The central idea behind the notion of "privatizing" part of Social Security is that stocks are a better investment than government bonds. If people are allowed to put part of their Social Security payments into shares in private companies, rather than having it all invested for them in U.S. Treasury IOUs (the argument goes), there will be more money available when they retire. Reveries of privatization tend to spend this extra money many times over. Last week's draft report of President Bush's Social Security reform commission hints that partial privatization could fix a looming Social Security deficit, supply working Americans with an extra retirement nest egg, and create an estate they can pass on to their children.
But do stocks pay more than government bonds? A familiar—and valid—criticism of privatization is that this may be true on average and over time, but will not be true for millions of individual privateers, many of them inexperienced sailors on the seas of finance. In fact, even the average stock market investor can be underwater for years at a stretch.
This familiar criticism actually understates the case. The prospect that investing Social Security tax payments in stocks instead of government bonds can be a free-lunch bonus for the average Social Security recipient or the system as a whole is not merely uncertain: It is mathematically impossible. OK, make that almost impossible. I have made this argument (for which I claim enthusiasm, not originality) in passing once or twice before, and defenders of privatization were either stunned into silence by its irrefutable logic or bored into silence by its obvious stupidity. So, let's try again, this time r-e-a-l s-l-o-w.
Start with this question. If stocks pay more than bonds, especially government bonds, why do people buy government bonds? Are they idiots? Well, yes, there is a school of thought that they are idiots. People buy bonds, especially government bonds, because they are considered safer than stocks. But some wizards believe that the riskiness of stocks has been misperceived and that by using sophisticated techniques it's possible to enjoy the higher return without the higher risk. They were even bringing out books a couple of years ago with titles like "Dow 36,000." If you think the whole Social Security system should take a flyer on those theories, I bid you adieu at this point in the argument.
The rest of us will stick to the conventional assumption that people who buy government bonds are not idiots. Which is fortunate, because if hundreds of billions of dollars in Social Security tax revenues are invested in stocks instead of helping to finance the national debt, the government will have to borrow that money somewhere else. Yes, yes, the government could also spend less, but the two issues are unrelated. If we want to cut spending, we should cut spending. Privatizing Social Security itself won't reduce total government borrowing by a nickel.
Therefore, privatization also will not increase the total amount of capital available for private investment. Every dollar of Social Security money that goes into private investment accounts instead of the U.S. Treasury will have an evil twin that is drafted out of the private economy to replace it in government service.
Now ask yourself: If privatization creates a bonanza for Social Security recipients, where did that money come from? There are only two possibilities. Either privatization must make the economy bigger somehow, or someone must be losing on the deal. There is a word for government policies that take money from some people to help others. That word, I'm afraid, is "tax." So, what about privatization making the economy bigger? There are also two ways this might happen. One is if the total amount of private investment increased. But we already have decided that dollars-out will equal dollars-in, for a net effect of zero. The other way is if the same-sized capital pool comes to be invested more wisely. You can believe, if you wish, that millions of financial innocents won't lose their shirts under privatization, but convincing yourself that their brilliant investment decisions will actually add vast billions to the economy is a more ambitious project.
So, if the Great Privatization Bonus isn't going to come from a bigger economy, either it will come out of someone's hide or it won't materialize at all. The most likely result is some combination of these. The stocks-are-better premise of privatization ignores the question of what happens when the government engineers a flood of capital from its own bonds into private stocks. Unless the laws of supply and demand are repealed at the same time, the return on stocks will go down, the return on government bonds will go up, and the gap between them will shrink. The more the gap shrinks, the more likely it is that people will be worse off than if they had settled for benefits based on government bonds. Meanwhile, those Social Security recipients who do end up better off will be enjoying a subsidy financed by what amounts to a tax on stock profits (which will be lower than if the government had not interfered).
One final point. The draft report of Bush's commission argues that Social Security's pile of government "bonds" is just a bunch of government IOUs to itself, which have no real value. It's certainly true that real stocks are better than phony bonds. But the premise of privatization is that real stocks are also better than real bonds. Furthermore, even "real" U.S. government bonds are just promises backed by nothing but the government's "full faith and credit." If the President of the United States takes up his Commission's argument and starts going around saying that the government might rat on its financial commitments, Social Security will be the least of our problems.