« Intellectual Property... | Main | Brunnermeier and Morgan on Clock Games »

January 30, 2005

Outlines of a Social Security Deal?

Joshua Micah Marshall reports:

Talking Points Memo: by Joshua Micah Marshall: January 30, 2005 - February 05, 2005 Archives: So there you have it. Rep. Harold Ford (D) of Tennessee, fainthearted no more.

I think that the outlines of a potential deal on Social Security that would be good for the country are becoming clear:

  1. Shift responsibility for maintaining actuarial balance off of the Congress and onto the Social Security Administration--have it gradually raise (and lower) the retirement age or the benefit-rule bend points in order to keep the system in projected balance.

  2. Uncap FICA and apply it to all wage income in order to top-off private add-on accounts for the poor and boost benefits for widows.

  3. Make enrollment in private accounts automatic (it's done automatically on your 1040) but voluntary (you can fill in an extra form to get the money the IRS earmarks for your account back as part of your refund).

  4. Use the government's existing Thrift Savings Plan as a vehicle for managing private add-on accounts--and keep its choices restricted: churning and extra administrative costs caused by asset shuffling are not your friend.
  5. Mandate that in fifteen years a commission consider and recommend whether or not two percentage points of FICA should be diverted and added to the add-on accounts as a forced savings program.

This would satisfy optimists who believe SSA's projections are much too pessimistic: if they are right, it would impose no benefit cuts. This would satisfy pessimists who worry that there is no mechanism to finance the existing level of benefits: if they are right, the SSA will then cut benefits. This would satisfy Congress: if there are benefit cuts, their fingerprints aren't anywhere nearby. This would satisfy believers in boosting national savings: the revenues from uncapping FICA and the money flowing into private accounts from people's tendency to choose the default option will boost national savings. This would satisfy those scared that private accounts would be churned and looted by unscrupulous brokers: the TSP is a good operation that provides powerful protections. And this should satisfy believers in ultimately moving Social Security from an unfunded defined-benefit to a funded defined-contribution model: fifteen years from now--when private accounts will have fifteen years of reliable performance as a track record--shifting a chunk of FICA to forced savings will look much more attractive.

Posted by DeLong at January 30, 2005 10:28 PM

Trackback Pings

TrackBack URL for this entry:

Listed below are links to weblogs that reference Outlines of a Social Security Deal?:

» Catching my eye: morning A through Z from The Glittering Eye
Here's what's caught my eye this morning: Lileks and Power and Control comment-fisking DU. Brad DeLong gives the highlights of a Social Security reform plan I'm completely in favor of (not surprising since I've been saying nearly all of these... [Read More]

Tracked on January 31, 2005 06:00 AM

» Let the Looting Begin from Mises Economics Blog
Harold Ford of Tenn., via Talking Points and Brad DeLong, points the way to a likely consensus on Social Security reform (politics being the art of a compromise to tax you more, and all that): in exchange for private accounts,... [Read More]

Tracked on January 31, 2005 06:36 AM

» A real deal on social security from Master of My Domain
Brad DeLong presents the outlines of a deal on social security that could (and should) really happen [Read More]

Tracked on February 1, 2005 09:05 AM


Okay, but I don't know how I feel about this aqua thing. Or is that the acid talking?

PS Just kidding.

Posted by: Robin the Hood at January 30, 2005 10:41 PM

That's basically giving into a the biggest portion of the Bush plan, but doing it years down the line?. Or so it seems.

I know you've said that you believe there's some merit in the Bush administration claims, but do you really think diverting some of the FICA tax is that worthy a goal? What's the biggest difference between doing that now and doing it then?

Posted by: Brian at January 30, 2005 11:04 PM

Finally a plan we can agree on. I particularly like this part, though I would modify it somewhat: "have (the SSA) gradually raise (and lower) the retirement age or the benefit-rule bend points in order to keep the system in projected balance". I would leave retirement age alone and address future benefits alone. Because more and more people are taking early retirement, tinkering with retirement age doesn't get you far, in the end it comes down to the level of benefits at time of initial eligibility anyway.

Under this system the pessimism of the Trustees actually works for me, depression of numbers on the front end creates upward pressure on benefits at the back end (that would be this Boomer: 19 years five months from retirement!!). And if I am really confident in my numbers I just take my money back as part of my refund.

This proposal advances the dialogue by light-years. It starts with the fixed point of currently promised benefits and works from there rather than arbitrarily tying those benefits to some other benchmark.

I am confident that America will grow faster in the future than it has on average in the past. (Well sue me for loving my country.) And this plan would allow me to put my money where my mouth is, and redistribute some income at the same time.

Kudos to Brad.

Posted by: Bruce Webb at January 31, 2005 12:03 AM

A couple comments.

