Over the past few years, the concept of income inequality has come up frequently in the news. I recently attended a conference in which this the topic of conversation.
Why Do We Care?
At the heart of this conversation is a basic question: why do we care about income inequality at all? In other words, is having a flatter income distribution in-and-of itself a social goal?
To some, the very fact that one person has more than another is repulsive. Being raised with the ethic of being rewarded with the just desserts of one’s labor, and knowing that people make different choices with different consequences, I cannot, in good conscience agree that one person owning more than another is inherently bad. Some people have preferences for leisure time, while others have preferences to work and spend more money in less time or to buy things rather than buy time. Having such preferences will then lead us to have differences in behavior and some outcomes. In addition, if there are no rewards for hard work, innovation, and advancement of the individual above the levels given to those for less work, society as a whole tends to take a downward turn. Some inequality is not only necessary, but is good.
To me, the central reason we care about income inequality is twofold:
1) We worry that inequality indicates losing people at the bottom to poverty and could have negative effects on social cohesion;
2) We worry that as the gap between rich and poor grows, the poor have fewer opportunities to advance.
These are legitimate concerns.
Other than the Gini coefficient, how is the US doing with the core reasons we care about income inequality? That depends on how you want to measure it.
Part 1: How We Measure “The Bottom”
For all intents and purposes, the poor in America feel stuck. Policy prescriptions such as raising the minimum wage and increases in Earned Income Tax Credits (EITC) have been proposed, but would likely not eliminate racial gaps in achievement. The minimum wage, in particular, is a blunt instrument that may have adverse employment effects if done too abruptly. Most social scientists agree that more fundamental changes are needed.
Education reform has always been on the docket. Given the political will and enough instruments like well-funded charter schools and community investment, the racial achievement gap can close, even in places with less than ideal family backgrounds. Closing such a gap can help alleviate the college graduation gap, the burden of student loans, and the psychological weight of debt and living paycheck to paycheck.
But other than the longstanding problems faced by the long-term poor is the question of who we actually consider poor.
If we compare the incomes of recent college graduates to people in their mid-forties, you will obviously be left wondering how we abandoned all these young people. Incomes are low (or even $0 if they have internships) at the moment, but lifetime incomes may be high. Someone moving up from a low-skill job at a call center to a supervisory position may be making much more net of debt than someone with an undergraduate degree in theater or the history of Greco-Roman Fingerpainting. And someone on the cusp of retirement may be living on less than $12/hour at a part time job, but they have $500,000 in equity in a home they plan to sell.
Which one of these do we consider “poor?”
Then there is the question of how to account for income.
Most measures try to account for after-tax and after-transfer income (that is, how much do people have after getting EITC, Medicare, Medicaid, etc).
But consider this question: how much of a worker’s prospective paycheck pays for fringe benefits such as health insurance, with its increasing costs? If we included that measure, how would their overall income look? Employers paying upwards of 75% of health insurance premiums sums to thousands of dollars per year in unaccounted pay. Yet most measures do not include these as income for comparison.
How do we take into account volatility in income? 50% of Americans see changes of greater than 25% of their income year over year.
For which year are we accounting their status? Or do we choose the mean over their lifetime, or over a 5 year period?
The picture of the bottom part of the income distribution is complicated and should not be simplified to a soundbite or uncritical comparison to somewhere like Sweden, Finland, or Brazil. These issues may be hinted at by a measure of income inequality, but they certainly are not defined by or fixed with it.