Social Security Archive

I haven't found the hack in blogger.com that allows me to make subject-based archives automatically, so I am trying a manual alternative. Listed below are all of the posts that I have made about Social Security, in reverse chronological order. I'll update this post as the thread continues, and I have put a permanent link to this post in the sidebar to the right (after the monthly arhives).

UPDATE: This now represents the first set of posts. Part II of the archive is here.

• A Social Security Debate
• FactCheck and Senator Reid's Calculator
• Well, at Least He's Engaging
• Why Should the Democrats Engage?
• Responding to John
• Senator Hagel's Social Security Plan
• Social Security on WNYC
• Senator Reid and Social Security
• Social Security at FSU
• Social Security in the SOTU
• Krugman's Unhappy Returns
• FactCheck.Org and Social Security's Unfunded Obligations
• Paul O'Neill and Social Security
• Social Security and Children
• The White House on Social Security
• CBO and Certainty Equivalents
• A New List on Social Security Issues
• Victor Does the Heavy Lifting
• More on Life Expectancy Projections
• Max Speaks, and He Clarifies
• The Sawicky Dialogues
• Democrats, Unity, and Social Security
• Progressivity in Social Security Reform
• An Answer from Max
• The Dean and Max Show
• More from Max on Framing the Social Security Debate
• Social Security in the New York Times
• Social Security Reform and the Budget
• Direct Questions that Merit Direct Answers
• Max Speaks, and He Listens and Replies
• Justifying and Administering Personal Accounts
• Social Security Reform in the Blogosphere
• Social Security in the President's Press Conference
• How to Reform Social Security, Part II
• How to Reform Social Security, Part I
• What Social Security Reform Should Democrats Propose?
• Social Security and the Legislature
• Senator Kerry's Statements about Social Security
• Why Is Social Security a Campaign Issue?

Other blogs commenting on this post

Social Security Archive, Part II

I'm getting ready to start a new series of posts on Social Security, and so I have created a second archive page with links to Social Security posts, beginning May 20, 2005.

Access the first set here. Changes in Blogger's technology also enable you to get all references to Social Security here.

• Krugman on Obama on Social Security [11/18/07]
• A Guide to Posts on Social Security Reform [10/07/07]
• Social Security in the Democratic Candidates' Debate [9/28/07]
• Fine Work by the Treasury on Social Security [9/25/07]
• Social Security in France [9/19/07]
• A New Approach to Social Security Reform? [9/13/07]
• Rove and Mythology [8/13/07]
• Quiet Discussions of Social Security Reform [6/09/07]
• The Social Security Coloring Book [3/01/07]
• What Did the Greenspan Commission Do? [1/22/07]
• Some Clarifications [1/21/07]
• Chairman Ben and the Long-Term Budget [1/20/07]
• Look Who's Talking about Social Security Reform [11/22/06]
• Social Security 2.0: Coming in 2007 [10/30/06]
• Which Budget Deficit to Target? [9/05/06]
• First Things First [8/28/06]
• LMS for Conservatives [6/30/06]
• Carveouts [6/20/06]
• Back from AEI [6/20/06]
• See you Monday [6/17/06]
• LMS at AEI [5/12/06]
• Samwick at RadioEconomics.com [4/17/06]
• Social Security Reform Lecture [3/16/06]
• Social Security in the President's Budget [2/08/06]
• Stitching a New Safety Net [2/07/06]
• Three Links for Social Security in the SOTU [2/01/06]
• Answering Questions on the LMS Plan [12/16/05]
• Greg Anrig Shoots the Messenger [12/14/05]
• Nonpartisan Social Security Reform Plan [12/14/05]
• It's Been Mostly Dead All Day [8/01/05]
• In Praise of Senator Robert Bennett [6/28/05]
• But I Can't Follow the Money [6/24/05]
• Inflation and Social Security Finances [6/22/05]
• Testify [6/21/05]
• Gene-tic Differences on Social Security [6/04/05]
• In Praise of Gene Steuerle [6/02/05]
• Getting Back to Bibamus [5/25/05]
• Voluntary Acccounts and Solvency [5/24/05]
• What Is the Social Security Trust Fund, Anyway? [5/23/05]
  

Other blogs commenting on this post

Chairman Ben and the Long-Term Budget

Dean Baker's post this week criticizing Fed Chairman Bernanke's Senate Budget Committee testimony has generated quite a buzz (see here, here, here, and here). I don't think the criticism is reasonable.

