Arnold Kling  

The Three Percent Solution?

PRINT
Energy Conservation... Ayn Rand, Wise Philosopher Des...

This briefing on the Administration's proposals for Social Security personal accounts cleared up a number of issues, such as the way that accounts would be administered and the type of investment options available. In both cases, the Federal Employees' Thrift plan was used as a model.

However, the briefing created some confusion concerning how to adjust Social Security benefits for people who choose to divert some of their payroll taxes into the accounts.

Suppose that worker B chooses to stick with the current system, and worker A chooses to put 30 percent of her payroll taxes into a personal account. Then the logical outcome should be that when they retire worker A gets 30 percent less in retirment benefits than worker B.

That logical outcome can be easily calculated if A and B make permanent decisions at the start of their working lives. But what if worker A only decides at age 45 to use a personal account? Or what if personal accounts were not available historically to worker A, who is now 45 years old and has that opportunity?

In order to adjust worker A's benefits to make them fair relative to worker B's benefits, some sort of interest rate or discount factor is going to have to be applied that equates the benefits that A foregoes at retirement to A's choice to put money into personal accounts when she works. Here is what the briefing says about this:

in return for the opportunity to get the benefits from the personal account, the person foregoes a certain amount of benefits from the traditional system.

Now, the way that election is structured, the person comes out ahead if their personal account exceeds a 3 percent real rate of return, which is the rate of return that the trust fund bonds receive. So, basically, the net effect on an individual's benefits would be zero if his personal account earned a 3 percent real rate of return. To the extent that his personal account gets a higher rate of return, his net benefit would increase as a consequence of making that decision.


The choice of interest rate involves a trade-off. The lower the interest rate that the Administration and Congress select, the more profitable it will be for individuals to select private accounts. On the other hand, the higher the interest rate that they select, the more fiscally conservative will be the transition to personal accounts.

If the Administration were to choose any interest rate above zero, then they become open to Paul Krugman's charge that


people are expected to take a loan from the government and use it to buy stocks, and if that turns out to have been a mistake - well, too bad.

Experts usually tell people to plan for their retirement by investing in a mix of stocks and bonds. They disapprove strongly of speculation on margin: borrowing to buy stocks. Yet Mr. Bush wants tens of millions of Americans to do exactly that.


On the other hand, if the interest rate were less than 3 percent, then the "actuarial scoring" of the personalization plan would show that it increases the deficit.

I think that I would have opted for a lower interest rate. I would rather err on the side of giving personal accounts a generous return, and let the "actuarial scoring" take the hit. I think that the real interest rate of 3 percent is too high. I would have picked 2 percent, or perhaps even a lower number, and I think this could be justified on the basis of historical real returns on long-term bonds. Furthermore, the most recent calculation of the yield on inflation-indexed government bonds by J. Huston McCulloch is about 2 percent.

In any case, it strikes me as elementary economics that some sort of interest rate has to be used in the calculation. If Paul Krugman and Peter Orszag want to imply otherwise, then they ought to not to represent themselves in the media as economists.

For Discussion. Should the interest rate be fixed permanently by legislation, be variable based on a formula, or be set periodically be some sort of board?


Comments and Sharing


CATEGORIES: Social Security



COMMENTS (57 to date)
Randy writes:

Perhaps I'm missing something here, but wouldn't it be easier to adjust the Social Security wages paid based on the percentage chosen for the personal account? An individual who diverted some percentage of income would get credit for a proportionally lower Social Security wage contribution, resulting in a lower standard benefit - which would hopefully be made up for by the value of the personal account.

James writes:

Randy, I think you missed this:

...what if worker A only decides at age 45 to use a personal account? Or what if personal accounts were not available historically to worker A, who is now 45 years old and has that opportunity?

The math gets somewhat complex -- what if A starts and stops personal account contributions 6 times throughout their life?

I think the answer is to just track actual contributions to the "old system" and calculate benefits as normal.

E.g.: normally, a worker who retires at 67 might contribute $200,000 (made up number) and get a benefit of $30,000 (again, made up number). So, if that worker only contributes $100,000 (because the rest is diverted into a private account), the benefit should be $15,000. If the worker starts late on PA contributions, and as a result contributes $150,000, they should get a benefit of $22,500. This isn't hard, and I'm pretty sure that they have to keep track of everything you've ever contributed anyway, in order to calculate what you get.

Basically, since a high-wage worker who diverts part of their tax into an account contributes the same to the "old" system as a worker who makes half as much, he should get the same benefit as the low-wage worker.

Have I missed something?

Arnold Kling writes:

"Basically, since a high-wage worker who diverts part of their tax into an account contributes the same to the "old" system as a worker who makes half as much, he should get the same benefit as the low-wage worker.

Have I missed something?"

My guess is that a worker could game this system by opting for private accounts until age 60 or so, and then going back into social security at age 60. My guess is that using the "simple" methodology you could get a huge amount of Social Security benefits compared with what you would get using a present value calculation.

spencer writes:

Wait a minute -- do I understand that you want the govt to subsidize stock investments?


Wow -- explain how that goes with your economic
beliefs.

J Mann writes:

Arnold, I think James makes a decent point. If you revised the SS system so that it calculated interest on contributions when determining each worker's final benefits, you would have effectively flipped the clawback plan.

Boonton writes:

Basically Krugman's (as well as Brad de Long) analysis is correct. This plan is no different than getting a loan at 3% (or whatver rate you want to pick) against your Social Security Benefits. If you do better than 3% you keep the profits but if you do worse you incur the loss.

