¹ Commons, John R., Races and Immigrants in America (New York, 1907), pp. 69–70Google Scholar.

² The following section relies on the Reports of the Immigration Commission, especially Abstracts, vol. 1 (Washington, 1911), pp. 12–37.

³ Ibid., p. 171, Tables 3 and 4, and p. 182, Tables 16 and 17.

⁴ Ibid., p. 183, Table 18.

⁵ Jones, Maldwyn Allen, American Immigration (Chicago, 1960), p. 252Google Scholar; Hourwich, Isaac A., Immigration and Labor (New York, 1912)Google Scholar, observes that similar arguments had been advanced in the 1880s against German and other “old” immigrants.

⁶ Jones, American Immigration, p. 295. For references on nativism, see p. 338.

⁷ Edward A. Ross, “The Significance of Emigration: Comment,” American Economic Review (Mar. 1912), 86, as cited in Lowell E. Gallaway, Richard K. Vedder, and Vishwa Shukla, “The Distribution of the Immigrant Population in the United States: An Economic Analysis,” Explorations in Economic History 11 (Spring 1974), 213–26, 215.

⁸ Oscar Handlin submitted the Immigration Commission Reports to a detailed critique in his Race and Nationality in American Life (Boston, 1957), Chapter 5Google Scholar. He concluded that the Reports' comments were largely dictated by an a priori assumption and “… not from any evidence—whatever that was worth; sometimes indeed they ran altogether against such evidence” (p. 103). Richard K. Vedder and Lowell E. Gallaway in “Settlement Patterns of American Immigrants, 1850–1968,” peripherally observe that their results indicate that “… the differences in behavior patterns of the various nationality groups do not seem to lend much credence to the distinction between ‘old’ and ‘new’ immigrant behavior,” in Fifth international Congress of Economic History, Papers, vol. III, Hermann Van der Wee, Vladimir A. Vinogradov, and Grigorii G. Katovsky, eds. (Moscow, 1976), p. 142.

⁹ A limit of ten nationalities was effectively imposed by considerations of computational manageability. The nationalities are treated individually rather than as aggregates of all new and all old immigrants since, as shown below, behavior across nationalities was sufficiently varied as not to warrant aggregation.

¹⁰ Data on income and employment are found in Everett S. Lee, A. R. Miller, C. P. Brainerd, and R. A. Easterlin, eds., Population Redistribution and Economic Growth, 1870–1950, Methodological Considerations and Reference Tables (Philadelphia, 1957), pp. 609–21, 753; the South classification is the census region; population and migrant stock data are available from the 1900 Census.

¹¹ These data are contained in the Annual Reports of the Commissioner-General of Immigration to the Secretary of the Treasury. Since these data cover only steerage arrivals, they have the advantage of economic homogeneity in addition to the grouping by nativity. The loss of observations on the settlement plans of the more affluent and of those who entered overland through Canada—primarily the British and Irish—may be, for those nativities, a serious shortcoming. For a discussion of these points see Richard K. Vedder and Lowell E. Gallaway, “The Geographical Distribution of British and Irish Emigrants to the United States After 1800,” Scottish Journal of Political Economy, 19 (Feb. 1972), 19–35; Gallaway, Vedder, and Shukla, “The Distribution …”; and James A. Dunlevy and Henry A. Gemery, “British-Irish Settlement Patterns in the United States: The Role of Family and Friends,” Scottish Journal of Political Economy, 24 (Nov. 1977), 257–63.

¹² For example, Richard K. Vedder and Lowell E. Gallaway, “The Settlement Preferences of Scandinavian Emigrants to the United States, 1850–1960,” Scandinavian Economic History Review, 18 (Fall 1970), 159–76, and Vedder and Gallaway, “British and Irish Emigrants,” have expressed a belief that migrants avoided the South because of an allegedly hostile social climate. Empirical support for this position, however, has been mixed.

¹³ An extensive literature has developed regarding the family and friends effect. See, for instance Nelson, Phillip, “Migration, Real Income and Information,” Journal of Regional Science, 1 (Spring 1959), 43–74CrossRefGoogle Scholar; Greenwood, Michael J., “Lagged Response in the Decision to Migrate,” Journal of Regional Science, 10 (Dec. 1970), 375–84CrossRefGoogle Scholar; and Levy, Mildred B. and Wadycki, Walter J., “The Influence of Family and Friends on Geographic Labor Mobility: An International Comparison,” Review of Economics and Statistics 55, (May 1973), 198–203CrossRefGoogle Scholar. The measure of family and friends, 1900 migrant stock, used in this study was in part generated by the 1898 flow, the dependent variable. The resultant misspecification, however, was not found to be serious. This problem is discussed in Dunlevy, James A. and Gemery, Henry A., “The Role of Migrant Stock and Lagged Migration in the Settlement Patterns of Nineteenth Century Immigrants,” Review of Economics and Statistics, 59 (May 1977), 137–44CrossRefGoogle Scholar.

¹⁴ The interpretation of the migrant stock variable has generated controversy with some claiming that it captures a lagged adjustment process rather than the family and friends effect. The main issues of this controversy and an empirical interpretation of the migrant stock and lagged migration flow are discussed within the context of late nineteenth-century U.S. immigration in Dunlevy and Gemery, “The Role of Migrant Stock … .” It is there that the recommendation to use the lagged variable is found.

¹⁵ The South dummy variable was not transformed to its logarithm. A discussion of the Zellner technique is found in Zellner, Arnold, “An Efficient Method of Estimating Seemingly Unrelated Regressions and Tests for Aggregation Bias,” Journal of the American Statistical Association, 57 (June 1962), 348–68CrossRefGoogle Scholar.

¹⁶ The basic model was also estimated using 1897 flow data. These (unreported) estimates do not differ qualitatively from those reported in Table 7. Preliminary tests of the model failed to reveal any heteroskedasticity, hence weighted least squares was not employed.

¹⁷ A similar conclusion based on a different methodology was drawn by Gallaway, Lowell E. and Vedder, Richard K., “The Increasing Urbanization Thesis—Did ‘New Immigrants’ to the United States have a Particular Fondness for Urban Life?,” Explorations in Economic History, 8 (Spring 1971), 305–19CrossRefGoogle Scholar. Further consideration of the similarity of response of the new and old immigrant nativities to this and other variables appears later in the text.

¹⁸ See Greenwood for consideration of points bearing on the decline of estimated coefficients when lagged adjustment is recognized.

¹⁹ These F-statistics are the squares of the more commonly used t-statistics associated with tests of equality between two random variables. Since the weights are the ex post relative numbers of immigrants, they are non-stochastic. The test is therefore that described by Theil, Henri, Principles of Econometrics (New York, 1971), p. 313Google Scholar.

²⁰ An erroneous alternative explanation is that earlier new immigrants had not as yet had time to disperse inland as much as had the majority of earlier old migrants. This explanation would attribute the measured pull of family and friends to the disequilibrium concentration of the new immigration migrant stock in eastern port states. This explanation, however, is largely discredited by the estimates of the Koyck lag model which allows for such disequilibrium. These estimates also support the hypothesis that the dependency of new immigrants on their migrant stock was stronger than was the dependence of old immigrants.

²¹ Gallaway and Vedder, “The Increasing Urbanization Thesis….”