Social Security and Deficit Reduction: Some Fun with Numbers

by Bruce Webb

Well it looks like we are going to get a Deficit Commission and one way or another Social Security will be on the table. But what exactly does that mean for either Social Security or the total deficit picture? Before answering that lets review a couple of basics.

In talking deficit reduction we need to distinguish between ‘deficit’ and ‘debt’. A ‘deficit’ is basically an accounting convention, it is worthwhile to calculate various gaps between income and cost. But not everything labeled ‘income’ actually represents cash extracted from the economy, in particular interest on Intragovernmental Holdings is not in fact financed by such dollars, although it is real as real in the long run, in the immediate term it is simply a bookkeeping entry. Which leads to some really odd results. In order to keep things straight the CBO maintains two different measures for deficits: ‘primary deficit’ which measures actual cash flow and ‘deficit’ which includes that interest. Now in general newspapers report the deficit and not the primary deficit, when we talk about an Obama $1.6 trillion deficit for FY2010 that figure silently includes interest on the Intragovernmental Holdings as a positive number, whereas a calculation of primary deficit wouldn’t include it at all.

Turning to ‘debt’. Last week Congress raised the debt ceiling from $12 trillion to close to $15 trillion. But this does not mean Treasury is free to borrow up to $3 trillion in cash prior to going back to Congress, because there are two different components of debt subject to the ceiling. You can check out both the total debt and its two components via a web application maintained by Treasury called Debt to the Penny which would tell you that as of Monday ‘Debt held by the Public’ was $7.849 trillion while ‘Intragovernmental Holdings’ were $4.499 trillion for a total of ‘Public Debt’ of $12.349 trillion, which is the number you normally see reported in the newspapers.

Mostly when people talk ‘deficit reduction’ they do so not in the context of opening borrowing room in the current year or its effects on interest rates, those those are pretty important things indeed. Instead they tend to think of it in terms of debt reduction or perhaps just a slowing in the growth rate of Public Debt. And one way to accomplish that is to cut spending. Or so you would think, once you start factoring in the curious treatment of interest on Intragovernmental Holdings things start getting strange. Follow me below the fold if you dare.

More than half of the $4.499 trillion in Intragovernmental Holdings are held by the two Social Security Trust Funds, in fact they have $2.5 trillion in Treasury obligations that score as part of Public Debt. But for the sake of this argument I want to simplify that a little:

Suppose you have a Trust Fund with a balance of $2 trillion in notes earning 5% a year which operates along side a benefits program with its own dedicated revenue stream that either will or will not pay all costs in any give year. Now imagine three different scenarios:

A1: the benefits program self-funds with non-interest income equalling cost
B1: the benefits program runs with an operating loss of $100 billion
C1: the benefits program runs with an operating surplus of $100 billion
What are the subsequent impacts on total Public Debt?

Under A1 there is no call on funds from Treasury nor is there any cash flow to it. But Treasury does have to create $100 billion in the form of new Special Treasuries to credit to the Trust Fund to account for interest. This $100 billion adds to Intragovernmental Holdings which in turn adds to Public Debt for a net increase in that debt of $100 billion.

Under B1 there is a call on $100 billion in funds from Treasury to meet costs. To meet that call Treasury has to borrow (or find revenue, same thing for our purposes) $100 billion in Debt held by the Public. In turn it offsets that by not giving any credit to the Trust Fund for accrued interest, instead ‘taking’ it in exchange for that other dept. The end result is that $100 billion is added to Debt held by the Public and so to Public Debt while nothing is added to Intragovernmental Holdings with a net increase in total Public Debt of $100 billion, i.e. the same amount as A1.

Under C1 the operating surplus of $100 billion flows to Treasury and in principle pays down that same amount in Debt Held by the Public (or reduces borrowing by that amount, same thing) which reduces Debt Held by the Public and hence Public Debt by that amount. But under the rules in play Treasury is required to issue $100 billion in Special Treasuries to ‘pay’ for that borrowing and another $100 billion in such Treasuries to account for the interest which increases Intragovernmental Holdings and so Public Debt by that amount for a total net increase in total Public Debt of $100 billion. Same as A1 and B1.

