Understanding the 2010 Social Security Trustees’ Report
Click here for a printer-friendly version of this article
Everything We Know Looks Much Worse (But What We Hope For Is Better)
The Social Security Trustees have finally released their long-delayed 2010 report projecting the program’s future finances. At first glance, the report seems to contain little new; the projected insolvency date is the same as in last year’s report, and the actuarial imbalance has barely changed. The Social Security Administration’s own press release downplayed the report’s significance, stating that the “long-range financing outlook remains unchanged.”
But closer inspection reveals a far more concerning picture: virtually all of the updated hard information in the report looks significantly worse than was projected last year. The effects of this worsening have simply been balanced over the long-term by more optimistic assumptions going forward.
The Trustees’ Report – the Basics
Before we delve into the changes relative to last year’s report, let’s review some basics about Social Security operations. Social Security pays benefits to millions of retirees, disabled, survivors, spouses and dependents, with retirees composing the biggest slice of program beneficiaries. As American society ages, program costs generally rise. Right now, as the enormous baby boom generation heads into its retirement years, program costs are rising faster than ever before.
This much has long been expected. But a severe recession hit in 2008, just as the first baby boomers were hitting the retirement rolls. Until that happened, it was anticipated that the program would run annual cash surpluses (excesses of incoming tax revenue over benefit obligations) well into the late 2010s. The recession, however, depressed payroll tax collections and wiped out most of the remaining anticipated surplus.
As a result, we are entering the era of Social Security deficits sooner than previously thought. Below are the latest projections of annual finances from the 2010 Trustees’ report (for the combined operations of the program’s retirement, survivor and disability components).
Far from an “unchanged picture,” the 2010 report is the first official recognition by the program’s Trustees that Social Security is now in deficit.
Everything on this chart is expressed as a percentage of taxable worker wages. It can be most readily thought of as the number of cents out of each worker’s wage dollar absorbed by Social Security. Because the Social Security payroll tax is 12.4%, the revenue line on this chart stays close to that level going forward (actually a little bit higher because of additional revenue from benefit taxation).
The cost line, however, rises dramatically in the years ahead. By the mid-2030s, fully one-sixth of workers’ taxable wages would be needed simply to pay for the Social Security program alone among all federal expenditures. Most of this cost growth is attributable to the rising number of baby boomers on the retirement rolls. The rest is due to the program’s wage-indexed benefit formula, which pays higher benefits (relative to inflation) to later cohorts as national wages grow.
Also visible on this chart is an even sharper short surge in costs (relative to worker wages) that began in 2008. Part of this was because costs increased as Americans filed for retirement and disability benefits in greater numbers (disability benefit claims typically rise during a recession). But an even more pertinent reason was simply that total taxable worker wages dropped as a consequence of the recession. So, program costs rose, but costs per taxpaying worker have risen even more.
As a result of all of these adverse trends, both the Social Security Trustees and the Congressional Budget Office are projecting that Social Security will operate a cash deficit this year.
This is where the Social Security Trust Fund comes in. The Trust Fund consists of Treasury bonds held by the Social Security system, which are also (and in equal measure) debt obligations of the general federal government. The Social Security system holds these bonds because of the surpluses it has generated in the past. Whenever such a surplus has previously been run, that amount of bonds has been issued to the Social Security Trust Fund. These bonds earn interest and to date have accumulated to a total balance of roughly $2.5 trillion.
As long as there is a positive balance in the Social Security Trust Fund, payments to beneficiaries continue. If taxes fall short of obligations – as anticipated this year – the general government must put up the cash to make up any shortfall. This year, the rest of the government is basically making a $41 billion cash payment of interest owed to the Social Security system. In future years, still larger payments would embody both interest and principal until the Trust Fund is fully redeemed (read: depleted) in 2037. Once that happens, the program’s authority to pay benefits would be limited to the amount of tax revenue generated annually (now projected to be 78% of benefits promised at that time).
