Social Security does not provide future retirees with a safe “return” on taxes paid into the system on their behalf, given the program’s funding problems. It’s not even clear that it provides a decent return to many current retirees, and it will get worse as younger age cohorts become eligible. Demographic changes worked in the system’s favor in its early years, but no more: the number of eligible retirees is growing faster than the working-age population. This has led to cash flow deficits since 2010 that will widen in the years ahead. The unfunded liabilities of the system are currently estimated to be $26 trillion. The so-called “Trust Fund” for retirement holds about about $2.8 trillion of government securities, but those can’t be “cashed out” without a raid on general tax revenue or new borrowing by the Treasury.

Michael Tanner reveals the absurdity of some of the myths surrounding SS, such as claims that there is “no crisis” (and even more absurdly, that benefits should be expanded), that the Trust Fund will “save” the system, and that SS payroll taxes are “saved” for retirees. They are not saved; it is a “pay-as-you-go” system with current payroll tax collections paid out to today’s retirees. Here is Tanner on the woeful state of the system’s finances:

“According to projections by the Congressional Budget Office, for workers born in the 1980s, there are only enough funds to pay 76 percent of their schedule benefits; for today’s children born in the 2000s, this falls to 69 percent. And, taxes are already so high relative to benefits that young people will receive far less than they could receive if they invested their taxes privately.“

Measuring the return on Social Security (SS) payroll taxes (otherwise known as FICA) is not without controversy. The Social Security Administration (SSA) performs its own analyses of the returns on payroll taxes periodically. They analyze individuals at different income levels for each of four circumstances: single men, single women, one-earner couples and two-earner couples. They do so under different scenarios about future payroll taxes and benefits. The benefits include cost-of-living adjustments. These calculations show that today’s younger workers, singles and high-income workers can expect to receive the lowest returns. According to the most recent report, from December 2014, annual rates of return for those not yet drawing benefits under present law varies from less than 1% to 6.5%. Of course, the promised benefits are not sustainable under present law.

Reforms are not optional, as the program cannot run a deficit under its current authority once the Trust Fund is exhausted. SSA attempts to analyze steps that might close the gap and the impact of those changes on returns to retirees. One scenario involves higher payroll taxes and another lower benefits. These changes reduce the calculated returns in all cases, though even the lowest returns remain positive, if barely. These alternative scenarios involve no changes until 2033, however.

At the time of the SSA report, the most recent Congressional Budget Office (CBO) predicted that the SS Trust Fund would be exhausted in 2033. More recently, the CBO predicted that the fund will run dry in 2029. (The Disability Trust Fund is projected to run dry in 2017.) Therefore, the returns calculated by SSA under the alternative scenarios are over-estimates, since more drastic and earlier measures are required to restore balance. It’s likely that some of those returns would turn negative using SSA’s methodology. And it’s not unreasonable to suggest that reforms, whatever shape they might take, should be implemented sooner than 2029. After all, the need for reforms is well known, and we’re talking about it now! As for the SSA’s alternative scenarios, changes much sooner than 2033 would cause even lower returns.

While the SSA’s effort to provide the estimates is laudable, there are several aspects of the methodology that are questionable. SSA claims that the returns are real (inflation-adjusted) internal rates of return (IRRs), but they do not offer a detailed explanation of the inflation adjustment that must take place after calculating the nominal IRR. Using projected cost-of-living increases to inflate future benefits does not make the calculated IRRs “real”, if that’s what they have in mind. Second, the cost-of-living adjuster is the Consumer Price Index for Urban Wage Earners, which underestimates inflation experienced by the elderly. Third, they do not attempt to account for the probability of death before retirement, which would obviously reduce the return on contributions for single earners.

The “present-law” returns are essentially irrelevant, given the unfunded projected benefits. But the calculations under the alternative scenarios fail on other grounds: they are not “dynamic” in terms of adjusting for the economic impacts of the policy changes. In particular, higher payroll taxes are likely to reduce employment and slow the economy. A slowdown might even lead to additional claims on the system from earlier-than expected retirements. Thus, the higher payroll tax rates used by SSA will not be sufficient to close the gap. Likewise, reduced benefits would have a negative impact on the economy, reducing payroll tax collections. In both cases, dynamic economic effects would cause a wider funding gap; closing it will reduce returns more than suggested by SSA’s calculations.

An analysis by the Urban Institute in 2012 made somewhat arbitrary assumptions about rates of return. They used a 2% real rate of return to compound past contributions and discount future benefits (presumably with no cost-of-living adjustment). Under their assumptions, the value of payroll tax contributions at retirement often exceeds the discounted value of SS benefits for age cohorts turning 65 in 2010 and 2030. That implies that the real IRR must be lower than 2%.

As a hypothetical exercise, if individuals could invest their own payroll contributions over their working lives, significantly better returns could be earned than the IRRs discussed above, even if workers were forced into low-risk investments as they approach retirement. Therefore, the implied value of payroll contributions at retirement inherent in the IRR calculations is far too low. And while the discounting of retirement benefits at a relatively low rate reflects an appropriate conservatism, the level of SS benefits would not be competitive with the dollar returns on safe investments funded by a larger pot at retirement. The IRR calculations show only that the SS program is about as good as stuffing money into a mattress.

Unfortunately, the mattress might burn. The risks inherent in future SS benefits are substantial, and none of the reform alternatives are very popular. Some of the opposition is rooted in unreasonable criticism: No one has suggested programatic changes that would affect the benefits of anyone over the age of 55. Still, cuts in benefits for future retirees, delayed eligibility and higher payroll taxes are not easy sells. Another solution is to phase out the pay-as-you-go system, allowing younger workers the option of a a self-directed account (subject to certain restrictions), including a discounted “cash value” credit as a buyout for previous contributions. This was discussed in a recent post on Sacred Cow Chips.

Social Security is unsustainable and is an inter-generational rip-off in its current, pay-as-you-go form, as younger, less affluent workers subsidize current retirees, who are relatively wealthy as a class. Rather than shutting-down debate over reforms with exaggerated political claims, those interested in assuring a viable public retirement program should consider proposals that would give workers more choice and control, taking advantage of the higher returns available on private investments. Only this type of program can take advantage of the economy’s ability to convert savings into productive investment and real growth. Simple transfers from young to old do not leverage this process, and can never hope to compete with it.