非同质需求转移与通货膨胀不平等

Federal Reserve Board, Washington, D.C.ISSN 1936-2854 (Print)ISSN 2767-3898 (Online) Non-homothetic Demand Shifts and Inflation Inequality Jacob Orchard 2025-085 Please cite this paper as:Orchard, Jacob (2025). “Non-homothetic Demand Shifts and Inflation Inequality,” Financeand Economics Discussion Series 2025-085. Washington: Board of Governors of the FederalReserve System, https://doi.org/10.17016/FEDS.2025.085. NOTE: Staff working papers in the Finance and Economics Discussion Series (FEDS) are preliminarymaterials circulated to stimulate discussion and critical comment.The analysis and conclusions set forthare those of the authors and do not indicate concurrence by other members of the research staff or theBoard of Governors. References in publications to the Finance and Economics Discussion Series (other thanacknowledgement) should be cleared with the author(s) to protect the tentative character of these papers. Non-homothetic Demand Shifts and Inflation Inequality Jacob Orchard Federal Reserve Board First Version: February 2021This Version: August 2025 Abstract This paper shows that adverse macroeconomic shocks systematically increase in-flation for low-income households relative to high-income households. I document twokey facts: (i) during every U.S. recession since 1959, aggregate spending shifts towardproducts disproportionately purchased by low-income households (necessities); and (ii)relative prices of necessities rise during recessions. These patterns can be explained bya model with non-homothetic demand and a concave production possibility frontier:shocks that reduce expenditure induce households to reallocate spending from luxuriesto necessities, raising their relative prices.I empirically show that this mechanismoperates for both demand and supply shocks, using monetary policy and oil price newsshocks.Incorporating this mechanism into a quantitative model reproduces most ofthe variation in necessity prices and shares from 1961 to 2024. The model shows thatthe fall in expenditure due to a recessionary shock similar to the Great Recession leadsinflation to increase by more than 1.5 percentage points for low-income householdsrelative to high-income households.The results suggest that low-income householdsare hit twice by adverse shocks: once by the shock itself and again as their price indexincreases relative to that of other households. JEL Classification: E30, D12KEYWORDS: inflation, non-homotheticity, real income inequality, business cycle Recent literature has established that over the past several decades, inflation has beenhigher for low-income households than for high-income households.1 In this paper, I showthat the gap in inflation between low- and high-income households (inflation inequality)varies over time and increases following adverse macroeconomic shocks. Shocks that loweraggregate expenditure also induce households to reallocate spending from luxuries to neces-sities, raising the relative price of necessities, which disproportionately affects low-incomehouseholds because these products occupy a larger share of their budgets. This paper makes three main contributions. First, I show empirically that necessities—product categories with a larger budget share for low-income than high-income households—have countercyclical relative demand. This holds in the aggregate time series, where the shareof necessities rises in every recession since 1959, and following plausibly exogenous shocksthat reduce aggregate expenditure. Second, I show that these same product categories alsohave countercyclical relative prices. In the aggregate time series, necessity relative prices arepositively correlated with unemployment and other indicators of economic contraction, andthey increase in response to shocks that lower aggregate expenditure.Third, I develop aframework with non-homothetic demand and a concave production possibility frontier thatrationalizes these facts. Embedding this mechanism in a standard business cycle model showsthat it explains most of the variation in relative necessity prices and expenditure shares since1961 and that non-homotheticity materially amplifies inflation inequality during recessions,including by over 1 percentage point in the Great Recession, the COVID-19 recession, andthe Volcker recessions. In order to study differences in household-level price indices across time, I match prod-ucts in the BEA’s Personal Consumption Expenditures (PCE) with equivalent spending inthe Consumer Expenditure Survey resulting in 148 product sectors for which I have aggre-gate expenditure and price data, alongside household-level purchasing patterns.I definenecessities as products where low-income households have a higher budget share than high-income households. Although this definition is based on household-level data, these productsalso behave as necessities in the aggregate: their expenditure share increases during everyrecession since 1959. Figure 1 shows the inflation