Arnold Kling  

Social Security Privatization

PRINT
Government vs. Private Debt... Fine-tuning Immigration...

In this piece, I argue that the "transition cost" argument against privatizing Social Security is bogus. I also make my usual case against what I call the Stock Market Scenario. Finally, I conclude by saying that privatization is the ultimate lockbox.


The real benefit of Social Security privatization is never discussed by either side in the debate. That benefit would be that Congress could no longer arbitrarily raise benefits and thereby increase the obligations of future taxpayers. Today, Congress can pass a benefit increase, such as the recently-enacted prescription drug benefit for Medicare, without specifying how to pay for it. The incentive will always be to expand benefits to the point where the system is bound to collapse.
...
Thus, what the Left sees as the biggest drawback of privatization -- the "transition" from pay-as-you-go to fully-funded -- is what I see as its most attractive feature. Privatization is the ultimate "lockbox" for Social Security, in that it would keep Congress from expanding future benefits without setting aside money to pay for them.

For Discussion. What are the real issues at stake in Social Security privatization, as opposed to the issues that will be raised politically?


Comments and Sharing


CATEGORIES: Social Security



TRACKBACKS (2 to date)
TrackBack URL: http://econlog.econlib.org/mt/mt-tb.cgi/45
The author at Winds of Change.NET in a related article titled Carnival of the Capitalists writes:
    Jay Solo's Carnival of the Capitalists has just made its scheduled arrival at Winds of Change. Please prepare for boarding, and ensure that your mental baggage is securely stowed in the overhead compartments.... [Tracked on January 26, 2004 1:16 AM]
The author at Deinonychus antirrhopus in a related article titled Carnival of the Capitalists writes:
    Carnival of the Capitalists is up at Winds of Change. Of personal interest to me is Arnold Kling's post on [Tracked on January 26, 2004 9:27 AM]
COMMENTS (75 to date)
Mcwop writes:

The single biggest issue at stake in Social Security privatization is the effect on the current federal budget.

Without the payroll taxes flowing into the treasury, the budget deficit would be larger. Even though the Budget Enforcement Act of 1990 requires that the deficit be calculated excluding Social security, the press reports the lower deficit numbers. For example, this year's deficit will be around $550+ billion excluding Social Security payroll tax revenue.

Furthermore, the debt added as a result of absent payroll tax revenue will require immediate interest payments, as it cannot be deferred through accounting as the Social Security "IOU's" are. This will add $250 billion to government bond issuance.

The crux of this problem is that politicians lose a lot of financial leverage. Example: how can any serious candidate roll out an expensive program like Medicare drug benefits, or universal healthcare without Social Security payroll tax revenue flowing into the treasury? Politicians will actually have to resort to options such as cutting spending (as tax hikes will never do the job).

There are ways to privatize and deal with this, but this issue is what scares politicians, and others that desire more government involvement in our lives.

Eric Krieg writes:

"Bush hating economist Brad DeLong"...

I LOVE IT!

The best thing the government has done to move along privatization was to start sending out "statements" showing how much you have paid into SS and what benefit they project you will get at retirement.

Seeing how much I've paid in has been very sobering. It is a LOT of money, money that I would much rather have in a 401(k) working in the market.

I have a feeling that my reaction is not the one the Clintonistas had in mind when they implemented the statements!

Eric Krieg writes:

"As I see it, there is a plausible scenario in which stocks yield a 7 percent return for many years going forward. That scenario is one in which Moore's Law, biotechnology, and nanotechnology prove fruitful. In such a scenario, the economy will see growth closer to 7 percent than to the 2 percent commonly forecast"

I want to know from Arnold, and any other degreed economist, what they think of Milton Friedman's thesis that full SS privatization, in and of itself, will raise the growth rate of the economy.

Eric Krieg writes:

Or was is Marin Feldstein's thesis?

Eric Krieg writes:

Google sez:

"Professor Martin Feldstein of Harvard University, for example, estimates that "the combination of the improved labor market incentives and the higher real return on savings has a net present value gain of more than $15 trillion, an amount equivalent to 3 percent of each future year's GDP forever."

Eric Krieg writes:

NBER working paper 5281:

A funded social security retirement program would imply a larger capital stock and a higher level of real income than an unfunded program that provides the same level of benefits. The transition from an unfunded program to a funded program that does not reduce the benefits of existing retirees or the present value of the benefit entitlements of existing employees would, however, require substituting explicit government debt for the equally large implicit debt of the unfunded program. This paper shows that such a debt financed transition from an unfunded program to a funded program is not just a change of form without economic effects. Such a debt financed transition would raise economic welfare if three conditions are met: (1) the marginal product of capital exceeds the rate of economic growth; (2) the capital intensity of the economy is below the welfare maximizing level (i.e., the marginal product of capital exceeds the appropriate consumption discount rate); and (3) the rate of economic growth is positive. Illustrative calculations based on U.S. experience since 1960 suggest that the present value of the gain from a debt financed transition to a funded program would substantially exceed the current level of GDP. More explicitly, even with a relatively high real consumption discount rate of 4.4 percent, the present value gain would be about 1.5 dollars per dollar of current net social security wealth or about $17 trillion.

Eric Krieg writes:

What does PK think about all this?

http://www.wws.princeton.edu/~pkrugman/socsec.html

I think that, as usual, Prof. Kling needs to staighten Prof. Krugman out.

Eric Krieg writes:

Goggle rules.

These guys disagree with Arnold completely...

http://www.findarticles.com/cf_dls/m1093/n2_v41/20513693/p1/article.jhtml?term=

Boonton writes:

***********
The incentive will always be to expand benefits to the point where the system is bound to collapse.
***********

Odd that SSI has not collapsed and doesn't appear close to collapse (something like a few % points increase in the tax or raising the retirement age a few years restores it to stability). Why is this?

Because the incentive to raise benefits is no different than any other incentive for the gov't to spend. It gains a politician nothing to vote for giving people free healthcare 50 years from now. Bush's drug plan only gains him to the extent that someone's grandmother gets something for her drug costs right now!

Sure Bush could enact a law saying $5B will go for healthcare now and $500 Trillion in 2050. What will that gain him? If the $500T entitlement is repealed in 2005 no one will care because no one is losing anything they already had.

