I wrote yesterday that Progressives should lead the way to fix Social Security now, both because it needs fixing as well as to focus the deficit debate on health care costs (the main driver of the long term deficit). With Social Security fixed, opponents of the Affordable Care Act (ACA) will no longer be able to get away with only saying what they are against, but instead will have to make clear what they are for. Then perhaps a policy-oriented health reform debate will ensue.
Several wrote to me asking that I say more about what I like about Jed Graham’s suggested Social Security reform. What I most like about Graham’s ideas:
- His old-age risk sharing proposal provides increased protection against outliving your private retirement savings. Old age risk sharing sets a lower level of benefits at younger retirement ages but the amount of your benefit rises with age, addressing the very real problem of persons outliving their private savings.
- It incentivizes people to work longer but maintains a more flexible safety net for those who need to retire younger. In short, the average person would get a bigger benefit penalty for early retirement, but the penalty wouldn’t remain constant for the rest of their life. For example, if the retirement age is moved to 70 without old-age risk sharing, someone who retires at age 62 would receive 81.4% of the full (age 70) retirement benefits for the rest of their life (receive same amount at age 62 as age 92). Under old age risk sharing with a standard retirement age of 68, a person retiring at age 62 would receive 57% of full benefits in that year, but at age 70 it would be 87%, rise to 96.9% at age 80, and be 100% of full benefits from age 85 on. So, retiring early wouldn’t ‘haunt you’ for the remainder of your life. Old age risk sharing is simply more flexible than an across-the-board raising of the retirement age. These figures are based on a person with average earnings defined as $42,000 in 2009.
- It is a progressive reform given the constraint of no payroll tax increase. My understanding of Graham’s plan is that it fixes Social Security without raising the payroll tax in any way (rate or amount of wages to which it applies). The early retirement penalty would be less for lower income workers as compared to the average worker example noted above, and higher for workers with higher incomes. Because lower income persons tend to have a shorter life expectancy, this lessens that differential amount of lifetime benefits they will receive, certainly as compared to an across the board retirement age increase. These changes are solidly progressive.
Under current law, when all the trust fund bonds have been redeemed (sometime around 2037) then Social Security outflows will have to match payroll tax receipts which would require across the board benefit cuts of around 20-25% with the retirement age at 67 where it is currently set to rise by 2025. Similarly, it would take an increase in retirement age to between 71 and 72 (I think I have that correct; a bit back of the envelope) to maintain current benefits with no increase in taxes. Given the constraint of no payroll tax increase, Graham’s plan is certainly preferable to both of these options, and far more progressive.
What would I change about Graham’s proposal? I would add Richard Pozen’s suggestion of increasing the amount of wages subjected to the Social Security payroll tax. He suggests the tax should max out at the 90th percentile of wages, which is where is was tagged in 1983 by the Greenspan Commission (but has not kept pace because of how it was indexed). This year, that would mean an increase from $106,800 to $170,000. What could be done with this extra revenue?
- Minimum benefit levels could be increased
- The degree of early retirement penalty could be lessened
- The extra revenue could be used to reduce the size of the annual deficit
- The payroll tax rate could be lessened for lower income workers, which would make the payroll tax less regressive
My preferred alternative would be to boost the minimum benefit for low income persons and use the remainder for deficit reduction. I am operating on the assumption that a substantial tax reform is needed and inevitable, and further that if we are ever going to have a balanced budget the proportion of GDP collected in taxes will have to rise over what is currently collected today. In the short run, the increase in the wage limit of the payroll tax is a placeholder for that larger discussion. These are my preferences for Social Security reform, but my strongest preference is that we develop a compromise Social Security fix sooner rather than later so that we can turn our full policy attention to the biggest long range deficit problem: health care costs.
Note: the post uses Graham’s numbers, I have not done any empirical analysis. I have also ordered, but not read his book A Well-Tailored Safety Net.