The pervasive effects of family on entrepreneurship

Introduction

One hundred years ago, ‘‘business’’ meant ‘‘family business,’’ and thus the adjective ‘‘family’’ was redundant. In the interim, the two social institutions have become more highly differentiated from each other. Today, scholars studying ‘‘family business’’ feel compelled to use the adjective ‘‘family,’’ even though they note that 90–98% of all businesses owned by households are family businesses, using the broadest of definitions (Heck and Trent, 1999).

Why is this the case? We believe the massive sociohistorical changes of the past century have led us to think of the two institutions as disconnected systems, needlessly fragmenting the study of each. This distinction has become enshrined in academia, as the scholarly study of ‘‘families’’ and ‘‘business’’ now takes place in different departments, even colleges. As Stafford et al. (1999, p. 198) noted: ‘‘The prevailing view claims that families and businesses are believed to be two ‘naturally separate’ institutions or systems.’’

Although understandable in terms of disciplinary identities and career needs, this unwarranted separation has blinded scholars to the significant causal processes connecting family systems and entrepreneurial phenomena. This separation persists despite observations that ‘‘entrepreneurs are usually family members manoeuvring in concert or disharmony with an array of other family members’’ (Heck and Trent, 1999, p. 209). Building on the pioneering work of others (Heck, 1998; Rosenblatt et al., 1985; Stafford et al., 1999), our primary objective is to encourage scholars to embrace a family embeddedness perspective on entrepreneurship, in which family factors figure more prominently in conceptual models and empirical investigations. To establish the case that families do, in fact, warrant greater research attention, we describe how the processes involved in two fundamental entrepreneurial phenomena—the emergence of new business opportunities and the emergence of new business entities (Davidsson et al., 2001; Shane and Venkataraman, 2000)—are inextricably linked to transformations in family composition and family

members’ roles and relationships.

Research on opportunity and venture emergence: the neglect of family embeddedness After years of vigorous debate, a growing consensus holds that entrepreneurship can be defined as the process by which people discover and exploit new business opportunities, often through the creation of new business ventures (Davidsson and Wiklund, 2001; Shane and Venkataraman, 2000; Timmons, 1999). As noted by Davidsson et al. (2001), for example, the field has converged around the view that entrepreneurship is about emergence, albeit with some scholars emphasizing the emergence of entrepreneurial opportunities and others emphasizing the emergence of new organizations. Gartner (2001) argued that entrepreneurship

researchers were studying such a diverse set of organizational forms that they were forced to use divergent theories. He suggested that entrepreneurship research should be organized into distinct communities of practice, with specific research questions guiding them, including the area of family businesses. Although research on the emergence of new business opportunities and the emergence of new business ventures has grown considerably over the past two decades, scholars within both streams have paid comparatively little attention to the role of the family (Stafford et al., 1999; Upton and Heck, 1997).

New business opportunities

Opportunity emergence

The lack of attention to families is particularly evident in the literature on the genesis of entrepreneurial opportunities. Whether interested primarily in the emergence of new products and markets (Shane and Venkataraman, 2000; Venkataraman, 1997) or in the emergence of new processes and ways of organizing (Abrahamson, 1996; Dijksterhuis et al., 1999; Lewin et al., 1999), scholars posit that entrepreneurial opportunities are spawned when environmental changes create information asymmetries, gaps, or ‘‘other vacuums in an industry’’ (Timmons, 1999, p. 81). Although some of these researchers include demographic and social changes as environmental triggers of entrepreneurial opportunities, most work has emphasized the effects of technological, regulatory, political, and economic changes. Scholars often

assert that technological change drives entrepreneurship (Shane, 2000), in spite of Drucker’s (1985, p. 35) assertion that demographic change is a more reliable and predictable source. Interestingly, even when demographic change is presented as a trigger of entrepreneurial opportunities, transformations within the family system are rarely mentioned.

Opportunity recognition

In addition to being absent from writing on the emergence of entrepreneurial opportunities, family considerations rarely appear in research on why, when, and how these opportunities are discovered by some individuals but not others. Shane and Venkataraman (2000) persuasively argued that an individual’s idiosyncratic prior knowledge strongly influences the process of opportunity recognition. Rather than being perfectly distributed, information about ‘‘underutilized resources, new technology, unsated demand, and political and regulatory shifts’’ is dispersed ‘‘according to the idiosyncratic life circumstances of each person in the population’’ (Shane and Venkataraman, 2000, p. 222). Studies have tended to focus on the

idiosyncratic knowledge obtained through prior work experience and education (see, for example, Shane, 2000). Although personal events are often acknowledged as another source of idiosyncratic knowledge (Venkataraman, 1997), researchers have not yet considered how family embeddedness can trigger events that stimulate the recognition of entrepreneurial opportunities.

