Third Time Lucky

Andrew Samwick has what will eventually stand up as the definitive rebuttal to the Krugman-endorsed Dean Baker idea about long term stock performance.  [The Big Finish to this post is in a new post.]

PGL of the Angry Bear Patch offers a similar rebuttal; Dean Baker responds.

Having scrolled through most of the comments, my impression is that for many folks, the confusion comes with the notion that, if the economy is growing steadily at 2%, it is not possible for stocks to earn steadily a return on capital of 7%.  Krugman himself may have contributed to this confusion when he says this:

The actuaries predict that economic growth, which averaged 3.4 percent per year over the last 75 years, will average only 1.9 percent over the next 75 years.

In the long run, profits grow at the same rate as the economy. So to get that 6.5 percent rate of return, stock prices would have to keep rising faster than profits, decade after decade.

That does sound like an unstable situation, doesn't it?  Of course, you might wonder why share prices grew at roughly 7% for the last 75 years, when the economy grew more slowly.  But history need not guide us here - a bit of simple economics will serve perfectly.

Imagine a numbingly steady-sate economy: Births = Deaths, New Job Entrants = New Retirees, and labor productivity, the capital stock, and the income shares of labor and capital are constant.  Just to toss in some numbers, let's imagine that Labor provides 90 units and Capital provides 100 units.  GDP is $100, with income divided as $90 to labor and $10 to Capital.

In that world, GDP will not grow.  Nor will profits.  However, despite growth in neither the economy or profits, the annual return on capital will be constant at 10%.  What gives?

Well, since we are not going to increase the capital base in our static world, the entire capital return must be paid out as dividends (if we were to buy a few extra machines, there would be no one to operate them anyway, so neither output nor the Income to Capital would rise - all that would happen is that the $10 to Capital would be divided over a broader base, resulting in a lower realized return).

So, no growth in profits or share price, but a constant 10% annual return - not so bad.

Well then, what is the intuitive appeal behind Krugman's argument?  Part of it may be a confusion about expectations.  If we all *expect* future growth of 4%, and actually experience growth of just 2%, we may over-invest, and hence end up with lower- than-targeted returns.

However, if we revise our expectations downward to 2% growth, we can adjust accordingly.  And the adjustment, in the form of reduced investment, can simply be a matter of not fully replacing machinery and equipment as it depreciates - in an extreme case, we could disinvest, with dividends exceeding income by the amount of depreciation.

What I have described is a bit of a truism, but it holds up in more complicated models.  Well, I say that, but as the Angry Bear points out (in a post I may not find again), when the model factors in increased labor productivity it is possible to reduce Capital's *share* of income even while the return on capital is constant.  That, BTW, is good news for workers but does not invalidate the point that Capital can earn its target in a growing economy.

So where, in his well-thought through letter to Martin Feldstein, did Dean Baker go awry?  Don't rush me!

After a bit of reflection, I am stumped by what looks like an important assumption he has made.  The Soc Sec Trustees project growth in the labor market, and in productivity, to project a growth in GDP.  Dean Baker observes, quite sensibly, that new workers need some capital to work with, as do more productive workers.  This implies a level of new investment.

Dean Baker then argues that all of this new investment must come out of profits, and is not available for dividends.  Why?  We have capital markets, and Franco Modigliani has a Nobel Prize, so I'm not sure we should get too hung up on dividends and the financing decision.

Setting that aside, my guess is that somewhere amongst these key assumptions excerpted below (my numbering) lies the answer.  My *strong suspicion* is that by assuming an earnings yield of 3.3%, Dean Baker implicitly (or explicitly) has assumed that the target return on capital is 3.3%; the rest of his analysis then circles back and demonstrates the implications of that assumption.  But why bother proving an answer that has already been assumed?  Of course, if his letter were titled "By Assuming That Target Equity Returns Are 3.3%, I Can Show They Won't Be 7%", who would read it?

Well, that is my late-night guess, but I need to reflect.  And I should nod to the thought that the earnings yield may be a useful predictor of equity returns.  And let's keep Victor involved here - he will either rally to this argument, or skewer me gently.

MORE:  "Third Time Lucky"?  Well, here I argued (endlessly) that Republicans aren't actually touting high stock market returns as a panacea (with new support from the LA Times); here, I wondered how a fellow who came to fame with the globalization issue could absolutely ignore it in the context of global growth and returns on capital.