Mcwop writes:

SSI won't “collapse”, but the general budget will be severely lacking those payroll taxes that it now taps. The general budget is what will “collapse”, or in better terms “be crowded out by Social Security and Medicare”.

The taxes and premiums collected for Medicare as a whole are already less than its expenditures (To the tune of about $78 billion, or about a fourth of the budget deficit).

Hiking payroll taxes will make it difficult to raise general revenues for discretionary spending in the future. Raising the retirement age simply makes the program a bigger rip-off for a large % of the population, considering most take benefits early at age 62 (about 70% of people). I would agree to increase the age if I can divert 4% of my payroll to a private account. That would reduce the governments 12.4% take to 8.4%.

Nice article slamming both parties on Social Security

Boonton writes:

It's interesting that the 'collapse' is demonstrated by combining both SSI and Medicare yet the privatization chiefly concerns SSI. Medicare is a very different animal than SSI and should be treated as such.

Basically you are saying that having a larger portion of the population retired puts a larger burden on the working population to supply the goods and services. That's only a moderately interesting population.

Whether or not SSI is a 'rip off' is an odd question. Since the benefits are coming from a payroll tax they are linked to the underlying growth (or lack of growth) of the overall economy. A privatized SSI will also be linked to the overall growth of the economy. The difference will be that some people will win big and others will loose.

Mcwop writes:

Boonton, you summed it up pretty well. I threw in Medicare because the budget effect being felt now is similar to the future effect to be experienced by SSI.

Regarding some people winning big and losing big under privatization, that is already the case under the current system. Though as time passes, more will lose than win.

Boonton writes:

SSI works nicely as an insurance system. If you lose big in your 401K you can fall back on SSI to keep you from having to try to find full time work at 75. Otherwise SSI's benefits are extra icing if you did very well in your 401K.

Medicare is quite a different animal since is it tied to purchasing a good for a specific population. Since health costs are rising in general, it would become a problem even if the population of elderly remained constant.

In a '401K only' world and a 'SSI only world', the only thing that really matters is economic growth. A growing elderly population means their share of the economic pie must be cut into smaller pieces...it cannot mean anything else. This really isn't a bad thing, it says a lot that our economic strength is such that we can have 40% or more of our population not in the labor force.

Eric Krieg writes:

>>Otherwise SSI's benefits are extra icing if you did very well in your 401K.

Well, I would do EVEN BETTER in my 401(k) if I could AT LEAST get my hands on my portion of the payroll tax (an extra 6.2% of my salary would really juice up my 401(k) nicely).

Here is another reason to like full, immediate privatization: the SS related bonds would compete for funding with all the other elderly entitlements, ESPECIALLY Medicare.

For the 8 people who think that Medicare should be transformed into a means tested catastrophic health insurance program, this is a probable outcome.

Lawrance George Lux writes:

The real Issues involved with privatization can be presented as these:
a) It ties the Social Security system to the performance of the Economy, to the injury of the Economy. Recessions will incite reduction of payments, so Consumer Demand will be further depressed.
b) It will not provide a 'Lockbox', unless it is completely privatized. A mixture will lead to escalating benefits, with diversion of significant taxes to personal investment.
c) The cost of Investment will likely absorb 5% of total Investment and earnings, as Investors pursue short-term advantages rather than long-term strategies.
d)The politicians will simply increase spending on curtailed programs of the Social Security system (threat to security).
e) It will increase taxation without added benefit, because of the scale of privatization.
lgl

Eric Krieg writes:

>>In a '401K only' world and a 'SSI only world', the only thing that really matters is economic growth

And according to a REALLY SMART ECONOMIST, growth would increase greatly if SS were privatized.

Boonton writes:
Well, I would do EVEN BETTER in my 401(k) if I could AT LEAST get my hands on my portion of the payroll tax (an extra 6.2% of my salary would really juice up my 401(k) nicely)

Well yea but 401K's as a whole could only do as good as overall economic growth. The same is true of revenue raised by taxes. The returns that you could have made would not be the same if SSI didn't exist at all (as opposed to just you, Eric, being given an exemption).

Here is another reason to like full, immediate privatization: the SS related bonds would compete for funding with all the other elderly entitlements, ESPECIALLY Medicare.

I suggest you read Arnold's piece.

Under 100 percent privatization, or under any transition from today's "pay as you go" system to a fully-funded system, today's payroll taxes would not go to pay current benefits, but instead would be used to fund the retirement benefits of today's workers. As a result, the government would have to find another source of revenue to pay current retirees. In practice, this means that the government would issue debt, putting the burden of repaying that debt onto future taxpayers.

15 years from now the gov't will still be carrying the debt from privatization which would compete with attempts to borrow more to fund Medicare.

And according to a REALLY SMART ECONOMIST, growth would increase greatly if SS were privatized.

Why?

Eric Krieg writes:

Boonton, I gave you the links above! Look at them!

Martin Fedstein says that full and immediate privatization would increase economic growth by 3 percentage points PER YEAR! His thinking on why is in the link.

Now, you can agree or disagree with the man. But by Paul Krugman and Arnold Kling's arguments, he is a degreed economist and REALLY SMART, so his opinion is automatically more valid than yours.

>>15 years from now the gov't will still be carrying the debt from privatization which would compete with attempts to borrow more to fund Medicare.

Which is my point. That competition is GOOD for you and me. It restrains the choices that we have in reforming Medicare, and forces us towards market solutions rather than more socialism.

Daniel Lam writes:


The 'transition costs are bogus' argument would be flawless if privatized Social Security accounts were allowed to hold nothing but Treasury bonds, as current workers are effectively forced to do in their 'accounts' under the existing system. Their claims on future government revenue are not marketable, and they are prevented from shorting Treasuries by credit and liquidity constraints.

If, upon privatization, they elect to move some of their funds out of Treasuries into other investments, the interest rate on government debt would have to rise by enough to induce someone else to hold it. Of course, other investment projects will be getting cheaper capital as a result of this shift, so it is by no means clearly a bad thing overall. But it will be an extra cost for the government.

