Veto Players, Globalization and the Redistributive Capacity of the State: A Panel Study of 15 OECD Countries

Globalization is said to restrict severely the state’s capacity to fulfill its welfare function in advanced industrial societies. This paper tests empirically the redistributive capacity of the state operationalized as the difference in percent of households who live below 50% of the median income in their respective country before taxes and transfers and after taxes and transfers, based on the latest Luxembourg Income Study data. Besides globalization, specific sets of domestic political institutions predictably and systematically affect the redistributive capacity of the state: what is termed ‘collective veto points’ buoy redistribution by the state, while ‘competitive veto points’ have the opposite effect. Partisan coloration is introduced as a control variable which, surprisingly, does not affect this critical function of the welfare state. The research design is a cross-sectionally dominant panel design (N=15, t=2). This study finds evidence that globalization and collective veto points both positively affect the redistributive capacity of the state while the converse is true for competitive veto points. ‘There is the alarming prospect that globalization will eventually emasculate democratic choice’ (Gøsta Esping-Andersen, 1996). ‘. . . consensus democracy may be considered more democratic than majoritarian democracy . . .’ (Arend Lijphart, 1999). How do the two forces of globalization and domestic political institutions affect the redistributive capacity of the state? Globalization, so the argument goes, is increasingly circumscribing the autonomy of the modern state, leading to what some observers call ‘neo-liberal convergence’. However, domestic political institutions mediate and filter these pressures of globalization, creating nation-specific, idiosyncratic outcomes despite globalization. The purpose of this article, is to examine how these two powerful forces, globalization and domestic political Markus M.L. Crepaz 2 institutions, affect a central tool of the modern state: its ability to redistribute incomes and thus, the degree to which citizens are capable of taking part in the political process. The definition of ‘redistributive capacity of the state’ and the explanandum of this article then, is the percent difference between pre-tax/ pre-transfer income and post-tax/post-transfer income of households whose disposable income is less than 50 percent of the median income. Thus, this article examines the change in percent between the number of households who make less than 50% of the median income before taxes and transfers and the number of households who make less than 50% of the median income after taxes and transfers. ‘Redistributive capacity of the state’ thus captures the varying capacity of states by means of taxation and transfer payments to affect the number of households who live in poverty (defined as the percentage of households who live below 50% of the median income of the respective country). This study uses a panel design of fifteen countries (Australia, Belgium, Canada, Denmark, Finland, France, Germany, Italy, Luxembourg, Netherlands, Norway, Spain, Sweden, United Kingdom, and the United States) at two different time points: the first of which occurs roughly between the late 1970s and the middle 1980s and the second roughly between the late 1980s and the middle 1990s. Since the early 1990s a veritable cottage industry has sprung up arguing the decline of national sovereignty as a result of the homogenizing forces of international capital (Strange, 1995; Schmidt 1995; Kurzer 1993, Cerny 1994, 1995; Cable 1995; Rodrik 1997, Ohmae 1995, and others). According to this view, cross border flows of trade, foreign direct investments, people, and particularly finance capital, have increased to such magnitudes that the nation state is said to lose its ability to meaningfully respond to citizen’s demands through the crafting of policies. Globalization, so the argument goes, is the death knell of the state since its Diktat will leave no room for any policies not designed to maximize economic efficiency, such as welfare policies, national styles of regulation, industrial policies, collective bargaining schemes, or nation specific workplace arrangements. There is no gainsaying that globalization is occurring – by now, most observers are familiar with the various empirical manifestations of globalization such as dramatic increases in trade, improvements in transportation technology, large migration flows, dramatic drops in capital controls and the corresponding increase in capital mobility, quantum leaps in communication and transportation technology, and last but not least, the indefatigable work of international organizations such as the World Trade Organization dedicated to eradicate tariff and non-tariff barriers to international trade. Veto