Deinonychus antirrhopus

The only fix possible is raising the retirement age. You get to retire the last 13 years of your (expected average) life, not after the first 65. As the average lifetime goes up, so must the retirement age. Because then you're supporting yourself, and the number of workers supporting retirees remains tolerable.

It doesn't matter whether it's Social Security of some privatization! What happens if you privatize? Then the average market return falls until age of retirement that results is supportable.

Why? It's the fallacy of composition (``If everybody stands on their toes, everybody can see better.''). You can only save now and spend later if you find an opposite partner to take the other side of the trade, wanting to spend now and save later. For one person to save to retire at 65 is easy. For everybody to do it is impossible, because everybody is the same, wanting to save at the same time and spend at the same time.

How does this work? The rare persons on the opposite side are able to bid really great returns for themselves, because the majority bid against each other to save now and spend later. It doesn't matter whether the economic pie grows or not. Growth just gets an even better deal for the rare current-spender and future-saver.

The return falls until the retirement age you can in fact save enough for, results in few enough retirees so that the remaining workers can support them.

Incidentally, Social Security is an annuity, so it's not really unfair to people who die early. It's insurance like an annuity is insurance, that you don't outlive your assets. In return, the issuer gets the assets when you die. Apples to apples and oranges to oranges.

So anyway, the looming Social Security financial disater isn't a financial disaster at all. It simply points to a political change that absolutely will happen. The retirement age goes up.

There is a huge difference between government bonds and other forms of savings that you just fly right over. I don't know what the rate of return on the nonmarketables in the trust fund is, but it really doesn't matter because that interest will just be taxed away from the population. It could be 20%, but it still doesn't mean much, except a heavier tax burden and larger transfer from worker to retiree. At least if accounts are privitized and allowed to be actually invested in the private marketplace (not just used to feed government consumption), that can increase the capital stock of the counry and provide for true gains.

Jayson,

There is no interest on the Social Security bonds. It's the accounting entry that allows the government to spend back into the economy the dollars that would otherwise disappear from the money supply. The government must spend those dollars because it cannot save, in principle. The truth of the situation is that the future retirees will be supported by future workers' taxes _at that future time_, regardless of how the bookkeeping is done. That's why you can't have as many retirees as a 65 year retirement age would produce. That simply cannot be made to work with any financing scheme whatsoever.

If the government put the dollars in a mattress, the Fed would instantly buy debt to offset it; but government spending would then cause the Fed to sell debt, resulting in a wash exactly equivalent to what happens now, namely the dollars sent in are instantly spent.

If the government tried to buy a market index fund, what I said way above would happen, namely the rate of return would go to zero, but with the added disadvantage that the government would be planning the economy in effect.

Arnold Kling wrote a good article about transition costs to a privatised system.

Ron has good comments above. I would like to add that There is currently scheduled an age increase of about 2.5 years (and remember the age is only for full benefits) over the next couple of decades. There is no real reason to stop before it hits the real average age of death in this country.

There are other side doors to benefit decreases that need to be looked at. Greenspan highlighted the CPI increases which he believes have over adjusted benefits. Also, either benefits need to be fully taxable under the federal income tax, or what would be better, a VAT should become the Federal governments primary form of revenue.

And on a final note the Medicare swamp needs to be drained.

to prevent crazy uncle Eddie from trying to time the market

Be nice, you young whippersnapper.

Whoa...Uncle Eddie what you doing reading my blog!?!?! Sorry about that.

As for the idea that privatization will drive the rate of return to zero I don't believe that argument for a minute. The problem with the argument is that we don't currently have just one generation, but several of them (this is one reason why economists use over-lapping generations models). Some will want to buy, some will want to sell. As the number of buyers and sellers changes the price of securities will change, but the idea that it will go to zero and stay there is not something that sounds reasonable to me.

Arnold Kling's article on transition costs is a very good one as it highlights another benefit of privatization that most ignore.

As for the CPI, yep its biased, but trying to fix that is tough. It'll basically have to be a political decision, but politicians are reluctant to do so for obvious reasons.

Why not lower the retirement age to ten? The necessary savings could be supplied by grandparents, maybe with an assist by the government for the poor.

One thing is that every worker would be less than ten. They'd supply everybody else with goods and services. Where's the hole in the idea? They'd be getting paid, after all. It's just a question of financing, right? Nothing about reality has to come up.

