Is Social Security a Fiction?

Beat the Press is an economics blog that attempts to deal with the problem of how badly the subject of economics is dealt with in the press. It often has some interesting commentary altough it also frequently starts with a liberal bias. The moderator of the blog, Dean Baker, who is from the Center for Economic and Policy Research, authored a couple of books including the Conservative Nanny State and Social Security, the Phony Crisis.

Earlier in the week the blog opened a commentary on whether Social Security is a fiction. Baker starts from the position that Social Security is just like any other government program. He opened his comments with the following "Nothing like some comments on the trust fund to get the blogging juices flowing. It is amazing how metaphysical these discussions on the trust fund get. I don’t really see anything very complicated here. I am simply referring to the law as it stands."

This morning Baker made a long commentary responding to most of the commenters who believe Social Security is a fiction. In part he said "First, none of the "fiction" advocates here has explained how the bonds held by the trust fund are different from any other government bonds in that they are only backed by claims on future tax revenues. If this somehow makes them a "fiction," then all government bonds are a fiction. " The concepts here are pretty simple. First, are the bonds that cover the obligations in Social Security different from other government bonds? In a very real sense I think they are. But second, if this is just another government program, why maintain the fiction that this is a "retirement" or "insurance" program when the program fails to follow any reasonable actuarial or fiduciary standards?

There are three kinds of bonds in government finance. The closest analogy for Social Security is either a bond or an insurance program or an annuity. If Social Security is like a bond (a pledged asset in the future) then it must be either asset backed or not. Asset backed bonds are linked to a specific project or tax source. For example, when you build a construction project and borrow the money to do it, you pledge a fixed asset that if the government were to default, could be seized. (Unlikely, but it is possible.) The second are fundamentally borrowing against the good name of the government. When a government slows its payment or defaults there are consequences. In a state government, the credit rating is lowered and the ability to borrow more money is limited. In the federal government, if the government begins to not honor its obligations, the currency is devalued. I would contend that the Social Security Trust Fund is a third type of bonds. Here is where Baker and his types fundamentally misunderstand or mis-state the nature of Social Security. When the program was established it was set up with a specific tax source to create the impression (I would call it a fiction) of a reliable funding source. The sponsors wanted us to believe that it functions like a retirement program or an insurance program - when the thing gets actuarily out of whack it is corrected. Baker makes no difference between Social Security and any other program supported by taxes - but were this true we would not use a separate taxing source to fund the program. Baker would probably be fine with moving this to a general tax revenue based program. There was a reason for creating the program in the way that we did. For other programs created at the same time, with separate revenue streams - things get adjusted when there is an imbalance - but for many of those programs the long tail (how long it takes liabilities to be actualized) is simply not there. For Social Security, as opposed to programs like Unemployment Compensation, one might wait 30-40 years for a claimant to demand money.

When the program was originally established it was called the OASDI - Old Age (Pension), Survivors (Partially Pension) and Disability Insurance (Welfare payment) program. The first two parts had elements of a pension system. The third - less so. But the founders of the program understood the need to link this to a specific set of revenues.

How have the politicians treated the program. For the last several years they have used the "excess" contributions as a check book. Funding which was in excess of current needs was spent in the unified budget to reduce the appearance of a deficit. That compounds a problem. One the one hand the deficit appears lower than it is. On the other we keep building up chits in the retirement accounts which are funded with unbacked bonds. Is there a problem with increasing the deficit? I am sure Baker would argue that there is. But for some reason on this program he seems to think that the issue is neutral.

The way the federal government accounts for some asset based transactions is a subject of debate among some experts in public finance. That is an esoteric argument that adds to the problems described here - but it is a bit off track. But the increasing reliance on debt to finance government activities is a problem. In the case of Social Security there is a double problem because we have made the explicit promise that sometime in the future we will get something back. Baker argues that this is no different than any other program - but that is not correct. If we spend too much on something like environmental protection we can curtail it if the deficit is too large. We could do that here too but not without some substantial political fights. Is not the AARP one of the most virulent defenders of the entitlement mentality of the Social Security program? If the added expectations of this program were not greater than other programs, why would the Social Security Administration send out those annual calculations about how much money each of us has put into the system and how much money we can expect at 65. Do the Treasury notes which finance our other debts make similar claims?

