Arnold Kling

Comment of the Week

PRINT
The Risk Premium... Stock Prices and Scandals...

Because the comments that people are submitting here are unusually good, I think I'll feature at least one every week. This week's "comment of the week" is from Jim Glass on the Deficit Argument IV thread. He links to an interesting article by Milton Friedman suggesting a way to immediately transition from Social Security to private retirement accounts.

I don't mean to slight any of the other interesting comments, including others on that thread and several on the Consumption Tax thread. Thanks for contributing.


Comments and Sharing


CATEGORIES: Social Security



COMMENTS (38 to date)
GT writes:

I won't fully repeat what I wrote in response to Jim's point but let me summarize.

Friedman's position is based on millions of people willingly dropping any claims to SS while they continue to pay other's benefits.

Does anybody believe that is in any way realistic?

David Thomson writes:

"Does anybody believe that is in any way realistic?"

I agree that Friedman's proposal is going nowhere. We must also stop deluding ourselves about "retirement." Most white collar professionals really never really retire until they are completely physically incapacitated. We are truly discussing the blue collar worker who held a very boring job.

Has anybody ever heard of Peter Drucker? Also, Jack Welch retired from GE---and is constantly busy giving speeches and advice.

Patrick R. Sullivan writes:

GT wrote in the original comments:

" If you can manage that politically then why don't we ask those people to pay for everything else."

Are you being facetious? We already do that, Friedman is simply saying we should treat SS the same as any other program.

GT writes:

Patrick,

I think you missed my point.

As I understand him Friedman says there is a cost-free way to move from the current SS system to a fully private one.

It is based on millions of people willingly dropping any and all claims to SS benefits while AT THE SAME TIME continuing to pay for the SS benefits of others.

Obviously if Friedman can find this goup that is so selfless why don't we ask them to take the remaining fiscal burden from the rest of us?

The point is that this group does not exist and, therefore, Fridamn's proposal is not a real one.

I could be missing something because I found Friedman's article a bit confusing. Am I wrong? Is Friedman saying something else?

Jim Glass writes:

GT writes...

"As I understand him Friedman says there is a cost-free way to move from the current SS system to a fully private one. It is based on millions of people willingly dropping any and all claims to SS benefits..."

Not so. Everyone would get back the full value of the benefits they've earned to date. Nobody gives up anything. As Friedman says...

~quote~
"For everyone [not retired] this would be an obligation due when the individual would have been eligible to receive benefits under the current system. And the maturity value would equal the present value of the benefits the person would have been entitled to, less the present value of the person's future tax liability, both adjusted for mortality."
~end~

So everybody gets in full everything they've earned to date. Nobody gives up a claim to what they've earned under the present system. The cost to the government of paying these claims doesn't change at all.

But the means of paying moves from regressive payroll taxes to progressive income taxes, which is unambiguously better for the poor working classes, who are relieved of paying a direct transfer to a prior generation of people richer than themselves. The obligation simply shifts to Warren Buffett, Bill Gates and the other "rich", who can much more easily bear it and surely will gladly do so.

"... while AT THE SAME TIME continuing to pay for the SS benefits of others. "

Exactly backwards. The majority of Americans who now pay more in payroll than income taxes would *receive in full* the value of all the SS benefits they have earned to date ... while AT THE SAME TIME being *relieved* of paying any more payroll tax transfers to the prior generation. That sure looks like coming out ahead to me.

"Obviously if Friedman can find this group that is so selfless why don't we ask them to take the remaining fiscal burden from the rest of us?"

This group is not "selfless", it is *benefitting* from the Friedman plan, obviously. Given time, this group, which is already in the labor market, may selfishly insist on the Friedman plan. (If SS's "defenders" don't smarten up soon and move to a Feldstein-, Moynihan-, or Swedish-style reform first, while there is still enough time to do it.)

Remember how Krugman called SS a "Ponzi game" that is running out, with today's young workers getting back no more or *less* than they put in? Well, if you don't believe in such selflessness, then how long do you think this Ponzi game is going to last?

Here on current law is the lifetime value of benefits to be received from SS relative to the amount of payroll taxes paid in, discounted to current value, as computed by the Social Security Administration's actuaries for persons entering the work force in 1994:

Single males:
low income - 2.6% (Ouch, the poor poor)
medium income - 27.5%
earnings limit income - 52.1%

Single women:
low income +12.9%
medium income -16.0%
earnings limit income -44.0%

Married couples with kids, SS's big winners.
low income +99.8% (Jackpot! That's over 2.5% a year!)
medium income +49.7% (2nd place, 1% a year)
earnings limit income -1.1%

[From "Ask an Actuary" at www.ssa.gov]

So six out of nine demographic groups will receive *negative* returns. And the largest return projected for anybody will be a little over 2.5% a year over a 45 year working life (age 22 to 67), say the SSA actuaries.

And that's with these benefits 25% underfunded! Of course, the unavoidable closing of the funding gap by reducing benefits and/or increasing taxes must drive these returns still lower in the real future world. So reality is going to be worse than the numbers above.

"The point is that this group does not exist".

Well, if a generation that is happy to get back from SS less than it puts in does not exist, then clearly SS is doomed, for that generation has already entered the work force. Just give them time to accumulate sufficient numbers....

GT writes:

Jim,

I think you are missing my point.

Let me make clear that I support moving from the current SS system to a defined contributions system like the IRAs.

But what I don’t support is the idea that there are no costs involved in this transition. That’s simply not true.

Friedman’s idea is as you say, that everyone gets in full what they have earned AS OF TODAY. The capitalized words are the crucial ones. That’s where the transition costs comes in because any money paid after the new system is put in place will not result in any new claims. That’s why I wrote that for it to work there must be millions of people willing to pay for the benefits of others while they themselves never receive a dime. People get paid in full what they put in ONLY as of the day of the conversion but they don’t get paid anything of what they put in AFTER the day of the conversion.

Maybe an example will make it clearer. Imagine a 21-year-old who starts working for the first time the day the new system is put in place. He has no claims on SS and receives none of those new bonds. But he has to pay for those bonds. Whether through a FICA tax or an increased income tax it makes no difference. He still has to pay.

Somebody in his 40s will get bonds covering what he paid in the past. But he will still have to continue making payments (to cover the new bonds for himself and others) and on that NEW money he will not get a dime. And so it goes in a continuum all the way to those who have just retired and get their full benefits without paying anything more.

Another way of looking at it. Suppose that the amount of bonds that the government needs to hand out is $1 trillion dollars (just to use a round number). Somewhere, somehow you need to find taxpayers willing to pay $1 trillion in taxes and ALSO willing not to receive a dime back in benefits on those $1 trillion they paid.

We can debate who would pay and how (greater income taxes, a new estate tax, a sales tax, whatever) but somebody has to pay and that somebody has to be willing to do so without getting anything, not a single dime, back.

