Since the nature of price adjustment plays a crucial role in determining how monetary and fiscal policies affect the economy, a longstanding literature in macroeconomics has focused on trying to document the nature of price dynamics. In recent years, major advances in this literature have arisen from the growing availability of data on prices. One important new source of such data is the research database of prices collected by the U.S. Bureau of Labor Statistics (BLS) to construct price indexes such as the CPI and PPI. The PI was among the first academic researchers to use these datasets and have for several years been involved in developing and analyzing these data. In this Career Proposal, the PI proposes several projects to study issues related to price dynamics. A major component of the proposal involves the creation of new variables that will allow the data to be used in new ways. The PI proposes the creation of standardized product identifiers at the level of individual brands/model numbers or UPC codes that can be used to link the CPI and PPI research databases. The PI also proposes the creation of new links within the PPI research database between upstream and downstream producers, as well as new product characteristic variables. This new linked dataset will make it possible to follow the full chain of price responses resulting from an initial change in upstream costs, through subsequent price adjustments by downstream manufacturers, to any eventual changes in the final retail price, allowing for a broad-based analysis of how firms respond to movements in costs. It will also allow for new analyses of other issues related to price dynamics, including strategic pricing interactions, price dispersion, and hedonic demand models. The unique product identifiers will also allow the PI to link the BLS data on domestic prices with the corresponding data on international prices, making it possible to develop improved estimates of how off-shoring affects the US economy. An important challenge in constructing measures of US input costs is that if a US firm switches from purchasing its inputs from a domestic to a foreign supplier, the price change that occurs at the time of the switch will not be recorded in the BLS data (essentially, it is dropped as the product transitions from the PPI to the IPP database). As a consequence, associated declines in the cost of US firms-inputs may simply be lost in transit-yielding incorrect estimates of the value-added of US manufacturing firms. Recent evidence suggests that this measurement problem may lead to a substantial overestimation of recent US productivity growth. The new linked database will allow me to develop new estimates of how outsourcing affects the US economy that address this measurement challenge. Finally, the PI proposes four related projects aimed at analyzing nominal rigidities using other sources of data. The first project focuses on studying how prices react to underlying movements in costs using firm data that include prices at multiple stages of production, and a measure of underlying production costs. The second project proposes to study how consumer shopping behavior varies over the business cycle. The third project proposes to study wage rigidity using data from a large payroll firm. The fourth project studies how the terms of credit contracts for new vs. continuing customers adjust to large macroeconomic shocks, such as the recent financial crisis, and the implications of this evidence for understanding the linkages between the financial sector and the real economy. Broader Impact: The proposed research is relevant to policymakers such as the US Federal Reserve and Treasury, who employ macroeconomic models of monetary and fiscal policy in which price adjustment plays a central role. The effects of outsourcing on the US economy, and the sources of recent productivity growth, also have important policy implications that have been emphasized in recent high profile articles in the business press. Finally, this project involves creating a new linked dataset that will be made available to the academic community through the BLS outside researcher program. The PI's earlier research on this topic has been published in top academic journals, and it is to be expected that she will continue to disseminate her work in this way. Her work has also been disseminated to the broader public through articles in major news outlets such as The Economist and BusinessWeek.
