Wednesday, December 14, 2005

Nonpartisan Social Security Reform Plan

Along with Jeff Liebman of Harvard University and Maya MacGuineas of the New America Foundation, I am pleased to announce the "Nonpartisan Social Security Reform Plan." Jeff was a Special Assistant to President Clinton's National Economic Council, where he worked on Social Security, and Maya was a Social Security adviser to Senator McCain's 2000 presidential campaign. Combined with my experience on the staff of the CEA in the Bush administration, we cover the political spectrum of recent years.

We've all spent plenty of time worrying about the looming fiscal crisis associated with the demographic shift toward an aging population, of which Social Security is the tip of the iceberg. Push finally came to shove, and we bound ourselves together via months of conference calls, and this is the plan that emerged. It's not what any one of us would have come up with on our own, but those sorts of plans never become legislation anyway.

What is unique about the plan is that it is designed around the broad areas of likely compromise across the political landscape on how to restore solvency to the system. What makes the plan important is that the Office of the Chief Actuary has evaluated it and certified that it would "easily satisfy the criteria for attaining sustainable solvency."

The plan contains four primary elements: a gradual reduction in future benefits; an increase in the payroll tax cap; an increase in the retirement age; and the establishment of personal retirement accounts. The plan puts great emphasis on fiscal responsibility – there are no transfers from general revenues to achieve sustainable solvency. Specifically:

1) Pay-as-you-go benefits would be gradually reduced to keep the costs of the traditional system to what can be afforded by the 12.4 percent payroll tax. The cuts are structured such that cuts are larger for high earners than for low earners.

2) The plan would establish mandatory personal retirement accounts (PRA) in the amount of 3 percent of taxable payroll. The accounts would be funded by a combination of diverting 1.5 percent of taxable payroll from the Social Security trust fund and requiring workers to contribute an additional 1.5 percent of payroll into their PRAs.

3) The funds diverted from the trust fund would be replaced, once the Social Security surplus was not adequate, by raising the cap on earnings subject to the Social Security payroll tax so that 90 percent of earnings were taxed. Workers would receive no incremental benefits for paying these additional taxes.

4) The plan would gradually increase the normal retirement age (currently scheduled to reach 67 in 2017) to 68 and the earliest age at which retirees could collect Social Security benefits from its current 62 to 65. People would be able to tap into their PRA assets beginning at age 62.

5) In order to minimize risks and administrative costs, accounts would be tightly regulated and full annuitization of account balances would be required.

6) Total replacement rates from the remaining traditional benefits and the new PRAs are comparable for most workers to those promised but currently underfunded in present law.

I invite your comments and questions on the plan, and I will be blogging more about the plan in the days and weeks to come. It was a fascinating experiment--we were trying to walk the very thin line between compromising our principles, which serves no one, and the principle of compromise, which is essential to moving public policy forward. It is a plan that respects political differences but not entrenched political interests. We believe that we have staked out the center of the political spectrum--the challenge now is to capture enough of the people just left and right of center to build the necessary coalition to see it through.

Blogsearch Technorati


Ritholtz said...

There were 2 aspects to the WH SS plan: Reform and Privitization.

While the former is necessary, the plan died because it was mostly about the latter . . .

Anonymous said...

It sounds like serious, responsible, consensus proposal. Too bad there aren't any serious repsonsible adults in Washington.

I can't imagine the AARP signing up for this

William H.

Anonymous said...

Let me get this straight, you want those of us in our thirties (and twenties) to add another year to our work careers (and payments), have our taxes raised another 1.5 percent for the individual accounts, and have our benefits cut (more so if we are successful).

Where I come from that is known as "throwing good money after bad." So, it sounds like a winner to me.

I bet all three of you are in your fifties.

Anonymous said...

It wouldn't be 90% of each persons income, but 90% of all income. That would currently translate to taxes on about the first $135k. The cap should be raised gradually starting immediately. Retirement age changes should enhance flexibility. People will be working longer; it is not necessary to raise the age to require them to. With little choice, PRAs are only a fig leaf. Will the annuity be indexed? Actually with current productivity growth, there may be no problem which makes this a hard sell.

Timothy said...

So in addition raising the cap on my taxable payroll income (thusly costing both myself and my employer more, if I earned over the cap, which I most certainly don't...but will one day, one hopes), you want to force me to put another 1.5% of my income into some government-run IRA? I don't like it, but it's probably better than other proposals.

