GDP: "Data for 1869 to 1888 from Kuznets (1961), T. R-25, pp." "561-562, col. 3, GNP, Variant III, five-year centred moving" "average and T. R-23, pp. 557-558, col. 1, B. Variant III," "billions of dollars. Using annual data from 1889-1892, the" five-year moving average number for 1890 is used to deduce the level of GNP in 1888 as five times the five-year centred "moving average for 1890 less the actual levels for 1889-1892. Using a similar procedure, the annual data for 1869-1888 are" derived by disaggregating the five-year moving averages. Data "for 1889 to 1928 are from Kendrick (1961), T. A-IIb, col. 11," "pp. 296-297, Gross National Product, Commerce concept," millions of dollars. Data from 1929 to 1945 are from United "States, Bureau of the Census (1975), Series F47, Gross" "National Product, current prices, p. 229. GDP data are" calculated by subtracting net income from investments abroad "and, net unilateral transfers from abroad from the figures for" GNP. The data on income on investments abroad are from United "States, Bureau of the Census (1975, series U13, p. 865 then" "series U5, U6, and U13, p. 864), income on investments abroad," private and public (U5 + U6) less income on foreign investments in US (U13). The data on unilateral transfers are "from United States, Bureau of the Census (1975, series U16 and" "U17, pp. 866-867), unilateral transfers, net private (U16)" "plus unilateral transfers, net public (U17). " Capital Formation: Data for 1869 to 1888 from Kuznets (1961). Kuznets’ figures on capital formation include net changes in claims against "foreign countries, so we subtract changes in those claims to" "get our own figure for gross capital formation. Thus, our" "capital formation figures are calculated as Kuznets (1961, T." "R-29, pp. 572-574, col. 1), gross capital formation, five-year" "centered moving average, less col. 3 from T. R-34, pp." "599-600, net changes in claims against foreign countries, five-year centred moving average. Annual data are derived from" the five-year centred moving average using the same procedure that was applied to the GNP series. Data from 1889 to 1928 are "from Kendrick (1961, T. AIIb, col. 7, pp. 296-297), gross" "private domestic investment, Commerce basis, millions of" "dollars. Data from 1929 to 1945 are from United States, Bureau" "of the Census (1975, Series F52, p. 229), gross private" "domestic investment, current prices. Data on stocks for 1869" "to 1888 from Kuznets (1961, T. R-34, pp. 599-600, col. 1), net" "changes in inventories, current prices, five-year centered" "moving average, and T. R-4, p. 490, col. 3, net changes in" "inventories, billions of dollars. Annual data are derived from" the five-year centred moving average using the same procedure "that was applied to the GNP series, except the data are" unscrambled from 1920 backwards. Data from 1889 to 1928 are "from Kendrick (1961, T. AIIb, col. 8, pp. 296-297), change in" "business inventories, millions of dollars. Data from 1929 to" "1945 are from United States, Bureau of the Census (1975," "Series F60, p. 230), total net change in business inventories," current prices. Current Account: "Data for 1869 to 1945 from United States, Bureau of the Census" "(1975), balance on goods and services, series U15, pp." "866-868, plus series U16 and U17, net private and government" unilateral transfers. Data from 1869 to 1899 are for fiscal year ended June. Data are converted to calendar year basis by adding each year to subsequent year and dividing by two. For "instance, fiscal year data for 1869 are added to fiscal year" "1870 and divided by two, yielding calendar year 1869. Data" from 1900 are for calendar year. Data to 1873 include exports and imports of gold. Data from 1874-1945 include nonmonetary gold exports. Data for nongold current account for 1869-1873 are calculated by taking the current account balance less net exports of gold. Data for nongold current account for 1874-1945 are calculated by taking the current account balance less net exports of gold less change in the monetary gold stock (equivalent to subtracting nonmonetary gold exports). Gold: Exports: Data for 1869 to 1914 are from NBER Macrohistory "Database, series 14112 net gold exports, thousands of dollars," monthly data. Annual data are derived by adding the sum of monthly net exports for each year (fiscal or calendar). Data "from 1915 to 1945 from United States, Bureau of the Census" "(1975), series U197 less U198, gold exports less imports, pp." "884-885, calendar year, millions of dollars. " Monetary gold stock data for 1869 to 1878 are from United "States, Bureau of the Census (1975, series X417, p. 993)," "billions of dollars, annual average. Monthly data from 1879 to" "1945 are supplied by the NBER Macrohistory Database, series" "14076, billions of dollars. To construct our own annual data" we select the December level of the monetary gold stock. Refs: "Kendrick, John W. (1961), Productivity Trends in the United States," "National Bureau of Economic Research Number 71, General Series," "Princeton University Press, Princeton NJ." "Kuznets, Simon (1961), Capital in the American Economy Its Formation and Financing" National Bureau of Economic Research Studies in Capital Formation and Financing "Number 9, Princeton University Press, Princeton NJ." "US Bureau of the Census (1975)," "Historical Statistics of the United States, Colonial Times to 1970, Bicentennial" "Edition, US Government Printing Office, Washington DC."