In fact, as I was re-reading your post I saw where you got it wrong. You say that : "The majority of Americans who now pay more in payroll than income taxes would *receive in full* the value of all the SS benefits they have earned to date."

And yes, you are right about that.

But then you say: "while AT THE SAME TIME being *relieved* of paying any more payroll tax transfers to the prior generation."

And that’s where you are wrong. They are not relieved of paying taxes. They may not be called payroll taxes, but whatever they are called, they WILL have to make transfers to previous generations because somebody has to pay for those new bonds. And that’s the crucial point (again).

So that’s why I, somewhat facetiously, suggested that if you can find those taxpayers why not ask them to pay for other stuff as well. If there are taxpayers that are willing to pay $1 trillion without getting anything in return –and I want to make sure that it’s clear that under Friedman’s proposal they get NOTHING in return for that $1 trillion; if they get any bonds that’s for money they paid in the past, not for the NEW $1 trillion—then let’s see if they will pay another trillion and use that to pay for prescription drug benefits and all the other programs people want.

I have other problems with Friedman’s proposal, even if you could find those very generous taxpayers. But I’ll leave that for another post.

Paul Zrimsek writes:

It's you who are misunderstanding, GT. Yes, you have correctly identified a cost. But it is not a cost-- i.e., a new cost which will exist because of changing from the status quo to Friedman's plan-- it is merely the already-existing unfunded liability of the system, which can be avoided only by defaulting on its existing obligations. Now there's a political non-starter!

Your 21-year-old, therefore, is going to be paying taxes in return for nothing no matter what. If we don't adopt the Friedman plan, the taxes will go to keep SS limping along once outlays start to exceed receipts sometime around 2016. Perhaps he can comfort himself with the thought that it won't exactly be the first time someone's sent money to Washington and not gotten it back.

I suppose you could make a case that the Friedman plan is at a political disadvantage because it takes away our ability to pretend that the unfunded liability can be made good without screwing anyone over.

Paul Zrimsek writes:

Second sentence in the preceding should begin, "It is not a TRANSITION cost...."

GT writes:

“Your 21-year-old, therefore, is going to be paying taxes in return for nothing no matter what.”

How so?

If the system continues as is the 21 year old will be paying in taxes but will be getting SS benefits in return. It may not be a great return (which is one of the reasons I support getting rid of SS as it stands today) but it is a helluva lot better than a ZERO RETURN ON PRINCIPAL which is what he gets under Friedman’s plan.

The bottom line remains the same. If we are going to issue $ 1 trillion in bonds to pay for all existing SS liabilities (or whatever the actual amount is) that money will have to come from taxpayers and those taxpayers will not see a single dime back of that $1 trillion they pay.

The 21 year old pays today knowing that the 5 year old will pay for him in the future. Friedman is telling the 21 year old to continue paying but to not expect anything from the 5 year old.

Bernard Yomtov writes:

I don't think Friedman is suggesting that his plan be adopted as a practical matter. What he is doing is using it to illustrate that the sum usually referred to as "transition costs" is really just the unfunded liability of the current system.

He confuses his case somewhat by being careless about the term "present value," not specifying when the PV is to be calculated. As I understand it, the system is discontinued, and everyone gets a bond. The bond given a retired person has a present value equal to the difference between the PV of expected benefits and that of expected future payroll taxes. So the retiree is not helped or hurt, in expectation. The worker's bond is designed to have exactly that same value, at the worker's normal retirement age. So the worker is no better or worse off either.

On the other hand, the govenment has had to issue bonds equal to the amount of the unfunded liability, and these will have to be paid off by future taxpayers - GT's 5 year old.

If we permit these bonds to be traded we have created a private system, where the "transition cost" is exactly equal to the unfunded liability in the present system.

That's all.

GT writes:

Bernard,

I can’t say whether Friedman proposed this as a real alternative or not.

But I disagree with your definition of transition costs. They are not the same as the unfunded liabilities. Those are two separate concepts.

The unfunded liabilities are a cost, yes. They are the cost of keeping the system as is. They are the result, mainly, of changing demographics. They will have to be paid, one way or another if we want to keep SS on its current path.

The transition costs, on the other hand, refer to the cost of MOVING from the current system to a private account one. They are the result of millions of people no longer paying into SS or , alternatively, not receiving SS benefits.

There’s some overlap between the two concepts but they are distinct.

Maybe the real argument to be made in favor of moving to a privatized system is that the transition costs are lower than the unfunded liabilities so that it pays to make the change. Or, if they are not less than the unfunded liabilities they may be close enough. If Friedman’s numbers about the unfunded liabilities are correct than maybe that’s the argument to make.

Has anybody seen any study comparing these two costs?

Patrick R. Sullivan writes:

GT, try thinking of it as though the unfunded liability is a sunk cost. Which it will be if we issue bonds that have to be redeemed some day.

Then you should see that it makes no sense to base a decision about how to fund your 21 year old's retirement with an eye on that sunk cost. This is what Friedman's idea clarifies beautifully.

And your 21 year old will come out way ahead, as he can immediately begin investing with his P/R tax savings. Unless we change current income tax law, he probably won't be paying much, or even anything, in income taxes to redeem those bonds issued to retirees.

At least for awhile. When he begins to earn enough to get hit with income taxes, part of which will go to redeeming those SS bonds, he's had his investment in an IRA like vehicle accumulating interest or dividends. Each dollar he put away while 21, being worth--at 6% annually--about $18 when he's 71.

And as Paul Zrimsek suggested, he's doing the same thing (paying taxes) for all kinds of programs that are of no benefit to him.

David Thomson writes:

“Jeremy Rifkin numbers among current hydrogen zealots--while skipping over the small matter of where we get the hydrogen.”

Shucks, I guess I’m just a cynical dude. It seems to me that those who are most infatuated with the idea of hydrogen power could care less whether it makes any sense. On a gut level, they merely conclude: “Water isn’t yucky like that nasty oil. It’s clean and pure. Therefore, it’s got to be a good thing.” Why let a few facts get in the way of one’s utopian schemes?

David Thomson writes:

Whoops! please ignore that last post. It should have been placed elsewhere.

David Thomson writes:

We must reexamine the idea of what it means to be retired. Our present consensus understanding is bewildering and results in much confusion. I personally believe that few professionals truly retire in the sense that they never again significantly contribute to society. The vast majority that I’ve met in my life remain very active. For instance, when is the last time you’ve heard of a writer retiring? Can anyone even point out a single major author who woke up one day declaring “Damn it, I’m sixty-five and I don’t want to anything else but play shuffleboard in the mornings and watch reruns of Gunsmoke in the evenings?”

I sense that few are comfortable discussing the phenomenon of the blue collar worker who retires--and simply goes out to pasture to die. We must strongly encourage them to update their skills and knowledge. Ongoing formal and informal education can no longer be the lifestyle limited to those who attended college.