Players, Globalization, Redistributive Capacity of the State 3 However, as these forces of globalization reach the national shores, they do not have identical consequences in every country. This article contends that whatever happens in the international arena will be filtered and refracted through domestic political institutions – and since they are vastly different, it is only prudent not to expect identical outcomes. This is not to say that ‘institutions’ are the only thing that differs between nations. Obviously, culture, historical experiences, structural constraints, and a multitude of other factors impinge on the consequences of globalization and are dealt with in idiosyncratic ways, leading to idiosyncratic outcomes. Thus, as of this moment, we are far away from ‘neo-liberal convergence’ or, from a ‘MacDonaldization’ of the world. As long as governments’ survival depends on the support of geographically determined constituencies, we should heed Tip O’Neill’s insight that ‘All Politics is local’, although, ‘local’ may not mean narrowly circumscribed constituencies, but rather regions, or even, the nation state itself. This article then, examines to what extent these two factors, globalization and constitutional features, affect the redistributive capacity of the state. The argument is rather straightforward: if indeed globalization has the hypothesized effects, we should expect less redistributive capacity in open economies; on the other hand, examining Lijphart’s two dimensions of democracy (1984, 1999) we should expect to find a higher redistributive capacity of the state driven by an ‘enabling’ institutional structure based on collective veto points, while we should observe a negative effect of competitive veto points on the redistributive capacity of the state, since they tend to ‘restrain’ government. In addition to the external influence argument it will be argued that formal domestic institutional structures such as proportional representation, and ‘informal’ institutional structures such as executive powersharing in broad multi-party coalitions, executive-legislative balance of power, multi-party systems, and corporatist systems of interest representation tend to expand state activity, while restraining institutions, i.e. separate institutions with mutual vetoes, such as federalism, equally strong legislatures, rigid constitutions, the presence of judicial review, and more independent central banks should restrict state activity. The purpose of this study is to examine whether differently structured constitutional elements, that is, formal constitutional rules such as the type of electoral system, but also less strictly defined features such as the effective number of parties, or the degree to which labor and capital coordinate public policy making, have systematic and predictable effects on the redistributive capacity of the state, while at the same time, controlling for the effect of globalization on the capacity of Markus M.L. Crepaz 4 the state to redistribute incomes. Thus, this study adds to the ‘veto points’ literature (Tsebelis, 1999; Colomer, 1995; Schmidt, 1996; Huber et al. 1993; Hallerberg and Basinger, 1998) by arguing that simply counting the number of veto points is not enough; that instead, one must examine the particular nature of the veto points to understand their effects. Political power is diffused in different ways: if institutions are structured such that political actors interact with each other based on separate agencies with mutual veto powers, we would expect difficulties in changing from the status quo. However, if political actors interact with each other on a face to face basis, such as in multi-party systems, or in institutions which share governmental responsibility, there will be a tendency to expand state activity based on logrolling behavior. Thus, not all veto points are created equal. What needs to be examined is the nature of the structure of veto players and their interrelationships. This study will also add to the growing literature on globalization with the surprising result that ‘globalization’ is positively related to the redistributive capacity of the state. The central point of this article is to analyze both forces simultaneously: the global as well as the institutional one. This is quite in keeping with Gourevitch’s (1978, 911) assertion that ‘International relations and domestic politics are . . . so interrelated that they should be analyzed simultaneously, as wholes’. What follows is a more detailed deductive analysis as to why these things should occur as stated. Then, the data and the research design are introduced. The conclusion will offer an assessment of the empirical analysis and the connections and interactions between the forces of globalization, political institutions, and the redistributive capacity of the state. The Forces of Convergence According to apostles of globalization, the state is