On Sam Williamson's SocSec.net list today, an interesting piece about Germany's pension system, during the 1920s through 1950s, was posted. Basic summary was that following catastrophic shocks:

"social insurance has proven the most reliable
means of securing one's retirement because governments possess the moral/political obligation and taxing power to reestablish
social-insurance payments after an economic or political shock and to protect payments against inflation."

Theory: private investment can deliver an equity premium that gives you a reliable exponential gain in wealth *most of the time for many of the people*. But the tail of losers may be long enough and big enough that universal guaranteed benefits are always desirable, and catastrophic events which might occur only once every couple or several generations can wipe out unlucky cohorts. That is, there is a meaningful risk premium which makes private investment unattractive for those dependent upon a predictable return of capital for basic life support.

What are the merits of the theory; and where does that lead for an outline of a safe "risk adjusted" SS system?

Second comment. Fifteen years from now is right about when SS stops subsidizing the rest of government, and perhaps 25 years before the trust might run out. That might not be the best time to start diverting revenue from the SS trust, and might be too late to scale down traditional SS benefits to cope with trust exhaustion. If we'd built up mandatory private SS savings for fifteen years, it might be reasonable to then start to phase in cuts to cover diverted funds. But, if there is no mandatory private saving, it does not seem reasonable to abruptly change benefit formulas then.

There would probably be a fairly strong constituency at that time to divert funds - perhaps so strong that 2% wouldn't satisfy the hungry beast. And, odds of one of those nasty catastrophies by that time is probably low, while odds of such a catastrophe seems higher going out to around 2040, if only because of demographics. So, having a commission designated to deal with that question 15 years from now might not be prudent.

How can you reconcile the theory that a funded private investment system can normally give you a higher rate of return and probably higher economic growth, while also acknowledging the theory that only the government, backed up by a pay as you go tax, can guarantee benefits during predictable infrequent catastrophes, as well as for the tail of the distribution in unlucky cohorts?

The natural political urge would be to let people put all of their money into the magic money tree, once they've seen the magic working. But if the magic sometimes fails, that may be a slippery slope.

I'm rambling, but what about the alternative of public investment of SS trust fund assets, using the same limited set of asset classes as would be available for personal accounts? This is probably even more DOA than Bush's plan, but how would that compare with privatization?

Posted by: ChasHeath at January 31, 2005 01:55 AM

Hi Brad:

Your proposal is sound, and actually somewhat similar to one of the three proposals from the late, unlamented social security reform commission.

All three commission proposals were, of course, rejected because they did not satisfy Bush's fantasy-land requirements, which were:

(1) No tax increases (your contains them)
(2) No benefit cuts for future or near retirees (yours contains at least the possibility of that)
(3) It must include private accounts (your proposal contains that)

I guess the point is that rational arguments have been tried with this administration, and even tried on this issue, and typically they are shouted down by partisans, or the President himself, saying something is strange as "I won't negotiate with myself". I'm still unclear as to why presenting something realistic is "negotiating with yourself". That, undoubtably, is because I'm such a horrid and patriotic person who disagrees with the President - and I've learned, over the last couple of years, that this offence is nearly as bad as genocide.

Reading the commission report, (which can be found online at http://www.csss.gov/reports/Final_report.pdf ) can be most interesting, in a "I can't believe this is so silly" kind of way. If you don't believe me, take a look at the cutsie visual metaphors (like on page 37) and the inclusion of discredited data (such as that African Americans get short-changed by the current system) and unrealistic assumptions (such as that a private account can reliably provide a net return of 3.5 percent over inflation - that is NET not GROSS!).

A pushover commission stacked with proponents of private accounts was unable to provide what the White House wanted.

If you want to have a good laugh, read the commission's minutes and see how many howlers you can count per sentence. The average seems to be about 3.5 (Net, of course).


Posted by: SZ at January 31, 2005 04:01 AM

Interesting ideas, but I do not wish the Social Security administration to determine retirement age and benefit levels. I would prefer Congress do this, for there is at least potential accountability.

Posted by: anne at January 31, 2005 04:24 AM

"...the revenues from uncapping FICA and the money flowing into private accounts from people's tendency to choose the default option will boost national savings."

AAAaaahhhhh!!! Never, ever say "will boost national savings." That's like saying "I'll be right back" in a spooky movie. Knock on wood just as fast as you can.

We are really good at shifting national savings through various government programs, but boosting savings is another matter.

Posted by: kharris at January 31, 2005 04:33 AM

Time for the reliable "is there one major policy initiative that this Administration has not completly screwed up?" (or something like that). Leave it alone and let those who vote in 30 years figure how how to handle it.

Posted by: JM at January 31, 2005 05:00 AM

sigh. this notion may be brilliant but it would suit the politics of Mars better than America. What's going on here is a debate over whether the Administration succeeds in dismantling the Democratic base, and, secondarily, whether the Administration can take more from the poor to give to the rich. Proposing this sort of plan unwittingly helps them, and of course nothing of this kind would ever be passed.