First, let's start with the testimony. This is almost exactly what I would have given as testimony in Bernanke's position (assuming I had his staff of excellent economists to help me prepare it). You should read the whole thing, but if I had to pick out my favorite two paragraphs, here they are:

An important element in ensuring that we leave behind a stronger economy than we inherited, as did virtually all previous generations in this country, will be to move over time toward fiscal policies that are sustainable, efficient, and equitable across generations. Policies that promote private as well as public saving would also help us leave a more productive economy to our children and grandchildren. In addition, we should explore ways to make the labor market as accommodating as possible to older people who wish to continue working, as many will as longevity increases and health improves.

Addressing the country's fiscal problems will take persistence and a willingness to make difficult choices. In the end, the fundamental decision that the Congress, the Administration, and the American people must confront is how large a share of the nation's economic resources to devote to federal government programs, including transfer programs such as Social Security, Medicare, and Medicaid. Crucially, whatever size of government is chosen, tax rates must ultimately be set at a level sufficient to achieve an appropriate balance of spending and revenues in the long run. Thus, members of the Congress who put special emphasis on keeping tax rates low must accept that low tax rates can be sustained only if outlays, including those on entitlements, are kept low as well. Likewise, members who favor a more expansive role of the government, including relatively more-generous benefits payments, must recognize the burden imposed by the additional taxes needed to pay for the higher spending, a burden that includes not only the resources transferred from the private sector but also any adverse economic incentives associated with higher tax rates.

The first sentence of this excerpt is what motivates me on matters of national fiscal policy. So what's got Dean so agitated? You should read his whole post, but I think these two paragraphs capture his argument:

The most recent projections from the non-partisan Congressional Budget Office show that Social Security will have enough money between projected taxes and the bonds in the trust fund to pay all benefits through the year 2046, with no changes whatsoever.

This is very important to understand when someone like Federal Reserve Board Chairman Ben Bernanke proposes cuts to Social Security. Workers have already paid for these benefits. The Social Security tax is very regressive. Its regressivity can be justified by the progressive payback structure of the program. However, if the benefits are cut, at appoint when the program can still easily afford the benefits (e.g. 10-20 years), then the government has effectively stolen from the people who paid Social Security taxes.

You will note that in Bernanke's testimony, he never mentions the magnitude of the Social Security trust fund as a guide to budget policy. Dean leads with it. That's a big difference. Dean's statement in the first paragraph that uses the phrase "has enough money ... to pay benefits" would more appropriately be stated as "has the authorization ... to pay benefits." By law, the Social Security Administration can continue to pay full benefits on schedule for as long as there is a positive balance in the trust fund. When that balance goes to zero, it can only pay benefits as it gets in tax revenues. Dean is describing what will happen to benefit payments. Ben's discussion focuses on what it will take for future generations of workers to actually make those payments.

As I discuss in this previous post, the Social Security trust fund balance represents the current value of the Social Security surpluses that have been run to date. When the surpluses are run, the trust fund is augmented with special issue Treasuries, and over time, the trust fund is credited with appropriate interest on those Treasuries. However, the magnitude of this trust fund balance does NOT represent the extent to which debt held by the public is now lower because of these historic surpluses. That would only be the case if every special issue Treasury bond put into the trust fund was associated with the repurchase of a Treasury bond from the public.

In practice, that's not what happens. The federal government simply puts a new Treasury in the trust fund and spends the Social Security surplus on things other than buying back its existing debt from the public, as if the Social Security surplus were just like any other tax revenue at its disposal. How can I assert this? The federal government targets the unified budget deficit, which treats the Social Security surplus in this way. In my memory (which may not be perfect), the only time in the last 25 years when we did not do this was when the on-budget deficit tipped into balance and then surplus in the Clinton Administration. President Bush did it when he pledged "to cut the deficit in half in 5 years" (see this earlier post.) His Administration is doing it again with the more recent statements about budget balance by 2012. In all cases, the deficit in question is net of the Social Security surplus, and thus the policy presumes that the Social Security surplus is available to spend on general government expenditures.

