Powered by TypePad

« Terri Schiavo, R.I.P. | Main | The Man Sans Q Is Back »

March 31, 2005



Even if one assumes a strong link between economic growth and stock market returns, it's only a contradiction to predict low GDP growth now and high returns after privatization if one assumes that GDP growth and privatization are independent variables. One might plausibly argue that privatization leads to more investment and that investment leads to greater growth, so that growth will be anemic if we don't privatize and robust if we do.


Amusingly, the SocialSecurityChoice web site dumped on Shiller's using global equity returns as a benchmark thus:

Then, to make life cycle accounts look even more "disastrous," Shiller uses historical returns that are far less than those actually achieved by the investments he is simulating. Instead of using returns from United States markets, he uses returns from an average of 15 countries. What, exactly, is the relevance of those 15 countries? Today the country ranked 15th by gross domestic product is Indonesia -- is there really any point in including the investment history of such a country, except that doing so makes the President's proposal look bad? Shiller simply says it is "more realistic."

As for a US government run index fund succeeding in capturing worldwide market share for growth driven by growing population overseas, I think it is reasonable to assume the biggest opportunities practically everywhere and always go to the locals, and our six percent of GDP trade imbalance suggests if anything, we've got a long way to go just to get to something like a neutral position in the world.

There is perhaps some sliver of hope in the fact that, though the US exports debt, we still buy more equity overseas than overseas types buy equity here. But I somehow suspect that the leverage, sooner or later, will work against the US rather than in our favor. And at any rate, it is doubtful that a US government run index fund would come up on the winning side for global equity investments.

David Walser

One of the many problems with trying to forecast market returns is trying to determine which of the current trends reflect a relatively permanent change in the market and which are temporary in nature. For example, over the last decade, a growing share of our national savings have moved into the equity market as more and more firms make 401(k) plans and similar opportunities available to employees. How much of the overall increase in the equity market is due to the increased level of demand for equities created by this flow of funds? Each pay period, about 10% of our firm's total payroll goes into equities because of the 401(k) elections our employees have made. When combined with the 401(k) dollars from all the other firms across the nation, this cash has to pressure the market upward. Is this a permanent change, or will the trend reverse when the baby boomers retire and start taking cash out of their plans? I hope for the first, but I believe the last. If this is not a permanent change, equity returns may be lower than they have been in the past.

Another change that might be more permanent is the way equity markets compensate investors. In the 1960s the average dividend yield from stocks was much higher than it is today. I forget the exact numbers, but IIRC, it was 4%+ back then and less than 1% today. Rather than pay dividends, corporations have retained the assets and deployed them to produce more growth. This allowed investors to reap more of their investment return as tax advantaged capital gains rather than tax punished dividends. Has this change led to a change in the economy? Perhaps, if the corporations were able to put the cash to a more productive use than the individuals would have, GDP should be higher than it otherwise would have been. If this is a permanent change, equity returns may be higher than they have been in the past.

Of course, while whether equity returns are higher or lower than expected is an important question, of greater interest is how the equity market performs compared with its alternatives. Historically, equities have out performed bonds for virtually any period greater than five years. I don't think that's about to change anytime soon.


I finally saw someone quote figures on the fed employes fund, which is supposed to be prudent low risk investing. It has averaged about 12 percent per annum gains over the last 15 years. This type of fund has been suggested as a model for SS investing regarding reasonable risk.
Is this better than t-bills? I think so. I can't understand why the admin is not saying more about this, and other funds / groups which have a special deal the rest of us do not have.

creepy dude's evil friend

"It has averaged about 12 percent per annum gains over the last 15 years"

I also am baffled why the admin is not saying more about this, since they lie about everything else.

Jim Glass

When one realizes that the DeLong/Krugman/Baker logic of lower stock returns equally implies lower bond returns (as Arnold Kling has pointed out)... and one considers the effect that lower-than-projected bonds returns would have on the solvency of the SS status quo (as Arnold has also noted) ...

Hmmm.... it looks to me like whether stock returns fall or not, the D/K/B analysis creates a pretty strong Catch-22 argument supporting SS reform.

Near seals the deal for me! Thanks, guys.


I may in a bit over my head here, but I think there is one benefit to personal retirement accounts that we're not addressing, and that is ownership. Even if the only qualified private investment was US Treasury Bonds, at least the so-called Social Security Trust Fund would be owned by American workers rather than the Chinese government.

I mean, let's get honest here. Isn't that why any attempt at reform Social Security, in current and previous administrations, has been resisted? Right now, surpluses of Social Security tax revenues are spent and back-filled with US Treasury bonds. So, any attempt to take those surpluses out of the "petty-cash" box is met with vociferous resistance.

