Skip to main content
SpringerLink
Log in

Race, Interstate Migration, and Employment Discrimination Law

  • Chapter
  • First Online:
Societal Agents in Law

  • 131 Accesses

Abstract

During World War I, migration by Blacks from the South to other regions of the United States increased substantially, and in the decades that followed, this migration remained above its prewar level. This chapter contends that the enlarged migration of Blacks to states outside the South created the foundation for law that prohibited race discrimination in employment. From 1945 until the enactment of the federal Civil Rights Act of 1964, 23 states in the continental United States approved legislation that (1) banned job-related race discrimination and (2) provided a mechanism to enforce the ban. An analysis of state-level data with logistic regression supported the hypothesis of the study: Net of other independent variables, the odds that a state adopted an enforceable statute prohibiting race discrimination in employment rose as the number of Black male migrants into a state between 1910 and 1940 became larger relative to the number of White male residents of the state. Furthermore, this migration had a greater impact on the odds than societal rationality (measured by the percentage of all state residents aged 25+ who as of 1940 had completed four or more years of college), an increase which also raised the odds. Finally, quantitative research on the degree to which such state legislation benefited Blacks economically is examined.

This is a preview of subscription content, log in via an institution to check access.

Access this chapter

Log in via an institution
eBook
USD 16.99
Price excludes VAT (USA)
  • Available as EPUB and PDF
  • Read on any device
  • Instant download
  • Own it forever
Hardcover Book
USD 54.99
Price excludes VAT (USA)
  • Durable hardcover edition
  • Dispatched in 3 to 5 business days
  • Free shipping worldwide - see info

Tax calculation will be finalised at checkout

Purchases are for personal use only

Institutional subscriptions

Notes

  1. 1.

    Steven A. Ramirez, The Law and Macroeconomics of the New Deal at 70, 62 Md. L. Rev. 515, 522–25, 535–61 (2003).

  2. 2.

    In re Relafen Antitrust Litigation, 231 F.R.D. 52, 89 (D. Mass. 2005).

  3. 3.

    Larry D. Barnett, The Place of Law: The Role and Limits of Law in Society 268 (2011) [hereinafter The Place of Law].

  4. 4.

    Brown v. Bd. of Educ., 347 U.S. 483 (1954). In Brown, the Court held that “[s]eparate educational facilities are inherently unequal,” and hence impermissible under the Equal Protection Clause of the Fourteenth Amendment, when they stem from a government-required, race-based sorting of students. Id. at 495. The Equal Protection Clause instructs states that they are not to “deny to any person within [their] jurisdiction the equal protection of the laws.” U.S. Const. amend. XIV, § 1; text accompanying note 81 in Chap. 1 of the first volume.

    The Court has indicated that it considers Brown to be one of its two most significant decisions in the last half of the twentieth century. Planned Parenthood of Southeastern Pennsylvania v. Casey, 505 U.S. 833, 866–67 (1992). Scholars, too, assign Brown a high rank among U.S. Supreme Court cases. Martha Minow, In Brown’s Wake 6 (2010) (considering Brown to be “[t]he most famous decision of the U.S. Supreme Court”); J. Harvie Wilkinson III, From Brown to Bakke 6 (1979) (attributing the importance of Brown to the “the immensity of law it both created and overthrew”).

  5. 5.

    Under Title II of the Civil Rights Act of 1964, “public accommodations” are retail businesses that offer lodging, food, motor-vehicle fuel, or entertainment. Pub. L. No. 88–352, § 201(b), 78 Stat. 241, 243 (1964) (codified at 42 U.S.C. § 2000a(b) (2012)). Retail businesses that sell other types of items are not within the category of “public accommodations” and hence are not covered by the Act. Jeremy D. Bayless & Sophie F. Wang, Racism on Aisle Two: A Survey of Federal and State Anti-Discrimination Public Accommodation Laws, 2 Wm. & Mary Pol’y Rev. 288, 291–93 (2011).

  6. 6.

    Rebecca E. Zietlow, To Secure These Rights: Congress, Courts and the 1964 Civil Rights Act, 57 Rutgers L. Rev. 945, 957, 988–89 (2005).

  7. 7.

    In the present chapter, see: infra Table 5.2; infra note 44 and accompanying text.

  8. 8.

    The present chapter is limited to law that is embodied in race-focused statutes and court decisions. A federal entity that existed from 1941 to 1946 and that had the goal of combating race discrimination in employment is outside the scope of the chapter, because it was created by executive orders of the President. The entity, known as the Committee on Fair Employment Practice, lacked subpoena power and was unable to enforce its findings of discrimination through judicial proceedings; its sole tools for dealing with discrimination it unearthed were conciliation and publicity. Michael I. Sovern, Legal Restraints on Racial Discrimination in Employment 9–15 (1966).

  9. 9.

    Bell v. Maryland, 378 U.S. 226, 242, 284 app. V (1964) (Douglas, J., concurring); Paul Finkelman, Civil Rights in Historical Context: In Defense of Brown, 118 Harv. L. Rev. 973, 981–85 (2005) (book review).

  10. 10.

    Pub. L. No. 88–352, 78 Stat. 241 (1964) (codified as amended at 42 U.S.C. §§ 2000a to 2000 h (2012)).

  11. 11.

    Supra note 5.

  12. 12.

    William N. Eskridge & John Ferejohn, Super-Statutes, 50 Duke L.J. 1215, 1237 (2001).