New business ventures

Venture creation decision

Research on the emergence of new business ventures exhibits a similar lack of attention on families. The classic perspective on venture creation, for example, focused primarily on the traits and dispositions of founders (Aldrich and Wiedenmayer, 1993). In light of minimal empirical support for the trait approach (Gartner, 1988), researchers have looked in different directions for insights into the venture formation process. Some have adopted a cognitive perspective (Baron, 1998, 2000; Herron and Sapienza, 1992; Shaver and Scott, 1991; Simon et al., 1999), whereas others have adopted an ecological approach (Aldrich, 1999, pp. 75–112; Mezias and Kuperman, 2001). A third group favors a social embeddedness stance (Aldrich and Zimmer, 1986; Cramton, 1993; Larson and Starr, 1993). The embeddedness perspective argues that people are not atomized decision-makers, but rather, are implicated in networks of social relations. Thus, individuals do not decide to start a business in a vacuum; instead, they ‘‘consult and are subtly influenced by significant others in their environment’’ (Aldrich and Zimmer, 1986, p. 6). Although the social embeddedness approach on organizational emergence ‘‘appreciates the embeddedness of economic relationships within social settings (Larson and Starr, 1993, p. 11), it has paid little attention to the influence of one fundamental social institution—the family. As Cramton (1993, p. 233) noted, researchers have not yet focused their attention on how a new venture might spring from family relationships.

Resource mobilization

Much research on the resource mobilization process also reflects a social embeddedness approach, emphasizing the importance of founders’ social ties in constructing a firm’s base of financial, physical, human, and other resources (see Brush et al., 2001 for a review). Within this literature, some scholars have explicitly highlighted the role of family connections. Aldrich (1999) and Aldrich and Zimmer (1986), for example, argued in support of the resources provided by the ‘‘strong ties’’ of family members. Similarly, Starr and MacMillan (1990) suggested that kinship ties—in addition to previous work relationships, volunteer connections, and community ties—‘‘lay the groundwork for new ventures’’ (p. 81). More recently, Chrisman et al. (2003) asserted that family represents a critical and often used resource for startups. Nevertheless, very few studies have expanded upon the role that families play in resource mobilization.

The neglect of family considerations in research on the new venture creation process is rather surprising, particularly in light of three sets of empirical findings. First, a sizeable proportion of new organizations are founded by two or more related individuals. Ruef et al. (2002), for example, found that married couples or cohabitating partners constituted about one-fourth of the nascent entrepreneurial teams in their nationally representative sample, and kin ties were involved in another 27% of the teams. Similarly, Reynolds and White (1997) found that people related by marriage or kinship initiated one-third of the start-ups in their sample. Second, Cramton’s (1993) case study provided compelling evidence that organizational foundings may represent responses to changing family relationships rather than outcomes of the rational assessments of discovered economic opportunities. Third, several studies indicate that, during the start-up process, family plays an important role in the mobilization of financial resources (e.g., Aldrich and Waldinger, 1990; Steier and Greenwood, 2000), the provision of human resources (Aldrich and Langton, 1998), and physical resources in the form of space in the family household (U.S. Bureau of Labor Statistics, 2002).

In sum, mounting empirical evidence suggests that families play an important role in the venture creation process and thus deserve greater consideration in the entrepreneurship literature. In the following section, we identify the major transformations in the North American family system over the past half-century and discuss the relevance of these trends for the entrepreneurial phenomena of opportunity emergence, opportunity recognition, venture creation decisions, and resource mobilization processes.

Sociohistorical changes in the family system: implications for opportunity and venture emergence

In the mid-20th century, ‘‘family’’ usually meant a nuclear two-generational group with parents and children sharing the same household. Few women worked outside the home, children were in school, and, if they worked, contributed their wages to the family purse. Strong ties bound multiple generations together, and limited geographic mobility kept many extended kinship groups in the same or nearby communities. In North America at the beginning of the 21st century, almost everything has changed. Below, we describe two sets of the most significant transformations in the family system. The first set pertains to changes in the composition of households, both family and nonfamily, and the factors contributing to these compositional transformations. The second set pertains to changes in the roles and

relationships of family members. Table 1 lists the trends we discuss. Table 2 summarizes examples of how these changes can influence, or have already influenced, the emergence of new business opportunities and new business ventures. In organizing our review, we were guided by research and writing from a life course perspective, which provides a model of how to study family and workplace interaction (Elder et al., in press; Moen, 1998).