So for the specific purpose of calculating Public Debt A1=B1=C1. I suggest this result is pretty damn counter-intuitive but absent some mechanism I am missing that is how the numbers run.

Okay now say in the interest of deficit and debt reduction we cut our benefits program by $100 billion a year while leaving its revenue constant.

A2. The benefits program, previously revenue neutral, now provides a $100 billion cash surplus to Treasury reducing Debt Held by the Public by that amount. In turn Treasury issues $200 billion in Special Treasuries to account for the borrowing and the accrued interest. Net increase in the Public Debt is still $100 billion.
B2: The benefits program, previously running at a $100 billion loss, is no revenue neutral with no cash effect on Treasury meaning that Treasury only has to issue a Special Treasury for $100 billion meaning a net increase in Public Debt of $100 billion.
B3: The benefits program, previously running a $100 billion surplus now is running a $200 billion one, reducing Debt Held by the Public and hence Public Debt by that same amount. But Treasury has to issue $300 billion in Special Treasuries for a net increase in Public Debt of $100 billion.

Now this is getting kind of spooky, not only does A1=B1=C1 for the purposes of the number on the Debt Clock, they are also precisely equal to A2=B2=C2. That is no matter whether the current system is running cash surpluses, cash deficits or is cash neutral cutting $100 billion in spending doesn’t move the Debt Clock at all.

Does this mean there is no real world effect? Well of course not, between the six scenarios we have impacts on Debt Held by the Public from up $100 billion to down $200 billion depending on the before and after states, that is real money. But all it really is is a transfer from Social Security beneficiaries to holders of Debt Held by the Public, it is just robbing Grandma so as to make it easier to pay off the Chinese Central Bank.

Making the attempt to attach at least this component of Entitlements Reform to raising the Debt Ceiling kind of a sham. Because for the specific purpose of calculating total Public Debt subject to the limit it is a wash.

Now lets turn to Deficits. Under Unified Budget scoring any surplus to Social Security including accrued interest counts as a positive meaning a $100 billion annual cut in Benefits will yield a $1 trillion ten year score plus any additional interest effects on the Trust Funds which actually would be substantial. Because every $100 billion added to or not subtracted from the Trust Fund will be generating 5% compound interest. But those same interest dollars which score as a positive on a 10 year deficit total actually score as an equal increase in Public Debt. This too is pretty damn counterintuitive, we have the same factor driving deficit and debt in opposite directions.

And just to insert some final confusion what does $100 billion in cost cuts per annum do for our old friend Unfunded Liability. Well it wipes it out, the problem being that it is simply replaced by Public Debt in the form of Special Treasuries even as the dollars used to build that fund were spent long since. Given all this I can’t see why any worker would support cuts to Social Security, all that does it cut current obligations while replacing them with Public Debt in the future, a debt that we can’t expect the capitalists of the future to be any more willing to pay back than the capitalists of today. As long as the Social Security Trust Funds are throwing off enough interest dollars to cover the gap between Income excluding Interest and Cost there is absolutely no reason why workers should simply sacrifice their own interests here. In this scenario all of the benefit simply flows to the bond market. Now theoretically there would be positive impacts on interest rates which might create some indirect benefits for consumers, and if we were experiencing Carter/Reagan double digit interest rates then maybe a case could be made. But under current conditions I just don’t see it.

Anyway the next time some one comes waving that $12 trillion figure as a reason to cut Social Security be sure to ask him to show his work, in this particular case cutting spending actually serves to increase debt. And weirdly enough so does increasing revenue. Such are the weirdities of Trust Fund accounting.

Final bonus nugget. What would a perfectly balanced Social Security system look like?
One it would run a small cash flow deficit, Treasury would be transferring some money to SS each and every year.
Two still Social Security would overall score as being in surplus for Unified Budget purposes.
Three beyond whatever borrowing was needed for One, Social Security would be adding to Public Debt every year.

Yep a system that is cash flow negative, in permanent surplus AND adding to debt. Perfect! Wrap your head around that for awhile.