David R. Henderson  

Social Security: The Privatization Shell Game

Austerity for Liberty... The Cost of High Taxes...

Bryan posits a policy in 2005, had Bush succeeded beyond his wildest dreams, of privatizing Social Security. Bryan gives an argument against that. But there's a more basic argument that I made on a late 1990s talk show with Roger Hedgecock where the other guest was Representative Mark Sanford of South Carolina. Hedgecock, who now has a national talk show, had a show in San Diego. The way I put it then was that this kind of privatization is a shell game that ends up actually reducing economic freedom even if it works.

Why? Look at how Bryan worded it:

1. Workers are free to divert 100% of their future payroll tax into private accounts.
2. Workers are free to immediately ask for the present value of their future SS checks, as long as they put their money in reputable investments.

So the government is going to have to come up with revenues to pay two groups of workers: those who want the present value of their future SS checks and those who don't want it but, instead, want to get the checks. The government hasn't come up with any new revenues to do this. So what must it do? Raise some other taxes and not just a little, but a lot. Much of the future payroll tax will be off limits because of point #1: Workers will divert it into private accounts.

So what just happened here? The government imposed a new forced saving scheme. It, in effect, said to workers, "We were already taking X from you. Now we're going to let you use X the way you want, within limits--don't dare buy a refrigerator or pay for life-saving surgery. Oh, and we're going to have to take another almost-X to pay SS claimants." The government didn't increase economic freedom because it still needs to raise other taxes to pay for future SS promises. It just sliced off some current freedom.

Interestingly, when I made this point, Mark Sanford didn't object and he stayed friendly in the interview. In my experience with Congressmen, once you disagree with them in public on one of their most passionately held ideas, the knives comes out. I was so impressed with him that when I saw him running for governor in South Carolina and promising to get rid of the income tax over (if I recall correctly) 20 years, I sent him money.

Comments and Sharing

CATEGORIES: Fiscal Policy

COMMENTS (11 to date)
Phil writes:

It makes things even more expensive, because people with terminal cancer will immediately choose make that withdrawal from SS. Under the current system, they would just die.

I assume that only the spouse is entitled to some of your accumulated SS benefits if you die? That's how it works in Canada, I think.

Steve writes:

Sanford's affair is one of the worst things (within the realm of reasonable probability) that could have happened for the liberty movement in the short term.

mick writes:

Social security is itself a revenue device, not a forced savings scheme by any means. Savings implies that something is preserved for future use, social security does not do this.

Sure there is the "trust fund" that was embarrassing even by 1930's propaganda standards. In the internet age it just sounds stupid.

David R. Henderson writes:

That's my understanding of the law too, except there might be some benefit for children under age 18. When you die with no surviving spouse, you get $255.
I agree. He was my hope for Prez. Interestingly, he had insisted that the "faithful" part be removed from his wedding vows. So he didn't apparently break his word.
You're right about social security. The "forced saving" scheme I referred to was the one proposed by Bush.

John Goodman writes:

I don't understand what the surprise is here. All social security privatization schemes (there are more than 30 in the world right now) involve forced saving. And they are preferable to government-run Ponzi schemes because at least they are funded and do not require future taxes on others.

For what it's worth the economic rationale is the same as that given for Ponzi scheme social security -- to keep people from consuming all their income during their working years and imposing a negative externality on everyone else in old age.

Forced saving (called provident schemes in the old days) does involve loss of freedom. But it puts more constraints on politicians than Ponzi schemes do.

Mr. Econotarian writes:

A comparison I've heard recently is between the Chilean private accounts and US state sponsored college savings accounts permitted under Section 529 of the Internal Revenue Code. State agencies offer a choice of no-load mutual funds as investments.

Regarding "losing your retirement in the stock market", I think that people need to understand that there is a real risk of trying to invest in equities with a short time-horizon. We need to educate people about risk vs. reward. Equities have higher reward, but are more volatile. People should be out of equities in their retirement plan around 10 years before they retire, then move to bonds and other safer funds.

David R. Henderson writes:

@John Goodman,
John writes:
All social security privatization schemes (there are more than 30 in the world right now) involve forced saving. And they are preferable to government-run Ponzi schemes because at least they are funded and do not require future taxes on others.
John is right on both counts. However, we aren't starting from scratch. With government paying off the Ponzi scheme recipients, we are taxed to do so and then forced to save. So the new piece is the "being forced to save" part.