Posted by: rod at January 31, 2005 05:08 AM

I keep hearing people talk about doing #2-removing the cap as something that could do alot. I'm fairly skeptical about that point. Since SS benfits are still assumed to be capped at 90K or whatever they will be an added incentive for those who can (and once you get past say 500k it would be most everyone) to chnage how they receive thier income. As if the small capital gains tax wan't enough to do so already we'll have more people taking salary in options and in kind transfers. And I think that could make corporate governance problems even worse.

Posted by: Rob at January 31, 2005 05:20 AM

Should we change the color of the dollar to aqua while we're at it?

Posted by: praktike at January 31, 2005 06:07 AM

The only problem with this plan is that it is logical, and therefore has no chance of survival with this administration.

Posted by: weinerdog43 at January 31, 2005 06:08 AM

I really dislike the uncapping from a political standpoint - despite the whining from our wingers (leading me to believe that most of them are employed in pizza delivery or have serious emotional problems and should seek help) once you start getting into the 200K plus wage range your SS payments are just not a big deal. Six percent at 180K, 3 percent at 360K....a good luxury auto salesman can make >300K I bet.

You can write them off as assuaging your social conscience or giggle over them as not having exactly zero value "well if I get sued by the Santorums I'll retire on SS and pet food..."

But once they are attached to every dollar you make you won't think of them as charity or cheap insurance. And the cycle will start over...

Leave SS alone and push the discussion back to Bush's Credit Card fiscal irresponsibilty re. the General Fund.

Posted by: a different chris at January 31, 2005 06:47 AM

ChasHeath wrote, "Theory: private investment can deliver an equity premium that gives you a reliable exponential gain in wealth *most of the time for many of the people*."

Well, yes, an exponential gain, but what's the exponent?

As William Bernstein (efficientfrontier.com) pointed out (click on my name for link) in his essay "The Heisenberg Equity Principle":

"In short, **the aggregate national investment return will be approximately the same no matter what the overall stock/bond mix of the capital markets.** To the extent that debt tends to decrease agency conflicts, a small nod may go to an increase in the overall debt/equity ratio. If everybody issues/invests in stocks, then stock returns must fall to the aggregate return rate. Which may actually already have happened. If all of the nation’s pension funds and newly-privatized social security accounts shifted to stocks, they most decidedly would not obtain the historical 7%-8% real return." [Emphasis in original]

Posted by: liberal at January 31, 2005 06:48 AM

Question: who appoints the members of the SSA ?
And are they subject to Senate confirmation ?

If you pass the responsibility to the SSA, and let
Bush appoint the members, my bet is his first pick
would be Grover Norquist. And then it would go
downhill from there ...

Posted by: Richard Cownie at January 31, 2005 07:05 AM

Why would people be inclined to take the default option for a private account? H&R Block is already constantly advertising that they will do all the paper work to save you up to $10,000 with the Earned Income Tax Credit. And they will do you the favor of giving you a Refund Anticipation Loan (TM) so you don't have to wait two weeks for your electronic refund. Wouldn't that be great to add on a few thousand more to the loan.

Why would one think people are too stupid or too lazy to fill out a form? It takes only seconds on a computer.

And if you made the contribution tax deferred as an incentive (and I don't think you are proposing that because that would be neutral for national savings) it would mostly benefit the wealthy.

Posted by: JackM at January 31, 2005 08:09 AM

This point:

3) Make enrollment in private accounts automatic (it's done automatically on your 1040) but voluntary (you can fill in an extra form to get the money the IRS earmarks for your account back as part of your refund).

Sounds a lot like the "Libertarian Paternalism" advocated by my former business school professor Richard Thaler. The idea is that you can improve a host of public policy outcomes and still respect the dignity of free adults to make their own choice if you give people control over their lives but make the default option on every choice the "eat your vegetables" option.

If people are lazy and irresponsible they'll be saved from themselves by being forced to do the mostly right thing. If people are diligent and responsible they'll be saved from government meddling by having the freedom to decide that the mostly right thing isn't always the totally right thing.

Posted by: sd at January 31, 2005 08:14 AM

I'll join in objecting to the actuarial movement of the "retirement age"--the costs of the policy and the disconnect between the public policy and the private (e.g., university, factory, service industry business) rules of retirement will make adjustments impractical.

For instance, Bruce Webb is, by my rough calculations based on his post above, slightly less than a year older than I am. Yet his scheduled retirement will be three years before mine, by virtue of his being roughly six months on the "older" side of the 65/67 split.

This is fine with me; I plan on working until I die. But my ability to do that will be impacted significantly by any "mandatory retirement" rules of the firm or firms for which I am working at the time.

Actuarial juggling will produce a gap between when the government says you may retire and when the company says you must. And the argument that one should simply accept the reduced payments of "early retirement" (if available) is disingenous at best.