Question, however: under this scheme would PRA contributions be taxable with regard to the income tax? If so would gains be taxed again upon withdrawl?

If they're taxed now, but not taxed later it's about the same as a RothIRA, if they're not taxed now but taxed later it's like a mandatory IRA...if you have to include them no AND pay capital gains when you get the benefit, this PRA is in no way preferable to other savings vehicles.

Bibamus said...

A couple of 'high level' comments after a first reading of your plan:

1) The experiment itself (seeing what you, Liebman, and MacGuineas could propose as mutually acceptable) is pretty interesting. So kudos there.

2) Overall, I though many of the particular features of this plan were agreeable: increase in the retirement age, increase in the payroll tax cap, etc. I am not a huge fan of personal accounts, but...

3) I understand the policital economy arguments for the necessity for private accounts (as you hinted at in an earlier comment), and I am quasi-persuaded that you are correct. But it seems to me that proposals that include them nearly always fail to discuss the way in which these accounts shift risk on to individuals. Although, as I said, these are just impressions from my first look at this, this proposal does not appear to be an exception to this rule.

For instance, in your Table 1, you just add the 'traditional benefit' and the 'expected yield on a mixed portfolio for the PRA.' This sort of exercise is useful, to be sure. But in the absence of any discussion of the relative riskiness of these accounts versus traditional benefits, it also strikes me as slightly dishonest. From a social welfare perspective, we are interested in the effect of plans like this on expected utility, not the utility of the expected value.

4) Finally, I would like to offer some rank political speculation (what would a comment from me be without it?). But just something for you to chew on: This proposal strikes me as essentially a center-left proposal. I would predict that if this, or anything like it is to become law in the next, say, ten years, it will be at the hands of a Democratic president. I think that you could persuade a substantial fraction of the modern Democratic party to eat the benefit cuts and the personal accounts in exchange for the implicit guarantees that this is not a 'slippery slope to total privatization'. I don't think you could persuade a substantial fraction of the modern Republican party to eat the tax increase or what they would see as the too-limited nature of the privatization aspects. Put another way, I could imagine a Clinton-type Democratic president pushing something like this through. I cannot imagine a Bush-like Republican doing the same. As for a McCain-like Republican, well, it might be closer, but I don't see McCain-like Republicans winning, anyway.

Nathan said...

"essentially a center-left proposal"

Wasn't the goal to be non-partisan on this?

Timothy said...

Income tax treatment of PRA contributions and withdrawals is exactly the same as current payroll tax payments and Social Security benefits.

Meaning you have to count the part taken for PRA in your 1040 this year, and you'll be taxed again on the benefits when you collect them. This treatment is one of the things I find most infuriating about Social Security, other than being expected to pay for some indigent older's cruises.

Having read the whole plan now, however, I have to say that as far as politically feasible Social Security reform plans go this is a good one. In an ideal world I'd be allowed just to keep the money to invest myself (or, , but that'll never fly. Thinking in the real world, this is likely the best solution that doesn't raise taxes exorbidantly and gives young folks like me a reasonable probability of actually getting something out of it.

Franco said...

People talk about lowering the fraction of salary you receive in retirement, or increasing the retirment age, as benefit cuts. These are just cuts in the parameters, not necessarily in the total benefit. The real way to evaluate whether benefits are cut is to consider present value, accounting for longer life expectancies. Basically, the current system has a hidden benefit increase which accounts for some of the imbalance we currently have. Since we're not going to cut life expectancy (I hope) we have to "cut" other parameters in order to keep the total benefit the same.

Personally, I think its key to have some sort of indexing of key parameters to changes in life expectancy.

PGL said...

Let me lead by: (1) thanking you for the email; and (2) noting Duncan Black (Atrios) is not happy with this proposal. On this one, I disagree with Duncan as I sort of like it. Two basic points:

(1) I really like the no transfer part of the proposal as my main fear regarding the Bush crowd is that they wish to abuse Soc. Sec. for a backdoor employment tax, that is, solve the General Fund crisis ala robbing from the Trust Fund. If one wishes to lower capital income taxes and raise employment taxes, just do so - HONESTLY.

(2) I could live with the idea of a small movement towards personal accounts on one proviso - we do so honestly as in Robert Barro's 2000 Business Week oped. In other words, the free lunch Cato crowd is either seriously confused about financial economics (as in risk and expected return) or flat out lying to folks with there opinion pieces on this issue.

