Sen. Lindsay Graham is an entitlement reform hero – a guy who’s been willing to get out there on crucially important but politically unpopular issues. But in a discussion of his potential presidential campaign (probability 0.925, he says) he makes what I now consider to be a mistake on entitlement reform: fixing Social Security, he says, “so simple you could do it on the back of a napkin.”
I used to think that. You can go to a list of reform options – raising the retirement age, cutting COLAs, whatever you like – and patch them together until the savings are enough to erase the long-term deficit. And you’re done.
Here’s the problem: that approach assumes that Social Security is working perfectly, with the exception of being underfunded. In other words, Social Security policy effectively has one lever, with one end labeled ‘More Taxes’ and the other ‘Less Benefits.’
But if Social Security isn’t working perfectly then the Chinese-menu approach will cement in place problems that could be fixed. And, by being fixed, make Social Security work better even while we’re lowering costs.
For instance, Social Security leaves almost 1-in-10 retirees in poverty, despite spending $900 billion per year. We could give every retiree a poverty-level benefit and take the elderly poverty rate to zero for about half that amount. That’s pretty much what I’ve proposed.
Likewise, Social Security penalizes people who choose to delay retirement. If you continue to work, you’ll continue to pay taxes. But, on average, you’ll receive only 3 cents in extra benefits for each dollar of taxes you pay. So why not cut the payroll tax on older workers?
Similarly, the Social Security disability program could be reformed to encourage rehabilitation and re-employment rather than disability and dependency. But those reforms have nothing to do with tax rates or benefit formulas.
In other words, Social Security is an underfunded government program. Meaning, it’s underfunded and it’s a government program. We can make improvements on both ends of that equation.