. . . [A] new analysis by the Center for Economic and Policy Research (CEPR) confirms that means-testing would yield very little in savings … unless we took benefits away not only from rich retirees, but also from many who are solidly middle-class.
The reason, as the CEPR analysis shows, is that there aren’t enough rich retirees — and they don’t collect enough in Social Security — to make much of a difference. Only 2 percent of Social Security benefits go to retirees with other (non-Social Security) income of $100,000 or more each year. (See chart.) Only about 10 percent of benefits go to people with outside income of $40,000 or more a year — a figure that most of us would regard as middle class.
Ms. Ruffin also understands this important point:
Means-testing would also penalize people who planned prudently for their senior years. Income security during retirement is often likened to a three-legged stool — consisting of Social Security benefits, employer pensions, and personal savings — sometimes supplemented by part-time employment. Reducing Social Security for people whose non-Social Security income is higher because they built up savings during their working years or take a part-time job would sap incentives to work and save.
Indeed it would.
Means-testing social security is a tax by another name. All taxes reward and penalize certain types of behavior (it comes with the territory) but hiking taxation in way that further penalizes both savings and work seems more than a little unwise. Far better, of course, to look at consumption . . .