Arnold Kling  

Social Security Transition Cost

PRINT
Social Security Reform... Whose Debt is it, Anyway?...

Responding to a comment on the previous post, I write,


Suppose that your house has a very dilapidated roof. The next big winter storm is likely to cause huge damage unless it is repaired. So you borrow $20,000 and fix the roof.

What is the cost of this transaction? The "cash flow cost" is that you have to pay back the loan. Until you complete the payments, you will have less money available to buy a fancy home theater system or a new Jacuzzi.

The economic cost of this transaction may be zero. In an economic sense, you have exchanged a certain, visible cost -- the cost of repaying the loan -- for an uncertain, invisible cost -- the cost of trying to get through winter with a dilapidated roof. Chances are, if you did not fix the roof, you would still not be able to afford the home theater system or the Jacuzzi, because a storm could cause water damage and force you to pay huge repair bills.

...There are serious decisions to be made concerning the best mix of methods to finance the transition that extinguishes Social Security's unfunded liability. Borrowing all of the money to fund the transition is probably not the best choice. But the economic costs of the status quo are, if anything, worse than a transition financed 100 percent by borrowing. If spencer or his father-in-law were to complain about the cost of fixing the roof, I would remind them that the cost of not fixing the roof is at least as great.

For Discussion. How would various methods of funding a transition to privatization affect intergenerational fairness?


Comments and Sharing


CATEGORIES: Social Security



TRACKBACKS (6 to date)
TrackBack URL: http://econlog.econlib.org/mt/mt-tb.cgi/163
The author at Catallarchy in a related article titled Truly Understanding Social Security Privatization Costs writes:
    Truly Understanding Social Security Privatization Costs. Or, Why I Want To Be Arnold Kling When I Grow Up. [Tracked on December 2, 2004 7:39 PM]
The author at The Entrepreneurial Mind in a related article titled Welcome to the Carnival of the Capitalists! writes:
    Welcome to this week's edition of Carnival of the Capitalists from my office at Belmont University in Nashville! This is the second opportunity I have had to host COTC. My has it grown! I hope you enjoy what my... [Tracked on December 5, 2004 8:16 PM]
COMMENTS (20 to date)
Lawrance George Lux writes:

They will not affect intergenerational fairness from its current point, unless the liability is renounced. Rationale: Current or near-Retirees have ceased providing economic contribution. The only element which could alter this formula, is to start charging wealthy Retirees for contributions to the more needy Retirees, which no one will do.

Privatization, though, can be a disaster for Young Workers. They will have to assume additional debt, pay incremental charges for investment, and gain much lower benefits in the SS system. lgl

Mark Horn writes:
Privatization, though, can be a disaster for Young Workers. They will have to assume additional debt, pay incremental charges for investment, and gain much lower benefits in the SS system.
I don't understand this. Why would a young worker be required to assume additional debt? They're just taking money that they had been paying in payroll tax and putting it into a private account. Where is the additional debt?

Are you talking about the "transition cost"? If so, isn't the entire point of Arnold's article to identify that this cost is a wash? Doesn't that liability exist in both cases? One is on the books and one is off the books, but the liability still exists. As such, how young workers have to assume additional debt?

If you're not talking about the "transition cost", what additional debt are you talking about?

Jim Glass writes:
Privatization, though, can be a disaster for Young Workers. They will ... gain much lower benefits in the SS system. lgl

Private investments will provide higher returns, of course.

Investment markets have their risks -- but it takes a government to guarantee loss on investment over 40 years, as the Social Security tax-to-benefits formula does for young workers.

Boonton writes:
Private investments will provide higher returns, of course.

Which brings up the following question: If private investments guarantee a higher return, then why couldn't gov't simply borrow $500B today and purchase an index fund of stocks. 50 years later sell that fund off and use the 'higher returns' to fund social security and other expenditures?

Brad Hutchings writes:

Forget the question. It's already completely unfair. Any change that lowers the cost of future benefts makes it more fair.

To put fairness into perspective... My 81 year old grandfather drives a 5-series BMW. It's paid for. He also collects Social Security at the highest draw level. He also has a nice pension and investment portfilio to draw from. I can accept that that's the deal these people made with themselves and current workers are bound by it. I love him to death and wouldn't think to disparage him for it. But _any_ fair system would not have future BMW owners drawing Social Security until every worker could afford an Escalade.