UPDATE:  The bare outline of a new post is emerging in the comments below; I will put a bit right here:

Dean Baker is right - the Soc Sec forecasts are unkind to capital, and probably embed assumptions that are not consistent with current equity valuations.

However, Paul Krugman is wrong - growth is not the issue, and high growth scenarios in which Social Security does well need not be associated with high returns on capital and high stock market returns.

The gist:

As noted, a stable, no growth economy can have a positive return on capital. In the context of the current discussion, the "high growth" forecasts that save Social Security do *NOT* also produce high returns on capital.

Surprised? Look, if the higher growth is due to a larger work force (e.g., more immigration=more folks paying payroll taxes), then we have a larger tax base, and more workers employing more capital. Growth is higher, Soc Sec is saved, but the forecast return on capital does not change.

Or, suppose growth is due to higher productivity - if labor captures the benefit of increased productivity (not a bold assumption - skilled workers get raises), then higher growth through higher productivity does not improve the return on capital even though it will save Soc Sec (all those higher wages translate into higher payroll taxes, which is good).

My point - growth is not the issue per se; the key is, what method is used to forecast returns on capital. For example, if we assume that capital captures the income associated with productivity gains, then the income share of capital rises, and higher growth due to higher productivity should lead to a higher return on capital.

So, in the higher growth models that save Soc Sec, the returns on capital can be dismal if, for example, it is assumed thatcapital intensity is constant and labor cpatures productivity gains.

Conversely, it is easy to conjure low growth scenarios where Soc Sec goes bust quickly, but returns on capital increase. Imagine a world where Reps restrict immigration and roll back current labor protections (I do not advocate this, but am simply identifying a possiblity). In this scenario, we could easily see low labor market growth leading to low GDP growth. However, the reduced bargaining power of labor *may* mean that more of the gains of productivity go back to "capital", i.e., the income share of capital rises. Does this sound implausible? To me, it sounds like the Kerry campaign, with Benedict Arnold CEOs exploiting outsourcing to hold down wages and earn more.

Developing...

My Excerpt of the Key Assumptions of the Dean Baker Soc Sec Letter

(1)  The first step in deriving the projections for dividend yields is to start with the current earnings to price ratio. This is assumed to be 0.033 to 1 in this analysis. After the latest market run-up, the earnings to price ratio is probably somewhat lower than this, but the actual number will depend on the stock market aggregate and the period picked to measure earnings. For this analysis, the earnings to price ratio is assumed to remain constant throughout the period.

(2) These projections therefore assume that net investment is equal to the projected rate of growth of the labor force, multiplied by the capital stock (which is assumed to be 60 percent of the value of corporate equities), plus an amount equal to 1.2 percent of the value of equities, which is needed for the capital deepening implies by the Trustees productivity projections.

After deducting the fraction of earnings needed for investment, it is assumed that all other profits are paid out as dividends. For example, in the decade from 2030 to 2040, the Trustees assume that the labor force will grow at 0.2 percent annual rate. This means that an amount equal to 0.12 percent of equity prices will be needed to keep the capital labor ratio constant, while net investment equal to an additional 1.2 percentage points of equity prices will be needed for capital deepening. This implies that net investment will be equal to 1.32 percent of equity prices, while earnings will be 3.3     percent. This implies a dividend yield of 2.0 percent.

(3)  Capital Gains

    The Trustees assume that the capital share of income remains constant.[4] This means that domestic profits will grow at the same rate as GDP.

Comments

[A Tiny]

TM,

I admire your willingness to think this through. But you (and Samwick) should deduce something from the fact that Feldstein stopped replying to Baker. It's because he didn't have an answer. That's not because he wasn't smart enough to consider all the things you and Samick are thinking of. Nevertheless, at some point he gave up. At some point you and Samwick will give up too.

I also suggest you consider how it is that Social Security has been a giant issue in the US for years, yet the stock return question has never been addressed. Does this indicate to you there may be some very serious flaw in American politics?

[Patrick R. Sullivan]

"...there may be some very serious flaw in American politics?"

You think? When Phd's in economics blind themselves to the difference between capital's share of GDP and the growth in that share, because they can't bear the thought that their political opponents might be right about something?

