Arnold Kling  

Social Security Reform

PRINT
Unstable Information Economy... Cost of Sarbanes-Oxley...

Michael Tanner of the Cato Institute has a proposal for weaning young workers off of the Social Security system.


workers who chose the individual account option would receive a bond in recognition of their past contributions to Social Security. That bond would be a zero-coupon bond calculated to provide a benefit based on accrued benefits under the current Social Security system as of the date that the individual chose an individual account.

The so-called "transition cost" of privatization would be handled in part through continuing to collect the employer portion of payroll taxes, regardless of whether employees opt out of Social Security. This seems really onerous for today's workers, but that reflects how out of balance the system is currently.

I think that the political problem that all privatization plans run into is that they force recognition of the huge off-balance-sheet liability of the government that consists of unfunded promised future benefits. The Cato proposal faces this issue squarely. If anything, Tanner's paper bends over backwards to give an honest accounting.

For Discussion. If you were forty years old, would you elect to take a "recognition bond" in lieu of all future Social Security benefits in exchange for the right to take your 6.2 percent in employee payroll taxes and invest them yourself?


Comments and Sharing


CATEGORIES: Social Security



COMMENTS (30 to date)
Cap'n Arbyte writes:

Young workers like myself don't need to be weaned off. I'm one of those people who would jump at the chance to totally opt out and would let the government keep every cent I've paid in so far.

What irritates me about this plan is that it's still a forced savings plan. I want to be individually responsible for my own retirement. I want to save at a time and a pace and in assets of my own choosing. No safety net. I don't want the government-approved mutual fund -- I want my money!

Joe Kristan writes:

Hmm. A 4% rate of return for 25 years, if you are earning the FICA base (I used $90,000), gets you a fund of about $237,000 at age 65. Where do I sign up?

Eric Krieg writes:

Forget the bond, forget everything I paid into Socialist Insecurity. Keep it. Just let me out of the program, and drop my portion of the payroll tax. Just let me increase my 401(k) by a similar amount.

I feel the same way about my pension, by the way. My company is one of the few left with a defined benefit pension. I think it would be better for them to close it out, distribute the funds to our 401(k)s, and greatly increase the company match to the 401(k).

Mcwop writes:

I am pushing 40 and I like the idea a lot. My feeling is that the risks associated with a private account are far less than the risks facing the social security system, and risks emanating from the politicians that control it (current and future).

Boonton writes:
I am pushing 40 and I like the idea a lot. My feeling is that the risks associated with a private account are far less than the risks facing the social security system, and risks emanating from the politicians that control it (current and future).

The risks associated with SSI are minimial. As long as there is an economy producing goods and services in the future, SSI can pay benefits since it is funded off of taxing the current economy. Therefore your only risk is that there won't be an economy in the future. (Why would that be? Nuclear war, astroid impact I suppose?). With a private account you run that same risk plus the risk that your personal investments will falter.

By definition private accounts must be riskier than SSI.

Lawrance George Lux writes:
If you were forty years old, would you elect to take a "recognition bond" in lieu of all future Social Security benefits in exchange for the right to take your 6.2 percent in employee payroll taxes and invest them yourself?

Yes, because of the immediate evasion of tax. This done knowing the SS issue is a national issue affecting all, and that any promise foregoing future SS benefits will be rescinded and negated at future date, when the real elements of viable security appear.

Most here buy into the concept of Private Accounts. They fail to perceive the true nature of all financial instruments--up to and including Stocks. All are a differentiated form of Money. They simply reflect a repayment feature for the process of Production, and are constrained by hard Costs which cannot be eliminated. Expression may be best by example: The famous words of a Doctor of the future:

"Of course that Procedure cost only 2.4 Shares of GM stock in 2004, but Today it costs 28.6 Shares of GM stock. It is not like expenses have not risen!"