I have in an earlier post argued that the government should be targeting the on-budget deficit and have Social Security in long-term balance. Bernanke stops short of saying this. That's the first way in which his testimony was not exactly what I would have said. There are two other things that I hope he stresses in his future public statements:

First, it is inconsistent for would-be Social Security reformers to be preoccupied with the debt burden placed on future generations due to Social Security's projected annual deficits but not with the debt burden placed on them by continued deficits in the General Fund. What is the rationale for running any deficit in the on-budget account when the economy is in the up side of a business cycle?

Second, it is inconsistent for would-be Social Security reformers to be preoccupied with the debt burden placed on future generations due to Social Security's projected annual deficits while at the same time enacting legislation, like Medicare Part D, that will generate even larger annual deficits to be financed by these same future generations.

Readers of this blog know that I don't exhibit these inconsistencies. I have precious few compatriots among would-be Social Security reformers on the political right.

Returning now to Dean's second paragraph, he regards cutting Social Security benefits as theft, asserting that "workers have already paid for these benefits." I might believe that if the Social Security surpluses were actually being saved rather than spent. But they aren't. It would be more appropriate to say that what the workers--Dean, me, you, all of us--have paid for is all of the government services that the Social Security surpluses have purchased in the past 20 years. We've already consumed them. We have no compelling justification to assert that future generations of workers, who were not party to these decisions, should have to pay higher taxes to honor promises that our generation has made to itself.

But something has to be done, and the sooner it happens, the less disruptive it will be. As always, I recommend that policy makers start here.

A Guide to Posts on Social Security Reform

Welcome to any readers who arrived here from the link in the local Valley News' article, "Democrats Raise Issue of Raising Social Security Income Cap," by John Gregg. His article makes reference to this post and an interview from last week.

If you would like additional information (and opinion) about Social Security reform, there are a few ways to navigate the site to find it.

1) John's article quotes me in one part as follows:

“As a … conservative, I would prefer to do almost everything on the benefit side, that's how I'm wired,” said Samwick, who supports funneling some Social Security payments into “personal” (or what opponents call “private”) accounts for recipients. “But if you are going to raise (revenue) from taxes, raising the cap is probably the most reasonable way to do it.”

That's an accurate quote, but it could use some clarification that is permitted by a blog if not the space constraints at the Valley News. To see a conservative approach to reforming Social Security, see these three early posts on the topic from October 2004:

• Why Is Social Security a Campaign Issue?
• How to Reform Social Security, Part I
• How to Reform Social Security, Part II

For a very good reform proposal that would appeal to those on the political left, see this post from the same month:

• What Social Security Reform Should Democrats Propose?

2) Along with Jeff Liebman, a former advisor to President Clinton, and Maya MacGuineas, an advisor to Senator McCain in his 2000 Presidential campaign, I have proposed a Nonpartisan Social Security Reform Plan. The plan documents are here, and my initial post on it is here. It represents the sort of compromise that we think could emerge from a genuine bipartisan effort to restore long-term solvency to Social Security.

3) There's plenty of other writing about Social Security on this blog. You can find it:

• A post-by-post listing in Part I (through May 2005) and Part II (since then) of the Social Security archive.
• A single link to all posts tagged as being about Social Security.
• A search for the term "Social Security," linked here.

Enjoy!

What Is the Social Security Trust Fund, Anyway?

Periodically, an economist working on Social Security reform has to try to succinctly explain to non-economists what the Trust Fund is. I think I'm getting better at it, so let me try the newest version on you.

Think of the Trust Fund as a line of credit that the Social Security system extends to the rest of the government. The balance in the Trust Fund is simply the current value--principal plus interest credited at the Treasury bond rate--of all the withdrawals that the rest of the government has made historically on that line of credit to pay for things other than Social Security. Its projected balance at the end of the year is $1.85 trillion. Under current law, that balance is projected to peak at $3.61 trillion in 2022 before declining to zero in 2041. During those 20 years, the Social Security system will be calling in the loans that it has made to the rest of the government.