[Max Sawicky]

While we're doing arguments from authority, I might as well post links to Peter Diamond's article in the SSA Bulletin, where he supports Baker's argument:

http://www.ssa.gov/policy/docs/ssb/v63n2/v63n2p38.pdf

He updates this in a paper Samwick links to but has the same basic conclusion. In re: the non-U.S. GOP thing, he devotes a paragraph:

"Another factor to consider in assessing the connection between growth and rates of return is the increasing openness of the world economy. Currently, U.S. corporations earn income from production and trade abroad, and individual investors, while primarily investing at home, also invest abroad. It is not clear that putting the growth issue in a global context makes much difference. On the one hand, since other advanced economies are also aging, increased economic connections with other advanced countries do not alter the basic analysis. On the other hand, although investment in the less-developed countries may preserve higher rates, it is not clear either how much investment opportunities will increase or how to adjust for political risk. Increasing openness further weakens the argument for a significant drop in the marginal product of capital, but the opportunities abroad may or may not be realized as a better rate of return."

Of course, maybe he and DeLong (who is sympathetic to private accounts) and PK are wrong, and TM and Jim Glass & the pseudonymous/anonymous Sullivan are right. Stranger things have happened, but not very often.

[Paul Zrimsek]

Though Krugman's column, for some reason, presents the constant dividend yield as an assumption, in Baker's analysis it's actually a derivation from the simple macroeconomic model he uses. I believe the crucial assumption is the last one: that capital's share of national income remains constant. Given the demographic fact driving the slowdown in GDP growth-- that in the future there will be relatively fewer workers and relatively more retirees living off their savings-- this assumption strikes me as suspect. But if we grant it nonetheless, disappointing returns to captial follow more or less automatically.

Speaking of suspect assumptions: if we're going to assume that whoever has the last word must be right, what conclusions shall we draw from the fact that "A Tiny" never replied to Maguire at all, but instead changed the subject to Feldstein?

[Bruce Berger]

TM, With respect to your simplified economy I think you need to take into account that a capital investment depreciates over time. I don't mean in an accounting sense but in a real economic sense. That would change your return on capital, because either some of the ongoing return on capital has to be saved to replace worn-out machines at some point in the future or at some point there are no machines and therefore the terminal value of the economy is zero.

[TM]

But you (and Samwick) should deduce something from the fact that Feldstein stopped replying to Baker. It's because he didn't have an answer.

You will be interested to learn that Paul Krugman has never replied to me on any point.

However, I have resisted your inference.

Max, I appreciate your acknowledgement that, at this point at least, you are offering nothing but an argument from authority. At some point, the mathematical question identified by Samwick, as to whether low growth and high returns are mathematically incompatible, might merit your attention.

Also, the possibility that Dean Baker simply used some economic models to conceal the fact (perhaps also from himself) that he assumed his conclusion probably ought to be addressed, although he would be the proper one to address it.

And let's not forget that I have the surely estimable Barry Bosworth on my side. Also Nobel Laureate Franco Modigliani.

Actually, I am pretty happy with my team, and very happy with my argument. And, under the new "Tiny" rule, it looks like we win if no one responds.

Oh, I appreciate the Diamond excerpt - his paper would not load on my PC last night. Since it is "not clear" to him that international flows will be decisive, it is "not clear" to me that he disagrees with me.

[TM]

Bruce - I think that the "income" to capital is net of economic depreciation - it certainly ought to be.

Or, if I were doing this with more accounts, I would divide GDP up into Consumption and Savings (which equals Investment), and make sure that Investment=Economic Depreciation.

All of which could be done in a steady state, and the return on capital would still be constant and (since I am setting the capital stock quite arbitrarily), the return might be quite high.

Paul - I believe the crucial assumption is the last one: that capital's share of national income remains constant.

I am leaning strongly that way. For example, I understand the problems with steady state models where profits grow relative to everything else, and eventually exceed GDP.

However, I think I saw a headline flicker by that said protits rose 20%in the last year. Presumably, thatis not the steady state growth rate, but... a one time spike in profits by 25% could take them from 8% to 10% of GDP (probably not outside of historical bounds). If this earnings spike were fully anticipated and did not prompt an increase in prices, it would also raise the earnings yield by 25%.

