نابرابری درآمد و اقتصاد سیاسی محلی در ایالات متحده، 1970-2000

Previous studies of rising inequality in the United States have overlooked the potential role of subnational political economic variation as an institution that shapes earnings restructuring. This paper uses hierarchical linear models to examine how state right-to-work laws contribute to growth in inequality in 80 metropolitan labor markets from 1970 to 2000. Contrary to conventional expectations, labor markets in states with right-to-work laws experience relatively mild growth in earnings inequality, and are less unequal by 2000 than non-RTW labor markets. The trend cannot be fully explained by union density, job growth, uneven development or variation in racial inequality. The findings contribute to a distinctly sociological perspective on rising inequality that considers how social, institutional and economic factors interact at the local and state levels to shape earnings.

Scholars of comparative stratification often point to the importance of employment regimes for explaining national differences in earnings1 and inequality (i.e., Freeman, 1996 and Kalleberg, 1988), but they pay less attention to the consequences of political economic variation within nations. Since at least the 1940s, states and localities in the United States have attempted to fashion distinctive business climates to attract investment and create jobs (Cobb, 1993). Local economic differentiation accelerated in the 1970s and 1980s in response to the devolution of responsibility for economic development to the states (Eisinger, 1988 and Jenkins et al., 2006). This paper asks whether state business climates shaped the path of inequality growth in metropolitan labor markets from 1970 to 2000. I focus on one of the oldest and most enduring elements of state business climates, right-to-work (RTW) laws. Right-to-work laws limit union security, or the requirement that workers covered by collective bargaining agreements pay union dues (Baird, 1998). Right-to-work laws are an important feature of American political economy because they shape labor relations, promote a competitive model of economic development, and are meaningful in the business community as signals of a neoliberal orientation to growth (Cobb, 1993). Using multi-level growth curve analysis I find that metropolitan labor markets in RTW states are more unequal than non-RTW labor markets in 1970, but RTW labor markets experience milder inequality growth over the 30-year period of earnings restructuring. I use the existing literature on RTW laws to derive hypotheses of why metropolitan labor markets’ levels of inequality in 1970, and growth in inequality from 1970 to 2000, vary across RTW contexts. The literature suggests that RTW laws may shape earnings inequality directly via effects on union density and employment growth, and indirectly due to the laws’ correlation with levels of economic development and racial inequality. These conventional expectations explain the greater levels of inequality observed in RTW labor markets in 1970, but they are insufficient for understanding variation in inequality growth across RTW and non-RTW contexts from 1970 to 2000. The robust negative relationship between RTW laws and inequality growth is puzzling from the perspective of comparative research that associates higher levels of inequality and sharper inequality growth with weak labor market institutions (e.g., DiPrete et al., 2006, Kenworthy and Pontusson, 2005 and Pontusson et al., 2002). I theorize that the relationship between labor market institutions such as unions and inequality is curvilinear, with the greatest levels of inequality found in systems with labor market institutions that are stronger than those in adjacent areas, but too weak to prevent capital flight and institutional weakening (see Calmfors & Driffill, 1988). From this perspective, capital mobility and related elements of competitive development decrease the bargaining power of unions net of union density (Moore, 1998) and increase employment insecurity (Bourdieu, 2003 and Bronfenbrenner, 2000), therefore altering the dynamics of earnings determination within firms and producing sharp inequality growth in non-RTW labor markets. By considering the relationship between local earnings restructuring and subnational political economy, and conceptualizing earnings determination as a multi-level phenomenon, this paper offers a fresh perspective on one of the most socially significant but sociologically understudied recent trends in the United States, the growth of economic inequality (Myles & Myers, 2007).