This Trust Fund accounting is a source of endless confusion in the public Social Security discussion. To simplify, certain basic concepts should be borne in mind.
One is that the Trust Fund doesn’t itself reduce the cost of paying benefits. If for example benefit payments in a given year equal 15% of taxpaying worker wages, then the government must come up with total cash equal to that 15% of taxpaying worker wages. If there is a positive Trust Fund balance in such a year, then (hypothetically) the sources of revenue could be the 12.4% payroll tax, another 0.6% from benefit taxation, and the last 2.0% from general revenues. But the Trust Fund itself doesn’t change that 15% by even a penny.
Secondly, the accounting concept of the Trust Fund shouldn’t be conflated with other more operational concepts. The Trust Fund, for example, doesn’t represent the amount of actual saving done to prepare for the baby boomers’ retirement. To the contrary, most academic studies have found that the Trust Fund doesn’t represent saving at all, because past Social Security surpluses have mainly financed additional government consumption.
Nor does the Trust Fund’s projected duration tell us how long we should consider currently-scheduled benefit obligations to be inviolate; to the contrary, though the Trust Fund would theoretically last until 2037, never at any point would it amount to even four years’ worth of future benefits. The vast majority of the value of benefits now being promised would be paid from the taxes of our children and grandchildren – and have nothing to do with the balance of the Trust Fund.
A further interesting note on that point is worth mentioning. Defenders of Trust Fund accounting are fond of pointing out that the nominal balance of the Trust Fund is scheduled to grow for years into the future. But even the Trust Fund’s most ardent fans might be concerned by one finding in the report: that the maximum benefit-financing power of the Trust Fund is by any definition a thing of the past. Specifically, the Trust Fund ratio –the ratio of Trust Fund assets to years of full benefit payments – actually peaked in 2008 at 358 (i.e., 3.58 years’ worth of benefits). According to this year’s report, the Trust Fund under current law will never have that level of benefit-financing power ever again.
Third, the Trust Fund represents the government’s obligation – not its means. The Social Security system can theoretically call upon the rest of the government to make full benefit payments through 2037. This, however, assumes that the government is in a position to pony up that full amount. Anyone who believes this is a trivial concern needs to better familiarize themselves with the long-term projections of the Congressional Budget Office.
Most importantly, the Trust Fund doesn’t tell us how long we can wait to fix Social Security. If we did nothing until the date of insolvency, those in retirement as well as new retirees would face a sudden 22% benefit cut – a result widely regarded as unacceptable. Alternatively, such benefit cuts could then only be avoided if the Social Security payroll tax rate were effectively raised to 16.1% -- and gradually raised after that. If it’s unacceptable to cut benefits in this way for those already in retirement, or to have benefits decline relative to inflation, or to hit workers with payroll tax increases, then we must (unless we paper over the worsening problem with budget gimmicks) enact a solution within the next few years.
In any case, the dawning era of Social Security deficits does not signal immediate benefit cuts for recipients. It does mean, however, that our time is running out to find a solution that will enable Social Security to fulfill societal goals.
How Have the Projections Recently Changed?
In a nutshell, the near-term Social Security picture looks much worse than in previous Trustees’ reports.
To get a sense of this, let’s look at the projections from each of the last few Trustees’ reports for the annual balance in 2010 (in $2010).
In the 2007 report, a $97 billion surplus was anticipated for 2010. The 2010 imbalance is now projected to be a $41 billion deficit. For perspective on this $138 billion annual decline, consider this: the Trustees’ current projections for surpluses in 2012, 2013 and 2014 are a razor-thin $2 billion, $5 billion, and $4 billion respectively – effectively negligible relative to the recent fluctuations in the annual projections.
It would therefore take only a slight downward nudge into the economy to eliminate these projections of surpluses in 2012-14. If that happens, then Social Security will have entered permanent cash deficits in 2010 – sooner than projected in any single Trustees’ report since the 1983 reforms.