If the Dean Baker conundrum is that he assumed his conclusion, then assuming a higher steady state earnings yield (say, moving from 4% to 5%) solves a lot of the problem.

Folks who don't like that might settle for a five year period at the outset where stock prices rise a little while earnings rise a lot, thereby getting the earnings yield to "target". The steady state then takes over.

[A Tiny]

TM,

Come on. You're being silly, and you know it.

It's not just that Feldstein has never reponded to Baker. It's that no one -- not Feldstein, not Samwick, not a million other people, and certainly not Franco Modigliani -- has ever simply written down some numbers. If they believe this is possible, then it should be very easy to show how.

Baker has written numbers down. So has Peter Diamond. But everyone else simply makes vague conjectures about how things might be. Okay -- things might be all kinds of ways. Write down some numbers.

My point is that if no one has done it before you, it's likely it can't be done. But perhaps I'm wrong. Write down some numbers, and get Feldstein and Samwick to endorse them. If they can do so without shredding their reputations, they'll be delighted to do so.

Paul Zrimsek,

You write:

if we're going to assume that whoever has the last word must be right, what conclusions shall we draw from the fact that "A Tiny" never replied to Maguire at all, but instead changed the subject to Feldstein?

As noted, there is (as yet) nothing to reply to. TM and others have made lots of conjectures about how things might be. And that's all.

[TM]

Well, I don't think it is unreasonable for me to try and understand the model Dean Baker is using before I try to solve it. Barring a major reversal, I am going to stay with the thought that he has wrapped an elaborate cloak around the notion that the earnings yield equals the return on capital.

One intereseting consequence of that would be to note that Krugman's long exposition about projected growth being lower than past growth becomes irrelevant; my question to Dean Baker would be, if the earnings yield drives the return on capital, why would higher growth lead to a higher return on capital? One might think that it would simply lead to a larger economy with a larger capital stock earning the same return.

In which case, the place to look is the earnings yield. My quick and dirty solution - Krugman posits a world today with a PE of 20, or an earnings yield of 5%. Let the income share of capital have a one time spike of 20% (e.g, from 8% of GDP to 9.6% of GDP), and my revised earnings yield is 6%. Baker used .2% for foreign income; I'll use 0.5%. Bingo, I'm done.

And in my new world, the growth is still low, at roughly 2%; since capital gets more, labor gets less, so payroll taxes fall a bit; and the Soc Sec "crisis" is more severe.

The key insight here (if this holds up) is that growth is neither the problem nor the solution.

[A Tiny]

TM,

I don't think it's unreasonable at all for you to try to understand what model Baker's using. I wish many more people were like you.

But try to get Samwick or Feldstein to put their name on something endorsing those numbers you came up with. I don't think you can. That's my point: they understand all these issues too. But they know the implications of this kind of "solution" are so weird that (to date) they have not been willing to attach their reputations to them.

I can solve this problem too: I predict the p/e ratio will rise to 500/1 by 2060. In your face, Dean Baker! But I have no reputation to protect. Samwick and Feldstein do, which is why they have, to date, been unwilling to write down some numbers.

[A Tiny]

TM,

Also, don't give up! A couple more years of 0% stock returns and the privatization case becomes much more plausible. The irony, of course, is that with continued poor stock returns it will be politically dead.

[Jim Glass]

"Of course, maybe he and DeLong (who is sympathetic to private accounts) and PK are wrong, and TM and Jim Glass & the pseudonymous/anonymous Sullivan are right"

Now, Max, I can't speak for our host or the pseudo-anonymous, but I thought I made it clear that I have no problem with the idea that long-run stock returns will decline a couple points to like 5% real. Not in principle, for I rather think they will, nor in practice regarding the benefits they offer for private SS accounts in comparison to the sub-bond rates SS so safely assures us from now on.

My only objection was to the column-cartoon level analysis that domestic GDP growth is a straight-jacket on profit growth.

If the profits of Nokia, Ericcson, Volvo and the like were limited by Sweden's GDP trend during its bad decade there, there'd be a lot fewer cell phones in the world for one thing.

I won't even consider Lichtenstein.

[TM]

But they know the implications of this kind of "solution" are so weird that (to date) they have not been willing to attach their reputations to them.

Hmm, I have profits rising quickly from from 8% to 10% of GDP, and that is too weird? Profits rose 20% last year, unless my memory deceives me.