The life course perspective focuses on movements into and out of roles and relationships, such as moving from school into employment and marriage. As an inherently dynamic approach to study social life, the life course perspective ‘‘reflects the interweave of work, family, and community role trajectories, the interdependencies of paths among family members, and the changing circumstances and options of both families and family businesses’’ (Moen, 1998, p. 16). In our review, we thus focus on the timing of events, within the life course of individuals and their families. We also focus on process, emphasizing

transitions into and out of difficult life events. Finally, we focus on the changing context within which families and businesses are embedded, noting the importance of major historical changes over the past half century.

Trends in family composition

Characteristics of families: prevalence, nature, and size Between 1960 and 1999, the number of households in the United States roughly doubled, from about 53 million to about 104 million households (U.S. Bureau of the Census, 2000b). As the population grew, significant changes occurred in household composition (see Fig. 1). In 1960, families constituted about 85% of all households, as defined by the Census Bureau. By 1999, families made up only 69% of households, and nonfamily households had grown from only 15% to about 31% of all households. Single-person households comprised the great majority of the nonfamily households, and cohabiting couples constituted most of the remaining households. One consequence of the increase in single-person households was a decline in average household size from 3.33 in 1960 to 2.61 in 1999 (U.S. Bureau of the Census, 2000a).

With respect to opportunity emergence, shrinking household size is clearly creating markets for new products and services. Proctor & Gamble, for example, recently launched ‘‘Folger’s Singles,’’ a single-serving coffee bag for people who live alone and thus do not need to make a full pot (Solomon et al., 2002, p. 410). Furthermore, the substantial increase in the number of single-person households ‘‘suggests that apartments, appliances, and food containers should be produced in sizes appropriate for single individuals’’ (Hawkins et al., 2001, p. 195). This trend is also expected to generate an increased demand for social and recreational club memberships as a means of alleviating the isolation associated with living alone (Sheth et al., 1999, p. 264).

As for venture creation, shrinking family size may both facilitate and impede the genesis of nascent entrepreneurs, i.e., individuals who are taking action to start their own business. On the one hand, people from smaller-sized families may perceive venture creation as less risky. Such individuals may find it easier to forego their salary and launch a business because they are not also financially responsible for a spouse, children, and/or aging parents. Moreover, these individuals will have fewer kin members in their discussion networks who might talk them out of becoming an entrepreneur. Research indicates that the greater the proportion of kin in a nascent entrepreneur’s discussion network, the lower the odds of that person actually starting a business (Renzulli et al., 2000). On the other hand, individuals from smaller-sized families may perceive that they have inadequate potential resources available from kin members, and thus decide against starting their own firm.

Shrinking family size clearly has negative implications for the resource mobilization process, particularly for securing human resources. When it comes to attracting employees, many entrepreneurs rely on family members, whether paid or unpaid. Aldrich and Langton (1998), for example, found that 25% of the firms in their sample employed family members at the time of start-up. Although not restricted to start-ups, Heck and Trent’s (1999) study of business-owning households provides further data on the importance of family members as a supply of human resources. About 73% of the business-owning households in their sample had at least two residential household members working in the business, approximately 24% had one or more employed relatives working for pay who did not live in the household, and 27% had nonhousehold relatives who were unpaid workers. Given the extent to which business owners rely on family members as a source of employees, shrinking family size complicates the human resource mobilization process for many organizational founders.

Smaller families may also make the mobilization of financial resources more difficult. Although Aldrich et al.’s (1998) empirical findings dispute the common assumption that family members represent a frequently used source of start-up funding, other studies indicate that, within some ethnic communities, kin provide a great deal of financial capital (Aldrich and Waldinger, 1990). Moreover, Steier and Greenwood’s (2000) case analysis showed that although family members may not represent a direct source of funding, they are often an important source of indirect ties to other individuals capable and willing to inject financial capital. Their study indicates that family ties are sometimes part of the resource mobilization process, if nascent entrepreneurs are able draw on the network contacts of other family members (Brush et al., 2001, p. 69; Starr and MacMillan, 1990, p. 88). Our review suggests that shrinking family size might be hampering the ability of organizational founders to secure financial capital.

Contributing factors: marriage, divorce, and birth rates

Why has household composition changed so dramatically over the past four decades? Changing patterns of marriage and divorce provide a partial explanation. First, age at first marriage has gradually increased, from 22.8 in 1960 to 26.8 in 2000 for men and from 20.3 to 25.1 for women. Second, marriage rates have dropped slightly over time, as more people have chosen to remain single or cohabitate. Third, divorce rates have increased substantially, as indicated by age-specific studies of ever-married women. For example, the proportion of white women aged 40 to 44 who were divorced from their first marriage increased from 20% in 1975 to 35% in 1990; for black women, the proportion increased from 30% to 45% (Teachman et al., 2000).

