The trustees overseeing the finances of Social Security and Medicare issued their latest report on Friday, declaring that a) the Social Security Trust Fund is expected to run out of money in 2035, the same estimate as last year; b) Medicare’s hospital trust fund is expected to run out of money in 2026, a two-year improvement over last year’s estimate; and c) the Disability Insurance Trust Fund is expected to run out of money in 2016, just as projected last year.
Here’s why I don’t believe that’s the correct way to think about these numbers.
The Social Security trust fund ended 2012 with $2.6 trillion in assets. Under current law, the trustees anticipate this account to be drawn down gradually and become negative after 2035, an event that is sometimes referred to as “running out of money.” But the $2.6 T in current assets consist of nothing more than a big I.O.U. from the U.S. Treasury to the Social Security trust fund. Where are the assets that the U.S. Treasury is holding that would enable it to make these payments? They don’t exist. Taxes will have to be raised, other programs cut, or the Treasury will have to borrow more from the public in order to deliver the funds that Social Security is assuming it’s going to be receiving from the Treasury between now and 2035.
Where did the $2.6 T balance come from? It represents the accumulated amount by which Social Security taxes generated more money for the government than have been spent on beneficiaries up to this point. But that surplus from Social Security didn’t go to acquire any real asset that could be used at this point to pay beneficiaries. The money has already all been spent on other programs. If the Treasury is going to pay this $2.6 T to Social Security, the funds have to come from tax increases, spending cuts, or more borrowing.
The question of whether this $2.6 T should be counted as an actual asset of the Social Security trust fund is related to the question of whether it should be counted as an actual liability of the U.S. Treasury. Many of us prefer to think in terms of a unified federal government balance sheet. From that perspective, the $2.6 T represents money that the government owes to itself, and should be counted as neither an asset nor a liability of the federal government. This is why many economists summarize the debt load of the United States in terms of net debt, which excludes the money that the Treasury owes to other government programs like Social Security, rather than gross debt, which includes those intragovernmental obligations. Net U.S. federal debt came to $11.6 T, or 74% of GDP, at the end of 2012. Gross debt was $16.4 T, or 105% of GDP.
It is interesting that the same people who insist that net debt is the correct summary of the government’s indebtedness are also are the ones telling us that Social Security will be adequately funded through 2035. But you can’t have it both ways. It can’t both be true that the $2.6 T is an asset of Social Security and yet somehow not a liability for the U.S. Treasury.
What I believe is the correct perspective is that the $2.6 T is neither an asset nor a liability for the federal government. There is, however, a significant unfunded off-balance-sheet liability associated with Social Security. Current and future retirees are expecting payments for which a source of funding has so far not been identified. The $2.6 T is an estimate of the size of the present value of the additional funding that would be needed to take us through 2035 given current Social Security benefit and tax schedules.
And what about beyond that? Another interesting calculation is the following. Suppose you took all the people who are over 15 years old today and asked what would be the present value of the cost of providing Social Security benefits to that group based on the current schedules. The answer is given in Table IV.B7 of the report– $52 T. For comparison, the present value of Social Security taxes collected to cover those payments is calculated to be $25.5 T. In other words, providing all the benefits that have been promised to current program participants (both workers and retirees) requires coming up not just with an additional $2.6 T but instead somewhere finding an extra $26.5 T.
Now, you’re free to propose some other (big) numbers for the size of the unfunded liability associated with Social Security. My point is not to argue for a particular number, but rather to call attention to the fact that the growing ratio of retirees to workers poses a major fiscal challenge for the United States. This reality should give pause to those who insist that the debt that we run up today will somehow become easier to manage at some date in the future, particularly to those who base their calculations on the net rather than the gross debt figures.
And I haven’t even gotten to the numbers from the Medicare report yet.