Real-Time Insights into Consumption Growth Inequality since the Onset of COVID-19

December 03, 2024
SHARE THIS PAGE:

The recovery in consumption since the onset of COVID-19 has revealed stark contrasts across population groups. In this blog post, we outline a new methodology that provides real-time estimates of monthly real expenditures for U.S. households with different levels of spending. We found that during the early recovery years, households with higher levels of expenditure experienced much faster consumption growth. However, since early 2023, consumption growth has become strikingly similar across all groups of households.

Our new methodology combines monthly data from the Bureau of Economic Analysis and the Bureau of Labor Statistics (BLS) to produce detailed consumption estimates by quintiles, in which households are equally distributed into five groups based on the level of their spending adjusted by the size of the household (technically, spending by “equivalent adult”).To remove the effect of the number of family members on the amounts of expenditures, households are split into five quintiles based on their equivalized expenditures, computed as the total expenditures per household divided by the square root of the number of family members. Monthly personal consumption expenditures (PCE) and consumer expenditure surveys (CE) are central to this approach.

Grouping Households by Spending Levels

A 2024 study by Thesia I. Garner, Robert S. Martin, Brett Matsumoto, and Scott Curtin organizes households into five expenditure-based quintiles and determines their spending shares across 16 categories aligned with PCE classifications.See the Bureau of Labor Statistics webpage “Distribution of U.S. Personal Consumption Expenditures Using Consumer Expenditure Surveys Data: Methods and Supplementary Results” for details on their work. Households at varying consumption levels buy different baskets of goods and services, leading to significant variation in expenditure shares by quintile. Assuming these shares remain relatively stable over time, our methodology assigns monthly expenditures in each PCE category to specific quintiles. Summing these expenditures across categories yields total expenditures for each quintile.

However, the most recent PCE shares published by Garner and her colleagues extend only through 2022, posing a challenge. To address this, we created a forecasting model that integrates estimated quintile expenditure shares from CE data to project PCE shares up to 2024 using the latest available CE data—currently the first quarter of 2024.Details on our methodology will be available in a forthcoming St. Louis Fed working paper. With our model, we can make timely estimates without waiting for lagged CE data.

Disparities in Real Consumption Growth after the Onset of COVID-19

Our findings reveal significant shifts in consumption growth across quintiles during different phases of the recovery. Understanding such consumption patterns provides insight into the economic challenges facing different income groups. Though our focus is on spending, household rankings of consumption and income have a strong correlation.Using the CE survey from the second quarter of 2023, the rank correlation between income per equivalent adult and consumption per equivalent adult was 0.70. To read more about the relationship between consumption and income, see our 2024 Economic Synopses essay “Uneven Consumption Growth in the COVID-19 Economic Recovery.”

The table below presents estimates of annualized, seasonally adjusted real consumption growth from January 2020 to January 2021 (left column), January 2021 to January 2023 (middle column), and January 2023 to the last available observation, October 2024 (right column).

Inequality in Annual Real Consumption Growth since January 2020
Groups Sorted by Household Spending Level Uneven Recovery Accelerated but Unequal Growth Broad-Based Recovery
January 2020 to January 2021 January 2021 to January 2023 January 2023 to October 2024
Total -0.1% 4.4% 2.4%
Quintile 1 -0.8% 1.9% 2.1%
Quintile 2 -2.2% 4.3% 2.0%
Quintile 3 -1.2% 3.9% 2.2%
Quintile 4 -1.2% 5.7% 2.1%
Quintile 5 1.8% 4.7% 2.8%
SOURCES: Bureau of Economic Analysis, Bureau of Labor Statistics and authors’ calculations.
NOTES: Growth rates are annualized and seasonally adjusted. Spending levels per household are adjusted for size; see Endnote 1 for more information. Quintiles are sorted from lowest to highest, with Quintile 1 being the group with the lowest-spending households.

The left column shows that by January 2021, national consumption levels had nearly returned to their January 2020 levels (-0.1% real growth). However, the households in the highest quintile (those with the highest level of expenditure) experienced a substantial recovery with 1.8% growth, surpassing pre-pandemic levels. In contrast, the recovery in the quintile with the second-lowest household spending lagged significantly behind, with consumption declining by 2.2%. This period highlights sharp disparities in consumption growth across quintiles.

The middle column indicates that from January 2021 to January 2023, consumption increased across all quintiles. Yet, growth at the bottom of the distribution was notably slower. The households in the lowest expenditure quintile (Quintile 1) saw a 1.9% increase, while all other quintiles grew at more than twice that rate, underscoring a persistent gap.

Finally, the right column shows that since January 2023, consumption growth has been strong across all quintiles. Differences between quintiles have narrowed considerably, with all groups growing at rates close to the aggregate 2.4%, signaling a more uniform recovery phase.

Real Consumption Growth Is Now Stabilizing across Quintiles

The trajectory of real consumption recovery since the onset of COVID-19 reveals three distinct phases. Early on, the highest-spending households rebounded quickly, while lower-spending households faced significant challenges. In the subsequent period, growth accelerated across all quintiles, though disparities remained, with slower recovery at the bottom of the distribution. More recently, the recovery has become broad-based, with consumption growth stabilizing and aligning across quintiles. These trends highlight both the uneven initial impact of the pandemic and the gradual convergence in recovery as time progressed.

Notes

  1. To remove the effect of the number of family members on the amounts of expenditures, households are split into five quintiles based on their equivalized expenditures, computed as the total expenditures per household divided by the square root of the number of family members.
  2. See the Bureau of Labor Statistics webpage “Distribution of U.S. Personal Consumption Expenditures Using Consumer Expenditure Surveys Data: Methods and Supplementary Results” for details on their work.
  3. Details on our methodology will be available in a forthcoming St. Louis Fed working paper.
  4. Using the CE survey from the second quarter of 2023, the rank correlation between income per equivalent adult and consumption per equivalent adult was 0.70. To read more about the relationship between consumption and income, see our 2024 Economic Synopses essay “Uneven Consumption Growth in the COVID-19 Economic Recovery.”
ABOUT THE AUTHORS
Juan M. Sánchez

Juan M. Sánchez is an economist and senior economic policy advisor at the Federal Reserve Bank of St. Louis. He has conducted research on several topics in macroeconomics involving financial decisions by firms, households and countries. He has been at the St. Louis Fed since 2010. View more about the author and his research.

Juan M. Sánchez

Juan M. Sánchez is an economist and senior economic policy advisor at the Federal Reserve Bank of St. Louis. He has conducted research on several topics in macroeconomics involving financial decisions by firms, households and countries. He has been at the St. Louis Fed since 2010. View more about the author and his research.

Masataka Mori

Masataka Mori is a research associate at the Federal Reserve Bank of St. Louis.

Masataka Mori

Masataka Mori is a research associate at the Federal Reserve Bank of St. Louis.

This blog offers commentary, analysis and data from our economists and experts. Views expressed are not necessarily those of the St. Louis Fed or Federal Reserve System.


Email Us

Media questions

All other blog-related questions