Families are also choosing to have fewer children. At the beginning of the 19^th century, women typically had about seven children, only some of whom survived to adulthood (Ahlburg and DeVita, 1992, p. 18). Birth rates dropped steadily throughout the 19th century, and in 1900, the average number of children per woman was only 3.6. Fertility rates continued to decline to the population replacement level of 2.1 children per woman, but then jumped dramatically after World War II. The ‘‘baby boom’’ peaked

in the late 1950s, and the rate fell again, actually dropping below the replacement level for a few years in the 1970s. Today, the birth rate hovers near the replacement level for white women and is higher for black and Hispanic women (Bianchi and Casper, 2000).

Although the birth rate declined during the latter half of the past century, the proportion of births to unwed women increased dramatically during this period (see Fig. 2). In 1940, only 4% of all births were to unmarried women, and in 1960, the level was still only 5%. The rate rose dramatically in the 1970s and 1980s, but seems to have levelled off now. In 1999, the proportion of births to unmarried women was 33%.5 As Bianchi and Casper (2000, p. 17) noted, however, ‘‘a similar plateau in the early 1970s proved to be temporary, so demographers cannot predict whether the stability of non-marital birth rates in the late

1990’s is a temporary lull or an end to one of the most pronounced trends in the latter half of the 20th century.’’

The increasing prevalence of divorce, as well as the rising proportion of births to single women, has transformed the household context in which many children are raised. The stereotypical ‘‘married couple with children’’ household dropped from approximately 44% of all households in 1960 to only 24% in 1999. Today, one out of every five children under the age of 18 lives with a mother only (see Fig. 2). Approximately 50% will spend at least some time in a single-parent family, and ‘‘about one-third of U.S. children will live in a remarried or cohabiting stepfamily household before they reach adulthood’’ (Coleman et al., 2000, p. 1290).

These trends in the life course transitions of marriage, divorce, and childbirth have implications for the emergence and recognition of entrepreneurial opportunities. Rising divorce rates, for example, have generated several new business opportunities. Many professional service firms have arisen to help individuals manage the new role relationships created by separation and divorce. Other entrepreneurs have discerned unusual opportunities in the aftermath of family dissolution events. A Canadian entrepreneur, for example, has recently introduced ‘‘Divorce X,’’ a digital imaging service that removes ex-spouses from family pictures (Solomon et al., 2002). In response to the rise in nonmarital births to teenage girls, another entrepreneur has produced the ‘‘Baby Think it Over’’ doll: designed to convince

teenagers that having a baby isn’t all fun, this doll shrieks at random intervals and must be held for up to 30 min before stopping (Mowen and Minor, 1998, p. 535).

Life course transitions such as marriage, divorce, and childbirth often bring new business opportunities into focus. As individuals experience such family disruptions, their product/service needs and buying habits change. Some may discover that existing vendors or means are not adequately fulfilling their needs, and may realize that these unmet needs present attractive prospects for new venture creation (Bhave, 1994). In other words, while striving to resolve a personal problem, individuals may become aware of a ‘‘customer problem’’—one of the three major dimensions of knowledge considered important in the process of entrepreneurial discovery (Shane, 2000). Becoming a parent is one life course transition

likely to trigger such awareness. As noted by Debelek (2001), parents often devise novel solutions to the problems associated with caring for children. A case in point is the ‘‘Baby B’Air Flight Vest,’’ a safety vest for babies and toddlers to wear on planes. The idea for the product originated from a new mother’s complaints about the difficulty of carrying car seats onto planes and her concerns about holding a child on her lap without a restraining device. Market research and subsequent sales of over US$1 million confirmed the business opportunity associated with the product (Debelek, 2001).

Changes in marriage, divorce, and childbirth rates not only have implications for opportunity emergence and recognition, but may also have mixed effects on the new venture creation process. Consider the increasing number of couples choosing to cohabitate rather than get married. Many nascent entrepreneurs who are married pursue a start-up plan that involves one member of the couple working so that they are both covered by health insurance, and thus cover an important risk factor. Given that health insurance coverage to unmarried couples is not yet universal, how will cohabitation affect the venture creation

decision? And what about the venture creation implications of the rising divorce rate? On the one hand, such family dissolutions might have negative consequences for resource mobilization.