GT writes:

Patrick,

I'm afraid I don't really understand what you are saying.

I'm surprised there's even a debate on this. I have never heard anyone, on either side of the debate, say that there are no costs involved in a transition from SS to a private system. There are debates as to whether it's worth it (I think so) and to best way to pay for it but not whether it must be paid.

Now maybe I'm overestimating the resistance from those that will continue paying and get no benefit and ,if so, it may be a lot easier to make the jump than I expect.

But there's no way around that for Friedman's proposal to work there must be taxpayers that have to give up SS benefits that they were expecting to receive.

Paul Zrimsek writes:

Whether the Friedman plan would cost a 21-year-old SS benefits that he was expecting to receive is an open question. People in this age group very often opine that they don't expect to get anything from SS, ever. Perhaps this extreme viewpoint is nothing more than fashionable cynicism, but at the very least the kids seem to grasp the truth that their rates of return from the program are going to be very disappointing.

It must not be forgotten that if we go the Friedman route our 21-year-old will not be paying in (with no expectation of return) the full amount he's paying to SS currently, but only enough to support a dwindling cohort of incumbent retirees. The remainder of the money which, under the status quo, he would have paid into SS, he is now investing and earning a market return on, instead of the break-even-if-you're-lucky SS rate. Losing, say, 6.5% of your contributions outright but earning 7% on the remainder is the same as earning 0% on the whole lot. These are made-up numbers for illustration, but since the loss of principal under Friedman comes from the same source as the gap between SS and market rates of return under the status quo, the transition will in fact be a wash.

Whether this holds true for specific individuals as well as the class of taxpayers as a whole depends on the distribution of the general revenue tax burden as compared to payroll taxes. Since the former are far less regressive, the change to Friedman's system would be to the benefit of 21-year-olds, but to the detriment of some others. Here again, I'm willing to consider the possibility that those others will squawk and this will present a political obstacle.

Patrick R. Sullivan writes:

I'm trying to find a way to say this without sounding rude, but...

If you, GT, don't understand what I'm saying, then you don't understand what Friedman is saying. But, I can't think of another approach from the several offered you, to break you of your idee fixe. Unless Paul's most recent post does it.

And in support of what Paul says just above, remember that the bottom 50% of income earners are paying only 4% of income taxes. So, the very young, entry-level workers would greatly benefit. Ceteribus paribus, of course.

GT writes:

Paul,

I think you rightly point out that the devil is in the details. I doubt, no matter what the polls say, that anyone will willingly pay for decades for others SS benefits and accept nothing in return.

You point out that they will pay less and the 'remainder' can be invested at higher rates.

Is that really so?

Would the 21 (or 35 year old) really pay less? After all he has one or two generations ahead of him that expect to be paid in full. Wouldn't that cover the remainder of his working life? Has anybody calculated this?

I think all these obstacles are surmountable but we have to start by recognizing they exist and not pretend we can move to a new system at no cost at all.

Following our host's methodolgy let me leave a question for discussion:

If we moved to a private system how much leeway would we want to give individuals in deciding where to invest their money? Since we are forcing them to save for their old doesn't it follow that we should also impose limits so that at least some amounts are prefectly safde, no matter what?

Bernard Yomtov writes:

GT,

I think it would help if you described what you think the transition cost are that differ from the unfunded liability. Where do they come from? (Of course we all understand that there would be some purely administrative costs associated with a change, but I presume that's not what you're talking about).

A coupleof further points:

I'm not sure Friedman's argument, by itself, supports privatization as much as it explains that transition costs are not an argument against it.

There have been claims that there are no transition costs. Some of the more innumerate advocates of privatization simply want to stop payroll taxes, let workers fund their own accounts, and continue paying benefits to retirees. This is impossible.

GT writes:

Patrick,

I think Paul understood my point. I'm not sure you did.

Paul recognizes there is a transition cost (my point all along) but also thinks it may be washed out by the gains from a transition. He may be right but I'd love to see some numbers. And, as I wrote above, even if there are net costs it may be that the unfunded liabilities cost is such that the transition is cheaper than simply fixing SS.

But even if Paul is right that doesn't really help Friedman's argument since Friedman never mentioned the gains from conversion as offsetting the costs. He simply came up with a scenario which had no costs at all.

Same with my argument. Even if I suspect correctly and the costs of fixing SS are equal to or higher than the transition costs (so that we will have to pay no matter what) it also doesn't help Friedman's point since he never mentioned this.

GT writes:

Bernard,

Supposed we converted to a private system tomorrow. People receiving their SS checks would have to be paid. The money would come from workers who would receive no benefits form those payments.

That is a transition cost.

It may be, as Paul mentions, that workers won’t care because they will be paying less and the remainder will earn them so much more it will more than compensate. Or we may distribute the cost across generations so that it’s not just today’s workers that pay the transition cost.

But these costs are not what I understand to be the unfunded liabilities everyone talks about. Those refer to the amounts we expect to pay for SS benefits over and above what we expect to collect in the future. But today (and for several decades at least) we have no unfunded liabilities. According to the latest SS report the unfunded liabilities don’t start until 2044.

By moving to a private system we get rid of the future unfunded liabilities but we create new ones. In effect the transition costs are unfunded liabilities, just a different type than the unfunded liabilities we have been talking about in the past. It seems to me that a basic question in all this debate is which one is larger.

Does anyone know? Does our host have a link to any study on this?

Bernard Yomtov writes:

"Supposed we converted to a private system tomorrow. People receiving their SS checks would have to be paid. The money would come from workers who would receive no benefits form those payments.

That is a transition cost.

But these costs are not what I understand to be the unfunded liabilities everyone talks about. Those refer to the amounts we expect to pay for SS benefits over and above what we expect to collect in the future. But today (and for several decades at least) we have no unfunded liabilities. According to the latest SS report the unfunded liabilities don’t start until 2044."

Aha, GT. We're using different definitions. My notion of an "unfunded liability" is anything we are obligated to pay for which we do not at present have funds on hand. We do not right now have the funds on hand to meet the future obligations of Social Security. We intend to collect them via the payroll tax, and this will keep us out of trouble until some future day, maybe 2044.

Put another way, my definition is simply money promised in excess of what is on hand. Of course this excess is going to have to be made up by taxes. The current system plans to use payroll taxes, but there is no inherent reason why it couldn't be some other tax.

So Friedman's hypothetical plan, as I understand it, is to give everyone a bond whose worth is the Net Present Value of their future benefits and payments. This covers what you call transition costs, and what I call unfunded liabilities. It increases government debt, and that debt will be retired by future taxpayers, just as Social Security obligations will be met by future taxpayers under the current system.

Note that, for all the talk about privatization, Friedman's point is largely technical. Privatization does not inherently either incur or avoid the liabilities.

