How policy shapes competition: early railroad foundings in Massachusetts

Administrative Science Quarterly,  Sept, 1997  by Frank Dobbin,  Timothy J. Dowd

• E-mail
• Print
• Link

What factors influence industry competition in ways that affect the establishment of new firms? Organizational ecologists point to the environmental resources at hand and the level of competition for those resources. Industrial organization economists stress two characteristics of industry structure: concentration and capital accumulation. Business historians and organizational institutionalists point to two characteristics of the policy environment: state capitalization and policies regulating competition. While these three approaches address common outcomes, they are rarely considered together. We consider them together here, by analyzing time-series data on the founding of 317 railroads in Massachusetts between 1825 and 1922. During this period, there was great variation in the six factors of interest: founding resources, number of competitors, industry concentration, capital accumulation, state capitalization, and public policies regulating competition. There was also great variation in the frequency of foundings. Because public policy creates the property rights upon which market exchange is founded, it follows that different policy regimes should produce different varieties of competitive behavior (White, 1988; Zelizer, 1988).

Public Policy and the Construction of Business Strategy

While we now know a great deal about the effects of population characteristics and industry structure on competition within a stable policy regime, we know little about the effects of different policy regimes. This is surely because much organizational and economic scholarship comes from the United States, where a belief in laissez faire obscures the role of the state in markets and where the size and geographical isolation of the economy render the policies of other nation-states distant and intangible. It may also be a consequence of the fact that national policy styles tend to be consistent over time, making controlled analysis difficult. France has capitalized vital industries since the time of Louis XIV (Shonfield, 1965); Germany has encouraged cartels since the late nineteenth century (Chandler, 1990); and the United States has enforced antitrust since the beginning of the twentieth century (Fligstein, 1990).

Early American railroading provides a natural laboratory for studying the effects of policy on competition, because railroads experienced each of these three policy styles. Between 1825 and 1871, American railroads operated under public capitalization policies as generous as France's (Hartz, 1948; Dunlavy, 1993). Between 1872 and 1896, the industry operated under pro-cartel policies as friendly as Germany's (Kolko, 1965; Berk, 1994). From 1897, railroads operated under the antitrust policies that would be used to govern most American industries in the twentieth century. In short, American railroads experienced both public capitalization and exclusively private capitalization, both pro-cartel and antitrust policies (Dobbin, 1994). We expect these policy regimes to shape the two principal factors that population ecologists have found important: resource availability and competition among incumbents. Public capitalization expands the resource pool. Pro-cartel policies dampen competition. Antitrust enlivens competition.

Our aim is to build on neoinstitutional insights about the effects of policy on business practice. Institutionalists have been concerned with the social construction of both human rights and economic rationality (Meyer, 1994; Strang, 1994; Scott, 1995). In early studies tracing the effects of federal policies in such realms as equal employment opportunity, policy shows fairly direct effects, causing firms to adopt specific practices (DiMaggio and Powell, 1983, 1991) or encouraging managers to invent new practices within narrow guidelines (Meyer and Rowan, 1977; Edelman, 1990; Sutton and Dobbin, 1996).

Recent studies suggest that the role public policy plays in shaping economic rationality is sometimes more subtle. In these studies, public policies have been shown to influence corporate behavior by framing the competitive environment, rather than by promoting specific practices. Since Fligstein (1990) explored the role of policy in the construction of such core business strategies as conglomeration, institutional insights have been used to study thrifts' decisions to enter new markets (Haveman, 1994), deconglomeration among large firms (Davis, Diekmann, and Tinsley, 1994), the diffusion of new accounting practices (Mezias, 1990), and the founding of new firms (Baum and Oliver, 1992). We build on this foundation, seeking to identify the key policy characteristics that shape industry competition and thereby influence entrepreneurial decisions.

We argue that most new policies create constraints and incentives, rather than dictating firm behavior, and that managers construct new business strategies taking those constraints and incentives into account (Edelman, 1992; Dobbin et al., 1993; Fligstein, 1996). Managers reach consensus on how to respond to policies largely by observing and imitating peers, as both neoinstitutionalists (Strang and Meyer, 1993; Meyer, 1994) and network theorists (White, 1981; Granovetter, 1985) have suggested. Only after managers have reached some sort of consensus are effects of policy on business behavior predictable. Thus, railroad managers defined merger as the appropriate response to the Interstate Commerce (1887) and Sherman Antitrust (1889) acts, but only after a decade of experimenting with various covert forms of collusion. Understanding how managers devise new strategies is key to understanding the effects of policy, and in a companion piece to this (Dowd and Dobbin, 1998) we explore the range of strategies devised after each railroad policy shift and the process of consensus formation. Railroaders were pioneers in more than one sense in American business: the strategies they devised for dealing with pro-cartel and with antitrust policies soon spread to other industries. We expect the results reported below to be generalizable in part because other industries followed the lead of the railroads.