How well do the changes to Social Security proposed by the Commission stack up?
1) Keep full-career minimum wage workers above the poverty threshold. “Benefit adequacy,” an explicit goal of the current system, is politically difficult to ignore for the Debt Commission. But this goal properly belongs in welfare programs, including laws that set the minimum wage—not in Social Security which is a retirement, dependent, survivor, and disability program. Under the current minimum wage of $7.25 per hour, someone who works 40 hour weeks for 50 weeks each year would make $14,500 per year — more than 1‑person poverty threshold of $10,830 per year. Presumably, the Social Security benefit formula will return 90 percent of such workers’ average monthly indexed wage by way of annual benefits — which, if the minimum wage level kept pace with average wage growth, would provide a benefit of about $13,000 per year in today’s dollars — well above the federal poverty threshold. The problem is that the minimum wage is not indexed — not even for inflation leave along wage growth. Congress increases the minimum wage in an ad-hoc way from time to time so that the minimum worker’s Social Security benefit may fall well below the federal poverty limit. But why not deal with this problem in the same way as is done for workers — by topping up the benefit by welfare payments? Fixing the problem of old age poverty through Social Security improperly (some might say unfairly to other workers) mixes welfare goals with those of a primarily retirement insurance system.
2) Wage index the minimum benefit to ensure that it is effective now and in the future. This means, although we don’t wage index the minimum wage — paid for working — we will wage index the Social Security benefit for retirees of every generation. Workers entitled to more than a minimum wage would experience a phase-out of benefits from working. The Social Security benefit formula is already progressive and adds to the US fiscal system’s high progressivity at the low end of the wage distribution. This reform element worsens the fiscal system’s already sizable disincentives for low-wage workers to acquire skills, work, and climb up the income and wealth ladder.
3) Provide a benefit boost to older retirees most at risk of outliving other retirement resources. The condition of being most “at risk” of outliving retirement resources applies to two types of people. Those who mismanage their finances and those with long expected lifetimes. The former do not deserve, but will be provided a bailout in welfare states. The latter deserve, but should not receive a bailout because they are among the highest beneficiaries of the system—longevity is correlated with high education and wealth.
The lesson of the past 4 decades is that the provision of ever increasing economic security during retirement leads to lower personal and national saving, reduced work effort, and an dissolution of social structures (the family) that are conduits for transferring skills to the next generation — overall, an erosion of economic growth promoting factors that would weaken the future economy. Even if Social Security is a Samaritan’s response to old age poverty, the evidence on saving, retirement, and skill formation suggests that prospective beneficiaries behave strategically. If each time, the government’s response to too little retirement saving is a “bailout” by way of more generous and more secure tax financed social insurance benefits, prospective beneficiaries (current workers) reduce saving by even more, work even less, and acquire fewer skills. Multiple rounds of such personal (under-saving and under investing) and government (increasing benefit generosity) reactions may continue until we achieve zero retirement saving, very low skill levels, and high leisure enjoyment consistent with comprehensive government insurance during old age. The financing problem long-term welfare retarding problems that this presents should be obvious. Fortunately, the commission’s other Social Security reform elements help us to step back from this abyss.
Taken together, the first three items are projected to increase the Social Security system’s 75-year financial shortfall by 16 percent.
4) Increase the progressivity of the benefit formula: introduce a new bend point at the 50th percentile and reduce upper bend points gradually until 2050. This is a major Social Security reform element — it would plug 45 percent of the system’s 75-year financial shortfall. But because benefit reductions are larger for higher earners, it increases the fiscal system’s progressivity and adds to work disincentives: Working more won’t pay as much as before but would this encourage more work or less? Given that labor-supply elasticities are highest for upper earners, this reform element together with increasing the taxable maximum limit will increase, on net, marginal tax rates on the most skilled workers and is likely to be detrimental to economic growth in the long-term. A better way to realign benefits with revenues would be to shift from wage to price indexing in the benefit formula but leave the benefit formula’s progressivity unchanged.
5) Index retirement ages to life expectancy: This is a long overdue reform. Social Security’s retirement ages have remained relatively fixed whereas post-retirement lifespans have increase substantially. This has transformed Social Security from an erstwhile retirement insurance system to a retirement saving system…one that essentially substitutes government saving for private saving. But the reform does not go far enough. The proposal allows the current rate of increase in the normal retirement age to remain in place until NRA reaches age 67 for those born in 1960 (who will attain age 62 in 2022). Instead, the current rate of increase in NRA should have been accelerated and extended until the normal retirement age reached age 72 sometime by the mid 2030s. Given the considerably healthier and more active population of middle-aged and elderly today, this alternative policy would have made up for the long delay in increasing working lifespans and restoring the program to its original function of providing old-age insurance.
However, even better would be a complete elimination of early and normal retirement ages by introducing a personal accounts system — wherein the responsibility for determining one’s desired retirement living standard and generating the corresponding savings — beyond a minimum traditional Social Security benefit — would reside with individual workers. The reason for this is that statutory retirement ages together with retirement incentives prompt economically and socially inefficient (default-driven) early retirement decisions on the one hand, and deny adequate pay-back from Social Security to those with shorter expected lifespans.
6) Switching COLA calculation to the chained CPI index: Another sound and overdue reform element — one that prevents hidden increases in real Social Security benefits — the costs of which accumulate over time. This reform element would close 26 percent of the system’s 75-year financial shortfall.
7) Gradually increase taxable maximum to 90% of covered earnings by 2050: This reform proposal is motivated by the fact that the system’s taxable-to-total earnings ratio has declined in the past from 90 percent during the early 1980s to about 83 percent today. But official projections are notorious for the absence of projections (and even a reasonable methodology for making such projections) of how this share will evolve in the future. My independent and detailed analysis suggests that the share of taxable-to-total earnings will increase in the future under ongoing demographic changes — reaching as high as 88 percent by the mid-2020s. And would remain at that level until the 2060s. If that projection is accepted, the net improvement from this measure could end up being much smaller relative to the true financial baseline. The Commission’s estimate of a 35 percent contribution toward plugging the system’s 75-year shortfall from this measure is probably overstated.
8) Expanding system coverage to newly hired state and local workers: This is a standard element proposed by many other Social Security reform proponents to bring in more revenues from younger workers within the 75-year horizon. But scoring under a truncated 75-year budget window doesn’t count the benefit obligations created beyond that time horizon—leading to an overstatement of the improvement in the system’s financial condition relative to the true long-term improvement.
9) Slide 48 of the Commission’s reports but does not explicitly mention the interaction effects between the different reform elements: The overall improvement in the 75-year shortfall is reported to be 116 percent whereas improvements from individual reform elements add up to 119 percent. The difference probably arises from interaction effects when all elements are implemented together.