Patrick R. Sullivan writes:

In the face of such denial as:

"Patrick,

"I think Paul understood my point. I'm not sure you did.

"Paul recognizes there is a transition cost (my point all along) but also thinks it may be washed out by the gains from a transition."

When what Paul Zrimsek actually wrote was:

" It's you who are misunderstanding, GT. Yes, you have correctly identified a cost. But it is not a [TRANSITION] cost-- i.e., a new cost which will exist because of changing from the status quo to Friedman's plan-- it is merely the already-existing unfunded liability of the system, which can be avoided only by defaulting on its existing obligations."

I think I'm finished with any more attempts at clarification.

GT writes:

Paul's last post makes clear he understands my point although he counters that the transition costs may be more than compensated by transition benefits.

"It must not be forgotten that if we go the Friedman route our 21-year-old will not be paying in (with no expectation of return) the full amount he's paying to SS currently, but only enough to support a dwindling cohort of incumbent retirees. The remainder of the money which, under the status quo, he would have paid into SS, he is now investing and earning a market return on, instead of the break-even-if-you're-lucky SS rate. Losing, say, 6.5% of your contributions outright but earning 7% on the remainder is the same as earning 0% on the whole lot. These are made-up numbers for illustration, but since the loss of principal under Friedman comes from the same source as the gap between SS and market rates of return under the status quo, the transition will in fact be a wash. "

And I don't necessarily disagree, except to say that although it's a good theoretical argument I've yet to see any data backing it up, any calculations that show that the costs are really compensated by the benefits.

Finally, I will repeat that even if Paul is correct, or even if my own counter argument (that we may have to pay more to reform SS than the transition costs so let's just convert) is correct NEITHER of the two help Friedman's argument since Friedman's argument did not include any of these explanations.

Friedman's argument was that there is a cost free way to convert to private accounts. That is clearly false. It may be true that the costs are more than made up by the gains (Paul's point) or that the costs, while real, are still smaller than the alternative (my point). But what is not true is that there will be NO costs.

Paul Zrimsek writes:

If you insist on interpreting Friedman's zero-transition-cost assertion as "a transition cost exactly balanced by a transition benefit", I guess I can't stop you. But most people, I think, will find it seriously misleading to charge to the transition a cost which will have to be borne whether we make the transition or not. I'm certainly one of those people, which explains why I never once used the phrase "transition benefits".

Jim Glass writes:

^^ ... as I was re-reading your post I saw where you got it wrong. You say that : "The majority of Americans who now pay more in payroll than income taxes would *receive in full* the value of all the SS benefits they have earned to date."
And yes, you are right about that. But then you say: "while AT THE SAME TIME being *relieved* of paying any more payroll tax transfers to the prior generation."

And that’s where you are wrong. They are not relieved of paying taxes. ^^^

I said they are relieved of paying *payroll* taxes, and they are. The amount they are relieved of paying is shifted to another, generally richer, group through income taxes. Since this is a *shift* there is no net change in cost -- no net transaction cost results at all.

There is indeed a transaction cost imposed on the richer -- but this is exactly offset by a transaction *gain* enjoyed by the poorer.

And the latter have more votes than the former. ;-)

GT writes:

Paul, Jim

I still can’t figure out what it is that is being debated. Let’s number the arguments so it’s clearer. If any of you disagree with any of the points, or wonder what it is based on, having them numbered will make it clearer to respond.

Point 1- Friedman argues that there is a cost free way to fully move from the current SS system (where workers pay for current retirees and beneficiaries) to a private system. It involves issuing bonds covering the liabilities as of the day of conversion that will be paid by general taxes. So if the current projected liabilities are, say $4 trillion we simply issue to all the beneficiaries the appropriate amounts (based on age and what they have paid). Friedman argues that these bonds simply make explicit the implicit obligations which would also have to be paid from taxes so there is no new cost. But since the beneficiaries now have the bonds it has become privatized at no cost at all. Notice, and this is crucial, that Friedman is not assuming, nor is his argument dependent on, benefits from the conversion offsetting any costs. He simply says there are no costs.

Point 2- I argued that this is incorrect. To be more precise than I have been up to now it can only be (and even then only partially) correct in a very narrow (an ultimately useless) definition. It is correct only as far as the specific amounts covered by the bonds are concerned but not if you look at SS as a whole. Let’s use the $4 trillion example. Whether it’s privatized or not taxes will still have to be raised to cover those $4 T. In that sense one could say that a conversion as described by Friedman is cost free. But that is only a (very) small part of the story.

Point 2a- First, as Jim recognizes in his last post it may not be the same people who pay in both cases. It could be that some people pay less and others pay more (compared to what they would have paid if there had been no conversion). Jim argues that it will be the richer that will pay more since it will be done through an increase in income taxes. Let’s suppose that that is correct. Will the rich willingly accept this change? Or will they lobby against it? For them it’s a cost increase and they may not care that in aggregate nothing has changed. Is it as Jim says that there are more poor people and so it won’t matter politically? This sounds unlikely. If taxes were simply based on votes by the poor we wouldn’t have gotten rid of the estate tax nor would we be reducing top marginal rates. So right from the start there is a cost, as some people will have to pay more even if others have to pay less.

Ok, but what if we all paid exactly the same as before? This would defeat one of Friedman’s purposes (to get rid of the payroll tax) but at least one could argue that it really was cost free. We would all pay the same but we would have moved to a private system.

Point 2b- Here’s where the bigger issue comes in and the real transition costs show up. What Friedman conveniently forgets is that the $4 T is only a part of the overall SS obligations. $4 T is what we owe as of TODAY (remember I’m using $4 T as a proxy for the current liabilities). But that number changes everyday. The people that pay SS today do so not just because they want to pay off the liabilities of the past but also because they expect to get money in return in the future. When they pay money into SS they reduce SS liabilities (by paying current beneficiaries) but AT THE SAME TIME increase SS liabilities (by increasing their claims).

Please note that this is the CRUCIAL point and the one we keep coming back to. The transition costs come from all those people that have to pay for the $4 T but won’t get anything in return. You can’t say (as I understand Paul to be saying) that this is a cost that will have to be borne whether we make the transition or not. That is not true. It’s true that payments to cover the $4 T will have to be made one way or the other. But under the current system the people that paid the $4 T will get it back in the future as the next generation pays for their benefits. We can debate how much exactly they will get of those $4 T. What is not debatable is that they will get SOMETHING. But in Friedman’s model they get NOTHING. That difference are the transition costs. The transition costs are not the $4 T paid (which we have to pay in any case). The transition costs are the dropping of any future claims on those $4 T.

Friedman’s point only makes sense if you think that liabilities as of today are all that matter. But that is ridiculous. What matters are not just the current liabilities but also the expected ones. Friedman simply assumes that there are millions of voters willing to continue to pay and get nothing in return.