In general - well done to an interesting set of ideas!

Bibamus said...

Let me revise my 'center-left' remark:

You are correct that there is nothing inherently 'left' about this proposal. It is best described simply as centrist.

What I should have said is that center-left politicians will have more success with this plan than center-right politicians will, for two reasons:

First, in the near term, centrists are likely to remain more powerful in the Democratic party than they are in the Republican party. A plan like this is just more likely to come out of the Democratic party.

Second, I don't think the public trusts the Republican party with Social Security. I think this is a classic 'Nixon goes to China' sort of thing - in the near term, only a Democrat will be able to sell the public on private accounts.

Andrew Samwick said...

(Big Red) Anonymous,

Thanks for following up. I agree that all groups should be represented, but that's the rub. The political process on Social Security does not permit those generations yet to come to be represented. If we do only those things that will be financially beneficial to current voters, we invariably shift the fiscal burden to those unable to vote. When you suggested that we were "all" over 50, I was tempted to reply that, no, we are all parents of two small children. And I am now (after 15 months of blogging) much quicker to acknowledge that Social Security is only one area of the federal budget where our generation seems content to pass the buck. (See CR's comments and my reply above.)


Anonymous said...

Will there be a provision to reverse or slow down the increase in the payroll tax if it is unnecessary?

By this I mean that if the economy does better than expected, what is the likelihood that the government will "give back" the higher payroll tax rather than maintain it at the new 90% level?

How credible are the provisions? Can something be done so that the plan is fully implemented or will there be a temptation to change matters halfway through? [Dynamic inconsistency issues?]

For me, the real problems with social security have always been Public Choice ones. Does the govt fool with benefits when things work out? Do temporary taxes become permanent ones? Perhaps your plan answers these questions, but I'd like to hear your thoughts on these issues.

jkas said...

congrats - this is an excellent idea.

ech said...

First, if they want the hoi-polloi (i.e. those of us that will pay for this) to understand it, some of the jargon needs to be defined in the paper.

Second, it is my understanding that one simple change could fix the majority of the social security funding gap: Change benefit increases from being tied to wages to being tied to inflation. I was flabbergasted that this was not the case. As the current system is set up, the elderly get an increasing standard of living -since wage increases of the economy as a whole are tied to productivity in the economy and they usually rise faster than inflation. An increase solely due to inflation freezes their standard of living at their retirement year. If they want a higher standard of living, they need to save for it.

Why isn't anyone suggesting this as the first step?

phwest said...

On inflation vs wage indexing - an advantage of wage indexing is that it ties benifit changes to income changes - i.e. if wage growth is negative for some period of time, benifits are effectively cut. This hasn't been an issue of late, but tying your expenses to receipts is suitably conservative (small c) for a pension system. The real problem, as has been pointed out, is that changes in life expectancy add an unfunded liability that is not covered by this. This is better dealt with directly.

Kyle Markley said...

I would support your plan if it included one additional provision -- that individuals be given the ability to opt out of Social Security.

Are you aware that 22% of people would get out of the system if they had the choice?

dryfly said...

Are you aware that 22% of people would get out of the system if they had the choice?

I would guess an even larger percentage would opt out of all taxes if they could do so legally. So should they be allowed to do so?

CR has it absolutely right - the budget deficit is a far greater problem and after that medical costs (both privately & publicly funded)...

Those are much bigger fish to fry and the success or failure of tackling them will directly impact the success or failure of funding Social Security.

Jack Miller said...

I like the key feature, the mandatory payment of X amount toward ones own retirement fund. The same idea should be included in a health care solution.

One cannot drive a car without buying liability insurance, one should not be allowed to work without setting aside a minimal amount into a tax advantaged health savings account. The real cost to the purchaser would be negligible and numerous inefficient government programs could be canceled.

The empowerment of the poor to control their health care and retirement savings would be a nice "freedom thing".

Anonymous said...

Calculated Risk has interesting point on prioritization.

I think implicit to this discussion is a sense that the U.S. could balance the books somewhat quickly and easily with tax hikes. However, in the interest of growth and avoiding deep recessions, income tax cuts have been enacted.

The entitlement spending, on the other hand, is more fixed and not discretionary, and is potentially not as easily solved by tax hikes or future economic growth. Some slight curbing of the social security benefit per person could be used to fund the new medicare entitlement.