We should all sit down and draw a line in the sand at 60 (2 years older than my own parents). Means test, rapidly increase the benefits age from 65 to 80 by at least 6 months every year, tell folks my age that our parents are our problem so long as they are not in true poverty. Call Social Security what it is -- welfare -- and see if the stigma keeps more people from going on the dole.

Imagine being 34 today, self-employed (so you see the employer contrbution) and wondering what you could do with the money being sucked up and how bad it will be 10, 20, 30 years out if this folly continues. Enough to make you sick...

Bill Fellers writes:

Brad,

You nailed me. I'm 34 and self-employed. Paying self-employment taxes makes me sick. I feel like a (partial) slave.

The world is full of primitive control freaks. Will it ever end?

David Bennett writes:

I'm inclined to believe reports that social security will last into the late thirties, if the federal government pays back instead of saying as Greenspan proposed, wage earners are fully responsible for this portion of the debt and the lower your wage the bigger your percentage.. The big problem is Medicare, but that's another issue and I expect the health care system to go through some major restructuring.

First thing is context. Assume 3% real growth, by 2030 per capita wealth will have doubled, so if we give the elderly a fixed amount of wealth then much of the problem will be solved.

To me the problem is low wage earners, and personally I'm tired of enterprises which pay decent wages being hurt by those that don't, I disagree with a system where certain people make 70% or less of what they made for less production (their have been improvements in technologies) 30 years ago.

So quite simply every employer is required to put a buck an hour into an indexed diversified retirement account with costs under 1%. Those who are already paying retirement will be covered, but those who don't will have to move to modern methods of production. If hamburgers go up a nickel fine (labor is about 25% of costs in most fast food.)

I feel the same about health care. Taxpayers and private insurance payers are subsidizing employers who don't cover costs. I'm tired of this welfare for "small business." The jobs won't go away, most of these things need to be done. It is simply feudal to punish those who can't protect themselves through unionism or the informal conspiracy which decides jobs that high school graduates could once do require a college degree and twice the pay.

Take care of the bottom and the big part of the problem goes away.

Jim Glass writes:
"Private investments will provide higher returns, of course."

Which brings up the following question: If private investments guarantee a higher return, then why couldn't gov't simply borrow $500B today and purchase an index fund of stocks. 50 years later sell that fund off and use the 'higher returns' to fund social security and other expenditures?

First, the comparison wasn't between the returns on stocks and gov't bonds.

It was between the return on market investments -- including in gov't bonds -- and the SS benefit formula which fixes by law negative returns for young workers.

It takes a government to guarantee negative returns over 40 years.

Second, regarding your comparison, George Ball, the noted former head of the Social Security Adminstration and long-time liberal champion of Social Security, proposed back in the early 90s that the SS trust fund be invested in stocks to shore up SS's solvency. That would today be not $500B but $1.5 trillion.

Lawrance George Lux writes:

Mark and Others,
The first dose of reality to be faced stands as the fact Government debt is never transitional. It will be paid off when American Taxpayers are forced to pay it. Arnold's assertion that it is a Wash, is not a Wash.

Boonton mentioned in a previous Post that actual Profitability in the Investments will not be present, due to the blend of Winners and Losers, and the gain will be incremental.

The last point to be mentioned is the injury to the Stock market itself. The real increase in monthly contributions since the 1970s from Mutual Funds,etc., intensify all market trends--creating Balloons and Grand Canyons. The introduction of Private SS Accounts will create Science Fiction on Wall Street. lgl

Boonton writes:
It takes a government to guarantee negative returns over 40 years.

Second, regarding your comparison, George Ball, the noted former head of the Social Security Adminstration and long-time liberal champion of Social Security, proposed back in the early 90s that the SS trust fund be invested in stocks to shore up SS's solvency. That would today be not $500B but $1.5 trillion.

This doesn't really answer the question Jim. Social Securities surplus is effectively invested in gov't bonds by reducing the overall debt the US gov't has in the public hands it cuts its interest payments thereby generating a return. In reality you should add to SSI's 'negative return' the return of reduced interest costs to taxpayers for the decades of surpluses generated by SSI.

After all, if SSI taxes had been reduced back in the early 70's to a level where the trust fund would never hold a surplus but also match outflows to inflows down to the dollar, then people would have had more disposable income for the last 30 years but the gov't interest costs to taxpayers would have been higher.

spencer writes:

I am glad to see my comments have an impact.