Attempts to evade the Costs of supporting Seniors will fail. It is a question of the profitability and total amount of Production achieved. Cato insists present investment will vastly expand future capacity to support Seniors. It is not viable without Seniors providing Productive labor at advanced Age, or younger Workers becoming vastly more productive--i.e., by a factor of 3-5. Otherwise, that Procedure will cost 147 Shares of GM stock. lgl

Mcwop writes:

Boonton,
SSI does have significant risks. The economy is not the only factor. The general federal budget is a big factor because that competes for taxes with outlays like SSI. Politicians may cut SSI benefits, or hike the age to an unreasonable level to fit general budget needs or fund new programs or cover shortfalls in non-SSI programs.

Definitions do not make risk. Both private accounts, and SSI have risks. Albeit some different ones, and some similar ones. In my view the risks with SSI are higher than a private account. Heck, if the US economy looks that bad, and China looks really good – I could allocate assets there offsetting the risks of a faltering U.S. economy. If I had a private account option, I would allocate assets internationally (including international bonds)to deal with the risks of the US economy declining relative to others. What guarantee do I have that the Gov will pay me any benefits at all? I have none, just like I have no guarentee with a private account.

Social security risks:
-Out of control government spending
-SSI surpluses have already been spent, creating a future funding issue
-Political risk, politicians decide to not pay benefits or cut benefits for some reason. Maybe they decide that only people without (or that have low savings) private savings deserve benefits.
-SSI is mostly tied to the U.S. economy’s ability to produce goods and services in the future, and it is difficult to diversify this risk to a manageable level
-The return on benefits can continue to decline relative to contributions

Mcwop writes:

Almost forgot the Demographic risks, which are very high right now.

Boonton writes:
SSI does have significant risks. The economy is not the only factor. The general federal budget is a big factor because that competes for taxes with outlays like SSI. Politicians may cut SSI benefits, or hike the age to an unreasonable level to fit general budget needs or fund new programs or cover shortfalls in non-SSI programs.

You can't escape the risk of gov't policy with 401K style accounts. Gov't can:

1. Impose capital controls, thereby making it hard to invest in overseas places like China.

2. Revoke or modify the tax favored status of such accounts thereby disrupting your expected returns.

3. Cause economic havoc by reckless deficit spending, inflation, bad policy etc. which will ruin your 401K investments.

Gov't policy risk, therefore, exists in both accounts but private accounts add market risk. In other words, you can end up poor even if the next 50 years is as rosy as the 90's were.

We've discussed this before abot gov't spending, I've seen no evidence that increased revenues causes more gov't spending. In fact, quite the opposite seems to happen. When gov't revenues go up the Congress stalemates on how to spend it and spending growth slows (the 90's)...when revenues go down spending seems to increase because it is perceived as an 'emergancy'.

Mcwop writes:

Of course you can't escape the risks if rules such as those are put in place. That still does not make SSI low risk, or change the facts that the program is not flexible, and has upcoming funding issues that must be dealt with.

Boonton writes:

blockquote>Almost forgot the Demographic risks, which are very high right now.

Demographic risks are also inescapable in a 401K world. A society that increases the amount of retired people will have to either increase the burden on workers or decrease the rewards of retirement.

Of course you can't escape the risks if rules such as those are put in place. That still does not make SSI low risk, or change the facts that the program is not flexible, and has upcoming funding issues that must be dealt with.

Upcoming funding issues? Probably, although the forecasts of SSI's 'failure' are highly sensitive. Raise the retirement age slightly, have slightly higher economic growth and suddenly the program is quite solvent. Name anything else that you can prove will be solvent for the next 50 or 75 years? Microsoft, IBM, Disney?

My point, though, is not that SSI is riskless. It is that a private system is riskier. SSI's risk:

1. Entire US economy is destroyed.
2. Bad gov't policy.

Private System's risk:

1. Entire US economy is destroyed.
2. Bad gov't policy.
3. Bad individual investments.


You want to assert the private system is less risky by assuming away #1 and #2. Why would that be a reasonable assumption? If the US economy implodes is it really rational to assume the gov't, in desparation, may not go after some of the great returns you made by investing in developing nations?