Lee A. writes:

But you haven't tallied everything. Don't ignore the ongoing transactions costs of (1) search and information, (2) bargaining and decision, and (3) policing and enforcement (all as defined by Coase) incurred by consumers in the market for financial instruments, and do not forget insurance and hedging. Perhaps you will want to ignore these costs because they are mostly non-monetized, or because the SEC (i.e., policing and enforcement) may some day do its job properly, after being properly funded by the crooks in Congress. (But that would be "political economy", and don't hold your breath.) It depends on how you price your time, or perhaps, other people's time. After all is said and done, the net gain of privatization may be closer to zero.

Still, those who wish to opt out should have that freedom. So long as they don't come whining when the next bubble breaks.

Boonton writes:
Boonton, I gave you the links above! Look at them!

Martin Fedstein says that full and immediate privatization would increase economic growth by 3 percentage points PER YEAR! His thinking on why is in the link.

Maybe I missed it but I didn't see any link detailing his thoughts on the matter. Only the statement that he believes growth would increase by 3%

Monte writes:

Boonton,

Here is a more detailed account of Feldstein's position on social security privatization. It's a bit dated, but still pertinent for discussion.

BTW, Chili transitioned to a pension savings account (PSA) system some years back (1981??). Does anyone have any current performance data on this?

mcwop writes:

a) It ties the Social Security system to the performance of the Economy, to the injury of the Economy. Recessions will incite reduction of payments, so Consumer Demand will be further depressed.

Isn't social security already tied to the performance of the economy, which dictates how much tax revenue flows into the treasury?

Eric Krieg writes:

I don't think that SS privatization should be implemented in a vacuum. It should be part of a broad program to make the US economy more competitive. In the end, it is probably most conservative to assume that private accounts won't grow any faster than the overall economy. Thus, increasing economic growth would ensure the success of privatization.

In addition to privatization, we need policies such as tort reform, market oriented health care reform, and a conversion of all defined benefit pensions to defined contribution ones (thus shifting the burden of retirement from businesses to individuals).

Boonton writes:

Of course, in one very real sense Keynes was correct; there is no such thing as savings. Whether people have 401K's, SSI or both the fact remains that tomorrows retired people will have to live off of the goods and services produced by the economy of tomorrow.

In a 401K world, what would happen if a larger portion of people were retired? To fund their retirement they would have to sell their 401K assets (i.e. stocks). To get the funds for a good retirement, those assets would have to be purchased by workers who are currently building their own 401K's. If there are fewer workers then each worker would have to devote a higher portion of their income to their 401K, otherwise retirees would have to accept a smaller retirement benefit.

Its basically the same problem we have now. The only difference is that instead of the retired benefit being roughly uniform it will be highly variable since some will have lost badly in their 401k and others will have been amazingly lucky.

Eric Krieg writes:

B, I am more interested in "second order effects". For instance, there will be a mass psychological change when people have to rely on themselves rather than the government for their retirement. Larry Kudlow talks about this often when he talks about the "investor class" that has been created by 401(k)s.

I think that you are partially correct about public pensions having similar problems to 401(k)s, in that there has to be someone buying if everyone is selling, or else valuations will go into the crapper.

We are having the opposite problem now. Everyone is investing in their 401(k)s, but there are only a limited number of stocks out there. This is one reason why P/E ratios have become so lofty.

The same dynamic is going on with real estate and college tuition. If everyone is trying to own a home and get a college education, but there are a limited supply of houses and colleges, then the price of both is going to go through the roof.

Ironically, immigration, high productivity growth, and working women all make the problem worse. If supply is constrained, but the number of people trying to buy goes up, or those people have more income to throw at the desired product, the price will go up strongly.

I can see the opposite dynamic occuring with stocks when the Baby Boomers retire.

Boonton writes:

I'm reading through Eric's link:

Our current Social Security system is acting as a drag on economic growth in two important ways. First, the payroll tax distorts the supply of labor and the type of compensation sought by workers. These losses are inevitable because of the low return implied by the pay-as-you-go character of the unfunded Social Security system. Second, the system reduces national savings and investment.


Here is how he attempts to estimate the impact of the tax versus a regular 401K type account:

A simplified example will indicate the magnitude of the tax wedge implied by the Social Security program. Consider an employee who contributes $1,000 to Social Security at age 50 to buy benefits to be paid at age 75. With a 2.6 percent yield, the $1,000 grows to $1,900 after the 25 years. In contrast, a yield of 9.3 percent would allow the individual to buy the same $1,900 retirement income for only $206. Thus, forcing individuals to use the unfunded system dramatically increases their cost of buying retirement income. In the example, a funded plan would permit the individual to buy the same retirement income with a 2.5 percent contribution instead of the 12 percent payroll tax. The 9.5 percent difference is a pure real tax for which the individual gets nothing in return.

The alternative 9.3% yield is coming from comparing the average wage growth since 1960 to the average return on corporate capital:

"The average growth of real wages since 1960—2.6 percent—can serve as a reasonable estimate of what an unfunded Social Security program can yield over the long-term future. In contrast, the real pretax return on nonfinancial corporate capital (i.e., profits before all taxes plus the net interest paid) averaged 9.3 percent over the same period.[4] "

The fundamental error here is that this is a meaningless comparision. First of all, by driving the return on Treasury Bonds lower SSI boosted other types of return. As Arnold showed in his article, it is impossible for an element of the economy to grow faster than the entire economy forever. If one individual happened to have been allowed out of SSI in 1960, he might have very well seen a return of 9% rather than 2.6%. If the hundreds of billions of SSI dollars started flowing into corporate capital it is impossible that return would remain constant.

Boonton writes:

If there's too many people buying stocks right now then the easiest solution is to simply issue more stocks. There's already a sense of having to look out for yourself among the general public. 401K's already exist and the gov't is bending over backwards to induce people to use retirement savings accounts. In a slightly different manner home ownership is another form of gov't favored retirement accounts since many people use their home equity to help them out during their retired years.

Eric Krieg writes:

>>If there's too many people buying stocks right now then the easiest solution is to simply issue more stocks.

And if there are "too many" people buying houses, then the answer is to build more houses. Look at the news. Housing starts are at record levels. But the housing industry simply cannot keep up with demand. Ditto the stock market.