  13. 13.

    Supra note 4 and accompanying text.

  14. 14.

    Infra notes 57 & 114.

  15. 15.

    U.S. Dep’t of Homeland Security, Yearbook of Immigration Statistics: 2013, at 5 tbl. 1, available at http://www.dhs.gov/yearbook-immigration-statistics (last visited July 5, 2018). Counts of persons reported in the Yearbook are by fiscal year. To simplify the discussion, I do not distinguish fiscal years from calendar years.

  16. 16.

    Id. at 5 tbl. 1.

  17. 17.

    Samuel R. Williamson, Jr., The Origins of World War I, 18 J. Interdisc. Hist. 795 (1988).

  18. 18.

    Calculated from U.S. Dep’t of Homeland Security, supra note 15, at 5 tbl. 1.

  19. 19.

    Calculated from id. at 5 tbl. 1.

  20. 20.

    The armed conflict of World War I came to an end in November 1918. Cyril Falls, The Great War 287, 416 (1959); Wikipedia, Armistice of 11 November 1918, https://en.wikipedia.org/wiki/Armistice_of_11_November_1918 (last visited July 5, 2018). Subsequent to the armistice, treaties that formally terminated the state of war were signed and ratified between 1919 and 1923 by the nations that had been adversaries. Walter Consuelo Langsam, Documents and Readings in the History of Europe Since 1918, at 12, 34 (1939); Wikipedia, World War I, http://en.wikipedia.org/wiki/World_War_I#Formal_end_of_the_war (last visited July 5, 2018). The treaties of the United States were concluded in 1921. Falls, supra, at 423; Wikipedia, Treaty of Versailles, http://en.wikipedia.org/wiki/Treaty_of_Versailles (last visited July 5, 2018).

  21. 21.

    Daniel M. Johnson & Rex R. Campbell, Black Migration in America: A Social Demographic History 71–77, 80–86 (1981); Carole Marks, Farewell—We’re Good and Gone: The Great Black Migration 93–94, 121–22 (1989); Douglas S. Massey & Nancy A. Denton, American Apartheid: Segregation and the Making of the Underclass 28–29 (1993).

  22. 22.

    Jennifer Roback, Southern Labor Law in the Jim Crow Era: Exploitative or Competitive?, 51 U. Chi. L. Rev. 1161, 1163–70 (1984).

  23. 23.

    Larry D. Barnett, Explaining Law: Macrosociological Theory and Empirical Evidence, at 308–10 (2015) [hereinafter Explaining Law].

  24. 24.

    Johnson & Campbell, supra note 21, at 95–96, 108–110, 122–23;

  25. 25.

    Townsand Price-Spratlen, Between Depression and Prosperity? Changes in the Community Context of Historical African American Migration, 77 Soc. Forces 515, 516–18, 527–28, 530 tbl. 2 (1998).

  26. 26.

    This thesis is embodied in propositions 3 and 4 of my macrosociological framework for law. Explaining Law, supra note 23, at 10.

  27. 27.

    Massey & Denton, supra note 21, at 19–24, 26–32, 37–39, 42–48; Stanley Lieberson, A Piece of the Pie: Blacks and White Immigrants since 1880, at 259 tbl. 9.2, 291 (1980); Stewart E. Tolnay et al., Race, Regional Origin, and Residence in Northern Cities at the Beginning of the Great Migration, 67 Am. Sociol. Rev. 456, 471–72 (2002).

  28. 28.

    Stewart E. Tolnay, Educational Selection in the Migration of Southern Blacks, 1880–1990, 77 Soc. Forces 487, 502–05 (1998).

  29. 29.

    1 St. Clair Drake & Horace R. Cayton, Black Metropolis: A Study of Negro Life in a Northern City 203–04 (rev. & enlarged ed. 1970); Florette Henri, Black Migration: Movement North 1900–1920, at 122–24 (1976); Marks, supra note 21, at 147.

  30. 30.

    Henri, supra note 29, at 85; Johnson & Campbell, supra note 21, at 110–11; Kenneth L. Kusmer, A Ghetto Takes Shape: Black Cleveland, 1870–1930, at 166 (1976); Marks, supra note 21, at 145–46; Gilbert Osofsky, Harlem: The Making of a Ghetto 127–31, 135–36, 140 (1966).

  31. 31.

    Drake & Cayton, supra note 29, at 204–05; Henri, supra note 29, at 108–09, 111–14; Johnson & Campbell, supra note 21, at 111; Kusmer, supra note 30, at 221–22; Marks, supra note 21, at 147; Osofsky, supra note 30, at 141–43.

  32. 32.

    The percentages in Table 5.1 were computed from data contained in Table P-1 of 1 Everett S. Lee et al., Population Redistribution and Economic Growth. United States, 1870–1950. Methodological Considerations and Reference Tables, at 107–231 (1957) [hereinafter Methodological Considerations]. Table P-1 uses the label “Negro” rather than the label “Black,” and it applies the signal “…” to designate an estimate that was “[b]elow the level for rounding.” Id. at 361. I recorded an entry of “…” as zero.

  33. 33.

    The measures in Table 5.1 were computed with Stata IC version 12.1.

  34. 34.

    Infra note 75 and accompanying text.

  35. 35.

