Shaping the debate on Social Security reform

• Article
• Comments ()
• Click-2-Listen
• Videos

By

Politics aside — ha, ha, ha — the Social Security debate over private accounts comes down to a question of an unfunded liability vs. a partially (or fully) funded liability. For those who oppose change, the unfunded liability of existing government promises is good enough: Though the Supreme Court has noted that there is no legal right to Social Security, in this view, there is a sort of "social contract" that promises Americans an old-age pension that grows from current levels with average wage growth.

That's that: Government will make good on its promises through the existing pay-as-you-go system. Anything else somehow "weakens" the system by diminishing political support for the universal old-age benefit (as well as the system's other benefits for survivors and the disabled).

But exactly how a system in which private accounts partially or fully fund benefits undermines this social contract is not clear. The idea is perhaps that those with fatter private accounts will be less inclined to support sustaining current benefit levels for those with smaller private accounts, specifically, those at the lower end of the income spectrum. I find that hard to follow. Wealthier retirees, those with substantial "private accounts" utterly apart from the current system, nevertheless seem to support existing benefit levels. I know that many Democrats, even rich Democrats, have an instinct to regard the rich as heartless, but seriously: What is the evidence?

So we come back to the question of whether, all things being equal, we're better off with a system in which we have a partially or fully funded liability, or an entirely unfunded liability. The question answers itself. If a national pension system were being created today from scratch, no one could propose with a straight face that it be funded with a payroll tax on a pay-as-you-go basis in a world in which there will soon be only two workers for each retiree. No, of course you'd fund it. You'd want to harness the compelling power of compound interest. That, finally, is the best argument for private accounts, ones that invest in real assets that will appreciate over time — with appreciation on the appreciation as well.

The problem with the current proposals for reform, in my view, is that they don't take as much advantage of compounding as they could or should. Essentially, under current proposals, you begin funding your private account when you start paying Social Security taxes, which is to say, when you start working. Say you're 22, a new college grad and, lucky you, you get a starter position at a Washington think tank for $25,000 a year. You get to put (say) 4 percent of that sum into a private account: $1,000. Say you'll be retiring at 67. Say between now and then an index fund returns on average (a very modest) 6 percent. Your money doubles every 12 years. That $1,000 becomes $2,000 at age 34, $4,000 at age 46, $8,000 at age 58, and more than $13,000 when you turn 67.

Not bad — but: Suppose that same $1,000 dollars had another 22 years to compound? Rather than $13,000, it's more like $48,000. No, I'm not proposing that people work until age 89 before they get Social Security. I'm proposing starting the accounts at birth.

One of the things children have in common is, generally speaking, parents — at least one, often two. Why not fund a personal retirement account for the kid out of the social security taxes due from the parent? There are many ways you could do this. The maximum Social Security tax in 2005 will be $5,580 from the individual and the same amount from the employer. (That would be on an income of $90,000 or more.) Lower incomes are taxed at a rate of 6.2 percent from the individual matched by another 6.2 percent from the employer, so our 22-year-old making $25,000 would pay $1,550 and her boss would pay the same.

You might, for example, fund the kid's account with $1,000 per year from mom's or dad's Social Security taxes for the first five years of the life of the kid. Now, if you managed to do that, by age 67, the $1,000 dollars from year one would amount to, back of the envelope computation, $48,116 (more if compounding is computed more frequently than annually); from year two, $45,392; from year three, $42,823; from year four, $40,399; from year five, $38,112. That's a total of $214,842 — which is a superb funding base for future liability. For those making less than an amount that would fund $1,000 a year, government could step in to equalize the playing field.

Or, you might fund the account with the full maximum Social Security contribution of the parent the year the child is born: $5,580 from year one. That alone turns into $268,488 at age 67. There are about 4 million babies born each year. If the whole cost was a dead loss in revenue to government, you could do it for $22 billion a year.

The point is simple: The more money the sooner, the more dramatic the result upon retirement.