Close
Stay informed: Sign up for eNews Subscribe
Read Chapter 1
Close

1

Institutions and Development

A Conceptual Reanalysis

Alejandro Portes

Revised version of Alejandro Portes, "Institutions and Development: A Conceptual Re-analysis," Population and Development Review 32(June): 233-62.

Recent years have brought a significant change in the evolution of economics and sociology, including an unexpected convergence in their approach to issues like firms and economic development. This convergence has pivoted around the concept of "institutions," a familiar term in sociology and social anthropology but something of a revolution in economics, dominated so far by the neoclassical paradigm. This development has been accompanied by confusion about what the new master term means and, importantly, by a failure to mine prior theoretical work that sought to order, classify, and relate the multiple aspects of social life that are now brought under the same umbrella concept.

This chapter seeks to reverse these trends by recalling key concepts and distinctions in sociological theory and illustrating their analytic utility with examples from the recent literature on economic development. The argument is that recourse to these concepts and distinctions enhance our ability to analyze economic and "economically relevant" phenomena (Weber [1904] 1949).

The New Institutionalism

As Peter Evans (2004b) has pointed out, the long-held consensus in economics that equated increasing capital stocks with national development has given way to an emerging view that the central role belongs to "institutions." He approvingly quotes Hoff and Stiglitz (2001: 389) to the effect that "development is no longer seen as a process of capital accumulation, but as a process of organizational change." Sociologists of development, including Evans himself and several nonorthodox economists, have been saying the same thing for decades without their arguments succeeding in swaying the economic mainstream (Evans 1979, 1995; Hamilton and Biggart 1988; Portes 1997; Hirschman 1958, 1963). Not until two Nobel laureates in economics, Joseph Stiglitz and Douglass North, elaborated the same arguments were some of those in the mainstream convinced. When North declared that "institutions matter," other analysts started to take them into account.

By 2004 the development economist Gerald Roland (2004: 110) declared that "we are all institutionalists now." Sociologists have generally welcomed this "institutional turn" (Evans 2004a; Nee 2005) as a vindication of their own ideas, albeit with a critical omission. Swayed perhaps by the promise of interdisciplinary collaboration in the wake of the new ideas, they have overlooked a fundamental fact: economists do not routinely deal with the multiple elements of social life or their interaction, and, in their attempts to do so, they often confuse them, producing impoverished or simply erroneous perceptions of reality.

Other observers have noted the same problem and put it in still more critical terms. Geoffrey Hodgson (2002: 148) states, "The blindness may be partial, but the impairment is nevertheless serious and disabling. What is meant by this allegation of blindness is that, despite their intentions, many mainstream economists lack the conceptual apparatus to discern anything but the haziest institutional outlines.... [They] have not got adequate vision tools to distinguish between different types of institutions, nor to appraise properly what is going on in them."

This judgment may be too harsh because, after all, institutional economists have taken the first steps toward incorporating key elements of social reality into their analyses. However, the level of interdisciplinary collaboration needed to do this optimally is still lacking. The first question is what institutions actually are. The answer that emerges from economics is a disparate set of factors that range from social norms to values and all the way to "property rights" and complex organizations such as corporations and agencies of the state (Haggard 2004; Williamson 1975, 1985). North (1990: 3) defined institutions as "any form of constraint that human beings devise to shape human interaction," a vague definition that encompasses everything from norms introjected in the process of socialization to physical coercion.

From this thin definition, all that can be said is that institutions exist when something exerts external influence on the behavior of social actors: the same notion that Durkheim identified as "norms" more than a century ago and not sufficient to capture the dynamics of communities and societies.

Neoinstitutionalism has also traveled to the realm of politics, where it has been used, as in economics, to denote the constraints that the social context puts on the actions of "rational man," thus leading to "bounded rationality" (Dolsak and Ostrom 2003; Elster, Offe, and Preuss 1998). While itself unimpeachable, this assertion leaves open the question of what are the features of social context that actually "bound" rational action. Saying simply that everything depends on time and place leads us nowhere theoretically, as this statement is nonfalsifiable.

Moving things further, Elinor Ostrom has proposed a neoinstitutional analysis of the "Commons," seeking to solve the dilemma between self-interest and the collective good among users of the same readily available, but exhaustible common property resources. Ostrom (1990; Ostrom et al. 2002) argues that neither the state nor the market does a very good job in these situations, since they seek to impose external rules on the relevant actors. Rather, actors can devise their own enforceable institutional arrangements (i.e., norms) to escape the tyranny of atomized self-interest. These norms again vary with time and place. As we will see shortly, Ostrom's analysis is compatible with a sociologically informed analysis of institutional development, but the latter has the advantage of going beyond the simple assertion that such arrangements vary with the local context.

In sum, development economists and neoinstitutionalists seek to flesh out North's insight that social constraints matter. But in the absence of a solid theoretical framework, the practical results of this institutional turn have been what might be expected. In the hands of development practitioners, the new consensus has led to the attempted export of legal codes and organizational blueprints to the global South. The dismal results of such attempts have already been recognized (Evans 2004b; Hoff and Stiglitz 2001). However, we can do more than point out that such efforts are doomed from the start. Economists and other social scientists can draw on established theoretical traditions to sharpen their conceptual tools and devise a more sophisticated and useful mapping of social life. Sociologists can contribute to this enterprise by refining their own conceptual legacy. The resulting "thick institutionalism" seems preferable, in most instances, to the "thin" version now making the rounds in several disciplines.

The basis for interdisciplinary collaboration is already at hand and consists of a body of knowledge containing key elements for the analysis of what actually takes place in society and for the proper placement of the concept "institution." These elements include (1) a distinction between the symbolic realm and the material reality; (2) an understanding of the hierarchical character of both realms; (3) an identification of the linchpin concepts linking both; and (4) a theory of social change that goes beyond current institutionalist understandings of this process.

Culture and Social Structure

From its classical beginnings, modern sociology developed a central distinction, consolidated by the mid-twentieth century, between culture and social structure. There are good reasons for this distinction. Culture embodies the symbolic elements crucial for human interaction, mutual understanding, and order. Social structure is composed of actual persons enacting roles organized in a status hierarchy of some kind. The distinction is analytic because only human beings exist in reality, but it is fundamental to understand both the motives for their actions and the consequences. Culture is the realm of values, cognitive frameworks, and accumulated knowledge. Social structure is the realm of interests, individual and collective, backed by different amounts of power. The symbolic distinction provides the basis for analyzing the difference between what "ought to be" or "is expected to be" and what actually "is" in multiple social contexts (Merton 1936, 1968a).

