Social Security From The Ground Up

I'm not sure it makes sense to take a position on Social Security privatization per se because "privatization" could mean any number of things, some of which might be better than the status quo and some of which might be worse. It occurs to me, though, to think of this. Say we didn't have Social Security or some equivalent program and you were a liberal trying to design some government programs to help out old people. What would you do?

One thing you would definitely do is put some kind of health care plan into place. It's unlikely that you would come up with the current Medicare system, since it treats prescription drugs so oddly and has other problems, but we'll just bracket that and call it "Medicare." What else would you do? What you almost certainly wouldn't do is create a huge intergenerational wealth transfer scheme funded by a very high and regressive tax that pays out money according to a formula whose mild progressivity is undermined by the way it favors the (predominantly wealthy) long lived over the (predominantly poor) short lived.

Instead, you would want to put in place some kind of "welfare for old people" such that a citizen who has worked hard all of his or her life and now is too old to work can live out the remainder of his or her days in dignity.

Such a program would be like Social Security in many respects -- pay-as-you-go financing out of current tax receipts and based around guaranteed monthly checks. But instead of the benefit level being calculated based on how much you used to pay in taxes when you were working, it would be based around how poor you are. The poorest people would get the most generous checks, and it would phase down so that well-off people weren't getting any public assistance. If you indexed benefit-growth to the growth in prices rather than the growth in wages you could make this "welfare for old people" program relatively insensitive to demographic shifts toward an aging population since wages grow faster than prices (barring severe macroeconomic problems that would stress any social spending program) thus making the burden something the economy could support more-or-less come what may.

At this point, if you were smart, you'd worry that you'd created a serious moral hazard problem. Rich people would still save for retirement, since they'd have nothing better to do with their money. But low- and middle-income people who periodically have trouble making ends meet during their working years would save little or nothing for retirement, secure in the knowledge that future taxpayers would foot the bill for their golden years. One good way to address this would be through a forced savings scheme. All workers would be made to contribute a fixed portion of their payroll income to a tax-free retirement account. Since the idea here would be to try and minimize the extent to which your "welfare for old people" scheme needs to be utilized, you would want to prevent people from making extremely high-risk investments unless their forced nest egg was very large. Thus, the first X number of dollars per year in forced savings would have to be invested through a limited menu of schemes. The Thrift Savings Plan for federal employees is, or so I understand, a good model for such a plan.

You could make the forced savings plan progressive by incorporating some fully refundable tax credits into the income tax system so that, de facto, lower-income people would be having their savings subsidized by higher-income people. You would make it so that people couldn't ordinarily draw down their savings account until they reached a certain age, and you would limit the amount of money that could be withdrawn in any given year up until then. You might also want to make it possible to borrow against your retirement account for the purpose of other equity-building activities such as purchasing a moderately-priced first home.

You're looking, then, in present terms at the conjunction of the following proposals -- partial privatization, means testing, and price indexing -- all of which are elements of the contemporary debate. The upshot would be a sustainable, progressive program that was, in large part, pre-funded. It would, moreover, have the benefit of giving relatively short-lived people real assets that could be passed down to their children or other family members.

Now by abstracting away from the fact of actually-existing Social Security you make the task easier in several ways. Fiscally, this simply eliminates the transition costs which, in the real world, are large and important. Second, you alter the political expectations so that a move from the hard-to-justify wage-indexing system to the price-indexing system isn't read as a "cut." Third, and perhaps most importantly, you're assuming the plan would be designed by well-meaning people and not the employees of George W. Bush who will doubtless put forward something similar, but much more scam-ish. So, as I say, better to withhold judgment until we see what, if anything, the president has in mind. But I'm not unsympathetic to something in the neighborhood of what he seems to be contemplating.