On the other hand, this type of family disruption may provide the impetus for organizational formation. Neider’s (1987) research, for example, indicated that many women start their own firms after becoming divorced. Moreover, the increasing prevalence of stepfamilies arising out of divorce and remarriage suggests that a growing number of organizational founders have access to an expanded pool of current and former ‘‘family’’ members, which may facilitate the resource mobilization process. However, people often lose touch with nonresident family members after a divorce.

Trends in family members’ roles and relationships

Women’s roles

One of the most dramatic changes affecting the North American family system during the latter half of the 20th century was the huge increase in the proportion of women working outside the home (see Fig. 2). In 1960, 90% of adult men were in the labor force, compared to only 30% of adult women (Bianchi and Casper, 2000). By 1998, men’s participation rate had declined to 78%, while women’s participation rate had increased to 60%. As a result, dualincome households are becoming more prevalent: in 1998, both spouses were working in 60% of all married couples.

Historically, women with children withdrew from the labor force, especially if their children were less than 6 years old. Currently, young children are much less of a deterrent to women’s employment: ‘‘in 1998, 71 percent of married mothers of children under age six did some work for pay during the year’’ (Bianchi and Casper, 2000, p. 29). Only about one-third of them, however, worked full-time. Bianchi and Casper (2000, p. 29) summarized these changes by noting that ‘‘although U.S. mothers of young children are much more likely to work in the 1990’s than they were in the 1970’s, which implies an increasing attachment of women to market work, married mothers tend to scale back their hours during their children’s preschool years.’’

One might expect the growing number of employed women, and their increasing contribution to household income, to lead to more sharing of household tasks with husbands and children. Although Goldscheider and Waite’s (1991) research suggested that this is indeed the case, women continue to carry the bulk of the responsibility, and the division of household labor remains traditional. Coltrane (2000, p. 1208), for example, noted that: ‘‘On average, women perform two or three times as much housework as men, and the vast majority of men, as well as most women, rate these arrangements as fair.’’ The primary response of women who work full-time has been to do less housework. The number of hours per week

that women spend doing household work declined from about 30 in 1965 to about 17.5 in 1995 (Bianchi and Casper, 2000, p. 29). Over this same period, men only increased their time on housework from 4.9 to 10 hours per week, clearly leaving some work simply undone (or done by nonfamily members). Women in dual-income families with young children face an especially difficult burden. Presser (1998) noted, ‘‘For the modal family type in the United States, two-earner couples, one in four includes at least one spouse who works evening, night, or rotating schedules. The presence of children, and particularly young children, increases the likelihood of a split-shift arrangement among spouses. So, in one out of three two-earner couples with children under age 5 in the United States, one spouse works in the day and the other works evening, night, or rotating schedules.’’

These changes in women’s roles have certainly spurred the growth of several industries, and have created opportunities for organizations within them to develop new products and services (Solomon et al., 2002). Consider, for example, the outsourcing of four common and time-consuming household chores: childcare, cooking and cleaning, and shopping. Private daycare employment has increased by more than 250% over the past two decades (Mowen and Minor, 1998, p. 519), reflecting the current reality that 48% of working families with children under the age of 13 have childcare expenses (Giannarelli and Barsimantov, 2000).

The demand for food prepared outside the home has also increased: in 1970, only 26% of total food expenditures was spent on away-from-home foods; by 1996, this figure had risen to 39% (Lin et al., 1999). In 1999, expenditures on away-from-home food totalled over US$340 billion in the United States alone (Clauson, 2000). The housecleaning industry has also grown: in 1986, only 6% of households employed a house cleaner or housekeeper; by 1999, that figure had reached 11% (Sheth et al., 1999, p. 63). Several academic studies have directly linked the consumption of away-from-home foods and housecleaning services to women’s increased labor force participation (Cohen, 1998; Hochschild, 1989; Oropesa, 1993).

Shopping services for people too busy to shop for themselves have also sprung up. The increasing number of working women has interesting implications for the recognition of entrepreneurial opportunities. For example, working women who become mothers may recognize that individuals on temporary child leaves, such as themselves, represent ‘‘underutilized resources’’ (Shane andVenkataraman, 2000, p. 222) who might be available to work on a limited-period, work-from-home basis. Acting upon such a realization could trigger the emergence of novel work arrangements and processes that ultimately transform the production processes within an industry—a form of entrepreneurship identified by Schumpeter (1934).

With respect to venture creation, the growing number of working women may both facilitate and inhibit the launch decision. By gaining work experience, an increasing proportion of women might perceive that they have secured sufficient human, social, and financial capital to launch a business. Moreover, the work-related inequities experienced by many women, such as hitting the ‘‘glass ceiling,’’ may actually provide the motivation for venturing out on their own.

