Social capital and household income distributions in the United States: 1980, 1990.

1. Introduction

Income inequality has been on the rise in the U.S. since the 1970s. Evidence of income inequality is the percentage of total income earned by the highest income families and the percentage of total income earned by the lowest income families. According to recent Census Bureau data, the 25% highest income families now receive 44.6% of U.S. income. The 25% lowest income families earn 4.4%. This is the widest rich-poor gap since the Bureau revised their data collection methods in 1947 (Bernstein, 1994).

In addition to the increase in income disparity there have been other changes in American society. Putnam (1995) claims that Americans volunteer less, are less engaged politically, have declining education standards, face rising crime rates, and have lost the sense of security due to changes in the work place. Putnam suggests that the social changes just described indicate a decline in social capital or in the quality of our relationships. Other scientists dispute Putnam's findings claiming that instead of a decay in civic engagement, we are experiencing shifting civic and community engagements (Clark, 1996).

At issue is the following question: Do changes in social capital influence the level and disparity of household income in the United States? The logical connection between changes in social capital and changes in income distributions follows. Income distributions are largely dependent on the terms of trade at which one exchanges his or her goods and services. If relationships (social capital) influence terms of trade and terms of trade determine income distributions, then social capital must also influence income distributions. The evidence presented in this paper suggests that changes in social capital and changes in income distributions are related.

1.1. Relationships and terms of trade

Terms of trade can be defined as the agreement between economic agents that determines the quantity, quality, risk, price, information content, timing, and location of goods and services traded. In many experiments, relationships appear to have altered the terms of trade. The evidence suggests that friends and family trade more and at different prices than do the estranged and strangers. Examples of relationships altering the terms of trade follow.

When farmland sales are recorded, a distinction is made between land sales between family members and "arms-length" sales made between unrelated individuals. The distinction is made because realtors recognize that the sale price of land depends on the relationship between the seller and the buyer (Gilliland, 1985). Nepotism laws restrict government employers from hiring their close relatives. These laws recognize the tendency of some government employers to grant employment advantages to their relatives. Civil rights laws preclude employment being denied because of one's race/ethnicity. These laws recognize that race/ethnicity often changes the relationship between employers and potential employees. Finally, our judicial system emphasizes the role of relationships by placing a blindfold on our symbol of the court, Lady Justice. The blindfold helps her make impartial judgments free from the bias created by knowing who is to be judged.

Families represent an organization in which a special relationship exists. One way this special relationship manifests itself economically is in the formation of business agreements. Gwilliam (1993) found that 89% of Michigan farmland leases were between friends and family members. Nelton (1990) noted that family businesses account for 76% of Oregon's small companies. Calonius (1990) wrote that 75% of U.S. companies are family-owned or controlled.

Relationships between individuals and causes represented by particular organizations account for substantial amounts of voluntary donations. Despite a sluggish economy, philanthropic giving across the nation in 1991 exceeded donations in 1990 by 6.2%. Voluntary donations in 1991 equaled $124.7 billion, of which individuals contributed 89%. The largest recipients of philanthropic giving included religious and educational organizations. Other recipients included environmental groups, the arts, health organizations, and other nonprofit groups providing human services (Tetsch, 1992).

Other studies demonstrating how relationships change the terms of trade include the following. Graduate students in the Department of Agricultural Economics at Michigan State University would sell a used car valued at $3,000 for $420 less than its market value if the buyer were a friend in need. However, these same graduate students would require $697 above the market price if the buyer were an unpleasant wealthy neighbor (Robison and Schmid, 1991). A survey of 103 Michigan bankers serving communities of less than 10,000 found that good business and social relationships increased the probability of loan approval in some cases by 60% (Siles, Hanson, & Robison, 1994). Survey respondents reported that their willingness to bear risk depended on the consequences of their risk decision on important others (Robison and Hanson, 1996). Finally, relationships have been significant factors in customer retention, tipping behavior, data perception, and willingness to cooperate (Robison and Hanson, 1995).

1.2. Income levels and income disparity

It is generally agreed that income inequality may have serious social and economic consequences. Addressing the question, "What are the five biggest challenges Clinton faces in his second term in office?," Professor George J. Borjas of the John F. Kennedy School of Government wrote that Clinton's number one challenge was to address income disparity. Professor Jeffrey Rosensweig of the Goizueta Business School at Emory University wrote that Clinton's number one challenge was to ease the gulf between rich and poor and keep the middle class from hollowing out.