    U.S. Bureau of the Census, Sixteenth Census of the United States: 1940. Population. Vol. I, Number of Inhabitants vi (1942) (map), available at http://www.census.gov/prod/www/decennial.html (last visited July 5, 2018) (select “Census of Population and Housing, 1940” and then “1940 Census of Population”). The following are the geographic divisions that comprise each region named in Table 5.1:

    Region

    Geographic divisions

    North Central

    West North Central and East North Central

    Northeast

    New England and Middle Atlantic

    South

    East South Central, South Atlantic, and West South Central

    West

    Mountain and Pacific

  36. 36.

    Wikipedia, American Civil War, http://en.wikipedia.org/wiki/American_Civil_War (last visited July 5, 2018).

  37. 37.

    The states were identified from Shelby Foote, The Civil War, A Narrative 52 (1958); Wikipedia, Confederate States of America, http://en.wikipedia.org/wiki/Confederate_States_of_America (last visited July 5, 2018).

  38. 38.

    Bell v. Maryland, 378 U.S. 226, 242, 284 app. V (1964) (Douglas, J., concurring).

  39. 39.

    David Freeman Engstrom, The Lost Origins of American Fair Employment Law: Regulatory Choice and the Making of Modern Civil Rights, 1943–1972, 63 Stanford L. Rev. 1071, 1072–73 (2011).

  40. 40.

    United Steelworkers of America v. Weber, 443 U.S. 193, 202–03 (1979).

  41. 41.

    Engstrom, supra note 39.

  42. 42.

    Because the study was confined to states, it omitted the District of Columbia.

  43. 43.

    Proclamation No. 3269, 24 Fed. Reg. 81 (Jan. 6, 1959) (Alaska); Proclamation No. 3309, 24 Fed. Reg. 6868 (Aug. 25, 1959 (Hawaii).

  44. 44.

    New York was the first state to approve such legislation. Anthony S. Chen, The Party of Lincoln and the Politics of State Fair Employment Practices Legislation in the North, 1945–1964, 112 Am. J. Sociol. 1713, 1717 (2007) [hereinafter The Party of Lincoln]. The New York statute became law on March 12, 1945, and was in force as of July 1, 1945. Act of Mar. 12, 1945, ch. 118, 1945 N.Y. Laws 457–64. A historical account of the New York legislation can be found in Anthony S. Chen, “The Hitlerian Rule of Quotas”: Racial Conservatism and the Politics of Fair Employment Legislation in New York State, 1941–1945, 92 J. Am. Hist. 1238 (2006).

  45. 45.

    Pub. L. No. 88–352, 78 Stat. 241, 268 (1964); President Lyndon B. Johnson, Radio and Television Remarks Upon Signing the Civil Rights Bill, July 2, 1964, http://www.presidency.ucsb.edu/ws/?pid=26361 (last visited July 5, 2018).

  46. 46.

    The search for these states began with lists that had been assembled by credible sources. The lists were found in Bell v. Maryland, 378 U.S. 226, 242, 284 app. V (1964) (Douglas, J., concurring) (reprinting list developed by U.S. Commission on Civil Rights); Duane Lockard, Toward Equal Opportunity: A Study of State and Local Antidiscrimination Laws 21–22 tbl. I, 24 tbl. II (1968); Arnold H. Sutin, The Experience of State Fair Employment Commissions: A Comparative Study, 18 Vand. L. Rev. 965, 965 n.1 (1965); U.S. Equal Employment Opportunity Comm’n, Part 1—Historical and Legislative Background, at 5–6, in Legislative History of Titles VII and XI of Civil Rights Act of 1964 (U.S. Equal Employment Opportunity Comm’n ed. [1968]). Because of some inconsistencies between the lists, I consulted current codified statutes for every state that in any of the lists appeared to satisfy the two criteria, reviewed the notes supplied by these codified statutes on the history of the statutes, and located qualifying session laws. In finding the session laws of some states, the following source proved helpful: W. Brooke Graves, Fair Employment Practice Legislation in the United States, Federal–State–Municipal (1951). A further check involved the seven states that, according to one list, adopted legislation in 1965 or 1966 that banned job-related race bias and that authorized a state agency to enforce the ban. Lockard, supra, at 24 tbl. II. The seven states on this list were checked to ascertain whether any of them might have adopted legislation that met criterion 2. For each of the seven states, I looked up the current codified statute that is relevant to the study and read its notes on the history of the statute; if the notes were not helpful, I attempted to find a qualifying session law that was passed in the year designated by the list. The process uncovered one error in the list. Specifically, although Utah is shown in the list as having adopted qualifying law in 1965, it did not do so until 1969. 1969 Utah Laws 391, 448–57, §§ 160 to 167. Even the earlier year (1965), of course, would not have satisfied criterion 2.

  47. 47.

    Title VII applies to employers and labor organizations that meet the criteria specified in section 701 of the Civil Rights Act. Pub. L. No. 88–352, § 701, 78 Stat. 241, 253–55 (1964) (codified at 42 U.S.C. § 2000e (2012)).

  48. 48.

    George Rutherglen, Employment Discrimination Law: Visions of Equality in Theory and Doctrine 6 (3rd ed. 2010); Ted Gittinger & Allen Fisher, Nat’l Archives & Records Admin., LBJ Champions the Civil Rights Act of 1964, Part 2, Prologue Mag., Summer 2004, https://www.archives.gov/publications/prologue/2004/summer/civil-rights-act-2.html (last visited July 5, 2018).

  49. 49.