The diverse elements that compose culture and social structure can be arranged in a hierarchy of causative influences from "deep" factors, often concealed below everyday social life but fundamental for its organization, to "surface" phenomena, more mutable and more readily evident. Language and values are the deep elements of culture, the first as the fundamental instrument of human communication and the second as the motivating force behind "principled" action, individual or collective. The importance of values can range, in turn, from fundamental moral imperatives of a society to traditions prized mostly out of custom. In every instance, values point toward a clear continuum between the good and desirable and the bad and abhorrent. "Neutrality" is the exact opposite of this basic element of culture (Durkheim [1897] 1965; Weber [1904] 1949). Values are deep culture because they are seldom invoked in the course of everyday life. Values come to the fore only in exceptional circumstances (Weber [1904] 1949; Merton 1989). Yet they underlie, and are inferred from, aspects of everyday behavior that are the opposite of unrestrained self-interest, the "constraints" that North, Ostrom, and others refer to.

Norms are such constraints. Values are not norms. The distinction is important: values represent general moral principles, and norms embody concrete directives for action (Newcomb, Turner, and Converse 1965; MacIver and Page [1949] 1961). Values underlie norms, which are rules that prescribe the do's and don'ts of individual everyday conduct. These rules can be formal and codified into constitutions and laws, or they can be implicit and informally enforced. The concept of norms has been used, at least since Durkheim ([1901] 1982), to refer to this restraining element of culture. Neglect of these classical analyses has led to lumping norms with the term institution, which has another, and important, connotation, as seen below. The significance of the values embodied within norms is reflected in practice in the level of sanctions attached to the latter. Thus life in prison or the death penalty awaits those found guilty of deliberate murder, while loud protest and insulting remarks may be the lot of those seeking to sneak ahead of a queue (Cooley 1902, 1912; Simmel [1908] 1964; Goffman 1959).

Norms are not free floating but come together in organized bundles known as roles. This sociological concept has been neglected in the institutionalist literature, which thus deprives itself of a key analytic tool. For it is as role occupants that individuals enter into the social world and are subject to the constraints and incentives of norms. Roles are generally defined as the set of behaviors prescribed for occupants of particular social positions (Linton 1945; Newcomb 1950: chap. 3). Well-socialized persons shift from role to role effortlessly and often unconsciously as part of their daily routines. The normative blueprints that constitute a role generally leave considerable latitude for their individual enactment. An extensive literature in both sociology and social psychology has analyzed roles as the building blocks of social life and as one of the linchpin concepts linking the symbolic world of culture to real social structures. The same literature has examined such dynamics as the "role set" enacted by given social actors and the "role conflict" or "role strain" created when normative expectations in an actor's role sets contradict each other (Cottrell 1933; Linton 1945; Merton 1957; Goffman 1959, 1961; Goode 1960). None of these concepts has made its appearance in the sociology being created in economics. Roles are an integral part of institutions, but they are not institutions, and confusing the two terms weakens the heuristic power of both concepts.

Along with normative expectations, roles embody an instrumental repertoire of skills necessary for their proper enactment. Language is the fundamental component of this repertoire for, without it, no other skills can be enacted. These cultural "tool kits" also contain many other elements-from scientific and professional know-how to demeanor, forms of expressions, manners, and general savoir faire suitable for specific social occasions. In the modern sociological literature, these elements are referred to by the concepts "cultural capital" and "skills repertoires" (Bourdieu 1979, 1984; Swidler 1986; Zelizer 2005).

Power, Class, and Status

Parallel to the component elements of culture run those of social structure. These are not made up of moral values or generalized do's and don'ts but involve the specific and differentiated ability of social actors to compel others to do their bidding. This is the realm of power, which, like that of values, is situated at the "deep" level of social life influencing a wide variety of outcomes. Weber's classic definition of power as the ability of an actor to impose his or her will despite resistance is still appropriate, for it highlights the compulsory and coercive nature of this basic element of social structure. It does not depend on the voluntary consent of subordinates, and for some actors and groups to have power others must be excluded from access to it (Weber [1922] 1947; Veblen [1899] 1998; Mills 1959). While values motivate or constrain, power enables. Naturally, elites in control of power-conferring resources seek to stabilize and perpetuate their position by molding values so that the mass of the population is persuaded of the "fairness" of the existing order. Power thus legitimized becomes authority, in which subordinates acquiesce to their position (Weber [1922] 1947; Bendix 1962: chaps. 9, 10).

In Marx's classic definition, power depends on control of the means of production, but in the modern postindustrial world this definition is too restrictive (Marx [1939] 1970; [1867] 1967: pt. 7). Power is conferred as well by control of the means of producing and appropriating knowledge, by control of the means of diffusing information, and by the more traditional control of the means of violence (Weber [1922] 1947; Wright 1980, 1985; Poulantzas 1975). In the Marxist tradition, a hegemonic class is one that has succeeded in legitimizing its control of the raw means of power, thus transforming it into authority (Gramsci [1927-33] 1971; Poulantzas 1975). Power is not absent from contemporary institutional economics, but the emphasis is on authority relations-what Williamson (1975, 1985) calls "hierarchies." Although these analyses are important, they neglect more basic forms of power. This omission supports Hodgson's argument on the lack of tools in modern economics to understand what institutions really are. For, as we shall see shortly, actual institutions are molded, to a large extent, by power differentials.

Just as values are embodied in norms, so power differentials give rise to social classes-large aggregates whose possession of or exclusion from resources leads to varying life chances and capacities to influence the course of events. Classes need not be subjectively perceived by their occupants in order to be operative, for they underlie the obvious fact that people in society are ranked according to how far they are able to implement their goals when confronted with resistance (Wright 1985; Wright and Perrone 1976; Poulantzas 1975). Class position is commonly associated with wealth, but it is also linked to other power-conferring resources such as expertise or the "right" connections (Hout, Brooks, and Manza 1993; Bourdieu 1984, 1990; Portes 2000a). As emphasized by Bourdieu (1985), dominant classes generally command a mix of resources that include not only wealth but also ties to influential others (social capital) and the knowledge and style to occupy high-status positions (cultural capital).

