in the city of Bangalore, a large urban center in South India. Though India's emerging wage-
Kin Investment in Wage-Labor Economies
87
labor economy has included an accelerating proportion of the population over the past hundred years or so, the transformation to a Western-style labor market is far from complete (Srinivas 1984; Visaria 1993). This means that while some groups emphasize human capital investment, other groups, whose jobs depend on different types of human capital (or on other types of capital altogether), do not. Though urban parts of the nation have had Western-style schools and education-based job markets for a century or more, it is only in the past several decades that these institutions have been open to participation by a significant and growing portion of the population (Srinivas 1984; Visaria 1993). The people I interviewed differ significantly in their access to education, cash, credit, and stable incomes; consequently, I would argue that in urban India members of different social classes effectively function at different levels of integration into the market economy. Some have access to many economic advantages and safeguards, whereas for others their social network is their only significant economic safety net. Specifically, in my study area, among members of what I will call the working class, education is limited, jobs are typically skills-based and involve apprenticeships rather than formal educational training, people earn small to moderate incomes, cash flow is highly unpredictable over time, and access to large amounts of cash or credit is frequently limited to informal mechanisms, the most common (and least risky) of which is taking "credit" from relatives and friends. In contrast, among a group I will call the professional class, education is common, occupations are education-based and higher-paying, income is typically stable over time, and access to both secure savings and credit through regulated institutions (e.g., banks and credit unions) is widespread. The characteristics of these two social classes have implications for their optimal resource sharing and kin investment behavior. Following the framework laid out by Kaplan and Hill (1985), which was discussed above, it can be argued that economic conditions among the working class in urban South India parallel those of food-sharing foragers in two primary ways. First, their incomes can be quite unpredictable over time. For instance, whether their income comes from selling produce from a cart on the street, working as a laborer on construction sites, or working as an assistant in a shop, employment is often at the daily discretion of a wealthier employer who has little or no legal obligation to continue to pay for the worker's services from day to day. Moreover, seasonal and market fluctuations of various sorts can substantially change the amount of money earned from one month or year to the next. Second, their lack of educational attainment and generally small amounts of cash make participation in institutions of economic storage (banks) problematic. Thus they have little choice but to rely on social networks to help cover unusual expenses. Members of the professional class, however, have the advantages of both jobs with stable salaries and access to banks and other means of accumulating income over time. They have less risk over time and more ability to
88
Human Nature / Spring 2005
store wealth; thus, they are in an economic position consistent with a strategy of reducing sharing with a larger social network. Thus we can predict that integration into a wage-labor economy should result in a decrease in the importance of kin network characteristics vis-?~-vis family characteristics in their effects on many types of investment. Specifically, in the professional class, where income is education-based and reasonably stable, and where access to credit is possible, characteristics of the family and/or individual child should have generally stronger effects on a child's or child's spouse's characteristics than should kin network characteristics. In contrast, among the working class where education is less common, jobs are not education-based, and access to credit through formal means is problematic, family network variables should continue to have a generally greater effect on child's and child's spouse's characteristics even if some family and individual characteristics also become important. These predictions are summarized under Item 1 in Table 1. Kin Network Characteristics and Labor Market Investment
For the purposes of this paper, three characteristics of kin networks will be examined: their aggregate wealth (as measured by adding the occupational ranks of their members), their aggregate level of human capital (as measured by adding the educational levels of their members), and their size. Each of these characteristics is predicted to have separate effects on kin investment behavior, many of which differ by social class. Predictions are discussed below and summarized in Items 2-4 in Table 1. First, as human capital differentials increase and are increasingly associated with differences in income, greater wealth differentials will arise between both individuals and kin networks. However, wealth of the kin network should remain an important predictor of outcome variables long after the effects of human capital have diminished because wealth is able to circulate more freely through social networks than is human capital. On the other hand, in a market economy it is possible to maintain wealth over time by storing it in investments and banks; it is no longer necessary to store it in the form of reciprocal goodwill in the kin network. Thus, we can predict that when kin networks become wealthy enough that their members are individually able to rely on stable incomes and the ability to access credit reliably, their individual motivations for maintaining economic obligations to the kin network will decrease, causing a decrease in their expected effects on child outcomes. Although we can predict that wealth of the kin network should remain an important predictor in both the working class and the professional class, the effects should be felt more strongly in the working class where there is reduced access to banks and thus greater need for wealth to circulate. Second, since human capital investment is a labor-intensive process involving significant amounts of direct contact (Kaplan 1996; Kaplan and Lancaster
~
,~^~ Q- ""
0
~.~ 9
rJl
.o~
~-~ ~
.~
~-~
~ ~.~ e~1.~ ~.~ ~ a~-~ ~_~
~ 0 ~
~o~
=,~
te r
r,.) 0
~ e.,O
.~.=
~'.~
o
~ ~.-~ 9~t ~
. , ~ ._- ,~
~ ~.~ ~3
~
0
~.~ e
~
.,,
e
.~ 9~ ~_~ i
o
.0
" ~ 9=0
9
'
.~ . 0~
~oo
'~2
~,.