Keeping track of the total amount outstanding on these loans is the accounting purpose of the Trust Fund balance. It also has a legal purpose. Specifically, as long as the the Trust Fund balance is positive, then the system can pay the benefits implied by current law. It would require the Congress and the President to execute a new law to interrupt this process. When the Trust Fund hits zero, then it would take a new law to get full benefits paid on time--they would be paid only as tax revenues flow into the system.

This summary suggests three things to keep in mind about the current debate:

1) Contrary to what I suggested in a post three months ago, current beneficiaries and those nearing retirement age do have a stake in these debates, even if their benefits as implied by current law would not change. Any deviation in the path of the Trust Fund from its currently projected path makes the timely payment of their benefits less certain than it appears today. Policy makers will likely have to view this as a constraint on possible reforms if they are to assure those in or near retirement that their benefits are not becoming less secure as a result of reforms. Bye bye carveout PRAs.

2) It appears to be an official White House talking point that the Trust Fund is merely "a file cabinet full of IOUs." Here's an excerpt from a speech last week:

Now, secondly, Social Security is not a savings account. In my travels around the country I hear people say, why don't you just give us the money back we put in. But that's not the way Social Security works. It's a pay-as-you-go system. You pay; we go ahead and spend. (Laughter.) You pay through payroll taxes; we spend on paying for the beneficiaries, the retirees for that year. But if we've got any money left over, we didn't save it for you, we spent it on government. That's the way it works. It's a pay-as-you-go. And then there's -- all that's left over is a file cabinet full of IOUs. I have seen the file cabinet in West Virginia firsthand, and I saw all the IOUs. But the system is not the kind of system where we're holding the money for you. That's not the way it works. We're spending your money and left behind some paper that can only be good if the government decides to redeem the paper. That's a pay-as-you-go system.

Similar phrasing also appears on the Treasury's Social Security website:

There Are More People Collecting Benefits. As the Baby Boom generation begins retiring in 2008, there will be a dramatic rise in retirees who will be living longer. Social Security is a pay as you go system that leaves workers with IOUs, not personal accounts.

Ignore the non sequitur for a moment and just consider how silly the last sentence is. Every financial security--from dollar bills to Treasury bonds to corporate stock--is an IOU, and there would be no more physical evidence of a personal retirement account than there is of the Trust Fund. This is a talking point that needs to be dropped.

(As an aside, we often hear this phrase attributed to the President as having called the Trust Fund "a bunch of worthless IOUs," but I cannot find a link to an official document. Can anyone send me such a link?)

3) The sentiment being expressed in this reference to the filing cabinet is that the Trust Fund is an unreliable way to pre-fund future Social Security obligations, in the sense of paying more taxes now so that future generations of workers will have to pay fewer taxes later. We spend every dime of the Social Security surplus that is building up the Trust Fund. OMB Director Josh Bolten admits as much in this interview on C-SPAN's Q&A program:

LAMB: As you know, Mr. Walker has a 15-year appointment, so he doesn’t have to worry about the job he’s in. Is he saying -- I mean, this whole business of using the Social Security surplus as a way to keep the deficit down, first of all, what do you say about that?

BOLTEN: Well, he’s right, that what this administration has been doing, what this Congress has been doing, and in fact, going back in time, is that as Social Security money has been coming in, there has been a paper IOU sent over a file cabinet in West Virginia, somewhere to the Social Security system, and then government has been spending the money. That’s one reason why I think the president’s Social Security reforms are so important, because I think we put ourselves much more into a system where people get to keep more of their own money rather than relying on a government promise. ...

Republicans have, for quite a while, been saying that the Trust Fund is unreliable, and when they do, they attribute it to "the government spending the money." This may have been okay rhetorically when Democrats were in power, but it is a sham when they control both houses of Congress and the Presidency.

There is nothing inherent in Trust Fund accounting that requires the government to spend the Social Security surplus. The President could announce a policy of balancing the on-budget deficit (i.e., not the unified deficit, that includes the Social Security surplus) over the business cycle, and Congress could pass annual budgets consistent with that policy. Voila! The surplus is saved, and the IOUs in the Trust Fund actually do represent the extent to which the government has repurchased its own debt on behalf of future generations. This policy would have allowed the government to run a deficit during a recession, but it would have also required the government to run on-budget surpluses afterwards (in other words, now) after the recession. It is this last part that we seem to be unable to do.