Why are we doing this again? Why not simply mandate 2% of payroll (1% employee, 1% employer) go into an IRA/401K of the individuals choosing on top of social security. This requirement can be refunded against those who are already contributing to a 401K/IRA. Simply tweak the Social Security benefits, if necessary, to keep the program in long term balance.

Such a plan would actually increase national savings. Bush's plan, amazingly would cause a net decrease in national savings! It's as if the man hates the future.

Edge writes:

The interest rate should be set to match the real cost of the borrowed funds, plus the cost of the life insurance policy to recover the funds for those funds beyond any portion of funds not expected to convert to an annuity, to be revenue neutral.

You could implment the life insurance premium as a surtax, perhaps a quarter of a percent.

Or, you could acknowledge that you are giving people who opt for inheritable personal accounts a better deal than people who don't.

I think you'd also need to include some premium to cover the cost of the capital tied up for 50 years. If the US government borrows a rolling 25% of GDP into the indefinite future, even if there is a matching asset at the end of 50 years, that still has a real economic cost. So, there should be a premium.

Again, unless you're acknowledging that you're letting one group of people forgoe part of the obligation to pay current beneficiaries.

Boonton writes:

This is an interesting question: Should the gov't borrow at, say, 2% and loan that money to people at 3% in the hopes that their investing will yield a profit to both the Treasury & themselves?

Here's a modest proposal, why not implement automatic tax loans? The way it would work is if you owe the gov't, say, $1500 in taxes you could automatically defer paying that provided you accepted terms such as an interest rate 1 or 2 points above the US Treasury rate and you paid all the interest plus a certain portion of the principle each year.

If a business could obtain a 5% return on their internal investments then they could take their payroll tax liability and turn it into an instant loan. This would put the gov't in the position of being a huge banker issuing de facto loans to the private sector where private banks will not (go ahead, how many people could get personal loans to invest in the stock market from a normal bank at a rate that wouldn't make it prohibitive). Questions:

1. Why is the private sector unable to fulfill this need when there is obviously profit to be made?

2. Keeping #1 in mind, is it because the risks are actually much larger than they appear?

Randy writes:

James,

What I'm saying is, on any given payday, if I choose to contribute 4% of my social security tax to a private account, then that paydays total social security contribution would be reduced by 4% (aprox). At time of retirement, my total contribution would be less, so my standard benefit would be less, but I would have my personal account to make up the difference. The formula applied to determine the standard benefit would be the same whether or not I had a private account - and it is this formula that would be changed to keep the system as a whole balanced.

In my opinion, this would be completely flexible. A person could start, stop, or change his or her election almost at will.

Randy writes:

Arnold,

I agree, you could game the system by contributing the max to the private account in the earning years outside of the high 35. But, the private accounts, due to their $1000 limit, aren't going to make up much of the system in any event. Thus these accounts will benefit the lowest wage earners mostly, and if these should choose to game the system and make a few extra bucks, that's a good thing.

Lawrance George Lux writes:

Boonton has the basically correct idea, which in Investment terms, remains a good mix between assured Returns and Venture capital. Real data has not been given to define the effect of economic performance on Private Accounts.

What would a Depression like the 1930s do to the rates of Return? What would a Downturn such as 2001 do to the rates of Return, and how long would be the impact? What would rates of Return be if a severe economic shock took the air out of current Stock Balloon (50-50 chance), and the Market returned to a Dow 5000?

All plans to Privatize SS rely upon a sustained Market without sharp alterations. A Baby Boomer has seen the Market move up and down sharply four times in their Worklife. What happens when you have to buy high, and are forced to sell low?

These are the questions which make Privatization risky. lgl

Paul Zrimsek writes:

By Krugman's "logic", anyone who buys stock is taking out a margin loan at an interest rate equal to the yield on the bonds he could have bought instead. (While there are a few differences between passing up the chance to put money into bonds and passing up the chance to put money into traditional SS, they're generally not of a nature to make SS look very attractive by comparison.)

Randy writes:

Lawrance,

Re; "What would a Depression like the 1930s do to the rates of Return?"

Good question - but another good question is, what would a Depression like the 1930s do to the current system? What happens when unemployment hits 25%, the trust fund empties in a year, and the tax base collapses? The current system is also built on the assumption of a steady economy. Indeed, the key to understanding Social Security fully is this; there was a first generation, and there will be a last generation.

Mcwop writes:

LGL,
Don't those same forces affect the taxes collected by government? A depression means that payroll taxes could plummet, and the government be forced to borrow more, cut benefits, and/or cut spending. The SS projections use a sustained GDP figure without sharp alterations. I wonder what a 10+% unemployement rate would do to payroll tax revenues. Probably the same effect as it might have on the stock market.

Also we already know what the downturn of 2001 has done to returns. I am looking at a fund with a portfolio mix of 60-30-10 (Stock, Bond, Cash) that has a 10 year avearge annual return of 10.56%. A 40-40-20 mix with a 10 year avearge return of 9.53%. Or an equity income fund with a return since inception of 13.53% (Inception Date October, 1985). That return includes 3 sharp market drops.

The market has its risks, but so does the SS system. SS faces certain demographic risk. A demographic risk that has already triggered several payroll tax increases, which reduces the pay a worker keeps (the return they get on their labor).

Lastly, when you retire you do not sell assets all at once.

