The purpose of this project is to create and analyze new estimates of gross national product for the U.S. for the period 1834-1928. As a first step, the study will involve an extensive analysis of the accuracy of current GNP estimates for that period. Preliminary research suggests that the estimates greatly exaggerate the size of business cycles because very volatile production series are used to interpolate between benchmark GNP estimates. This research is important because it will provide new insights into the nature and causes of business cycles both in the past and in the modern era. Furthermore, this study will also provide a better understanding of how the economy has changed over time and how those changes affect macroeconomic performance. In order to deal with the excess volatility in the existing data series, regression procedures will be used to convert available fragments of production data into new estimates of gross national product. Explicitly accounting for the differences in volatility between GNP and the interpolating series will yield estimates of GNP for the 19th and 20th centuries that represent cyclical movements much more accurately. This new data series will be used to measure the size of business cycle fluctuations and to estimate the cyclical behavior of prices over this period.
