Politics & Policy

Promises, Promises

The Social Security system is loaded with a confusing mishmash of IOUs.

If somebody owes you $10, the next best thing to having the cash itself is probably having a written IOU. After all, the IOU will make it harder to forget the liability, and also might give you evidence to use in court should the borrower attempt to shortchange you.

The Social Security system is loaded with a confusing mishmash of IOUs, both implicit and explicit. Reforms tend to rearrange the liabilities in this crazy quilt. Do the labels on the IOUs matter?

Social Security faces an unfunded liability of $4 trillion over the next 75 years. The liability is $11.1 trillion in perpetuity. Some dismiss infinite timelines as absurd. The concept is an easy target: It’s difficult enough to get people to care about the fiscal situation over the next 75 years; why should they care about Social Security benefits in the year to 2500? But that’s not the point. The infinite horizon is necessary to ensure that any Social Security fix is a permanent one as opposed to the many temporary patches that are floated — patches that would leave the system out of balance beyond the truncated window.

As the Social Security Trustees point out, “Overemphasis on summary measures for a 75-year period can lead to incorrect perceptions and to policy prescriptions that do not move toward a sustainable system.” The 75-year unfunded liability increased by $300 billion over the past year and the infinite gap increased by $600 billion. So right now the American people have an implicit liability of about $11 trillion. The liability is implicit since there is not an explicit IOU for the whole shortfall sitting in any filing cabinet.

The unfunded liability is the amount of money that we would need to put away today in order to generate a large enough revenue stream to fill in the growing gap between program benefits and revenues. It is important to understand that this implicit liability is completely different from other types of debts, such as the government debt owed the public. Unlike a government debt resulting from borrowing, this implicit debt is not binding because participants have no legal right or claim to their Social Security benefits. While the benefits have been promised (or implied), they are not owed, need not be paid, and can be changed at any time.

That is exactly what all of Washington is talking about. The most straightforward way to eliminate the unfunded liability would be to reduce those promised benefits, bringing them into line with what the program can afford. (The other option is to raise taxes.) Price-indexing benefits, for instance, would reduce the unfunded liability to essentially zero.

Most of the plans that have been put forward to eliminate Social Security’s unfunded liability also include an individual account component. And since most do not include a plan for how to pay for the accounts, they feature some additional level of borrowing.

The borrowing that would take place to jump-start a system of privately owned individual accounts would create a new explicit debt. A dollar diverted into such an account would, ignoring dynamic effects, lead to the issuance of a dollar’s worth of Treasurys on the open market — an additional dollar of debt for the government. Opponents of reform often point to this and assert that the government’s asset position would be worsened.

In response, proponents have made the case that the new dollar of explicit debt would not be problematic because the accounts would include offsets to traditional benefits; every new dollar of explicit debt would be counterbalanced by a dollar less of implicit debt. Furthermore, the dollar borrowed by the government would also be saved in the accounts, so again, barring feedback effects, national saving wouldn’t be altered.

On one hand, this argument makes perfect sense. There is some level of explicit debt that is appropriate in creating accounts that are part of a comprehensive reform plan that achieves sustainable solvency. Such borrowing would spread over time some of the costs of reform and would likely have positive effects on separating Social Security from the rest of the budget.

However, a one-for-one trade between explicit and implicit debt is not clearly a deal worth making. For one thing, financial markets may be pricing in the assumption that unsustainable benefit levels will be reduced by more than that. If that’s the case, borrowing to create accounts could have adverse effects on markets.

Secondly, explicit debt is for real. Unless one is willing to devalue it with high inflation, it can never be changed. There exists the very real risk that the part of the plan that reduces future benefit levels would not materialize and the deal would end up with a new explicit debt and the continuation of a staggeringly high implicit debt. Never underestimate the power of the senior lobby — especially with the baby boomers joining their ranks.

An increase in explicit debt is not clearly problematic. The bigger the corresponding reduction in the implicit debt the better the deal — if one wants to be confident that the long-run balance has truly been reached.

Some reform plans also include “guarantees” of some level of benefit in the future. The purpose of the guarantees is to insulate account holders from excessive financial risk by putting a floor on the downside of owning accounts. If the purpose of fixing Social Security is to remove some of the risk for future participants being forced into a plan with large and unaffordable liabilities, the effect of creating guaranteed benefits would be to put that risk right back on participants — this time in the form of a contingent liability. After all, taxpayers will foot the bill if necessary.

It is worth noting that “guarantee” in this situation is a meaningless term. Congresses today cannot bind the hands of future Congresses. Benefits will always be subject to change whether they are promised, guaranteed, chiseled in stone, or written on embossed paper. This goes for current benefits just as much as it does for so-called guaranteed benefits. Benefits can be altered at any time.

When it comes to Social Security there’s no such thing as a free lunch and no such thing as a guarantee.

Nonetheless, for political reasons guarantees would likely be honored — again because the constituency that would be claiming them is so powerful. On the liability spectrum, these contingent liabilities would then fall somewhere between the system’s current implicit liabilities and the explicit liabilities that would result from borrowing.

Recently, some have been floating the idea of fixing only part of Social Security’s shortfall. That is not a high enough standard. If we end up with an explicit liability and limited and possibly tenuous future savings then the job of putting Social Security on a sustainable path will be left undone.

Kevin A. Hassett is director of economic-policy studies at the American Enterprise Institute. Maya MacGuineas is the director of the fiscal-policy program at the New America Foundation.


The Latest