Posted by: Ken Houghton at January 31, 2005 08:16 AM

Step 1: Get a "reasonable deal" on the table

Step 2: Persuade some Democrats to sign on

Step 3: Get bill to conference committee

Step 4: Jack up opening paragraph; remove all text; substitute the Radical Right's preferred wealth transfer language

Step 5: Send W out to the husting to cry far and wide that "Democrats are going back on their word and torpoeding 'reform'"

Step 6: Sit back and watch your money work for you. If your income is over 500,000/year, that is.


Posted by: Cranky Observer at January 31, 2005 08:38 AM


Employers Can Get Medicare Subsidies for Lower Benefits

WASHINGTON - The Bush administration has touched off a furious debate with new rules allowing employers to collect billions of dollars in federal subsidies for prescription drug benefits less generous than what many retirees were expecting under the new Medicare law.

In theory, those retiree benefits should be at least equal in value to the new Medicare drug benefit. But that will not always be the case, according to Medicare officials, labor unions and specialists in employee benefits.

In comparing retiree benefits with Medicare, the administration said, many employers will be able to ignore Medicare's catastrophic coverage, which helps people with high drug costs and accounts for about one-fourth of the annual value of the standard Medicare drug benefit, $300 out of $1,220.

Final rules for the new program were published Friday in the Federal Register. The new drug benefit becomes available next January.

In issuing the rules, Dr. Mark B. McClellan, administrator of the Centers for Medicare and Medicaid Services, said the federal subsidies would reverse the erosion of retiree health benefits and enable employers to "offer high-quality retiree coverage at a much lower cost." To qualify, Dr. McClellan said, employers must provide coverage "as good as or better than" the standard Medicare drug benefit.

But JoAnn C. Volk, a health policy analyst at the A.F.L.-C.I.O., said, "The rules allow an employer to get the subsidy for a benefit that is less valuable to retirees than what they would receive if they signed up for the Medicare drug benefit and the employer dropped coverage altogether."

Retirees can sign up for Medicare drug coverage if they think it is better than an employer's plan. Employers get no subsidy for such retirees. But it may be difficult for beneficiaries to compare the options available to them, which are likely to have different premiums and co-payments and to cover different medicines....

Posted by: anne at January 31, 2005 08:42 AM

JackM wrote, "Why would people be inclined to take the default option for a private account?"

Who knows? But contrary to your post, there's been a lot of research on this topic, and that's what a lot of people do.

Posted by: liberal at January 31, 2005 08:54 AM

Benefit cuts and retirement age increases become readily possible, but once adopted limits to Social Security will come to be repeatedly expected. This will severely limit support for the program.

Administrators rather than elected officials should not be responsible for a program in which public accountability is essential.

There is a tax increase suggested, at a time when the public has repeatedly opted for no tax increases, and Republicans can not be expected to go along. Democrats have been much harmed by suggesting tax increases.

Posted by: anne at January 31, 2005 08:59 AM


Why a lot of people think using the Thrift Savings Plan as a model for Social Security is a bad idea.

Posted by: Susie from Philly at January 31, 2005 09:00 AM

Can we please stop using the oxymoron "mandatory private savings". It's fine when the Administration wants to mangle language and meaning (that's their strategy), but here we should know better.

Posted by: peBird at January 31, 2005 09:16 AM

How about a 50% inheritance tax on all inheritances from 100k to 1,000k; 75% above 1,000k. No exceptions or deductions. That should maximize each generations' desire to better itself through individual effort.


Posted by: Cranky Observer at January 31, 2005 09:22 AM

Since the current SS setup is only questionably in trouble, it doesn't need much messing with. That said, the most straight-forward adjustment, if one IS needed, would be to look at the original situation of the population in terms of retirement age and life expectancy when SS was introduced. Since we have expanded the life expectancy, if it's needed to retain solvency until employee/retiree ratios balance better in the future, what's wrong with considering a further increase in the retirement age? I don't view this as a fundamental reduction in benefits if the average period of retiree benefit stays roughly fixed. For those that do, lets consider the SS Plus idea, but only in a single massive fund to keep costs at a minimum.

Posted by: Blue Ohio at January 31, 2005 09:48 AM

I think it sounds really elegant, though I am not clear about #2, capping FICA..."top-off private add-on accounts." Could you elaborate? The no-churn, no-slick-brokerage (TSP) is great.

Posted by: PW at January 31, 2005 10:00 AM

C'mon Brad. You're assuming some level of good faith from the administration--that they actually want to develop a sound program that would strengthen SS. You of all people (the "economic clown show")should know better. SS "reform" is being pushed by Bush because he and his patrons want to destroy the program--NOT strengthen it--partly for ideological reasons, and partly because it is positively associated with the Democrats. In the process of destroying it, they want to allow their Wall Street allies to loot it. Your program does none of this. And do you REALLY expect Bush's people to agree to uncapping FICA?? That's a tax hike on the rich, which is the ultimate evil from this administration's point of view.

