View all Articles
Commentary By Charles Blahous

What's In the Social Security Trust Funds, Or: Why Continuing the Payroll Tax Cut Could Eventually End Social Security as We Know It

Economics Healthcare

The ongoing effort to partially convert Social Security from payroll-tax-financing to income-tax-financing – by further cutting the payroll tax as a stimulus measure and replacing the funds with general revenues – may in short order put an end to the longstanding conception of Social Security as a benefit earned by worker contributions. The demise of this conception would also threaten the special political protections Social Security benefits have long enjoyed.

Most Americans do not know all of the details of Social Security finances. They do, however, retain a strong sense that Social Security participants somehow paid for their benefits, and that the program’s Trust Funds represent “their money” in a way that the financing for other government programs does not. This sense gives Social Security benefits an extra political protection relative to other programs. It would likely end if we abolished the Social Security payroll tax, did away with its trust fund, and funded the program with general budget revenues.

The proposed payroll tax cut extension would take a major step toward ending this longstanding special status. There’s no way to know where exactly the tipping point is, but it will come sooner than most observers now realize. Continuing and expanding this policy would likely soon turn bipartisan perceptions of Social Security into something more like welfare or at least like Medicare Part B: that is, benefits continually open for political renegotiation because they’re known to be subsidized from the general fund -- that beneficiaries themselves did not really pay for them.

As I’ve previously written, even under a “no action” scenario Social Security risks an eventual merger into the general budget (ending the special separation originally envisioned by FDR). A recent Ralph Bristol column persuasively argues that continuing to cut the payroll tax will make it inevitable even sooner that Congress “will formally adopt a ‘reform’ that is least disruptive by simply transferring a large portion of the legal liability for Social Security benefits to (the general fund) ‘where it will have been for the past many years anyway.’”

A detailed inspection of the Trust Funds and their income sources may help explain how rapidly the payroll tax cut might end the “earned benefit” rationale for Social Security.

The picture above divides Social Security’s $2.68 trillion Trust Fund balance (projected for the end of this year) into three pieces:

A: Surpluses of past program tax collections over expenditures;

B: Subsidies from the general government fund;

C: Interest credits.

Each of these three financing categories bears differently on the question of whether Social Security is a self-financed, earned benefit program.

Category A is the piece unambiguously consistent with the principles of self-financing. Whenever Social Security generates more in tax income than it spends on benefits, the surplus is deposited in its Trust Funds. If the program later draws on these funds to meet its benefit obligations, it is clearly just drawing on income it previously generated.

The vast majority of the money in Category A (98% over Social Security’s history) comes from the payroll tax, and the remainder from benefit taxation. Both tax sources contribute positively to the U.S. Treasury, and both thus contribute to the program’s self-generated income.

Category B, by contrast, is unambiguously inconsistent with the idea that Social Security benefits are earned and that the program is paying its own way. By the end of this year only 5% of the total Trust Funds balance will be attributable to such subsidies from the general fund. The lion’s share of this amount is in turn attributable to this year’s payroll tax break.

Category C – interest credits – is where fierce analytical arguments take place. Interest payments to the Trust Funds are made from the general fund, and viewed narrowly do not represent net additional income to the U.S. Treasury generated by Social Security. There are, however, two possible attitudes one can take towards these interest payments:

Attitude #1: The government’s ability to borrow from Social Security reduces the amount of other borrowing it must do, and thus its interest owed. Payments of interest to the Social Security Trust Funds thus represent actual net income generated by past program surpluses.

Attitude #2: Surplus Social Security taxes were spent rather than used to reduce government debt service. Payments of interest to the Social Security Trust Funds thus do not represent actual income generated, but are instead simply a further liability facing subsequent income taxpayers.

If one takes Attitude #1, one could justify grouping the interest payments in C with category A – that is, as consistent with the principles of self-financing. If one takes Attitude #2, then we would group C with B, because the interest credits would be an additional subsidy to Social Security provided by income taxpayers.

It’s beyond the scope of this article to explore this issue at length, but it has often been studied within academia and the evidence consistently supports Attitude #2. That is, the interest payments in category C represent, like category B, a further extent to which income taxpayers subsidize Social Security. But unlike category B, the question is arguable – that is, if the evidence pointed the other way one would reach the opposite conclusion.

One important additional technical point before moving on. Occasionally some commentators argue that interest payments to Social Security simply represent the repayment of money it previously loaned to the general fund -- that everything is “made even” once the money is repaid with interest.

This is incorrect. The reason it is incorrect is that the generations who must make the interest payments are not the same ones who borrowed the money. Basically, when one generation uses its surplus payroll taxes to buy additional government services for themselves, no additional money flows into the Treasury. When that same generation later receives Social Security benefits funded in part by credits of interest from the general fund, it is receiving a pure subsidy financed by younger income taxpayers.

It should also be remembered that while most interest earned by the Trust Funds to date is based on past tax surpluses, a small piece of it is attributable to general revenue transfers. This piece is another unambiguous subsidy from the general fund.

To sum up:

It is unequivocal that piece A represents an extent to which Social Security participants have paid for benefits.

It is unequivocal that piece B undercuts the view that Social Security is self-financing.

Piece C is arguable: one could construct a theoretical argument either way, though the empirical evidence favors the view that it is, like piece B, a subsidy from the general fund.

So where does this leave us? At the end of 2011, we will in effect be “a little bit pregnant.” That is, our enforcement of Social Security’s self-financing ethic will have been breached somewhat, but the total general fund subsidy to date would still be much smaller than net tax surpluses contributed by program participants. (Most of the Trust Funds would consist of interest credits, which are surrounded by analytical ambiguity.)

Now let’s look forward. Suppose that the President’s proposed policy were adopted of extending and deepening the current payroll tax cut. By the end of 2012, the Trust Fund balance would look like this:

This picture differs from the previous one in a very important way: already, subsidies from the general fund are beginning to compete in size with past net tax surpluses. Indeed, by the end of 2012 fully 37% of the Social Security Trust Funds’ non-interest balance would be attributable to general fund subsidies.

At what point would perceptions of Social Security being an “earned benefit” be irretrievably changed? Two approaching milestones might be worth noting:

  • If the President’s proposed policy were adopted and then later extended, then by June 2013 the share of the Trust Funds attributable to general revenue subsidies would actually exceed the share attributable to past net surpluses of tax income over expenditures (i.e., B > A).
  • If, hypothetically, the policy were adopted and extended indefinitely, by early 2015 the entire $3 trillion balance of the Social Security Trust Funds would be attributable to general revenue subsidies and interest credits (i.e., A < 0).

How would the public regard a Social Security Trust Fund that consisted entirely of general revenue subsidies and payments of interest between government accounts? Would they continue to regard Social Security as a truly “earned” benefit?

There’s no way to know. But one thing is certain: further continuation of the policy of replacing payroll taxes with income taxes will test perceptions of Social Security as an “earned benefit” to a degree that they have never been tested before.

Charles Blahous is a research fellow with the Hoover Institution, a senior research fellow with the Mercatus Center, and the author of Social Security: The Unfinished Work.