withering away, but not in the manner predicted by Marx. Rather, the state has been likened to ‘. . . an old tree, still sprouting new leaves and branches and apparently still alive, but actually hollow in the middle, inactive and ineffectual in the really basic matters of security and money for which it was designed’ (Strange, 1995, 304). Others argue that exposure to the international economy will hollow out the state, destroy idiosyncratic state-society relationships such as social democratic corporatism, where ‘deliberative democracy may suffer’ (Schmidt, 1995, 77), leading to ‘erosion of state capacity’ (Cerny 1994, 334); in short, leading to a situation where the state is becoming ‘defective’ (Strange, 1995, 55). Extending this line of thought to the fate of democracy itself, Claude Ake (1997, 285) asserts that the ‘. . . process of globalization attacks Veto Players, Globalization, Redistributive Capacity of the State 5 democracy in a manner that is uniquely different and extremely threatening . . . It [globalization] is simply rendering democracy irrelevant and in this it poses the most serious threat yet in the history of democracy’. And under the rubric of ‘demise of the nation state’ Ake also (1997, 287) argues that ‘. . . the transnationalization of more and more things, especially economic activities, decisions which affect people’s lives and shape public policies decisively are made in distant places, often anonymously, by agents and forces we can hardly understand, much less control’. Not that long ago we were told to ‘bring the state back in’ (Skocpol, 1985), and now we are supposed to throw it back out already? On the other hand, there are those who are less sanguine about the debilitating effects of globalization on the sovereignty of nations. Two excellent edited volumes by Berger and Dore (1996) and Boyer and Drache (1996) argue that the claims of the subversive impact of globalization on the autonomy of the state are exaggerated and premature. Hirst and Thompson (1996) make a similar argument. Also, Duane Swank (1998) finds that the theory of ‘diminished democracy’ as a result of globalization is overstated, as there is no empirical connection between the structural power of international capital and its pressure on domestic policy makers to reduce taxation of capital. For an observer of corporatism, the debate about globalization and its impact on the national political economy, must at this point of our debate, become rather puzzling. Was it not precisely the process of openness itself that has historically created the nation specific forms of corporatist interest intermediation? For comparativists, the impact of the international system on domestic politics has always played a major part in their scholarship (Gerschenkron 1963, Hintze, 1975, Gourevitch 1978, Cameron 1978, Rogowski 1987, Almond 1989, and others). Most of this literature argues that as a result of exposure to international business cycles, nations developed specific ways to deal with these challenges. Practically all of these nation-specific corporatist arrangements were fostered either during the Great Depression or during WWII. Consequently, international challenges did not weaken the respective states; if anything, they have learned to dynamically adjust to these international challenges in such a way as to ensure economic efficiency with a minimum of social dislocation. Maybe some of us have forgotten the fable of the snake, the frog and the owl, Peter Katzenstein (1985, 200) tells us at the end of his book entitled Small States in World Markets. Its gist is that: ‘The small European states adapt domestically to economic change imposed by an international economy that they cannot hope to control’. It is true that today capital is more mobile than ever before; capital Markus M.L. Crepaz 6 controls have been virtually abolished, and the state, so the argument goes, will find it increasingly difficult to tax capital (but, Swank 1998). If capital perceives to get taxed ‘excessively’, it will become ‘footloose’ and flee to ‘safer’ environs. Similarly, high skill, mobile labor will move to wherever opportunity knocks harder, while low skill labor will stay put. If the argument of the traditional ‘corporatists’ above is correct, i.e. that corporatism and the welfare state are a cushion to soften the blow of globalization, then the state faces a cruel choice; it either continues the welfare state and risks falling behind in competitiveness, efficiency and productivity, or it opens up to the global economy and becoming more competitive, productive and efficient. The downside of that policy, however, is higher exposure to the vagaries of the market, rising inequality, increasing social dislocation, and breakdown of social cohesion and community. New labels of ‘outsiders’ and ‘losers’ are created all over Europe. The Germans speak of a ‘two thirds society’; the Danes speak of the ‘A-team’ (the winners), the ‘B-team’ (the losers), and also of a ‘C-team’ (the thoroughly marginalized). France has become a ‘societé à deux vitesses’, a society of two speeds. The British have their ‘underclass’. However, according to Esping-Andersen, these marginalized groups do not make up a coherent and homogenous stratum (Esping-Andersen, 1999). However, there is little empirical evidence that the welfare state itself, or its institutions, such as social democratic labor market institutions and corporatism, have been reduced or undermined. Stephens, Huber, and Ray (1999, 189) find that over time, ‘. . . only relatively moderate [welfare state] rollbacks . . .’ have occurred between 1945 and 1989. Similarly, a close examination of union density, union coverage, union monopoly and concentration, and authority in unions and employer’s organizations for the period from 1945 to 1989 finds that ‘. . . industrial relations institutions and trade unions have by and large proved quite resilient in the face of considerable domestic and international economic pressures in the past two decades’ (Golden, Wallerstein, Lange, 1999, 223). Thus, it does not appear that over time, social democratic labor market institutions have significantly been dismantled. A panel analysis below will shed light on the performance of these institutions with regard to their ability to redistribute incomes. The debate above, thus, has isolated a central puzzle: does globalization undermine or increase the state’s capacity to redistribute incomes to such an extent to maintain social cohesion, equality, and prevent social dislocations as a consequence of globalization? According to Dani Rodrik, ‘. . . globalization presents this dilemma: it results in increased demands on the state to provide social insurance while reducing the ability of the state to perform that role effectively’ (Rodrik, 1997, 53). Veto Players, Globalization, Redistributive Capacity of the State 7 To check this claim will be precisely the task in the empirical section below, as the impact of economic openness on the redistributive capacity of the state will be explored. But before we can move on to that, another factor weighing in on the redistributive capacity of the state has to be examined – the role of domestic political institutions. This is the task to which we turn now. Forces of divergence There is a second major contender which could assist in explaining the redistributive capacity of the state: that contender is political institutions. Using political institutions as an additional predictor, we rely on a fundamental assumption: the pressures of internationalization are filtered, mediated and refracted through nation-specific political institutions. Since these institutions are different, we should expect different impacts of internationalization, even if the international forces, which impinge on the states, are of the same kind and intensity. The most appropriate model specification for such moderating effects is an interaction model which will be discussed below. This is not to say that the only forces explaining policy choices in advanced industrial democracies are political institutions and globalization. Besides institutions, ‘cultures’ differ. Gourevitch (1996, 258) is very doubtful for neo-liberal convergence to take place given that: ‘Cultural traditions are powerful; countries have different traditions in networking, personal relationships, conceptions of authority, models of organizations and individuals’. There are also nation specific differences in ‘production cultures’. Streeck (1996) finds that lean production principles based on broad rather than specialized skills and company specific skills rather than portable skills between companies is consistent with Japanese culture, but would not transfer well to German auto plants because of the different cultural occupational ethos of the German workforce. Thus, this article does not deny these variations in ‘culture’; however, given the small number of countries examined and the space limitations, the emphasis in this article will be placed on globalization and domestic political institutions. There is a myriad of evidence indicating that political institutions such as parties, forms of interest group representation or constitutional features, have predictable effects on a variety of policy fields, such as macroeconomic outcomes (Kirschen, 1964; Hibbs, 1977; Castles, 1982; Schmidt, 1982; Olson, 1982; Lijphart 1999, and others), income inequality (Cameron, 1988; Mueller, 1988, 89; Lijphart, 1999), strikes (Korpi and Shalev, 1979; Humphries, 1990; Birchfield and Crepaz, 1998; Lijphart 1999), environmental outcomes (Enloe, 1975; Jänicke, Markus M.L. Crepaz 8 1990; Lijphart, 1999, and others), voter turnout (Powell, 1986, Jackman, 1987; Lijphart, 1999, and others), welfare effort (Wilensky, 1975; Huber, Ragin, and Stephens, 1993, Lijphart, 1999), and a host of other