And in footnote 4 to the Feldstein letter, I see this:

CBO projects that the profit share of GDP will fall from 9.4 percent in 1999 to 8.2 percent in 2009 (The Economic and Budget Outlook: Fiscal Years 2000-2009, p 21).

From which I infer that a swing of 1.2 percent does not beggar the imaginations at CBO.

Now, as to why Samwick and Feldstein have been ducking the Baker challenge, you seem to know more about it than I do. However, my impression was that Samwick only came across this a month or two back - is it possible that Baker is being ignored, rather than ducked?

[GT]

Tom,

I see many differnt opinions but no rebuttal, much less a definite one, of Krugman's points. Much of Andrew Samwick's analysis is based on "if A happened" without ever explaining why we should assume A.

Jim,

Given the relative size of the US economy I suspect that yes, GDP does act as a contraint. Where do you think the investment opportunities will come from? Mexico?

[TM]

You don't see a rebuttal there? I wonder what Max saw.

Anyway, to pick up Jim's point, the World Bank reports (last page) that in 2003, US GDP was just above $10 trillion.

They also helpfully sum what they call Low and Middle income countries, and come to about $7 trillion. (Upper middle income is $1.8 trillion, so the two lower categories sum to $5.2 trillion).

Since you ask, Mexico is about $600 billion, or roughly 20% of the target market.

So, is a $5 trillion global developing market likely to be significant relative to a $10 trillion US economy? It is not clear to either Diamond or myself that it won't be.

[Paul Zrimsek]

If nothing else, the SS debate means the end of offshoring as an issue. Perhaps someday the developing world will once again be an irresistible opportunity for conscienceless American capital to make indecent profits by destroying American jobs. But for now, it can't be anything but a money pit.

[Dean Baker]

Since there appears to be some confusion on the issue, let me clear up one point in this debate -- the assumption of constant capital shares comes from the SS trustees, not me. In general, it is probably a reasonable assumption over the long-term, although there are fluctuations over the business cycle. In 1999, as at the present, we are probably near a cyclical peak in the pfoit share -- that was CBO's assessment then (and now), which was my reason for citing it in a footnote in the letter to Feldstein.

The issue of foreign profits can be easily dealt with also -- plausible numbers will not change the picture in any substantial way. I raised the rate of profit growth by 0.2 pp annually in the calculations in my letter to Feldstein. This assumption had close to 30 percent of corporate profits coming from developing countries by the end of the projection period. Does anyone think a higher percentage is plausible?

[bob]

Why not just ask naysayers if they are invested in the stock market via any vehicle. End of argument.

[TM]

Hello, Dean, thanks for joining in.

In the course of doing whtever it is I do when I am not doing this, it did dawn on me that the constant income shares assumption was not yours. Anyway, let me pick up one point on foreign income, and then present a Brisk Summary:

This assumption had close to 30 percent of corporate profits coming from developing countries by the end of the projection period. Does anyone think a higher percentage is plausible?

According to Jim Glass, the *current* share is 17%; so, if the US grows slowly while the rest of the world does not, then yes, I think the foreign share could rise to more than 30%. As I noted, the low and low-middle income countries have economies that are already roughly half the size of the US. If their growth rate exceeds ours by 2%, they will be our equal in 36 years (A cleverly disguised use of the Rule of 72).

Now, Brisk Summary: As I understand Krugman's column, he says that forecast growth is too low to justify projected equity returns of 6.5%, given a current P/E of 20.

My seemingly relevant point - "growth" is not the issue, per se. Perhaps it was a word count restriction; if he had a 701 word limit, maybe Krugman could have put the word "productivity" in his column.

But, staying with the open-ended "growth", my first point is that growth is not necessarily the problem; consequently, higher growth may not be the solution.

For example, and perhaps If we can tax a bit more of Dean Baker's time he would clarify this, my impression is that if I increased the growth in the US economy by having the labor force grow by an additional 3% per year, with no changes in the productivity assumption, the end result (as per the Feldstein letter) would be (a) higher growth); (b) the same current low return on capital; and (c) numbers indicating that stock returns of 6.5% are not achievable.

If that is true, than can we agree that "growth" is not in itself the issue? In which case, can we agree that Krugman's column spends a lot of time on the wrong issue?

