How Progressive is the Tax Code?

The federal tax code has been a big topic of discussion this election cycle. A lot of this discussion has been driven by the increased focus on income inequality. Whatever your views on this subject, I think it's useful to look at the way the tax burden is distributed across income groups. In this post, I'll present data from the Congressional Budget Office (CBO), which you can download here.

The charts below were created using three data series: market income, transfers, and federal taxes. Each series runs from 1979 to 2011 (unfortunately, 2011 is the most recent year for which this data is available) and is broken down into quintiles by market income. The top 1% is also split out into its own category. Every data point is a household average for the group. The data set linked above breaks the top quintile into a few more groups if you're interested in that.


Here are some definitions taken from the data set.

Government transfers are cash payments from Social Security, unemployment insurance, Supplemental Security Income, Temporary Assistance for Needy Families (and its predecessor, Aid to Families with Dependent Children), veterans’ programs, workers’ compensation, and state and local government assistance programs. They also include the value of in-kind benefits, such as Supplemental Nutrition Assistance Program vouchers (formerly known as food stamps), school lunches and breakfasts, housing assistance, energy assistance, and benefits provided by Medicare, Medicaid, and the Children’s Health Insurance Program. (The value of health insurance is measured on the basis of the Census Bureau’s estimates of the average cost to the government of providing such insurance.)

Federal taxes are the amount of taxes a household owes based on income earned in a year, regardless of when the taxes are paid. Individual income taxes are allocated directly to households paying those taxes. Payroll, or social insurance, taxes are allocated to households that pay those taxes directly or indirectly through their employers. Excise taxes are allocated to households according to their consumption of the taxed good or service. Corporate income taxes are allocated to households according to their share of capital and labor income. 

Market income includes labor income, business income, capital gains, interest and dividends, retirement income, and income from other sources. The chart below summarizes market income from 1979 to 2011. Note that income has risen in real terms for all 5 quintiles and the top 1%. Income for the highest quintile and top 1% has increased more, but is also more volatile. It rose dramatically during the dot-com and real-estate bubbles and crashed afterward.

(click to enlarge)

But what about the tax burden? Who pays the most? Do the rich pay more? The chart below shows the amount of federal taxes (net of transfers) paid for each group. Note that as incomes rose and fell for the richest Americans their tax burden did as well. The story is different for the other 4 quintiles. While the poorest 2 quintiles have received more in transfers than they paid in taxes since at least 1979, the middle quintile didn't have a negative tax burden until around the time of the "Bush tax cuts." The tax burden for the 4th income quintile also began trending down around that time and fell to 0 in 2009. This implies that nearly all of the tax burden at the federal level is borne by the top income quintile.

(click to enlarge)

Another way to look at the distribution of the tax burden is to calculate the net effective tax rate. This is based on actual taxes paid, so it has nothing to do with marginal rates or tax brackets. The graph below shows the following ratio for each income group:

(Federal Taxes - Transfers) / Market Income

Note that this number is negative when transfers are greater than federal taxes paid. The story we told above is more obvious here: tax burdens were negative for the bottom 2 income quintiles over the entire time period, the middle and 4th income quintiles have declining tax burdens, the middle income quintile's tax burden went negative in 2001, and the 4th income quintile's tax burden went to zero in 2009.

The increase in the effective tax rate for the lowest income quintile in the mid-1990s is interesting and I don't have a specific policy change to explain it. The data shows that transfers fell off in the mid 1990s while taxes remained about even. Transfers for the lowest quintile didn't increase much after that until the mid-2000s but were at a record high in 2011 (13.5% above the average). Federal taxes on the lowest quintile were at a record low (60.7% below the average) in 2011.

Also interesting here is that the highest income quintile and the top 1% both have essentially flat tax burdens over the 1979-2011 period. Both groups' tax burdens topped out in the mid-to-late 1990s and were 3% and 2% below the average in 2011. Federal taxes net of transfers in 2011 averaged $47,000 for the highest income quintile households and $411,000 for the top 1% households.

(click to enlarge)

With all the rhetoric floating around about tax breaks for this or that group, loopholes that need closing (or not), and the ever-increasing focus on inequality, I think it's important to take a hard look at the data before engaging in these discussions.