A popular press refrain is that the US suffers from income inequality. Progressive politicians translate this to a call for high income households to pay their “fair share” in taxes. This blog evaluates the current degree of income inequality and addresses the notion that making the tax system more progressive would redress this issue.
The primary source of income inequality measures is the US Census Bureau and Bureau of Labor Statistics. They use the Current Population Survey (CPS) data to regularly measure and report household incomes by quintile. These data give the press and the public the widely quoted measure of income inequality in our economy. Figure 1 presents the most recent 4 years of the reported, mean household income, by quintiles, and the top 5 percent of the distribution.
Figure 1. Mean Income, by Quintile, in Current Dollars (from US Census: Table H-3)
The ratio in 2019 of the highest quintile to the lowest is about 17, a number that is seen by many proponents of the income disparity movement as excessive. That one-fifth of US households should earn 17 times that the lowest fifth is proof, to such critics, that a largely free market economy produces “unfair” outcomes. There are numerous reasons why this comparison is naive, at best.
For one thing, this tabulation is of households and thus does not speak directly to the incomes of individual workers. Each quintile has 25,000 households. Households in the lowest income quintile have fewer people in them and more individuals reporting zero earnings. On the other hand, there are 4.5 times as many earners in the highest income household quintile as in the lowest income household quintile.
Additionally, the Census does not include income received by the lowest quintile households through many, existing redistributive programs. This includes such major programs as “negative taxes” such as the Earned Income Tax Credit (EITC), the value of food stamps, medical care, rental assistance, or other existing free or subsidized services targeting low income households. John F. Early, a former assistant commissioner of the Bureau of Labor statistics, estimated in 2018 that as a result of these omissions, the incomes of the lowest quintile households are understated by “a factor of two or more”.
Most importantly, the Census data does not account for taxes paid by households with taxable income. Focusing on the distribution of income without accounting for the associated tax liability exaggerates the disparity of financial resources available to the various household quintiles. Taxes paid by higher income households reduce those households’ disposable incomes. Failure to credit the redistributions of the income households’ tax revenue to lower income households exaggerates further the appearance of income disparity. The greater is the redistribution from high income households to lower income households, the more progressive a tax system is said to be. On these grounds, the US tax system is one of the most progressive in the world.
Although income is concentrated in the top percentile groups, tax paying responsibility is even more concentrated. This concentration of both incomes and taxes is illustrated in Figure 2. The greater concentration of the share of taxes paid than the shares of income is the signature of tax progressivity. In the US case, a very small fraction of tax payers pay virtually all of federal income taxes, and their tax share is significantly greater than their revenue share. Moreover, both distributions are very different from a perfectly equal share (each household has the same income), which is represented by the Flat Distribution line in the Figure 2 graphic
Figure 2. Distribution of Taxable Income and Tax Obligations in the US, 2019
For example, while the top 3 percent of households earn 30 percent of all income, they pay 53 percent of all taxes. Similarly, the top 10 percent of households earn 47 percent of all income, but pay 71 percent of all taxes.
As Figure 2 illustrates, a perfectly “fair” system— where income and tax obligations were equal for all (displayed in the figure as a flat distribution)— would result in the need to shift a huge portion of the total tax burden onto lower income households. Presently, all of households together that are below the 50th percentile income bear only 3 percent of the total federal income tax burden.
Figure 3. The Tradeoff between Income Equality and Tax Progressivity
From Figure 3, there is a clear tendency for there to be a trade-off, in practice, between income concentration and taxation progressivity. Countries like the US and Ireland, for example, have noticeably unequal distributions of income, but compensate for this by also having highly progressive income taxation structures.
Conversely, although no country has a perfectly equitable income distribution (i.e., an inequality index of zero on the X-axis), the ones that come closest, such as Switzerland, Iceland, Sweden and Denmark, are associated with unprogressive tax structures. However, that approach, especially when coupled with the need to fund government monopolies in the provision of medical care, education, child care, etc., results in the need to broaden the tax base. The result is that low income households often end up facing high tax rates. Poor and low income households in Sweden, for example, face very high total income tax rates (in the range of 60 percent). This results in relatively low standards of living as measured by private consumption spending.
This approach seems to be favored by the Biden Administration, which plans to introduce a $3.5 trillion social welfare “infrastructure” plan. To pay for the plan, the Biden administration has stated its intent to raise marginal income and capital gains tax rates significantly on households with incomes in excess of $400,000 dollars. However, by my estimates, in 2018, this group of taxpayers was already paying approximately three-quarters ($5.5 trillion) of all federal income taxes.
Even assuming that the spending plans are accurate, therefore, we can expect middle- and low-income households will have to be taxed at higher rates in order to finance the planned spending. It is difficult to estimate how much additional taxation of lower income households will have be to be. Increasing the progressivity of the tax system may not necessarily make the distribution of income more equitable.
Indeed, the economy can adjust so as to defeat the aims of tax reform. For example, as Laura Jackson and her colleagues at the Federal Reserve Bank of St. Louis have pointed out, when tax progressivity increases, the bottom of the income distribution and the economy as a whole may initially benefit. This is because lower income households benefit from an increase in disposable income because of the income transferred from higher income households.
However, the increased marginal tax rates levied on high income households may have adverse effects on investment and employment of others. As higher income households save, invest and employ less, there will be less taxable income to tax. To maintain the high Federal spending, the tax base will have to expand to lower income households.
This may explain why, in Figure 3, there are no countries that, in equilibrium, are able to simultaneously maintain both a progressive tax system and equality of income before taxes. In my view, the emphasis on disparities of income alone, without considering taxation, misleads policy makers.