So, perhaps sadly, the Administration's conclusion is correct: if Social Security is to pre-fund any future liabilities, then this pre-funding needs to be done in a new system such as personal accounts that deprives the government of the cash surpluses to spend. But the Administration's reasoning is all wrong--it is not because Trust Fund securities (like all other financial securities) are IOUs, but because the federal government--including this Administration--chooses not to implement a sensible budget policy.

Other blogs commenting on this post

Greg Ip Earns a Voxy

Brad DeLong often titles his posts "Why Oh Why Can't We Have a Better Press Corps?", and Andrew Sullivan often names his posts after and gives awards in (dis)honor of journalists who make outlandish statements. I would like to introduce my own award--the Voxy--to be bestowed occasionally on journalists in the mainstream media who make exceptionally lucid and thoughtful contributions to the public discussion. Feel free to e-mail me with nominations.

The inaugural award goes to Greg Ip, for his article in yesterday's Wall Street Journal, Medicare Ills Make Social Security Look Fit. Read the whole thing. I'm just going to focus on some excerpts that show why the article is noteworthy. Greg begins with an observation:

Reforming Social Security occupies countless scholars, commissions and legislators. Reforming Medicare, the program that could really wreck the budget, gets almost no attention at all.

He's right. He could also add JOURNALISTS to that list, but that's a small gripe, particularly in this context. He continues:

The mismatch between the programs' problems and the energy devoted to them is striking. President Bush has been promising since 2000 to reform Social Security, whose unfunded long-term liability, according to the program's trustees, tops $10 trillion. Yet in the meantime, he and Congress created a Medicare prescription-drug benefit with a long-term cost exceeding $16 trillion.

Yes, that's basically right, too. According to the 2004 Medicare Trustees Report (see Table II.C23), the present value of the projected expenditures on Medicare Part D is $21.9 trillion, or 2.4% of GDP. (I would have called this the long-term cost.) Beneficiariy premiums and state transfers are projected to offset $3.6 and $1.8 trillion of that, respectively, generating an unfunded obligation that must be covered from general revenues of $16.6 trillion (after rounding), or 1.8% of GDP.

There are two caveats to comparing this $16.6 trillion directly with the $10.4 trillion in unfunded obligations for Social Security. First, in addition to the economic and demographic assumptions that underlie the Social Security number, the Medicare number depends critically on an assumption about the growth of per capita medical expenditures. The disparity could be higher or lower than $6.2 trillion even if the $10.4 trillion projection is completely accurate. Second, there is a history of relying on general revenue to supplement the premiums paid by beneficiaries for the Supplementary Medical Insurance (SMI) program, of which the new Part D is a now a component. Some general revenue financing appears to be part of the design.

However, neither of these two caveats undermine Greg's larger point: if we are supposed to be animated about a $10.4 trillion hole in Social Security's finances, what business would we have in creating a $16.6 trillion hole in Medicare's finances? And for pointing out that inconsistency, Greg earns a Voxy. Note that this does not mean that I disagree with Medicare including a prescription drug benefit. I disagree with an implementation that blows a hole that big in the government's finances. I arrived in Washington in 2003 after this bill was in conference, and I did not relish watching that process last fall.

In fact, Greg retains the Voxy despite including a quote from me in his article that will render yours truly unconfirmable for future positions in government:

So how to fix Medicare? One way is to raise the age at which retirees qualify for benefits, as is often proposed by Federal Reserve Chairman Alan Greenspan and others for Social Security. "Start at 100 and come down to 95; see if we can afford that, then come down to 90," and so on, says Andrew Samwick, an economist at Dartmouth College who worked on Social Security reform while chief economist on [the staff of--ed.] President Bush's Council of Economic Advisers. "There is some age at which the system is in balance."

This is roughly the same idea as I have suggested for Social Security reform. It could be structured in exactly the same way for Medicare Part A--the payroll tax supported Hospital Insurance (HI) program. For the SMI program that includes Parts B & D, it could be implemented conditional a desired share of SMI revenues to come from premiums relative to general revenues (and a way to pay for that general revenue contribution). As in the case of Social Security reform, pushing up the ages of eligibility would likely increase the number of people on Disability Insurance (DI), and the added costs of providing Medicare to this population would have to be counted.

