Postindustrial Indicators, Human Development, and Red–Purple–Blue States

Abstract

This chapter asks: What factors explain the close relationships between human development (HD) and RPB, and which of these two variables is more prior than the other? To answer this two-part question, this chapter probes the extent to which indicators of postindustrial economies are relevant explanatory factors. States scoring high on indicators of postindustrial economies emphasize access to telecommunications and the internet, the production and application of knowledge, and the growth of services rather than manufacturing or the extraction of raw materials that may be polluting. Applying partial correlation analyses, this chapter finds that these postindustrial indicators tend to explain the correlation between states with higher human development and blue political color (RPB); both are relevant to practical voting. Moreover, applying structural equation models (SEMs), this chapter finds that higher HD influences RPB more than RPB influences HD. By clarifying some of the tangible characteristics of a state that are associated with the RPB typology, this chapter provides a substantive interpretation of this classification and clarifies how HD may limit its effects.

The value-making locales offer greater opportunities and better services, and this offer attracts talented and innovative professionals. And because there is money, there is a thriving market and there are better cultural amenities, educational facilities, and health services, and therefore jobs which are still the main source of urban growth. Since jobs are appealing globally, these metropolitan regions also become the hubs for immigration. They develop as multi-ethnic places…

—Manuel Castells (2010: xxxvii)

Notes

• 1. Gelman et al. (2008, 175).

• 2. These indicators of postindustrial economies are broadly consistent with the views of Bell ([1973] 1999), Brint (1994), Castells (2010), Kalleberg (2011), and Moretti (2012). There is considerable conceptual overlap among the views of these authors but there are some differences: Castells emphasizes the flows of information through the internet; Bell, Kalleberg, and Moretti emphasize the importance of theoretically based knowledge and cognitive skills; and Bell and Brint emphasize the growth of services and the services sector.

• 3. For these macrolevel data go to www.measureofamerica.org/maps. Click on “download data” along the top for any available geographical breakdown you want. The data were accessed and the initial merged SAS file was created on January 10, 2012.

• 4. Haq (1995, 13–28) and Sen (1999, 13–24) define human development as the enrichment of the choices people have for leading a decent, secure life.

• 5. Berlin (1969, 122) defines negative freedom as implying the absence of external constraints on agency: “I am normally said to be free to the degree to which no man or body of men interferes with my activity.” A person with minimum levels of health, education, and income has severe constraints on his power to exercise choice and agency. He defines positive freedom as (1969, 131): “the wish on the part of the individual to be his own master. I wish my life and decisions to depend on myself, not on external forces of whatever kind.” Positive freedom implies human agency. Handlin and Handlin (1961, 9–22) also distinguish the absence of constraint (negative freedom ) from personal power (positive freedom ).

• 6. These definitions are those of Lewis and Burd-Sharps (2010, 272–281), who modified the human development measures of the United Nations Development Program (2010) for use in the U.S. These definitions differ from those of Porter (2012, 1364–1365) who analyzed county-level data on the southern states of the U.S.

• 7. Moretti (2012) analyzes the separate effects and interrelationships of education, economic well-being, and health in the U.S. He concludes that human capital based on educational achievement is pivotal for enhancing innovation and job creation (2012, 215–249).

• 8. The sources are en.wikipedia .org/wiki/Red_states_and_blue_states, 1992–2008 (Accessed 28 October 2011) and Population Reference Bureau , “Swing, bellwether, and red and blue States, 1998–2004” by Kevin Pollard (2004) (Accessed 28 October 2011). Earlier, Table 3.2 defined and Table 3.3 validated the preferred trichotomous classification of states as red, purple, or blue.

• 9. Lazarsfeld and Menzel (1972, 230–231) refer to such properties as contextual because each person is described by a property of his or her collective; namely, a person’s state of residence. These correlations are between two contextual properties of the individuals and are interpreted as correlations between two properties of the states.

• 10. The reliabilities would be with health deleted, alpha = 0.892; with education deleted, alpha = 0.739; and with income deleted, alpha = 0.777.

• 11. Lewis and Burd-Sharps (2010 provide this indicator of public assistance and the other data from the American Human Development Project.

• 12. See Lewis and Burd-Sharps (2010, 34–35) for their dashboard of risks.

• 13. These findings corroborate and extend Gelman et al. (2008) by linking the political color of a state explicitly to its human development and to the risks to human development.

• 14. The rates of Tea Party membership circa spring 2011 are from Skocpol and Williamson (2012, 91) and their personal communications about Tea Parties in Washington D.C., Hawaii, and Alaska. Their more refined rates have been dichotomized here as less than two Tea Party groups per million people (1) versus two or more Tea Party groups per million people (0).