Posted by: Rebecca Allen, PhD at January 31, 2005 10:00 AM

Oh great, I'm almost to the point where my income exceeds the cap and you want to elimnate it.

Posted by: section321 at January 31, 2005 10:02 AM

Blue Ohio wrote, "Since we have expanded the life expectancy, if it's needed to retain solvency until employee/retiree ratios balance better in the future, what's wrong with considering a further increase in the retirement age?"

In the abstract, nothing, but people who work at physically demanding jobs "wear out" earlier and pushing back the retirement age is a significant imposition on them. (Doesn't mean I think the status quo on the retirement age is a good thing.)

Posted by: liberal at January 31, 2005 10:55 AM

> C'mon Brad. You're assuming some level of good
> faith from the administration--that they actually
> want to develop a sound program that would
> strengthen SS. You of all people (the "economic
> clown show")should know better. SS "reform" is
> being pushed by Bush because he and his patrons
> want to destroy the program

This of course has been Josh Marshall's point from the beginning. There cannot be an open and honest discussion of alternatives, nor any sort of meaningful policy analysis, when one party has an unalterable agenda that is will not reveal.


Posted by: Cranky Observer at January 31, 2005 10:59 AM

I don't really have a problem with Brad's points, but I feel like all of us have gotten conned into worrying about Social Security's future while ignoring the blue whale in the middle of the room: the general fund. Sometimes, I think that the privatization itself isn't as bad as the illusion of responsibility it creates ("Bush is willing to step up to the plate for America's future rather than kicking the problem down the road!") in the Bush administration even as they steadfastly ignore general fund problems.

Posted by: Julian Elson at January 31, 2005 11:38 AM

I knew all along that the Democrats would roll on the private accounts.

Posted by: pragmatic_realist at January 31, 2005 11:46 AM

Uncap FICA and apply it to all wage income

why would anyone ever work under such a scheme?

take a two professional family, doctors, making $150K each. With the employer match, you have just raised their taxes 30K

There will not be an educated person left in the democratic party

[Perilously close to trolling here...]

Posted by: Moe Levine at January 31, 2005 12:09 PM

Howabout just solving the problem by changing the old-young ratio? Not only legalize every drug known to man for retirees, but add on to the SS benefits with free crack and heroin. The ensuing drop in lifespan will fix our problems posthaste.

Posted by: psetzer at January 31, 2005 12:26 PM

Slightly off topic --

Treasury has just announced projected borrowing needs for calendar Q1 and Q2 of this year. The total $159 bln. That compares to an actual net $177 bln borrowed in the same quarters in fiscal 2004. Kick in the first quarter of the current fiscal year (calendar Q4 of 2004) and Treasury thinks its net borrowing needs for the first 3 quarters of this fiscal year amount to $257 bln. The actual borrowing need in the same period in fiscal 2004 was $290 bln. So unless Treasury anticipates a scary deterioration in the fiscal stance in Q4 of this fiscal year, then Treasury just disagreed in public with the White House/OMB forecast of a widening in the deficit this year.

Now, after getting the deficit projection so embarassingly wrong last year, why would the White Hous not find out from real pros (like the guys at Treasury) what the near-term fiscal outlook is?

Posted by: kharris at January 31, 2005 12:30 PM


One possibility is that game they played before: overestimate the deficit on purpose, then boast how you brought the budget deficit in under the line.

Posted by: liberal at January 31, 2005 01:03 PM

"take a two professional family, doctors, making $150K each. With the employer match, you have just raised their taxes 30K"

Sounds good to me. If you weep over the plight of the $300,000 family having their taxes raised, you shdn't be in the Democratic party. Really.

"Question: who appoints the members of the SSA?"

This is right to the point. This sort of decision cannot be depoliticised. It affects too much of the economy and is directly redistributive between voting blocs. Congress will appoint the members and vote on any rule changes it proposes.

"Mandate that in fifteen years a commission consider and recommend whether or not two percentage points of FICA should be diverted and added to the add-on accounts as a forced savings program."

I didn't expect Brad to be channeling Sandra Day O'Connor's affirmative action judgements...

Posted by: Otto at January 31, 2005 01:15 PM

social insurance is not private or personal think about it.

Posted by: slothrop at January 31, 2005 01:28 PM

JM wrote:

Time for the reliable "is there one major policy initiative that this Administration has not completly screwed up?" (or something like that). Leave it alone and let those who vote in 30 years figure how how to handle it.

I agree. We cannot let Bush co define our agenda -- if anything needs to be fixed, it is the Bush tax cut.

Another point: after reading Krugman's little black lie, I start to dislike the option of raising retirement age: it does cut the benefit of Afraican Americans more if the life expectancy gap does not close quickly (meaning in the next decade or so).