>>There's already a sense of having to look out for yourself among the general public.

A "sense" is not good enough. People need it burned into their cerebral cortex.

>>401K's already exist and the gov't is bending over backwards to induce people to use retirement savings accounts.

It would be better if they didn't have a choice to do so. It should be either 401(k) or no retirement.

>>In a slightly different manner home ownership is another form of gov't favored retirement accounts since many people use their home equity to help them out during their retired years.

True, and it is the consequences of high home ownership that gives me pause. The phenomenon of personal bankruptcy over the last generation is largely caused by the high cost of home ownership. Government programs that try to increase the rate of home ownsership do nothing more than increase home demand when supply is limited. This, in turn, does little more than increase home prices.

This is great if you are elderly and bought your home before the price was bid up. If you are at the leading edge of the baby boom, you made out like a bandit also. But for the rest of us, it is not so clear that we are better off as a society. Certainly, there wouldn't be as much personal bankruptcy if banks still required 20% down, for example.

I fear that privatizing SS could do something similar to stocks. It would be great for today's elderly, because the stock market would go through the roof in the short term. It would be great for the leading edge of the baby boom. And if Feldstein is right, it would be great for all of us. But if he isn't right, it might not be so great for some of us (but then again, neither is the current system).

Eric Krieg writes:

What investment is going to make a more productive use of the money: a corporate bond or a T-bill?

That is the question that must be answered if we are going to figure out if Feldstein is right or not.

If you are like me and think that the majority of what the Feds do is unproductive, you welcome the fact that T-bills will need to pay a higher interest rate. That means that interest costs will compete with Medicare for scarce tax dollars.

There is a theory out there that one reason federal spending is so out of control right now is that interest rates are so low. Why cut spending when you can float bonds at 4%?

Eric Krieg writes:

I wonder if economists like Arnold and Feldstein have though about the effect of privatization on the interest rates that T-bills pay, and how so many financial instruments like CDs and mortgages are tied to those rates.

Privatization will have an impact on interest rates. I just don't know if the impact would be up or down.

Boonton writes:
What investment is going to make a more productive use of the money: a corporate bond or a T-bill?

It doesn't really matter. By holding T-Bond rates lower than they otherwise would have been, investment that would have otherwise gone to T-Bonds goes to other forms (such as corporate bonds). On the flip side, removing SSI's effect of holding T-Bonds rates down will make corporate investment less attractive.

No new investment is generated in the overall system.

Daniel Lam writes:
It doesn't really matter. By holding T-Bond rates lower than they otherwise would have been, investment that would have otherwise gone to T-Bonds goes to other forms (such as corporate bonds).

This is true if you assume that the amount of government borrowing is exogenous. But Congress may respond to cheaper funding by borrowing and spending more.

Boonton writes:
This is true if you assume that the amount of government borrowing is exogenous. But Congress may respond to cheaper funding by borrowing and spending more.

Perhaps but it may be better to model gov't spending as exogenous. Borrowing costs fell in the 90's yet gov't spending was rather stable...even falling to some degree. Spending is exploding now even though rates are not all that low compared to the 90's.

Eric Krieg writes:

>>Spending is exploding now even though rates are not all that low compared to the 90's.

What the hell are you talking about? The 10 year T-bill was UNDER 4% exactly 1 week ago! That is lower than just about any time in the 1990s.

In fact, I can remember back in 1996, on some taking head show on CNBC, a debate between Larry Judlow and "The Investment Biker" (what's his name? Jim something).

The Investment Biker said that the 30 year interest rates would never fall below 6%. What is the rate now? Around 5% or so?

It is only because of the most recent recession that rates are as low as they are. Coincidently or not, spending has surged during this era of historically low rates.

There might be a connection.

Eric Krieg writes:

Spending during the 1990s was held in check by one man: Newt Gingrich.

Okay, ir was more than one man. It was a Republican Revolution.

What the hell happened to us Republicans? Slick Willey beat the snot out of us in 1996, that's what. If you can't beat the New Democrats, join the New Democrats.

Dubya is such a good New Democrat that he has driven them to extinction!

Bernard Yomtov writes:

Eric,

Better citations please. Google doesn't say anything. It just provides a list of sites.

The Feldstein paper does not say privatization would increase growth by 3% a year. That would be a ridiculous claim.

Eric Krieg writes:

Bernard, nothing more than the abstract is available on the net for free.

I e-mailed Feldstein about it, to see if he would send me a copy. No reply. I bet he's not big into e-mail.

I think technically Feldstein was saying that, in 1994, if SS had been privatized in 1980, the increase in GNP growth at that point moving forward would be 3% for all time. There would be a transition period.

And, of course, he had 3 qualifiers: (1) the marginal product of capital exceeds the rate of economic growth; (2) the capital intensity of the economy is below the welfare maximizing level (i.e., the marginal product of capital exceeds the appropriate consumption discount rate); and (3) the rate of economic growth is positive.

Eric Krieg writes:

>>Better citations please. Google doesn't say anything. It just provides a list of sites.

I want to make Boonton work!

Eric Krieg writes:

Bernard, sorry, I will do better.

FYI, I googled: feldstein "Social Security" privatization "economic growth"

I got 1490 hits!!! This was hit #1:

http://www.socialsecurity.org/pubs/articles/mt-art-tanner1.html

"Professor Martin Feldstein of Harvard University, for example, estimates that "the combination of the improved labor market incentives and the higher real return on savings has a net present value gain of more than $15 trillion, an amount equivalent to 3 percent of each future year's GDP forever."

Eric Krieg writes:

>>The Feldstein paper does not say privatization would increase growth by 3% a year. That would be a ridiculous claim.

I am pretty sure that is what it says. Like I said, I e-mailed Feldstein to get his take, to make sure that he still believes it. No response.

Bernard Yomtov writes:

Eric,

I think that what the item you quote is saying is that there is, in effect, a one-time jump of 3% in GDP, not that the growth rate goes up by 3%. So if GDP were going to be $10 trillion this year it would instead be $10.3 trillion, but then grow at the normal rate thereafter.

eric krieg writes:

What don't you understand about "an amount equivalent to 3 percent of each future year's GDP forever."