Bob Murphy writes:

Good post David. Yeah, I had the hardest time getting Republican conservatives to see that Bush's privatization scheme was smoke and mirrors. He was promising to let workers invest their SS contributions in the stock market, without increasing the deficit, without cutting any other spending, and without raising taxes.

I spelled out the specific flaws at the time, here and here. (I like to point to those articles when today's Democrats accuse me of being a Republican when I criticize Obama.)

Jim Glass writes:

Milton Friedman (of course) described essence of Bryan's scenario most clearly and simply.

SS is a paygo program, so tax receipts must equal benefits in total. Thus if from Time 0 to Time 1 participants receive benefits exceeding the SS payroll taxes they pay by $x, then after Time 1 participants must receive benefits that total less than the payroll tax they pay by $X.

The Social Security Trustees say that with Time 1 being today, $X is about $15 trillion. That is, if SS were terminated on say 9/15/10, with the payroll tax terminated cold turkey, current participants would still be owed another $15 trillion on promises made to them. (Participants from 1937 to 9/15/10 would be owed $15 trillion more than the payroll tax they've paid.)

Friedman, e.g. in debunking the idea of the "transition cost" as an obstacle to privatization schemes, made the implications clear:

This $15 trillion transfer owed to pre-2010 participants is a sunk cost. Whether it was a good idea or bad one for the polity to incur the debt resulting from this gift to the workers of the past is irrelevant today -- it is a debt incurred just the way the polity incurred the debts to fight WWII, for the moon progam, VietNam war, Obama stimulus, etc. We must pay it or renege upon it.

But there is no logical reason why this debt incurred by the nation must be financed with payroll taxes coming from a worker retirement pension program. We'd never imagine financing debts incurred by the nation to fight WWII or to fund the Obama stimulus only from the worker payroll tax base!

For instance, in the simplest case, if SS as it is were terminated on 9/15/10, with the $15 trillion debt converted to normal Treasury debt financed with general revenue, then as of 9/16/10:

[] Workers could get a "new" SS program providing bigger benefits at a fraction of the cost. E.g., a 6% wage contribution to a plan fully-funded with real investments earning a positive return for 40 years is a whole lot better than a 12.4% contribution earning a guaranteed negative return.

[] Because the income tax base is so much larger than the payroll tax base, the tax rate used to pay the $15 trillion backward transfer would be slashed (with the tax burden shifting up to the rich from the poor).

Most people would consider all those to be good things, improvements over the status quo -- if they ever thought about them.

I.e., all of SS's current problems result from trying to shoehorn the cost of that $15 transfer to the past into payroll taxes on the workers of the future. Which is totally illogical. But the popular psychological fixation with payroll tax and "trust fund" financing for SS stops people from realizing it.

In Friedman's words: "The link between the payroll tax and benefit payments is part of a confidence game..." It is a political ploy used to paralyze people out of considering logical reforms. (FDR himself famously said as much when creating the program).

So what just happened here? The government imposed a new forced saving scheme.

Well, yes, but the bottom line is that the $15 trillion is a sunk cost, there is no way to avoid it, it must be paid one way or another. Future taxes must exceed future benefits by $15 trillion with any scheme one comes up with, raising taxes, cutting benefits, whatever complexity one can devise.

It is compelled. But it is too late to change that now.

If SS were totally repealed today and *not* replaced with anything at all -- a libertarian dream? -- someone would still take that same $15 trillion hit.

In short, the real issue with SS is not its future solvency or bankruptcy, for SS's future net financial position is already determined.

The real issue is who is going to take the $15 trillion loss ... and how.

Oh, and the reason why issuing the $15 trillion of bonds on 9/15/10 in the cold turkey privatization above isn't a "transition cost" is because the $15 trillion gap is in the status quo. The status quo must pay it too -- status-quo-ers just hide from the fact by not issuing bonds to represent its unfunded promises. Something existing in the status quo isn't created by a transition from it.

Any status-quo-er who objects that the US govt can't afford to finance the $15 trillion from general revenue has to then explain how it can afford to do so using payroll taxes alone -- or admit that the future that the status quo promises is reneging on its promises.

Jim Glass writes:

Bob Murphy writes:

"Bush's privatization scheme was smoke and mirrors"
I was no fan of Bush's privatization model, and even less a one of the rationales his people tried to use to sell it.