The deep character of power seldom comes to the surface of society, for its holders aim to legitimize it in the value system in order to obtain the voluntary consent of the governed. For the same reason, class position is not readily transparent, and it is a fact, repeatedly verified by empirical research, that individuals with very different resources and life chances frequently identify themselves as members of the same "class" (Hout, Brooks, and Manza 1993; Grusky and Sorensen 1998). Legitimized power (authority) produces, in turn, status hierarchies. Most social actors actually perceive the underlying structure of power on the basis of such hierarchies and classify themselves accordingly. In turn, status hierarchies are commonly linked to the performance of occupational roles (MacIver and Page [1949] 1961; Newcomb, Turner, and Converse 1965: 336-41; Linton 1945).

The various elements of culture and social structure, placed at different levels of importance and visibility, occur simultaneously and appear, at first glance, as an undifferentiated mass. Their analytic separation is required, however, for the proper understanding of social phenomena, including economic phenomena. Not everything is "constraints on behavior"; some elements constrain, others motivate, and still others enable. Economists have not done the conceptual spadework required to understand these differences. The framework outlined thus far is summarized in figure 1.1. As the citations accompanying the text suggest, this framework is neither new nor improvised but forms part of a classical intellectual legacy neglected in the current enthusiasm for the institutionalist turn. 

Institutions in Perspective

As indicated in figure 1.1, status hierarchies with attached roles do not generally exist in isolation but as part of social organizations. Organizations, economic and otherwise, are what social actors inhabit, and they embody the most readily visible manifestations of the underlying structures of power (Powell 1990; DiMaggio 1990; Granovetter 2001). Institutions represent the symbolic blueprint for organizations; they are the set of rules, written or informal, governing relationships among role occupants in social organizations like the family, schools, and other major institutionally structured areas of organizational life: the polity, the economy, religion, communications and information, and leisure (MacIver and Page [1949] 1961; Merton 1968b; North 1990; Hollingsworth 2002).

This definition of institutions is in close agreement with everyday uses of the term, as when one speaks of "institutional blueprints." Its validity does not depend, however, on this overlap but on its analytic utility. The position advanced here is nominalist: Concepts are mental constructs whose usefulness is given by their collective capacity to guide our understanding of social phenomena, including the economy. If North and his followers denominate norms "institutions," then they must cope with the conceptual problem of the relationship between such "institutions" and the roles in which they are embedded, as well as the symbolic blueprints specifying relationships among such roles and, hence, the actual structure of organizations.

A "thick" institutionalism that limits the scope of this concept, while systematically relating it to other elements of social life, gives us the necessary analytic leverage to understand phenomena that otherwise would be obscured. For example, the distinction between organizations and the institutions that underlie them provides a basis for analyzing how events actually occur in social and economic life. For it is not the case that once institutional rules are established, role occupants blindly follow. Instead, they constantly modify the rules, transform them, and bypass them in the course of their daily interaction.

No doubt, "institutions matter," but they are themselves subject to what Granovetter (1985, 1992) referred to as "the problem of embeddedness": The fact that the human exchanges that institutions seek to guide in turn affect these institutions. That is why formal goals and prescribed organizational hierarchies come to differ from how organizations operate in reality (Dalton 1959; Morrill 1991; Powell 1990). Absent this analytic separation, as well as the understanding that institutions and organizations flow from deeper levels of social life, everything becomes an undifferentiated mass where the recognition that "contexts matter" produces, at best, descriptive case studies and, at worst, circular reasoning. The following sections seek to put this conceptual framework into motion on the basis of two recent examples from the development literature.

The Failure of Institutional Monocropping

The most tangible practical result of the advent of institutionalism in the field of economic development has been the attempt to transplant the institutional forms of the West, especially the United States, into the less developed world. The definition of institutions employed in such attempts is in close agreement with that advanced here: blueprints specifying the functions and prerogatives of roles and the relationships among their occupants. Institutions and the resulting organizations may be created from scratch-as a central bank, a stock exchange, or an ombudsman office-or they may be remolded-as in attempts to strengthen the independence of the judiciary or streamline the local legislature (Haggard 2004).

Many authors have noted that these attempts to put the ideas of North and other institutionalists into practice have not yielded the expected results and have frequently backfired. Evans (2004b), in particular, calls these exercises in transplantation institutional monocropping, whereby the set of rules constructed by trial and error over centuries in the advanced countries is grafted into different societies and expected to have comparable results. Roland (2004) diagnoses the cause of the failures of such efforts as lying in the gap between "slow-moving" and "fast-moving" institutions, but the actual forces at play are much more complex.

Institutional grafting takes place at the surface level of society and, as such, faces the potential opposition of a dual set of forces grounded in the deep structure of the receiving countries: those based on values and those based on power. Within the realm of culture, consider the different bundles of norms and cultural tool kits that go into formally similar roles. That of "policeman" may entail, in less developed societies, the expectation to compensate paltry wages with bribe taking, a legitimate preference for kin and friends over strangers in the discharge of duties, and skills that extend no further than using firearms and readily clubbing civilians at the first sign of trouble. The role of "government minister" may similarly entail the expectation of particularistic preferences in the allocation of jobs and government patronage, appointments by party loyalty rather than expertise, and the practice of using the power of the office to ensure the long-term economic well-being of the occupant.

Such role expectations are grounded in deeply held values that privilege particularistic obligations and ascriptive ties and that encourage suspicion of bureaucracies and seemingly universalistic rules. When imported institutional blueprints are superimposed on such realities, the results are not hard to imagine. Blueprints do not necessarily backfire, but they can have a series of unexpected consequences following from the fact that those in charge of their implementation and the presumed beneficiaries view reality through very different cultural lenses (O'Donnell 1994; Portes 1997, 2000b).

Several authors, including economists who have analyzed these dynamics, recognize the importance of power. Karla Hoff and Joeseph Stiglitz (2001: 418-20) note, for example, that imposing new sets of formal rules without simultaneously reshaping the distribution of power is a dubious strategy. Similarly, Roland (2004: 115) recognizes that "whatever group holds power will use that power in its own best interest." Less well understood are two other key features of social structures. The first is that "power" is not free floating but depends on control of certain strategic resources-capital, means of production, organized violence-that vary from country to country. Second, and more important, the existing class structure may be legitimized by the value system in such a way that change is resisted not only by those in positions of privilege, but by the mass of the population. As Weber and Gramsci recognized, legitimized power is particularly hard to dislodge because the masses not only acquiesce to their own subordination, but stand ready to defend the existing order. The experiences of "modernizing" regimes seeking to dislodge entrenched theocracies in the Middle East and elsewhere illustrate the decisive role of this kind of power (Lerner 1958; Levy 1966; Bellah 1958).