    Pub. L. No. 88-352, §§ 705 to 707, 78 Stat. 241, 258–62 (1964) (codified as amended at 42 U.S.C. §§ 2000e-4 to 2000e-6 (2012)). In 1972, the enforcement powers of the Attorney General were moved to the EEOC. Pub. L. 92–261, § 5, 86 Stat. 103, 107 (1972) (codified at 42 U.S.C. § 2000e-6 (2012)).

  50. 50.

    Sutin, supra note 46, at 965.

  51. 51.

    Pub. L. No. 88–352, § 706(c), 78 Stat. 241, 260 (1964) (codified as amended at 42 U.S.C 2000e-5(d) (2012)).

  52. 52.

    Congress believed that Title VII was needed in order to, inter alia, “remove obstructions to the free flow of interstate and foreign commerce.” H.R. Rep. No. 88–914, at 9 (1963), as reprinted in 1964 U.S.C.C.A.N. 2391, 2402. For example, employers as well as labor organizations are not covered by Title VII unless they are “engaged in an industry affecting commerce.” 42 U.S.C. §§ 2000e(b), 2000e(d) (2012). The terms “commerce” and “industry affecting commerce” are defined broadly. 42 U.S.C. §§ 2000e(e), 2000e(g), 2000e(h) (2012). Congress thus evidently rested Title VII to a large degree on the authority it derives from the Commerce Clause of the Constitution. U.S. Const. art. I, § 8, cl. 3 (“The Congress shall have power … To regulate Commerce with foreign Nations, and among the several States, and with the Indian Tribes”). That authority is extensive. Heart of Atlanta Motel, Inc. v. United States, 379 U.S. 241, 257–58 (1964) (Congress is authorized by the Commerce Clause to prohibit an activity that, even though local, is regarded as “a moral and social wrong” as long as evidence exists that the activity has “a substantial and harmful effect” on interstate commerce).

  53. 53.

    A state could have adopted a qualifying statute after Title VII had become law, of course, but it may have done so because of the federal Act. A state whose statute was due to Title VII is not relevant to the instant study, which focuses on whether conditions and forces within states mold the content of state law.

  54. 54.

    A state, within the time interval specified by criterion 2, may have enacted legislation that, although prohibiting job-related race discrimination, did not meet criterion 1, but it may have subsequently adopted legislation that satisfied both criteria. In such cases, Table 5.2 cites the subsequent, not the earlier, legislation. Indiana and Kansas are illustrative. The initial legislation in Indiana and Kansas, respectively, was 1945 Ind. Acts 1499, ch. 325, and 1953 Kansas Sess. Laws 469, ch. 249.

  55. 55.

    For instance, a state that had adopted, within the time period of criterion 2, a statute banning race discrimination in private-sector employment, and placed responsibility for advancing the goals of the statute on an administrative agency, would be coded 0 if the statute permitted the state agency to engage only in conciliation and authorized no court-imposed penalties for discrimination. Three states—Nevada, Oklahoma, and West Virginia—were coded 0 for this reason. 1961 Nev. Stat. 731, ch. 364, § 7; 1963 Okla. Sess. Laws 470, ch. 322; 1961 W. Va. Acts 692, ch. 135, §§ 1, 4.

  56. 56.

    Logistic regression must be used, and least-squares regression cannot be used, when cases have been quantified as either 0 or 1 on the dependent variable. Fred C. Pampel, Logistic Regression: A Primer 1–10 (Paper No. 132, Sage University Paper Series on Quantitative Applications in the Social Sciences, 2000).

  57. 57.

    The Party of Lincoln, supra note 44, at 1747, 1749 tbl. 3; William J. Collins, The Political Economy of Race, 1940–1964: The Adoption of State-Level Fair Employment Legislation (Nat’l Bureau of Econ. Research, Historical Paper No. 128, 2000), available at http://www.nber.org/papers/h0128. However, both studies estimated the relationship between the percentage of a state population that was Black and whether the state passed FEP legislation. With other independent variables held constant, a higher percentage was found to reduce the likelihood that such legislation would be adopted. The Party of Lincoln, supra note 44, at 1750 tbl. 4; Collins, supra, at 20, 36.

  58. 58.

    Supra Sect. 5.1.2.

  59. 59.

    Christoph Brumann, Writing for Culture: Why a Successful Concept Should Not Be Discarded, 40 Current Anthropology S1 (1999).

  60. 60.

    Vytautas Kavolis, Civilization Analysis as a Sociology of Culture, 3 Sociol. Theory 29, 34 (1985). Accord, Explaining Law, supra note 23, at 68–69.

  61. 61.

    Isaac Ariail Reed & Julia Adams, Culture in the Transitions to Modernity: Seven Pillars of a New Research Agenda, 40 Theory & Soc’y 247, 250 (2011).

  62. 62.

    Explaining Law, supra note 23, at 68–69. For quantitative evidence that the culture of a jurisdiction has a bearing on whether certain concepts and doctrines of law are present in the jurisdiction, see id. at ch. 2 pt. 2, ch. 3 pts. 2 & 3. See also Sect. 2.3.2.2 of Chap. 2 in the first volume.

  63. 63.

    David Lockwood, Social Integration and System Integration, in Explorations in Social Change 244, 245 (George K. Zollschan & Walter Hirsch eds., 1964).

  64. 64.

    Id. at 245 (characterizing the distinction as “wholly artificial”).

  65. 65.

    The sociological concept of institution is explained in The Place of Law, supra note 3, at 43–44, 394.