But even assuming* that every dollar put into market investments would require the government to borrow another dollar, it in no way precludes such from providing real benefits. Belief to the contrary treads close to the "transition cost fallacy" repudiated by Friedman (as per my prior comment, which I hope will soon escape the moderators' icy grip).

The debt cost associated with SS is indeed fixed, and there is no magic wand to get rid of it -- but different methods of financing it can result in significant differences of result (as per the prior comment), just as different methods of financing any debt can make a difference.

Another potential benefit of pre-funding benefits not mentioned in the prior comment is that with pre-funding one borrows (or taxes) earlier rather than later. In light of the coming fiscal crunch circa 2030, see chart, earlier is better -- cheap rather than dear. (Well, it would have been when the 1994 Advisory Report first seriously considered the idea. Too late now.)

Also keep in mind that the return on SS contributions going forward is negative compared to the risk-free rate, projected by the SS Trustees at about 1% less than the T-bond rate. Over the 60 years or so one may be a SS participant that compounds up (or down). So the relevant investment comparison isn't between "safe" T-bonds and "risky" other stuff that should risk-adjust to equality, it is between negative and positive.

A schematic example of implications:

Say the real return on SS contributions is 2% (since the Trustees project real 3% for T-bonds) and market investments earn a very conservative 4% (if markets don't beat the risk-free rate over 30 years we'll have other problems to worry about than SS.) Also say under current rules one's work record for this year entitles one to $100 present value of benefits 30 years from now, upon retiring.

With a 2% discount rate $55 of payroll tax paid now legally "earns" that future $100 benefit, although it is isn't actually invested to do so. Instead you and the govt both risk the govt's ability to come up with the $100 in cash on its own then, amid the future financial crisis.

With a 3% rate, $41 earns that $100. With a 4% rate, $31 earns the $100.

So to pre-prefund that $100 benefit today the govt takes $41 of payroll tax, uses it to fund market investments for SS participants, then borrows $41 to pay the current retiree benefits that would have been paid with the tax.

Even if the market investments produce *only* the risk-free 3% (if markets produce less than the risk-free rate over multi-decades we are really doomed) the borrowing is OK for the govt, a wash, as $41 borrowed today = $100 borrowed then -- but $14 of payroll tax today is unneeded! That's a $14 gain, to be used for whatever, which at 3% compounds to $34 in 30 years.

If the market investments earn 4% (or more), the $41 in them produces $133 (or more). That $33 can be taxed and split between the govt and participant, say 50/50, $16.50 each. Now the SS particpant is 16.5% or more ahead on his $100 benefit, *and* the govt is $16.50 per $100 of future financing cost ahead, and that original $14 of tax savings is still freed up and compounding to another $34 as a bonus.

This is all because even T-bonds beat the below T-bond return on future SS contributions.
*Multiple studies find that the SS surplus has increased the debt held by the public, since the govt's credit line increases with its revenue. For instance, Kent Smetters:

"... each dollar of Social Security surplus appears to have actually increased the debt held by the public in the past by $1.76..."
Thus, eliminating the SS surplus by investing it in market assets might have reduced the borrowing of the govt. So the assumption that each $1 put in a market investment would have caused the govt to borrow $1 is not necessarily true.

And that is the end of my over-long comments on this topic.

Jim Glass writes:

Fixing some "writing at 3 am" sloppiness above:

"but $14 of payroll tax today is unneeded! That's a $14 gain, to be used for whatever", doesn't mean it is totally free money, profit to the govt.

It means the $14 is available, free for use one way or another, depending on circumstances.

It might become "free" money if SS is in surplus at the time, as I don't automatically assume that investing it in markets would cause govt debt to increase dollar for dollar (as per Smetters).

If SS is in deficit, the $14 could be applied re other benefits as described above, to get the advantage of the 3% risk-free rate versus SS 2% "earnings" rate, as described.

Even without that rate advantage it might be borrowed against to prefund benefits generally. I certainly wouldn't do that today, but in the mid 1990s, when the idea was first seriously examined, it could have made sense on the idea that 35 years of pre-funding while rates/taxes are low could provide a significant fiscal benefit during the coming fiscal crunch of the 2030s, by saving borrowing/tax needs then, when there might well be a fiscal crisis. But it is probably too late to do that without taking a big risk now.

Comments for this entry have been closed
Return to top