Sources and Techniques Used in the Construction of Annual GDP, 1790-Present

by

Louis Johnston
Department of Economics
College of Saint Benedict at Saint John's University
LJohnston@csbsju.edu

and

Samuel H. Williamson
President of MeasuringWorth
and
Department of Economics, emeritus
The Miami University
sam@mswth.com

This note describes the sources that were used and the techniques applied to those sources in constructing these series. There is no doubt that the accuracy of these series varies, and can be different for each of the series and for the different periods. In most cases the earlier the period, the less accurate are the observations. One reason for this is simple: GDP data were not collected or even defined before the 1930s and thus any measures for years before 1932 rely on sources that were not collected for the purpose of constructing national income and product accounts.

The construction of the GDP series before 1909 starts with an estimate of the observation for each benchmark year. The benchmarks are either the census or calendar year that comes every ten years beginning in 1800. The values for the years between benchmark years are computed by interpolation. Part of this interpolation is based on annual observations of related series, and part is based on an assumption of constant growth.

The data for these benchmark years can be assumed to be more accurately measured than the nine years between each of them. Because of the method of construction, we do not recommend the data be used for sophisticated time series analysis.

The data from 1790 to 1868 are for census years, and from 1869 to the present for calendar years. Each census year includes the first five months of the following calendar year. The reason is that most of the data used are based on information from the censuses that were collected every ten years. The early census data were collected from June 1 of years ending in 9 through May 31 of years ending in 0. Thus the pre 1869 benchmark years are based on these twelve months.

I. Benchmark Years 1790 to 1909

1793-1829

Benchmarks observations (real 1840 dollars) are from McCusker (2000) for 1793 and Weiss (1993) for 1799, 1809, 1819, and 1829. We are using Weiss's narrow definition of GDP.

1839-1859

Benchmark observations are based on Gallman's (1966) GNP numbers (real 1860 dollars and nominal) for 1839, 1849, and 1859. His nominal GNP numbers are adjusted for the flow of services to consumers using data from Weiss (1975). They are also adjusted for the flow of government purchases using data from Weiss (1975) and Trescott (1960), and NFI from North (1960). The results are the nominal GDP benchmarks.

Gallman's real GNP numbers are adjusted in the same method after the Weiss and Trescott data are deflated by Gallman's GNP deflator and NFI by the terms of trade from North (1961). The results are the real GDP benchmarks.

1869-1909

Benchmark observations are based on Gallman's (1966) GNP numbers (real 1860 dollars and nominal) for 1869, 1879, 1889, 1899 and 1909. His nominal GNP numbers are adjusted for the flow of services to consumers using data from Weiss (1975) and for the flow of consumer durables from Olney (1989). They are again adjusted for government purchases, in 1869 using the same manner as above, for 1879 to 1899 using data from Weiss (1975) and Kendrick (1961), and for 1909, from Kendrick (1961). NFI are from Simon (1960). The results are the nominal GDP benchmarks.

Gallman's real GNP numbers are adjusted in the same method after the Weiss, Trescott, Kendrick and Olney data are deflated by Gallman's GNP deflator and NFI by the terms of trade from Simon (1960). The results are the real GDP benchmarks.

II. Annual observations 1790 to 1908

For 1790 to 1868, agricultural output, the value of shelter, government purchases and net factor income are subtracted from the real GDP benchmarks described above. Annual observations of the residual are created by interpolation using Davis (2004). The annual real GDP is the sum of this residual, federal government purchases and net factor income, and computed values of agricultural and shelter components that are assumed to grow at constant rates between each of the benchmarks years.

For 1869 to 1909, agricultural output, the value of shelter, government purchases, net factor income and consumer durables purchases are subtracted from the real GDP benchmarks described above. Annual observations of the residual are created by interpolation using Davis (2004). The annual real GDP is the sum of this residual, federal government purchases, net factor income and consumer durables purchases, and computed values of agricultural and shelter components that are assumed to grow at constant rates between each of the benchmarks years.

Nominal observations are computed by the product of these real observations and the GDP deflator described below.

III. Annual observations 1909 to Present

1909 to 1928: Kendrick (1961), Table A-III

1929 to present: Bureau of Economic Analysis at http://www.bea.gov.

IV. Implicit GDP deflator

1790 to 1909: The benchmark year deflators are the ratio of the nominal to real values of GDP that we computed. The annual values for the years between these benchmarks are interpolated using David and Solar (1977.)

1909 to 1929: Kendrick (1961), Table A-II.

1929-present: Bureau of Economic Analysis at http://www.bea.gov.

V. Population (in millions)

1790 to 1928: Carter, et al. (2006), Series Aa9

1929 to present: Bureau of Economic Analysis at http://www.bea.gov.

VI. Nominal GDP per capita

Nominal GDP divided resident population.

VII. Real GDP per capita

Real GDP divided by resident population.

References

Carter, Susan B., Scott Sigmund Gartner, Michael R. Haines, Alan L. Olmstead, Richard Sutch,and Gavin Wright, editors. Historical Statistics of the United States, Millennial Edition. New York: Cambridge University Press, 2006

David, Paul A. and Peter Solar. "A Bicentenary Contribution to the History of the Cost of Living in America." Research in Economic History 2 (1977): 1-80.

Davis, Joseph H. "An Annual Index of U. S. Industrial Production, 1790?1915." Quarterly Journal of Economics 119 (November 2004): 1177-1215.

Gallman, Robert E. "Gross National Product in the United States, 1834-1909." In Output, Employment, and Productivity in the United States After 1800, Dorothy S. Brady, editor, 3-76. New York: Columbia University Press (for NBER), 1966.

Kendrick, John W. Productivity Trends in the United States. Princeton: Princeton University Press (for NBER), 1961.

McCusker, John J. "Estimating Early American Gross Domestic Product." Historical Methods 33 (Summer 2000): 155-162.

North, Douglass C. "The United States Balance of Payments, 1790-1860," in Trends in the American Economy in the Nineteenth Century, William N. Parker, editor, 573-627. Princeton: Princeton University Press (for NBER), 1960.

North, Douglass C. The Economic Growth of the United States, 1790-1860. New York: Norton, 1961.

Olney, Martha L. "Consumer Durables in the Interwar Years: New Estimates, New Patterns." Research in Economic History 12 (1989): 119-150.

Simon, Mathew. "The United States Balance of Payments, 1861-1900," in Trends in the American Economy in the Nineteenth Century, William N. Parker, editor, 629-711. Princeton: Princeton University Press (for NBER), 1960.

Trescott, Paul B. "The United States Government and National Income, 1790-1860," in Trends in the American Economy in the Nineteenth Century, William N. Parker, editor, 337-361. Princeton: Princeton University Press (for NBER), 1960.

Trescott, Paul B. "Federal Government Receipts and Expenditures, 1861-1875"

Journal of Economic History 26 (June 1966): 206-222.

Weiss, Thomas. "Estimates of Gross Domestic Product for the United States, 1800 to 1860." Mimeo (University of Kansas) 1993.

Weiss, Thomas. The service sector in the United States, 1839 through 1899. New York: Arno Press, 1975.

Please read our Note on Data Revisions.