Has inequality declined over the course of the recession and the recovery? In a recent paper, Stephen Rose of the George Washington Institute of Public Policy contends that the answer is yes. He claims that the top 1% have faced the largest losses in income over the Great Recession, which should result in declines in inequality. In contrast, recently updated data and analysis by Emmanuel Saez suggests that the top 1% has captured 91% of income growth over 2009-2012. In other words, Saez concludes that most of the gains from the recovery have gone solely to the top 1%. This might lead one to conclude that inequality has risen over the course of the recession. So what is really going on with inequality?
In this article, I evaluate trends in income inequality since the start of the recession using recently updated data from the Congressional Budget Office (CBO) and from Emmanuel Saez’s database, which is based on tax filing statistics from the Internal Revenue Service. I supplement this analysis with data from the Consumer Expenditure Survey to study consumption inequality as well. I find that income and consumption inequality declined over the period 2007-2011, however inequality has followed a U-shaped pattern, declining initially and then rising but not yet back at 2007 levels.
Let us first turn to the database compiled by Emmanuel Saez for studying trends in income inequality. These series are based on tax filers and exclude government benefits as well as taxes. As per numbers recently updated in January 2015, top earners experienced declines in real incomes over the period 2007 to 2013. Average incomes (in constant 2013 dollars), including realized capital gains, of the top 1% were $1,119,315 in 2013, relative to $1,533,064 in 2007 -- a drop of 27%. For the top 0.1%, incomes declined from $8,006,789 in 2007 to $5,279,695 in 2013, and for the top 0.01%, the decline was from $39,371,876 to $24,988,251. As a result, the share of incomes earned by these groups went down between 2007 and 2013. The share of the top 0.01% went down from 6 to 4.5%. The steepest decline happened over the recession when the share dropped to 3.9% in 2009. For the top 1% as a whole, the share went down from 23.5% in 2007 to 18.1% in 2009 and then rose to 20.1% in 2013.
This decline is mirrored in CBO data, which was recently updated as well, but is only available till 2011. The CBO definition of before-tax income includes all capital and labor income, as well as government transfers. After-tax income subtracts out federal taxes. Average before-tax incomes for the top 1% (in constant 2011 dollars) declined from $1,986,000 in 2007 to $1,453,100 in 2011. Average after-tax incomes have declined from $1,423,100 in 2007 to $1,031,900 in 2011. Over the same period, the share of after-tax incomes for the top 1% has declined from 16.7% to 12.6%. The decline in before-tax income shares was from 18.7 to 14.6%. Again, it is true that the big dip in shares took place between 2007 and 2009, but since 2009 income shares have been recovering, though they have yet to reach 2007 levels.
One measure of income inequality looks at the ratio of incomes for the top and bottom quintiles i.e. the 80/20 ratio. Between 2007 and 2011, shares of the bottom quintile, the bottom 20%, have gone up from 4.8 to 5.3% in before-tax income and 5.6 to 6.3% in after-tax incomes, as per the latest tables from the CBO. For the top 20%, the corresponding change has been a decline from 54.6 to 51.9% in before-tax income and 51.4 to 48.2% in after-tax income shares. However, the declines were steepest between 2007 and 2009. Since 2009, these top shares have been rising. For all quintiles aside from the top, there has been an increase in after-tax and before-tax income shares over this period. These data suggest that income inequality initially declined over the recession but then has started to rise again, even though it is still lower than in 2007. As corroboration, the Gini index tabulated by the CBO declined between 2007 and 2009 but has risen since then.
Consumption may be considered a better measure of standards of living or of permanent income than annual income since people can hedge against fluctuations in annual incomes through saving and borrowing. Consumption inequality is calculated as the ratio of consumption expenditures for high income households relative to low income households. Using data from the Consumer Expenditure Survey, we find that in 2007, average annual expenditures for the bottom 20% were $20,471 and in 2013, $22,393. For the top 20%, the equivalent values were $96,752 in 2007 and in 2013 were $99,237. In other words, the ratio declined from 4.7 in 2007 to 4.4 in 2013. While both the top and bottom quintiles saw increases in consumption, the increase at the bottom was higher than for the top in percentage terms. However, the big dip happened in 2011 when the ratio averaged 4.3. Since then the ratio has inched marginally higher.
A recent paper by Meyer and Sullivan also finds that consumption inequality decreased between 2007 and 2011. However, they find that income inequality in fact increased continuously over this period, when measured using the 90/10 ratio and when income is defined as net of taxes, plus food stamps and housing and school subsidies. Since the CBO data do not allow for tabulation of the 90/10 ratio and the Saez definition of income does not include taxes and transfers, it is hard to reconcile why the 90/10 ratio would show different trends than the 80/20 ratio. One possible explanation of course, is that the increase in incomes since 2009 has been much more pronounced for the top 10% of earners than for the top 20% as a whole.
Another measure of inequality is wealth inequality. A 2014 paper by Edward Wolff finds that wealth inequality rose sharply over the period 2007 to 2010, but remained largely unchanged over the period 2010 to 2013, despite the rebound in asset prices. Median wealth declined sharply by 44% between 2007 and 2010. The share of wealth held by the top quintile increased 4 percentage points, while shares of other quintiles dropped, with that of the bottom 40% falling from 0.2 to -0.9%. However, the same paper finds that, income inequality measured using the Gini coefficient, contracted sharply over 2007 to 2010. This decline was reversed in the period after 2010.
What explains these trends? The decline in incomes at the top is likely driven by the decline in capital gains income and the general drop in asset values. Capital gains as a share of total income dropped from 31.7% to 18.2% between 2007 and 2011 for the top 1%, as per the CBO data. Tax and transfer policy is also likely to have played a role. Average federal tax rates on the top 1% have gone up from 28.3% in 2007 to 29% in 2011, as per the CBO. At the same time, average transfer incomes have increased from $7,900 to $9,100 between 2007 and 2011 for the lowest quintile. This may have helped households at the bottom of the income distribution sustain their standards of living.
To summarize, the recession and recovery paint a complicated picture for recent changes in inequality. While data from the CBO suggest that income inequality declined between 2007 and 2011, a more nuanced picture suggests that income inequality first declined between 2007 and 2009, but has been increasing again since then, though inequality has not yet returned to 2007 levels. Wealth inequality rose in the first few of years of the Great Recession, but has stayed relatively constant since. Consumption inequality declined over the course of the recession, but is showing a marginal uptick in recent years.
So yes, income inequality did decline over some years of the recession as top incomes took a hit and transfer programs kicked in to boost incomes at the bottom. However, both of these “automatic stabilizers” are temporary artifacts of the recession. Recessions are a good reminder that even the very wealthy are not immune to economic busts. However, transfer programs are also only a temporary, albeit useful, solution to helping the poor insure against economic shocks. A more meaningful and, long-term, response to alleviating poverty and income inequality requires investments in human capital and training, encouraging workforce participation, and stemming the decline in traditional, stable family structures, which is highly correlated with poverty and adverse economic outcomes across generations.