Posted by: pat at January 31, 2005 02:10 PM

From Richard Berner in Morgan Stanley:


... Social Security has another key dimension: It combines wealth and insurance, because in addition to providing a defined retirement benefit; it also insures against disability, inflation and outliving one’s resources. The disability component is especially valuable for low-income workers in physically-demanding jobs. The life insurance component exists because benefits in retirement are paid until death. Relative to private annuities, Social Security is inexpensive insurance against outliving one’s resources. For example, according to Life Insurance Marketing Research Association data, insurers typically return between 80 and 90 cents per dollar of cost for single-premium immediate life annuities which closely parallel Social Security benefits — except they are not indexed to inflation. The inflation-indexed annuity feature in Social Security is less expensive because the system mutualizes mortality risks across a broad population base. For many households, Social Security is wealth, but except for survivors, it cannot be passed on to heirs.

If Social Security is wealth, albeit of uncertain value, economic theory suggests that reform that reduces the future benefit stream will boost current saving. The so-called life-cycle theory of saving developed by Friedman, Modigliani, Brumberg and Ando, a cornerstone of modern macro, posits that consumption is a positive function of income and wealth. A reduction in wealth thus should reduce consumer spending in relation to income, and thus boost saving. Empirical estimates of this effect for Social Security center on a 25% offset: Each dollar of reduced benefits over time might boost accumulated saving by 25 cents — and trim spending by a similar amount. To be sure, the estimates are far from precise, and both the timing and magnitude of such an offset is unclear. However, most current proposals to reform Social Security that claim to boost saving rely on this offset.

Some proposals still rely on three other ideas to justify personal accounts as a prop to saving. First, many, including some thoughtful analysts, believe that personal accounts would represent a more certain asset than future Social Security benefits, because Congress could not take them away. As a result, they are thought to have a higher expected value. That still begs the question of how the accounts are funded, however. Unless personal accounts were funded by cutting benefits or raising taxes, introducing them would not increase either saving or wealth.

A second school of thought relies on the equity risk premium, arguing that to borrow at today’s 2% real rate and shift the funds into equities earning a 5-6% real return will boost wealth. But that process itself would erode the equity premium to low levels, allowing only a thin margin of excess return for the future. Moreover, as individuals watched the wealth in their personal accounts grow, they might not feel the need to save as much from current income, and national saving could actually decline ...

Posted by: pat at January 31, 2005 02:40 PM

Very interesting suggestions and a far cry from those reading materials the GOP retreat featured - as noted by Joshua Marshall over at TPM (with link).

Posted by: pgl at January 31, 2005 02:41 PM


Interesting posts; thanks.

Posted by: anne at January 31, 2005 03:11 PM

Unless I am mistaken, eliminating the FICA cap would allow S.S. to go on a pay as you go basis. There has been a 60% jump in overall income since 1973 but how much extra tax ended up in Social Security coffers when up to 70 percentile wages are the same today in real terms as 1973 and the cap is at the 84 percentile level ["State of Working America, 2000/2001?, table 2.6]. Doesn't income historically grow four times as fast as population -- couldn't that be enough for S.S. revenue to keep up with the boomers.

If so, we could dispense with the trust fund -- leading to an additional 25% FICA cut -- in additon to the 25% cut allowed by taxing an addtional 33% more income. The double tax reduction could yield a FICA rate a little more than half today's rate -- forever going down.

Please tell my why it cannot be this easy. :-)

Denis Drew

Posted by: Denis Drew at January 31, 2005 03:17 PM

Very interesting suggestions. However, I am not sure that Congress would really be up for giving away authority over Social Security to the SSA. Now this might sound kind of backwards, since Social Security is indeed a political nightmare for anybody in Congress, but giving the ability to basically run the program to a "non-elected" body would surely make many people skeptical of how much or how little the authority can do. And I am also very doubtful that the wage cap can be completely lifted, as there is a good argument to be made why there should be one. Nonetheless, raising the cap a little could get bipartisan support.

Posted by: AJ at January 31, 2005 03:20 PM

Denis Drew

The idea is interesting, but politically untenable. The point is that Social Security is indefinitely sustainable as is if economic growth continues at historical rates.

Posted by: anne at January 31, 2005 03:33 PM

CORRECTION to my post, 2 above: HOURLY wages -- not yearly wages -- are what have remained unchanged for 32 years in real terms. Yearly wages of course advanced -- but only with extra hours of work. That extra hours may have reached the limit is another factor I wonder whether S.S. trustees have in mind.

Denis Drew

Posted by: Denis Drew at January 31, 2005 04:19 PM

Moe Levine
About educated people leaving the Democratic party if it raises taxes on high income people back toward what they used to be. Incorrect. It is only high income, not high education, people who will have their taxes raised. If you are a very highly educated person working as a security guard you will not have your taxes raised.
There are some very highly educated people who didn't learn anything salable, or who don't want to work, or who otherwise don't make a lot of money. Like my cousin Edmund, and others I know.
He spent his life in school working hard at education (got into Berkeley) till he was thirty, and then more or less retired to live off his inheiritance from his father.
He told me that he had worked hard all his life in school and he didn't want to do it anymore. Even though he had passed the bar, he wasn't going to practice and was more or less going to retire.
Guy had a point. He is also a socialist and supports Democrats.

