The first thing to understand in any discussion of the government is that there is no promise that the government makes that can't be broken. The promise may be important enough to enough voters that the government will try not to break it, but there just aren't any guarantees.
The assets that have accumulated in the Social Security Trust Fund are simply a promise by the Federal government that the moneys that it has collected through taxes will someday be used to pay Social Security benefits. It is not impossible that this will actually occur. But it's only a promise and it will only be honored if honoring that promise is less painful than the alternatives.
Sometime around 2018 (if I remember correctly), the current projections show that we will start taking in less money in Social Security taxes than we have to pay out in current benefits. Although there will be a certain amount of pain before then (because the current Social Security surplus is being used to finance deficit spending and the available annual surplus is going to be declining), that's when the real pain is going to start. The Federal government is going to need to start generating dollars to pay those benefits with.
One option is to just run the printing presses and inflate the currency. But that's a pretty painful option, as students of history and hyperinflationary economies in more recent times are familiar with. It's probably the least likely choice, just because it sucks so badly. But it's always a possibility, so I may as well enumerate it.
Alternatively, we could just borrow the money from someone else. Of course, this requires that there be someone who is willing to lend us the money. It's not clear that there will be anyone who wants to lend us that money on terms that we'd like. If we didn't have such a large national debt, then we'd have more debt capacity available for borrowing when we need it in the future. But we do have a nasty amount of national debt, we're running large deficits, and there's no sign on either side of the aisle of the necessary will to try to do something about those current deficits.
This is, by the way, why deficits matter -- at least, to my mind. While there's certainly been some justification for substantial deficit spending during the first term of Bush's administration (as in, a recession and the problem of terrorism), excuses start to wear thin after a while. Actually, it's not so much the excuses that start to wear thin -- it's that not doing something about the deficit is going to be more painful than doing something about it.
Reforming the tax code and cutting wasteful spending are both good ideas. Sadly, the definition of "reforming the tax code" means raising taxes on someone who isn't me; while "cutting wasteful spending" means getting rid of spending that doesn't benefit me. Since we're an entire nation full of "me", this makes these projects difficult.
But I digress. The point is that we could make some difficult choices now that would improve our chances of borrowing money to pay Social Security benefits in the future, but that just doesn't seem likely to me unless we can convince people that they should put up with the pain now.
We could raise taxes or cut spending at some future date in order to pay Social Security benefits. This has the advantage -- for current politicians and retirees -- of postponing the pain, but it will probably just make it more painful when we get to that point than it would be if we could fix the problem now.
Of course, one way to cut spending is to reduce Social Security benefits in the future. The most likely way of doing this is through "means testing". What does this mean? In the simplest form, it means that if the Federal government decides that you don't need Social Security benefits, you won't get them. When this is discussed, you will hear that "we'll just take them away from the rich". The problem with that statement is that there aren't enough "rich" people to make a difference in the budget. Willie Sutton robbed banks because "that's where the money is". I believe that people who are relying on Social Security benefits to be the difference between an "ok" retirement and a "comfortable" retirement are going to find that they have an "ok" retirement, because means testing is going to slide down the economic curve once it's in place. That's where the money is, sitting in the fat part of the bell curve of retirement incomes.
(This sort of logic also makes me worry that the government will use some sort of means testing to tax our supposedly tax-free gains in our Roth IRAs. But I could be excessively cynical. Or not.)
All of these options (save for current fiscal and tax discipline) have one wonderful feature for current politicians: they'll be retired when the bills come due. (And their retirement plan is a good bit better than Social Security. Why is this not a surprise?)
This is why finding Social Security reform that forces current politicians to accept some of the pain now is a good idea, because they will not do it if their constituents give them a choice in the matter. The only Social Security reform that I have heard of that might do this is to divert some of the current tax receipts from FICA into assets that are not simply a promise that the Federal government will give the money back some day.
Note that this does not necessarily imply any form of privatization of Social Security. We could gain many of the same benefits that privatization promises simply by having the Federal government run its own index fund, buying shares across a wide swath of the American economy. This would have its own set of problems, not the least of which would be that I'm scared of how those yahoos in Washington DC might decide to vote the shares. (But there are ways around this problem, which would introduce different bureaucratic problems -- for instance, "Congratulations, Mr. Smith. You're voting 100 shares of GM for the next year as your share of the Social Security Trust Fund. Try to do something sensible, ok?")
Because the current Social Security receipts would no longer be available to finance deficit spending, this improves the chances that Congress and the President will have to make some of the hard budgetary and tax choices now. I'd think that would be a good thing all by itself.
But there's another advantage: in the long run, the stock market will earn a higher rate of return than Treasury securities. Yes, I know that the market cratered in 2000 when the Internet bubble burst and that there are risks involved. If there was no risk involved, then you would not get a better long-run return. That's one of those key economic principles that people would like to forget. (And if you'd like to find out more about it, go Google on "Capital Asset Pricing Model". That'll get you started.) Over the long-run, those risks tend to cancel out, giving you a higher long-run return than you'd get by investing in less-risky assets. In this case, risk means "volatility", the chance that the asset will be worth either more or less than you would forecast at some time in the future. Risk-free securities -- like Treasury bonds -- are completely predictable. You know how many dollars you'll be getting back if you hold the bond to maturity. (That's not true if you sell before maturity, due to interest-rate risk, but that's a different problem altogether.)
On average, you're better off buying equities. You just have to be able to wait out the ups and downs of the market. And on average, Social Security would be better off buying equities. The trick is to figure out the best way to do it.
We still don't know all of the details of Bush's privatization plan. It might be good. It might be just awful. But I can put together some criteria to judge it by.
1) Does it reduce the long-run obligations of the Social Security system?
2) Does it cut (or raise!) the expected average benefit?
3) Does it cut the expected minimum benefit?
4) How are the risks in the plan shared between the government and the people?
5) How much current pain will it cause? (As in, more or less than we can stand?)
If the plan doesn't do (1), there's no reason to look at it at all. If the expected average benefit goes up, that's a feature. If the expected minimum benefit leaves people impoverished, that's a bug.
I'll toss out an idea here that I've tossed out elsewhere:
Suppose that the government guaranteed that you could not lose by setting up a private Social Security account. The government will share in the risks and rewards of your account. If your account would allow you to receive a higher benefit than Social Security would have provided, the government will tax away half of that surplus. If your account would leave you with less than Social Security would have provided, the government will make up the difference.
I believe that such a scheme would leave everyone better off. On average, the Social Security Trust Fund will earn higher returns than it would have if left in Treasury securities. And the government will share in those returns and use part of them to help the "losers" that we hear about when privatization is criticized.
It probably doesn't match up with Bush's plan. But it doesn't have to.
It's my plan.
And I like it.
(I hope I like Bush's plan nearly as well.)