Following the argument of another Nobel Prize winner, Amartya Sen, Evans (2004b) offers an alternative to institutional monocropping, which he labels "deliberative development." Sen's argument for participatory democracy starts with the notion that "thickly democratic" initiatives, built on public discussion and free exchange of ideas, offer the only way to reach viable developmental goals. For Sen (1999), thickly democratic participation is not only a means to an end, but a developmental goal in itself. Evans agrees and cites the "participatory budgeting" process in Brazilian cities dominated by parties of the left as examples of the viability of deliberative development (Baiocchi 2003).

The conceptual framework discussed previously is useful for envisioning the contrast between institutional grafting and deliberative development. As shown in figure 1.2, the idea of importing institutions begins at the surface level and tries to push its way "upward" into the normative structure and value system of society. For reasons already seen, such efforts are likely to meet resistance and frequent failure. The participatory strategy begins at the other end, by engaging the population in a broad discussion of developmental goals (values) and the rules (norms) and technical means (skill repertoires) necessary to attain them. Although the process is messy and complicated, the institutional blueprints that eventually emerge from these discussions are likely to be successful because they correspond to the causal directionality of culture itself. [Figure 1.2 about here]

A key problem with deliberative development proposals is that they ignore the right-side elements of society, as outlined in figure 1.1, namely, those grounded in power and crystallized in the class structure. Unless the dominant classes are somehow persuaded or compelled to go along with such experiments, the latter are not likely to succeed. If implemented against elite resistance, they are bound to be derailed into just talk-deliberation as an end in itself. When the population mobilized to take part in such meetings sees that they lead to nothing or produce outcomes predetermined by the authorities, participation drops rapidly and generalized discontent sets in (Roberts and Portes 2005; Roberts 2002).

As Sen (1999) himself recognizes, technocrats (i.e., technically trained elites) prefer to impose institutional blueprints that enhance their power and external image rather than subordinate themselves to the messy deliberations of ordinary people. Evans (2004b: 40) acknowledges as well that the dynamics of power are likely to be the biggest impediment to the "institutionalization of deliberative institutions." Not surprisingly, only when parties of the left have gained solid control of governments have experiments in participatory democracy been given a reasonable chance of success. This occurs because authorities can mobilize the resources of government to neutralize resources possessed by local elites, persuading them that it is in their interest to join the deliberative process (Biaocchi 2003; Agarwala 2004).

The Privatization of the Mexican Economy

Starting in 1982, the Mexican government began a massive program of divestiture of the many companies that it had created. This program amounted to a radical departure from the previous state-centric model of development and touched the interests and life chances of almost everyone in Mexican society (Centeno 1994; Ariza and Ramirez 2005). The shift came in the aftermath of the Mexican default of 1982 and the conditions imposed by the International Monetary Fund (IMF) and the U.S. Treasury to bail out the country. Over the next three sexenios (presidential terms), the Mexican state divested itself of almost everything-from the telecommunications company to the two national airlines (Mexicana and Aeromexico).

This massive economic realignment was not accomplished without resistance. A great deal of money was to be made from privatizations, but there were also a number of actors who lost power, wealth, or their jobs. Mexico's privatization of the economy amounted to drastic institutional change-a profound modification of the legal/normative blueprints under which firms operate and their internal organization. This transformation, however, could not have been accomplished at the level of the institutions themselves.

State-owned firms operated with a logic of their own, creating constituencies around themselves. Although frequently inefficient, they gave secure employment to many and political capital to the line ministers and managers that operated them (Lomnitz 1982; Eckstein 1977). Thus Aeromexico operated with a staff of 200 employees per airplane at a time when the inefficient and about-to-be-bankrupt Eastern Airlines had 146. Yet the minute that plans for Aeromexico's restructuring were announced, its employees struck, arguing that the firm would be profitable "if only" management were more efficient (MacLeod 2004: 123, 133).

The battle for divestment and market opening pitted the unions, the managers of state-owned industries, and the ministries that supervised them against a group of reformers imbued with the new neoliberal doctrines at the Treasury Ministry and other strategic places in the Mexican bureaucracy. Large capitalists, foreign multinationals, and the IMF supported divestiture and opening; small firm owners who had much to lose from the removal of state protection opposed it (MacLeod 2004: 96).

During President Miguel de la Madrid's sexenio, only smaller and relatively marginal firms were privatized. Defenders of the status quo could keep faith that the strong corporatist traditions of the ruling party, the PRI, would in the end prevail. Despite sustained external pressure, institutions (i.e., state-owned corporations) would not reform themselves, and attempts at reform were effectively resisted: "From their positions on the executive committees and boards of directors of parastate firms, line ministers could keep a watchful eye on the efforts of would-be reformers... line ministers withheld data or presented contradictory or incorrect data, making it virtually impossible to evaluate the company" (MacLeod 2004: 75-76, see also 71).

True reform, as the IMF and the multinational corporations envisioned it, could only come from the top of the power structure. This actually happened during the next sexenio under President Carlos Salinas de Gortari. A convinced free marketeer, Salinas appointed economists of the same persuasion to key positions in the Central Bank and the Treasury Ministry. Once there, they created new, compact, and powerful agencies to ensure that privatization would move forward. Not willing to believe that things would take such a turn for the worse, union leaders and firm managers bypassed the new bureaucratic structures to take their case directly to the president. To no avail: "When the UDEP (Unit for the Divestiture of Parastate Entities) began the process of privatizing parastate firms, labor leaders, line ministers, and executives of parastate firms often sought to circumvent the authority of UDEP by appealing directly to the President. President Salinas regularly sent these supplicants back to the director of UDEP.... [T]his process quickly consolidated UDEP's authority within the Mexican bureaucracy" (MacLeod 2004: 81-82).

The "sale of the state" engineered by UDEP in subsequent years amounted to a major case of institutional transformation; it also represents a clear example of the dynamics of power. As shown in figure 1.3, reforms initiated from the outside barely made a dent in the Mexican corporatist structure. It was necessary for the country's top political and economic leadership to get involved in order to overcome strong and organized resistance from various classes. Unionized workers and national entrepreneurs became the losers in this giant power struggle that saw the Mexican labor market become far more "flexible" and the Mexican corporations far more open to external competition and takeover (Shaiken 1990, 1994; Ariza and Ramirez 2005). As elsewhere, significant institutional and organizational change did not originate with organizations themselves but required major transformations at deeper levels of the social structure.