As Brush (1992) noted in her literature review, female organizational founders are often motivated by job frustration. Members of a dual-income family might also perceive less risk in new venture creation than those who depend on only one income, as a dual-earner family would still have the employed spouse’s salary as a source of income. All of these factors point to women’s employment as facilitating venture creation decisions.

Nevertheless, working women may have difficulty finding the time to launch a business. Working women still bear primary responsibility for household chores, even when they also work full-time outside the home. Perhaps this inhibiting factor partially explains why proportionately fewer women than men attempt to start businesses, and fewer succeed in actually doing so (Carter, 1994). The dual burden of work and household responsibilities may also explain Ruef et al.’s (2002) finding that a large proportion of women involved in business start-ups do so as part of a spousal team rather than on their own. However, it is also possible that spousal pairs exert reciprocal positive influences on one another, increasing the likelihood that both will become involved in a new venture. For example, Budig’s (2002)

dynamic analysis of the 1978–1998 years of the National Longitudinal Survey of Youth found that having a spouse who was self-employed substantially increased the likelihood of someone being self-employed.

As with the venture creation decision, the increasing number of working women (and the corresponding rise in the number of dual income families) has mixed implications for the securing of start-up resources. One positive implication of this trend is an increase in the proportion of families with ample entrepreneurial capital that can be mobilized during the startup process (Marshack, 1993). A less obvious negative implication of this trend is that fewer start-ups will be able to rely on the human capital provided by female spouses. Female spouses are a commonly tapped source of unpaid (or at least underpaid) labor during an organization’s early years (Aldrich and Langton, 1998). It is unlikely, however, that already-employed female spouses would have either the time or the willingness to forego their salaries to help their husbands’ firms in this traditional manner.

Children’s roles

Children’s and young adults’ roles in families have changed as dramatically as wives’ roles during the latter half of the previous century. ‘‘Until the mid-20th century, most young people lived in their parents’ home until marriage. Since people married later before the baby boom, they remained at home for a decade or more after completing schooling to contribute to the family economy’’ (Goldscheider and Waite, 1991, p. 17). Only in the 20th century did men, as family heads, make enough money to allow their children not to work, and thus to attend school full-time. In the early days of the industrial revolution, children’s incomes were a significant contributor to family income (Zelizer, 1994). Not having to work outside the home made children available, potentially, as household and small business help.

However, children have been contributing less and less in recent years to the family purse (Goldscheider and Waite, 1991). Many of them have taken on paid work outside the household, thus enhancing their own economic welfare. A 1997–1998 Consumer Expenditure Survey revealed that about one-third of all teenagers were employed sometime during that year (Johnson and Lino, 2000). Most did not work for family necessities. Rather, they were earning money for their own personal needs and to buy work-related items. Even in lowincome families, a teenager’s earnings accounted for only a modest proportion of the family’s budget: about 9%.

The greater proportion of teenagers working away from home for their own disposable income may by creating a large pool of nascent entrepreneurs who perceive that they possess adequate human, social, and financial capital to one day launch their own business ventures. At the same time, teenage employment outside the home may be reducing the likelihood that organizational founders can rely upon their children as an inexpensive source of labor during the start-up period. Employed teenagers may be reluctant to trade the guaranteed hours and pay of an outside job with the variable hours and deferred pay of a job in their parent’s business.

Relationships between family members

The final type of transformation in the North American family system that we identified pertains to the ties between family members. The media frequently contains statements announcing the ‘‘death of the family unit’’—a claim based not only upon the declining prevalence of traditional family households but also upon data documenting the deterioration of social bonds within even intact nuclear families. Putnam (2000, p. 101), for example, pointed out that ‘‘between 1976 and 1997, according to Roper polls of families with children aged 8 to 17, vacationing together fell from 53 percent to 38 percent, watching TV together from 54 percent to 41 percent, attending religious services together from 38 percent to 31 percent, and ‘just sitting and talking’ together from 53 percent to 43 percent.’’ These figures suggest that parents are spending less time interacting with, and thereby socializing, their children. As noted by Bianchi and Casper (2000), other institutions have taken over many of the traditional social functions of families, and some functions have simply disappeared, such as parental regulation of teenage sexual behavior.

