Association for Public Policy Analysis & Management

Growing interest in economic inequality continues to dominate the headlines.  In 2014, Janet Yellen, Federal Reserve Chair, stated “The extent of and continuing increase in inequality in the United States greatly concern me.” This echoed President Obama’s remarks the previous year that increasing inequality “challenges the very essence of who we are as a people.”  Studies of economic inequality usually conduct separate examinations of income inequality, consumption inequality, and/or wealth inequality, and findings are almost always limited to one variable at a time.  Studying these measures separately, however, misses the synergy between the three measures explicit in the life-cycle budget constraint.  An increase in income held by the top of the distribution means that consumption and/or wealth of the top must also increase.  In addition, retirees are under-represented in the bottom of the income distribution but greatly over-represented in the top of the consumption distribution and the wealth distribution.

We are the first to study inequality in three dimensions using the same dataset, focusing on the conjoint distributions of income, consumption, and wealth for the same individuals. We impute consumption to the 1989-2013 Survey of Consumer Finances using the Consumer Expenditure Survey from the same years, which is the first time consumption is imputed to the SCF to study overall inequality.  We measure the level and trend in inequality in income, consumption, and net worth, comparing our results to existing research.  We examine the pairwise distributions of our measures and how these have changed over time, and relate it to the life-cycle budget constraint of income less consumption equaling the change in net worth.

For the joint distribution of income and consumption, we see the twin peaks phenomenon seen in the relative mobility literature.  In our context, it means that those in the bottom (top) of the income distribution are more likely to also be in the bottom (top) of the consumption distribution.  There is less of a correlation between income and wealth in the bottom of the distribution, but there is a strong correlation between wealth and income at the top of the distributions.  Similar results are seen when we look at the correlation of income and consumption by wealth quintile.  The correlation between income and consumption is lower in the bottom wealth quintile than in the top wealth quintile.

We also use the Panel Study of Income Dynamics (PSID) because it contains income and wealth since the mid-1980s and consumption since 1999, which has been imputed back to the 1980s (see Attanasio and Pistaferri, 2014).  The PSID allows for longitudinal analysis and intra- and inter-generational mobility issues not feasible with the SCF, albeit without the top end wealth and income inequality found in the SCF.

These correlations have important social policy implications, especially as they affect equality of opportunity, intergenerational mobility, transfers across generations, poverty, and social exclusion.   In closing we will address policies that may help stem the tide of ever growing wealth and income inequality.