¹ This is one of the major themes of their City Politics (Cambridge: Harvard University Press and the M.I.T. Press, 1963)Google Scholar; see also Wilson, James Q. and Banfield, Edward C., “Public-Regardingness as a Value Premise in Voting Behavior,” this Review, 58 (12, 1964), 876–887Google Scholar.

² (New York: Knopf, 1955), pp. 8–9.

³ Banfield and Wilson, p. 46.

⁴ Here, with Banfield and Wilson, we refer not to the stylish young liberal club members of California and Manhattan, but to the more conservative “good government” forces in many cities.

⁵ Throughout this article the “percent foreign stock” or “percent ethnic” refers to the proportion of a city's 1960 population that is foreign born or native born with at least one parent born abroad. Nineteen per cent of the total U. S. population is of foreign stock. Data on nativity and parentage are from U. S. Bureau of the Census, County and City Data Book, 1962 (Washington: U. S. Government Printing Office, 1962)Google Scholar.

⁶ Perhaps because so much of the best scholarly research on local politics has been conducted close to the great universities of the Northeast and perhaps also because most serious nonacademic writers live in a few northeastern cities, political organizations in these cities have been described at great length, while very little is known about existing machines in other parts of the country.

⁷ Banfield and Wilson, p. 330; see also ibid., pp. 92, 95, 154, 170.

⁸ See, e.g., Schnore, Leo F. and Alford, Robert F., “Forms of Government and Socioeconomic Characteristics of Suburbs,” Administrative Science Quarterly, 8 (06, 1963), 1–17CrossRefGoogle Scholar.

⁹ See especially Kaufman's, Herbert review of City Politics in this Review, 58 (06, 1964), 422–423Google Scholar. Kaufman asks, “are those measures designated ‘public-regarding’ by Banfield and Wilson really manifestations of selfless fellow feeling or are they the self-serving policies of a particular group in society that is trying to hold on to what it has?” (p. 423).

¹⁰ In these computations the Midwest includes Ohio, Indiana, Illinois, Iowa, Michigan, Minnesota, and Wisconsin.

¹¹ Banfield and Wilson, pp. 38–46.

¹² Wilson and Banfield, p. 885.

¹³ Banfield and Wilson, p. 123; see also ibid., p. 329.

¹⁴ Wolfinger, Raymond E., “The Development and Persistence of Ethnic Voting,” this Review, 59 (12, 1965), 896–908Google Scholar.

¹⁵ In other areas of political behavior the upward mobile tend to have characteristics midway between their old and new classes; see Barber, James A. Jr., , Social Mobility and Political Behavior (Chicago: Rand McNally, forthcoming)Google Scholar.

¹⁶ Banfield and Wilson, p. 123.

¹⁷ Ibid., p. 40.

¹⁸ Ibid., p. 235.

¹⁹ See, e.g., Berg, Louis, “Peddlers in Eldorado,” Commentary, 07, 1965, pp. 64–66Google Scholar.

²⁰ Pomeroy, Earl, The Pacific Slope (New York: Alfred A. Knopf, 1965)Google Scholar.

²¹ Similarly, an analysis of local voting returns in the 1960 presidential election showed considerable regional variation in the responses of similar Protestant voting groups to President Kennedy's candidacy; see Davidowicz, Lucy S. and Goldstein, Leon J., Politics in a Pluralist Democracy (New York: Institute of Human Relations Press, 1963)Google Scholar.

²² Kessel, John H., “Governmental Structure and Political Environment: A Statistical Note about American Cities,” this Review, 56 (09, 1962), 615–620Google Scholar.

²³ Schnore and Alford, op. cit., p. 12.

²⁴ Edgar L. Sherbenou found that Chicago suburbs with higher priced homes are very likely to use the manager form, while more modest towns all have mayors; see his “Class, Participation, and the Council-Manager Plan,” Public Administration Review, 21 (Summer, 1961), 131–135CrossRefGoogle Scholar.

²⁵ Cutright, Phillips, “Nonpartisan Electoral Systems in American Cities,” Comparative Studies in Society and History, 5 (01, 1963), p. 218CrossRefGoogle Scholar.

²⁶ One proposition of the ethos theory states that public-regardingness is manifested in support for expansion of many government services that do not benefit the individual, but that will increase his tax payments. In their own work Banfield and Wilson have found that, among homeowners subject to property taxes, support for such measures to be financed by property taxation rises with the median income and home value of the voting unit. They also found that voting units with large foreign-stock populations were less likely to support such measures. They believe that the controlling element in these relationships is ethnicity, and attribute the findings to the prevalence of the private-regarding ethos among ethnic group members; see Banfield and Wilson, pp. 237–240; and Wilson and Banfield.