Ray writes:

Boonton,

“SSI works nicely as an insurance system. If you lose big in your 401K you can fall back on SSI to keep you from having to try to find full time work at 75.”

First of all, Boon’s initial argument is basically that SSI isn’t broke enough to fix. Proof that it is; the fact that this hypothetical 75 year old worker would have to find a job if the only thing he has to depend on now is SSI. Some have argued that SSI wasn’t meant to live off of but a privatized plan could support a retired worker with no other income.

This is also misleading in another way. This hypothetical 75 year old worker would have been investing in his retirement plan over the course of decades and a privatized social security plan would not allow said worker to “bet” the whole thing at any time, much less right before retirement. So even conservative investing in the market over decades yields 5+% in the least which is much better than what SSI does at its best.

All this even misses the more salient but overlooked point, SSI is socialism pure and simple. I know we’ve bred several generations that have come to expect a certain amount of socialism in their lives to the point where such things are now a “right” but the bottom line is that we’re talking about my money. Remember that concept, individual rights to personal property? Of course we can’t just rip the nipple out of America’s mouth but we can surely give some degree of control back to the individual and get a better return at the same time.

So Boon’s argument that SSI isn’t broke enough to fix simply holds no water from an economic freedom stance or even just talking about the return we get on each dollar.

Ray writes:

“A growing elderly population means their share of the economic pie must be cut into smaller pieces...it cannot mean anything else.”

I don’t get this part of Boon’s post. Why must their share of the economy be cut into smaller pieces? First of all, in a free market, the economic pie is not a fixed number of pieces. Secondly, it is the money that works, not their actual labor. So the retiree’s money can be hard it work and contributing to the economy even when grandma is a convalescent down at “the home.”

Or maybe I just missed something there.

Ray writes:

Boonton,

“In a 401K world, what would happen if a larger portion of people were retired? To fund their retirement they would have to sell their 401K assets (i.e. stocks). To get the funds for a good retirement, those assets would have to be purchased by workers who are currently building their own 401K's. If there are fewer workers then each worker would have to devote a higher portion of their income to their 401K, otherwise retirees would have to accept a smaller retirement benefit.”

This is just wrong and really misses the mark in both basic economics and basic investing knowledge.

First of all there is no reason to assume that people are going to have to, or even want to liquidate their entire portfolio immediately upon retirement. Even assuming that this was the case (which it wouldn’t be) it would average out over time and there would be no great burden upon the fund managers to suddenly buy hundreds of millions of dollars of liquidated retirement packages. Securities that were known to be heavily held by pension packages would already have that guaranteed “sell” volatility factored into them and there would be no great fluctuation in the overall market.

What would actually happen, is most plans would have a certain % earmarked for liquidation (maybe a laddered bond position or just the most volatile allocation in the portfolio) and the rest of the portfolio would be designed to provide an indefinite amount of dividend income.

The 2000 market plunge was not the result of blue chip and brick & mortar stocks (dividend paying stocks) suddenly devaluing but it was the result of a certain over blown sector. The more traditional sectors of the stock market did not fall proportionately with the market.

What the non-investors around here are assuming is that these pensions would have a handful of mutual funds tied up in a few industrial sectors that would be constantly at risk of crashing through the floor and this simply is not so.

“The only difference is that instead of the retired benefit being roughly uniform it will be highly variable since some will have lost badly in their 401k and others will have been amazingly lucky.”

Another display of many misconceptions about investing in general and how such plans would be administered. No one would lose “badly” and the only ones that could be considered lucky would be the ones who took no interest in their investments and just happen to make the right choices.

Boonton seems to think that every worker would receive a copy of the Pink Sheets in with their payment stub and then off to the races with the grocery money.

dsquared writes:

Estimating the effects on behaviour of a tax wedge is a balck art; there's no valid way of doing it statistically, because you are comparing to a counterfactual. Feldstein is entitled to his opinion, but he would have to admit that his numbers are very sensitive to assumptions (Feldstein was also responsible for an *awful* piece of "SS is bust" scaremongering from the CEA once, so I'd be reaching for the salt cellar). It would not be hard to come up with similar breathless estimates for the potential benefit to the US economy from a National Health Service, and if I can be bothered I'll do it.

I thought Arnold's piece was good and made important points. I agree with the point made above that the movement back and forth between "SSI" and "SSI and Medicare" is far too quick and particularly confusing given that Medicare is the vast majority of the problem.

Finally, the way I would have written this piece would be to point out that the "transition problem" is the same problem as the "bankruptcy of SSI" and both are bogus. Because there are enough future taxpayers to pay the coupons on Arnold's debt, there are enough future taxpayers to keep the transfer payments flowing. This entire cock-up is FDR's fault for misleading the American public by dressing up a sensible transfer program as an investment scheme, but that doesn't mean we have to stay wrong.

Eric Krieg writes:

>>Feldstein is entitled to his opinion, but he would have to admit that his numbers are very sensitive to assumptions

Do you people read? Feldstein laid out those assumptions!

"(1) the marginal product of capital exceeds the rate of economic growth;

(2) the capital intensity of the economy is below the welfare maximizing level (i.e., the marginal product of capital exceeds the appropriate consumption discount rate); and

(3) the rate of economic growth is positive. "

Now, Mr. Brilliant, Degreed Economist, tell me what you think of those assumptions.

I'm sick of people giving me these nitpicky, speculative criticisms (that means you, Boonton). There is some substance to Feldstein's thesis, and I want to know what real economists think of it.

Eric Krieg writes:

>>This entire cock-up is FDR's fault for misleading the American public by dressing up a sensible transfer program as an investment scheme,

This gets back to psychology and second order effects. Privatizing Social Security and abolishing defined benefit pensions changes peoples' outlooks from a socialist one to a capitalistic one. If you are interested in fostering personal responsibility, privatization is a good way to do it.

Boonton writes:
“A growing elderly population means their share of the economic pie must be cut into smaller pieces...it cannot mean anything else.”

I don’t get this part of Boon’s post. Why must their share of the economy be cut into smaller pieces? First of all, in a free market, the economic pie is not a fixed number of pieces. Secondly, it is the money that works, not their actual labor. So the retiree’s money can be hard it work and contributing to the economy even when grandma is a convalescent down at “the home.”