Mcwop writes:

So, because you listed only 2 risks for SSI and 3 for private accounts that makes SSI less risky. Yes, this amp goes to eleven.

Risk can be assessed systematically and at the individual or family level.

A private system can be more flexible at the individual level, and have less risk at that level. What if my life expectancy is lower because of my genes than the averages? What if my goal is to retire earlier, or scale back on work?

Boonton writes:
A private system can be more flexible at the individual level, and have less risk at that level. What if my life expectancy is lower because of my genes than the averages? What if my goal is to retire earlier, or scale back on work?

Define risk here. Your life expectancy is not changed by having a 401K or SSI. If you die the day after you retire then you lose the utility of enjoying the bulk of either your 401K or your SSI checks.

I agree that social security lacks flexibility. Flexibility is not the same as risk, though. I wouldn't abolish 401K's in favor of expanding SSI. I think both should continue to co-exist with SSI providing a base of income regardless of what shannigans you get yourself into with your 401K account.

All public benefits suffer from a lack of flexibility. Taxing property owners for a fire department, for example, hurts the offbeat guy who would rather live in a fireproof cave and benefits the guy whose house is a firetrap since he fills it with old newspapers.

SSI's primary purpose is to guarantee a base retirement for those who are too old to work for themselves. This is necessary because, quite simply, people refuse to be rational about their final years and too often by the time they realize their mistake it is too late. Society made a collective choice in the Great Depression to guarantee that such people will not be left to starve on the streets.

If you accept that premise then you've basically accepted SSI. What is left is a very large (and oddly unspoken) debate about how much SSI we should have. Demographics has cause the SSI benefit to dramatically expand so the question should really be how much of this expansion can be paid for by economic growth and higher taxes on working people and how much of the expansion should be cut off. While a lot of privatization schemes are floated there's very little real investigation of ideas to balance SSI more towards the workers rather than the eldery. Ideas such as:

1. Indexing the retirement age to life expectancy.

2. Creating incentives to encourage people to remain to retire later (thereby expanding the labor force).

3. Greater means testing of SSI for the well off.

Yes that would lower SSI's 'return' but it would also lower SSI's cost thereby making it easier for workers to fund it. Demographics are destiny, if society wants to increase those who aren't working and decrease the portion who are...either the lifestyle of the non-workers will suffer or the lifestyle of the workers will.

Eric Krieg writes:

I would agree with you Boonton if Socialist Insecurity were less generous. Back before Nixon, it probably was a system of last resort, a base income to keep seniors off of dog food. But then Nixon and Congress started to play political football with the program, increasing payments as a way of vote buying.

And then Congress indexed the program to inflation, embedding in stone those gold plated SS payments.

So maybe if we rolled back SS payments to, say, 1968 levels, I would be on board with you. But I don't think there is a snowball's chance in hell of that happening, and I much prefer privatization.

Boonton writes:
So maybe if we rolled back SS payments to, say, 1968 levels, I would be on board with you. But I don't think there is a snowball's chance in hell of that happening, and I much prefer privatization.

Your very review of the history of SSI shows how your assertion is false. Retirement age was raised in the 80's. Benefits were indirectly cut under Clinton (by removing a portion of the tax exemption I believe) and also indirectly cut under Reagan.

So there's no reason there couldn't be rational benefits reform that is phased in gradually. If there isn't will for that there won't be will to scrap the whole system.

Eric Krieg writes:

Boonton, I am not talking about raising the retirement age by 2 years over a span of 20 years. That is a trivial change.

Benefits were raised by over 1/3 during the Nixon years. Do you think that any politician in America would even think of cutting benefits by 1/3?

BTW, part of the Clinton increases in the amount of SS income subject to income taxation was rolled back by the Republican Congress. It was a point in the Contract with America.

Boonton writes:

I don't think a 1/3 cut would be necessary. The fact is as society becomes richer we can afford more luxuries...including a retirement that's a notch above eating cat food. IMO, a gradual rise in the retirement age coupled with indexing that to lifespan & maybe a little means testing would be sufficient to put the system in balance.