²⁷ A number of other political forms, styles, and policies are said to be favored by one or the other of the two ethics. Data permitting intercity comparisons are not available on these other variables.

²⁸ Nolting, Orin F. and Arnold, David S. (eds.), The Municipal Year Book 1963 (Chicago: The International City Managers' Association, 1963)Google Scholar.

²⁹ Urban Renewal Administration, Urban Renewal Directory, December 31, 1963 (Washington: Urban Renewal Administration, 1964)Google Scholar.

³⁰ For information on state restrictions on home rule we are grateful to, among others, David S. Arnold of the International City Managers' Association, William N. Cassella, Jr. of the National Municipal League, Eugene C. Lee of the Institute of Governmental Studies of the University of California (Berkeley), and Keith Ocheltree of the Public Personnel Association.

³¹ Banfield and Wilson do not include Negroes in those groups they consider disposed to private-regardingness, although some political goals they attribute to Negroes, such as desires for representation and “recognition,” might be thought to incline them in that direction (Banfield and Wilson, pp. 158–159, 293–294, 307–308). On the other hand, Negroes are described as disproportionately public-regarding in their voting on bond issues (ibid., pp. 237–239).

³² See, e.g., ibid., pp. 42, 123, 330. While upper-middle-class Jews are often classified with Yankees as adherents of the public-regarding ethos, Banfield and Wilson do not say if they are any more likely to be public-regarding than Catholics of similar social status.

³³ We think it unlikely that most individuals' political knowledge has developed to the point of having opinions about, say, the relative merits of ward and at-large elections. Thus the ethics may exist at two levels: as general value systems for most people and as a set of specific political preferences for their leaders.

³⁴ Banfield and Wilson, p. 55. Our interpretation of this statement is based on its context, Banfield and Wilson's citation of the Sherbenou and Schnore—Alford correlational studies in support of their proposition about public-regardingness and the city manager plan (ibid., p. 169n), and passages like this:

“Many council-manager cities are upper-class or middle-class in character; few if any are predominantly lower-class. In the Chicago area, for example, … eighteen of the twenty cities with the highest home values had the [manager] plan, whereas none of the thirty-one cities with the lowest home values had it. Its popularity with people of the upper and middle classes explains its popularity in small communities, which are more likely to consist predominantly of those classes than are Urge ones” (ibid., p. 169).

³⁵ Ibid., p. 141. The International City Managers' Association considers nonpartisanship and at-large election “main features” of the manager system (ibid., p. 172).

³⁶ Cf. Banfield and Wilson, “The connection between the partisan and district systems, as between the nonpartisan and at-large systems, is of considerable significance, for, as we shall see later, the connected elements tend in both cases to produce the same style of politics and to reinforce one another” (p. 90).

³⁷ Ibid., p. 235.

³⁸ Cities that elect at least three-quarters of their municipal legislators from wards are classified as using the ward system; the same criterion is used with the at-large system. Cities that elect more even proportions by the two methods are classified as “combination” cities. Since all cities using the commission form elect the commission at large, they are omitted from these tabulations.

³⁹ “However important may be the evasions of the civil service system in particular cases, it is clear that in general the effect of the system everywhere has been to make it increasingly difficult for the parties to maintain effective discipline over their workers by giving and with-holding jobs” (ibid., p. 209).

⁴⁰ State law in Anglo-Saxon Indiana makes it almost impossible for cities there to deviate from the mayor form of government. Since form of municipal government is not subject to local choice, we have excluded the nine Indiana cities from the tabulation presented in Table 4. Several other states interfere with home rule in this respect, but their cities all use the commission form and on this ground are excluded from our tabulations.

⁴¹ Cf. Banfield and Wilson, “The larger the city, generally speaking, the more is at stake politically, and consequently the greater the effort that professional politicians will put forth to avoid being displaced. This is certainly a factor that generally tends to prevent adoption of the [city manager] plan in a large city” (pp. 182–183).

⁴² The western cities are those in the conventionally defined eleven western states, plus Honolulu. The Northeast includes the six New England states plus New York, New Jersey, and Pennsylvania. The Midwest includes Ohio, Indiana, Illinois, Iowa, Michigan, Minnesota, Wisconsin, North Dakota, South Dakota, Nebraska, and Kansas. The South includes the eleven ex-Confederate states.