  66. 66.

    Social disorganization is considered by sociologists and criminologists to be a major source of crime. E.g., Robert J. Sampson & W. Byron Groves, Community Structure and Crime: Testing Social-Disorganization Theory, 94 Am. J. Sociol. 774, 777–80 (1989). A pair of points meriting mention here is that (1) social disorganization has been conceptualized to include poverty and (2) poverty has been empirically shown to raise the incidence of crime. Barbara D. Warner, Whither Poverty? Social Disorganization Theory in an Era of Urban Transformation, 32 Sociol. Focus 99, 100–04 (1999). The conceptual connection of social disorganization to poverty and the empirical connection of poverty to crime has prompted the recommendation that poverty should be a key constituent of theory dealing with social disorganization. Id. at 110. However, poverty can be subsumed under socio-economic status, and socio-economic status and social disorganization can be treated as separate phenomena. The differentiation of socio-economic status from social disorganization is warranted because the impact of socio-economic status on the incidence of socially disruptive activities such as crime is mediated by, or occurs entirely apart from, some aspects of social disorganization. Sampson & Groves, supra at 777–78, 799; Warner, supra at 102, 104; Ivan Y. Sun et al., Neighborhood Characteristics and Crime: A Test of Sampson and Groves’ Model of Social Disorganization, 5 W. Criminology Rev. 1, 7–10 (2004). Socio-economic status, therefore, is a distinct phenomenon and as such is usable in conjunction with an independent variable other than social disorganization. In Chap. 5, a component of socio-economic status—viz., educational attainment—serves as the indicator of societal rationality. Infra Sect. 5.2.1.3.

  67. 67.

    See Kenneth D. Bailey, Sociology and the New Systems Theory: Toward a Theoretical Synthesis 44 (1994) (attributing to a system “an entropy value below the maximum” and defining entropy as “degree of disorder”).

  68. 68.

    Explaining Law, supra note 23, at 118–19.

  69. 69.

    Id. at 98–99, 122.

  70. 70.

    The U.S. Supreme Court has observed that such a stereotype is the product of “irrational or uncritical analysis.” Nguyen v. Immigration and Naturalization Service, 533 U.S. 53, 68 (2001).

  71. 71.

    The Place of Law, supra note 3, at 267–68, 342.

  72. 72.

    Id. at 197, 213–14.

  73. 73.

    The Bureau, when coding the race of individuals in the censuses that yielded the data, employed the term “Negro,” not the term “Black.” The latest (2010) decennial census, in its question on race, included the category “Black, African Am., or Negro.” Sonya Rastogi et al., U.S. Census Bureau, The Black Population: 2010, at 1, 2 (2011), available at www.census.gov/prod/cen2010/briefs/c2010br-06.pdf (last visited July 5, 2018).

  74. 74.

    Text accompanying supra notes 44 & 45.

  75. 75.

    U.S. Bureau of the Census, Sixteenth Census of the United States: 1940. Population. Vol. II, Characteristics of the Population. Part I, United States Summary and Alabama–District of Columbia, at 93 tbl. 40 (1943), available at http://www.census.gov/prod/www/decennial.html (last visited July 5, 2018). Labor force participation rates among persons who were age 14 or older are reported for nonwhites as a whole, not for particular categories of nonwhites. Id. Of all nonwhite males in 1940, 94.8 percent were classified as Negro, and of all nonwhite females in 1940, 96.4 percent were classified as Negro. Computed from id. at 19 tbl. 4 (using data on persons of all ages). Accordingly, the sex-specific levels of labor force participation among Negro males and females were unlikely to have diverged markedly from the corresponding levels that are given in the text for nonwhite males and females. The 1940 census uses the term “Negro” rather than the term “Black.” Id. at 9. I have employed its terminology here.

  76. 76.

    The data for computing the indicators were obtained from Table P-1 in Methodological Considerations, supra note 32. According to id. at 361, the signal “…” was used in Table P-1 to represent an estimate that was “[b]elow the level for rounding”; in the data compiled from Table P-1 for the instant study, “…” was treated as zero. Table P-1, when reporting data that involves race, uses the labels “Negro” and “White.”

    The extraction of all of the data for the two indicators from a single source maximizes the likelihood that the data were generated with consistently defined concepts and consistently followed procedures. Nonetheless, the data are unavoidably estimates due to errors that occurred in conducting censuses and changes that occurred between censuses in definitions of concepts. Everett S. Lee, Migration Estimates, in id., at 6–7.

  77. 77.

    Everett S. Lee, Migration Estimates, in Methodological Considerations, supra note 32, at 9.

  78. 78.

    The number of Blacks at the start of each decade is provided for every state in Methodological Considerations, supra note 32, at 352 tbl. P-4A. However, the data in table P-4A are for all Blacks, not just Black males, and are for all ages, not just Black males aged 10 or older. As a result, the data in table P-4A were not used for the study.

  79. 79.

    The number of U.S.-born Whites at the start of each decade is provided for every state in Methodological Considerations, supra note 32, at 249–97 tbl. P-3, 350 tbl. P-4A. However, the data in Tables P-3 and P-4A are for all Whites, not just White males; in addition, the data cover all ages, not just White males aged 10 or older. The data in Tables P-3 and P-4A, therefore, were not used in the study.

  80. 80.

