Comparing Estimates of Fixed Investment in Nonresidential Structures and Equipment

The U.S. Bureau of Economic Analysis (BEA) has relied mainly on the U.S. Census Bureau’s monthly Value of Construction Put in Place Survey (VIP) to measure levels of fixed investment in structures; the Annual Survey of Manufactures (ASM, with the commodity-flow method) to measure levels of fixed investment in equipment; and the Annual Capital Expenditures Survey (ACES) to estimate shares of fixed investment by industry. The estimated levels of fixed investment in nonresidential structures and equipment from these different surveys—with adjustments for known issues—should be similar. This paper compares estimated levels of investment in nonresidential equipment and structures from the ACES and from the other sources, using more than 20 years of published data. The results show that the estimates of fixed investment based on the ACES and the other sources are roughly similar and display similar cyclical trends. Nevertheless, some discrepancies between these estimates are worth noting. The ACES-based estimates of investment in structures are relatively higher in several years—as much as 20 percent higher—but not in all years. The ACES-based estimates of investment in new equipment are about 14–33 percent lower than the ASM/commodity-flow-based estimates. Despite these discrepancies, which could arise for many reasons, the rough similarity of the estimates from these surveys is reassuring and can be interpreted as generally supportive of BEA’s estimation methods and the value of these surveys.