Share |

The Misguided Arguments Against Social Security Reform

e21 | 09/01/2010 |

Click here for a printer-friendly version of this article

The ancient Greek playwright Aeschylus is often credited with the well-worn maxim, “In war, truth is the first casualty.” Judging by the commentary circulated whenever a hint of Social Security reform is in the air, Social Security politics are the equivalent of war for many people. Sorting out Social Security fact from Social Security fiction can become practically a full-time job for those who take it on.

For the past few months, it has been rumored that President Obama’s fiscal responsibility commission will tackle Social Security as part of their broader budget recommendations. Given that Social Security’s fiscal imbalance has been well documented by everyone from the Congressional Budget Office to the General Accounting Office to the program’s own actuaries and trustees, this would seem to be a rational step to take. But once again, a cacophony of voices has arisen to argue that nothing about Social Security warrants changing any time soon.

Because the assertions of reform opponents are once again receiving significant circulation, it is worthwhile to dissect some of their arguments in detail.

Paul Krugman’s recent column, sensationally titled, “Attacking Social Security,” contains many of the prototypical arguments of those opposing action. One usual argument from this quarter is to question whether the Social Security shortfall is even certain to materialize:

The program won’t have to turn to Congress for help or cut benefits until or unless the trust fund is exhausted, which the program’s actuaries don’t expect to happen until 2037 — and there’s a significant chance, according to their estimates, that that day will never come.” (Italics added).

Let’s deal with this question first. Is it really the case, according to the official estimates, that the Social Security shortfall might never happen? This has been a shibboleth on the left for some time, and it has remained remarkably persistent even in the face of overwhelming empirical evidence to the contrary.

At this point, it ought to be well-established that the frequent depiction of conservative actuarial projections is pure folklore. The mythmakers have failed to account for salient demographic realities, such as the slower net growth of the labor force as the baby boomers head into retirement. They have also misrepresented the trustees’ projection track record, which has not been at all conservative. Even before the recent recession, Social Security’s difficulties were arriving faster than previously foreseen. That some continue to propagate the myth even in the face of worsening Social Security finances demonstrates that the myth is heedless of empirical realities.

To demonstrate this yet again, let’s look more closely at the actuaries’ estimates cited by Dr. Krugman. What constitutes a “significant chance” that the program shortfall will never materialize? The trustees include a probabilistic analysis with their annual projections. What is their estimate of the likelihood of the problem vanishing by itself?

Actually, we don’t exactly know, because absolutely nowhere within the trustees’ 95% confidence band – from the 2.5thth percentile of their projections – percentile to the 97.5 does the problem actually go away. What Dr. Krugman calls a “significant chance” is seen by the actuaries as having a less than 1 in 50 probability of occurring. Meanwhile, the scenarios in which the problem is twice as large as currently projected are actually more probable than the scenarios in which the problem goes away. “Significant chance,” indeed.

This brings us to Krugman’s other chief argument, which is to mock those calling for promptly-enacted constraints on the growth of Social Security’s costs:

“And having invented a crisis, what do Social Security’s attackers want to do? They don’t propose cutting benefits to current retirees; invariably the plan is, instead, to cut benefits many years in the future. So think about it this way: In order to avoid the possibility of future benefit cuts, we must cut future benefits. O.K.”

Put in that disingenuous way, that does sound rather silly. But of course, this is because Krugman’s phrasing grossly misrepresents the choices involved. The choice is not simply one between possible and definite future Social Security benefit cuts. Instead, the choice is between:

  1. Sudden, precipitous benefit cuts that would affect those already in retirement, including low-income beneficiaries – which would be the result under inaction, or:
  2. Changes in the benefit formula that would allow benefits to still grow relative to inflation, and which could shield those already in retirement and low-income retirees generally from the threat of benefit reductions – the choice made under most reform proposals.

Krugman’s depiction implies that whether we act now or act later, the same people will be affected in the same way. This is untrue, as any examination of real-world Social Security proposals clearly shows.

Under current projections, if we do nothing, benefits would suddenly be cut by 22% in 2037. This is often represented on graphs that look like the following one. It shows the initial benefit payments that Social Security would make to a “scaled low-income retiree” retiring at 65 under current law.

Projected Initial Annual Benefits (Chart 1)

Does anyone believe that this bizarre outcome would make good policy sense? To allow payment costs to rise to unaffordable levels by 2037, and then to hit all retirees (including low-income ones) with a sudden 22% benefit cut?

Of course it doesn’t. But unattractive as this picture is, it actually understates the problem under current law. This is because the picture above depicts initial benefit payments to different individuals – that is, to a new set of retirees in each year depicted on the graph. Under current law, initial benefit payments for each succeeding cohort would rise through 2037, drop dramatically, and then start rising again.