Simple mathematics: Say you represent the economy's output in 2050 as $100. Say the portion of nonworkers to workers in 2050 is 40%. That means the portion of output going to nonworkers (whether provided by the gov't thru SSI like programs or by their liquidation of 401K type accounts or both) is limited by definition to the economy's output. Say nonworkers receive $60, by definition this means workers will only receive $40. Since workers produce 100% of the output, how can this be?

In a pure free market model the workers would have to be willing to buy the non-workers 401K assets for $60. In a pure SSI type model it would mean gov't would have to tax workers 60% (or gov't could tax the workers a smaller amount and sell them bonds).

This is a simplified model (it ignores the fact that not people gradually retire and it ignores international trade and other things) but it demonstrates the essence of the problem cause by an aging population. This problem exists regardless of the retirement system being used by the country.

It shouldn't be surprising. Imagine a primitive society with one eldery person and a dozen younger people. Its easy for them to each contribute a little to keep the 'wise old man' around. If there's 6 old people and 6 workers then the young are going to have a tougher time (and/or the old people will not have it as good as when there was just one old man).

Bruce Cleaver writes:

Nothing substantive to offer here, but seeing the volume of responses here leads me to quote the old NBC newsman John Chancellor...

"Would you like to hear my most controversial editorial, the one which always got me the most mail ? Ladies and Gentlemen, Social Security. Thank you and good night."

Boonton writes:
First of all there is no reason to assume that people are going to have to, or even want to liquidate their entire portfolio immediately upon retirement. Even assuming that this was the case (which it wouldn’t be) it would average out over time and there would be no great burden upon the fund managers to suddenly buy hundreds of millions of dollars of liquidated retirement packages. Securities that were known to be heavily held by pension packages would already have that guaranteed “sell” volatility factored into them and there would be no great fluctuation in the overall market.

That doesn't change the fact that in a '401K only' world a population that is gradually aging will put downward pressure on the price of their 401K assets as they liquidate their 401K's to fund their returement. Yes the process will not happen all at once, it will happen gradually and be averaged out over many years. That doesn't change the fact that as the liquidation of 401K's gradually increases the only way for prices received not to also gradually decline would be for someone else to gradually buy more and more of the assets. Since the portion of workers is declining in an aging society the only remaining possibilities would be if workers were willing to allocate more of their incomes towards their own 401K's or if the rest of the world was willing to buy more assets.

“The only difference is that instead of the retired benefit being roughly uniform it will be highly variable since some will have lost badly in their 401k and others will have been amazingly lucky.”

Another display of many misconceptions about investing in general and how such plans would be administered. No one would lose “badly” and the only ones that could be considered lucky would be the ones who took no interest in their investments and just happen to make the right choices.

It's not a misconception to note the following:

1. A huge portion of plans are not managed well. They are often highly overinvested in a single stock (usually the employer). Even diversified plans are having a huge portion of their returns siphoned away by administrative fees charged by mutual funds (way too often the funds with the poorest record of beating the market charge the highest fees).

2. Rules that would prohibit these abuses are unlikely to ever be accepted on the dubious grounds that they are restricting the investor's 'freedom to choose'.

3. Even if you exclude the two above abuses, you can't deny that even 'well managed' accounts will show a diverse array of returns...some well above the average and some well below it. That is unless your version of privatization would require everyone to invest in exactly the same thing (like T-Bonds)...in which case you've accomplished nothing but retain the current SSI system with massivly increased administration fees.

Boonton writes:

Feldstein appears to have also assumed that the marginal product of corporate capital is fixed and higher than the average growth of wages. Some problems here:

1. Corporate capital is risky, for every corporate investment the succeeds there's ones that fail. So one would expect the average return to be higher.

2. Putting money into the stock market or bond markets is not the same thing as putting it into corporate capital.

3. Even if SSI funds were directed towards corporate capital via privatization a surge in such investment would have to lower returns. When you talk about something as massive as SSI contributions, you can't assume the rest of the economy will stand still while that multi-hundred billion dollar train is redirected.

IMO, this pretty much makes Feldstein's 3% increase in GDP conclusion very suspect.

Eric Krieg writes:

>>IMO, this pretty much makes Feldstein's 3% increase in GDP conclusion very suspect.

I think that you would have a transition period where you would not get a 3 % point increase, yes. That transition period might be quite lenghty. I don't know, maybe 20 years? But get beyond that transition period, and you would get much increased GNP.

Look, every economist I have every been taught by has complained about the low savings rate of the US population. This fixes that problem overnight! And best of all, it gives every working person in this country a stake in the economy. That is a wonderful thing, and will transform the political landscape, perhaps more than the Great Depression did.

Eric Krieg writes:

>>Even if SSI funds were directed towards corporate capital via privatization a surge in such investment would have to lower returns.

It really depends on what companies do with the money, right?

Boonton writes:
It really depends on what companies do with the money, right?

Not really. Imagine all the possible investments a corporation can make and rank them in order of their return:

Investment A1 --- 100000%
Investment A2 --- 99999%
....
Investment Am --- 9%
Investment Am1 --- 8.9%
.....
Investment An --- 0.1%

Suppose a corporation can borrow at 9% because that's the market rate for corporate bonds at the moment. The corporation(s) has (have) an incentive to borrow and make all the investments from A1 through Am.

Say there's an influx of money into corporate bonds (imagine Congress directs a small portion of SSI funds towards private accounts) which causes rates to drop to 8.9%. The additional investment & its return is only the Am project.

If a corporation today has some magic idea that will pay off huge returns then it can already raise the revenue for it. Barring magic ideas one would expect an influx of a lot of money towards corporate investment to lower the marginal return.

Eric Krieg writes:

Feldstein Assumption Number 1: the marginal product of capital exceeds the rate of economic growth

Do you people read?

Boonton writes:
Feldstein Assumption Number 1: the marginal product of capital exceeds the rate of economic growth Do you people read?

Assumption Number 1.5

The marginal product of capital is fixed at a rate above the rate of economic growth.


This is where I believe his model breaks down. Of course corporate capital before taxes has a high rate of return. Its supposed to be higher because corporate capital is:

Riskier than the entire economy
Must pay off more than just the investor.