Under the current system taxpayers would pay the $4 T and get (roughly) that amount back. Under Friedman’s system taxpayers would pay the $4 T and get nothing back.

Point 3- Having said that there are some counter arguments to the transition costs. One is that there may be benefits (from greater returns) so that they outweigh the costs. This is not Friedman’s argument but may be a good argument. I haven’t seen any data.

The other counterargument is that the costs involved in fixing SS may be such that it makes sense to make the transition. Again, hard data would be useful.

Point 4- Finally, and as an aside, let me reiterate that the transition costs are not the same as the estimates of current unfunded liabilities. Those are two different numbers which would be paid by different people in different time periods and are the result of different causes. The unfunded liabilities, for example, don’t become an issue until 2044 according to the SS when many of today’s taxpayers will be dead or close to it.

GT writes:

Now, having said all that I wanted to address Jim’s and Paul’s points directly, even if it is a bit repetitious.

Paul says:

“But most people, I think, will find it seriously misleading to charge to the transition a cost which will have to be borne whether we make the transition or not. I'm certainly one of those people”

As I explained above if Paul is referring to the costs of paying today’s liabilities (the $4T example) he is correct. We have to pay that amount transition or no transition.

But that is not the cost I charged to the transition. That is not what I mean by transition costs. Under the current system taxpayers would pay the $4 T and get (roughly) that amount back. Under Friedman’s system taxpayers would pay the $4 T and get nothing back. That difference is the cost I am referring to. And that is certainly not a cost we have to incur whether there is a transition or not. It is a cost that only shows up with the transition.

Jim says:

“There is indeed a transaction cost imposed on the richer -- but this is exactly offset by a transaction *gain* enjoyed by the poorer.”

Same as above. Jim, like Paul, is only looking at the cost of paying the $4 T. But that’s not the transition costs I’m referring to.

Paul Zrimsek writes:

You guys can carry on if you like. My forehead is getting sore.

Jim Glass writes:

OK, here's my final shot at settling this, as wordy as Paul was brief -- freeing him to go bang his head against the wall at Brad DeLong's blog for a while. GT wrote...

"Friedman argues that these bonds simply make explicit the implicit obligations which would also have to be paid from taxes so there is no new cost. But since the beneficiaries now have the bonds it has become privatized at no cost at all."

Right, no additional cost -- the same cost as before. And now the crux of the matter:

"Here’s where the bigger issue comes in and the real transition costs show up ... The people that pay SS today do so not just because they want to pay off the liabilities of the past but also because they expect to get money in return in the future. When they pay money into SS they reduce SS liabilities (by paying current beneficiaries) but AT THE SAME TIME increase SS liabilities (by increasing their claims).

"Please note that this is the CRUCIAL point and the one we keep coming back to. The transition costs come from all those people that have to pay for the $4 T but won’t get anything in return. You can’t say (as I understand Paul to be saying) that this is a cost that will have to be borne whether we make the transition or not. That is not true. "

But it IS TRUE. IS! IS! IS! ;-)

"It’s true that payments to cover the $4 T will have to be made one way or the other. But under the current system the people that paid the $4 T will get it back in the future as the next generation pays for their benefits."

No they won't!! That's the point! Somebody in the future generation *must pay* an amount equal to the SS funding gap (I think its closer to $15T now) one way or another, in real time within that generation, and *not* get it back. Through payroll taxes, income taxes, higher admissions fees to the National Parks, some way. How can they not? The questions are "who?" and "how?"

Here's the simplest way to see it clearly, as per Friedman again (elsewhere):

Call SS's unfunded retirement liability as of TODAY, to today's workers and retirees ($4T, $15T, whatever) $X. Unless the gov't renegs, we are on the hook for it, no way out -- $X is a "sunk cost". The nation must pay it, though we have free choice as to how. But remember, though we must pay that $X in coming years, it provides $0, zip, nada, towards financing any retirement benefits earned as of tomorrow on.

Thus, starting TOMORROW, all additional retirement benefits must be separately financed. Call this amount $Y. So in future years workers must pay $X *plus* $Y to finance retirement benefits, but collect *only* $Y. They are going to be out $X! They *must* be, as long as the rules of arithmetic apply.

You seem to be thinking of SS as being some sort of perpetual rollover machine where today's young workers will at first be out $X, but later get it back from those who follow them. But it's not so! Because SS is actuarially unsound and insolvent -- the "make up" money does *not* arrive in the later years.

Remember, while the SS formula for their lifetime benefits/taxes is barely breakeven, that formula is 25% underfunded over the next 75 years. For today's 20-year olds that is until age 95, WITHIN their lifetimes, so they *don't* get it back. (And on a permanent actuarial basis things are even worse -- conditions are worst in year 75 and worsen every year out after that.)

That 25% SS funding gap is due to the cost of $X, of course. How to close it? Well over the next 75 years SS could increase payroll taxes by $X without increasing benefits, or reduce benefits by $X without increasing payroll taxes, or some combination of both. But, obviously, that leaves the next generation losing $X outright and *not* getting it back over those 75 years. It's just arithmetic, there's no way around it.

The other option, of course, is to use $X in general revenue to supplement SS payroll taxes over the next 75 years. Then SS benefits will be break-even relative to payroll taxes -- but income tax payers will be out the full $X while getting "no benefit" in return. SOMEBODY is going to be out $X for no benefit in return.

Krugman, Summers, Rubin & Co., want to go the "close the gap with general revenue to save SS" route. But Friedman points out this is in for a dime, in for a dollar. I.e., If it's better to fund the next marginal dollar of SS benefits with income tax revenue rather than payroll tax revenue, then why isn't it also true for the *previous* marginal dollar? All the way back to the first dollar?

He points out that the routine answer "because of the transition cost" is bogus --- there is NO TRANSITION COST, not in financial terms. There is only in political terms (to some people, yes indeed, a big one).

The real issue is *who* should pay that $X over the next 75 years and not get it back -- poorer payroll tax payers, as a result of trying to channel both $X and $Y into a constricted payroll tax (while also imposing a punitive tax on entry level jobs); or richer income tax payers, allowing poorer payroll tax payers to finance $Y with contributions half the size while getting healthy positive returns from them. Because SOMEBODY has to pay $X and not get it back.

Hey, maybe the hidden part of Bush's plan for privatizing SS is to sell off Iraqi oil for the Social Security Administration ....

GT writes:

Jim,

We may be getting closer to an agreement. Using your notation I propose the following definitions:

$X= Liabilities as of conversion date (what I called $4T in my previous post). It is the money that we would have to pay if all new claims to SS were stopped beginning today and only claims already in place were honored. It is NOT the same thing as SS’s unfunded liabilities. In fact if payroll taxes (or their equivalent) remain unchanged $X is most likely fully funded.


$A= What the taxpayers who paid $X expect to get in return JUST for paying $X. THESE are the transition costs.

