Aggregation Issues in Integrating and Accelerating Bea's Accounts: Improved Methods for Calculating GDP by Industry

Abstract

Aggregate measures of real GDP growth obtained from the GDP by Industry Accounts often differ from the featured measure of real GDP growth obtained from the National Income and Product Accounts (NIPAs). We find that differences in source data account for most of the difference in aggregate real output growth rates; very little is due to the treatment of the statistical discrepancy, differences in aggregation methods, or the contributions formula. Moreover, we demonstrate that with consistent data, use of BEA's Fisher-Ideal aggregation procedures to aggregate value added over industries yields the same estimate of real GDP as aggregation over final commodities. Thus, two major approaches to measuring real GDP -- "expenditures" approach used in the NIPAs and the "production" or "industry" approach used in the Industry Accounts -- give the same answer under certain conditions. This result enables us to show that the "exact contributions" formula that the NIPAs use to calculate commodity contributions to change in real GDP can also be used to calculate consistent industry contributions to change in real GDP. We also find that using some newly developed datasets would help to bring the aggregate real output measures into closer alignment.