Brad DeLong

Uncle! Uncle! Uncle! It's not "impossible"; it's just really unlikely...

Harry Arthur

VD, ownership is definitely the key - the ultimate "lock box" if you will. All the numbers are, see comments above, SWAGs, based on numerous assumptions many of which can't even be classified as "scientifical" (the S in SWAG).

At the risk of extreme redundancy, as in "that Arthur guy is a total broken record", Galveston has a totally privatized system with NO stocks or stock funds at all - a totally "banking instrument" system. It makes SS look positively anemic by comparison.

Sorry guys, I know this REALLY wasn't the subject but ...


I just drove by a huge Kodak factory in China last week.

Corningware is building a factory in Taiwan.

I assume all the local profits earned from these endeavors wouldn't assist those companies' stock prices but not affect US GDP?


So the "ownership society" means "own the whole world"! Who knew?

Harry Arthur

Sammler, I'd say it starts with owning your own retirement program, as even the least among us (economically speaking) arguably ought to be able to do.

Read the comments above - the rest of the world owns a pretty good piece of US too. And, honestly, I'd rather we owned "the whole world" than the Chinese own it, wouldn't you? Now if we can just figure out the hydrogen uel thing we could own the whole world ...

Patrick R. Sullivan

When I hear Professor DeLong retreat to "really unlikely", I consider his army to be routed. Maybe, on a subconscious level, things like:

"...the world in 2050 or 2100 contains a great many
people outside the U.S. whose productivity would be amplified if they were able to move to the U.S...."

uneasily co-exist with their admission that 19th century Britain took advantage of something like the above by overseas investment.


Points to note about the Galveston plan.

The plan was initiated when interest rates were at a peak. If the plan had begun 10 years earlier, participants in the plan for those 10 years might have returns of about HALF the level for those since 1981. As a defined contribution plan, people participating at different times can expect significantly different benefits.

The plan also sheds responsibility for the "transition cost". Galveston doesn't have to contribute anything to supporting the legacy cost of Social Security. If participants had to bear a fair share of that burden, returns would be considerably lower.


In response to sammler: "So the "ownership society" means "own the whole world"! Who knew?"

The U.S. government takes in a surplus in actual SS revenues, but it spends those surplus funds on other government activity. In place of those actual funds is a special promissary note to pay that back to Social Security. Though the Chinese government holds up to $1.1 trillion dollars of our national debt, none of it is the special promissary notes used for the Social Security Trust Fund.

In fact, by pilfering the Social Security surplus assets while declaring the debt used to backfill off-budget, the Clinton Administration was able to be in the bizarre situation of declaring "surpluses as far as the eye can see" while continuing a net increase in total public debt in order to finance continued deficit spending. In a true surplus, the national public debt would be decreased (by payments) or at least remain the same (if payments were interest only), but that has not occurred since 1960.

I would prefer that the American government and American citizens have the option of holding as much of that debt as possible, rather than some third world nation, like China. Also, though I eat well, don't smoke, and get moderate excercise, there is only so much I can do to fight genetics. There is not a very promising history of long life amongst males in my family. So, if I die at age 65 or 70, it would only be right that I transfer some portion of over 50 years retirement savings to my heirs.

Jim Glass

Mankiw has a response (.pdf) to B/D/K paper that does a number on it...

This paper is really three papers in one.

The first paper is a roundabout and not completely persuasive discussion of what return we should expect for the stock market, given current measures of valuation.

The second paper is a straightforward review of how population growth affects the return to capital in standard models of economicgrowth.

The third paper is a model of the equity premium that can be viewed as creative,bizarre, or vacuous, depending on your point of view.

What links the three papers -— beyond their common title page — is their motivation...

[snip the economics, down to the politics]

The case for moving Social Security from a defined benefit to a defined contribution structure is that it gives individuals more choice and control over their retirement income and the government greater transparency in its finances.

These arguments are not based on any particular estimate of the average return to capital or of the equity premium. I don’t think the key issue in the debate over Social Security is whether, over the next century, the risk-free return will be 1 or 3 percent, or whether the equity premium will be 3 or 5 percent. So even if I agreed with the arguments raised in this paper and lowered my estimates of rates of return, it would not change my mind about the need to reform Social Security or the kinds of reforms that are

I would guess that, in their hearts, the authors of this paper agree with me about this.

To see if I am right, I would like them to answer the following question: Suppose that next week, the stock market falls by 50 percent, so dividend and earnings yields double.

Would Baker, DeLong, and Krugman suddenly be in favor of President Bush’s proposal for Social Security reform?

I suspect they would not.

If I am right, this suggests that while the paper raises some interesting questions about the future of assets returns, as far as the debate over Social Security goes, it is largely a non sequitur.


