Dominic A Smith

Economists aiming to understand the behavior of inflation often rely on robust measures of inflation, which exclude volatile expenditure categories from headline inflation to reduce noise. We evaluate the forecasting performance of a wide set of such measures between 1970 and 2024, including core, median, and trimmed mean personal-consumption-expenditure (PCE) inflation. Core inflation, that excludes a fixed set of categories, performs significantly worse than median and trimmed mean inflation, that continuously change the excluded categories based on current price changes. We also evaluate alternative trimmed mean measures to minimize their forecasting error relative to standard measures of trend inflation. However, we find no grounds to select any single trim on the basis of forecasting performance: A wide range of trims have an average prediction error that makes them statistically indistinguishable from the best-performing trim. Despite indistinguishable average forecasting performance, different trims imply different predictions for trend inflation in any given month, within a range of 0.5 to 1 percentage points. This range and its midpoint outperform all trimmed mean inflation measures in tracking trend inflation, suggesting the use of the range of inflation implied by the set of near-optimal trims as a valuable complement to any single inflation measure.

The retail sector has changed from a sector full of small firms to one dominated by large, national firms. We study how this transformation has impacted productivity levels, growth, and dispersion between 1987 and 2017. We describe this transformation using three overlapping phases: expansion (1980s and 1990s), consolidation (2000s), and stagnation (2010s). We document five findings that help us understand these phases. First, productivity growth was high during the consolidation phase but has fallen more recently. Second, entering establishments drove productivity growth during the expansion phase, but continuing establishments have increased in importance more recently. Third, national chains have more productive establishments than single-unit firms on average, but some single-unit establishments are highly productive. Fourth, productivity dispersion is significant and increasing over time. Finally, more productive firms pay higher wages and grow more quickly. Together, these results suggest that the increasing importance of large national retail firms has been an important driver of productivity and wage growth in the retail sector.

Official Bureau of Labor Statistics (BLS) estimates of productivity growth in the retail trade sector indicate that productivity has grown at a moderate rate of 2.8 percent per year between 1987 and 2017, and that there is considerable variation in growth rates across 4-digit industries. But the official data, which can be thought of as weighted averages of establishment-level productivity, tell us nothing about what goes on within industries. Given the transformation of retail trade over the past three decades, this information could provide more insight. In this paper, we present productivity dispersion statistics for industries in the retail trade sector. These statistics are similar to the BLS-Census Bureau Dispersion Statistics on Productivity (DiSP) for manufacturing industries and complement the official BLS industry-level productivity statistics. We find that from 1987 through 2017, productivity dispersion increased slightly on average. Surprisingly, the tails of the retail productivity distribution have similar dispersion as we find in the middle. Firm dispersion has increased more than establishment dispersion.

Increasing national concentration has contributed to market power concerns. Yet local trends are more informative about market power in retail, where consumers have traditionally shopped at nearby stores. Using novel product-level census data for all US retailers, we find that local and national concentration increased in parallel and affect most markets, products, and industries. The expansion of multimarket firms into new markets explains most of the increase in national concentration, with consolidation via increases in local market shares increasing in importance between 1997 and 2007. Increases in local concentration can explain one-quarter to one-third of the observed rise in retail gross margins.

U.S. import and export price indexes replaced unit value indexes forty years ago, given quality concerns of mismeasurement due to unit value bias. The administrative trade data underlying the unit values have greatly improved since that time. The transaction records are now more detailed, available electronically, and compiled monthly with little delay. The data are used by academic researchers to calculate price measures, and unit value indexes based on trade data are used by other national statistical offices (NSOs). The U.S. Bureau of Labor Statistics is now evaluating whether replacing price indexes with unit value indexes for homogeneous products calculated from administrative trade data could expand the number of published official import and export price indexes. Using export transactions, the research calculates detailed unit value indexes from 200 + million trade records from 2012–2017 for 123 export product categories. Results show that 27 of the 123 unit value indexes are homogeneous and closely comparable to published official price indexes. This article presents the concepts and methods considered to calculate and evaluate the unit value indexes and to select the product categories that are homogeneous. Compared to official price indexes, export unit value indexes for the 27 5-digit BEA (U.S. Bureau of Economic Analysis) end-use product categories would deflate real exports of these goods by 13 percentage points less over the period. Incorporating these 27 indexes into the top-level XPI would increase the value of real exports of all merchandise goods by 2.6 percentage points at the end of 2017.

The “factoryless manufacturing” (FM) business model is employed by a rising share of U.S. firms. Factoryless manufacturers outsource the fabrication of products but maintain control of the production process, own the associated intellectual property, and bear the entrepreneurial risk. FM is an important component in the role of U.S. firms in global manufacturing value chains. We estimate the scope of U.S. factoryless manufacturing using three approaches. First, we use financial reports for S&P 500 companies to show that FM is prevalent and increasing in the United States and that FM, once only common in the production of apparel, electronics, toys, and pharmaceuticals, has spread to a broader array of products. Second, we use Economic Census microdata to estimate that manufacturing value-added would have been 5 to 20 percent greater for 2007 if all FGPs were reclassified to manufacturing. Third, using a list of FM semiconductor companies matched to Economic Census microdata, we estimate that value-added would be 20 to 30 percent greater for semiconductor manufacturing, an industry where FM is especially prevalent, if FGPs were included. These results suggest that outsourcing and offshoring of product fabrication by U.S. firms is coupled with significant domestic production management. Thus, identifying FGPs in economic data is important for the study of fragmentation and globalization.

In this article, we estimate consumption of automobiles by using a user-cost approach and data from the Consumer Expenditure Surveys Interview Survey (CE). The user-cost approach is a method for valuing the flow of services from long-lived goods. A key input into user cost is the depreciation of the good, which we estimate by using purchase data from the CE.

U.S. import and export price indexes (MXPI) replaced unit value indexes forty years ago, given quality concerns of mismeasurement due to unit value bias. The administrative trade data underlying the unit values have greatly improved since that time as have index calculation methods and processing capabilities of computers. The transaction records are now more detailed, available electronically, and compiled monthly with little delay. Academic researchers use the administrative trade data to calculate unit values and study firms’ pricing decisions. Moreover, other national statistical offices (NSOs) produce unit value indexes based on administrative trade data. The U.S. Bureau of Labor Statistics will begin using these data for about 40 percent of the MXPI beginning in 2025. Years of research show that the characteristics available in the data allow detailed items to be constructed that mimic a matched-item model. In fact, we estimate that using the administrative trade data for the MXPI between 2017 and 2023 would have had little impact on both import and export price index levels, changing them by less than one percentage point.

Results in this paper have been moved into other projects.