I'm sorry to hear that the GOP did that. I guess you can add it to the list of Acts of War they have committed against the nations long term fiscal health.

Eric Krieg writes:

There were (are?) supply side reasons that they did it. The argument was that by taxing SS so heavily (85% vs. 50% now), you made it economically unattractive for SS recipients to work.

When you think about the time frame, right before the Internet bubble, it probably was a good thing. It kept the boom going a little longer than it might otherwise have (for a lack of workers).

Of course, right now, it would probably be a good thing to have those benefits taxes at 85%. Retirees exit the workforce, and others take their place.

Mcwop writes:

Eric,
We have to take it like men. Basically, the "reform" will be phased in gradually until 50+% of the recipients get little in return for the money paid in. 2% payroll tax when it all began 12.4% now, and probably 17+% in the near future. The best part is that politicians keep their budget slush fund.

Eric Krieg writes:

Okay, Boonton, let's index the retirement age to the average life expectancy.

Let's see. When Socialist Insecurity was started, the average life expectancy was 61. Socialist Insecurity kicked in at 65. The percentage difference is 6.5%.

What's the average life expectancy today? 85? To make the retirement age equal on a percentage basis, the retirement age would be 90-1/2!

I daresay that that would solve our Social Security problem. But man, I would need some serious Prozac if I had to wait until 91 to retire!

David Thomson writes:

“But man, I would need some serious Prozac if I had to wait until 91 to retire!”

But do you really want to ever retire? I am sure that you may very well wish to do something else---but I doubt if you will ever play shuffle board and constantly drink beer on a daily basis. No, I’ve sized you up as someone who will remain busy until their health gives out. Bernard Lewis remains intellectually active at the age of about 87! You will almost certainly do likewise. The retirement crisis is really about blue collar individuals who are economically of little value when they reach their “golden years.”

Mark writes:

Would I opt out of Social Security? Not unless I saw a convincing case that the crisis is genuine. To be specific, I would want to see rigorous demonstrations that:

1) The economic projections that are the entire basis for claiming that "SS is going to go bankrupt" are sound, and not, as they appear to this economist, incredibly conservative. I want to see a reasonable basis for assuming that real GDP growth will average 1.7% annually over the next several decades, and that there will be virtually no labor force growth over a period in which we will see a population increase of over 100 million.

2) That government bonds will somehow become worthless between now and 2018, which is implicit in Tanner's claim that the true "day of reckoning" is then, and not in 2029--oops, it's now 2042, while the SSA was saying the trust fund would run out in 2029 only a few years ago. Isn' it interesting what a few years of growth exceeding the Trustees forecasts will do?

3) That it is possible for investors to escape not just firm-specific risk, but also market risk, so that there will not be any risk to private-accounters from a stock market crash.


Until I see some reasonable basis for accepting the Cato crowd's claims about the impending doom of SS and the vast superiority of private accounts, I will continue to call it like I see it right now--the move for privatization is fuelled by a mix of anti-government dogmatism and Wall Street greed.

Boonton writes:
Okay, Boonton, let's index the retirement age to the average life expectancy. Let's see. When Socialist Insecurity was started, the average life expectancy was 61. Socialist Insecurity kicked in at 65. The percentage difference is 6.5%.

Must you be so dramatic? No one said the index had to baseline lifespan 50 years ago. SSI's issues are very long term so solutions should be phased in very long term. Doing so modestly will both ease the burden of the retiring baby boomers on their children without throwing people to the wolves.

3) That it is possible for investors to escape not just firm-specific risk, but also market risk, so that there will not be any risk to private-accounters from a stock market crash.

SSI's funding depends upon the performance of the economy as a whole. It isn't quite the same as a tax on GDP but it is pretty close since it higher GDP usually gets directly filtered into higher payrolls. That's more diversification away of firm specific risk than you can achieve with any index fund.

Jim Glass writes:

"The risks associated with SSI are minimal".