=
9
..~ ~ ..~ 9
~= ~
0
"~' ~ ~=
9~
~
o
"~,.~
o
~
=
~ = ",~E.~0
9
.=~
9= ~ =.'~
o F~
L~
oo
t;
~.
Z 0
"G
< >
-E.~
0
-'=~
r
=..
~=~'B.
_
,.o
~0 " ~
=l
~ ~-r-,
~..~
9
~
.~.~,~.~
~
=-.=
T
Received March 18, 2004; accepted August 4, 2004; final version received October 5, 2004.
Address all correspondence to Mary K. Shenk, Centerfor Studies in Demography and Ecology, Box 353412, University of Washington, Seattle, WA 98195. Email: mshenk@u,washington.edu Human Nature, Spring 2005, Vol. 16, No. 1, pp. 8 1 - 1 1 4 .
1045-6767/98/$6.00 = .15
82
Human Nature [ Spring 2005
tions about how the use of kin networks should alter with progressive integration into a market economy. These predictions are then tested by analyzing the differential effects of kin networks on child's and child's spouse's characteristics and the means of financing weddings using data from members of two social classes in urban South India. I will begin with an overview of relevant theory from studies of food sharing among foragers and resource sharing among agricultural and urban communities. I will then give brief summaries of embodied capital theory and the basic tenets of parental investment theory, following which I will describe the urban Indian social context as it relates to both parental investment and investment by kin. I will then integrate these disparate topics to develop hypotheses and predictions regarding the effects of kin networks on child and spouse characteristics and mechanisms of marriage financing. After summarizing the dataset and the methods of analysis, I will present my results and discuss the relevance of these findings. I will finish by summarizing my findings and commenting on their importance in a broader theoretical context. THEORETICAL
BACKGROUND
Kin Networks and Resource Sharing A large body of literature in evolutionary anthropology addresses the question of the causes of food sharing among foragers. One typical interpretation of this phenomenon is that it is a form of risk-sharing which compensates over time for unpredictability in the rate of return to foraging (particularly hunting) and temporal fluctuations in the availability of resources (e.g., Gurven 2004; Kaplan and Hill 1985; Winterhalder 1997). In this way, the resource--which in many cases is not subject to storage--is converted from a tangible but perishable physical resource into an intangible but presumably more durable resource, variously either (a) a forager's stake in expected reciprocity (i.e., the right to be shared with by others in return) or (b) some form of social status which may never be directly repaid in calories but which may have other effects on the survival or reproductive success of a forager or his/her children (Bliege Bird and Bird 1997; Kaplan and Hill 1985). Among others, Kaplan and Hill (1985) argue that the unpredictability of returns to foraging as well as temporal fluctuations in food availability are important conditions leading to the adoption of food-sharing practices. Specifically, when unpredictability in foraging returns or food availability is high, food sharing becomes a beneficial strategy that will buffer against the effects of hunger at any particular time. Conversely, when unpredictability in foraging returns or food availability is low, food sharing will not benefit those who are more efficient foragers or better at storing food. In these circumstances we
Kin Investment in Wage-Labor Economies
83
should expect food-sharing practices to be less important. Furthermore, Kaplan and Hill discuss the ability or inability to safely store food as important indicators of whether or not these fluctuations will produce sharing behavior. If storage is difficult or impossible, foragers will be exposed to strong unpredictability risks and will be strongly motivated to share. On the other hand, if food can be stored, more-successful foragers may be able to buffer themselves against unpredictable returns to foraging and may thus have a reduced motivation to share. A potentially analogous situation has been described in numerous ethnographic accounts of the economic practices of the rural and urban poor, specifically those living on the margins of modern market economies (i.e., Dohan 2003; Lewis 1959, 1970; Scheper-Hughes 1992; Stack 1974). Authors describing environments as diverse as shantytowns in periurban Brazil, villages in rural Mexico, and modern Americans living in the inner city give details of remarkably similar systems of economic obligation and exchange. Perhaps most famously discussed by Oscar Lewis (1959, 1970) as part of the "culture of poverty," these individuals and households are described as existing at the center of wide-ranging kin networks, the members of which have strong reciprocal obligations to each other with regard