Income inequality and household debt

Published: 9 May 2024| Version 1 | DOI: 10.17632/wnx3n8vfch.1
Chu-Ping Vijverberg


The data set consists of a 1999-2015 state-year balanced panel of forty-eight states of the U.S. Except for Alaska, Hawaii and the District of Columbia, all U.S. states are included. Inequality is measured in four different ways that each captures a different aspect of inequality: Gini and Top-1, Top-5 and Top-10 income shares, labeled as Top1, Top5 and Top10 respectively. Values for these measures are provided by Frank et al. (2015) and are available up to 2015. For household debt, three different measures are used: household debt/income ratio (D/I), total household debt per capita (totD), and credit card debt per capita (CC). The year-over-year changes of these debt levels are considered as household borrowing: D.D/I, D.totD and D.CC. The state level D/I is obtained from the Federal Reserve Board Enhanced Financial Accounts (EFA). totD and CC are obtained from FEBNY Consumer Credit Panel/Equifax. totD includes mortgage debt, student loans, auto loans, and credit card debt. All these debt measures are available from 1999 onward. Furthermore, GDP per capita (pgdp) data are obtained from Fed St. Louis and U.S. government spending websites. The monetary policy indicators are referred to as MI.1 and MI.2. MI.1 is the difference between the shadow federal funds rate and inflation rate while MI.2 is calculated as the 10-year T-bond rate minus the 3-month T-bill rate. The credit market indicator—denoted as CSB—measures the yield difference between corporate BAA bonds and 30-year government bonds. All these indicators are from Fed St. Louis and are measured as interest rates in percentage terms.


Steps to reproduce

The "ReadMe" file explains the data and R scripts. Further information is available in the paper that is published in Economic Modelling (forthcoming as of May 2024).


City University of New York Graduate School and University Center, College of Staten Island