This paper identifies distinctive paths of earnings restructuring for local labor markets in the United States based on a subnational political economy that has been overlooked in previous studies of earnings and inequality. Labor markets in RTW states were more unequal than their non-RTW counterparts in 1970, largely due to differences in union density, but RTW labor markets experienced milder growth in inequality from 1970 to 2000. Earnings inequality was greater in non-RTW labor markets by 2000 than in RTW areas. This trend is largely unaffected by either the theorized direct mechanisms—union density and employment growth—or the alternative hypotheses, economic development and racial inequality. One question that emerges from this analysis is whether the shifting geography of high-wage work contributes to the remaining unexplained variation in inequality growth across RTW and non-RTW contexts. High-wage earnings growth drives national trends in rising inequality since the 1970s (Piketty & Saez, 2003) and high-wage workers have become more geographically concentrated during this period (Berry & Glaeser, 2005). But neither high-wage earnings levels nor the rate of growth in high-wage earnings vary across right-to-work contexts (see Appendix D). The large and growing concentration of high-wage work in cities like Miami, FL, Austin, TX and the North Carolina Research Triangle illustrates the point that that state right-to-work laws are not predictive of high-wage earnings concentration, which follows a different geographic political economy that is beyond the scope of this paper. The inequality trends examined in this paper are driven by the shifting dynamics of low-wage earnings growth across RTW and non-RTW states. The negative relationship observed between state RTW laws and local inequality growth in the United States is surprising from the perspective of research on inequality and institutions. Cross-national and comparative studies of earnings trends (e.g., Kenworthy and Pontusson, 2005 and Volscho and Fullerton, 2005) have consistently affirmed that labor market institutions such as unions temper the inequality-producing effects of the market, but efforts to connect the decline of labor market institutions to rising inequality in the United States have found mixed results (DiPrete et al., 2006 and Morris and Western, 1999).12 This study finds that metropolitan labor markets in weak-institution RTW states experience milder inequality growth from 1970 to 2000 than their strong-institution counterparts, and are less unequal by 2000. One possible interpretation of these results is the neoclassical economic view that labor market institutions such as unions heighten inequality because they stifle economic growth (e.g., Friedman, 1962). Proponents of measures such as RTW laws would recognize the slower inequality growth observed in those states from 1970 to 2000 as the intended consequence of those self-consciously pro-market, anti-union strategies. Some would go further, advocating right-to-work laws and similar measures that limit the strength of unions as a strategy for reducing inequality and promoting earnings growth. I do not interpret the results reported here as support for the neoclassical economic perspective on institutions and inequality for three reasons. First, neoclassical economics cannot account for the situation observed through 1970, when lower levels of inequality and higher earnings were observed in states where labor market institutions strengthened workers’ interests. Second, the inequality-reducing earnings growth associated with weak labor market institutions from 1970 to 2000 is focused on the lowest fifth of the earnings distribution, with no net effect on median or high-wage earnings growth. If institutions truly stifle economic growth we would expect higher wages throughout the earnings distribution in labor markets where institutions favor the interests of capital. Finally, the slower inequality growth associated with weak labor market institutions that I have identified follows a quadratic function, suggesting diminishing benefits to neoliberal economic governance in the future. Rather than challenging the conventional sociological wisdom on institutions and inequality, this analysis refines institutional analyses of rising inequality by placing them in a larger context of spatial inequalities and multi-level processes. The standard narrative in the inequality literature of (relatively) strong institutions weakening in the face of globalization and deindustrialization never accurately described labor markets in the American South and Southwest, where RTW laws limit the growth of labor unions. Sharp employment growth in RTW labor markets, which itself explains some of the negative RTW-inequality growth effect, may reduce remaining unions’ power to influence earnings at the bargaining table. The threat and reality of industrial relocation may increase employment insecurity and thus reshape power dynamics within firms, particularly in the non-RTW states that have suffered the most extreme industrial job losses (Bronfenbrenner, 2000). We might in this way expect the competitive development dynamics initiated by RTW laws to facilitate the erosion of workers’ rents across industries and occupations. Bringing spatial inequality and multi-level processes to the study of inequality is also suggestive of the role of organizational change. Sharper inequality growth in non-RTW labor markets may have developed as firms leveraged spatial inequalities to implement shareholder value strategies to reduce labor costs by reorganizing production. Massey writes that “[r]elations between economic activity in different parts of the country are now a function rather less of market relations between firms and rather more of planned relations within them” (1984: 297). Building on Fligstein and Shin (2007) and Sørensen (2000), firm-level strategies such as mergers, layoffs and outsourcing may have targeted the proportionately larger rents captured by workers in non-RTW labor markets by moving operations to RTW states and, increasingly, overseas. From a cross-national perspective the U.S. labor market is characterized not only by weak labor market institutions, but by internal political economic variation, both of which are important for understanding rising economic inequality. The trends identified here are consistent with research on globalization showing that dynamics of capital mobility accord short-term benefits to weak-institution labor markets at the expense of adjacent strong-institution labor markets in the form of rising wages and slower inequality growth (e.g., Hansen, 2006 and Rodrik, 1997). The implication is that labor market institutions do not shape inequality from within a closed system. The relationship between institutions and inequality within a particular labor market is determined as much by the institutional structure of competing labor markets as by that labor market's own properties and institutional propensities for reducing inequality. The potential for heightened inequality may be particularly acute in labor markets with institutions that are stronger than those in adjacent areas, but too weak to prevent painful employment restructuring and capital flight. The relationship between institutions and inequality may therefore be curvilinear, with the greatest wage penalties and levels of inequality found in intermediate systems (Calmfors & Driffill, 1988). Adopting a political economic perspective to examine within- and between-place inequality dynamics in the United States thus refines the conventional sociological wisdom on labor market institutions by placing it in the larger context of social, institutional and organizational dynamics that operate at the local and state levels. Earnings determination is an inherently multi-level phenomenon, but too few studies have engaged simultaneously with the multiple layers of social structure that shape economic activity. By considering variation in earnings restructuring across local contexts, this analysis ultimately reaffirms the importance of politics for understanding labor market outcomes such as economic inequality. Global macroeconomic shifts and financial advisors’ recommendations were the stimuli for firm-level earnings restructuring starting in the 1970s (Fligstein, 2001), but in the United States these pressures unfolded in a context of internal political economic variation and significant pre-existing spatial inequalities. Decentralized economic governance—the states’ authority to fashion distinctive political economies—may facilitate rising inequality nationally by promoting competitive development dynamics that encourage firm-level restructuring strategies that leverage spatial inequalities to reorganize production, weaken workers’ rents in strong-institution areas and increase employment insecurity. Detailed case studies of particular firms and labor markets are necessary to fully understand the between-place inequality dynamics identified here and to evaluate the theory suggested by this statistical analysis. But this paper contributes to the development of a distinctly sociological perspective on rising inequality by identifying a subnational political economy that shapes earnings restructuring in the United States, and by theorizing the interrelationship between social, institutional and organizational factors that shape earnings at multiple levels of social structure.