Boonton writes:
By Krugman's "logic", anyone who buys stock is taking out a margin loan at an interest rate equal to the yield on the bonds he could have bought instead. (While there are a few differences between passing up the chance to put money into bonds and passing up the chance to put money into traditional SS, they're generally not of a nature to make SS look very attractive by comparison.)

And that's exactly what it is, this concept is otherwise known as opportunity cost. The bond yield, BTW, is just a cover for a much more complicated calculation. Your SS taxes do not earn a bond yield, your social security benefits are in the form of an annuity whose true return is unknowable until you die. A person who lives very long will have to do much better to make up for the lost SS benefits.

I'm still waiting for someone to tell me why this is really a good idea. This seems to be a roundabout attempt to exploit the fabled 'equity premium'. The equity premium is a historical oddity, stocks have had higher returns relative to risk than other investments. If you believe in the market then you cannot seriously believe in the premium. There is no rational economic reasons for stocks to have higher returns than other investments after adjusting for risk. The premium is almost certainly a historical oddity.

If it isn't then the premium is a magical money machine. Gov't can borrow at 6%, buy stocks (directly or indirectly thru 'private accounts') and reap 9% returns. Rule One of economic analysis; when you stumble upon something that gives you a free lunch forever something is very wrong!

Good question - but another good question is, what would a Depression like the 1930s do to the current system? What happens when unemployment hits 25%, the trust fund empties in a year, and the tax base collapses? The current system is also built on the assumption of a steady economy. Indeed, the key to understanding Social Security fully is this; there was a first generation, and there will be a last generation.

Social Security would probably be the most secure thing in such an economy. The fund's revenue is tied to the underlying economy. A while ago I ran 10 and 15 year average growth rates of real GDP. There was only one year (1932 I believe) where the average 15 year GDP growth was negative...even there it was more than -1%.

I suppose there will be a last generation. So what? Are you seriously proposing SS must be able to survive the end of the world? If so private accounts won't help you all that much.

Boonton writes:
The market has its risks, but so does the SS system. SS faces certain demographic risk. A demographic risk that has already triggered several payroll tax increases, which reduces the pay a worker keeps (the return they get on their labor).

mcwop, risk is the inability to predict the future. The actuaries at SS were able to predict the portion of Americans over 65 this year to within 40 basis points back in 1934.

The problem is that no matter how you cut it Social Security, by being tied to the underlying economy, is less risky than any particular piece of the economy. For example, a corporate bond/stock has all the political risk of SS (if not more since its quite easy for Congress to blindly pass a law costing a company millions), all the risk that the economy may collapse AND the risk associated with corporate bonds/stocks as an asset class and finally the risk associated with that particular company.

The proper view is to see Social Security as a low risk leg of retirement that counterbalances the risks of private investment/savings. By its nature Social Security provides a fail safe back up that permits people to take more risk than they otherwise would.

Jim Erlandson writes:

Money in a "private account" would go into your estate upon death. If you die at age 61, money in Social Security is gone forever.

The real purpose of this plan is to take this money away from the government so it can't be spent on other programs and projects.

Mcwop writes:

Boonton,
You posted the numbers on your own site. Most males born in 1970 are certain to get back less than they put into social security - a negative real return.

That is why many people do not like the current system when 40+% of its contributors born after a certain date will lose money? I'd rather take my chances with some diversified investments.

Bernard Yomtov writes:

Arnold,

I think you have misrepresented Krugman's position. He does not urge that no interest be charged. he simply disagrees with the whole notion of the government lending people money to invest in stocks.

Paul,

Yes. In fact there is such a thing as opportunity cost of capital. Leaving Social Security aside, when you put money in the stock market instead of, say, a savings account, you are forgoing the interest you wold have earned on your savings. Similarly, if you buy stocks instead of paying off your highest-interest debts the cost is the rate you are paying on that debt.

Arnold, you have to choose between the two rules you enunciate for calculating how much to reduce Worker A's regular Social Security benefit payments. If A reduces his contributions this year by 3/10, say $2000, either we reduce his future benefits by the equivalent of 3/10 for a year, or we reduce his benefits by $2000(1 + i)^t, where i is the interest rate and t is the number of periods to retirement. The two rules are consistent only when benefits represent an implict rate of return of 3% on contributions. For most workers implicit rate of return is well below 3%.

I agree that a 3% real rate is probably too high. The Treasury's own website puts the real rate on 20-year T-bonds, based on the TIPS market, at just below 2%. That's a strong argument for lowering the price of opting out. A desire to subsidize stock investments is not a strong argument.

Lawrance George Lux writes:

Another Blog discussing the Social Security talked of the German Pension system after 1923, which was much like Social Security. Benefits were scaled down, but still sufficient for survival, while Equity holders lost practically everything because of no guaranteed minimum.

We have currently lost over a quarter of the value of the Dollar since 2001. Bush spending and taxing patterns assure another major devaluation if continued. The devaluation could reach 50%, placing Us in a position similiar to Germany in 1923. It might force Price-indexing instead of Wage-indexing, but SS would be sound. Bonds and Cash would be Crap, and Stock Dividends would be Crap by the time paid. How does this affect Private Accounts? lgl

Victor writes:

Why not calculate your private account contribution just like AIME, but separately? I.e., take your 35 highest contribution years, and directly subtract that from your AIME?

You couldn't really game that one by contributing early and then forgetting it later.