One challenge associated with reducing income inequality is the belief held by some economists that there is a tradeoff between increasing incomes and reducing income inequality. One view attributed to Kaldor (1978) is that a high level of savings is a prerequisite of growth. Moreover, since the rich save more than the poor, growth requires income be concentrated in the hands of the rich whose savings rates are high. A second view suggesting the tradeoff between income growth and income inequality is attributed to Kuznets. His view was that as labor shifts from a low productivity sector to a high productivity sector, aggregate inequality must initially increase substantially and only later decrease. This latter view, Robinson (1976) observed, has "acquired the force of economic law." The validity of this law, however, is not universally accepted. Commenting on Kuznets' law, Fields (1988) wrote:

Perhaps one of the greater ironies in the history of thought on economic development is that the economic law which today is most often associated with Kuznets and that has come to bear his name, the idea that income inequality increases in the early stages of economic development and decreases in the later stages, thus tracing out an inverted-U curve receives remarkably little empirical support, either from the evidence presented in Kuznets' writings or in subsequent data (p. 462).

Deininger and Squire (1997) used an expanded data set covering 30 years to test for the presence of the Kuznets' curve. They found no evidence of the Kuznets' curve in almost 90% of the cases they examined.

Other explanations for the increasing income disparity include falling wages for unskilled workers as automation spreads, low tax rates on the wealthy during the 1980s, low minimum wages, the decline of trade unions, and the rapid rise in the value of stocks and bonds during the 1980s, in which the wealthy are heavily invested (Bradsher, 1995). More recently, Williamson (1997) suggests migrations of unskilled workers can explain a significant portion of change in income disparity.

It appears that important work remains to be done before economists and others agree on the causes of income inequality. To improve our understanding, this study considers the possibility that relationships or social capital may be a significant factor in explaining income inequality.

The remainder of this paper examines the connection between social capital and income distributions. Section 2 defines social capital and describes its properties. Section 3 outlines opportunities for investments and disinvestments in social capital. Section 4 describes how social capital internalizes externalities and alters terms of trade. Section 4 also analytically connects social capital influences to income distributions and develops several hypotheses. Sections 5 through 11 develop measures of social capital and test empirically the connections between social capital and income distributions using U.S. Census data for 1980 and 1990 and indicator variables for the same period. Finally, section 12 summarizes our findings, that distributions of social capital and household income are related.

2. What is social capital?

Suppose person i perceives a change in the well-being of person j and as a result experiences a change in his or her own well-being. Then i is said to have a relationship with j. Person i's relationship with j may also produce in person i feelings of attachment toward objects associated with person j including places, communities, schools, clubs, animals, organizations such as churches and service clubs, and legal institutions. The relationship person i has with person j depends on at least two elements. The first element is awareness or "social distance." Social distance measures i's knowledge of j that may include information about j's behavior, consumption, wealth, values, or social bonds. As i's knowledge of j increases, i's social distance to j decreases. For those individuals, groups, communities, or institutions j about whom i has no knowledge, i's social distance to them is infinite and i has no relationship with them. Consequently, if i has no knowledge of j, then changes in j's well-being do not influence i's well-being and there is no relationship.

The second element that determines i's relationship to j is the degree of sympathy or antipathy that i holds toward j. Person i may develop toward j feelings of sympathy, antipathy, or neutrality (Bogardus, 1924). We have observed that stable relationships between persons tend to be symmetric; that i cares for j about the same as j cares for i.

Suppose i has awareness of and sympathy for j. Then, any improvement in j's well-being also benefits i vicariously. As a result, j can expect person i to extend favors, preferential terms of trade, and in other ways look out for j's interest as long as the favors, preferential terms of trade, and other benefits extended do not impose a cost on i greater than i's vicarious benefits earned as j's well-being improves. The relationship of sympathy (antipathy) i has toward j is called here j's social capital with i (denoted [k.sub.ij]) and is defined next.

2.1. Social capital: a definition

Definition. Social capital is the potential benefits, advantages, and preferential treatment resulting from one person or group's sympathy and sense of obligation toward another person or group. Social capital also includes the potential benefits, advantages, and preferential treatment that originate from one person's sympathy and sense of obligation toward his or her idealized self.(1)

Other definitions of social capital include: (1) the social obligations or "connections" which are convertible into economic capital under certain conditions (Bourdieu, 1986); (2) a resource of individuals that emerges from their social ties (Coleman, 1988); (3) the ability to create and sustain voluntary associations (Putnam, 1993); (4) trust (Fukuyama, 1995); and (5) the relationship or caring between persons and between persons and their institutions (Robison and Schmid, 1994).(2)

