New York Times Fudges Income Data

November 30, 2006

A New York Times article that appeared on the front page of the business section on Tuesday cited new IRS data to make some technically accurate, yet highly misleading statements regarding incomes and taxes.

First, the article states, “The tax cuts contributed to a big decline in individual income tax receipts, which fell at a rate 14 times that of the drop in incomes.” While this is technically true, one of the reasons that such a drop in tax receipts would exceed the overall percentage drop in incomes is because of the high degree of progressivity in the individual income tax code. Because the incomes of wealthier individuals tend to be more pro-cyclical (i.e. fluctuate with the economy, like the stock market), a progressive code that relies on most of its revenue from these wealthy individuals is going to be more volatile than the aggregate measure of income throughout all taxpayers.

In other words, since the lion’s share of tax revenue comes from the wealthy who have their money in investments, tax receipts will wax and wane depending on the health of the stock market. For this reason, it is natural that the IRS measure of income (Adjusted Gross Income, or AGI) fluctuates more than other measures of income because AGI includes capital gains, while, for example, personal income does not.

Also, the article attempts to show that low-income Americans rely on “$7 a day each” by looking at the bottom 20 percent of tax returns. There are three problems with this analysis. The first is that among the bottom 20 percent, over one-third of all tax returns are filed by dependents, a number of whom file a tax return just to get their money back from the IRS. These are mostly young people still living at home, declared as dependents on their parents’ returns. Obviously these young, part-time workers are not poor in the normal sense of the word.

The second problem is the way the article defines the total number of people in the bottom quintile. The article claims that 60 million people would fall into the bottom quintile by calculating 1/5th of the U.S. population (300 million). In reality, the number of people represented by total AGI in the bottom quintile is closer to 35 million due to the number of exemptions in that group. In addition, the income of non-filers is not represented in total AGI so they should be excluded when calculating per capita income.

Finally, the article flippantly mentions that the “$7 per day” figure ignores some government transfers like EITC and food stamps. Does it ever! In fact, this year the government will give low-income people $50 billion in refundable credits through the EITC and child credit – roughly equal to a quarter of the total AGI in the bottom quintile. Needless to say, $50 billion is a significant amount of income for the poorest Americans.

Astonishingly, it also fails to acknowledge Social Security benefits that are mostly tax-exempt for low-income individuals. The IRS does not count all of one’s Social Security income as part of Adjusted Gross Income, and for low-income individuals, a greater fraction of Social Security benefits are tax-exempt and thereby not included in AGI. The article flatly ignores a huge flow of money to retirees who may appear to be low-income when these benefits are not counted. In fact, retired couples who own a home outright and live mostly on Social Security are numerous in this group.

In order to craft sound tax policy, it is it vital that lawmakers have accurate and reliable information. Hastily thrown-together statistics that attempt to serve personal political agendas will do nothing to help those who need it the most.

For a sober analysis of IRS income data, please see:
Summary of Latest Federal Individual Income Tax Data, by Gerald Prante, September 25, 2006

New IRS Data Show All Income Groups Have Seen Tax Liabilities Fall Since 2000, by Gerald Prante, September 20, 2006

New IRS Data: Tax Code More Progressive in 2004 than in 2000, by Gerald Prante, October 17, 2006


Related Articles