But this highlights what I think is the main reason for why they have de-linked it from AIME altogether: the benefit structure is highly progressive. Rich folks would LOVE to take $1000+ of their income and subtract it from AIME if it also meant they only had to forgo 15% of their AIME, on the margin.

This also shows the problem with Arnold's original "30% reduction in benefits for a 30% reduction in payroll taxes" scheme. The notional account offset is a very clever way to keep many attractive properties of the benefit scheme intact. However, it comes at the huge cost that we are now discussing.

No, this is just a fundamental problem. If you had *guaranteed* constant real benefits, it wouldn't drive me so nuts. But this sort of proposal drops after-offset, non-privatized benefits down to extremely low levels (I think I posted the calcs on my blog a few weeks ago when I got started on this kick, although I may have only posted up the graph. I keep intending to get back to it).

Personally, I'd rather forget about the notional account offsets altogether.

Randy writes:

Boonton,

Re; "Social Security would probably be the most secure thing in such an economy."

I agree with you on that. The government could always socialize the entire economy, redistribute whatever it could lay its hands on as it saw fit. Then again, they could do the same if the system had been based on private accounts before the crash. I guess my real objection to Social Security in any form is that I fear it is hastening us towards that time.

Personally, I think the creators of this system were ethically challenged, to put it mildly. I'm giving my checks directly to my daughters. I'll pay for my parents, but I will not make my daughters pay for me.

Peter writes:
Bernard Yomtov writes:

Arnold,

I think you have misrepresented Krugman's position. He does not urge that no interest be charged. he simply disagrees with the whole notion of the government lending people money to invest in stocks.

I'm still trying to figure out how the government not taking away 4% of my income is equivalent to them loaning me 4% of my income.

Paul Zrimsek writes:

Very good: we all realize that investing in stocks always entails an opportunity cost. So why am I the only one to realize that the logical conclusion of Krugman's silly argument is that not just that people shouldn't invest their SS money in stocks, but that they shouldn't invest any of their other money in stocks either?

Since we've just observed Ayn Rand's centenary, and can't be too far from the second centenary of Boonton's "historical oddity", he might consider these words of wisdom from the former in connection with the latter: "Contradictions do not exist. Whenever you think that you are facing a contradiction, check your premises. You will find that one of them is wrong."

Edge writes:

Suppose there was an "opt out" option for Social Security. You could agree to "opt out" of the program, but you would have to pay a legacy tax to cover the historical transition cost.

Then, you would agree to get zero traditional benefits plus you would pay the legacy cost.

What would the legacy cost be? I'd guess it would be something like a 4% tax rate. You'd save 6.2% FICA tax, but forgo all old age benefits.

That would be the "Goldwater" option, fully opting out. You'd get zero benefit and pay a tax of 4% of payroll for legacy costs.

The "Bush" option lets you sort of opt out on 39% of old age benefits. You'd put 4% out of the 10.2% FICA tax for old age benefits into a personal account. Then you'd agree to cut your traditional benefits by an equal amount - 39%.

You should also deduct the extra legacy tax, at a rate on 39% of 4%, from your contribution to the personal account. You'd start with about 60% of the money you choose to divert, and agree to cut your traditional benefit by 39%.

Is this about right, to get to a neutral cost evaluation, where the private account owner wouldn't be subsidized either by other SS beneficiaries or by the general population?

Greg writes:

Opportunity cost is NOT LIKE A LOAN! Let's look at a margin purchase of $100 of stock, with a loan rate at 3%. If the stock only returns 2% per year the liability would be growing faster than the asset. As a result existing equity would at fall by around $100*(1.01)^T over T years. In this margin scenario WEALTH IS DESTROYED.

Opportunity cost states that our 2% returns will yield wealth that is about $100*(1.01)^T LESS THAN it would be under a plan that earns 3%.

See the difference? With margin investing underperforming destroys wealth. Opportunity cost just gives the amount of underperformane we have RELATIVE TO THE 3% BENCHMARK. Hey, the wealth is still going up by 2%. It's not going down!

So, to all of you who have asked, "Am I missing something?" The answer is, YES!

Randy writes:

It has become obvious to me that I am in over my head here, so I'm going to make one last point and then just sit back, listen, and learn.

Before the great depression, accepting welfare carried with it a sense of shame. It is now accepted with a sense of entitlement. My friends, the trust fund is not the problem, the sense of entitlement is. As long as this sense of entitlement exists, we will be fixing the trust fund every ten years until...?

Am I recommending that we reintroduce a sense of shame? Yes. Absolutely.

Bernard Yomtov writes:

So why am I the only one to realize that the logical conclusion of Krugman's silly argument is that not just that people shouldn't invest their SS money in stocks, but that they shouldn't invest any of their other money in stocks either?

Because that's not the logical conclusion. Suppose you divide your portfolio between safe treasury bonds and the stock market. The higher the percentage you put in the market the riskier your portfolio becomes. You might even go so far as to borrow money to put in the stock market, in effect leaving a negative percentage in safe instruments. This increases your risk even further. Krugman's point is that this is what the proposal encourages - the highly risky strategy of borrowing to invest in stocks, and that this is generally considered excessively risky when dealing with retirement funds.

The main point you are overlooking is that portfolio risk changes with allocation. Some risk is fine, butit is possible to overdo it. Krugman is not arguing that it is always unwise to have some stock investments. he is arguing that it is unwise to borrow money in order to invest in stocks for retirement purposes. These are two vastly different positions and to claim they are equivalent is simply wrong.

