迷失在聚合中：收入和支出的地理计量错误

Lost in Aggregation: Geographic Mismeasurementof Income and Spending☆Sinem Hacıo˘glu-Hoke∗Federal Reserve BoardCEPRLeo Feler†NumeratorJack Chylak‡NumeratorJuly 3, 2025AbstractUsing zip-code median income as a proxy for household income is common ineconomics but can mask heterogeneity and yield misleading conclusions.Usingzip-code median income and self-reported household incomes from a representa-tive panel of 150,000 U.S. households, we decompose average retail spending for2018-2024.When using self-reported incomes, we observe substantial divergencein spending between low- and high-income households starting in mid-2021. Whenusing zip-code aggregates as a proxy, this divergence disappears. Our findings indi-cate a 35 to 75 percent discrepancy between zip-code aggregates and self-reportedincomes, highlighting the limitation of zip-code aggregates as a proxy for householdincomes.Keywords: Spending, Income, Heterogeneity, Zip-code Average IncomeJEL Classification: E01, E2, E32∗sinem.haciogluhoke@frb.govWeb:www.sinemhaciogluhoke.com†leo.feler@numerator.comWeb:www.leofeler.com‡jack.chylak@numerator.com☆The authors would like to thank Tomaz Cajner,Ryan Decker,Aaron Flaaen,Chris Kurz and Michael Palumbo.The analysis and conclusions set forth herein are those of the authors and do not indicate concurrence by other membersof the research staff or the Board of Governors of the Federal Reserve System.The use of commercially provided datais for research purposes only and does not imply endorsement, recommendation, or favoring of any brand, product,service, or company by the Board of Governors or the Federal Reserve System. The views expressed in this paper are theresponsibility of the authors and should not be interpreted as reflecting the views of Numerator.All errors remain ourown. 1IntroductionIt is common in social sciences to categorize households as low-, middle- or high-income toassess the heterogeneity in, for example, consumers’ spending behavior. When householdincomes are not observed, a common practice is to use as a proxy the average or medianincome in zip codes where households live, generally obtained from population surveys.The reasons for such an approximation of household incomes using zip-code aggregates arebecause, first, where households live is correlated with their incomes, and second, it is rareto have access to datasets that provide information on disaggregated household incomes.In this paper, we assess whether zip-code aggregates are a valid proxy for disaggregatedhousehold incomes that accurately reflects heterogeneity in consumer behavior, usinga representative panel of U.S. households with disaggregated information on householdincomes.This paper has two objectives that lead to novel contributions to existing studies thatdocument heterogeneity in spending patterns. First, using a detailed micro dataset, weconstruct a measure of real average retail spending for low-, middle- and high-incomehouseholds using households’ self-reported incomes to construct income groups. Second,we use our micro-data to test the implications of using zip-code aggregates.Our re-sults indicate that using zip-code aggregates as a proxy for household incomes masksheterogeneity in consumer behavior and leads to misleading conclusions about changesin consumer behavior during the post-pandemic period.We use a panel dataset of 150,000 representative U.S. households from Numerator,a consumer data and survey company. Numerator obtains physical and online receiptsfrom this rolling static panel of U.S. households for whom we have detailed informationon household attributes. We first compare our overall measure of monthly average retailspending between 2018 and 2024 against the Census Bureau’s Advance Monthly Sales forRetail and Food Services report (MARTS) to establish our measure’s reliability.Having established that the aggregated Numerator retail spending series closely matchesthe Census Bureau’s retail sales series, we examine spending patterns based on house-hold income. We analyze the average monthly household retail spending for low-incomehouseholds ($0-$60K in annual household income), for middle-income households ($60K- 2 $100K), and for high-income households ($100K+) from 2018 to 2024. Published mea-sures do not provide details onwhichconsumers’ spending has remained resilient in thepost-pandemic period, and hence fall short of identifying vulnerabilities in the economyoriginating from specific groups, so our analysis fills this gap. Our results suggest thatretail spending growth evolves similarly for all households before the pandemic. However,starting in mid-2021, the spending of high-income households diverges from the spend-ing of low- and middle-income households.High-income households continue to spendstrongly while low- and middle-income households’ spending lags behind. This finding issupported by many analyses and can be rationalized by low-income households depletingtheir pandemic-era savings (Abdelrah