Please note that my definition of $X is, I think, a bit different than yours. You define $X as the “SS's unfunded retirement liability as of TODAY, to today's workers and retirees”. I’m not sure what you mean by unfunded liability. Do you refer to the expected gap between collections and payments in the SS in the next 75 years (or whatever timeframe you have in mind)? If I read your definition correctly that’s what you mean. If so I use a different definition. I only look at the TOTAL liabilities as of today if we stopped the SS system today so that any new payments would ONLY be used to cover the liabilities as of the conversion date.

Since we are not face to face, which would make it easier for me to explain myself, let me reiterate hoping to be clear. SS’s unfunded liability is normally understood to be the gap between collections and expenditures. It does not become a problem until 2044 (meaning that for most people over 40 it’s basically irrelevant, maybe even for those over 35). This doesn’t mean it shouldn’t be addressed btw.

$X on the other hand is not unfunded. In fact right now it’s running a surplus (and will continue for many years to do so). $X refers to the amount of money we owe TODAY if as of TODAY there are no more claims on SS. Notice the HUGE difference with what is commonly talked of as the unfunded liabilities. $X is different in time, size and impact from the unfunded liabilities. I make this point because you seem to be confusing both concepts a bit, as if the unfunded liabilities of SS are the same thing as the current liabilities as of today.


In fact let’s add another definition.

$U= SS’s unfunded liabilities, calculated as the difference between what SS will receive in payroll taxes in the next 75 years minus what it will have to pay out. The SS Administration has calculated that the liabilities don’t start showing up until 2044.

OK, we have the definitions.

We both agree that $X is a cost no matter what either under Friedman’s plan or under the current SS.

Where we disagree is on what $A means. There are two important points.

The first, you seem to imply that $A is zero and here’s where I can’t fully follow your argument.

You say:

“You seem to be thinking of SS as being some sort of perpetual rollover machine where today's young workers will at first be out $X, but later get it back from those who follow them."

Exactly. That is how the system works. Workers pay today expecting future workers (or younger current workers) will pay for them in the future and so on. I am fully aware that what future workers are supposed to pay will not be enough, beginning in 2044. But that doesn’t mean it will be zero.

So I don’t see where your next sentence comes from:

“But it's not so! Because SS is actuarially unsound and insolvent -- the "make up" money does *not* arrive in the later years.”

If by this you mean that after 2044 it doesn’t arrive 100% I agree. But substantial amounts will be still be paid after 2044 and 100% is covered before 2044.

But it’s the second point that’s more important. Notice my definition of $A. It’s what workers EXPECT to be paid JUST for paying $X. We can debate how much that is and whether they realistically expect to get all the principal plus a return, and whether it varies by age, and so on. But certainly they expect to get SOMETHING. Or do you think that the mass of SS taxpayers expects to get nothing at all, not a single dime? Whatever that something is, whatever amount you imagine THAT is the transition cost.

Suppose that $X is $4 Trillion. Now suppose that workers expect to only get back 75% of what they put in, implying a -25% return! That still means that $A, the transition cost, is $3 T which is about 30% of GDP. That’s huge.

Friedman is telling workers, pay $X but $A will be zero! Using my example above Friedman expects workers to simply forgive $3 T.

Hopefully I’ve made it clearer this time.

Finally there are several good arguments to convert to a private system. Using the notation, we can restate some of the points made earlier in this thread. Paul says that the return workers would get under a private system could be similar to, or even greater than $A since they would be getting higher returns in the portion they can now set aside for themselves.

And I make the point that maybe $U is greater than $A so that it makes sense to change in any case.

But notice that a) neither argument helps Friedman’s point since his article is not based on either one and b) even accepting my or Paul’s arguments we still have the issue of changes in who pays. After all, even if I am correct and the eventual costs of fixing SS ($U) are greater than the transition costs today ($A) that still leaves us with the political problem that $U won’t be paid until 2044 (when most current taxpayers will be dead or retired) while $A would have to be paid today.

Jim Glass writes:

" You define $X as the “SS's unfunded retirement liability as of TODAY, to today's workers and retirees”. ... I only look at the TOTAL liabilities as of today if we stopped the SS system today so that any new payments would ONLY be used to cover the liabilities as of the conversion date."

Fine, same thing, those are the future taxes that will have to be collected to pay SS's liabilities as of today if the program is ended tomorrow. And even if it isn't!

"$A= What the taxpayers who paid $X expect to get in return JUST for paying $X."

Um, no ... Nobody expects to get *anything* for paying $X. They expect a promised return for paying a specified amount of payroll taxes, as per the SS benefit formula. I believe this is important because there are two ways $X can be paid ...

1) Through income taxes, as per Friedman, about $11 trillion worth of $X (I looked it up) in which case the income tax payers get nothing in return, and expect nothing because they pay no payroll taxes.

2) Through payroll taxes under the current SS program, in which case the benefits SS participants receive over the next 75 years will be *LESS* than they expect from the promised SS funding formula by $11 trillion.

The two $11 trillion costs are not coincidence, obviously. SS is a "pay as you go" system with basically all payroll taxes collected used to pay benefits at the same time, so taxes and benefits must match over any given time period. *Except* if you extract $X to make a transfer to cover benefits credited to a prior period. Then taxes must *exceed* benefits for the current period -- you've made a transfer after all!

To quote the Treasury's _ Analytical Perspectives of the 2004 Budget_: "The current deficiency in Social Security is essentially due to the excess benefits paid to past and current participants"

That's the $11 trillion current deficiency.

I mean, you can't promise to finance $100 of benefits with $100 of taxes pay as you go, then transfer $25 of the taxes to third parties, and then pay the promised $100 with the $75 you have left -- which is functionally what is going to happen to SS relative to its promised benefits over the next 75 years, on current law.

SOMEBODY is going to be out $11 trillion in the next 75 years. Either income tax payers out of pocket, or payroll tax payers out compared to what's been promised to them by law. You don't seem eager to grasp that.

" THESE are the transition costs."

??? A transition cost is incurred *only* if a transition occurs. The $11 trillion is going to be incurred by somebody WHATEVER happens. C'mon, you can't seriously call something a transition cost when you have to pay it even if there is no transition. ;-)

Look, I don't know how to make it any more clear than this: There are only two types of retirement benefit costs -- $X, the cost of all retirement benefits earned under SS to date, which *must* be paid and is a *fixed* sunk cost; and $Y which is the cost of all retirement benefits to be earned in the future and can be whatever we decide.

There is no "$T", transition cost, which must be paid *only* if SS is terminated in a transition to something else. If you think there is, point out who must pay it to whom.

"You say:
"“You seem to be thinking of SS as being some sort of perpetual rollover machine where today's young workers will at first be out $X, but later get it back from those who follow them.""