Well, if we define "risk" as the probability of young workers receiving a lower return on SS contributions than from a very safe market portfolio -- such as one based on gov't bonds -- the SSA actuaries say this minimal risk is a "sure thing".

The formula benefit for young workers provides only a 1% to 2% return on average, with negative returns guaranteed for a great many, say the Actuaries.

And that formula return is 25% underfunded. The *only* way to close such a funding gap in a paygo system is to reduce benefits or increase taxes -- both of which make the 25% further decline on return a *reality*, pushing return well down into negative territory, and visibly so for everyone to see.

Which is why the politicians refuse to take the simple steps needed to close the "funding gap" as long as SS is cash flow positive so they don't really have to. So the gap grows larger every year, making more dramatic steps inevitable in the future.

"Therefore your only risk is that there won't be an economy in the future. (Why would that be? Nuclear war, astroid impact I suppose?). With a private account you run that same risk plus the risk that your personal investments will falter."

Let's see, a government account guarantees negative return to me. Private investment over 30 years offers extremely remote chance of negative return -- it's never happened yet. It takes a government to *guarantee* negative returns!

So yes, if the world ends, I will lose more by not living to collect from a private account. Assuming Armageddon, private accounts are more risky!

"By definition private accounts must be riskier than SSI."

The definition being: 'sure loss on investment is less risky than remote chance of loss'.

"...there's no reason there couldn't be rational benefits reform that is phased in gradually. "

True, aside from the powerful reasons of politics, of course.

And also noting that "gradually" means closing the discounted to current value 25% funding gap today. To the extent "gradual" reform keeps this from being done, the gap becomes 26%, 27%...

"If there isn't will for that there won't be will to scrap the whole system"

False. If there isn't the will for moderate reform today, reform will necessarily be that much more dramatic when it becomes inevitable in the future.

After all, what is more human than putting off facing an unpleasant reality as long as you have the choice to do so, until it worsens to the point where you have the choice no longer?

Jim Glass writes:

"Would I opt out of Social Security? .... I would want to see rigorous demonstrations that:

"... government bonds will somehow become worthless between now and 2018 which is implicit in Tanner's claim that the true 'day of reckoning' is then..."

Not become worthless by then, but remain as meaningless as they are now.

In every year after SS goes cash-flow negative relative to payroll taxes it will have a deficit to finance to be able to pay benefits. Call the deficits for these years $X1, $X2, etc.

Absent the bonds in the trust fund, the government would have to collect additional general revenue of $X1, $X2, ... etc, annually to close that funding gap.

But with the bonds in the trust fund, the government will have to collect additional general revenue of only $X1, $X2, ... etc, annually to redeem the bonds that will be used to close the funding gap!

Hey, wait ... if one can rigorously demonstrate that $X1, $X2, ... etc. = $X1, $X2, ... etc, then it follows that the bonds in the trust fund are meaningless to the funding of SS.

And as their effect then is seen to be the same as if they didn't exist, one might even say the 'true day of reckoning' for SS will be when it goes cash-flow negative relative to payroll taxes -- and Congress will have to start coming up with an ever growing amount of *extra cash* to pay for them, from income taxation or some similarly pleasant financing alternative. Which will occur around 2018 or so.

As to the "worthlessness" of the bonds that the Treasury has issued to *itself* in the trust fund as a means of saving to finance its future obigations, they are no more or less worthless than any debt that any party issues to itself as a means of saving to pay a future cash flow need.

Feel free to look forward your own future retirement financed by cash flow you will receive on a note that you issue to yourself as an act of saving. Set the interest rate real high and retire rich!

"... That it is possible for investors to escape not just firm-specific risk, but also market risk, so that there will not be any risk to private-accounters from a stock market crash."

Hey, if average return on a diversified portfolio over the next 30 years is a below-historic-norm 5%, and then the markets falls 50% the day before you retire (and you stupidly didn't diversify into secure investments during the decade before then) you will still come out *way* ahead of the return you are promised by the SS funding formula.