We maintain a fiction that the Social Security program is financed by tax contributions that are somehow dropped into a trust fund for a future draw. That gives the appearance to the program to be a retirement program. But the federal program does not follow the accounting rules of either pensions or insurance. In private companies the accounting rules require an actuarial estimate of the fund and its future draws. If there is not enough money in the fund to pay future draws, then the company's balance sheet gets dinged. Likewise, the pension systems for governmental employees, including the Civil Service Retirement System, are governed by actuarial rules. Assets have to equal future obligations. In the case of Social Security, those rules do not apply. Right now we are exchanging the tax revenues for bonds which are ultimately conditioned on a tax contribution which we can estimate. But the supposed deposits actually go into nothing and the obligations are real - with the continuing expectations created by those little slips sent out by the SSA each year.

There are a couple of alternatives available when we come up to the problem, sometime in the future, when the number of taxpayers putting money into the fund comes a lot closer to the number of people drawing out. First, we could slash benefits for those currently receiving payments. A second alternative, would be to increase fairly substantially the taxes paid by the current payors. That is what we did in 1983. In our current political environment neither of those options is likely to happen.

So you are left with a couple of misimpressions. On the Social Security Administration page - FICA taxes are described thusly "The payroll taxes collected for Social Security are of course taxes, but they can also be described as contributions to the social insurance system that is Social Security. Hence the name "Federal Insurance Contributions Act." The page on the history of social security also comments "Retirement programs for certain groups of State and local government employees—mainly teachers, police officers, and fire fighters—date back to the 19th century." There is a clear implication that Social Security is different from other programs. Indeed, there are multiple motives for the program - that include social welfare programs. But the main emphasis is the look and feel of a retirement program. If Social Security is not bound by insurance standards of actuarial standards then why refer to it as "social insurance?" Does government insurance work differently than other insurance programs? Does one government retirement program operate by entirely different rules than programs like the Railroad Retirement or Civil Service systems? If it goes by different rules yet calls itself the same thing - doesn't that make it a fiction?

Baker and others like him, ultimately would like to move Social Security into a general tax obligation. The FICA tax would be eliminated. But there was a reason why, when the Social Security program was created, we created a separate tax stream for this program. In part that distinction came with the recognition that people above the income ceiling of the tax would realize smaller benefits net than those who paid into the system for all of their income. Those above the ceiling would finance their retirements with private assets and programs. That is less true for Medicare (again a separate tax but again a long term problem where the tax will not cover the costs of the program). Politicians have complicated the distinctions by adding a lot of social welfare payments under the guise of Social Security. As opposed to Baker, I would separate the welfare payments from the rest of the system. One can calculate how much it costs to offer those payments and use a separate tax to fund them. But the real obligations of the system - which are health and retirement payments - should be treated like an insurance or retirement program.

Not doing that makes the program look like a Ponzi scheme. Indeed, if one looks at the payouts of the first 30-40 years of recipients versus the projected payouts of the next 30-40 years, the program looks like a classic Ponzi. The first recipients received huge rewards for their contributions. The current generation will barely be able to generate something close to passbook savings returns. Future generations, without a significant bump in contributions will see some form of either reduced benefits or increased tax contributions or more likely both. Again, that payout scheme, which Baker seems almost oblivious to, is a fiction. Why would any sane person contribute to a retirement system that guaranteed his father significantly higher returns than he would get and guaranteed that his children would potentially get negative returns?

In Alice and Wonderland characters are able to change the definition of things at their whim. That is a classic fiction. If this "social insurance" program is not bound by the standards of any other insurance or retirement program then why continue to refer to it in those terms? One cannot have it both ways.