Boonton writes:
Boonton, You posted the numbers on your own site. Most males born in 1970 are certain to get back less than they put into social security - a negative real return.

The total returns of the system are equal to the the growth of the economy. What you are describing is how those benefits are distributed via the benefit formula. I think that since most males appreciate the survivorship benefit that their wives are able to receive (well, maybe some divorced men won't :) ) the returns for males are probably underestimated. This is part of the nature of a progressive program. If we kept a progressive policy (say by matching funds for poorer workers) and funded it by progressive taxation the returns for the wealthy would be hurt as well.

I agree with you on that. The government could always socialize the entire economy, redistribute whatever it could lay its hands on as it saw fit. Then again, they could do the same if the system had been based on private accounts before the crash. I guess my real objection to Social Security in any form is that I fear it is hastening us towards that time.

If Social Security is hastening us towards that time its certainly taking its time. The projections, assuming poor growth, show that its net deficit is maybe 2% of GDP a half century from now....if you want to socialize the economy you'd be better off with 4 terms of Bush...he's at least doing it in clips of 4% per year!

I'm still trying to figure out how the government not taking away 4% of my income is equivalent to them loaning me 4% of my income.

Really? Bush has proposed reducing the payroll tax by 4%???

Greg:

See the difference? With margin investing underperforming destroys wealth. Opportunity cost just gives the amount of underperformane we have RELATIVE TO THE 3% BENCHMARK. Hey, the wealth is still going up by 2%. It's not going down!
So, to all of you who have asked, "Am I missing something?" The answer is, YES!

If the present value of your benefits are decreased at 3% per year while your 'private account' grows 2% you'll be falling 1% per year behind an alternate universe where you didn't opt for the private account.

Which leads to another serious question: Are we really going to assume that as more and more people accumulate assets in private accounts the demand for gov't to bail them out in bad years??

Jon writes:

Krugman and Orzag's comparison of Bush' plan to borrowing money to buy stocks falls along the lines of how many analysts examine investments in structured products. It is neither wrong or dishonest for them to point out this similarity.

Also, although Boonton is correct to point out that the equity risk premium is a historical oddity; it is not necessarily a sign of an irrational market. We are observing the returns ex post; the equity risk premium of earlier times accounted for risks of events that never happened. Other country's turbulence (e.g. World War II) produced different historical return patterns in their equity markets.

Steve writes:

Is this SS proposal much ado about nothing? It doesn't increase SS taxes or reduce benefits in any measurable way. My understanding is that if we don't do one or the other (or perhaps change the retirement age, which is a form of benefit reduction) then social security as we know it is not sustainable.

As to the interest rate question, my hunch is that most who opt for private accounts will invest conservatively (bonds) and are not likely to earn a 3% real rate of return. I would choose variable interest rates based on a formula.

Jim Glass writes:
Paul Zrimsek:

By Krugman's "logic", anyone who buys stock is taking out a margin loan at an interest rate equal to the yield on the bonds he could have bought instead....
~~~~

Boonton:

And that's exactly what it is, this concept is otherwise known as opportunity cost...

Great!

Now all you have to do to finish squaring this circle is find a dictionary of finance and/or economics that defines "opportunity cost" as "taking out a loan", and we'll be done.

Then the WaPa can retract its retraction and reprint its original story that made the "loan" claim.

And Krugman can retract what he wrote on the web yesterday, "My column this morning wasn't the finest...", and stop feeling guilty about his failings.

(After all, it's not like that column was anything like his worst.)

Bob writes:

The choice of the interest rate is an interesting strategic problem. Ideally, we should set it at zero to maximize the switch to private accounts. This will, of course, lead to massive borrowing to cover the transition. Given the willingness of Asian central banks to lend us money at extremely low rates, I don't see why anyone would object to the Treasury borrowing a ton. The private accounts need to be set up to encourage investment in real assets, ideally in developing countries with young populations and capital needs (if only Latin America could get its act together...). Future Congresses respond to the crushing debt load in the standard way by deflating it and screwing those that lent us the transition cost. The inflation period is painful, but not terribly so because the real assets financed by the original borrowing retain their value. Problem solved, at least for the US. Even the balance of payments problem is solved, only we export our problems rather than real goods.

Don't laugh, this is the road we're on. Life is good when you can denominate your debt in your own currency and the "preditory" export economies have tied their lifeboats to your sinking ship. The boomers are a one-time problem and you can screw your creditors once. The choice of an interest rate, and every other policy choice made for SS over the next decade, will be driven by politics, not economics, because the key to this out is leaving enough of a haze to let Asia continue to live in its fantasy world that somehow everything will turn out okay. I don't think that you can set it to zero without rocking the boat, but we should push the envelope.

Movie Guy writes:

This is a good discussion.

It's interesting that no one has acknowledged any relationship or need for a relationship whatsoever between the 3% fee and the overall transition costs of establishing the added on program, estimated by CBPP to be $6 trillion by 2030.

Source: http://www.cbpp.org/2-3-05socsec.htm

If our auto industry operated this way, a car division or brand unit within any manufacturer could design, test, produce, and market a new vehicle without any concern for recovering R&D, plant, tooling, and preproduction build (prototypes and Fed crash testing vehicles), and prebuild labor costs prior to producing the first marketable unit.

I'm sure that my corporate friends at GM will find this to be an interesting non-reality-based concept of operating. Or perhaps Ford will adopt it and simply go out of business. Their economists may be thinking along these lines and following the lead of innovative Government proposals, such as this one.

The absence of a bottoms up total costing analysis makes the currently proposed private accounts program look like what it is - a pipe dream. But one that will likely be implemented at considerable near term cost.

Social Security Plus, on the other hand, wouldn't have this problem. Nor a 3% fee loan fee to the users. Moreover, we would benefit from additional national savings.

Boonton writes:
Also, although Boonton is correct to point out that the equity risk premium is a historical oddity; it is not necessarily a sign of an irrational market. We are observing the returns ex post; the equity risk premium of earlier times accounted for risks of events that never happened. Other country's turbulence (e.g. World War II) produced different historical return patterns in their equity markets.

My pet theory is that the equity premium is the result of the stock market crash of 1929. Being the prelude to the Great Depression, I suspect that it gave stocks a bad name for quite a long time. That meant that stocks ended up offering a premium return since people were irrationally fearful of them. As the Great Depression recedes from our collective memory, people have become less fearful of stocks thereby bidding them up to generate an abnormal premium for the hypothetical investor who brought the market in 1929 and held it until today.

Is this SS proposal much ado about nothing? It doesn't increase SS taxes or reduce benefits in any measurable way. My understanding is that if we don't do one or the other (or perhaps change the retirement age, which is a form of benefit reduction) then social security as we know it is not sustainable.

This is the great laugh here, Social Security is perfectly sustainable. Again we are talking about maybe a 2% of GDP deficit in half a century from now. We are being told we have to move mountains to avoid this horrible fate while today we happily rack up 4% of GDP deficits!

I'm not surprised my old Social Security Blog Comment Debating Parter Jim Glass has shown up here! Too bad he doesn't seem to get the joke either. Bush hasn't proposed anything for social security aside from cutting benefits drastically. That's the first cut, the second cut comes if you opt fo the 2% account.

If you wanted a real reform simply take the 2% account and split it away from Social Security. Require 1% from employee and 1% from employer to go into an IRA of the employee's choice *unless* the employee is already putting 2% or more into a 401K or IRA. That would increase real savings and perhaps begin to address Bush's mission to bankrupt this country....but don't expect to hear anyone propose that any time soon.

Boonton writes:
And Krugman can retract what he wrote on the web yesterday, "My column this morning wasn't the finest...", and stop feeling guilty about his failings.

Krugman retracted nothing nor should he. Jim, why do you and so many other SS haters distort the truth so much? For those interested, they can read what he really wrote at: http://atrios.blogspot.com/2005_01_30_atrios_archive.html#110755213925785666

Robert Brown writes:

I don't understand why there has to be an interest rate associated with the benefit reductions associated with private accounts.

The current relationship between payroll taxes and benefits is simply a political decision with no fixed rate-of-return relationship. Likewise the benefit reduction should be chosen politically to be large enough to help solve some of the solvency problem while not discouraging workers from choosing private accounts. The reductions would likely have to be means tested is some way since high income workers receive much smaller benefits relative to their taxes than low income workers.

Jim Glass writes:

Boonton wrote:

Krugman retracted nothing nor should he.

Boonton, where did I say Krugman retracted anything?

I wrote (and you quoted me as writing):

And Krugman can retract what he wrote on the web yesterday, "My column this morning wasn't the finest...", and stop feeling guilty about his failings.

Do you see any difference between between the meanings of "K can retract" and "K did retract"?

This isn't the first time you've had trouble reading what I wrote.

Jim, why do you and so many other SS haters distort the truth so much?

Tut, name calling is rarely persuasive. And since you want to stand on the truth, I'd appreciate it if you represent what I've said in cases like the above accurately from now on, before slurring me about them.

BTW, speaking of respecting the truth, have you found that definition of "opportunity cost = margin loan" yet?