I continue to be facinated by how the critics of the SS reform nearly always detour into a debate about potential/future rates of return, or the assumptions thereof, while it's the dysfunction of the structure of SS that is causing its ruination.

Life expectancy has gone way up, and the birth rate has declined, resulting in a much lower worker-to-retiree ratio. The government has been spending the SS surplus since the federal budget was unified 30+ years ago, and there are no assets (only authority to tax income) backing SS obligations--obligations the Supreme Court has ruled are only promises and not binding on any future Congress.

The Ponzi scheme is being exposed.

Meanwhile Krugman, DeLong et. al. are trying to convince the rest of us that the game is really Musical Chairs--and their method of making the music last longer (tweaking here and there) is better than adding, rather than removing, chairs (retirement assets owned in private accounts).

It would be funnier if our beloved professors weren't so righteous--but then those of the left always know what's best for the rest of us.

The part I find so amusing is how Krugman, DeLong, et. al., have become staus quo conservatives, while those pressing choice and private account reform are truely the progressives on this issue.

Try as they might, the left never fails to disclose their affection for collectivization and socialism--we all should be suckled by the teat of the nanny state until the day we die--or until the left pulls your plug because your SS checks have become inconvenient.

snark off\

Patrick R. Sullivan

Dean Baker is now, over at Max Speak, conceding the entire argument. His fallback position is that it's extremely unlikely that investment overseas will rise. Mixed in with the sunk cost fallacy.

Unmentioned, is any forecast of how plausible 50-65% increases in tax revenues are. Which is what is needed, to believe that SS is gonna be okay.

Harry Arthur

CH, you said "The plan was initiated when interest rates were at a peak. If the plan had begun 10 years earlier, participants in the plan for those 10 years might have returns of about HALF the level for those since 1981. As a defined contribution plan, people participating at different times can expect significantly different benefits."

Fair enough, however, is your objection based on the level of analysis elsewhere in this thread or is it a generalized guess? Since the Galveston, Brazoria, Matagorda, et al, privatized program has been in effect for approximately 25 years, through periods of high inflation/interest rates and almost nonexistant inflation/interst rates, we should be able to come up with at least an analysis to date of how they've been doing down there on the Texas coast.

Here are some of the details that Judge Ray Holbrook testified to before the President's Commission to Strengthen Social Security in 2001:
Annual rate of return over 21 years: 7%-8% compared to less than 2% under the current SS program;
System based on commercial banking products, annuities and bonds;
Disability insurance pays up to 60% of worker's salary, up to $5,000 per month;
Life insurance program pays 3 times worker's salary up to $50,000 a year or a total of up to $150,000 compared to $255 from SS.

Even if we assume that your generalization is true and the rate of return experienced for the first 21 years of the Galveston, et al, program was cut in HALF, the return would still almost double the current SS rate of return, not to mention that other aspects of the privatized benefits far exceed SS.

Are annuities currently particularly bad retirement instruments? Bonds? I don't find your analysis particularly compelling, given that for today's young workers the projected returns on their SS forced "investment" will be negative.

You also said "The plan also sheds responsibility for the "transition cost". Galveston doesn't have to contribute anything to supporting the legacy cost of Social Security. If participants had to bear a fair share of that burden, returns would be considerably lower."

First, I don't share your assertion that it is somehow not "fair" that younger workers not bear a "fair share of the the burden" for the transition. Given your concern, however, I would suggest that transition costs are either not a cost at all as Lawrence Hunter suggests in his article in Human Events titled "No, Virginia, There Are No 'Transition Costs' for Social Security Reform", or they are at worst the movement of future unfunded liabilities from off balance sheet to on balance sheet. Transition "costs" will be paid by either the SS Administration borrowing or General Fund borrowing to offset the money being placed in personal accounts. At the very least this is borrowing to save as compared to borrowing to spend - a significant difference.

Hunter makes two essential points: "1) Any debt sold to the public in the short term during the so-called 'transition period' to pay all promised Social Security benefits simply refinances an existing liability; and 2) Refinancing Social Security's debt through personal accounts makes possible a reduction, and eventually an elimination, of Social Security's unfunded liability, which otherwise is scheduled to grow without bound."

My only difference with his last statement is: absent major SS tax increases, general fund tax increases, reductions in benefits, or significant increases in the retirement age.

I would submit, along with Judge Ray Holbrook, that it's high time we created "A Better Deal" for our workers, children, and grand children. The sooner we act, the easier to take advantage of long term compounding and the longer we wait the greater will be the costs. To think that the current SS system in anything like its current configuration is sustainable by some minor or even fairly major "do more of the same thing tweaking" here and there just doesn't pass the common sense test with me. Ownership is the only solution for the long term viability of this program.