to many sorts of goods and services. If one family or individual has an economic surplus of any sort, that surplus is expected to be shared with members of the family network who have more immediate need of it. Given the size of these kin networks and the general poverty of their members, there are, of course, almost always relatives in need. Moreover, in these settings families are interconnected in ways which make concealment of wealth flows difficult and the social consequences of refusing to share quite problematic. In fact, people may describe themselves as being forced to share to avoid negative social repercussions (e.g., Stack 1974), an idea reminiscent of Blurton Jones's (1984) notion of tolerated theft among food foragers--that sharing occurs because of the reduced marginal utility of large resource packets and the comparatively high costs of resource defense, rather than because people are individually motivated to share. Yet social attitudes toward sharing in these cultural settings tend to emphasize generosity, and there tends to be trust that others will share if one is in need (e.g., Farrer 1996; Stack 1974). The result is that individuals effectively store or save wealth in community goodwill and obligations rather than bank accounts, believing that it is their right, in return, to ask for money from relatives if and when they need it. In contrast, members of urban middle or upper classes in these cultural contexts are typically described as not relying as heavily on social networks as a means of gaining access to resources. Instead, their patterns of resource use are often more similar to the ideal of the independent household unit common in economic theory (see Becker 1991 for a sophisticated treatment).
84
Human Nature I Spring 2005
Embodied Capital and Parental Investment Kaplan (1996; also see Kaplan and Lancaster 2000; Kaplan et al. 1995, 2002), drawing on the work of Becker (1975, 1991) and basic tenets of behavioral ecology, argues that in allocating resources to having and raising children, parents should act to maximize the total lifetime allocations of their children to their children's children. To do this, parents must optimally allocate their resources among investments in the number (quantity) and the embodied capital (quality) of their children. Under Kaplan's model, embodied capital can be any physical capacity, skills, or knowledge that increases a child's lifetime income. Income can be measured in terms of calories, cash, or any other salient currency that humans use to fund their survival and reproduction. Embodied capital is Kaplan's term, and it implies a somewhat broader and more explicitly evolutionary version of Becker's concept of human capital; the two terms will be used interchangeably in this paper, as the outcome variables discussed clearly fall under both definitions. In behavioral ecology, parental or kin investment is typically understood to be any type of investment, whether in terms of resources or care, that increases one child's survival or ability to reproduce at the cost of the ability to invest in other offspring (Trivers 1972). Investment in a child can thus include investment in food, shelter, or health care, but in a developed or developing economy the largest and most variable aspect of investment is often in human capital (in most cases education)--the skills necessary to obtain employment in competitive labor markets. In such environments, Kaplan argues, the returns to investment in human capital do not begin to diminish until very high levels, resulting in an increase in optimal per-child investment and a consequent decrease in fertility among those who have access to education and market-linked employment. Furthermore, Kaplan argues that if those with more education are more efficient at investing in the educational achievement of their own children than those with less education, more-educated people should tend to have both lower fertility and higher levels of per-child investment than lesseducated people in the same economic and cultural environment.
Parental Investment in Urban India Investment in children in India can take two primary forms: investment in a child's labor market attributes and investment in a child's ability to compete in the marriage market. I define labor market investments as those that result in higher-quality child characteristics as measured in terms of either human capital endowment (measured here by amount of education) or earning power (measured here by occupational rank). In contrast, marriage market investments are those that result in a higher-quality spouse for the child in question; in this paper spouse quality is also measured in terms of both human capital endowment (education) and earning power (occupational rank).
