¹ For a more extensive discussion of the antebellum period see my “The American Common School Revival: Fact or Fancy?” in Rosovsky, H., ed., Industrialization in Two Systems: Essays in Honor of Alexander Gerschenkron (New York: Wiley, 1966).Google Scholar

² A complete disaggregation of private education in 1890 may be found in the U.S. Census of Schools at that date. The 1870 census categories differentiate for the first time between academies and other private schools and hence permit the comparison made here.

³ Bureau, U.S. of the Census, Statistical Abstract of the United States, 1964 (Washington, D.C.: Government Printing Office), p. 107Google Scholar , for 1960 private and public educational expenditures. Of the private elementary school enrollment of 4,286,000 and a secondary total of 1,035,000 in that year, Catholic Church officials claim 4,260,000 and 828,000 respectively. See The World Almanac, 1961, p. 706.

⁴ Schultz, Theodore W., “Capital Formation by Education,” Journal of Political Economy, LXVIII, No. 6 (12. 1960), pp. 571–83.CrossRefGoogle Scholar

⁵ By which I mean the average number of days actually attended per enrolled pupil.

⁶ Our technique, without adjustment for labor force participation, implies foregone earnings for college and secondary students of $80 million against Schultz's $105 million. If all students were engaged in nonagricultural activities, the resulting estimate is slightly greater than Schultz's.

⁷ As calculated from Bureau, U.S. of the Census, Historical Statistics of the United States (Washington, D.C.: Government Printing Office, 1960), p. 72.Google Scholar

⁸ From the individual standpoint, of course, such legislation removes the option of alternative earnings as a factor in the decision to attend school.

⁹ Report of the Commissioner of Education for 1910, II, xxv. Note also that in Massachusetts until 1889, poverty of parent remained a valid reason for child employment rather than school attendance; Ensign, Forest C., Compulsory School Attendance and Child Labor (Iowa City: Athens Press, 1921), p. 69.Google Scholar

¹⁰ See Stigler, George, Employment and Compensation in Education (New York: National Bureau of Economic Research, 1950), pp. 67–70Google Scholar ; and Ensign, for individual state studies.

¹¹ William Mather, a member of the British Royal Commission in Technical Education, quoted in a special report of the Bureau, U.S. of Education, Educational Exhibits and Conventions at the World's Industrial and Cotton Centennial Exposition (Washington, D.C., 1886), Part II, p. 121.Google Scholar

¹² The relevant equations, with 20 observations each, are as follows:

where En Rt is enrollment relative to population five to eighteen years of age; Y/P, income per capita; AgY/Y, the proportion of income emanating from agriculture; Att. Rt., the actual days attended relative to popultaion 5-18; and Ex/pupil, average expenditures per enrolled student. The educational data derive from the Commissioner's Report; income data per capita and by source are from Easterlin, Richard A., “Interregional Differences in Per Capita Income, Population, and Total Income, 1840-1950,” in Trends in the American Economy in the Nineteenth Century (Princeton, N.J.: Princeton University Press for NBER, 1960).Google Scholar

The other regions were excluded in order to emphasize the positive relationship of enrollment to the proportion of farm income, followed by sign reversal for expenditures, in a set of states similar in terms of attitudes and educational history. The low enrollment rates in the South would have obscured this relationship while intensifying the negative rural effect upon expenditure and length of school year.

¹³ Calculations are based upon data from the Commissioner's 1900 and 1901 reports. Inclusion of the South would exaggerate this urban-rural difference by a considerable margin.

¹⁴ The basic regression equation from which these estimates were derived is based upon 56 national observations pertaining to 1958. Expenditures per capita were related to income per capita and to the proportion of the population five to nineteen years in the following way:

Incomes per capita in 1860 dollars from Gallman were converted to 1958 dollars by the wholesale price index prior to estimation. Estimated expenditures in 1958 dollars were calculated in 1860 dollars and then converted to current dollars by Gallman's implicit price deflator.

¹⁵ The educational expenditures for Germany from Hoffmann, while labeled public, seem to be more than ample to encompass total outlays; indeed, reference to the sources used for France and Great Britain suggests they may be too large even for aggregate expenditures. Note as well that use of net national product exerts an upward bias on the German ratio. Conversely, the estimate for French educational expenditures may be understated due to apparent exclusion of capital account. More generally, these data are to be regarded only as approximate both because of deficiency in the original sources and of failure to convert to fully comparable dollars.

¹⁶ These expenditures and enrollment ratios derive from the data underlying Table 4. For convenience, the three components required to calculate the expenditure ratio in 1900 are as follows:

¹⁷ Calculated from Historical Statistics, pp. 208-9, and the 1900 Report of the Commissioner, I, 765 ff. For cash equivalent of housing privileges, see 1899 Report, I, p. 136.

¹⁸ Taken for France and England from the 1900 Report of the Commissioner, II, 2618; Germany estimated from information in 1903 Report, I, 651; U.S. calculated from Historical Statistics, pp. 207-9, after adjustment for public secondary pupils.

¹⁹ See 1903 Report, I, 249 and 591, for retrospective reports on Great Britain and France.