I agree with you completely that changes in the SS are needed and should be discussed on a rational basis. What I was objecting to was the technigue you were using to try to mis-define the problem. The arguments I keep hearing that there are no transition costs is a classic example of "the big lie". At least I forced you into an honest discussion of the issue rather than trying to get away with your silly debating techniques that may fool the typical newspaper reporter but not anyone that wants to have an honest discussion.

Edge writes:

An example for transition costs.

Suppose we issue $2 Trillion in public debt, to finance private accounts.

Then suppose there is some significant external event - a real war, plague, cheap energy runs out and technology doesn't rescue us soon enough. This happens somewhere in the short to medium timeframe.

Is there then any real transition cost? Perhaps the federal govt could reclaim all the private accounts to help finance a massive government action. But even if the govt could do that, assets reclaimed would have a depressed value, and meanwhile, we'd likely still be obligated to pay back the public debt.

I think we've got real transition costs no matter how you slice it. Just as well to make the transition pay as you go as to try to issue something akin to margin loans to be paid back by the government, with the assets put into private hands.

If we say there's a free lunch, then practically any government project can justify issuing bonds. "There is no transition cost, because it offsets future liabilities that weren't properly accounted for".

Jim Glass writes:
An example for transition costs.

Suppose we issue $2 Trillion in public debt, to finance private accounts.

Then suppose there is some significant external event - a real war, plague, cheap energy runs out ...

Hello??

And what do you imagine is going to happen starting about a decade from now when the goverment has to start issuing trillions of dollars of new public debt to pay for SS under the status quo? Will that be a "transistion cost" to you?

The one thing we know for sure is that the government's fiscal position will be much worse then than today, and be worsening fast due to the huge growing cost of Medicare, Medicaid and, under present law, Social Security. With the annual deficit heading to 20% of GDP by about 2040, says GAO, on current law.

Now in that fiscal position so much worse than today's, start adding the trillions of new debt and taxes that will be needed then to pay SS benefits under status quo rules.

Then suppose there is some significant external event - a real war, plague, cheap energy runs out ...

Will you be happy then that you didn't pre-finance that debt when borrowing was so much easier?

Jim Glass writes:
This doesn't really answer the question Jim.

What question? Do you mean:

If private investments guarantee a higher return, then why couldn't gov't simply borrow $500B today and purchase an index fund of stocks. 50 years later sell that fund off...?

Well, being that the head of the SSA wanted to do just that (and that state governments are required to do it with their retirement programs, and that foreign governments do it), the apprarent answer is that there is no reason at all why the federal government couldn't do it too.

Except politics. (See all the political objections being raised now.)

Jim Glass writes:
"Social Securities surplus is effectively invested in gov't bonds by reducing the overall debt the US gov't has in the public hands it cuts its interest payments thereby generating a return."

Oh, you keep making this claim as the rock upon which you build your church, but with no support whatsoever that I've seen.

What does it contradict? Logic, which says politicians will rapidly spend and tax-cut away money at hand to buy votes. History, in which they did just that with the surplus of circa 2000 and the big SS tax inflow of 1939. Authorities, like Rudy Penner, head of the CBO, who said spending exploded in 1998 because of the surplus (which was all SS then) and Moynihan, who knew both SS and government spending first-hand, and said Congress "embezzled" the SS surplus to spend it as free money.

And here's an academic paper:
http://irm.wharton.upenn.edu/WP-security-Smetters.pdf

"We find that there is no empirical evidence supporting the claim that trust fund assets have reduced the level of debt held by the public.

"In fact, the evidence suggests just the opposite: trust fund assets have probably increased the level of debt held by the public...

"Social Security surpluses not only appear to have failed to decrease the debt held by the public, each dollar of Social Security surplus appears to have actually increased the debt held by the public in the past by $1.76.

"At first glance, this dramatic overspending of Social Security surpluses seems entirely implausible. However, the next section presents a simple game theory model that is consistent with this result...."

OK, now if you can produce any papers or authorities documenting the contrary, I'll be glad to consider them. But until then ... c'mon, this is just a religious belief of yours.

After all, if SSI taxes had been reduced back in the early 70's to a level where the trust fund would never hold a surplus but also match outflows to inflows down to the dollar, then people would have had more disposable income for the last 30 years but the gov't interest costs to taxpayers would have been higher.

Aw, taxes were that low in the 1970s, there were no trust fund "savings" back then. That's why SS went bankrupt twice, in 1973 and the early 80s, and had to bailed out twice with higher taxes.

So you have to admit that "interest costs to taxpayers" were higher, by your logic, eh?