Personal accounts benefit poor people - arguably moreso than the rich. I do not know why liberals oppose personal accounts in knee-jerk fashion. Rich people are probably rich regardless of the 3% per year personal account. The personal accounts is for people of very modest means, and people who do not expect to live many years into retirement. This proposed change to social security would give poor people something they do not have, and this benefit would be funded by higher-income people than under the current system.

Anonymous said...

A question on full annuitization of accounts:

This benefits people who retire when the stock market and penalizes people who retire when the stock market is not at highs. Is this good? Why require full annuitization? Is there a way to provide some flexibility regarding the start of annuitization? Can some people opt out of annuitization based on demonstration of sufficient income or assets?

Anonymous said...


This benefits people who retire when the stock market "is high"

Anonymous said...

one more:

In my mind, the social security system is an "insurance" system as much as a "retirement" system. I have no problem with means testing, or curbing benefits, for retirees who have high incomes and wealth in addition to a social security check. No insurance company pays out on a policy when there is no reason for a claim.

JG said...

"The two most pressing fiscal challenges for the US are: 1) the health care system and 2) the General Fund Deficit (close to $600 Billion this year alone)."

$507 billion in 2005.

"Social Security is irrelevant when compared to those two problems.

"I suggest fixing the most serious problems FIRST, and then returning to Social Security."

The present value of the unfunded SS liability is $5.7 trillion (over 75 years, much larger open ended) say the Treasury's latest numbers.

This is non-trivial in anybody's book -- and includes an extra near $500 billion for SS in 2005 alone. Matching that general fund deficit for the entire rest of the government which is so bad.

As to why fix this very non-trivial problem first, the answer is because we can. It has been studied up and down and over since the SS Advisory Commission of 1994, and the answers are all there -- specific answers -- if only there is the will to do them.

In contrast, Medicare and Medicaid are a huge mess and nobody has any plausible answers for them.

So saying "Do health care first, not SS, even though we can do SS" seems pretty much like a ploy to try to preserve the status quo and do nothing.

Unless, of course, one actually has a solution to propose for health care first. Does one??

As Moynihan used to say, "Social Security is easy, Medicare is going to be hard, if we can't fix Social Security then God help us."

katzxy said...

I understand the grim arithmetic forcing higher taxes and lower benefits. But the idea of forced annuitizaiton is not good, as it prevents the accumulated assets from being passed to the next genera tion to build up wealth. You need to find a way to structure this so that wealth can transfer and build up over a few generations.

dryfly said...

As Moynihan used to say, "Social Security is easy, Medicare is going to be hard, if we can't fix Social Security then God help us."

Translated: "If we can't arrange the deck chairs what will we do when the ship sinks?"

First things first... even before Midicare... tackle the deficit NOW... then medical cost escalation including Medicare... then Social Security.

First things first always.

jm said...

Since the goods and services consumed by future retirees cannot be saved up, and must be produced in the future, the only way in which saving in advance can improve retiree living standards overall is for those savings to be invested so as to fund (or free other money to fund) activities that will increase the productivity of those who will be working in that future.

Although person A may improve his or her retirement living standard relative to person B's by saving more than B, and both may improve their future standard of living relative to future workers by accumulating durable assets which future workers may forgo other consumption to buy, there'll be little improvement if the savings are mostly lent to fund consumption of services and nondurable goods.

What matters is not how much we save up in SS Trust Funds or private accounts, it's how we invest those funds. Since I can see no reason why private accounts would be more wisely invested than SS funds, I can't work up any enthusiasm at all for them, regardless of form. As recent history so clearly demonstrated, and as we could reason from basic principles, few people have the knowledge and intellect to earn a return greater than they could get on US Treasuries. If private accounts were to be invested in common stocks by creating some kind of public index fund, then we would just create innumerable opportunities for the pros of Wall Street to front-run and otherwise game the system and pump out the money; and since the only way to get the money out would be to sell the stocks, it would guarantee a market decline when the baby boom peak hits retirement age. (Note also that hardly any of the money spent to purchase stocks goes to buy new shares and fund company operations -- almost all of the money goes to buy "used" shares from people who acquired them earlier, often at enormously lower prices.)

bakho said...

So they want to take one of the most regressive taxes we have, the FICA taxes and bump the rate from 7.5% to 9% for the lowest paid workers. That is a whopping 20% tax increase on those who can least afford to pay. And WTF does the other 1.5% come from once the SSTF expires?

