Regardless of whose numbers you use, Social Security needs to be fixed. This makes Sen. John McCain's (R-AZ) promise to deal with the program's coming deficits early in his administration, as well as Sen. Barack Obama's (D-IL) understanding of the program's fiscal situation, welcome news.

While updated Congressional Budget Office (CBO)[1] long-term projections for Social Security differ somewhat from earlier projections made by the Social Security Trustees[2]both show that the program has promised more in benefits than it can afford to pay. The main difference between the two projections is that CBO says Social Security will begin to run annual cash flow deficits in 2019 and its trust fund will be exhausted in 2049, whereas the Trustees' estimates are slightly earlier at 2017 and 2041, respectively. However, CBO's more optimistic result is certainly not an argument for delaying Social Security reform. No matter which date is correct, Congress will have to make painful choices if it fails to reform the program.

Between a Rock and a Hard Place

The salient feature of both the CBO's and the Trustees' estimates is that Social Security is headed for deficits in either 2019 or 2017. While CBO paints a somewhat rosier picture, failing to reform the program would still require severe benefit reductions or payroll tax increases.

CBO estimates projected revenues will be sufficient to cover only 84 percent of scheduled benefits in the year of trust fund exhaustion, whereas the Trustees place that figure at 78 percent. Either way, benefits would have to be reduced by a substantial margin in the future.

To prevent a benefit reduction and make the system solvent would require an immediate and permanent payroll tax increase from the current rate of 12.4 percent to 13.46 percent, under CBO's estimates. The Trustees estimate taxes would have to increase to 14.1 percent, but again, either scenario produces the same effect: a significant tax increase.

The Same but Different

Both CBO and the Trustees make it clear that the year of deficits is near, and it is encouraging that both presidential candidates have proposed plans to improve Social Security's finances during the next administration. Now that everyone has agreed on the important part-the program must be reformed-it is helpful to understand the disagreement between CBO and Trustees' estimates to frame the discussion that will ensue under the next administration.

As noted by Andrew Biggs of the American Enterprise Institute, CBO projections use several key alternate assumptions that explain the divergence from the Trustees Report[3]:

1. CBO Assumes Higher Productivity and Wages. Labor productivity has grown at 1.7 percent over the last 40 years. The Trustees assume productivity growth will stay at that rate in their estimates, whereas CBO assumes labor productivity will grow at 1.9 percent. This is a rosy assumption, and it has a significant impact on deficit projections. Even seemingly small changes in productivity can make huge differences in economic growth over a number of years. Higher productivity leads to higher wages, which will lower long-term deficits, since wages are the revenue base for Social Security;

2. CBO Assumes Higher Taxes. The Trustees assume taxes will remain constant as a share of GDP, whereas CBO assumes that taxes will rise dramatically as a share of GDP. The reason for this is that CBO follows current law, which assumes 2001 and 2003 tax cuts will expire and that the alternative minimum tax will not be indexed for inflation. The result is that income taxes on Social Security benefits will be higher, which has the net effect of lowering program expenditures. This difference results in lower deficits of 1.06 percent of taxable payroll in the CBO report, as opposed to 1.7 percent under Trustees' assumptions[4]; and

3. CBO Assumes Higher Interest Rates. The Trustees assume interest rates will be 2.9 percent above inflation, whereas CBO assumes it will be 3 percent above inflation. While, again, this may seem like a trivial difference, the fact that projections are made over a 75-year period means that this small change compounds to a large difference. For instance, under CBO's calculations, higher payments will be made to the Social Security trust fund, extending its life and pushing out the date of final exhaustion.

Outlook Does Not Look Good

Given the uncertainty involved in long-term economic projections, the fact that CBO and Trustees' estimates of the year of deficits and insolvency are actually fairly close together leaves little room to doubt that Social Security is in trouble. Even under CBO's rosier scenario, the benefit cuts or tax increases that would be necessary to balance the system would be severe. No matter what set of assumptions is used, it is clear that the next Congress and the next president must take the need for reform seriously.

Nicola Moore is Research Coordinator in the Thomas A. Roe Institute for Economic Policy Studies at The Heritage Foundation.