He keeps the Voxy because he shows where a "raise the eligibility age" strategy may come up short:

But it's not a cure-all. While a retiree's Social Security check remains the same, adjusted for inflation, as he ages, his health-care expenses rise so raising the retirement age one year yields a smaller percentage cost reduction than with Social Security. And it's politically unpalatable.

Greg's right again. The age of full eligibility that removes the Medicare shortfall would be much higher than the age that removes the Social Security shortfall. Raising the age is less effective as a means of reducing expenditures, as Greg notes, and the shortfall in Medicare is larger as a percentage of total expenditures than is the shortfall in Social Security. Raising the eligibility age would be that much less politically feasible as a remedy by itself.

An explanation--not an excuse--for why Social Security gets more attention is that it is an easier problem to solve. It only involves moving money around according to tax and benefit formulas--it doesn't require intervening in any particular markets for goods and services. This doesn't mean that it has gotten no attention. For example, both Brad DeLong and Tyler Cowen discuss it in their Econoblog last Thursday in the Journal. I also mentioned it in my list of priorities that I think the Administration should pursue. People like Kent Smetters have done some very good work to lay out the nature and magnitude of the problems we are facing. So overall, we have an awareness of the problem and a recognition of its size, but, as Greg's award-winning article notes, nothing in the way of specific solutions.

Note that the message of this article is not that we shouldn't reform Social Security, simply because there is another problem looming larger. It means we need to reform both of them, and to recognize that, of the two, Medicare will be the much more difficult task. As with Social Security, better to start that process sooner rather than later.

Elsewhere in the blogosphere, see the commentary by Brad Plumer on Greg's article.

Other blogs commenting on this post

Social Security and the Legislature

In my last post, I tried to clarify some of the misstatements the Kerry campaign has made about the President's Social Security policies. I noted that I think this issue (and the analogous problems in the Medicare system) should be receiving more attention in the campaign.

More importantly, programs that redistribute hundreds of billions of dollars per year should be addressed as part of the high-frequency activities of both the legislative and executive branches of government. We shouldn't think about reform only in the crucible of a partisan election campaign. The President convened a commission, thanked it for its report, but has done little in public to move the issue forward since then.

What has Congress done? The good news is that there are several Senators and Representatives who have put forward reform plans that are well specified enough for the Office of the Chief Actuary at the Social Security Administration to analyze and formally report on them. A list of these plans, with links to the reports, is available here. The list includes plans by people in and out of elected office. There are plans that include large personal accounts, small personal accounts, and no personal accounts. The idea that we would be starting from scratch in discussing reform is misguided. There already exist several plans that would be reasonable places to begin a new round of bipartisan discussion.

Some legislators have taken a leadership role on this issue. If it were up to me, and I had to pick one of these plans to start the discussion, I would pick Senator Lindsey Graham's plan, given its focus on solvency and its inclusion of personal accounts, but I am by no means endorsing that plan to the exclusion of others. There are also plans from Representative E. Clay Shaw Jr. and Representative Paul Ryan, who are Chairman and Member (respectively) of the Social Security Subcommittee of the Ways & Means Committee. Again, these plans are not perfect, but they do represent more of a contribution to the debate than the ambiguous statements of politicians who won't develop a specific plan.

Unfortunately, this latter group includes the senior Democrats on the Ways & Means Committee, and this, along with Senator Kerry's refusal to make any specific statement about restoring solvency while criticizing the Commission's plans, undermines the Democrats' credibility on this issue. Consider the following two examples.

Congressman Robert T. Matsui is the ranking Democrat on the House Ways & Means Subcommittee on Social Security. His website's page on Social Security also lists him as the Convening Co-Chair of the Social Security Leadership Task Force for the House Democratic Caucus and the "leading opponent" of Social Security privatization. The website says nothing about how he would plug the $10.4 trillion hole in Social Security's long-term finances. With respect to finances, it says only, "As a result, Matsui believes that today’s choices should be targeted and incremental improvements to the system that will also help increase very long-term solvency." Even these "incremental" improvements are not enumerated.