Now, can you explain how the trillions of dollars of unfunded liabilities that had already been piled up when SS went broke in 1982 had been financed by "savings" in a trust fund of $0?

If you can do that, it will be easy for you to explain how the $12 trillion current value in future liabilities over future payroll tax collections are being financed by a current trust fund balance of $1.5 trillion.

You know, a hallmark of religious belief is how it is impervious to factual reality. Trust fund savings of the 1970s ... sheesh.

Boonton writes:
Oh, you keep making this claim as the rock upon which you build your church, but with no support whatsoever that I've seen.

You must have missed the part where I actually went back and looked at spending, taxes and debt as a % of GDP and found no connection. You must have also missed the regression analysis I posted (which anyone else here is free to do) showing no connection except a possible negative relationship between tax revenue and spending.

What does it contradict? Logic, which says politicians will rapidly spend and tax-cut away money at hand to buy votes.

Key phrase there is 'money at hand'. Anyone even partially familar with the Federal Budget knows only a tiny bit of it is 'money at hand'. For that piece, your logic probably does hold. However the bulk of the budget is not 'money at hand' but 'money on auto pilot'. Tax receipts are based on the underlying economy and much of spending is on autopilot.

When times are good spending is pushed down in terms of GDP because fewer people apply for benefits, likewise tax revenue goes up. As I stated before it is quite easy to miss this relationship because those looking 'at the trees' do not see the larger effects of a growing economy, growing population, and inflation making a dollar in one year different than a dollar in another.

Edge writes:

No, we wouldn't be happy in that case, because we would have gambled and lost.

The gamble is on the margin loan. If the treasury issues a couple $Trillion in junk bonds now, and the offsetting asset is worth less at some future date because of a crisis, we'd be worse off. The treasury would still owe the old debt, and its ability to finance further debt might be limited, and would be more costly. Since much of the rest of the economy is leveraged around the "riskless" long term bond yield, the whole economy could be depressed.

It's like the old story of the ant and the grasshopper. We'd be better off increasing our savings rate rather than gambling on financial engineering.

Jim Glass writes:
"Oh, you keep making this claim as the rock upon which you build your church, but with no support whatsoever that I've seen."

You must have missed the part where I actually went back and looked at spending, taxes and debt as a % of GDP and found no connection. You must have also missed the regression analysis I posted (which anyone else here is free to do) showing no connection except a possible negative relationship between tax revenue and spending.

No offense, but I've repeatedly asked you for support from any authority in the world other than yourself.

I also pointed out the holes in your regression analysis that one could drive a truck through, and I haven't seen you do anything to close them.

Look, I'll glady post again an authority supporting my belief other than myself, with a professional regression analyis included!

Is the Social Security Trust Fund Worth Anything?, Kent Smetters, Wharton and NBER.

~ quote ~

We find that there is no empirical evidence supporting the claim that trust fund assets have reduced the level of debt held by the public. In fact, the evidence suggests just the opposite: trust fund assets have probably increased the level of debt held by the public....

... each dollar of Social Security surplus appears to have actually increased the debt held by the public in the past by $1.76.

At first glance, this dramatic overspending of Social Security surpluses seems entirely implausible. However, the next section presents a simple game theory model that is consistent with this result...

We show how this counterintuitive result can be explained by a simple "split the dollar game" where competition between two political parties exploits the ignorance of voters who don’t understand that the government’s reported budget surplus actually includes the "off-budget" Social Security surplus...

~ quote ~

Holy Cow! The the trust fund has increased the government's debt by $1.76 for every $1 it's collected! That'll give some people religion!

There, now it's your turn.

If you can't find one published quantitative study in the whole wide world, other than by your own self, to contradict this, Penner, Moynihan, the historical lessons of 1939 and 1998-2003, well ...

Edge writes:

Any study of the trust fund has to show a negative correlation between trust fund assets and publicly held debt.

The trust fund had practically no assets until 1983, after which it has been steadily funded up to the current level around $1.5 Trillion. We issued massive public debts between 1982 and 1993, and again from 2002 through 2004.

You don't need to do any academic study to establish that correlation. Just look at the spreadsheets posted by the treasury.

The data isn't sufficient to generalize; guilt can be assigned to specific Presidents and Congresses, with a nod to the timing of economic cycles.

Boonton writes:
I also pointed out the holes in your regression analysis that one could drive a truck through, and I haven't seen you do anything to close them.

I must have missed that post. Perhaps you can summarize them again or email them to me personally.

Comments for this entry have been closed
Return to top