"Exactly. That is how the system works... So I don’t see where your next sentence comes from:

"“But it's not so! Because SS is actuarially unsound and insolvent -- the "make up" money does *not* arrive in the later years.”"

Well, when the rollover doesn't arrive... ;-( Your A$ -- SS's promised return on payroll taxes -- is underfunded by $11 trillion over the next 75 years. The rollover machine is busted. Today's young workers are *not* going to get the $A that was promised to them via SS payroll taxes. [BTW, you do realize that saying "SS is fully funded until 2044" is saying to today's young workers "SS is fully funded until just about when you retire". I'm not sure that's a winning political slogan. ;-) ]

"But it’s the second point that’s more important. Notice my definition of $A. It’s what workers EXPECT to be paid JUST for paying $X. We can debate how much that is and whether they realistically expect to get all the principal plus a return, and whether it varies by age, and so on. But certainly they expect to get SOMETHING. Or do you think that the mass of SS taxpayers expects to get nothing at all, not a single dime?"

I think you've got your argument backward here. If you think it is so important to workers that they the get the $A that's been promised to them for paying all those payroll taxes, then I'd think you'd believe they'd be rather angry at getting $11 trillion *less* than that -- which is what they are going to get, no way around it if SS stays funded with payroll taxes, as a matter of the simple arithmetic I pointed out before.

SOMEBODY has to pay that $11 trillion -- payroll tax payers through higher taxes/reduced benefits compared to the "expected, promised $A" from SS, or income tax payers in exchange for nothing. Note that if income tax payers do pick up the $11 trillion, then payroll tax payers can get a rate of return much *larger than A$* on their retirement savings in the future. [So maybe income tax payers do get something -- good feelings about helping the less fortunate. ;-)]

THIS is the key to the whole discussion. I asked you before, but you seem to have ducked the issue. Who do you pick to pay the $11 trillion?

"Friedman is telling workers, pay $X but $A will be zero! Using my example above Friedman expects workers to simply forgive $3 T."

Well *you* are expecting workers to forgive $11T under the current system! That's a lot more than $3T.

Friedman is promising that all those who have paid payroll taxes will get 100% of the benefits promised to them. And he is saying that the $11 trillion shortfall in SS that must be paid by SOMEBODY should be picked up by richer income tax payers instead of poorer payroll tax payers, as a matter of equity and sound economics.

Who do you choose?

GT writes:

Jim,

It seems to me there is some confusion of concepts and a lot of arguments and side arguments that make it difficult to follow what we are saying to each other. That’s why I numbered my points previously, to make it easier to keep a tab on what we were saying. I think we are both a bit frustrated with the other in that we think (well, I know I do) that the other is trying to engage in real debate but somehow keeps missing the point.

My main point all along has been that there are transition costs in terminating SS as Friedman suggests. I have acknowledged that those costs may be smaller than the alternative of ‘fixing’ SS so it may make sense to pay the transition costs anyway. But what I don’t accept, what I think is wrong, is the idea that there are no transition costs which is what you, Friedman, and several other posters here argue.

Now I have been always referring to monetary costs, costs that can be measured in dollars that have to be either paid by someone or forgiven by someone else. If I owe you $100 the cost can be paid by me, by paying you the $100, or by you, by forgiving the $100, or a combination of both. But there are also huge political costs which I have , mostly, not dwelled on. If two people owed you $50 each, and you decided to charge one of them the full $100 even though there was no change in the total amounts there would be strong resistance from the guy who now has to pay double. This example is relevant to your discussion as to who pays what.

______________________--

I will try to focus on two points, which I think clearly delineate the differences in our views.

The first point is a confusion between $X (liabilities as of today) and $U (unfunded liabilities of SS in the next 75 years). This point is less crucial than the second but, if I understand what you have been arguing, seems to explain in part why we can’t agree on the second point.

The second point is my main one all along, the origins and the size of the transition costs. You asked me who pays for this cost. I’ll try to answer below.

___________--------

On the first point. You claim that $X and $U are both about $11 T. You also say that “The two $11 trillion costs are not coincidence, obviously.” This is a big point. If I read you correctly you are saying that $X and $U are related, maybe even the same thing.

But they aren’t. They are two separate, even unrelated concepts. An example will make this clear. Suppose that SS was perfectly balanced and expected to be perfectly balanced forever. In this example the demographics don’t change and the beneficiary-worker ratio never changes. What would $U be? Well, zero! There are no unfunded liabilities so $U is zero. What is $X? $X remains the same, $11 T or whatever the number is.

$X is the result of ONE number, the amount of money paid in the past. $U on the other hand is the difference between TWO numbers. The first number is the amount of benefits that will be paid in the future based on current law. It includes, at least in part, what I call $X but it also includes the new claims on SS that are added as time goes by. The second number is the expected SS revenues. The difference is what we call the unfunded liabilities or $U.

This is important. You say “The $11 trillion is going to be incurred by somebody WHATEVER happens. C'mon, you can't seriously call something a transition cost when you have to pay it even if there is no transition.”

You are confusing the two separate concepts. They are not the same thing. If there is no transition the only way there is a cost is if SS has a deficit. That's the ONLY way. But if there is a transition there is a cost no matter what, even if SS has no deficit.

So let’s stop bringing up the unfunded liabilities as if they are relevant to the main issue at hand, whether there are transition costs. They aren’t. Let’s be clear. Whether SS has unfunded liabilities or is in balance or even whether we project it will be in surplus has no bearing whatsoever on Friedman’s point. Friedman claims there is a costless way of basically unwinding SS and that is wrong.

______________________-----

So now to the transition costs. You asked me who had to pay for it and to whom. The costs must be paid by all those taxpayers that will pay the $11T under Friedman’s plan but will not get anything in return. In other words it’s a cost not of paying money but of giving up a claim (which is the same thing) similar to my example of the $100 dollars. If you forgive me a $100 debt it’s the same thing as if you just gave me $100. The taxpayers that will pay the $11 T and get nothing in return are the ones that must pay the cost. We can debate who, exactly, would pay that cost.

Your counter argument has been that why would they get anything in return. And here’s where you lose me a little. As I understand it you have two different arguments as to why paying the $11T in taxes needed to cover $X should not result in any claims for the tax payers.

First you say that SS is broke so they wouldn’t get the money back in any case. This, as I have recognized, is correct if you mean they won’t get 100% of it back. But it is incorrect if you mean they won’t get anything back. According to your numbers SS will have enough in the next 75 years to cover about 75% of the expected claims. If so that means that even if we ONLY include as expected claims the money that will actually be there then those paying the $11 T should normally expect AT LEAST $7.75T back (btw, I’m not sure how the calculations are done and whether the 75% return is comparing what individuals get back vs what individuals and companies pay in). Where does this come from? From future workers who, in turn will have new claims which will in turn be paid by future workers and so on. That’s how SS works. The fact that is broke, that it can’t cover 100% of expected claims does not mean it can’t cover anything at all.