    Explaining Law, supra note 23, at 70, 116–17; Peter J. Rentfrow et al., A Theory of the Emergence, Persistence, and Expression of Geographic Variation in Psychological Characteristics, 3 Persp. Psychol. Sci. 339, 351–54, 360 (2008); Peter J. Rentfrow et al., Divided We Stand: Three Psychological Regions of the United States and Their Political, Economic, Social, and Health Correlates, 105 J. Personality & Soc. Psychol. 996, 998–1001, 1004–08 (2013); Dara Shifrer & April Sutton, Region-Urbanicity Differences in Locus of Control: Social Disadvantage, Structure, or Cultural Exceptionalism?, 84 Sociol. Inquiry 570, 572, 575–76, 592 (2014).

  81. 81.

    Supra note 35 and accompanying text.

  82. 82.

    The states within each region are listed in supra Table 5.1.

  83. 83.

    Melissa A. Hardy, Regression with Dummy Variables 7–12, 26–28 (Paper No. 93, Sage University Paper Series on Quantitative Applications in the Social Sciences, 1993).

  84. 84.

    The percentages were in U.S. Bureau of the Census, Sixteenth Census of the United States: 1940. Population. Vol. II, Characteristics of the Population. Part I, United States Summary and Alabama–District of Columbia, at 97 tbl. 44 (1943), available at http://www.census.gov/prod/www/decennial.html (last visited July 5, 2018) (select “Census of Population and Housing, 1940” and then “1940 Census of Population”). The category “proprietors, managers, and officials” included all persons in these occupations, whether employed in the private sector or in the public sector. The specific occupations covered by the category are listed in id. at 17. Additional categories used for occupation in the 1940 census (but omitted from the instant study) were “Professional Workers,” “Semiprofessional Workers,” “Farmers and Farm Managers,” “Clerical, Sales, and Kindred Workers,” “Craftsmen, Foremen, and Kindred Workers,” “Operatives and Kindred Workers,” “Domestic Service Workers,” “Service Workers, Except Domestic,” “Farm Laborers (Wage Workers) and Farm Foremen,” “Farm Laborers (Unpaid Family Workers),” and “Laborers, Except Farm.” Id. at 17.

  85. 85.

    Id. at 91 tbl. 38 (males), 92 tbl. 39 (females).

  86. 86.

    The percentages were computed from data in U.S. Bureau of the Census, Sixteenth Census of the United States: 1940. Population and Housing: Families. General Characteristics: States, Cities of 100,000 or More, and Metropolitan Districts of 200,000 or More, at 57–63 tbl. 22 (1943), available at http://www.census.gov/prod/www/decennial.html (last visited July 5, 2018) (select “Census of Population and Housing, 1940” and then “1940 Census of Population and Housing”). The data from this source that were used for the calculation involved three numbers for each state: (i) the total number of families with a female head under age 45); (ii) the number of families with a female head under age 45 in which there were no children under age 18; and (iii) the total number of families. The calculation for each state was performed with the formula ((i − ii) ÷ iii) × 100, and the resulting percentages were rounded to three decimal places.

    A “family” and “family head” were defined by the Bureau of the Census as follows: “The term ‘family’ … is limited to private families. … A private family comprises a family head and all other persons in the home who are related to the head by blood, marriage, or adoption, and who live together and share common housekeeping arrangements. A person living alone is counted as a one-person private family. A family head sharing his living accommodations with one or more unrelated persons, or providing rooms for the use of lodgers, servants, or hired hands, is also counted as a one-person private family. A group of related persons residing permanently or for an indefinite period in an apartment hotel is counted as a private family.” Id. at 2. “One person in each family was designated … as the family head, that is, the person regarded as the head by the members of the family. … The head of a private family is usually a married man and the chief breadwinner in the family. In some cases, however, the head is a parent of the chief earner or is the only adult member of the household.” Id. at 3.

  87. 87.

    The percentages were calculated using data obtained from table 19 for each state in Bureau of the Census, Sixteenth Census of the United States: 1940. Population Vol. IV, Characteristics by Age: Marital Status, Relationship, Education, and Citizenship. Part 2: Alabama–Louisiana, Part 3: Maine–North Dakota, & Part 4: Ohio–Wyoming (1943), available at http://www.census.gov/prod/www/decennial.html (last visited July 5, 2018) (select “Census of Population and Housing, 1940” and then “1940 Census of Population”). To do the calculation, three numbers were extracted from table 19 for every state: (i) the number of state residents aged 25+ in 1940 who, as of 1940, had completed four years of college, (ii) the number of state residents aged 25+ in 1940 who, as of 1940, had completed five or more years of college, and (iii) the total number of state residents who were aged 25+ in 1940. The percentage for each state was obtained with the following formula: ((i+ii) ÷ iii) × 100. The percentage for a state thus included every person in the state who satisfied the age criterion (i.e., who was at least 25 years old), regardless of ethnicity, race, or sex.

  88. 88.

    All of the findings in Sect. 5.2.2 were generated with Stata IC version 12.1 (64-bit). Stata and its several “flavors,” including IC, are described in StataCorp., Stata User’s Guide Release 12, at 49–52 (2011). Regression analyses employed the logistic command. StataCorp., Stata Base Reference Manual Release 12, at 932–42 (2011).

  89. 89.

    Supra note 35 and accompanying text.

  90. 90.

    J. Scott Long & Jeremy Freese, Regression Models for Categorical Dependent Variables Using Stata 140 (2nd ed. 2006).

  91. 91.