Congressman Charles B. Rangel is the ranking Democrat on the House Ways & Means Committee. Social Security does not figure prominently on his website, but there are links to the Democrats' Ways & Means Committee website. That website has mostly political statements about current news and legislation, including my own favorite, a selective misrepresentation of the Economic Report that rivals the Kerry campaign's. It does eventually have a section on the challenges facing Social Security. In the answer to the question, "How big is the problem?" the website states:

The long-range actuarial deficit is 1.92 percent of wages that are subject to Social Security taxes. That is, raising Social Security’s revenues, or cutting its benefits, by an amount equal to 1.92 percent of wages would close the gap.

Very poor form. The 1.92 percent long-term actuarial deficit is measured only over the first 75 years of the projection period. (This was the figure from the 2003 Trustees Report, the latest one available when this webpage was last updated in June 2003.) The analogous figure is 1.89 percent in the 2004 Trustees Report. If all that occurred was a revenue increase of 1.89 percent of taxable wages for the next 75 years, then after the end of this 75-year period, the system would face annual deficits of 6 percentage points of payroll with a Trust Fund equal to one year's worth of benefits, or about 20 percent of taxable payroll that year. (See the last number in this table to verify the 6 percent annual deficits.)

The problems with using this 75-year actuarial deficit, along with other abbreviated metrics for the system's finances like the date of trust fund exhaustion, are clearly explained in Chapter 6 of the Economic Report of the President (pages 137-139, in a section appropriately named, "Misunderstanding the Financial Crisis"). The first page of the Economic Report of the President has the words "Transmitted to the Congress." I wonder.

Whether in the campaign or in their decades-long tenures in the legislature, the Democratic leadership appears to be capable of nothing but criticizing other people's positive contributions to the debate on Social Security reform. And every year that this issue goes unaddressed shifts more of the burden of resolving it onto future generations.

Gene-tic Differences on Social Security

When I read this column by Gene Sperling over at Brad DeLong's site, I did a double-take when I got to the last sentence in this part:

What non-political reason, I am often asked, could there be for someone like myself who supports Universal 401(k)s outside of Social Security to so stubbornly refuse to even consider private accounts within Social Security?

The answer is twofold. First, Social Security is simply the wrong vehicle for pushing the worthy and important goal of increasing ownership and savings among working Americans. In our three-legged retirement system -- which includes market-sensitive private savings, home equity and pensions -- Social Security is the only leg free of market and economic risk.

Did he just write that Social Security is free of economic risk? That's clearly not true, even if all benefits scheduled in current law are eventually paid. The amount I get from Social Security will rise or fall with the growth of the average wage in the economy (literally, the average wage in OASDI covered employment) over the rest of my career. This is obviously a growth rate that is subject to economic risk. And it is not likely that all of the benefits scheduled in current law will eventually be paid--the extent to which they are likely to be cut depends on how well the economy does, since that helps determine how much revenue there will be to pay benefits. Victor over at the Dead Parrot Society has two excellent posts showing the fallacy of Sperling's statement. The first is "Risk in the Current Social Security System" and the second is "Here We Go Again."

I think it is kind of ironic for Gene to be overstating the security of the current system in a post where he goes on to say:

That is why those of us who support new investment incentives like Universal 401(k)s should be the ones most adamant about the importance of keeping the Social Security leg of our retirement system completely risk-free.

The second substantive rationale for a hard "no'' on privatization is that virtually every private-account plan is designed to make Americans undervalue the social-insurance benefits of Social Security and overvalue their private accounts.

But I don't want to be too hasty to judge. I don't know Gene, and he didn't actually write anything that would be at odds with, say, this plan. I'd give up personal accounts if he'd give up tax increases--maybe there is room for compromise. And his sin is overstatement more than anything else--there is both less risk and less return in a system tied to the growth in average covered wages than there is in a system tied to financial market returns. The risk is just not zero. This is certainly not as bad a misrepresentation as suggesting that the bonds in the Trust Fund have no value.

But I wonder what the strategy is for Sperling, following similar advice given by Robert Rubin a week earlier, to admonish the elected Democrats in Congress not to engage in the policy process. Here's an interesting excerpt from the Hill News, covering Rubin's speech to Democrats:

“Putting out a Democrat plan on Social Security would be a horrible mistake because right now it’s the president’s principles against our principles,” Rubin said, according to a Democratic leadership aide. The aide added that Rubin told his party colleagues that it would be hard to win a battle of specifics.

