My neighbors recently decided to encrypt their wireless signal, so I no longer get internet at home. Thanks neighbors! It seems, then, that over the weekend I missed a nice discussion on the benefits of privatizing Social Security for national savings.

It's important to clarify exactly what we're talking about when we say "national savings." Some commentators (none of the smart ones, though) have assumed that privatization will just dump trillions into the stock market, thus energizing Corporate America and sending us on our way to Dow 30,000 or whatever. But how, exactly, will this happen? Do the finger exercise. By allowing workers to put part of their payroll taxes into private accounts, the federal government will divert trillions from low-yield bonds to stocks. This will raise the price of stocks, thus lowering the return. Bond prices, meanwhile, will fall, thus raising yields. Hence, a number of investors will divert their own money from stocks to bonds, until the market settles at roughly the equilibrium it was at before. So Corporate America won't reap much benefit (aside from mutal fund managers who will earn plum investing fees.) The only difference is that putting trillions into the stock market will likely reduce or eliminate whatever equity premium is supposed to exist because of large-scale market failure. But you don't "juice" the stock market simply by moving money around.

Okay, onto actual savings. As Kevin Drum points out, it does no good to increase household savings if you decrease federal savings by ramping up the deficit. You're still crowding out investment. The main way that privatization can increase national savings, then, is if privatization makes it impossible for the federal government to paper over its debt with pension funds—a transparency which usually results in fiscal austerity. This is Brad DeLong's argument, and it's basically what happened in Chile, and savings shot up. But Chile also didn't have to deal with the hefty transition costs that we'll have to deal with, since it was growing very rapidly at the time of the switch, and its unfunded liabilities were quite manageable.

Okay, now leave those problems aside. Let's talk about household savings. Will privatization boost them? Not by a whole lot. An unrecognized fact is that Americans already save quite a bit. The confusion here, I think, is that national savings are presently calculated in a rather odd way. The National Income and Product Accounts is determined by subtracting household consumption and tax payments from total personal income. This completely leaves out wealth accumulation—houses, cars—that generate implicit income and also usually retain some equity value. Anyways, the point is that households are saving plenty, and if you include wealth the household savings rate is at somewhere around 40 percent, I think. So do we really expect that private accounts will force people to save even more? Most likely, having a private account will allow people—even low-income people—to free up income elsewhere and spend more. Meanwhile, people who are already saving will simply be able to move money around without really increasing their savings.

The only actual gain that I can see is if corporations manage to save more because they no longer have to contribute as much to their pension plans. But this assumes, as Brad DeLong likes to say, that there are free $1000 bills lying on the table that the government can pick up. So really, it depends on whether you think those bills are there or not. Personally, I don't know. But it seems like we should actually, you know, find out before embarking on a half-brained scheme to gut our national safety net.

At any rate, the only surefire way I can see to increase national savings is to levy some sort of consumption tax that decreases the value of existing assets, forcing people to save more in order to cover the loss. I just don't see how privatizing Social Security is supposed to do this.