But the biggest problem under current law is that not only new retirees, but those already in retirement would be hit with benefit cuts. The situation facing a single individual retiree under current law would be far worse than seen on the above graph. Let’s look at the situation facing a single low-income woman who retires in 2025 at age 65 under current law. Her stream of benefit payments under current law would look as follows:

Projected Initial Annual Benefits (Chart 2)

As the above graph shows, this low-income woman would retire with an annual benefit (in $2010) of $11,313 in 2025 under current law. This benefit would stay constant (in inflation-adjusted terms) until 2037, when at the ripe age of 77, she would be hit by a sudden 22% cut in benefits, down to less than $9,000 a year. (This amount would rise very slightly through 2050, then decline still further in subsequent years.) This would vastly reduce her standard of living relative to her own previous level, and even relative to what low-income retirees receive today in 2010. To suddenly cut benefits for those of advanced age in this manner would sharply increase poverty rates among the future elderly.

Now, contrast this with a typical proposal to reform the Social Security benefit formula today. We’ll select the plan of Senator Robert Bennett, because it has a publicly available score that includes benefit projections, and because it reflects the general consensus among reform proponents to implement benefit changes that protect workers of lower incomes.

Under the Bennett proposal, initial Social Security benefits would continue to grow smoothly over time, as shown on the chart below:

Projected Initial Annual Benefits (Chart 3)

Under Bennett’s and many similar proposals, by acting soon to rationalize the Social Security benefit formula, we can avoid the sudden drops in benefits that would transpire under current law.

The advantages of prompt action can be seen even more clearly by examining the individual case of the low-income woman shown earlier. Instead of experiencing a sudden drop in the value of her benefits, she could continue to rely on her checks continuing without diminution for the rest of her life.

Projected Initial Annual Benefits (Chart 4)

Why is a reformed system able to avoid the sudden drops in benefits that would occur under current law? The answer is very simple: a given amount of revenue can only fund a given total amount of benefits. If we act today, we can spread the effects of any changes among the largest number of people, which allows the effect on any one person to be much smaller. Furthermore, we can tailor the adjustments to protect those already in retirement as well as those on the low-income end. If instead we leave the problem unsolved until 2037, then the full amount of the financing gap will be felt by anyone unlucky enough to be receiving benefits at that particular moment – meaning a sudden plunge in benefits for all beneficiaries, including the most vulnerable.

Advocates of the status quo might object that we would never allow these sudden benefit reductions under current law to take place. But that is exactly the point – the only way to prevent them from taking place is to take legislative action to prevent them – precisely the action that these advocates are arguing against. If opponents of action have their way, the benefit reductions will take place. If instead we want to avoid sudden reductions and instead have the smooth growth in benefits envisioned under Bennett’s and other proposals, we need to implement them soon.

Advocates of the status quo might also object that these examples only hold water if we fail to raise Social Security taxes. It’s true that these specific examples assume that there is no tax increase. But the principle is the same whether we raise taxes or not; no matter what level of benefits we choose to finance, we can minimize the disruptions of any (benefit or tax) changes on a single individual by acting sooner rather than later.

Does it make sense to defer action for a while, just in case the optimistic projection outlook of status quo advocates turns out to be correct? Not by a long shot. Let’s explore all the way out to the 90th percentile of the trustees’ projection band -- basically, to see what happens if a very lucky 1-in-10 shot comes in.

Even in that highly optimistic scenario, our representative low-income retiree would still face sudden benefit cuts. These cuts would hit in 2045 when she was 85 years old, an age where the risk of poverty rises significantly. Under even this highly optimistic projection, her sudden benefit cut would be roughly 11%, and would still bring her back below the benefit levels of low-income retirees in 2010. Thus, even if the naysayers are right about the Social Security projections being conservative, their recommended inaction would be disastrous for low-income seniors.

Not only that, but they are much more likely to be wrong. Of equal probability to the optimistic scenario described above is another in which Social Security finances are far worse than currently projected. It is thus equally likely that, if we do not act, our low-income retiree will face benefit cuts of 28% – and face them in 2032.

The stubborn facts are that the Social Security shortfall isn’t going away, despite the persistence of folklore alleging the contrary. But equally importantly, seniors’ future retirement benefit security hangs in the balance, no matter whose projections are right.

If we act today, we can protect all of today’s retirees, we can protect low-income beneficiaries today and in the future, and we can allow for a smooth growth in future benefit levels. But the longer that we postpone action, the more sudden and disruptive will be the inevitable changes that hit the workers and beneficiaries of the future.


e21 Projects & Partnerships