But of course I could be totally dead wrong. In which case I'm standing by to hear someone, anyone, from the "do nothing" side of this argument tell me logically and succinctly why Galveston, et al, is a bad idea, why Chile is a bad idea, and why it's such a bad idea to control your own retirement program instead of the government controlling it.

richard mcenroe

Growth will slow because the locust-like boomers will begin stripping all available capital out of the economy to support their indulgent retirements. Look for the Democrats to introduce legislation on behalf of AARP to require all GenX workers to assume direct fiscal responsibility for 2.5 retirees each to bring us into line with EU standards...

richard mcenroe

Growth will slow because the locust-like boomers will begin stripping all available capital out of the economy to support their indulgent retirements. Look for the Democrats to introduce legislation on behalf of AARP to require all GenX workers to assume direct fiscal responsibility for 2.5 retirees each to bring us into line with EU standards...

richard mcenroe

Sorry for the double-dip. Feel free to delete one.


DeLong and Krugman and the rest of them are just laughing their asses off at us. Look how we allow them to divert the discussion to the arcane Beavis Model projections and bicker over whether 5% or 6% is more realistic.

Barring huge tax increases, my generation can expect benefits to be cut by at least a third. That's not just negative return, it is hugely negative.

So comparing the imputed return of today's recipients to some model for future returns is ludicrous. Let them explain to us how they intend to keep earning the return today's recipients are supposed to be getting. What is it, 2%? Tell us, Slim, how would you fund this 2% return for future recipients?

We're making a mistake by indulging them on this point.

Harry Arthur

sw, I hope your generation starts discovering these facts soon and that you funnel some of your anger into the political process. Unfortunately the republicans are doing a HORRIBLE job making the case for ownership and the other side has their head in the sand or possibly somewhere where there's even less light. Meanwhile they have managed to divert us into picking fly droppings out of pepper as you suggest.

This whole thing ain't anywhere near as tough as the "do nothing" advocates claim, nor do I think we need the level of detail argued above. To me it's common sense that there's a mess coming our way - yours because you will have to pay and mine because I'll be retired and there's no way to pay the bill from the general fund - it's just that simple.

I'm still standing by for a rational discussion, based at least partially on common sense I hope, on why we shouldn't at least be looking very seriously into Galveston, Chile, etc.

Defined benefit, pay-go programs are a relic of the twentieth century. Increasingly companies are turning to defined contribution programs such as 401K and eliminating or vastly scaling back defined benefit because they are simply too expensive. Take a look at all the major US airlines for example.

In my opinion, Europe has a much worse problem looming than do we, though I note we haven't talked about their defined benefit retirement "success" at all in any of these threads. I wonder why.

Robert Musil

Don Luskin posts a wonderful and lucid demolition of the Baker, DeLong and Krugman nonsense, including this passage:

"Baker, DeLong and Krugman expend 41 pages of junk science trying to come up with [an inconsistency in the Social Security Administration actuaries' forecasts of how well personal accounts holding stocks might perform]-- and fail. Acting as the "paper's" discussant at Brookings last week, N. Gregory Mankiw (the widely admired Harvard economist and recent chairman of the Council of Economic Advisors) was merciless. Mankiw called it "the strangest contribution to the equity premium literature I have seen... [it] can be viewed as creative, bizarre, or vacuous..." He noted it relied "on the level and growth of a variable that, to a first approximation, does not matter." And he declared, "If I took this model seriously, it would do more than inform my view of the equity premium. It would shake my faith in corporate capitalism!""

In the past DeLong has pretty much admitted that his worst nightmare is Luskin - and this makes it all real. Read the whole thing, you can't help but feel the authors squirm:

Harry Arthur

Robert, superb link!

Here's the conclusion: "But wait -- it gets even worse for the leftist economists. Their 'paper' claims that, in a low growth environment, real stock returns should be about 4.5% annually. That may be lower than the 6.5% assumed by the actuaries, but it's higher than what a worker can expect to earn from the existing Social Security program. And it's a lot higher than what a worker could expect to earn from the program once its already shaky finances are rocked by the impact of lower economic growth.

"So what have we learned from our illustrious leftist economists? First, that it's a wonderful world -- high growth as far as the eye can see. And that's a great world in which to own a Social Security personal account that can invest in stocks. Second, if the world doesn't turn out to be so wonderful, that same account can be your personal lock-box to protect you from benefit cuts -- and stock returns will still probably beat the miserable returns of the current program.

"Class dismissed. Professors exit left."

Unfortunately the left has never let facts get in the way of any good argument.

Harry Arthur

And the article even quotes our friend Jim Glass!

The comments to this entry are closed.