Lawrance George Lux writes:

Read what I have posted today on my Blog, and provide commentary:
http://laglux.blogspot.com/

We are building Windmills to charge, and it is to Our long-term disadvantage. lgl

Jim Glass writes:

Boonton wrote:

Social Security is perfectly sustainable. Again we are talking about maybe a 2% of GDP deficit in half a century from now.

This is, of course, disregarding the 2.97 points of GDP deficit by 2030 that the Social Security Trustees warn will result due to the cost of operating the trust funds -- and increasing thereafter.

Some people seem still wed to the notion that just because these funds have already been collected by Congress once through payroll taxes (and been by consumed by it) it will be cost-free to collect them again through a 35% increase in income taxes from present levels. That this money to operate trust funds will come from some kind of "free tax money fairy". So taxpayers will nary notice nor care.

And thus that the Trustees are wrong when they warn that this rising demand for general revenue poses real risk to the future of the system -- especially when taxpayers will have to pay it simultaneously with a 28% increase in income taxes for Medicare, creating a 63% income tax increase total by 2030 -- and a 98% increase by 2050 (when that small, maybe 2% deficit supposedly will first arrive).

But the Trustees are not wrong.

In any event, the "sustainability of SS" is not the issue with private accounts, as the latest White House briefing made explicitly clear, so the whole topic is a red herring as far as private accounts are concerned.

Boonton writes:

Jim,

I see your point about what Krugman was saying, I suppose I should avoid name calling and playground tactics even though you seem to be comfortable employing them against Krugman.

I've already listed a host of deceptive tactics employed by Bush's supporters. One that you are really big on though is mixing Medicare projections into SS projections to make SS look a lot worse than it really is.

Edge writes:

But Jim G, your good friend over at the Skeptical Optimist tells us the current national debt, around 67% of GDP, is small relative to our historical debt. And guess what! The Social Security trust fund debt is included in that debt! The debt graphed there isn't just the publicly held debt; it includes the various trust fund obligations. I guess we don't have a problem!

Steve - Plan II cuts benefits significantly - much more than necessary simply to achieve solvency. Then, carve-out private accounts on top of that create an increment of debt something north of 23% of GDP between now and 2036. That's according to the President's Report to Congress in 2004. The system comes back to annual revenue balance around 2050, and then theoretically pays back the 23% of GDP debt out of excess payroll taxes during the next quarter century. As if people will be willing to pay excess payroll taxes when traditional benefit are cut in half from the current levels, in order to pay off this "transition" debt.

Robert - I think private accounts should be neutral, cost wise. They should bear the transition cost, rather than shifting the transition cost to a huge debt burden that is paid off either by reducing traditional benefits or by shifting future taxes somewhere else.

Make the carved out accounts revenue neutral; then evaluate whether they remain attractive enough on their own to be worth doing.

Robert Brown writes:

Edge,

I think the SS trust fund should be used to pay current benefits when payroll taxes no longer suffice. That's what it's for. General revenue will have to be raised by borrowing or taxing soon anyway. Implmentation of private accounts merely moves the date earlier.

The lower payroll taxes paid by current workers should be compensated for by lower future guaranteed benefits. It seems it would be possible to implement a sliding scale on benefit reductions based on the income of the taxpayer. Perhaps offering minimum wage workers private accounts with little or no benefit reduction and more substantian reductions for high income tax payers, possibly making a sort of front end means tested benefit that would be politically palitable and yet reduce the over all cost of the sytem

jaffer writes:

The whole issue of privatizing part of social security program is a complex subject. I'd like to understand the following: President Bush says that money saved in personal accounts would grow faster if it's invested in well-diversified equity funds rather than in government bonds. If so, shouldn't he persuade Congress to allow (perhaps require) the Social Security Trust Fund to invest the current accumulated surplus, and future surpluses, into the equity funds (the same ones that the government would approve for us to invest in) instead of investing in low-interest Treasury bonds, which it has been doing all along. This way the Trust Fund would likely become solvent and the benefits would not have to be reduced, nor would the payroll taxes have to be increased, nor would the government have to borrow, in the future, to meet its obligations.

If he can persuade Congress to make this modest change, perhaps privatizing can be the next step. But if he is not able to convince Congress to allow the Trust Fund to invest in randomly selected well-diversified equity funds (even if it means investing some of the accumulated surplus) then it is futile to push for partial privatization of the SS system

Edge writes:

Jaffer, I think Bush's personal accounts are doomed, and fortunately so.

But I thought of one way that might overcome conservative's reluctance to allow the government to own assets other than treasury bonds. For every dollar controlled by individual in private accounts, put one dollar of the SS surplus into the same asset pool.

The whole thing is pretty silly, though, assuming the Federal Thrift Savings model is used for private accounts. I suspect the real concern is that someone, anyone, will vote the shares held. These days, practically any shares held by mutual funds, anywhere, are not really voted. That would hold for personal accounts by way of the federal thrift savings plan.

It really shouldn't be hard, these days, to let people set up a proxy to vote their personal interests, even for assets held by a mutual fund, if people wanted to allow voting control by the masses other than just "buy" and "sell" as issued by a mutual fund director. Could probably even be done without having to disclose positions, although someone could probably reverse engineer it.

Robert Brown writes:

jaffer, Edge

One could argue that a portion of the SS trust fund is, indeed, invested in private assets.

The excess payroll taxes collected over the past two decades were used for general expenditures, thus reducing the level of general revenue that needed to be raised by borrowing or taxing. Since the payroll taxes are paid by lower income people who tend to spend most of their wages and general revenue comes mostly from higher income people who tend to save more of their income, one can make the case that cash flowed from consumption by lower income taxpayers to investments held by higher income taxpayers with bonds representing SS claims against those assets placed in the trust fund.

As an example, suppose we were to decide to replace the SS trust fund bonds with equity investments. And, suppose we raised income taxes on the top 1% of tax payers to raise the cash to redeem the bonds in the trust fund. The net result would be to divert cash that would otherwise flow into private investments held by that top 1% to the SS trust fund which would then invest the cash in private investments. From an economic standpoint I don't see any difference in the trust fund holding equities v.s. allowing the high income people to own the equities with the SS trust funds retaining the right to tax those assets in the future as needed to pay SS benefits (assuming the SS investment board could allocate the investments as efficiently as the private sector).

