The true measure of inequality

Standard figures on inequality are only loosely related to the concept that legitimately arouses concern

Inequality and what to do about it figure prominently in political debate. A clearer view of the problem and the possible remedies would be valuable—and a new study by Alan Auerbach, Laurence Kotlikoff and Darryl Kohler helps. It shows the hazards of jumping to conclusions about inequality, and also highlights some pernicious effects of America’s convoluted tax-and-benefit system.

The study begins by noting that inequality in lifetime spending power is more important than inequality in current incomes and wealth, the usual measures. Momentary snapshots of the kind provided by the standard estimates are misleading.

A new graduate just starting out isn’t poor in the way that a high-school drop-out is poor, though their current incomes may be the same. And while figures for wealth suggest that a retired person with meagre savings is richer than a new Harvard graduate with student debt, common sense says otherwise.

Auerbach and his co-authors first set out to measure US inequality using a broader life-cycle definition of incomes and spending. They then look at how inequality defined that way is affected by federal and state policies—income taxes, welfare payments, disability benefits, Medicare taxes and payments, social security taxes and payments, and so on, all acting together.

It’s complicated. In addition to masses of data and a minutely detailed model, a lot of assumptions and projections are needed for this exercise to work. But after much crunching of numbers, the authors find less inequality than the standard measures show. For instance, the top 1% of 40-49 year olds (ranked by pre-tax resources) has 18.9% of that cohort’s wealth, and 13.4% of its current income—but only 9.2% of its remaining lifetime spending.

In accounting for this difference, progressive taxes and benefits play a big role. Again looking at 40-49 year olds, each dollar of resources of those in the top 1% is taxed, on average, at 45%. The tax rate for the bottom fifth is negative—a net subsidy of 34.2%.

On the face of it, the system is working, by leaning heavily against pre-tax inequality. The problem is that the system is also inefficient and unfair. It applies vastly different tax rates to people in similar circumstances, and saddles many people, especially those with few resources, with crippling work disincentives.

To measure disincentives, look at marginal as opposed to average tax rates: what matters is the tax due on additional income. These marginal rates vary not just from one income group to the next or from one age group to the next, but within each of those categories.

For instance, the median marginal tax rate for 40-49 year olds in the lowest income quintile is 42.2%. That’s already high, but what’s more striking is the dispersion of marginal rates inside that narrow category. In the bottom quintile of 40-49 year olds, the lowest marginal rate is a subsidy of 17.9%, and the highest is a tax of more than 900%.

That staggeringly high rate results from the removal of Medicaid benefits following a small rise in income.

It’s the rate of a married couple earning roughly $21,000. The couple qualifies for Medicaid. But if the husband earns an extra $1,000 and smooths this extra income, they end up losing Medicaid for a year. As a result, their marginal net tax rate is 933.7%. That is, the extra $1,000 delivers an increase in remaining lifetime net taxes (equivalently a decline in remaining lifetime net spending) of more than $9,000.

That’s an extreme case, but the tax-and-benefit system has so many overlapping components, it’s full of anomalies similar in kind if not in degree. If nothing else, the study underlines the harm that follows from letting a needlessly complicated system proliferate unchecked. The oft-proclaimed goal of tax simplification should go far beyond reducing the number of bands in the income-tax schedule—a change that, by itself, would achieve almost nothing. What’s needed is a view of the fiscal system as a whole.

The other main lesson is that the standard figures on inequality are only loosely related to the concept that legitimately arouses concern. The paper doesn’t show that inequality isn’t a problem; it doesn’t address whether inequality is getting better or worse. It does suggest that measures of inequality based
on current incomes and wealth are seriously misleading, and that tweaking policy piecemeal can do more harm than good. BLOOMBERG

Clive Crook is a Bloomberg View columnist.

Comments are welcome at