The Honorable Pat Toomey
502 Hart Senate Office Building
Washington, D.C. 20510

Dear Senator Toomey:

I read through the talk on the budget that you gave at the Brookings Institution this week. The talk included several comments on Social Security that were at least misleading, if not actually wrong.

First, on page five of the transcript you lumped Social Security in with Medicare and other health care programs and said that collectively they are unsustainable. This is misleading for several reasons.

Most importantly, the cost of Social Security is projected to rise much less rapidly than the costs of health care programs. And, after the mid-2030s, Social Security’s costs are projected to remain virtually constant as a share of GDP through the rest of the century. There is nothing about these projections that imply this will be an unsustainable burden.

Furthermore, because of the way in which Social Security is financed, under the law it cannot contribute to the deficit. The Congressional Budget Office’s (CBO) most recent projections show that the program will be fully funded from its dedicated stream of tax revenues through 2038, with no changes whatsoever. (The Social Security Trustees project 2033 as the date of trust fund exhaustion.)

This means that for the next quarter-century, CBO projects that Social Security will be fully funded from its designated tax and the interest and principal from the bonds bought with surplus revenue from this tax. If we reach 2038 and the fund is depleted as projected, then under the law Social Security would not be able to pay full benefits. (The payable benefit would be about 80 percent of the scheduled benefit, which would still be considerably higher than what current retirees receive.) Social Security would not be able to make payments in excess of the money coming into the system, and thereby add to the deficit, unless Congress were to vote to change the law and allow Social Security to spend from general revenue.

Therefore it is simply wrong to claim that the baseline path is in any way unsustainable. It is possible to design reasonable changes that would allow the program to pay full benefits for the rest of the century. For example, removing the cap on taxable wages would largely accomplish this task, as would an increase in the payroll tax roughly equal to 5 percent of projected wage growth over the next three decades. Meanwhile, under current law the program cannot add to the deficit.

For this reason, lumping in Social Security with health care programs, whose costs are projected to rise rapidly, is seriously misleading. It is perhaps worth noting that the cost of these health care programs stems from projections of rapidly rising private sector health care costs. The government, through programs such as Medicare, pays for approximately half of the country's health care, almost all of which is actually provided by the private sector.  The baseline projections show health care costs exceeding 20 percent of GDP by 2022 and continuing to rise as a share of GDP throughout the century. In today’s economy that would imply health care costs of more than $34,000 for a family of four. This rate of health care cost growth is clearly unsustainable, whether it is paid by the public or private sector. It would seem that the responsible position would be to focus on bringing U.S. health care costs under control, something that every other country in the world has managed to accomplish, rather than cutting the public sector health care programs.

Second, your comments on the chained consumer price index were also misleading. You said, “it is almost universally acknowledged as a more accurate way to measure real inflation as it occurs to real consumers.”

This is misleading because we know that all demographic groups do not experience the same rate of inflation. The Bureau of Labor Statistics (BLS) has compiled an “elderly index” for more than two decades that tracks the cost of living for the elderly population. This index has consistently shown a somewhat more rapid rate of inflation than the overall consumer price index (CPI).

Congress’ original purpose in indexing benefits to the CPI was presumably to ensure that benefits kept pace with the expenses faced by seniors, not an abstract measure of inflation that might include many items that seniors never consume. It can be argued that the elderly index is simply an experimental index, and not a comprehensive measure of the rate of inflation experienced by seniors. (It does not examine the specific items purchased by seniors at the outlets they patronize.) However, it would seem that if Congress is interested in ensuring the accuracy of the cost-of-living adjustments, then it would have the BLS calculate a full elderly index, not simply switch to a chained CPI.

Switching to a chained CPI, as you propose, would undoubtedly result in lower Social Security benefits. With the average beneficiary receiving benefits for a bit more than 20 years, the switch to a chained CPI would be equivalent to a 3 percent across the board cut in benefits. Perhaps you and other members of Congress want to endorse such a benefit cut, but we should then be clear that changing the index that is used to make cost-of-living adjustments is about cutting benefits, not increasing accuracy.

I hope you find the points above informative. If you would be interested in discussing any of these issues, I would be happy to talk to you or your staff.

Site Maintenance

"The CEPR website currently takes longer to load than usual. We hope to have this and other issues addressed shortly. While this much needed site maintenance is taking place, our content is still available so please continue to slooowwwly surf the pages of our site. Thank you for your patience."