Now, as to productivity growth, suppose we increase that by 1% per year. Again, if I am following the Feldstein letter, that will *not* result in an increase in the return on capital, which seems to be controlled by the initial earnings yield.

I am not banging the table on the productivity point, although I think I am right. But I am banging the table on the labor force growth. And if I am right about the productivity side then it is absolutely the case that the point of Krugmam's column, that forecast growth is too low, is misdirected. A quick excerpt:

if the economy grows fast enough to generate a rate of return that makes privatization work, it will also yield a bonanza of payroll tax revenue that will keep the current system sound for generations to come.

Alternatively, privatizers can unhappily admit that future stock returns will be much lower than they have been claiming. But without those high returns, the arithmetic of their schemes collapses.

That does sum up his side - but the rest of my response is that, having set aside growth as the issue, let's focus on that assumption of constant income shares. If, in the low-growth forecast at hand, we increase that profit share, several things can happen:

(1) Labor gets less, payroll taxes fall, and Soc Sec looks even worse;

(2) there is room to jiggle the earnings yield so that a return on capital of 6.5% becomes the norm.

Now, my approach does *not* contradict Dean Baker's point that the assumptions in the Soc Sec projections must be changed, so it is not a rebuttal in that sense.

However, my suggestion would crater the Krugman position that (a) higher growth is always the answer and (b) any reasonable forecast that delivers a reasonable return on capital also "saves" Social Security.

Now, I am just typing this without even having had a cup of coffee, so don't anyone go singing it like Gospel. But I would be intrigued to pry a reaction out of Dean Baker on this, if, as noted, we can further tax your effort here.

It seems to me that his response could be something like, "Well, as I wrote to Feldstein, the assumptions need to be changed". But would Mr. Baker also care to defend Krugman's point, that any reasonable changes that save corporate returns also save Social Security?

[GT]

Tom,

You raise some good points on the issue of growth. To be honest I should sit down with a spreadsheet and play with the numbers. But chances are I will wait for others to do so. Lazy me!

On foreiegn investment I would love to see any calculation that took into account politcal and exchange rate risks. A lot of people invested in mexcio and lost a lot of meny after the 1994 devaluation. Same in SE Asia in 1997, and Brazil and Russia in 1998, Argentina in 2001. And so on. Most dollar investments in these countries are government bonds. Which brings up the other risk, default.

So I don't see the $5 trillion figure as very useful. In includes large number of basket cases and eve investing in the two biggest markets, India and China, has many risks. Any foreign investment that would significantly increase the expected returns would require a very large portion of all investments to be made in pretty risky countries. What do you do when there is a mjor crisis in one of them?

That is why I wrote that I saw no rebuttal. I read, from you and Andrew, some very smart comments of how things could work differently but no indication if this is realistic or not. I'm not sayiong you are wrong. Just that I don't see you are right, not yet at least.

You've spent some time on this and seem to know more about it than most on the blogosphere. Why don't you email Brad or Krugman? Both read and participate in some of these debates and maybe you will get a good response from them.

[Dean Baker]

On the point on redistribution -- of ocurse, you can have a redistribution from labor to capital which would allow corporate profits to grow more rapidly than overall GDP. Again, this contradicts the trustees assumptions and it implies slower wage growth (possibly much slower, if you assume a big enough shift). The point here is that we have to agree on a set of numbers when we discuss SS and accept their implications. If we want to bring different numbers to the debate, that's fun too, but we should be explicit about the implications of the alternative asusmptions (e.g. an unprecedented shoft from labor to capital).

On foreign earnings -- I encourage the enthusiasts to do a little arithmetic. Most of the foreign earnings from U.S. corporations at present are from rich countries (Canada, Europe, Japan), so having 30 percent come from DEVELOPING countries is in fact a huge increase. And, to make your hurdles even higher, I assumed that the return on physical capital was twice as high in developing countries as in the U.S.(20 percent as opposed to 10 percent), this is probably a reasonable ballpark number today, but presumably China, Mexico, etc. will still not be devel;oping countries with very hich risk associated with investment in 2080. So, I was being very generous to get my 0.2 pp of additional profit growth.

[TM]

Thanks, GT; since I have briefly caught the eye of Dean Baker himself, I may come yet up Aces.

