January 9, 2013

Social Security: Even More Insolvent Than You Thought

Its actuaries have been blowing the calculations, and there is no “buffer.”


This column went up at FrontPageMag.com with minor revisions early Tuesday morning.


At first, one doesn’t know whether to laugh or cry at research findings reported by two Ivy League profs in a co-authored column in the Sunday New York Times titled “Social Security: It’s Worse Than You Think.”

Actually, white-hot anger is more appropriate, given that what Gary King, a professor of government and director of the Institute for Quantitative Social Science at Harvard, and Samir S. Soneji, a demographer and assistant professor at the Dartmouth Institute for Health Policy and Clinical Practice, really told us, namely that the New Deal-era retirement system, thanks to its use of ossified actuarial calculations, is more insolvent than almost all of us knew.

The pair’s core finding, as presented in the Times: “[T]he Social Security Administration underestimates how long Americans will live and how much the trust funds will need to pay out — to the tune of $800 billion by 2031.”

There are at least three justifications for feeling righteously outraged.

The first has to do with the delay between when King and Soneji originally reported their core findings to the academic community and when they deigned to let the rest of the country in on their virtual secret.

The pair’s research paper, “Statistical Security for Social Security,” reported that “Social Security, especially the OASI program, may be in a considerably more precarious position than officially thought” because the government’s actuaries are ignoring the financial implications of life expectancy increases due to “the steady decline in smoking,” which more than offset decreases relating to “the rapid rise in obesity.” The estimated net effect they reported in their paper was “$730 billion less in the OASI and SSDI (Old Age and Disability) Trust Funds” by 2031 (I will get to the “Trust Funds” fiction and the $800 billion/$730 billion discrepancy later in this column).

The paper first appeared in Demography on May 17, 2012; the link from the their Sunday Times column to the web page containing the paper’s abstract is labeled “August 2012.”

This means that King and Soneji had credible evidence that Social Security is in far worse shape than most of us thought well in advance of the November 2012 elections. Apparently, they made no special effort to bring their work to the nation’s attention, even though virtually any credible center-right news outlet would have been extremely interested in their findings. In their Times column, King and Soneji almost seem to crow about the fact that Social Security was not a contentious item in the presidential contest between Barack Obama and Mitt Romney: “It was a rare issue on which both men agreed — and both were utterly wrong.” Well guys, if you knew they were wrong, why didn’t you say anything?

Further blame assignment is in order. How could Romney’s largely Northeast-based advisers have missed the article’s publication? Isn’t setting up Google Alerts on key terms supposed to part of Campaigning 101? Though it appears that Demography has been coding its web pages to avoid search engine detection, a Google Web search on the research paper’s first sentence shows that it should not have escaped detection last summer.

A very sad and worse possibility which cannot be ruled out is that Team Romney learned of the research and didn’t believe that bringing up Social Security’s unprecedented decay during Obama’s presidency would be a good idea. As we have so often seen but Republican campaigns have failed to learn, timidity does not win elections.

The second reason for outrage is the government’s decades-long failure to modernize its actuarial methodologies. King and Soneji show that the system currently uses assumptions based on “a combination of linear extrapolation and qualitative judgments,” which, though they were the best available frameworks many years ago, are now dangerously outdated and divorced from reality. Unfortunately, this is all too typical of what happens when entrenched government bureaucracies simply go through the same processes year after year after year without giving adequate or sometimes even any thought to whether they could or should be improved.

This debacle would never have happened if Social Security’s trustees had been outsourcing their annual actuarial reviews to one of the industry’s leading firms. These organizations live or die on their ability to develop credible and state of the art mortality estimates for their private and public clients. Social Security’s trustees should make outsourcing this work in future years while ordering the firm selected to follow the best available practices one of their first orders of business. Odds are they won’t — and that if they try, Obama and Harry Reid’s Senate won’t let them.

The final source of outrage relates to King’s and Soneji’s fictional presentation of how Social Security supposedly works in their research paper, their Times column, and that column’s accompanying flowchart.

Sentences from each of the pair’s three documents will demonstrate that what they tell readers about the Social Security’s “trust funds” is an exercise in sheer fantasy:

  • Their research paper engages in laughable historical revisionism: “As a result of payroll taxes that generated revenue in excess of annual benefit outlays over the last 25 years, the trust funds have amassed large surpluses in preparation for the aging population.”
  • Among their column’s claims: “(The trust funds are) a $2.7 trillion buffer built in anticipation of retiring baby boomers …”
  • They saved the worst howler for their column’s graphic: “Current workers’ payroll taxes go into the Social Security Trust Funds — a bank account for current and future beneficiaries …”

What rubbish. Jay Leno could use these assertions as jokes on The Tonight Show and have most of his audience rolling in the aisles.

The real facts are these:

  • The “trust funds” consist almost entirely consist of IOUs from the rest of the federal government. Congress has taken Social Security’s annual surpluses and spent them on other things for decades.
  • The rest of the government currently has a balance of $11.5 trillion in “debt held by the public” (i.e., excluding “intergovernmental holdings” like the “trust funds”).
  • Our supposed betters in Washington have run deficits of over $1 trillion during each of the past four years, and have added an even greater amount to the national debt during that time. Fiscal 2013, despite the fiscal cliff deal, appears to be on track to stretch that streak to five.
  • Part of the government’s annual deficit now includes money needed to cover Social Security’s annual cash shortfalls, which, as King and Soneji have noted, began in 2010.
  • Social Security’s ability to continue to pay benefits at current levels depends largely on the government’s ability to cover those annual cash deficits, which are projected to grow as far as the eye can see. Given that Obama insists that “we don’t have a spending problem,” there is substantial reason to believe that the government will note be able to cover those cash deficits several years from now.

The serious deterioration in Social Security’s viability we have seen in the past four years is partially due to the recession, but really has far more to do with the fact that the economy under Obama has failed to adequately recover from that recession. If the economy only grows by 2 percent per year, those paper entities known as the “trust funds” will run dry in an accounting sense well before 2031, perhaps by as much as a decade — assuming that the rest of the government’s finances don’t collapse sooner.

One indicator of how serious the system’s decay is can be seen in the fact that the $730 billion shortfall estimated in King’s and Soneji’s research paper ballooned to the $800 billion noted above in their Times column’s presentation. Though they may have refined their life expectancy data in the meantime, the higher number more likely occurred because their original paper used data from the 2011 Social Security Trustees Report containing calendar 2010 data, while their Times column instead used data from the 2012 report.

Parting question: What kind of impact would the failure to consider the life expectancy estimation problems King and Soneji cited have on the financial projections for Medicare and ObamaCare? None of us should be surprised if the twenty-year impact on those programs runs to several trillion dollars.



  1. Rose colored glassess dominate policy. The simplest example is the “doc fix.” An outdated formula sets MD rates for Medicare. Congress knows it is outdated. If implemented, it would cut MD reimbursemnent by 30%. Knowing they cannot do this, they pass a last minute bill, often yearly, for a short term delay. Why not simple fix it? Because then the real projected Medicare costs would be much higher, and no one wants to deal with this reality. They have been doing this FOR TEN YEARS. Obamacare estimates needed the lower projected costs to justify the bill’s existence. That would have been an opportunity for the Republicans in Congress to fix the formula and then tell people about the real cost. Wasted.

    Comment by Jim — January 9, 2013 @ 9:27 am

  2. Yeah, they’re good at wasting opportunities — and, as shown in the column, not even recoginizing them.

    Comment by Tom — January 9, 2013 @ 10:05 am

RSS feed for comments on this post.

Sorry, the comment form is closed at this time.