Some applications of social capital follow. Coleman (1988) discussed social capital and its application to sociology. Hyden (1993) discussed social capital in a political science setting. Putnam (1998) suggested recently that the supply of social capital in the United States has decreased. Fukuyama (1995) associated social capital with trust and suggested that trust or social capital is at the foundation of collective action. C. Flora and J. Flora (1996) discussed the importance of social capital in maintaining society's social contract. Robison and Schmid (1991; 1994), Robison and Hanson (1995; 1996), and Schmid and Robison (1995) discussed the role of social capital in economics. Finally, Evans (1996) and Fox (1996) have written about the role of social capital in development.(3)

2.2. Economic properties of social capital

Social capital and other forms of physical and financial capital have many features in common. Like physical capital, the potential benefits of social capital can be depreciated through neglect and the passage of time. Like physical capital, social capital can sometimes be enhanced or depreciated by providing or extracting services. For example, asking a friend for a favor or improved terms of trade in an economic transaction may reduce the likelihood the friend will extend favors in the future. On the other hand, granting favors and extending favorable terms of trade may increase one's social capital and increase the likelihood of receiving favors in the future. Finally, like financial capital, social capital may be fungible. For example, person A may use his social capital with person B to improve the terms of trade for person C who plans a trade with B but has no social capital to employ in the transaction.

Perhaps one of the most interesting aspects of social capital is that social capital provides a new perspective on goods and markets. Traditionally, economists have described goods as objects with properties wanted by consumers. For example, a good may be wanted because of its temperature, taste, sight, place, form, location, or ability to create physical sensation. Exchange prices and amounts of the goods exchanged were then said to depend on incomes, marginal utility for the good's properties, and the cost of supplying the good. Social capital theory suggests that the desirability of a good may be modified by relationships as well as physical properties of the good.

Social capital theory suggests that transactions between agents may include exchanges of money, goods, and social capital. In some markets, goods may be exchanged for social capital only. For example, one neighbor may supply another neighbor an item such as a cup of sugar and refuse money as payment. The neighbor supplying the sugar may refuse payment because the increase in social capital he or she receives by supplying the sugar is valued more than the money value of the cup of sugar. In other exchanges where the good is of significant value, a good may be exchanged for both money and social capital. Consider, for example, the sale of a used car. The seller of the used car may offer it at a discount to a friend or family member. The car seller receives less than the market value of the car plus social capital worth more than the discount offered the buyer.

Another important economic property of social capital is its ability to reduce transaction costs. Monitoring costs, threats of litigation, price, quantity, and quality discovery costs and the costs of writing contracts that consider many contingencies may all be reduced by increases in social capital between trading partners. These transaction costs are reduced by increases in social capital because each party to the trade has his well-being linked to the well-being of his or her trading partner. Thus, we expect to find reduced transaction costs and more successful contracts between friends and family. Supporting the assumption that social capital reduces transaction costs and improves the success of contracts, Johnson et al. (1987) found that farmland leases between related individuals were often oral (less costly to complete) and more successful than written leases between unrelated lessees and lessors.

The importance of social capital's ability to reduce transaction costs has important implications. In some markets, especially in less-developed countries, transaction costs are very high. As a result, opportunities for mutually beneficial trade between strangers are limited. Thus, in high transaction cost economies, we expect to find more trades between family and friends as a percentage of total trades than in low transaction cost economics.

Another economic property of social capital is its ability to change the terms of trade. Suppose person i has an object for sale and expects his/her well-being to be improved by exchanging the object for the object's "arm's-length" value. Next, suppose that i has a friend j who needs the object being sold. Selling to j would improve i's well-being in several ways. First, i would benefit from the money received from the sale. Person i would also receive some satisfaction from knowing j's well-being has improved as a result of the sale. Person i may also feel good about his/her relationship to his/her idealized self, knowing that j's well-being has improved as a result of his/her efforts. Finally, i may benefit because j's goodwill toward i has increased because of the purchase. The extra benefits associated with social capital i receives from a sale to j means that i could sell to j at a price below the market price and still be better off than selling the object at the market price to a stranger. Thus, because of social capital, i is likely to offer more favorable terms of trade to his/her friend or family member than he or she would offer to a stranger (Robison & Schmid, 1994).

Not all economic properties of social capital are beneficial economically. Consider some possible negative consequences of social capital. One negative consequence is that social capital may lead to agreements that are not economically sound. For example, a parent may employ a son or a daughter in the family business and pay them a wage …