    Supra notes 44 & 45 and Table 5.2.

  92. 92.

    The intercept had just a small influence on the accuracy rate. When the intercept was included with the independent variables that comprise Model I, the accuracy rate was 77.1 percent.

  93. 93.

    The standardization procedure used the standard deviations of the indicators of just the independent variables. It did not use the standard deviation of the indicator of the dependent variable, because that indicator was measured with the numbers 0 and 1, that is, as a dummy. Pampel, supra note 56, at 32–33.

  94. 94.

    The listcoef command in Stata was used to obtain the standardized measures. Long & Freese, supra note 90, at 178.

  95. 95.

    The factors for a one standard-deviation increase in propmgr, in disorder, and in coll4plus were 0.691, 0.642, and 1.319, respectively. A factor above 1.000 and a factor below 1.000 are compared in magnitude by employing the numerical value of one factor and the reciprocal of the numerical value of the other factor. Long & Freese, supra note 90, at 179. See also Sect. 1.1 in supra Chap. 1.

  96. 96.

    Section 1.5 of supra Chap. 1.

  97. 97.

    propmgr had the highest VIF score (6.90). Every other independent variable had a VIF score under 4.0.

  98. 98.

    The smallest Cook’s Statistic among the seven states was 1.74 (for Florida); the largest was 4.13 (for Vermont). For six of the seven states, the probability was below 0.50 that the state would be in the category of the dependent variable in which it fell. The probability was computed with the leastlikely command. Long & Freese, supra note 90, at 152–53.

  99. 99.

    After Florida was taken out of the data, the odds ratios for bmigwpop and bmigbpop were 2.742 and 1.021, respectively. The difference between the odds ratio for bmigwpop in the absence of Florida (2.742) and the odds ratio for bmigwpop in Model I (viz., 1.441) warrants comment. A percentage-point increase in bmigwpop raised the odds that a state would have an FEP statute by 174.2 percent when Florida was excluded from the data but by just 44.1 percent when Florida was included. However, Florida did not under Model I have a low probability of being where it was on the dependent variable, that is, being without an FEP statute. Indeed, as calculated by the leastlikely command, the probability that Model I had correctly placed Florida on the dependent variable was 0.80 and hence very high. Because the failure of Florida to enact FEP legislation is unsurprising, Florida was considered not to be an influential outlier. See Sect. 1.5 of supra Chap. 1.

    For the six other states that, like Florida, had a Cook’s Statistic of 1.00 or more, the probability under Model I of being where they were on the dependent variable was below 0.50. Four of the six states had a probability below 0.30, making them credible candidates to be influential outliers. However, when each of these four states was excluded (one at a time) from the regression analysis, bmigwpop and bmigbpop had odds ratios that were not markedly different from those in Model I—the odds ratios for bmigwpop were in the range 1.285–2.165; the odds ratios for bmigbpop were in the range 1.036–1.047.

  100. 100.

    Robert Wuthnow & Marsha Witten, New Directions in the Study of Culture, in 14 Ann. Rev. Sociol. 49 (1988).

  101. 101.

    Robert K. Merton, Social Theory and Social Structure 556 (enlarged ed., 1968).

  102. 102.

    Talcott Parsons: Sociological Theory and Modern Society 141–43 (1967).

  103. 103.

    See Parsons, supra note 102, at 145 (maintaining that the values of culture, “through legal or informal norms,” must justify the ends pursued by the operating units of a society); Merton, supra note 101, at 187 (postulating that culture, inter alia, “regulates and controls the acceptable modes of reaching out for” culturally endorsed goals).

  104. 104.

    Robert M. Stein et al., Reconciling Context and Contact Effects on Racial Attitudes, 53 Pol. Res. Q. 285 (2000).

  105. 105.

    Supra Table 5.3.

  106. 106.

    James Jaccard, Interaction Effects in Logistic Regression 30–34 (Paper No. 135, Sage University Paper Series on Quantitative Applications in the Social Sciences, 2001).

  107. 107.

    Id. at 32–33.

  108. 108.

    The recalculations yielded odds ratios for bmigwpop from 723.46 to 0.27.

  109. 109.

    Jaccard, supra note 106, at 33.

  110. 110.

    The interaction variable bmigwpop * south was omitted because the reference region in Model II was the South, as it was in Model I. Id. at 30–33.

  111. 111.

    In Model II, the standardized regression coefficient for bmigwpop (1.458) was above 1.000, but the standardized regression coefficient for bmigbpop (0.570) was below 1.000, evidencing opposite effects on the dependent variable. As before (supra notes 93 & 94 and accompanying text), the coefficients were standardized on the standard deviations of just the independent variables.

  112. 112.

    Although the two groups each classified 39 states on feplaw without error, their accuracy rates were not the same because they differed in the total number of states predicted to have an FEP statute, in the total number of states predicted not to have an FEP statute, and in the number of states within the former total as well as within the latter total that were correctly placed on feplaw.

  113. 113.

    The standard deviations of the independent variables were used to create standardized measures for Model III in the same manner that they were used to produce such measures for Model I. Supra note 93 and accompanying text; the standardized measures for Model I are in (i) the text accompanying supra note 94 and (ii) note 95. In Model III, the factor for a one standard-deviation increase in bmigwpop was 4.907, which materially exceeded the factor for a one standard-deviation increase in bmigbpop (1.879) and in coll4plus (2.331). It also exceeded, though not by a large margin, the factor for a one standard-deviation increase in propmgr (0.297). To compare a factor above 1.000 and a factor below 1.000 in terms of their magnitudes, see supra note 95.

