In a 2011 briefing paper, I estimated that weak and unequal wage growth accounted for more than half the projected Social Security shortfall that has emerged since 1983. Despite their importance, wages have received much less attention than life expectancy, a relatively minor factor hyped by alarmists like former Senator Alan Simpson, co-chair of President Obama’s Fiscal Commission.

Demographic trends affecting the worker-to-beneficiary ratio are often presented as inexorable, though many are influenced by public policies. Neo-Malthusian arguments exaggerating the challenges posed by a growing dependent (in this case elderly) population are used to argue that social insurance is increasingly unaffordable, ignoring productivity growth and other positive trends while falsely suggesting that policymakers have few alternatives to benefit cuts.

The Social Security system requires periodic adjustments to maintain fiscal balance and keep up with changing times. The only unusual thing about the current situation is the fact that a distant and relatively modest projected shortfall has been labeled a “crisis.” Though there have been 21 contribution rate increases in Social Security’s history, none have been legislated since the Social Security Amendments of 1983 ushered in a period of retrenchment.1

The 2012 Trustees Report puts the 75-year shortfall at just under 2.7 percent of taxable payroll, meaning that an immediate increase in employer and employee contribution rates of 1.4 percentage points (from 6.2 to 7.6 percent) would be more than enough to close the shortfall,2 as would a gradual increase in the contribution rate from 6.2 to 9.4 percent over the 75-year projection period.3 To put this in perspective, average real wages are projected to grow by a cumulative 140 percent in inflation-adjusted terms by the end of this period.4 Thus, an increase in the contribution rate would still leave future workers significantly better off than today’s workers while preventing a further erosion of the Social Security replacement rate, which is already declining from 49 percent in 1980 to 36 percent in 2080 for the prototypical “medium earner” retiring at 65. In contrast, the plan proposed by Simpson and his Fiscal Commission co-chair Erskine Bowles would result in Social Security benefits replacing just 28 percent of wages for medium earners in 2080.5

And this analysis assumes that the fix to Social Security financing has to happen simply by increasing contribution rates—a fix that ignores the distributional issues that have been the biggest cause of the increase in the 75-year shortfall. Earnings inequality has eroded Social Security’s taxable earnings because earnings above a cap (currently $110,100) are exempt from Social Security taxes. Since high earners have received the lion’s share of pay increases as well as life expectancy gains, it makes more sense to close most of the shortfall by raising or eliminating the cap on taxable earnings.

Another way to close the gap—and address many other problems—is to pursue full-employment fiscal and monetary policies and a pro-worker agenda, including raising the minimum wage and strengthening collective bargaining rights. Such policies would have multiple positive effects on Social Security’s long-term finances: raising real wages, reducing the share of earnings above the cap, and lowering the beneficiary-to-worker ratio, among other things.

In the long run, real wage growth is constrained primarily by productivity growth. But since real wages for most American workers have lagged productivity for the past 40 years, wages (especially those below the Social Security cap) could outpace productivity if these workers made up for lost ground. Conversely, for technical reasons related to the way inflation is measured for the economy as a whole and for consumers, the Social Security actuaries assume that a price differential will drive a small wedge between productivity and real wage growth.

According to the 2012 trustees report, productivity growth has averaged 2.06 percent per year in the last complete business cycle (2000-07), 1.75 percent in the previous cycle (1989-2000), and 1.68 percent in all cycles since 1966. The trustees report also provides a “low-cost” alternative scenario which is intended to be optimistic but plausible, estimating that wage growth of 1.71 percent rather than 1.12 percent would reduce the projected shortfall from 2.67 percent to 1.77 percent of payroll, or 34 percent.

Expansionary macroeconomic policies combined with a higher minimum wage and stronger worker bargaining power would not just increase average wages but would also reduce inequality and therefore the share of earnings above the taxable cap. Since growing inequality reduced the taxable share of covered earnings from 90 percent in 1983 to 84 percent in 2010, one way to frame the issue is to ask what would happen if the process were reversed.

The 2012 trustees report assumes the taxable share will continue to decline before leveling off at a slightly lower level, though it doesn’t provide detailed projections or an alternative “low-cost” scenario. However, the Social Security Advisory Board’s 2011 Technical Panel report suggests that increasing the taxable earnings share from 80.0 percent to 84.3 percent would reduce the actuarial deficit from 2.00 to 2.47 of payroll. Extrapolating from these estimates (which were done by the Social Security actuaries on behalf of the Technical Panel), restoring the taxable share to 90 percent could reduce the 75-year shortfall by 1.09 percent of taxable payroll, or 41 percent (1.09/2.67), assuming a linear relationship between the taxable payroll share and the 75-year actuarial balance.

It’s difficult to gauge the combined effect of slow wage growth and rising inequality since the two are related—the taxable earnings cap is indexed to average wages. It’s probably fair to say, however, that robust full-employment and pro-worker policies could eliminate half to three-fourths of the shortfall. These calculations don’t take into account other full-employment effects, such as reduced unemployment and disability rates, which are harder to estimate with available information.

Endnotes