Eric Krieg writes:

Yeah, well, our recent experience with the rate of return on corporate capital hasn't been so good. You know, that whole 1990's internet bubble thing.

B, thanks for a substantive criticism of Feldstein. That wasn't so hard now, was it?

I'm thinking that I can find some real numbers on the after tax return on corporate capital, so we can test Feldstein's hypothesis since he wrote the study in the early ninetees. I wonder how it has fared since then.

Eric Krieg writes:

What could we do with this?

http://www.economagic.com/em-cgi/pdf.exe/fedbog/aaa

Boonton writes:

Eric,

I think you're missing the guts of my criticism...it suffers from the same problem that most SSI criticisms suffer from.

It typically goes like this:

If I could only have invested my SSI tax money in the stock market in 1980 onwards I would have had more money today than all my social security checks put together!

What's missing is the problem of what would have happened to the stock market if hundreds of billions of SSI dollars were directed towards it. Prices would have gone up and when prices go up returns go down. You can't assume the return on stocks, bonds, corporate capital or anything will remain constant while you direct serious funds towards it.

Eric Krieg writes:

>>You can't assume the return on stocks, bonds, corporate capital or anything will remain constant while you direct serious funds towards it.

I agree with this criticism, in the short term at least. God knows that is what has happened in the market for housing and a college education.

But those markets are also illustrative of (I think) Feldstein's point. Even though the housing market has been bid up, and the gains going forward will not be anywhere near as strong as the gains in the recent past, no one is talking of there being NO gains in the future.

Ditto for a college education. The price of a college education is probably getting to the point where the gains in income of the degree are a wash. But would you advise anyone to skip getting a degree? And you can see certain "second order effects" from getting a college degree that don't directly show up as increased income.

And these markets show that the transition period can be very long, and during that transition period, gains can be very high.

Ray writes:

Boonton,

Actually, most plans, a huge majority, are managed quite well (I am a licensed securities broker still even though I'm not currently practicing). I won't fault you for assuming a large number of poorly managed plans as the only ones that make the news are the losers.

Secondly, a SSI type of pension plan would not be set up like what you are picturing as a typical stock portfolio. Most private retirement portfolios contain less than 40% stock (equities).

What the industry is moving to more and more, for ease of use and it is safer from a legal standpoint, are wrap accounts. So your portfolio isn't eaten up with numerous management fees but instead, you'll get an overall wrap fee of about 1% to 2%. If you're wondering how the brokerage makes their money by doing that, it's a bulk buy. The typical wrap account only has X amount of funds and outside managers to choose from. If you want to go buy a bucket full of your favorite stock, you have to go do that on your own.

And rules prohibiting risky trading in retirement funds are already in place. This is one reason why 401 rollovers are so popular these days. If you have a 401 yourself, read the legalese in the contract; some - not all 401's have a clause allowing "non-hardship withdrawal."

Something like that would probably be, and wisely so, verbotten in an SSI pension plan.

The most popular complaint about most 401 plans is not poor managment, as in fraud and incompetence but a lack of control and small returns. This lack of control is not a bad thing as most people are horrible investors and the small returns are only a product of the necessary caution most fund managers take in running a retirement fund.

The typical fund is not allowed by law to buy certain kinds of securities such as bond funds that contain non-rated municipals or to be over a certain %allocation in one sector. And companies loading their 401s up with their own stock is coming to a screeching halt after Enron. (I wonder if former Enron advisor Paul Krugman is the one who advised them to allocate so much of their stock for this purpose?)

Bernard Yomtov writes:

Ray,

It's not clear to me what you're referring to in your post, but 1-2% for portfolio management is absurdly high. Vanguard and other index funds typically charge substantially less than 1%, sometimes as little as .25%.

Ray writes:

Bernard,

We're talking about a complete portfolio, not an index fund or any single fund. 1% to 2% for a wrap fee on a managed account is good.

Depending on how much you have invested in that portfolio and your personal risk tolerance, typically less than half of your portfolio will contain equities. Over 50% equities is generally considered a little risky as of course the stocks follow the market whereas the various fixed instruments can at least lock you in a decent rate.

So you establish a good, conservative "nest egg" with X amount in fixed instruments and maybe even, depending on the size of your portfolio, some managed futures and things such as this (managed futures typically don't go over 5% of a portfolio).

Then, to dial in a little capital appreciation, you may choose to allocate X amount to some higher growth potential. Higher potential is also more loss potential and in a retirement plan, the most aggressive is still very conservative by anyone's normal measures.

This is why I was pointing out earlier that no one would even be able to "bet" their whole pension on Enron or tech stocks or whatever. There are already rules governing such activity on the existing qualified retirement plans (qualified means tax deferred).

If you trusted your retirement plan to one (or even a few) index fund such as Vanguard, you would be very much at risk of extremely poor returns and possibly losses. Just throwing together a basket of mutual funds is not financial planning and is not what anyone is suggesting we do with SSI.

Also, under a truly managed portfolio, if a wrap fee was not implemented and you had to pay all of the individual fund fees (.45 for this one, .77 for that one and so on) it would quickly add up to more than 1% or 2%.

This is why I said it is a bulk buy. The brokerage is not charged the full fund fees by the funds themselves in agreement that the brokerage will carry their fund as part of their XYZ retirement plan.

So decent managed plans can stay in the positive in the worst years and vastly outperform individual mutual funds every year. Paying 1.5% or so on such a portfolio is then a small price to pay.

Ray writes:

I should also mention that funds like Vanguard may be a good addition to an overall plan but you are paying such a low rate because you get what you pay for - no management.

The Vanguard managers may take care to stock their fund with solid companies but there is still a limit to what an index fund can do. Remember, an "index" fund by definition is going to either follow the market or a certain sector, depending on what it is "indexed" to. This brings about the scary scenarios that Boonton and so many others are assuming would happen to a privatized SSI.

