Privatization enthusiasts sometimes admit to this as a transitional problem. But it is not a transitional problem. It is the entire problem. As Brookings economist Henry Aaron pointed out in the "Committee" discussion, if you could pour in enough money to pay for the "transition" to privatization, the system would no longer be out of balance. The problem, or crisis, would be solved. Privatization would not be needed.
You have to watch closely as the privatizers describe their schemes. They play a shell game: Take this part of the Social Security tax and convert it into a mandatory savings contribution; take today's benefit and divide it into two parts; give everybody this new account and that minimum guarantee; take this shell and put it there and bring that shell over here, and--hey presto! But no matter how you divvy it up, the same money can't be used twice. That is the problem with Social Security now, and it is the problem with privatization. All you've said when you endorse privatization is that if there were water, it should be used to make lemonade. Not obviously wrong, but not terribly helpful.
Privatization is a shell game in a second way. It is supposed to bring more money into the system because returns on private securities are generally higher than returns on government bonds. Even members of the recent commission who oppose full privatization supported investing part of Social Security's accumulated surplus in the private marketplace. But (as Stein pointed out in the "Committee"), every dollar Social Security invests privately, instead of lending to the Treasury (as happens now), is an extra dollar the government must borrow from private capital markets to finance the national debt. The net effect on national savings, and therefore on overall economic growth, is zilch. Every dollar more for Social Security is a dollar less for someone else.
If Social Security manages to achieve a higher return, by investing some or all of its assets privately, the rest of the economy will achieve a lower return, by having more of its assets in government bonds. In essence, the gain to Social Security will be like a tax on private investors--an odd thing for conservative think tanks to be so enthusiastic about. Also, the arrival of this huge pot of money looking for a home will depress returns in the private economy, while the need to attract an equally huge pot of money into the Treasury to replace the lost revenue will increase the returns on government bonds. Result? The diversion will be at least partly self-defeating.
The size and security of future retirement benefits ultimately depend on the country's general prosperity at that time in the future. Checks to be cashed in the year 2055 (whoops! there's a number) will be issued in 2055, whatever promises we make or don't make today. The most direct way for Social Security to affect future prosperity (as Aaron pointed out in the "Committee") is to increase national savings, of which the Social Security reserve is part, by trimming benefits and/or increasing revenues. Since we have to do that anyway--even as a prelude to privatization--why don't we do it first? Then we can argue about privatization at our leisure.