However, just as attempted transformations of existing institutions can meet with resistance from power holders in the social structure, power plays that impose institutional change can produce generalized opposition when the underlying values remain unaltered. The Salinas reforms took place against a background of public skepticism about the need to denationalize the economy and strong opposition from many sectors of Mexican society (MacLeod 2004). Salinas ended his term in disgrace, becoming an unpopular figure and eventually being forced to leave the country.

The Problem of Change

Diffusion and Path Dependence

In his recent book, Institutional Change and Globalization (2004), John Campbell systematically describes the different schools of institutional analysis that exist today. These he labels "rational choice institutionalism," associated primarily with economics; "organizational institutionalism," associated with the sociology of organizations; and "historical institutionalism," based on political economy and certain strands of political science. Depending on the school, social change is seen primarily as an evolutionary process, developing gradually over time, or as a combination of evolution and "punctuated evolution" when drastic shifts occur.

Despite these differences, Campbell characterizes all three schools as favoring two major determinants of change. These are "path dependence," meaning the tendency of events to follow a set course where what existed yesterday determines what happens today and what is likely to occur tomorrow (Thelen 2004; North 1990); and "diffusion," meaning the tendency of established institutional patterns to migrate, influencing the course of events. Diffusion is identified by the school led by John Meyer as a master process in the contemporary global system in which the institutions of the advanced countries are commonly reproduced in weaker societies, either under the aegis of international agencies or out of the desire of local rulers to imitate the modern world (Meyer and Hannan 1979; Meyer et al.1997).

Campbell argues that "the problem of change" has been a thorny one for institutional analysis. This is not difficult to understand. First, with a vague and contested definition of institution, the analysis of change confronts an elusive target. When institution can be anything-from the incest taboo to the Central Bank-we do not have a sufficiently delimited object to examine how it changes over time. The definition advanced here-sets of rules that govern the regular relationships among role occupants-is sufficiently specific to allow consideration of how processes of change take place in this sector of social life. Thus defined, institutional change is not the same as change in the class structure or in the value system, processes that ultimately affect institutions but that occur at other levels.

Second, with concepts such as path dependence and diffusion as its main tools for the analysis of change, it is not difficult to understand how the predicted course of events for institutional analysis would be either evolution or punctuated evolution. It is a fact that, at the surface level of social life, change tends to be gradual, with patterned ways of doing things largely determining the future course of events.

Intersocietal diffusion of cultural elements operates at a deeper level, affecting the normative and skill contents of specific roles. Diffusion of new technologies and patterns of consumption from the advanced world to the less developed countries is indeed one of the most common and most important sources of change in these countries (Sassen 1988; Meyer et al. 1997).

But dynamics of change are not limited to diffusion and path dependence; they can also exist at deeper levels of the culture and social structure, producing drastic and nonevolutionary outcomes. To be sure, as argued by some institutionalists, radical change tends to have long periods of gestation, but this does not negate the fact that once such change occurs, consequences can be abrupt and often traumatic. Technological changes, to take one example, can be endogenous rather than brought about by diffusion. Once they occur, technological breakthroughs can affect, in a very short time, the skills repertoires and, hence, the roles of large numbers of social actors. One such example is the advent of the Internet, an innovation that has altered the content of occupational roles and the rules linking them in most institutions of modern society (Castells 1998, 2001).

Religion and religious prophecies can affect the culture in still more radical ways because they impinge directly on the value system (Wuthnow 1987, 1998). Weber's theory of social change focuses on the history of religion and specifically on the role of charisma and charismatic prophecy as forces capable of breaking through the limits of reality, as hitherto known, and providing the impetus necessary to dismantle the existing social order and rebuild it on a new basis. The influence of the Protestant Reformation, especially Calvinism, in revolutionizing economic life in Western Europe is perhaps the best-known illustration of the effects that charismatic prophecy can have on society (Weber [1922] 1964, [1915] 1958).

The advent of charismatic prophecy capable of revolutionizing the value system and, hence, an entire civilization occurs after a long period of historical gestation, but this does not prevent it from having immediate and profound consequences once it bursts onto the scene. After Calvinism had transformed the social order of much of Western Europe, historians had little difficulty tracing the concatenation of events that led to it. But they would not have bothered to engage in such an exercise had Luther not nailed his theses at Wittenberg and had Calvin not risen to power in Geneva. Post hoc reconstruction of revolutionary social change can always be "evolutionary."

For those who dismiss the role of religious charisma as a thing of the past, one need only point to the decisive influence that Evangelical Christianity continues to have in transforming large portions of American society (Wuthnow 1998; Roof 1999) and to the emergence of a fundamentalist brand of Islam set on ultimate confrontation with the West. The "war on terrorism" that is today the overriding concern of states in North America and Europe is interpretable as a direct consequence of a reenergized, charismatic religious prophecy seeking to remake the world in its own image (Kastoryano 2004; Kepel 1987).

Revolutionary change can also occur in the realm of social structure, as when power is wrested from its current possessors and vested in a new elite. The question of power and of class struggles has been addressed by a long line of historians and social scientists, classic and contemporary. Vilfredo Pareto's ([1902] 1966) theory of the circulation of elites and his remark that "history is but a graveyard of aristocracies" focus on the fact that dominant groups have never been able to maintain power indefinitely and on the analysis of the mechanisms leading to their demise. From very different theoretical quarters, Marx privileged the class struggle and, at a deeper level, the conflict between new modes of production and entrenched "social relationships of production" as the master mechanism leading to revolutionary change. For Marx and his numerous followers, the internal contradictions of feudalism that brought about its end were being re-created anew under capitalism, as rising social classes clashed with the dominant class structure (Marx and Engels [1847] 1959: 20).

Much of contemporary historical sociology-including the writings of Barrington Moore (1966), Theda Skocpol (1979), Charles Tilly (1984), Immanuel Wallerstein (1974, 1991), and Giovanni Arrighi (1994)-is concerned with the same question of revolutionary change and elite replacement. To take one example, Skocpol's (1979) theory of social revolutions highlights interelite conflict, external military pressure, and an oppressed peasantry as factors that, when coming together, can drastically transform the class structure and bring about a new social order. While structural change of this magnitude occurs after a lengthy concatenation of events, this is made evident only by the moment of social explosion itself and the ensuing events.

