Health insurance before the welfare state: the destruction of self-help by state intervention.

• Author: Chalupnicek, Pavel

Social scientists, especially sociologists and economists, are paying increasing attention to the concept of social capital. The expansion of its use has been so rapid that it has led some to warn against its misuse and against overstatement of its importance (see, for example, Portes 1998, 21). Bearing these caveats in mind, we show in this article how the concept of social capital can help us to understand some adverse effects of government social policy.

Since the publication of Charles Murray's Losing Ground (1984), if not earlier, it has been clear that Western-style welfare states are encountering deepening problems and that despite social scientists and politicians' efforts and an increasing amount of resources, these states' traditional measures are failing to achieve their main goals. Harvard sociologist Nathan Glazer, one of those who helped to formulate these welfare policies, summarizes the difficulties in his 1988 book The Limits of Social Policy. Sentences such as "It didn't work" and "[W]e seemed to be creating as many problems as we were solving" (2) are a leitmotiv of his account.

Today, analysts generally agree that the structure of incentives is crucial for the success of any such effort. To illustrate this point, we provide here a social-capital-based explanation of the origins and development of voluntary "social insurance," focusing on health (or "sick") insurance. We build on ideas presented in studies by Peter Leeson (2005, 2007) and by Anthony Carilli, Christopher Coyne, and Peter Leeson (forthcoming). We first describe the theory of government interventionism that provides a conceptual framework for the rest of the article. We then analyze historical cases of voluntary provision of social insurance by friendly and fraternal societies around the turn of the twentieth century.

Social Capital

Early uses of the term social capital can be ascribed to sociologists. Among them, Pierre Bourdieu is usually credited with the idea's elaboration that led to its widespread use. In 1986, he described social capital as "an attribute of an individual in a social context. One can acquire social capital through purposeful actions and can transform social capital into conventional economic gains. The ability to do so, however, depends on the nature of the social obligations, connections, and networks available to you" (qtd. in Sobel 2002, 139). Although Bourdieu's description explicitly ties social capital to an individual (and thus makes it an individual asset), the parallel emphasis on the social context has created much confusion. (1) In this article, we treat social capital as an individual asset, building on James Coleman's (1988, 1990) contribution. (2)

Employing an interesting analogy, Coleman writes about people who are investing in their social capital as exchanging "credit slips"--that is, confirmations of their mutual obligations (1990, 306). The investor (person A, who did something for person B) holds the credit slip "to be redeemed by some performance" (306) by person B later, and all such slips together constitute A's accumulated stock of social capital. Coleman also employs an analogy between these "credit slips" and "fiduciary money" (186), both of them having in common the need for trust between the trading parties inasmuch as they are not backed by any "real" values.

Leeson (2005, 2007) and Carilli, Coyne, and Leeson (forthcoming) take a similar approach. Building on a traditional Austrian approach to the business cycle (Hayek 1931, 1941; Mises 1996, esp. chap. 20), they argue that government creates "artificial" trust or distrust, thus generating fluctuations in social-capital accumulation similar to the changes in investment that result from altering the market interest rate. In both cases, government interference scrambles market information signals. People, the "social capitalists," then have difficulty distinguishing between trustworthy individuals with reliable ways of interacting and untrustworthy individuals with unreliable ways. If the government creates a false expectation of greater trust (for example, by declaring its backing of one of the parties), it can cause the second party to invest too much in this relation (and thus create "overinvestment"). (3) In the opposite case, it can create distrust, causing people to invest too little and making them lose potential benefits from unconsummated transactions. By generating such fluctuations, the government can "add" or "remove" trust from various private activities or individuals, or it can interfere with them by its own activity and crowd them out. In financial markets, the government in most cases obtains more favorable loan conditions than any other potential borrower. The most common explanation attributes this advantage to the government's power to tax. (4) As a result, it crowds out private investments that cannot compete. Similarly, some studies of crowding-out in the area of private philanthropy explain it with reference to private charities' reduced effort to raise funds from individuals after they receive a government grant (see, for example, Andreoni and Payne 2003).

Another theoretical explanation pertains to social capital. We might argue that private provision of "public goods" involves serious free-rider problems. In the absence of government, communities usually find a workable and efficient way to provide these services (see Beito, Gordon, and Tabarrok 2002 for an overview). To eliminate free riders, they set up a complex net of self-enforcing rules--that is, rules that do not require external (governmental) enforcement (Telser 1980). In game-theoretic jargon, these rules help to overcome prisoner's dilemma problems connected with the provision of public goods (Ostrom 2000).

Observable holdings of social capital--for example, one's engagement in social networks--have an important signaling potential and help to prevent free riding ex ante. According to Leeson, to eliminate possible problems of cooperation, each party can screen the other for signals that indicate its credibility. "Successful screening does, however, require two things: easily observable attributes or activities--signals ..., and signals with an appropriate cost structures ... that are cheap for cooperative types to send but expensive for cheaters to send" (2005, 79). The "cost structure" of signals based on social capital is obvious because the creation of such capital is connected with certain costs and because it is lost in the case of cheating (which serves also as an ex post punishment for the cheating party), thus constituting "selective incentives" (Olson 1965).

After the government steps in, these systems of self-enforcing rules break down in many cases, and the voluntary provision of public goods collapses. Such a breakdown may happen because of interference with information signals through the creation of artificial homogeneity; for example, when membership in a social network is compulsory, the membership's information signal ceases to convey any information about the members' credibility. After such intervention, the society becomes "fractionalized" (Leeson 2005, 87): ties between individuals in the particular social network are lost or weakened as the trust based on the information signaled by membership deteriorates.

Thus, government action is in many cases self-fulfilling: expected failures of private systems lead to interventions that cause their real failures (Ikeda 1997). Nathan Glazer complains, for example, about government social policy's effect on personal social ties: "In our efforts to deal with the breakdown of these traditional structures [which he earlier describes as being 'located in the family primarily, but also in the ethnic group, the neighborhood, the church'], our social policies were weakening them further and making matters in some important respects worse.... Our efforts to deal with distress are themselves increasing distress" (1988, 3).

Friendly and Fraternal Societies

We turn now to an analysis of the institutional setting and functioning of friendly or fraternal societies--voluntary organizations that provided diverse kinds of "social insurance"--with emphasis on health insurance. We rely on four main sources about these societies: J. M. Baernreither (1889), P. H. Gosden (1961), and Simon Cordery (2003) with regard to English friendly societies; and David Beito (2000) with regard to U.S. fraternal societies.

The first friendly societies evolved from other kinds of mutual associations, such as medieval guilds, and the first fraternal societies in their modern form appeared in England at the end of seventeenth century (Gosden 1961, 2; Cordery 2003, 13). (5) The nineteenth century was the golden age of friendly societies in England. At that time, friendly societies were officially defined as "institutions 'whose object is to enable the industrious classes, by means of a surplus of their earnings, to provide themselves a maintenance during sickness, infirmity and old age" (Gosden 1961, 15). The term friendly societies covers a wide array of associations that used various insurance methods. The oldest and simplest were "dividing societies," in which members contributed to a common fund that after a time was divided among the members who survived (Baernreither 1889, 172). Some friendly societies accepted only members of a particular trade ("particular trade societies") or living in a specified area ("local town societies"); some were established and endowed by local gentry or clergy ("patronized societies"). Others specialized in the provision of burial insurance ("burial societies") or experimented with "individual accounts" for their members ("deposit friendly societies"). The friendly societies' variety and flexibility led Baernreither to the conclusion that "[a]ny one can make provision in the friendly societies for himself or his family, in the manner that suits himself best" (1889...