It makes no sense whatsoever to say that those that pay into SS can expect nothing in return. They can and they do. Even in the worst case scenario, that no changes are made, they will get 75% of what they put in back. That’s more than the 0% they get under Friedman’s plan.

As for looking beyond 75 years that’s completely useless. We have no way of knowing what will happen demographically 75 years from now. Imagine if someone in 1900 had tried to predict the demographics of 1975. To be able to predict the Baby Boomers, for example, he would have had to know about the Great Depression and World War 2. For all we know in 75 years we will have merged with Mexico and Canada and our demographic profile will be completely different. Even looking at the next 75 years is a stretch.

The second argument you offered explaining why those that pay the taxes should not expect anything is that under Friedman’s plan the current liabilities would be paid out of general taxes (maybe income taxes, maybe something else) and when taxpayers pay those taxes they never expect a specific service in return. But this assumes that taxpayers are complete idiots and won’t realize that, although their SS taxes have disappeared other taxes have increased by the same amount so that they continue to pay the same total but now they have lost any claims on SS and all its services (which includes much more than retirement, it also includes insurance for widows and orphans as well as disability insurance) services which if they want they will now have to pay for on their own. And if SS taxes are replaced by income taxes millions of tax payers not only will lose any new SS benefits but they will have to pay MORE than before. You really think voters will accept this?

Let’s look at the numbers. SS taxes collect about $600 billion a year. By paying this $600 billion taxpayers expect many services and the numbers show that there will be enough to at least return them about $450 billion (or maybe $225 B depending on how the numbers are calculated, since the companies that pay half of the SS never get anything in any case). But now we implement the Friedman plan. Overnight SS taxes fall by $600 billion which means that income taxes or some other tax must rise by that same amount. (Since no new claims are being added that income tax surcharge will be able to fall a little bit every year but will be substantial for many years). Can you imagine a $600 billion tax increase in JUST ONE YEAR?

Remember that some people will end up paying more than they did before, even if they don’t pay SS anymore. The only way for nobody to pay more would be for everybody to pay the exact amount as before. In other words we would change the name of the payroll tax to income tax surcharge (or whatever) but it would be the same thing.

I’ll finish with this.

You wrote:

“There are only two types of retirement benefit costs -- $X, the cost of all retirement benefits earned under SS to date, which *must* be paid and is a *fixed* sunk cost; and $Y which is the cost of all retirement benefits to be earned in the future and can be whatever we decide.
There is no "$T", transition cost, which must be paid *only* if SS is terminated in a transition to something else”

Your mistake is that you forget, or leave out, that under the current SS plan the tax dollars that pay for $X CREATE new claims that will be paid by future generations (and we know there will be enough to pay back at least 75% of that back in the future). In Friedman’s plan the tax dollars that pay for $X create no claims whatsoever. That’s why those transition costs ONLY show up if SS is terminated. If SS continues those that pay $X will get 75% back. But if they terminate SS they get 0% back. That difference is the transition cost, a cost to those that pay $X.

It is true that on an intergenerational basis Friedman’s plan has no cost since the money that those that pay $X give up (by not making any new claims) is money that other taxpayers won’t have to pay. But that’s simply because the $X taxpayers have, in effect, forgiven moneys that would be due them. That’s the cost.

Ral;ph King writes:

I retired 2 years ago, and get a social security retirement chech each month, and thats fine. The problem is I can't work and make but so much money of earned wages, or I start loosing benefits. The real pain is that those limited wages I can make, I still have to pay Social Security taxes on. So I get a check each month from them, and they get a chech each month on my wages earned. It doesn't compute to me to be fair. I retired at age 62.

Mark Licholat writes:

I am looking for a clear defination on the greek term "ceteribus paribus". It is utilized in many economic research papers and is ofter refered to.

Does this translate into a clear english defination?

Hi, Mark.

You asked:

>I am looking for a clear defination on the greek term "ceteribus paribus". It is utilized in many economic research papers and is ofter refered to.
>
>Does this translate into a clear english defination?

"Ceteris paribus" is translated as "other things equal". If in English you subsititute the phrase "holding other things constant," you will understand it right every time.

For example, if someone says "Increasing the money supply causes inflation, ceteris paribus," he means that if you increase the money supply but hold other things (say, annual output, production, income, employment, or even smaller factors like banking regulations or money definitions, or average frequency of bank trips for getting cash) unchanged, then inflation will occur.

The speaker is saying that those other things change less frequently or are less likely to change under the circumstances you and he agree you are discussing, so for simplicity, he is going to set aside all those possible objections for the moment, however reasonable they may be in special circumstances worth considering next.

Usually, when a trained economist uses the phrase "ceteris paribus," which "other things" are to be held constant are clear to educated readers after a moment's thought. But sometimes when the term is used, it accidentally brushes aside some important objections. Sometimes holding some particular thing constant goes against the grain of the original question, so if you've given it some thought and still don't know which things are being held constant, it's fair to ask!

The Latin/Greek roots of the term stem from a few sources, but these familiar current usages may help: "Par" may be familiar from common current usage in golf: it evokes the concept of a norm, an average, or the expectation. The root "ceteris" may be familiar from "et cetera" (or "etc.," meaning 'and any other related things.' "Et" in Latin means 'and'.)"

In economics, the term "ceteris paribus" is usually pronounced with a hard "c" and a long next "a": kay'tur-ees pah'ree-boos. It roughly means "let's focus on changing this one item and hold other possible items unchanged, with the goal of pinning down the marginal contributions attributable to what happens when just this one item changes."

Lauren

Al Hleileh writes:

Friedman's proposals are realistic: In answering critics who have warned that making the transition to a privatized Social Security system would be too burdensome on today’s young workers, Nobel Prize winner Milton Friedman says, “Given a proper understanding of Social Security’s current unfunded liabilities—variously estimated at from $4 trillion to $11 trillion—there are no real transition costs to privatizing Social Security, merely the explicit recognition of current implicit debt.” In “Speaking the Truth about Social Security Reform”, Friedman explains that one of the myths underlying Social Security is that it is a form of social insurance equivalent to private insurance. Friedman points out that the reality is that taxes paid by today’s workers are used to pay today’s retirees. “If money is left over, it finances other government spending—though, to maintain the insurance fiction, paper entries are created in a ‘trust fund’ that is simultaneously an asset and a liability of the government.” Friedman argues that a privatized Social Security system should not be mandatory. “It makes no more sense to specify a minimum fraction [of income that all people must save for retirement] than to mandate a minimum fraction of income that must be spent on housing or transportation. Our general presumption is that individuals can best judge for themselves how to use their resources.” This seems to be the efficient and realistic system, and critics should try and come up with a better alternative ?

Comments for this entry have been closed
Return to top