If I can ask onbe quick questioin of the Dean, here they are:

(1) Am I right in my interpetation that the earnings yield becomes the forecast return on capital?

(2) Am I right in saying that growth may solve the Soc Sec problem without imcreasing the return on capital?

For example, if we allow much higher immigration, the labor force and capital employed will grow. This will be a higher growth scenario that "saves" Social Security, but it won't improove the return on capital.

Put another way, do the "optimistic" scenarios that save Social Security also show a high return on capital, or is Krugman linking things that aren't really linked when he says that:

if the economy grows fast enough to generate a rate of return that makes privatization work, it will also yield a bonanza of payroll tax revenue that will keep the current system sound for generations to come.

If the growth is due to higher productivity, but the gains to productivity are captured by labor, than there are plenty of ways for more growth to save Soc Sec, but no way for more growth (sas modeled here) to improve returns on capital.

If I am right about this, big parts of Krugman's column are wrong (as I suspect you are beginning to suspect).

Cheers.

[GT]

Tom,

Correct me if I am wrong. If I understand your argument is is this:

Krugman said A. You agree that A is true historically but think that it need not be true in the future. You have provided several arguments why A could in fact be B in the future.

The probelm is that, even if you are correct logically, you have provided no reason why we should think B is more likely than A. If years ago we were debating what the long term P/E ratio should be and you had argued that it could be higher due to changes in the perception of relative risk you would have been right. But would you have had enough information at the time to bet on that? I suspect not.

I think this is important because it goes to the heart of Krugman's argumenmt, as well as that of many others. To say that something can happen is not the same as saying something may happen. SS is supposed to be, its only reason to exist is to be, the risk-free portion of any retirement. It is that part of our retirement money which we know we will always have. For many it is the ONLY retirement money they have.

So it makes no sense to gamble that things will be so different in the future from what they were in the past. Maybe stocks will grow fast even if GDP slows down. But should we bet on it? Particularly when there is no need? That is the crucial point IMO.

[TM]

Well, Krugman's argument as I understand it is that privatizers are playing a shell game - low growth forecasts in which Soc Sec go bust are also forecasts in which the retrun on capital and subsequent share price perfromance are dismal; high growth forecasts in which Soc Sec does well are also high return forecasts in which share prices do well.

Obviously, this has strong intuitive appeal - most folks will mentally substitute "growth below expectations" for "low growth" and "growth above expectations" for high growth, in which case we are all in agreement.

However, in fact, the high growth forecasts also embed (depending on the modeling assuptions, obviously) a dismal return on capital - for example, higher immigration "saves" Soc Sec, but does not improve the return on capital.

Or, low growth scenarios could be great for investors, but ghastly for Soc Sec - imagine a world where Reps restrict immigration and roll back current labor protections (I do not advocate this, but am simply identifying a possiblity). In this scenario, we could easily see low labor market growth leading to low GDP growth. However, the reduced bargaining power of labor *may* mean that more of the gains of productivity go back to "capital", i.e., the income share of capital rises. (I have scare quotes there because, in a different model, we would want a third category, which I will call "rent-seekers", who hire capital and labor, and eke out a profit. Those are the folks who would really benefit if labor lost bargaining clout - they would hire at capital at previous rates, refuse wage hikes as productiviy rose, and earn more for themselves. But in a two category scheme, we will put them on capital's team).

Does that seem hopelessly farfetched? It sounds like the Kerry campaign to me, with Benedict Arnold CEOs engaging in outsourcing with helpless workers unable to benefit from the recovery.

And in that world, Soc Sec has financing problems, but personal accounts do great. No shell game at all.

Now, this does not mean that, because there is no shell game being played, that privatization is a great idea - there are plenty of reasons to oppose it. The shell game theory is simply not one of them.

Hmm, evidence of "shell game" argument from Krugman's column:

...the numbers the privatizers use just don't add up.

Let me inflict some of those numbers on you. Sorry, but this is important.

Schemes for Social Security privatization, like the one described in the 2004 Economic Report of the President, invariably assume that investing in stocks will yield a high annual rate of return, 6.5 or 7 percent after inflation, for at least the next 75 years. Without that assumption, these schemes can't deliver on their promises. Yet a rate of return that high is mathematically impossible unless the economy grows much faster than anyone is now expecting...

"mathematically impossible". Really?