Not surprisingly, social bonds between parents and children are even weaker within families disrupted by divorce. Divorce seems to particularly strain a father’s relations with his biological children. Using figures from the 1981 National Survey of Children, Furstenberg (1988, pp. 202–203) argued that ‘‘many, if not most, non-custodial fathers are only weakly attached to their children.’’ Almost half the children living in mother-headed households had not seen their biological father in the previous year.7 The increasing prevalence of divorce, as well as the growing number of births to unwed women, has also weakened mother–child relationships. Bianchi’s (2000) review revealed a growing gap between two groups of households: those with fathers present and those without. When fathers are not present, ‘‘mothers have neither adequate time nor money to invest in children’’ (Bianchi, 2000, p. 412).

Intergenerational family relations have also been undergoing a fundamental change (Rossi, 1989). Even as longevity was increasing substantially, creating more families in which grandparents were still alive when children reached adulthood (Uhlenberg, 1996), residential co-location across generations has been declining. Over the past century, a very sharp drop occurred in the proportion of elderly persons residing with their adult children, especially among whites (Ruggles and Goeken, 1992).8 In 1900, among whites aged 60 and over, about 53% were living in multigenerational families, defined as sharing a residence with children who are older than 21 or ever married. By 1980, only 14% were living in such arrangements. Although most parents and children ‘‘live near one another and continue to interact on a

regular basis throughout life’’ (Bumpass, 1990, p. 491), the frequency of contact appears to be diminishing. Bumpass (1990, p. 492), for example, noted that between 1962 and 1984, ‘‘the proportion of the elderly seeing a child at least once a week declined by 25 percent.’’

Although unfortunate, these changes in the social bonds between family members are creating entrepreneurial opportunities. For example, one company has capitalized on the diminishing frequency of joint parent–child activities by developing the ‘‘Weemote’’—the first programmable TV remote control designed for kids (InventiveParent, 2002). Similarly, the increasing amount of time spent by children without adult supervision has created sizeable markets for such products as convenience foods, security systems, and appliances that young children can operate on their own (Hawkins et al., 2001, p. 196; Solomon et al., 2002, p. 407).

As an example of how the process of opportunity emergence and recognition can also be stimulated by changing intergenerational relations, consider the experience of an individual who recently found himself responsible for his aging father’s care.10 Consistent with the societal trend of not wanting his father to reside with him, but not knowing what options were available, he searched for information. Frustrated by the lack of easily accessible material to help him with this important family decision, yet aware of the prevalence of this major demographic change, he realized that the personal problem he was facing actually represented a potentially lucrative business opportunity. He subsequently published a book to help others with the process of arranging suitable eldercare.

Weakening bonds between family members surely carry implications for the venture creation process, but the nature of these implications is not clear-cut. On the one hand, for example, declining parental interaction with their offspring may reduce the preparation and motivation of children from business families to continue in their parent’s entrepreneurial footsteps. On the other hand, an increasing proportion of children may be willing to make themselves available as a source of inexpensive human resources for their parent’s business, as a means of potentially increasing the level of parent–child interaction.

Toward a family embeddedness perspective: implications for entrepreneurship and family business research

During the past two decades, the notion that entrepreneurs are embedded in social relationships has become almost axiomatic in the entrepreneurship literature (Aldrich and Zimmer, 1986; Burt, 1992; Larson and Starr, 1993). Rather ironically, however, the embeddedness approach has virtually neglected the one social institution in which all entrepreneurs are embedded—the family. This gap persists in spite of evidence that: family transitions can trigger organizational emergence (Cramton, 1993), founding teams are frequently composed of family members (Ruef et al., 2002), and families play an important

role in resource mobilization process during start-up (Aldrich and Langton, 1998). Guided by the literature on life course theory (Elder et al., in press; Moen, 1998), we have argued that family considerations should figure more prominently in entrepreneurship research. We illustrated how major sociohistorical transformations in family composition and family members’ roles and relationships can influence, or have already influenced, the entrepreneurial phenomena of opportunity emergence and recognition, the venture creation decision, and the resource mobilization process. We turn now to some conceptual and methodological suggestions for researchers interested in connecting family systems with

organizational emergence, and summarize the implications of a family embeddedness perspective for the literature on entrepreneurship, family businesses, and business families.

An emergent ‘‘family embeddedness perspective’’ on new venture creation In Fig. 3, we present a conceptual framework to help guide research adopting a family embeddedness perspective on new venture creation. The left-hand side of the framework suggests that the characteristics of entrepreneurs’ family systems (i.e., transitions, resources, and norms, attitudes, and values) influence the processes involved in venture creation (i.e., opportunity recognition, the launch decision, resource mobilization, and the implementation of founding strategies, processes, and structures). The right-hand side of the framework suggests that venture creation processes result in outcomes for the new organization (i.e.,

survival, objective performance, and subjective perceptions of success) that may alter the entrepreneurial family’s resources, potentially trigger certain family transitions, and ultimately even change family members’ norms, attitudes, and values.