⁴³ Most western cities examined here are in California, where all cities use the nonpartisan ballot. California law does not require non-partisan local elections, however. In 1913 the legislature required “general-law” cities (those without charters) to use the nonpartisan ballot. This did not apply to charter cities, of which there were 70 in 1960, including the communities in our sample. Any California municipality of more than 3500 persons may adopt its own charter. See Lee, Eugene C., The Politics of Non-partisanship (Berkeley and Los Angeles: University of California Press, 1960), pp. 13–15, 23Google Scholar.

⁴⁴ For a description of Mexican-American political apathy in one Texas border city see Banfield, Edward C., Big City Politics (New York: Random House, 1965), pp. 76–78Google Scholar.

⁴⁵ The significance of differences between means was measured with a two-tailed t test.

⁴⁶ In this and all other analyses of relationships between form of government and socioeconomic variables, the nine Indiana cities have been omitted.

⁴⁷ Controlling for social class by means of analysis of covariance did not produce greater differences in ethnicity between the two types of city. Whenever relationships between a dependent variable and a demographic independent variable are described in this paper, we have also controlled for the possible influence of other demographic variables by analysis of covariance. Except where noted, this procedure did not increase or diminish the explanatory power of the independent variable being examined.

⁴⁸ The same is true for the other structural dependent variables; the “public-regarding” north-eastern cities are more ethnic than “private-regarding” cities in other parts of the country (see Tables 11 and 12).

⁴⁹ The five cities with populations of more than a million persons have been excluded from these tabulations involving methods of electing councilmen.

⁵⁰ Banfield and Wilson, pp. 183–184.

⁵¹ Significance of correlations is measured by the F value.

⁵² We counted all projects classified in the execution stage, or where a loan and grant contract was approved but not yet formally executed, in the December 31, 1963 directory. A renewal project in the execution stage at the end of 1963 is the product of prior decisions extending at least two or three years into the past.

⁵³ Several western and southern states were very slow to pass enabling legislation to permit their cities to undertake urban renewal projects; one or two states still impose severe restraints in this regard. We excluded southern states from our analysis; the western states that followed this policy do not, with one exception, contain cities with populations over 50,000.

⁵⁴ Amos H. Hawley has found an inverse relationship between the proportion of businessmen and professionals in a city's population and the likelihood that it has begun to carry out a renewal project. See his “Community Power and Urban Renewal Success,” American Journal of Sociology, 68 (01, 1963), 422–431CrossRefGoogle Scholar. Hawley's study does not take account of the magnitude of a city's program. It has been criticized on other grounds by Bruce C. Straits, “Community Action and Implementation of Urban Renewal,” ibid., 71 (July, 1965), 77–82. See also Hawley's rejoinder, ibid., pp. 82–84.

⁵⁵ An earlier study found no relationship between form of local government and magnitude of urban renewal. See Duggar, George S., “The Relation of Local Government Structure to Urban Renewal,” Law and Contemporary Problems, 26 (1961), 49–69CrossRefGoogle Scholar.

⁵⁶ This southern concern with unity may also explain why Mexican-Americans in Texas have been so apolitical, in contrast to the political involvement of immigrants in the North. Perhaps the political participation of newly arrived minority groups is enhanced when the political environment is competitive. European immigrants to northern cities typically arrived in political arenas where partisan competition motivated both sets of contenders to appeal for their votes. As the immigrants acquired political skill, their prices rose until they had attained political influence at least somewhat in proportion to their numbers. But if immigrants come to a political system where the elites shun conflict with each other, they are likely to find that the interest of those elites is to exclude them from politics rather than appeal for their support. The impact of immigrant political values obviously would be far greater in the former case.

⁵⁷ Regional differences in ethnic politics and salience are discussed at greater length in Wolfinger, op. cit., p. 898.

⁵⁸ Banfield and Wilson, p. 95.

⁵⁹ As we noted earlier, the scanty scholarly knowledge of nonmetropolitan political machines may be due to the fact that most of the best research on local politics has been done at schools like Harvard, Columbia, and Chicago. These universities are located in cities where politics is conducted almost entirely by members of ethnic groups. From the vantage point of Hyde Park or Harvard Square, almost everyone not connected with the university is an ethnic. Political practices in such cities typically are “private-regarding.” Much of the opposition to these practices comes from people connected with or attracted to universities, that is, mostly from Jews and Protestants. College faculties and their social satellites are scarcely typical of the Protestant middle class. Banfield and Wilson may be overgenerous to the bulk of the Anglo-Saxon population. While professors and account executives are not particularly interested in patronage, ticket fixing, and the like, these and other elements of the private-regarding style may be quite congenial to people who have little in common with college faculties but nonmembership in the Catholic Church.