Kin Investment in Wage-Labor Economies
85
In urban India, the primary form of investment in sons is in education. Men are expected to be the primary (if not the exclusive) breadwinners for their families, and education is an increasingly important means of ensuring that a son has a good income. Furthermore, from the perspective of human capital theory, a son's education should not only increase his lifetime income but also make him more efficient at investing in the education of his own children. Finally, an educated son with a good income is in great demand in the marriage market, able to command an educated, beautiful, and/or wealthy bride, an expensive wedding, and potentially significant property transfers from his bride's family. Educating daughters has a somewhat different set of rationales. First, it may help the daughter's chances in the marriage market because better-educated grooms often demand better-educated brides (e.g., Caldwell et al. 1983). Second, education may enable a woman to get a job. If this is the case, her income will generally be under the control of her parents before she is married (and may be used by them to help defray the costs of her wedding) and under the control of her husband or her in-laws after marriage. Both of these reasons for educating daughters have evolved in the changing social and economic circumstances of the past several decades (Srinivas 1978, 1984), and even today working brides are an anomaly in some social classes. The second major form of parental investment in South Asia takes the form of marriage costs, a broad term that I use to apply to all types of expenses associated with weddings. In Indian society, it is virtually necessary to pay high marriage costs to get one's daughter married. This is especially true if she is to be married in a socially acceptable manner and to a socially acceptable groom. Most explanations for this "dowry marriage" system emphasize the economic dependence of women as key to the logic of the institution (e.g., Boserup 1970; Goody 1973); however, this social practice has become so normative that even women who are not in an economically dependant position are still expected to pay high marriage costs for grooms who match their status (Caldwell et al. 1983; Srinivas 1984). Although investment in marriage costs is not of direct economic benefit to a bride's parents or other kin, it is in keeping with a human capital framework to treat marriage costs as investment because they will affect a daughter's lifetime income (in the sense that she will share in the benefits of her husband's income and that of his family) and thus her ability to raise quality offspring. Parents also invest in the marriage costs of their sons, though usually significantly less than they invest in the marriage costs of daughters (Shenk 2004). These gifts and functions must be of a style that is appropriate to the status of the bride's family and the style in which the bride's family is celebrating the wedding. If a bride has wealthy parents or other relatives or a large income the wedding is likely to be lavish, and the groom's family has little choice but to match the style or lose face.
86
Human Nature ] Spring 2005
Kin Networks and Kin Investment in Urban India Traditionally in India kin are thought of as playing a significant role in the raising of any particular child. On the paternal side, these obligations are an important component of the joint family system, in which adult brothers continue to live with their parents even after their marriages. Consequently, parallel cousins often grow up in the same household, effectively as siblings (Harrell 1997). Although the joint family system is becoming less widespread, the social obligations of paternal kin toward each other's children are still significant. On the maternal side is the Indian system of dowry marriage, which has been widely practiced among higher castes in the country for the past several centuries (Srinivas 1984). This system entails not only significant economic contributions at the time of marriage but also significant, continuing economic and social obligations on the part of the bride's family to both the bride and the children that result from her marriage. While dowry (politically defined as a sum used to induce a groom into marriage) is technically illegal in modern India, gift-giving is not restricted and most families are happy to discuss details of property transfers (gold, saris, houses, furnishings, etc.) as long as undue stress is not placed on the topics of dowry "demands" or transfers of cash. Extensive ethnographic interviewing among respondents in my sample suggested that both maternal and paternal kin, especially aunts and uncles, make significant contributions of time, influence, and cash to aid both the labor market and the marriage prospects of their nieces and nephews. Common examples of labor market investment included: giving cash to the child's parents to help pay school fees, taking on a child as an apprentice in one's occupation, caring for a child while parents are working or ill, helping a child with his or her schoolwork, helping to pay a child's medical bills, and using one's influence to help a child obtain a job or school admission. Common examples of marriage market investment were: aiding in finding and/or vetting prospective spouses, aiding with marriage negotiations, giving gifts or loans to help with marriage expenses, and aiding with time-consuming and ritually intensive wedding preparations. In addition to these direct forms of aid, the characteristics and behavior of the members of one's kin network may indirectly influence a child's position in the marriage market. Wealthy or otherwise high-status kin may aid a child merely by their presence at marriage-related functions, while impoverished or low-status relatives may make a child unattractive to prospective spouses who fear loss of either status or resources by establishing a kin relationship with them through marriage. Social Classes, and Resource Sharing in an Emerging Wage-Labor Economy The research on which this paper is based was conducted in the city of Bangalore, a large urban center in South India. Though India's emerging wage-
Kin Investment in Wage-Labor Economies
87