If you can see it's reasonable that that can't be done no matter how you finance it, then you can ask, well, what age will it work at?

You don't have to look at financing at all. You need enough workers to support themselves plus the retirees.

Then you can ask, still being reasonable, how does financing fail when you try to make it work where it can't? In the case of privatization, it's in a collapse of the rate of return. Too many of one kind of people, too few of another.

Ron,

Citing a patently ridiculous example as support for your argument wont persuade me.

Your argument would be valid if we had only one generation, we don't so your argument falls flat. Now if you want to argue the rate of return could fall so low as to make retiring at a given age problematic that is another thing all together now.

It's like calculus. You look at how things change. Start with a situation you can analyze directly, and then ask how the financing scheme fails for it. It fails by the fallacy of composition. One ten-year-old can retire easily, a gift from his grandparents. They all cannot.

The problem in social security is that the relative populations of kinds of people prevent you from financing what you want. There are too many going-to-retire, and too few very-young to make it work. That's a fallacy of composition. The whole population can't do what any single individual can easily do alone.

If everybody stands on their toes, everybody can't see better, even though if one person does, he can.

It isn't the fallacy of composition. Society is not comprised of a single generation, but at least 3 maybe 4.

The problem in social security is that the relative populations of kinds of people prevent you from financing what you want. There are too many going-to-retire, and too few very-young to make it work. That's a fallacy of composition. The whole population can't do what any single individual can easily do alone.

The whole population is not going to be doing what an individual does. This is where your argument falls down. Further, it isn't like the elderly are just going to turn around and sell everything at 65. They'll sell it off gradually. Sure the price of stocks will move around, but the idea that the imputed rate of return goes to zero implies that nobody values future consumption. That alone shows why your argument is just plain not going to happen.

If everybody stands on their toes, everybody can't see better, even though if one person does, he can.

Faulty analogy, as noted several times now.

I don't see what you're seeing in ``multigenerational.'' The question is too few and too many of different kinds. As you lower the retirement age, you get too many people selling for the number of people buying. That drives down the return, no matter how many generations you have. All it needs is that the number off retirees increases as the retirement age decreases. How is that hard to see?

The return conversely increases as the retirement age increases. At some age there is a magical equality, where the return is enough to support retirement at exactly the same age. That's the retirement age dictated by the dynamic system.

Like the Laffer curve, you can display the dynamics by looking at endpoints where the result it obvious. If the retirement age is 100, the system has no difficulty. If the retirement age is 20, the system won't work no matter how it's financed. Somewhere in the middle is an age where the system can be financed and just works.

You can't just pick ``65'' and say that it's there. It's not there. It's higher.

Well which is Ron? First the analogy was everybody was doing exactly the same thing (e.g. standing on their toes) now it is too few and too many.

And what is it with the lowering of the retirement age? Seems like a total red herring? I haven't advocated lowering it, let alone raising it. Further, retirement age isn't like a magic number.

"Ah-ha, I'm 65 dump my portfolio and WAAHHH-HOOOO, I'm off to life in the fast lane."

Its like you are mixing apples and oranges here. On one hand you want everybody to do the same thing (buy or sell stocks, bonds, and other insturments) and then on the other you are saying the retirement age is...is...I don't 10 years old or something.

Lets start simple, you have 3 generations, each with 10 people. Where is the problem if the people in this economy are financing their own retirement via savings? Consider this the baseline case, we can change it around later. Right now keep it simple: 3 generations, with each generation of equal size.

Is there a problem here, if so what is it? If not, why isn't there a problem?

Ron Hardin, reading from here: http://www.ssa.gov/pressoffice/pr/trustee03-pr.htm

"This report is yet another reminder of what we have known for some time: Social Security's long-term financing problems are very serious, and will not be fixed by wishful thinking alone," said Jo Anne Barnhart, Commissioner of Social Security. [...]

"The release of this report is a good time to remind people how the Social Security program works. Social Security taxes pay the benefits of today's retirees. Money in excess of what is needed to pay today's benefits is invested in special issue, interest-bearing Treasury bonds. This system works well when there is a relatively high ratio of workers to beneficiaries. For instance, in 1965, there were 4 workers for every Social Security recipient. [...]

Interest earned on the invested assets of the combined Trust Funds was $80.4 billion in 2002. The combined trust fund assets earned interest at an effective annual rate of 6.4 percent.

Jason, nobody pays the interest. The money does not exist.