They can rescue their happy vision for stock returns by claiming that the Social Security actuaries are vastly underestimating future economic growth. But in that case, we don't need to worry about Social Security's future: if the economy grows fast enough to generate a rate of return that makes privatization work, it will also yield a bonanza of payroll tax revenue that will keep the current system sound for generations to come.

Alternatively, privatizers can unhappily admit that future stock returns will be much lower than they have been claiming. But without those high returns, the arithmetic of their schemes collapses.

Look, he is arguing a shell game. But in fact, the growth assumptions are a separate issue; the real question is, what return on capital assumptions are made or embedded in things like the income share of capital, or the capital intensity required as the economy expands.

And it may well be that the Soc Sec trustees have not fully thought this through - in that respect, Dean Baker has clearly made a very helpful suggestion. However, Krugman has headed to a wrong conclusion - high growth scenarios that save Soc Sec do not necessarily produce high stcok market returns, and low growth scenarios that bankrupt Soc Sec do not require low stock market returns.

OK, I am repeating myself now - someone in the above is the draft of a new post.

[Jim Glass]

"...if the economy grows fast enough to generate a rate of return that makes privatization work, it will also yield a bonanza of payroll tax revenue that will keep the current system sound for generations to come."

Illogical. This is mixing clam shells and eggplants.

Private accounts are *not, not, not* about SS's "solvency" -- as the latest White House press briefing at last makes explicitly clear.

Private accounts are fundamentally neutral regarding SS's finances, as Mr Anonymous Senior Adminstration Official said out loud so the press could hear. Perhaps they can have some modest benefit -- but that is not their point so the argument above is irrelevant.

The point of private accounts is that Social Security, after 60 years of "succeeding" and becoming hugely popular as a result of giving everybody **huge positive returns**, is now making an historic shift into giving workers ever-worsening *negative returns*.

As a matter of policy it is a bad thing to make workers poorer on a lifetime basis.

As a matter of politics, one doesn't have to be a genius to see that SS's doing so will make it as unpopular among them as its former positive returns made it popular for earlier generations -- with predictible result.

Private accounts put positive returns back into SS to avert these consequences. It is becoming harder and harder for me to take seriously the progressiveness of "progressives" who oppose these goals.

Sweden has private accounts in social security.

Swedish social insurance policy is too right-wing for US Democrats.

[GT]

Jim,

Do those return calculations include the disability and survivors benefits?

[Joe Mealyus]

"The point of private accounts is that Social Security, after 60 years of "succeeding" and becoming hugely popular as a result of giving everybody **huge positive returns**, is now making an historic shift into giving workers ever-worsening *negative returns*."

But you ignore much of the true returns people recieve from Soc Sec. Everyone (or most everyone) has parents and grandparents and has recieved utility from seeing them better off. And it is not just a private good (or an arrangement by which private goods are transferred), it is a pure public good. You can not be excluded from the fact that Mr. Jeter down the street was forced to cough up 12.4% and is now not going to be indigent after his retirement - and your fellow citizens won't be taxing you to pay him welfare benefits.

[Joe Mealyus]

"As a matter of politics, one doesn't have to be a genius to see that SS's doing so will make it as unpopular among them as its former positive returns made it popular for earlier generations -- with predictible result."

I suppose you're right, but since the rise in the payroll tax to 12.4% hasn't most of the "popularity" of Soc Sec simply been a combination of political inertia plus the fact that old people actually make it to the polls?

And is it really "negative returns" (think about those cookies Grandma used to bake for you, and weep!) that are going to make Soc Sec unpopular? I say it's welfarization. As Robert Brown points out in a different thread, it's already been heavily (and increasingly) welfarized by playing with the benefit schedules and with taxes, and more welfarization of some sort is clearly going to be the Dem magick fixit. You have to appreciate that the Dems believe that the R's are the ones who are out to destroy it.

Anyway, perhaps the decreasing popularity on both counts means that defeating Bush now, besides being largely irrelevant (if the program can't start till 2009 because of the deficit, who's to say it can't start till 2029 because of the deficit?), is going to be one more gift to the Republicans.

[TM]

I have a new post up wrapping this up. I await the verdict of the judges (and we still have roster spots on our team).

http://justoneminute.typepad.com/main/2005/02/lets_help_paul_.html