From the perspective of the profound consequences wrought by transformations in a society's value system or class structure, a theory of change based on path dependence and cultural diffusion looks limited indeed. Change-whether revolutionary or not-at deeper levels of the culture and social structure filters upward to the more visible levels, including institutions and organizations. Thus it is possible to distinguish at least five forces impinging on institutions and leading to their transformation: path dependence, producing evolutionary change at the more visible institutional level; diffusion, also leading to evolutionary and sometimes "punctuated" change at the intermediate levels of culture; scientific/technological breakthroughs affecting the cultural skills repertoires and normative order; at a deeper level, charismatic prophecy-religious or secular-capable of transforming the value system and, hence, the rest of the culture; and interelite and class struggles with the potential for transforming the distribution of power. The last three sources hold the potential for profound institutional change, of the type seen in the aftermath of social revolutions and epoch-making inventions.

Figure 1.4 summarizes this discussion. Campbell (2004) concludes his review of institutional change by recommending that we consider such a process only within well-limited time frames and "in its multiple dimensions." These recommendations are unobjectionable, but they do not go far enough. While limited time frames prevent infinite regress into history, they do not distinguish between evolutionary change over a given period and abrupt, revolutionary transformations. Similarly, the "multiple dimensions" to be considered in the analysis of change are left unspecified.

A conceptual framework such as that outlined here helps distinguish between different elements of culture and social structure and the relative impact of processes of change taking place at different levels. An institutional analysis of change limited to institutions themselves produces an impoverished account of these processes, relative to what the social sciences in general and sociology in particular have already accomplished.

Conclusion

Disciplinary amnesia is perhaps inevitable as new generations of scholars seek to make their mark in the world. The unfortunate consequence, however, is the rediscovery or reelaboration of what had already been found.

Advocates of the "institutions are everything" approach may reply that the conceptual framework proposed in this chapter is dated since it is based on the work of nineteenth- and early-twentieth-century precursors. They may add that there has been progress since then, and that "thin institutionalism"-a loose definition of the concept-is more flexible and, for that reason, preferable in many circumstances. To this, we answer that progress is indeed desirable but that, with the exception of debunking the patently implausible assumptions of neoclassical economics, neoinstitutionalism is still far from achieving its potential. We would attribute this failure, first, to the neglect of a rich theoretical heritage and, second, to careless definitions. It is impossible to cumulate scientific knowledge when master concepts can mean practically anything. No better conceptual framework has been devised to replace that which is our legacy from earlier generations of thinkers and researchers. For that reason alone, thick institutionalism-a precise definition of the concept placed within a systematic framework-is preferable as the basis for cumulative progress.

The theoretical synthesis presented here is tentative and subject to modification. No intrinsic truth is claimed for it, save its utility for delimiting the scope of the concept of institutions and moving us away from an impoverished understanding of social change. The comparative study described next puts this agenda into motion by analyzing really existing institutions and their bearing on national development.

References

Agarwala, Rina. 2004. "From Work to Welfare: The State and Informal Workers' Organizations in India." Working Paper #04-07. Center for Migration and Development, Princeton University.

Ariza, Marina, and Juan Manuel Ramírez. 2005. "Urbanización, mercados de trabajo y escenarios sociales en el México finisecular." In Las ciudades latinoamericanas a comienzos del siglo, edited by A. Portes, B. R. Roberts, and A. Grimson, 299-361. Buenos Aires: Prometeo Editores.

Arrighi, Giovanni. 1994. The Long Twentieth Century: Money, Power, and the Origins of Our Times. London: Verso.

Baiocchi, Gianpaolo. 2003. Radicals in Power: The Workers Party and Experiments in Urban Democracy in Brazil. London: Zed.

Bellah, Robert N. 1958. "Religious Aspects of Modernization in Turkey and Japan." American Journal of Sociology 64 (July): 1-5.

Bendix, Reinhard. 1962. Max Weber: An Intellectual Portrait. Garden City, NY: Anchor Books.

Bourdieu, Pierre. 1979. "Les trois états du capital culturel." Actes de la Recherche en Sciences Sociales 30: 3-6.

---. 1984. Distinction: A Social Critique of the Judgment of Taste. Cambridge, MA: Harvard University Press.

---. 1985. "The Forms of Capital." In Handbook of Theory and Research for the Sociology of Education, edited by J. G. Richardson, 241-58. Westport, CT: Greenwood Press.

---. 1990. The Logic of Practice. Stanford: Stanford University Press.

Campbell, John L. 2004. Institutional Change and Globalization. Princeton, NJ: Princeton University Press.

Castells, Manuel. 1998. End of Millennium: The Information Age. Oxford: Blackwell Publishers.

---. 2001. The Internet Galaxy. New York: Oxford University Press.

Centeno, Miguel A. 1994. Democracy within Reason: Technocratic Revolution in Mexico. University Park: Pennsylvania State University Press.

Cooley, Charles H. 1902. Human Nature and the Social Order. New York: Charles Scribners and Sons.

---. 1912. Social Organization. New York: Charles Scribners and Sons.

Cottrell, Leonard S. 1933. "Roles and Marital Adjustment." Publications of the American Sociological Society 28.

Dalton, Melville. 1959. Men Who Manage: Fusions of Feeling and Theory in Administration. New York: Wiley.

DiMaggio, Paul. 1990. "Cultural Aspects of Economic Action and Organization." In Beyond the Marketplace, edited by R. Friedlander and A. F. Robertson, 113-36. New York: Aldine de Gruyter.

Dolsak, Nines, and Elinor Ostrom. 2003. The Commons in the New Millennium: Challenges and Adaptation. Cambridge, MA: MIT Press.

Durkheim, Emile. [1897] 1965. Suicide, a Study in Sociology. Translated by J. A. Spaulding and G. Simpson. New York: Free Press.

---. [1901] 1982. The Rules of the Sociological Method. Translated by W. D. Halls. New York: Free Press.

Eckstein, Susan. 1977. The Poverty of Revolution: The State and the Urban Poor in Mexico. Princeton, NJ: Princeton University Press.

Elster, Jon, Claus Offe, and Ulrick K. Preuss. 1998. Institutional Design in Post-Communist Societies: Rebuilding the Ship at Sea. Cambridge: Cambridge University Press.