The problem with having the upper crust design these plans is they don't know what it is like to live on the minimum wage.

It would be much better to make the retirement account savings a fixed amount benefit. Otherwise, the lowest paid workers will have the least nest egg. At $5.15 /h 3% is a whopping $309/ y in a savings account. If you assume the IRA doubles every 10 years, a person paid $5.15 /h would not even have $100,000 in retirement account. Meanwhile, those making close to $100,000 per year would get a hefty almost $1 million IRA account, half of it paid by the SSTF. I think this is unacceptable because it represents a huge transfer of wealthy from the bottom quartile to the top quartile, doesn't it?

We already tax SS benefits paid out to the wealthy and that is paid back in to SS. Why not just up the tax rate on wealthy SS recipients? That way if they don't need it, it can go to people that do. Then if catastrophe strikes, the tax rate automatically goes down and they get full SS benefits.

Also, keeping money in an IRA is a bad idea for people with credit card debt. It would be much better to pay off the credit card debit that is charged at 15%+ interest than to keep the funds in an IRA making 4%. Maybe part of the intent is to create a savings fund the credit card companies can collect when the borrower dies?

Anonymous said...

I notice there is no mention of the raid that happens every year on the SS surplus. The press also never mentions this as well. The way I understand it, more money gets paid into the SS fund every year than is spent. That money is SUPPOSED to be set aside for future SS payments as they are needed. Unfortunately, what happens is that congress "borrows" this money and spends it on other things. Of course, they are borrowing it with no real intention of EVER paying it back, so we end up with a stack of IOUs.

SS is supposed to be a specific tax, that goes into a specific fund, for a specific purpose (read: fiscal conservatism). What congress has done is turn this into another general slush fund (read: tax-and-spend liberalism).

Again, I almost never hear this point debated. What will your plan do about protecting any surpluses that may occur? Will you put it into an account that CANNOT EVER be diverted to other purposes? If not, then you can basically chalk up any money raised from your proposal as just another stack of IOUs.

My gut feeling is that any congressional opposition to SS reform comes from the simple fact that they may no longer have this slush fund to spend on pork and the myriad of other wasteful federal programs. Without the ability to waste money, they are nothing, which is why SS reform scares them to death.

Andrew Samwick said...

The plan allocates all of the near-term surpluses to the personal accounts, with exactly this issue in mind.

In order to achieve sustainable solvency, the plan eventually has the system returning to surplus. If I live to see the day, I would advocate some other use for those surpluses--like greater contributions to the accounts or a payroll tax rebate.


Anonymous said...

That sounds like a great idea, glad to see someone took notice that we are being robbed blind by the current system.

bakho said...

"If I live to see the day, I would advocate some other use for those surpluses--like greater contributions to the accounts or a payroll tax rebate."

So WTF propose to raise the tax from 7.5% to 9%??? Doesn't this make the problem you are trying to correct worse and not better? If you want to drain the SSTF in favor of private accounts, why not just transfer the whole SSTF into private accounts tomorrow (equal amounts per person) and then cut SS taxes by 30%???

Both the SSTF and the personal accounts only exist on paper. The SSTF is invested in the equivalent of Government securities. Private accounts could potentially invest in more risky options and be subject to pricy management fees.

The 30% reduction in SS taxes would have to be replaced by borrowing an additional $184 Billion in 2006 or by raising taxes else where to collect another S184 Billion.

At the end of the day it is still the same pot of money. You are just shuffling paper.

Ken Houghton said...

I'm confused now.

Right now, we pay 12.4% (6.2% from the employee, matched by the employer) on the first $90K. (As noted above, Medicare is not being addressed here.)

Is the additional 1.5% really 3.0% (as people seem to be treating it), or is it an unmatched 1.5%?

On the demographic question, and reminding you of the Disability Insurance aspect of the program, what percentage of people now retire before 65, and why would you assume this percentage would go down?

Ken Houghton said...

And, by the way, what would the harm be of leaving the fourth item (mandatory PRAs) out of the plan?

Anonymous said...

Cut the entire system. I am 23 and hate the social security and don't give a crap about senior. They already take 40-50% of the fedeal budget(medicare, social security), how much do you guys want?

Goldwater was right!
And no, i am not a republican. I am a Democrat---a Liberterian Democrat(Bush is a moron too).