Inequality in America II – Income, Taxation and Spending

Part II of our series, Inequality in America, focuses of inequality of income.

By: George Noga – May 8, 2016

  There are numerous and mind-numbing statistical methods for calculating income inequality. The Census Bureau alone reports the Gini coefficient, Theil index and MLD (mean logarithmic deviation). Many of these statistics do indeed show more inequality now than in past decades; however, peeking inside the numbers is revealing. Note: Most data herein are from US Census Bureau and BLS reports published in 2013-2014.

   By every measure extant, inequality rose more under Clinton than Reagan – Theil at double and MLD at triple the rate. The same is true with Obama’s first six years vs Bush 43. The Gini coefficient rose triple the rate under Obama; MLD rose 37% more; and Theil is up sharply while it fell under Bush 43. It is not a giant leap to deduce that most of the putative increase in income inequality results from progressive policies.

   Despite all the esoteric statistics, we really know very little about income inequality because all the data are – to use a highly technical term – crapola! Every study is fatally flawed by inconsistencies and limitations affecting source data; the major flaws are:

  1. Statistics are based on AGI (adjusted gross income) and not on all income. Much income is not included in AGI, such as contributions to IRA and 401(k) plans. AGI excludes the non-taxable portion of Social Security, EITC, Medicare, Medicaid and SNAP. Every one of these, if included in AGI, would significantly reduce inequality.
  2. Data use household instead of individual income. This renders all comparisons between time periods and income quintiles meaningless because the number of people per household changes over time and also changes between quintiles. For example, the number of one-person households has sharply increased in recent years (mostly in the bottom income quintile) making it appear there is more inequality among households even though, in reality, there is much, much less inequality among individuals.
  3. Quintiles are inconsistent. The top quintile has 3.2 people per household whereas the bottom quintile has 1.7; the income in the top quintile must be spread among twice as many people as the bottom quintile. It also means there are 25 million more people in the top quintile versus the bottom. Use of household data paints a deeply flawed picture of increased inequality between income cohorts – inequality that doesn’t exist.  
  4. Aggregate statistics don’t compare the same groups. Statistics showing inequality increasing over time don’t track the same people. New people (most poor immigrants) keep entering the back of the line, skewing all data downward. If they tracked the exact same people (and excluded new people), the data would show decreasing inequality.

   Does our tax system result in inequality? In one word, no. The US has one of the most progressive tax regimens in the world. Even Social Security and Medicare are somewhat progressive when taking (as should be done) the benefits into account. Moreover, increasing marginal tax rates on the wealthy would not result in their paying more in taxes – a principle well documented and even codified in Hauser’s Law.

   No discussion of income inequality would be complete without genuflecting to the so-called gender gap. However, economic analysis shows that the gap between incomes of men and women completely disappears when properly adjusting for level of education, type of degree, experience, hours worked and level of danger.

   The Census Bureau also reports data on spending by income quintile. The lowest quintile spends $2 for every $1 of reported income. Some of this comes from the underground economy – which logically is the province mostly of that cohort. If we were to gauge inequality based on actual spending rather than on fatally flawed measures of income, the effect would be a signal decrease in inequality in America.    


The next post in this series on May 15th addresses the $15 minimum wage.