Evans, Peter. 1979. Dependent Development: The Alliance of Multinational, State, and Local Capital in Brazil. Princeton: Princeton University Press.

---. 1989. "Predatory, Developmental, and Other Apparatuses: A Comparative Political Economy Perspective on the Third World State." Sociological Forum 4: 561-87.

---. 1995. Embedded Autonomy: States and Industrial Transformation. Princeton: Princeton University Press.

---. 2004a. "The Challenges of the 'Institutional Turn': Interdisciplinary Opportunities in Development Theory." In The Economic Sociology of Capitalism, edited by V. Nee and R. Swedberg, 90-116. Princeton: Princeton University Press.

---. 2004b. "Development as Institutional Change: The Pitfalls of Monocropping and the Potentials of Deliberation." Studies in Comparative International Development 38 (Winter): 30-52.

Giddens, Anthony. 1987. Las nuevas reglas del método sociológico. Buenos Aires: Amorrortu.

Goffman, Erving. 1959. The Presentation of Self in Everyday Life. Garden City, NY: Doubleday.

---. 1961. Encounters: Two Studies in the Sociology of Interaction. Indianapolis, IN: Bobbs-Merrill.

Goode, William J. 1960. "A Theory of Role Strain." American Sociological Review 25 (August): 483-96.

Gramsci, Antonio. [1927-33] 1971. "State and Civil Society." In Prison Notebooks, translated and edited by Q. Hoare and G. N. Smith, 206-76. New York: International Publishers.

Granovetter, Mark. 1985. "Economic Action and Social Structure: The Problem of Embeddedness." American Journal of Sociology 91: 481-510.

---. 1992. "The Sociological and Economic Approaches to Labor Market Analysis: A Social Structural View." In The Sociology of Economic Life, edited by M. Granovetter and R. Swedberg, 233-63. Boulder, CO: Westview Press.

---. 2001. "Coase Revisited: Business Groups in the Modern Economy." In The Sociology of Economic Life, 2nd ed., edited by M. Granovetter and R. Swedberg, 327-56. Boulder, CO: Westview Press.

Grusky, David B., and Jesper B. Sorensen. 1998. "Can Class Analysis Be Salvaged?" American Journal of Sociology 103 (March):1187-1234.

Haggard, Stephan. 2004. "Institutions and Growth in East Asia." Studies in Comparative International Development 38 (Winter): 53-81.

Hamilton, Gary, and Nicole W. Biggart. 1988. "Market, Culture, and Authority: A Comparative Analysis of Management and Organization." American Journal of Sociology 94 (Supplement): 552-94.

Hirschman, Albert O. 1958. The Strategy of Economic Development. New Haven, CT: Yale University Press.

---. 1963. Journeys toward Progress. New York: Twentieth Century Fund.

Hodgson, Geoffrey M. 2002. "Institutional Blindness in Modern Economics." In Advancing Socio-Economics: An Institutionalist Perspective, edited by J. R. Hollingsworth, K. H. Muller, and E. J. Hollingsworth, 147-70. Landham, MD: Rowman and Littlefield.

Hoff, Karla, and Joseph Stiglitz. 2001. "Modern Economic Theory and Development." In Frontiers of Development Economics, edited by G. Meier and J. Stiglitz, 389-460. New York: Oxford University Press.

Hollingsworth, J. Rogers. 2002. "On Institutional Embeddedness." In Advancing Socio-Economics: An Institutionalist Perspective, edited by J. R. Hollingsworth, K. H. Muller, and E. J. Hollingsworth, 87-107. Landham, MD: Rowman and Littlefield.

Hout, Michael, Clem Brooks, and Jeff Manza. 1993. "The Persistence of Classes in Post-industrial Societies." International Sociology 8 (September): 259-77.

Kastoryano, Riva. 2004. "Religion and Incorporation: Islam in France and Germany." International Migration Review 38 (Fall): 1234-55.

Kepel, G. 1987. Les Banlieues de l'Islam. Paris: Ed. du Seine.

Lerner, David. 1958. The Passing of Traditional Society: Modernizing the Middle East. New York: Free Press.

Levy, Marion. 1966. Modernization and the Structure of Societies. Princeton: Princeton University Press.

Linton, Ralph. 1945. The Cultural Background of Personality. New York: Appleton, Century, Crofts.

Lomnitz, Larissa. 1982. "Horizontal and Vertical Relations and the Social Structure of Urban Mexico." Latin American Research Review 17: 51-74.

MacIver, R. H., and Charles H. Page. [1949] 1961. Sociologia. Translated from English by J. Cazorla Perez. Madrid: Tecnos Editores.

MacLeod, Dag. 2004. Downsizing the State: Privatization and the Limits of Neoliberal Reform in Mexico. University Park: Pennsylvania State University Press.

Marx, Karl. [1939] 1970. The Grundrisse. Edited and translated by D. McLellan. New York: Harper and Row.

Marx, Karl, and Friedrich Engels. [1847] 1959. "Manifesto of the Communist Party." In Marx and Engels: Basic Writings on Politics and Philosophy, edited by L. S. Feuer, 1-41. Garden City, NY: Doubleday.

Merton, Robert K. 1936. "The Unanticipated Consequences of Purposive Social Action." American Sociological Review 1: 894-904.

---. 1957. "The Role-Set: Problems in Sociological Theory." British Journal of Sociology 8 (June): 106-20.

---. 1968a. "Manifest and Latent Functions." In Social Theory and Social Structure, edited by R. K. Merton, 73-138. New York: Free Press.

---. 1968b. "Social Structure and Anomie." In Social Theory and Social Structure, edited by R. K. Merton, 175-214. New York: Free Press.

---. 1989. "Unanticipated Consequences and Kindred Sociological Ideas: A Personal Gloss." In L'Opera di R. K. Merton e la sociologia contemporanea, edited by C. Mongardini and S. Tabboni, 307-29. Genova: Edisioni Culturali Internazionali.

Meyer, John, John Boli, George M. Thomas, and Francisco Ramirez. 1997. "World Society and the Nation State." American Journal of Sociology 103 (July): 144-81.

Meyer, John, and Michael T. Hannan. 1979. National Development and the World System: Educational, Economic, and Political Change, 1950-1970. Chicago: University of Chicago Press.

Mills, C. Wright. 1959. The Power Elite. London: Oxford University Press.

Moore, Barrington. 1966. Social Origins of Dictatorship and Democracy. Boston: Beacon Press.

