Social Security’s finances are deeply troubled. In order to pay all of Social Security’s benefits on an ongoing basis, the average American worker needs to immediately start paying a nickel more of every dollar earned in payroll taxes. This assessment comes from Social Security’s own actuaries and is based on their “intermediate” assumptions. Under more pessimistic, but arguably more realistic assumptions, average workers need to pay an additional 7 cents per dollar earned. Since Social Security is currently taxing average workers 12 cents per dollar earned, the immediate and permanent tax hike needed to avoid defaulting on benefit promises ranges between 40 and 60 percent!

The magnitude of Social Security’s fiscal problem invites careful attention to the actuaries’ forecasting method. It also invites attention to the System’s likely ultimate treatment of America’s workers. This paper looks at both issues. It first lays out Social Security’s projection methodology. It then compares this methodology with a dynamic microsimulation approach to forecasting future benefits and taxes. Next, it presents a specific dynamic microsimulation model that combines two tools: CORSIM -- a socioeconomic microsimulation model -- and SSBC -- a detailed Social Security benefit calculator. The paper uses this model to examine Social Security’s benefit projections based on its low, intermediate, and high cost assumptions. Finally, the paper uses the model to compute internal rate of return and lifetime net tax rates for postwar Americans under a) the actuaries’ low, intermediate, and high cost assumptions, b) the assumption of an immediate and permanent tax hike sufficient, under intermediate assumptions, to meet benefit commitments, and c) the assumption of an immediate and permanent benefit cut sufficient, under intermediate assumptions, to eliminate the System’s long-term fiscal imbalance.

Our findings suggest three things: first, there may be a somewhat smaller imbalance between long-run benefits and payroll tax revenues than is being forecast by the actuaries. Although our projections of aggregate taxes are quite close to those of the Social Security actuaries for years when comparisons are meaningful, our projected benefit payments are roughly 15 percent smaller, for years when we can form comparisons given the nature of our data. Second, assuming no change in current law, postwar Americans will earn less than a 2 percent real rate of return on their contributions to Social Security. If one assumes that postwar Americans could otherwise invest their contributions at a 5 percent real rate, this below-market rate of return translates into lifetime net Social Security tax rates of more than 5 cents per dollar earned. Third, implementing the tax hikes or benefit cuts that the actuaries suggest would resolve fully Social Security’s long-term fiscal imbalance will leave today’s newborns receiving a real return on their contributions of only 1 percent or thereabouts.

Steven Caldwell
Cornell University

Alla Gantman
Boston University

Jagadeesh Gokhale
The Federal Reserve Bank of Cleveland

Thomas Johnson
Cornell University

Laurence J. Kotlikoff
Boston University

[Home]