Means Testing Social Security: Income versus wealth
10 July 2017, byAlan L. Gustman, Nahid Tabatabai and Thomas L. Steinmeier
The U.S. Social Security System is a defined benefit retirement system that uses a progressive formula to calculate own benefits based on past indexed covered earnings, the date benefits are claimed and a cost of living adjustment after retirement. Covered earnings have changed over time both with aggregate wages and with earnings history. There also are provisions for spouse and survivor benefits. Tax revenue is based on earnings up to a taxable maximum.
U.S. Social Security finances are out of balance. Future tax revenues will not be enough to cover promised benefits. To remedy this situation, some policy makers recommend increasing taxes. Others recommend reducing the growth of benefits, with some advocating means testing Social Security benefits.
Popular discussions conflate income and wealth as measures of means. For those policy makers who have thought about means testing Social Security, their proposals typically focus on incomes, but their rhetoric often suggests their concern is wealth. Many seniors with highest incomes do not have highest wealth, and similarly, those with highest wealth may not have highest incomes. Because income and wealth are not perfectly correlated, and neither is perfectly correlated with Social Security benefits, any means test will have different distributional effects depending on the criteria used to define high means. In addition, there will be substantial differences when a Social Security means test based on income is evaluated in terms of its effects on individuals arrayed by their wealth rather than their income. Similarly, a means test based on wealth will be evaluated quite differently by policy makers who believe that income is the appropriate basis for a means test than by those who believe that means tests should be based on wealth.
We examined the distribution of income and wealth for a sample of individuals from the Health and Retirement Study who were aged 69 to 79 in 2010. Income and wealth are far from perfectly related. We find they are correlated .50 among individuals, and .53 among households. Of our sample, 35.4 percent was in either the top quarter of the income distribution or the top quarter of the wealth distribution, but only 14.5 percent of the sample was in the top quarter of both distributions. Consequently, suppose that a means test was applied to those in the top wealth quartile. Now consider a policy maker who believed that income falling in the top quartile was a better basis for means testing. Such an individual would find that only 58 percent of those whose benefits were reduced fell in the top quarter of households ranked by highest income (14.5/25.0 = .58).
Read the full article at Austaxpolicy blog.