Randy writes:

1. Imagine the government manipulating the actions of a corporation by threatening to dump it's stock. Need I say more?

2. An investment in government is also an investment in the economy, as the government's ability to tax is directly proportional to the state of the economy.

jaffer writes:

Brown,

Tax will have to be collected when the bonds mature, in any case. For now, investors can swap the their equity with the bonds held by the Trust Fund.

jaffer writes:

Randy,

"1. Imagine the government manipulating the actions of a corporation by threatening to dump it's stock."

An independent agency, which has the integrity of the Fed Reserve Board, would randomly select 5000 companies to invest the SS Trust Fund surplus. The holdings could be rotated every 5 years.

Edge,

The SS Trustees would be restricted from voting the shares held in the Trust, nor could the company's managers vote the shares held by the Trust.

Boonton writes:
The whole issue of privatizing part of social security program is a complex subject. I'd like to understand the following: President Bush says that money saved in personal accounts would grow faster if it's invested in well-diversified equity funds rather than in government bonds. If so, shouldn't he persuade Congress to allow (perhaps require) the Social Security Trust Fund to invest the current accumulated surplus, and future surpluses, into the equity funds (the same ones that the government would approve for us to invest in) instead of investing in low-interest Treasury bonds, which it has been doing all along. This way the Trust Fund would likely become solvent and the benefits would not have to be reduced, nor would the payroll taxes have to be increased, nor would the government have to borrow, in the future, to meet its obligations.

Of course it could. This nicely demonstrates why we should not rely on the so-called Equity Premium (the higher returns stocks have historically shown compared to other investments when you control for risk!). Conversly, if you really thought the equity premium was real you could simply finance social security by taxing stock profits (capital gains). If stock returns were really outstanding then the tax revenue would be as well.

Boonton writes:

Gov't ownership of stocks wouldn't be a problem either. The gov't could be forbidden from voting the shares (it's not like much get done by voting shares relative to selling shares anyway). There's a very simple way to have the gov't buy & sell in an unbiased manner:

You take the market cap of all stocks at a particular moment and rank them:

IBM $2.50 1M shares...$250M 71%
Apple $1 50M shares..$100M 29%

The gov't would buy 71% IBM and 29% Apple (with thousands of stocks in the real world, of course, this list would be much longer). The next day the portfolio would be rebalanced as the share prices change by the market.

Robert Brown writes:

jaffer: "For now, investors can swap the their equity with the bonds held by the Trust Fund."

I'm not sure what this means. The bonds in the trust fund are a liability to public investors. They could retire that liability by "exchanging" their equities for trust fund bonds. The bonds then are retired and no longer exist.

Jim Glass writes:
But Jim G, your good friend over at the Skeptical Optimist tells us the current national debt, around 67% of GDP, is small relative to our historical debt. And guess what! The Social Security trust fund debt is included in that debt! The debt graphed there isn't just the publicly held debt; it includes the various trust fund obligations. I guess we don't have a problem!

Well, if you don't want to have your Social Security benefits paid to you, then we don't!

You see, there are just two tiny little points you miss that my good friend doesn't talk about.

One is that it's one thing to carry a debt indefinitely, which costs the US gov't an average of only about 2% real of its total amount annually -- but quite another thing to pay off that debt. Such as to pay SS benefits post 2018.

Think of the difference between having a mortgage, and having your mortgage called by the bank.

Or of the difference between today's income tax rates, and a 35% increase in today's income tax rates.

The other thing is that my good friend's charts don't include the accruing implicit debt representing benefits newly promised by the government but for which no bonds have been issued in any trust fund.

What was the increase in that debt in just 2004? Oh yeah, $11 trillion.

But those promises are every bit as real and binding as SS promises, aren't they?

So when they come due all at the same time, since it's pretty self-evident we won't be able to pay them all -- even Krugman says that! -- I wonder which ones the Congress of the time will decide not to pay.

Edge writes:

OK, so Bush and the GOP added $8 Trillion of Medicare obligations last year.

You're arguing that cashing in the SS trust fund bonds will cause a cash flow problem. Indeed, it will require a couple hundred $Billion a year of current value when withdrawals peak.

Compare that, though, with Bush's privatization plan, which would have practically the same cash flow deficits to finance current SS obligations, and then add on another couple hundred $Billion more each year from now until after the SS trust fund is exhausted.

If your concern is about cash flows, then Plan II makes the problem worse.

I guess we can thank our lucky stars that these cash flows, at least, will be made in the most trusted, most liquid bond market in the world. If cash flows like this had to come out of the stock market, that, indeed, could cause trouble.

At least we've seen, in the past three years, that the US treasury market is capable of handling this sort of incremental supply of publicly traded debt.

Perhaps we should take the opportunity over the next 15 years to pay down as much of the treasury debt as we can in advance, and shift the average maturity of the remaining debt into something that won't need to be refinanced every three years, like the current public debt obligations.

If this isn't the most effective way to handle the cash flows, then our Federal Reserve chairman ought to let us know. But certainly, if cash flows are the problem, the federal reserve chairman would HAVE to acknowledge the status quo would have better cash flows out to mid-century than diverging one-third of SS revenue and issuing transitional debt in its place.

Boonton writes:

Jim nicely ignores the fact that the bonds do not have to paid off with taxes in 2018. There's no reason the Treasury simply couldn't turn them over and borrow them from the market like it does with all its debt.

Christine writes:

I just read (well I skimmed many of the comments)the thread on SS and the interest rate etc.. I can only come up with one solution that I like. We should completely phase out Social Security and go private a la Chile. I know it won't happen in my lifetime, and my husband and I are not anywhere near being able to put away enough money to not need SS when we retire. I am going to redouble my efforts, however. (Just think what we could have done with all that payroll tax we've been paying.)

The whole thing is a mess, and Medicare is in even worse shape. But it is now nearly impossible to opt out of that when we retire and buy our own insurance. No one will insure you when you reach age 65. (And if they do the cost is prohibitive.)

The entire subject is just too depressing. It's easy enough for some to say, well, we could save SS without making any other changes if we just cut benefits 20 to 30 percent. Thanks a lot! I think some are right that social security will have a last generation and it's the boomers.

Comments for this entry have been closed
Return to top