This week, we were told by David Leonhardt (in his summary of recent writing by Emmanuel Saez and Gabriel Zucman) in the New York Times that “The Rich Really Do Pay Lower Taxes Than You”. This week, we also heard from Joint Committee on Taxation economist David Splinter that “U.S. Taxes are Progressive.” How do you reconcile these two facts?
It all comes down to what is meant by “rich”, and what are meant by “taxes”. The New York Times article focuses on the 400 individuals with the most income. They are certainly rich, but, are also very different than other people most would consider “rich”. To me, the guy a few neighborhoods over that pulls in $2-3 million a year is rich. So are the joint doctor/lawyer couple in the next neighborhood that earns a joint $800K a year. These people have far, far more income than most people, and anyone in the 0-90% of the income distribution would look at these folks and consider them very rich. Most folks like this pay a lot of their income in taxes (as is tabulated, but not extensively discussed, in the NYT article). But, when you get to the top 400 households, they are a different sort of “rich” person. They earn their wealth through capital income, they have unrealized gains, etc. Specifically, in 2014, the top 400 households earned $5,682,344,000 in salaries and wages, $5,387,334,000 in interest, $13,839,214,000 in dividends, and $76,843,035,000 from the sale of capital assets. It should not startle us that if we have a lower top capital gains tax rate than we do top income tax rate, that a group which largely lives off of capital gains will have lower rates than some might hope. It is a conscience tax policy choice whether we want to cater policy to 400 people, or be bothered by policies that optimize some objective function for others have a different impact on a small set of people.
Just as it matters who the rich are, it matters what you mean by “taxes”. The earned income tax credit is one of the largest welfare programs we have in the US. It makes a difference whether you consider that credit, and others like it, into the taxes people “pay” (or allow them to reduce the amount paid). If you consider those types of credits, you get a very large, and increasing, set of people who pay nothing in federal income taxes. State taxes matter. Payroll taxes hugely matter. Including payroll taxes in this analysis will result in much lower tax rates for the wealthy, and one should consider them. But, those payroll taxes are a factor in allowing people to receive social security and other similar benefits—and you have to decide what to do with that income. Corporate taxes matter. In the end, real individuals (mostly wealthy ones, if you believe the incidence literature) will be made worse off by the corporate tax being remitted. Those taxes should be allocated to someone. Implicit taxes matter a lot, especially if consider the type of wealthy individuals that invest in tax-free assets like municipal bonds. But, implicit taxes are extremely difficult to measure. Which taxes you take into account, and how, has a huge ability to change the answer.
David Splinter points out how some of these factors make a big difference, and can lead to very different results. While some analysis done on this topic makes design choices and shows a specific result, David Splinter (and in a far more comprehensive view, David Splinter and Gerald Auten) look at all the reasonable design possibilities, and show how the different choices make a huge difference. If your goal is to actually understand the economy and design policy motivated by data, this is an incredibly useful exercise.