It takes a government to *guarantee* negative returns over 30 years!

Boonton writes:

So SSI taxes are sky high. Yet SSI benefits are negative! SSI's administrative cost (assinging everyone numbers, printing checks, keeping track of it all etc.) are a tiny (less than 1% I seem to remember). Odd isn't it that the rate of return could be negative.

Boonton writes:

I've created a simple spreadsheet to illustrate my thinking, I can email it to whoever wants it...

I begin with the following:

Salary Contribution Balance

I start salary at $30K and let it increase 3% per year (let's assume 0 inflation). I set the contribution at 10%. That may seem high but remember many employers provide partial matches. Balance is simply a 5% return on the previous years balance plus the current years contribution.

From age 30 to age 65 our hypothetical worker retires at an ending salary of $86,948.35 with a whopping $464,427.04 in his 401K style account! I then begin negative contributions of $24K per year to represent his withdrawls to fund his retirement ($2K per month). Because his account continues to earn 5% on the remaining balance, even at age 95 he has $412,694.56 left.

Now onto part II

Boonton writes:

At age 65 my Everyman worker has retired and starts to make withdrawls from his account. But just like SSI's 'account', there is no box with $464K in it at the bank. His account is in the form of bonds, stocks, and other assets that he must sell to turn into $24K cash per year.

Every time something is sold, by definition it must also be purchased. So in my hypothetical simulation at age 65 I added a 2nd generation's salary column. I started this column at $30K*5 (to represent 5 workers for each retired person). Again I let that 2nd generation salary grow by 3% per year. Now instead of using a 10% contribution rate I figured out how much this 2nd generation would have to put into their plans to purchase assets from our retired guy. To buy $24K of assets each year, the 2nd generation must contribute 16% of their salaries to their retirement programs. For 16 years they must contribute more than 10% to guarantee that $24K of cash will be available to buy the retired guys assets.

So the Social Security bind has not been resolved. If the 5 working people refuse to buy the retired guys assets then the value of his assets will fall until their price makes it worthwhile for the workers to pick them up for $24K. By assuming a 5% constant rate of return we prohibited price drops but in the real world assets do fall which means the return the investor made on them would also fall. It doesn't matter if this retired person keeps his money in a bank that guarantees him a constant 5%. That just shuffles the problem to someone else. The bank will have to do something with those assets that will earn them 5% to fulfill their obligation to the depositor.

So again we are stuck with either the retired person getting less of a retirement than he thought, the workers contributing more than their parents had to, or some combination of both.

Mark writes:

But with the bonds in the trust fund, the government will have to collect additional general revenue of only $X1, $X2, ... etc, annually to redeem the bonds that will be used to close the funding gap!

The fiscal impact of bond issuance has nothing to do with having to "redeem" them, since as long as the government retains its credibility in financial markets, it can roll over its debt indefinitely. Rather, the fiscal impact is the interest cost of servicing the debt. We are already paying interest on the bonds held by the SSA. Selling those bonds, or selling others to roll them over, will not increase the total indebtedness of the government by one dollar. There might be one minor fiscal impact--as our bond sales to the public will have to increase somewhat, there might be a slight impact on the interest the government has to pay on new bond issues in 2018 or whenever the "day of pseudo-reckoning" winds up being.

Of course, the government has to keep its fiscal house in order between now and then--this really should go without saying. If a presidential administration were to start running massive deficits to finance regressive tax cuts, they might raise the debt/GDP ratio high enough that rolling over debt would become too costly, and thereby forcing tax increases or benefit cuts circa 2018. But that is not an argument for privatization, but an argument for electing a 44th President of the US later this year.

Boonton writes:

Assume for the sake of argument that the SSI 'surplus' does not cause other gov't spending (or SSI benefits for that matter) to be any higher than what they otherwise would have been.

What would be the effect of SSI surpluses either eliminating gov't debt or reducing what it otherwise would have been up until the trust fund starts needing money?

Comments for this entry have been closed
Return to top