Posted by: walter willis at January 31, 2005 05:06 PM

"take a two professional family, doctors, making $150K each. With the employer match, you have just raised their taxes 30K"

There would be a tax increase but it would be 6.2% of the difference between about 90,000 and 150,000, or 6.2% of 60,000 for each person and the match from the employer.

Posted by: Ari at January 31, 2005 07:14 PM

I'd put the cap back at 90% of all income - where it was after the Greenspan reform in 1983.

I believe that would cover about half of the projected long term shortfall for SS.

But before doing that, the only short term problem is in the general budget. Income tax revenues have fallen far below the lowest they've been as a fraction of GDP since WWII, and we've got a massive structural deficit.

Thankfully, or prescient legislators have built in sunset clauses, so all of these tax cuts will expire over the next ten years.

This, in turn, will help to assure that Social Security will remain solvent out to around mid-century.

We could come back around 2020 and re-evaluate SS's funding needs then. Right now, the last thing we need to do is raise SS taxes, given that the GOP says FICA tax revenue can go into US tresury bonds, but can't come back out.

Posted by: ChasHeath at January 31, 2005 07:51 PM

If you weep over the plight of the $300,000 family

no, but I understand two propositions:

1) you cannot tax people whose income comes from being highly educated at these rates and not expect to drive them out of the party

2) we are already in trouble in this regard--this was once the source of many of our state/local candidates--they have all stopped running for public offices

Posted by: Moe Levine at January 31, 2005 09:10 PM

Regarding the disappearing equity premium if a portion of Social Security is invested in the stock market.

Total Domestic Stock Market Capitalization in the US was approx $15.7 Trillion as of 30 Nov 2004.
Total OASDI Payroll taxes collected in 2003 was $533.5 Billion.
If 1/6 of that amount were diverted to the stock market, that would represent an infusion of capital equal to approx .6%.

How much of an effect on the equity premium can we expect from this additional market investment each year?



Posted by: Neil S at January 31, 2005 09:28 PM

Social Security is fine, and needs no tax increase or benefits cut if simply left alone for several years. The general government budget deficit is a problem now. A combination of a small income tax increase and low interest rates would help the general budget problem immensely. But, an income tax increase is not a political possibility at this time.

Posted by: anne at February 1, 2005 03:16 AM

"For instance, Bruce Webb is, by my rough calculations based on his post above, slightly less than a year older than I am. Yet his scheduled retirement will be three years before mine, by virtue of his being roughly six months on the "older" side of the 65/67 split."

Very rough calculations. I was born on Jan 1, 1957 and am due to collect full benefits at 66 1/2, that is sometime in June 2023. Under current law nobody has to wait more than the age of 67 years to collect full benefits. There are some people disadvantaged compared to me, but nobody by three years, and in fact on the spectrum of "screwed out of benefits by Moynihan/Reagan" I am pretty far out on the "screwed" end.

But to answer some random questions above: the Social Security Trustees consists of three cabinet secretaries (Treasury, Labor, and HHS), the Commissioner of SS and two "Public" Trustees. All to my knowledge are Presidential appointees.

But Brad's proposal actually leaves the calculations to the career professionals in the Office of the Actuary. This is not to say they are immune from political pressure, there has been some raw work done with the numbers in the last three Annual Reports, but under the DeLong plan benefits will be tied to economic outcome. And they can only screw around with the numbers so far.

Right now under Intermediate Cost (the officially prefered and used data set) Social Security can only return 80% of promised benefits after depletion in 2042. But then again Intermediate Cost will have to be adjusted to reflect 2004 growth of 4.0% compared to projected 2.7%, so under the 2005 report we can expect that 80% to be adjusted upwards. And for that matter 2042 to be shoved back. (How much we leave to those people who actually enjoy spreadsheets). For the sake of argument lets say it can return 85% at depletion which we will put at 2045 (consistent with past out-performance of the numbers) http://www.epinet.org/content.cfm/issueguides_socialsecurity_changes
EPI: Changes in Projections

So we continue to set benefits at a rate indexed by wage growth but discount it by exactly that much by which projected income doesn't meet projected benefits. Or we recognize that Intermediate Cost has been pessimistic and set the discount at some point in between. In the end it doesn't matter much, we wait and watch. Every year real growth beats Intermediate Cost we ratchet projected benefits up. And the more pessimistic the initial assumption the more dramatic the jump.

Under this scenario nobody loses. Current beneficiaries continue to have their checks adjusted by a COLA. New beneficiaries get their initial check set by real world wage levels at retirement. Future beneficiaries get to sit back and see if the economy outperforms the predictions. And every year they do their projected initial check gets boosted. The more dismal the initial economic projection the more dramatic the projected boost once the real number rolls in. And given the truly dismal projections of Intermediate Cost, full fillup of the "gap" takes no time at all.