Morrill, Calvin. 1991. "Conflict Management, Honor, and Organizational Change." American Journal of Sociology 97: 585-621.

Nee, Victor. 2005. "The New Institutionalisms in Economics and Sociology." In The Handbook of Economic Sociology, 2nd ed., edited by N. J. Smelser and R. Swedberg, 49-74. Princeton: Princeton University Press and Russell Sage Foundation.

Newcomb, Theodore M. 1950. Social Psychology. New York: Holt, Rinehart, Winston.

Newcomb, Theodore M., Ralph H. Turner, and Philip E. Converse. 1965. Social Psychology: The Study of Human Interaction. New York: Holt, Rinehart, and Winston.

North, Douglass C. 1990. Institutions, Institutional Change, and Economic Performance. Cambridge: Cambridge University Press.

O'Donnell, Guillermo. 1994. "The State, Democratization, and Some Conceptual Problems." In Latin American Political Economy in the Age of Neoliberal Reform, edited by W. C. Smith, C. H. Acuña, and E. A. Gamarra, 157-79. New Brunswick, NJ: Transaction.

Ostrom, Elinor. 1990. Governing the Commons: The Evolution of Institutions for Collective Action. Cambridge: Cambridge University Press.

Ostrom, Elinor, Thomas Dietz, Nives Dolsak, Paul C. Stern, Susan Stonich, and Elke Weber. 2002. The Drama of the Commons. Washington, DC: National Academy Press.

Pareto, Vilfredo. [1902] 1966. "Les systèmes socialistes." In Vilfredo Pareto: Sociological Writings, edited by S. E. Finer, 123-42. New York: Praeger.

Portes, Alejandro. 1997. "Neoliberalism and the Sociology of Development: Emerging Trends and Unanticipated Facts." Population and Development Review 23 (June): 229-59.

---. 2000a. "The Resilient Significance of Class: A Nominalist Interpretation." Political Power and Social Theory 14: 249-84.

---. 2000b. "The Hidden Abode: Sociology as Analysis of the Unexpected." American Sociological Review 65: 1-18.

Poulantzas, Nicos. 1975. Classes in Contemporary Capitalism. London: New Left Books.

Powell, Walter W. 1990. "The Transformation of Organizational Forms: How Useful Is Organization Theory in Accounting for Social Change?" In Beyond the Marketplace: Rethinking Economy and Society, edited by R. Friedland and A. F. Robertson, 301-29. New York: Aldine de Gruyter.

Roberts, Bryan R., and Alejandro Portes. 2005. "Coping with the Free Market City: Collective Action in Six Latin American Cities at the End of the Twentieth Century." Latin American Research Review 41 (June): 233-62.

Roberts, Kenneth. 2002. "Social Inequalities without Class Cleavages in Latin America's Neoliberal Era." Studies in Comparative International Development 36 (Winter): 3-33.

Roland, Gerard. 2004. "Understanding Institutional Change: Fast-Moving and Slow-Moving Institutions." Studies in Comparative International Development 38 (Winter): 109-31.

Roof, Wade Clark. 1999. Spiritual Marketplace: Baby Boomers and the Remaking of American Religion. Princeton, NJ: Princeton University Press.

Sassen, Saskia. 1988. The Mobility of Labor and Capital: A Study in International Investment and Labor Flow. New York: Cambridge University Press.

Sen, Amartya. 1999. Development as Freedom. New York: Knopf.

Shaiken, Harley. 1990. Mexico in the Global Economy: High Technology and Work Organization in Export Industries. Monograph Series 33. San Diego: Center for U.S.-Mexican Studies, University of California.

---. 1994. "Advanced Manufacturing and Mexico: A New International Division of Labor?" Latin American Research Review 29: 39-72.

Simmel, Georg. [1908] 1964. "The Stranger." In The Sociology of Georg Simmel, edited and translated by K. H. Wolff, 402-8. New York: Free Press.

Skocpol, Theda. 1979. States and Social Revolutions, A Comparative Analysis of France, Russia, and China. Cambridge: Cambridge University Press.

Swidler, Ann. 1986. "Culture in Action: Symbols and Strategies." American Sociological Review 51: 273-86.

Thelen, Kathleen. 2004. How Institutions Evolve: The Political Economy of Skills in Germany, Britain, the United States, and Japan. New York: Cambridge University Press.

Tilly, Charles. 1984. The Contentious French: Four Centuries of Popular Struggle. Cambridge, MA: Harvard University Press.

Veblen, Thorstein. [1899] 1998. The Theory of the Leisure Class. Amherst, NY: Prometheus Books.

Wallerstein, Immanuel. 1974. The Modern World-System: Capitalist Agriculture and the Origins of the European World-Economy in the Sixteenth Century. New York: Academic Press.

---. 1991. Geopolitics and Geoculture: Essays on the Changing World-System. Cambridge: Cambridge University Press.

Weber, Max. [1904] 1949. The Methodology of the Social Sciences. Translated by E. A. Shils and H. A. Finch. New York: Free Press.

---. [1915] 1958. "Religious Rejections of the World and Their Directions." In From Max Weber: Essays in Sociology, edited by H. H. Gerth and C. Wright Mills, 323-59. New York: Oxford University Press.

---. [1922] 1947. "Social Stratification and Class Structure." In The Theory of Social and Economic Organization, edited by T. Parsons, 424-29. New York: Free Press.

---. [1922] 1964. The Sociology of Religion. Translated by E. Fischoff. Boston: Beacon Press.

Williamson, Oliver. 1975. Markets and Hierarchies. New York: Free Press.

---. 1985. The Economic Institutions of Capitalism. New York: Free Press.

Wright, Erik O. 1980. "Varieties of Marxist Conceptions of Class Structure." Politics and Society 9: 299-322.

---. 1985. Classes. London: Verso.

Wright, Erik O., and Luca Perrone. 1976. "Marxist Class Categories and Income Inequality." American Sociological Review 42: 32-55.

Wuthnow, Robert. 1987. Meaning and Moral Order: Explorations in Cultural Analysis. Berkeley: University of California Press.

---. 1998. After Heaven: Spirituality in America since the 1950s. Berkeley: University of California Press.

Zelizer, Viviana. 2005. "Culture and Consumption." In The Handbook of Economic Sociology, 2nd ed., edited by N. J. Smelser and R. Swedberg, 331-54. Princeton: Princeton University Press and Russell Sage Foundation.