Many of the relationships depicted in the left-hand side of the framework were implicit in our discussion of the dramatic transformations in the North American family system during the previous century and their implications for entrepreneurship. For example, we presented examples of how experiencing a family transition, such as childbirth, can provide family members with new information about unmet customer needs. This change in the family’s informational resources may subsequently trigger the recognition of an entrepreneurial opportunity, and possibly the other processes involved in venture creation. As another example, the increase in a family’s financial resources associated with a woman’s transition to employment might facilitate her spouse’s decision to start his own firm.

Other relationships within the left-hand side of the framework were not derived directly from our sociohistorical essay but are plausible, based on existing research on either organizational emergence or family systems, and consistent with a family embeddedness perspective on new venture creation. Several entrepreneurship scholars, for example, have suggested that the venture creation process is not limited to opportunity recognition, the launch decision, and resource mobilization, but also includes the implementation of founding strategies, processes, and structures. The norms, attitudes, and values held by entrepreneurial family members likely influence these founding strategies, processes, and structures.

Similarly, the postulated relationships depicted in the right-hand side of the framework represent extensions of the arguments implicit in our sociohistorical review of family system changes. If families and businesses truly are as inextricably intertwined as we have claimed, then not only should family systems affect business systems, but the reverse should also hold.

That is, business systems should also affect family systems. In the context of organizational emergence, this logic of interdependency implies that venture creation processes—particularly the outcomes associated with these processes—are likely to contribute to changes in the characteristics of founders’ family systems.

Methodological implications and suggestions

Operationalizing ‘‘family’’

The family embeddedness perspective encapsulated in Fig. 3 also raises a number of methodological implications. For example, what does the term ‘‘family’’ mean for organizational founders? Does this construct refer to founders’ immediate nuclear families, which was the focus of Cramton’s (1993) research, entrepreneurs’ families of origin, which was the focus of Aldrich, Renzulli, and Langton’s (1998) study, or entrepreneurs’ extended and step family members? Researchers will need to explicitly define the notion of ‘‘family’’ in their conceptual models and empirical investigations.

The trends in family composition that we have discussed in our sociohistorical review highlight the importance of working with as inclusive a definition of ‘‘family’’ as possible. Researchers should not limit themselves to narrow definitions that are constrained by historical conceptions of what constitutes a family. We suggest focusing on households, regardless of size, and not simply on multiperson units in which two or more people are related. Many of the new business opportunities that will emerge in the 21st century will reflect a changing mix of household forms, which increasingly include single people and multiple unrelated people under the same roof. Similarly, patterns of family formation and dissolution have created extended kinship networks that cut across household boundaries, creating ‘‘business families’’ that may include multiple single-person households and family members living elsewhere.

Developing dynamic designs

Entrepreneurship research will need to be much more process oriented to capture the kinds of relationships suggested by Fig. 3. Investigators will need to build more longitudinal study designs so that they can make dynamic inferences about the effects of family system changes on the new venture creation process, as well as about the effects of new venture outcomes on family system changes. We suggest that more resources be invested in ethnographic approaches in particular, in which investigators spend considerable amounts of time within families, conducting systematic observations and collecting detailed field notes (Gartner and Birley, 2002). Disciplined field research gives us the possibility of making sense of what can only be indirectly inferred through other methods.

Capturing the reciprocal causality of family and business system changes

Conceptual frameworks such as that presented in Fig. 3, which explicitly include characteristics of family systems, should enhance the comprehensiveness of our explanations for entrepreneurial phenomena and deepen our understanding of the complexities and compromises involved in entrepreneurial processes. Designing rigorous studies consistent with a family embeddedness perspective, however, will undoubtedly require research teams comprised of scholars with interests and skills in both ‘‘families’’ and ‘‘businesses.’’ We hope that our paper will encourage at least some researchers to bridge these currently separate domains.

Conclusion

We need more research on how family systems affect opportunity emergence and recognition, the new venture creation decision, and the resource mobilization process. We need to learn more about the role that family characteristics and dynamics play in why, when, and how some people, but not others, identify entrepreneurial opportunities and decide to start business enterprises. In turn, we need to better understand the effects that venture creation can have on family systems. Such research projects would extend the scope of the existing social embeddedness approach to entrepreneurship, paving the way for a more holistic (and thus more realistic) approach to our understanding of entrepreneurial phenomena. The family embeddedness of most entrepreneurial activities—particularly new venture creation—necessitates broadening our persp