labor economy has included an accelerating proportion of the population over the past hundred years or so, the transformation to a Western-style labor market is far from complete (Srinivas 1984; Visaria 1993). This means that while some groups emphasize human capital investment, other groups, whose jobs depend on different types of human capital (or on other types of capital altogether), do not. Though urban parts of the nation have had Western-style schools and education-based job markets for a century or more, it is only in the past several decades that these institutions have been open to participation by a significant and growing portion of the population (Srinivas 1984; Visaria 1993). The people I interviewed differ significantly in their access to education, cash, credit, and stable incomes; consequently, I would argue that in urban India members of different social classes effectively function at different levels of integration into the market economy. Some have access to many economic advantages and safeguards, whereas for others their social network is their only significant economic safety net. Specifically, in my study area, among members of what I will call the working class, education is limited, jobs are typically skills-based and involve apprenticeships rather than formal educational training, people earn small to moderate incomes, cash flow is highly unpredictable over time, and access to large amounts of cash or credit is frequently limited to informal mechanisms, the most common (and least risky) of which is taking "credit" from relatives and friends. In contrast, among a group I will call the professional class, education is common, occupations are education-based and higher-paying, income is typically stable over time, and access to both secure savings and credit through regulated institutions (e.g., banks and credit unions) is widespread. The characteristics of these two social classes have implications for their optimal resource sharing and kin investment behavior. Following the framework laid out by Kaplan and Hill (1985), which was discussed above, it can be argued that economic conditions among the working class in urban South India parallel those of food-sharing foragers in two primary ways. First, their incomes can be quite unpredictable over time. For instance, whether their income comes from selling produce from a cart on the street, working as a laborer on construction sites, or working as an assistant in a shop, employment is often at the daily discretion of a wealthier employer who has little or no legal obligation to continue to pay for the worker's services from day to day. Moreover, seasonal and market fluctuations of various sorts can substantially change the amount of money earned from one month or year to the next. Second, their lack of educational attainment and generally small amounts of cash make participation in institutions of economic storage (banks) problematic. Thus they have little choice but to rely on social networks to help cover unusual expenses. Members of the professional class, however, have the advantages of both jobs with stable salaries and access to banks and other means of accumulating income over time. They have less risk over time and more ability to
88
Human Nature / Spring 2005
store wealth; thus, they are in an economic position consistent with a strategy of reducing sharing with a larger social network. Thus we can predict that integration into a wage-labor economy should result in a decrease in the importance of kin network characteristics vis-?~-vis family characteristics in their effects on many types of investment. Specifically, in the professional class, where income is education-based and reasonably stable, and where access to credit is possible, characteristics of the family and/or individual child should have generally stronger effects on a child's or child's spouse's characteristics than should kin network characteristics. In contrast, among the working class where education is less common, jobs are not education-based, and access to credit through formal means is problematic, family network variables should continue to have a generally greater effect on child's and child's spouse's characteristics even if some family and individual characteristics also become important. These predictions are summarized under Item 1 in Table 1. Kin Network Characteristics and Labor Market Investment
For the purposes of this paper, three characteristics of kin networks will be examined: their aggregate wealth (as measured by adding the occupational ranks of their members), their aggregate level of human capital (as measured by adding the educational levels of their members), and their size. Each of these characteristics is predicted to have separate effects on kin investment behavior, many of which differ by social class. Predictions are discussed below and summarized in Items 2-4 in Table 1. First, as human capital differentials increase and are increasingly associated with differences in income, greater wealth differentials will arise between both individuals and kin networks. However, wealth of the kin network should remain an important predictor of outcome variables long after the effects of human capital have diminished because wealth is able to circulate more freely through social networks than is human capital. On the other hand, in a market economy it is possible to maintain wealth over time by storing it in investments and banks; it is no longer necessary to store it in the form of reciprocal goodwill in the kin network. Thus, we can predict that when kin networks become wealthy enough that their members are individually able to rely on stable incomes and the ability to access credit reliably, their individual motivations for maintaining economic obligations to the kin network will decrease, causing a decrease in their expected effects on child outcomes. Although we can predict that wealth of the kin network should remain an important predictor in both the working class and the professional class, the effects should be felt more strongly in the working class where there is reduced access to banks and thus greater need for wealth to circulate. Second, since human capital investment is a labor-intensive process involving significant amounts of direct contact (Kaplan 1996; Kaplan and Lancaster
~
,~^~ Q- ""
0
~.~ 9
rJl
.o~
~-~ ~
.~
~-~
~ ~.~ e~1.~ ~.~ ~ a~-~ ~_~
~ 0 ~
~o~
=,~
te r
r,.) 0
~ e.,O
.~.=
~'.~
o
~ ~.-~ 9~t ~
. , ~ ._- ,~
~ ~.~ ~3
~
0
~.~ e
~
.,,
e
.~ 9~ ~_~ i
o
.0
" ~ 9=0
9
'
.~ . 0~
~oo
'~2
~,.
=
9
..~ ~ ..~ 9
~= ~
0
"~' ~ ~=
9~
~
o
"~,.~
o
~
=
~ = ",~E.~0
9
.=~
9= ~ =.'~
o F~
L~
oo
t;
~.
Z 0
"G
< >
-E.~
0
-'=~
r
=..
~=~'B.
_
,.o
~0 " ~
=l
~ ~-r-,
~..~
9
~
.~.~,~.~
~
=-.=