Steve: with 3 generations, say child parent grandparent, each family supports 2 children and 2 grandparents, if you retire at grandparent age. You have people dying in one generation (no great grandparents), which is (if both parents work) called a one-to-one ratio, which is too much and it won't work.

Another check: grandparents didn't stop working at grandparenthood when there was no Social Security and people did provide for their own retirement. In addition they died well before greatgrandparenthood.

I should have responded to ``lowering the retirement age.'' The number of retired decreases if you raise the retirement age, and increases if you lower it. It's the same fact about the mathematics. If you raise it enough, the system is solvent. If you lower it enough, the system is insolvent. That's what I meant to say.

That variable, retirement age, is easy to control in Social Security. So that's the solution. Draw up a retirement age schedule, born in X-Y year, you retire at age Z, and so forth for other birth years.

If you privatize, the same sort of schedule comes out of the mathematics of the rate of return and its response to ``everybody investing'' as required or suggested by the new law, whatever the new law may be. My point is to say that privatizing doesn't eliminate the fallacy of composition. The rate of return does not go unchanged when everybody invests, but falls to resolve the contradiction it entails. Contradictions are resolved in real life by violating an assumption, and that's a general law.

Ron,

I'm having a hard time seeing the problem you are seeing here.

Again.

We have three generations. Lets simplify it so that each generation lives for three periods. Period one is childhood/learning basic work skills. Period two is the working phase. Period three retirement. Now, if each person is financing their own retirement then I don't see a problem. You say the workers have to support the child and the retired, but this isn't the case, the retired are supporting themselves via drawing down their savings. These savings are replaced by current workers who are saving for their retirement. I see no problems.

Steve, savings aren't real. They're trades with somebody else. ``I did such and so in excess of my consumption for society, and now society does such and so much in excess of my production for me.'' But the savings themselves aren't anywhere.

The other guys though are not willing to trade on those terms. So it fails, in one of the many ways it can fail. One way is inflation (so the owed support is not enough to live on, but on the same hand is not much of a burden any longer to those supplying services). Another way is forcing you to postpone your retirement, via reduced earnings, until enough retired people die off.

At some point, at a high enough retirement age, it works; everybody is happy, more or less. Workers can support the retirees, with the prospect that they themselves will receive similar support - that's why they're happy. But they're not going to do anything to support too many retirees, and some political change will happen to dump them.

Take your example, and look what happened before Social Security. People did save for their own retirement. When their children had their first child, did they retire? No. They worked on quite a while, more or less until they were senile or sick or dead. Then their family took care of them. It worked out. Average that out and that's the level of support the working population is willing to supply. More than that counts as ``not working.''

It doesn't matter how you finance it, that's the constraint. Savings are a medium to get family support; not any particular family, but it averages out exactly the same over the families still working.

What about the fallacy of composition? It's still there. A single individual can save for lavish retirement. It doesn't average out to much over all families, so it can work. It can't work for everybody though. It screws up the average to too high a level.

I suppose you could print money with food value.

Ron,

I completely disagree with much of what you have written above. I think it is best that we simply agree to disagree on this.

Steve,

As someone who is caring for one eighty-five year old grandmother and three kids, I understand exactly what Ron is talking about.

If too many people need to sell their investments (to pay for their retirement) and too few people need to buy those investments (to save for their retirement) the value of the saved investments will go down. That's why he is trying to decrease the number of sellers and increase the number of buyers.

Yours,
Wince

Wince,

Right. Notice when I wrote,

As for the idea that privatization will drive the rate of return to zero I don't believe that argument for a minute. The problem with the argument is that we don't currently have just one generation, but several of them (this is one reason why economists use over-lapping generations models). Some will want to buy, some will want to sell. As the number of buyers and sellers changes the price of securities will change, but the idea that it will go to zero and stay there is not something that sounds reasonable to me.

The idea that the price will go to zero however is extreme and highly unlikely. Further, this isn't a fallacy of composition fallacy. The fallacy of composition is: everything in this item is small and light weight, hence the item is small and light weight.

Also missing is a notion of equilibirium. If there are not enough buyers, then either the price drops or if the person selling doesn't want that price they don't sell. The retirement age could also be allowed to change to allow for this. In fact, I'd prefer to let people decide when to retire vs. penalizing them for making a choice.

Finally, the price of stocks and so forth are not simply decided on the numbers of buyers and sellers. While that is a factor there are other factors as well.