So we simply call them on their bluff: reduce my projected benefits to meet your model, but increase them annually by exactly as much as we beat the numbers. (And there simply is no room to adjust out year numbers down, they are already ridiculously low.)

But we resist until day's end the attempt to change the basis of indexation. We'll take 90% or whatever of wage growth to meet the "gap", we will not accept a tie to inflation that guarantees a 42% lower relative check. Because we love America and believe that it will continue to grow as much in the future, and more, than it has in the past. And we are willing to bet our retirement check on that.

Posted by: Bruce Webb at February 1, 2005 04:59 AM

Crancky O said "Get a "reasonable deal" on the table -- Persuade some Democrats to sign on -- Get bill to conference committee -- Jack up opening paragraph; remove all text; substitute the Radical Right's preferred wealth transfer language -- Send W out to the husting to cry far and wide that "Democrats are going back on their word and torpoeding 'reform'"

The Washington Post says Bush is looking for ways to reshape his package to blunt criticism from those who think Social Security is actually valuable.


Oh, crap.

Posted by: kharris at February 1, 2005 05:32 AM

What will happen if we start investing SS funds into stocks?

One thing will be that the additional liquidity will drive up stocks prices quite significantly in the short term. (http://www.trimtabs.com/news/liquidity/latest.html)

Jeff Bezos and the other smart-money insiders who have been selling about $50 billion a year et al will be delighted with this. (http://finance.yahoo.com/q/it?s=AMZN)

As the market skyrockets, the administration will crow about the fantastic success of the program, the budget will again approach balance thanks to tax collections on the insider cashouts, and the apparent success will be used as a reason to cut taxes further, or reduce savings elsewhere, as everyone will be getting rich on their stocks. (Am I the only one who remembers all those people a few years ago who were going to retire at 45 on their 401(k) tech-stock profits?)

Then when the boomer retirement wave hits in full force, and those stocks have to be sold, it will be "Sell? To whom?" as in 1929.

How can people who think there won't be enough money around to buy back the SS trust fund bonds think there's going to be enough to buy back those stocks? Do they really expect such a tsunami of liquidity to be invested productively???????

Posted by: jm at February 1, 2005 06:04 AM

Much is made of the fact that if one started buying them in the '70s or '80s, S&P 500 index funds have been a great investment. By around the mid-90s this had become common knowledge, and many 401(k)s were set on autopilot into those funds. Was it mere coincidence that the index began rising even faster around that time? Has anyone ever tried to adjust the rise of the S&P 500 for self-fulfilling prophecy effects? Has anyone tried to calculate what will happen to the index when those holders start retiring, and a net outflow begins???

Posted by: jm at February 1, 2005 06:46 AM


Self-fulfilling prophecies and bubbles have much in common. As a rough guide to the influence of "it's gonna keep going up" thinking and its aftermath, look at the S&P in September of 1996, September of 2000 and September of 2002.

Posted by: kharris at February 1, 2005 07:28 AM

My colleague and I have been simulating some of the past ephisodes assuming a fixed amount of money is invested every period and examining the liquidating value of the full portfolio at retirement. Here is one of the worst ephisodes:

Worse than investing in stocks right before a market crash is liquidating stocks shortly after the crash. This situation faced investors whose holding periods ended in late 1974—the parents and grandparents of many of today’s baby boomers. Take, for example, an investor who planned to retire in 1974 after investing $100 in the stock market every month since the beginning of 1955. Even with all dividends reivested, that 20-year portfolio had a total cumulative real return of -13 percent. In other words, by the end of 1974, the investor would have contributed a total of $24,000 to the retirement account, but the entire portfolio could only purchase an equivalent of $20,900 of 1955’s goods and services—not much of a nest egg.

Posted by: pat at February 1, 2005 08:46 AM

Is uncapping the amount of wage income subject to FICA taxes the most efficient way to increase revenue? It seems like some sort of tax aimed at the uber-wealthy might be a better idea. After all, people making over $87,900 are certainly doing well for themselves, but they're not exactly rich.

Posted by: Adam at February 1, 2005 08:53 PM

How about making all earnings by those above pensions age tax & social security payment free & cutting pensions payments by no more than 15% of total earnings?

It would cut pensions payments slightly & increase elderly incomes. Assuming, a big but defenceable assumption in a growing economy, that jobs taken by pensioners, often part time ones, create as many new jobs as they fill, it would not lead to more unemployment or have any cost to the Treasury.

On the other hand it would inspire arguments about exploiting pensioners & about giving pensioners an unfair advantage.

This also depends on the assumption that pensioners today, being healthier than in previous generations, would, in significant numbers, like to work.

Posted by: Neil Craig at February 2, 2005 02:04 PM

Post a comment

Remember Me?