This chart shows how income tax liabilities have changed each year for the five income quintiles over the past half century. Here are a few things to notice:
- We love cutting income taxes. Almost all of the changes are tax cuts (i.e. they fall below zero in the chart), especially for the bottom 80%. Note that this pattern does not hold for payroll taxes.
- Tax cuts almost always benefit top income earners most. They pay the most taxes in dollar terms, so they get the most back. For example, the top 10% paid roughly 70% of federal individual income taxes in 2009. Increases in income inequality tend to increase this share.
- Some tax cuts are much more broad-based than others. Notice that blue, red, green and yellow lines are more pronounced for some tax cuts like 1964 and 2001 compared to others like 1982 and 2003, which shows that 1964 and 2001 were more broad-based cuts. In 1993, Clinton raised taxes on the top group from 36 to 39.6% and decreased taxes for lower income groups.
- Subsequent job creation and economic growth varies based on the composition of tax changes. A natural question to ask is whether or not these compositional differences affect the aggregate impacts of tax changes. I find that heterogeneity is quite important, that almost all of the stimulative effect of tax cuts results from tax cuts for the bottom 90%, and that there is no substantial link between tax cuts for the top 10% and subsequent job creation. The notion that raising top rates slightly leads to substantially lower job creation and economic growth has no empirical support from the last half century, although it may have the power to push us over the fiscal cliff. For more information on this evidence, see here for a popular summary in NYTimes Economix, here for the paper, and here for slides.
Note that the “Read my lips, no new taxes” change under G.H.W. Bush shows up as a tax increase when finer income groupings are used (and in versions of this chart that include payroll changes, reflecting the doubling of the cap for health insurance).
owenzidar 
What would really be helpful is a chart showing by year since the end of World War II what percentage of the national income each quintile earned (and by earned, I include here capital gains and dividends, not just “earnings” in the sense used in the tax code) and what percentage of all government revenue (federal, state and local) each quintile paid, along with the same data for the top 5%, top 1% and top 0.1%. Such a chart would enable one to determine how tax burden has changed along with the change in income distribution. It seems to be fairly well admitted that income distribution has widened over the decades so that the top 20%, 5%, 1% and 0.1% now receive more of the total income than they did in the past. Even if those groups also pay a higher percentage of the total taxes (and I have no idea if they do), has that increase kept up with or even exceeded their increase in percent of total income or has it fallen behind.
Okay, so you chart the changes.. but there is no corresponding analysis of growth. Not a single regression. How can you draw any conclusions here?
Hi Sal, the last sentence of the post provides links to a full paper (as well as a popular summary and presentation) with many regressions to support the claim.
Pingback: Broad-Based Tax Increases are Rare | owenzidar
Pingback: Tuesday’s tax increase is the biggest in decades
Pingback: Gerri McDaniel – Tuesday’s tax increase is the biggest in decades