At his blog today, Ezra Klein takes issue with a suggestion by Andrew Biggs that we improve government finances by raising the early eligibility age (EEA) for receiving Social Security benefits. The EEA has remained at 62 despite the ongoing increases in the full benefit age (FBA, usually called the normal retirement age) from its historic level of 65 to 66 today and 67 within a decade. Both Ezra's and Andrew's contributions are worthwhile. I can add the following six points to the discussion:
1) Social Security exists to prevent people from outliving their means of supporting themselves.
The core program is for Old Age and Survivors Insurance. None of the ages being discussed -- 62, 65, 67, even 70 when all age-related aspects of the program max out -- constitute old age. What matters for the effectiveness of Social Security as an insurance program is whether it keeps the truly old -- think age 85 and almost certainly unable to work -- out of poverty.
2) If the FBA is going up, eventually the EEA will have to go up as well.
Retiring at 62 when the FBA is 67 reduces benefits by an additional 10% relative to doing so when the FBA was 65. You might think that people can weigh those tradeoffs appropriately and decide if the additional 10% is worth it. Biggs writes, "The evidence indicates that most Americans could work longer and would benefit from doing so." Klein writes, "The words 'would benefit from doing so' are doing some heavy lifting here, and in a way even I find worryingly paternalistic."
In that case, welcome to the party. I find a lot of what Social Security does to be worryingly paternalistic. That doesn't mean it isn't necessary social policy to have mandatory Old-Age and Survivors Insurance. As I said above, I judge Social Security's effectiveness by whether it helps avoid poverty for those who are so old that they cannot reasonably be expected to work, not whether it accommodates a preference for many 62-year olds (or even a need for some 62-year olds, see #4 below) to claim benefits at that age. The additional 10% benefit cut applies as well to benefits that the worker (or, more likely, a surviving spouse) will be collecting at age 85. I'm not in favor of allowing that cut at that age, because the rest of society is on the hook for the resources that then must be applied to keep the 85-year old beneficiary out of poverty.
3) If we won't consider raising the EEA, then at least consider changing the indexation of benefits after intial claiming.
To find room for a compromise, perhaps what is required is better indexation of post-retirement benefits. Apply not a 30% reduction (the total decline from FBA at 67 to EEA at 62) but, as an example, a 40% reduction. But then allow for indexation at CPI+1% in each subsequent year. By age 72, the cut would be the same 30% as under current law. By age 82, when it really matters, the cut will only be the same as the 20% that prevailed historically when the FBA was 65 and the EEA was 62. This allows us to accommodate the "preference" that workers have for retiring as soon as possible without jeopardizing the ability of Social Security to do what we count on it to do.
These numbers are just illustrative -- we would likely want to ensure that this is revenue-neutral, and my view about post-retirement indexation doesn't apply to just the benefits that are taken starting at age 62.
4) Do we trust the Disability Insurance program as a screening device?
Social Security includes not just the Old-Age and Survivors Insurance but a Disability Insurance program as well. The DI program enables workers to leave the workforce and claim benefits immediately and without the reductions to benefits that are applied for early eligibility. The DI rolls have been increasing over the past couple of decades. The counterpart to DI in many Western European countries already serves as a de facto early retirement program. And the pressures on DI are likely to increase as the new health care law becomes operational -- particularly, the parts of that law that make health insurance more widely available for older workers outside of employment relationships. If you trust the DI application process to get unreduced benefits to those who are unable to work (and not to those who are able), then you should be less concerned about raising the EEA.
5) The revenue gains from raising the EEA won't be found in Social Security as much as in the rest of the budget.
Yes, people will be paying payroll taxes at ages 62-64 if they are unable to claim (and in many cases unable to retire) until age 65. But their Social Security benefits will also be going up, so the net effect is small. However, while working, they are also paying income tax on their earnings -- that's the big item.
6) The idea of raising the EEA isn't new.
As just two examples, Jeff, Maya, and I proposed it in the Liebman-MacGuineas-Samwick plan back in 2005. And here's a post from 6 years (and 1 day) ago in which I describe increases in the FBA (beyond what is currently being phased in) as "How to Reform Social Security." I laid out a number of the same concerns as above, along with this addtional one:
3. The Bigger Hit to Low-Earners. People with higher incomes over their lifetime tend to live longer. Raising the Normal Retirement Age therefore disadvantages lower income workers more than higher income workers. This impact can be offset by changes in the benefit formula to give higher replacement rates for lower earnings, paid for by lower replacement rates at higher earnings levels.
More about Social Security in the weeks to come.