• 15. As in Table 5.1 these macrocorrelations are based on two contextual properties of the 2000 voters and are interpreted as correlations between two properties of the states.

• 16. The correlations of the individual-level survey responses of the voters with the macrolevel classifications of states as BPR or according to their human development rank are based on contextual properties of the voters (their positions on the macrolevel variables) and on absolute properties of the voters taken from the survey. Lazarsfeld and Menzel ’s (1972, 229) definition is “Absolute properties are characteristics of members which are obtained without making any use either of information about the characteristics of the collective, or of information about the relationships of the member being described to other members.” These correlations are best viewed as fully standardized bivariate regression coefficients of ranks that capture the asymmetric effect of the macrolevel typologies on the absolute properties of the individual voters. For example, the individual has a value on the absolute property y _i  = “attitude toward Bush’s job performance is poor.” That individual also has a value on an ordinal contextual property x _i  = “living in a state with a certain political coloration [R = 1, P = 2, or B = 3].” The Spearman correlation of +0.112 across the 2000 individual voters quantifies the fully standardized effect of the ordinal political coloration variable (blue rather than red) on the voters’ attitude that Bush’s job performance is poor.

• 17. See Smith (1996, 1997, 1999, 2003) for previous research on three core political interests.

• 18. Other previous supporters of health care reform included Presidents Truman and Nixon .

• 19. Lewis and Burd-Sharps (2010, 235, 237, 243, 245) provide these raw measures, which the author has interpreted based on his review of the literature on societies with postindustrial economies (Smith 2008, 217–222), and these other sources: Bell ([1973] 1999); Brint (1994); Castells (2010); Kalleberg (2011); and Moretti (2012).

• 20. All of the bivariate correlations in Table 5.3 are based on two contextual properties of the 2000 individual voters and are interpreted as correlations between two properties of the states.

• 21. Lewis and Burd-Sharps (2010, 256) provide these social inequality indicators. Stepan and Linz (2011, 843) report that U.S. disenfranchisement occurs at a rate 60 times higher than the average of their comparison set of 22 advanced democracies.

• 22. Lipset (1981, 509–510) notes that “Post-industrial politics is increasingly concerned with noneconomic or social issues —a clean environment, a better culture, equal status for women and minorities, the quality of education, international relations, greater democratization, and a more permissive morality, particularly as affecting familial and sexual issues.” Lipset’s initial interpretation is corroborated and advanced by Clark (1994), Clark and Hoffmann-Martinot (1998), Clark and Lipset (2001), and Clark (2014). New social problems of postindustrial society that tend not to appear as issues in election campaigns are high incarceration rates of African American men (Manza and Uggens 2006; Western 2004); the imposition of the death penalty (Garland et al. 2011); and the privatization of prisons in the U.S. and its consequences (Pattillo et al. 2004).

• 23. The burning of limestone to 1450 °C to produce cement is responsible for at least 5% of man-made carbon dioxide (CO₂) emissions worldwide. About 60% of these emissions attributable to cement are due to chemical reactions when the limestone is burnt; heating the kiln to that temperature with fossil fuels is responsible for the remaining 40%. See Paul Miles (2012).

• 24. For studies of the impacts of the great recession see Grusky et al. (2011).

• 25. The different component indexes of human development have slightly different consequences; comparisons of the correlations underscore the importance of education and health for favorable outcomes. Higher scores on the education index have the strongest favorable correlations with the postindustrial and children’s health clusters. Higher scores on the health index have the strongest favorable correlations with the pollution and housing and homelessness clusters. Higher scores on the economic index have mixed results.

• 26. The four intervals of this human development typology are (1) lowest human development, 3.80961–4.67980; (2) 4.67981–5.07048; (3) 5.07049–0.555520; and highest human development, 5.55521–6.30347.

• 27. The realignment of the South from the Democrat party to the Republican party in large part due to the Democrat’s civil rights legislation changed the party choices of southern states so that their voting choices are now more consistent with their prior lower levels of human development and discontent about enhancing minority rights.

• 28. If b _yx is the unstandardized path-regression coefficient for any variables x and y, then the standardized path coefficient is b _yx (s _x  / s _y ).

• 29. When energy consumption (in BTUs per capita) is deleted from Model 1 leaving human development rank as a quasi-instrument, then the positive effects of practicing physicians (per 10,000 population) are almost statistically significant, the p = 0.053.

• 30. Kline (2005, 137) suggests that values of $\chi^{2}$/df ranging from 2 to as high as 5 may indicate reasonable fit of a SEM.

• 31. Jill Tao of the SAS Institute suggested using the Mod option to improve a model’s fit.

• 32. Kline (2005, 133–145) provides explications of these goodness-of-fit measures.