Why has the near-term picture become so much worse? Mostly because payroll tax collections have lagged far behind previous projections, due to the recession. The 2008 trustees’ report foresaw $787 billion in tax revenue arriving in 2010 – the actual amount is now expected to be $673 billion.
But benefit costs are up, too. In 2008, cost obligations were expected to be roughly $700 billion in 2010. Largely because benefit claims are up, actual costs this year are somewhat higher – now expected to be $715 billion.
The Trustees acknowledge that the short-term picture has grown significantly worse. But they are offsetting that story by portraying the long-term picture as somewhat better. This can be appreciated by looking at a picture reproduced directly from the 2010 report.
Social Security Annual Balances: 2009 and 2010 Trustees Reports
[As a percentage of taxable payroll, based on intermediate assumptions]
The most concerning aspect of this chart is this: most of the near-term worsening seen in the picture above is based on hard data. Most of the long-term improvement is instead based on more optimistic assumptions. As in the title to this piece – virtually everything we know looks much worse; it is only what we are hoping for that looks better.
The factors that have worsened the near-term Social Security outlook are raw facts. Payroll tax collections are lower than previously projected. Benefit claims (especially disability benefit claims) are higher. Another year has gone by without a legislated solution (the addition of one year in the valuation outlook by itself worsens the long-term imbalance by about 3%). Also tucked away in the Trustees’ report is the information that updated mortality rates are lower than previously anticipated; that’s good news for Americans generally, but unhelpful for program finances.
But in the latest Trustees’ report, these factors are more than offset over the long term by changes to future assumptions.
The biggest of these changes is an assumption that real wage growth going forward will average 1.2% annually. Over the last 40 years, the historic average has been 0.8%, and last year’s report projected future growth of 1.1%.
Why the increase? The trustees have incorporated assumptions made by others about the positive fiscal effects of recent health care reform legislation. As the Trustees’ report states, “Most of this change is due to the assumption that an excise tax on employer-sponsored group health insurance, effective beginning in 2018, will lead to slower growth in the total cost of employer-sponsored group health insurance. “
That assumption of slower growth in health care costs in turn, means that more of worker compensation would be received in the form of taxable wages, and less in the form of non-taxable health benefits. This in turn increases the amount of tax revenue projected for Social Security.
Modeling health care cost growth is not within the province of the Social Security actuary; accordingly, the apparent improvement here results purely from the incorporation of health care cost projections made elsewhere. It is difficult to find encouragement in this.
The Trustees have also incorporated a number of other small methodological changes that improve the long-term outlook. One of them changes the composition of the projected workforce to include “fewer teenage and more female and older workers.”
These changes in the long-term assumptions might, for all we know, prove to be more accurate. It is, however, concerning that the minor changes to this year’s trustees’ report are primarily a consequence of real worsening in the near-term being balanced by assumed improvement in the long-term.
Other Odds and Ends from the Trustees’ Report
As in previous years, the Trustees’ report demonstrates that the Social Security problem will not go away by itself. The 95 percent confidence band around the trustees’ intermediate projection includes not a single scenario in which the shortfall resolves by itself. It’s actually more likely that Social Security costs will ultimately absorb more than one-quarter of workers’ taxable wages than it is that legislated reforms will be unnecessary.
Another useful table in the Trustees’ report shows that future generations stand to lose, net, 3.3% of their taxable wages to Social Security. This would be the net loss by all future generations even after they receive a lifetime of retirement and/or disability benefits. The longer we delay action to reform Social Security, the more net income the program will subtract from future generations.
Another item is that the trustees again project that there will be no COLA at the end of this year. In fact, under the full range of the trustees’ assumptions – intermediate, low-cost, and high-cost – there will be no COLA until December 2011.
Altogether, the latest Social Security Trustees’ report is a compendium of sobering news. The main optimistic news in the report is due to assumptions of the benign effects of health care reform. More on that subject will be provided in the next summary, of the Medicare Trustees’ Report.