I forget all of the actual rules but Peter Lynch has an interesting story about the negative side of the limitations on mutual funds. Google Lynch and you'll probably find the story but in short, Lynch saw the '87 crash coming on (quite a few people did actually) and could not, by law, reconfigure his mutual funds fast enough. The limitations on how and when a manager can move things in and out are there for good reason but in a fast moving market, those same limitations can bite back. Thus an entire retirement plan consisting of a few basic funds would indeed produce, as Boonton said, some losers and some lucky gainers.

Thankfully however, this is not the case with privatized SS plans.

Bernard Yomtov writes:

What don't you understand about "an amount equivalent to 3 percent of each future year's GDP forever."

The part about how a 3% increase means the growth rate is going to go up 3%. That is what you are claiming:

"I think technically Feldstein was saying that, in 1994, if SS had been privatized in 1980, the increase in GNP growth at that point moving forward would be 3% for all time. There would be a transition period."InsertTextHere

You are wrong for two reasons.

First, the paper (I coughed up the $5 and read it) says nothing about GDP growth rates. I don't know where you got that 7% business, but it wasn't from Feldstein. Tell us please.

Second, "3% forever" is not the same as a 3% increase in growth. Consider these numbers:


100.000 103.000
102.500 105.575
105.063 108.214
107.689 110.920
110.381 113.693

The second column is 3% higher than the first "forever," yet the growth rates are the same.

Bernard Yomtov writes:

Ray,

The problem with your argument is that there is no evidence managed funds perform better than the indexes, and quite a bit of evidence the other way, which is what anyone familiar with finance theory would expect.

I agree that it is unwise to put all your investment funds into equities, but for the portion that does go there an index fund is usually the way to go. Even if you find the very rare manager who can outperform the market (and remember, we're not talking about a few years, but about a working lifetime) the outperformance has to be around 1 to 1.5% per year, compounded. That's huge.

Of course you can avoid even the index fund fees by building and holding a diversified portfolio of equities yourself. Once you get to about twenty or twenty-five stocks, in different industries, you've gotten most of the diversification benefit.

Monte writes:

Ray,

“So decent managed plans can stay in the positive in the worst years and vastly outperform individual mutual funds every year. Paying 1.5% or so on such a portfolio is then a small price to pay.”

Bernard is right to point out that index funds, net of fees and expenses, consistently outperform actively managed portfolios on a risk-adjusted basis, making them preferable to managed plans that charge a wrap fee of 1.5% or more. That’s not to say, however, there’s no role for active portfolio management, even if we subscribe to the notion that markets are efficient with limited opportunities for earning abnormal profits. There will always be a large contingent of investors willing to defer asset allocation decisions to the professional for a reasonable fee. What’s more, your point regarding “the worst years” is well taken, as the evidence shows that managed portfolios can minimize losses that the passive index fund is otherwise vulnerable to in economic downturns.

But if SSI ever becomes partially or fully privatized, I’d prefer to manage my own assets and avoid the 1.5% wrap fee, which, charged to accumulated annual savings of 5% on a salary of $50,000 compounded over the course of 40 years (assuming a 5% return) approaches $40k. Pocket change to some, maybe…

Eric Krieg writes:

All right, Bernard. You have the paper and I don't. What does it say?

In particular, Feldstein trots out this "$15 trillion" number that he says is the NPV gain of privatization.

There is no way that $15 trillion is 3% of GDP, especially in 1994 or whenever that paper was written. That's like 3 times GDP in 1994. So you can see why I thought that it was a 3% point increase in GDP per year, forever.

dsquared writes:

Eric; I have a problem with 1) , the assumption of a rate of return on capital that is fixed no matter how much you throw at it. In more conventional models, there are diminishing returns to a fixed factor.

Eric Krieg writes:

>>1) , the assumption of a rate of return on capital that is fixed no matter how much you throw at it.

I don't think that that is being assumed by Feldstein. Is he were, why would he state that he is assuming that the marginal product of capital exceeds the rate of economic growth?

The assumption is probably that, after privatization, the marginal product of capital is STILL higher that the rate of economic growth. You are in a better position than I am to say if this is a good assumption or not.

Monte writes:

AARP PPI briefs on Sweden and Chili sizing up social security reforms in each country. Key learnings appear to confirm the belief by some that transition costs could be high and that the public’s general lack of investment skills places a severe constraint on the system’s ability to guarantee an adequate level of retirement funds for future beneficiaries.

Other issues notwithstanding, these two could easily be overcome by:

1) Setting aside a large cash reserve prior to implementing reforms to meet the pension obligations of the first wave of boomers.
2) Education, education, education!

So far, the overall performance of the two models has vindicated privatization advocates (albeit Sweden’s model needs more time to fairly assess). I, for one, am encouraged by the results and would take advantage if offered a choice.

Bernard Yomtov writes:

Actually, Feldstein does assume a fixed rate of return to capital. He also seems to assume that a privatized program would not reduce savings outsidethe retirement program. This also seems dubious to me, given that a fair amount of individual savings is intended for retirement, even if it is not part of a 401(k) or the like.

Essentially Feldstein derives an equation for the present value gain from privatization and then multiplies it by what he calls Net Social Security Wealth (NSSW) - the PV of future benefits less future tax payments.

The parameters include things like certainty equivalents on investment returns and intergenerational discount factors, not exactly precisely known numbers.

The multiplier he gets is very sensitive to his estimates of the parameters. He generates a range of PV estimates running from .6 to 14.5 times NSSW. The high end comes to more than $150 trillion, the low is just under $7 trillion. Feldstein does not, in this paper, cite a $15 trillion figure, though he does cite an $18 trillion figure under certain parameter values, using no risk adjustment.

I think it's fair to say that Feldstein expects a gain, but this strikes me as more of a "first shot," than a realistic estimate. It is interesting that he does not give these sorts of numbers in later non-technical articles on SS reform.

I am not sure what this article's status as a "working paper," rather than a published article, implies about the weight one ought to give the results.

Eric Krieg writes:

Did anyone read the WSJ editorial page commentary regarding the trade deficit, the aging population, and how they are related?

The thesis is that the developed world is aging, the developing world is young, and it makes sense for the old to buy "stuff" from the young by selling them assets.

I thought that the article had relevance to the debate over privatization. If there is a question as to who will buy our assets when we all retire, the answer is the Indians and the Chinese.

Comments for this entry have been closed
Return to top