  114. 114.

    The societal agents behind state FEP statutes were the subject of two earlier studies, but while the share of the state population that was Black was an independent variable in both studies, neither study employed data on the volume of state-level interstate migration by Blacks. In each study, an inverse association was found between the percentage of the state population that was Black and the likelihood that an FEP law would be adopted by the state; that is, as Blacks became a larger share of the population, the likelihood of an FEP statute became smaller. The Party of Lincoln, supra note 44, at 1750–51; Collins, supra note 57, at 20, 36.

  115. 115.

    A rise in the mean number of years of school completed by White adults in a state, according to a prior study, may have deterred the enactment of FEP legislation by the state. Collins, supra note 57, at 17, 22, 36. The finding by Professor Collins may diverge from mine because our studies used very different measures of education and/or because our studies used markedly different models.

  116. 116.

    Supra note 113.

  117. 117.

    Larry D. Barnett, The Roots of Law, 15 Am. U. J. Gender Soc. Pol’y & L. 613, 627–35 (2007).

  118. 118.

    Explaining Law, supra note 23, at 300–04.

  119. 119.

    Bureau of Labor Statistics, Glossary, http://www.bls.gov/bls/glossary.htm (last visited July 6, 2018) (definition of “unemployment rate”).

  120. 120.

    The data for State A are the national unemployment rates of Black males aged 20 and older during the period 1987–1997; the data for State B are the national unemployment rates of Black males aged 20 and older during the period 1978–1988. The data were obtained from the Bureau of Labor Statistics, U.S. Dep’t of Labor, Labor Force Statistics from the Current Population Survey, http://www.bls.gov/cps/cpsatabs.htm (last visited July 6, 2018) (follow “Table A-2. Employment status of the civilian population by race, sex, and age” hyperlink; check box “Not seasonally adjusted” for “Unemployment rate” of “Men, 20 years and over” under “Black or African American”; select button “Retrieve data”; choose years from drop-down menu; and click “Go”). The Bureau of Labor Statistics does not publish race-specific rates of unemployment for years prior to 1972.

  121. 121.

    The coefficient of variation is based on the ratio of the standard deviation of a set of numbers to the mean of the numbers. When the ratio itself is used, the coefficient is expressed as a proportion; when the ratio is multiplied by 100, the coefficient is expressed as a percentage. To avoid confusing the coefficient with the unemployment rate (which the text discusses in terms of percentage points), the coefficient is measured here as a proportion.

  122. 122.

    William M. Landes, The Effect of State Fair Employment Laws on the Economic Position of Nonwhites, 57 Am. Econ. Rev. 578, 587–90 (May 1967). The study apparently used data from the censuses of 1940, 1950, and 1960.

  123. 123.

    David Neumark & Wendy A. Stock, The Labor Market Effects of Sex and Race Discrimination Laws, 44 Econ. Inquiry 385, 411–15 (2006). The data for the study were from the censuses of 1940, 1950, and 1960. Id. at 410.

  124. 124.

    Id. at 414.

  125. 125.

    William J. Collins, The Labor Market Impact of State-Level Anti-Discrimination Laws, 1940–1960, 56 Indus. & Lab. Rel. Rev. 244 (2003). Data from the censuses of 1940, 1950, and 1960 were used in the study. Id. at 250.

  126. 126.

    Id. at 254 tbl. 1.

  127. 127.

    Id. at 257 tbl. 2.

  128. 128.

    Id.

  129. 129.

    The Place of Law, supra note 3, at 198–200.

  130. 130.

    Explaining Law, supra note 23, at 14–15, 55–58, 252 n.45.

  131. 131.

    The Place of Law, supra note 3, at 59–60, 395, 398, 403. Symbols probably foster social integration as well as system integration. See text accompanying supra notes 63 to 65.

  132. 132.

    This reasoning has been advanced in a study that attributed economic gains among Blacks in the South to the ban that Title VII of the federal Civil Rights of 1964 imposed on employment-related race discrimination. John J. Donohue III & James Heckman, Continuous Versus Episodic Change: The Impact of Civil Rights Policy on the Economic Status of Blacks, 29 J. Econ. Lit. 1603, 1605, 1640 (1991).

  133. 133.

    Supra note 118 and accompanying text.

Author information

Authors and Affiliations

  1. Widener University Delaware Law School, Wilmington, DE, USA

    Larry D. Barnett

Authors
  1. Larry D. Barnett
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Larry D. Barnett .

Rights and permissions

Reprints and permissions

Copyright information

© 2019 The Author(s)

About this chapter

Check for updates. Verify currency and authenticity via CrossMark

Cite this chapter

Barnett, L.D. (2019). Race, Interstate Migration, and Employment Discrimination Law. In: Societal Agents in Law. Palgrave Macmillan, Cham. https://doi.org/10.1007/978-3-030-02004-0_5

Download citation

  • DOI: https://doi.org/10.1007/978-3-030-02004-0_5

  • Published:

  • Publisher Name: Palgrave Macmillan, Cham

  • Print ISBN: 978-3-030-02003-3

  • Online ISBN: 978-3-030-02004-0

  • eBook Packages: Social SciencesSocial Sciences (R0)

Publish with us

Policies and ethics

Search

Navigation