You don't want to put your principles up against the other side, and you don't think you could win a battle of specifics? And your compelling reason for the American electorate to return you to office is what?

I wonder what future electoral success the Democrats are contemplating that would allow them a better shot at restoring solvency to Social Security than they would get by engaging today. Suppose that they win control of both houses of Congress in 2006. They would then have more bargaining power with a President intent on legacy-building in Social Security. But that seems like a slim possibility. Suppose that, by 2008, they have control of both houses of Congress and win the Presidency. Even then, Democratic majorities in Congress would be slim, and Republicans would have as much bargaining power as the Democrats have now. So compromise is inevitable if any reform is to get done, and that compromise is likely to include personal accounts.

So why not engage now, and look for a deal that included these three elements:

• Accepting personal accounts, but getting the President to agree to smaller accounts (like 3% up to a ceiling),
• Accepting reductions in the growth rate of future benefits, but smaller or more progressive than what the President has suggested,
• Insisting that, as a trade for the first two points, the President raise the maximum taxable earnings or impose a surtax above it to fund a portion of the accounts.

And then we could move on.

Other blogs commenting on this post

The White House on Social Security

There have been several posts around the blogosphere on President Bush's recent remarks on Social Security and commentary by Rex Nutting of CBS Marketwatch, by Brad DeLong, Max Sawicky, and Angry Bear. The title of Nutting's article is, "Fact-checking Bush on Social Security: President exaggerates problems in retirement system." It is true that there are some statements that are easily misunderstood in the President's interview, and they all do go in the direction of making the problem more rather than less urgent. But the headline is overkill. Let's take a look at the article.

Here's the introductory material:

WASHINGTON (CBS.MW) - President Bush made several factual errors Tuesday about Social Security's long-term financing problems at a photo op event designed to educate the public about the retirement system.

Bush is expected to offer a plan in the next few weeks to cut future benefits and to divert about one-third of Social Security's tax revenues into individual private savings accounts in order to "save Social Security." See full story.

Before a specific plan is unveiled, the White House is holding a series of events to convince the public that the system must be radically altered to prevent a crisis.

According to the Social Security Administration and the Congressional Budget Office, the retirement system faces long-term funding problems, amounting to about 0.7 percent of gross domestic product over the next 75 years, or $3.7 trillion. Read the Social Security trustees report here. Read the CBO's analysis here.

The SSA says the system's trust fund, financed by payroll taxes and interest payments, will probably be exhausted in 2042, requiring the government to reduce benefits by about a fourth or a third. The CBO says the fund will be exhausted by 2052.

Nutting loses points for looking only at the next 75 years, knowing full well that the system has collected taxes during that period on earnings that generate benefits after that period. He hasn't misstated anything that's in the Trustees Report, but he is using an incomplete measure of the program's projected financial shortfalls. His ability to accurately use these terms to convey something incomplete is partly due to the imprecision of the language that is routinely used in these discussions. In the glossary of the Trustees Report, we find:

Long range: The next 75 years. Long-range actuarial estimates are made for this period because it is approximately the maximum remaining lifetime of current Social Security participants.

I would prefer that he say 1.2 percent of GDP, acknowledging that the annual shortfalls are projected to continue rising after the 75-year horizon. (Both numbers are shown here.)

He then starts in with a "Bush v. facts" section:

Bush: "As a matter of fact, by the time today's workers who are in their mid-20s begin to retire, the system will be bankrupt. So if you're 20 years old, in your mid-20s, and you're beginning to work, I want you to think about a Social Security system that will be flat bust, bankrupt, unless the United States Congress has got the willingness to act now."

The facts: The Social Security system cannot go "bankrupt," for it has no creditors. By law, the trustees will continue to pay reduced benefits even if the trust fund is exhausted. Payroll taxes will continue to come in and benefits will continue to be paid.

According to the trustees' intermediate economic forecast (neither doom nor boom), the trust fund will be able to pay about 73 percent of scheduled benefits in 2042 and about 68 percent of scheduled benefits in 2078.

Future presidents and Congresses could also choose to fully fund scheduled retirement benefits from general tax revenue.

If he has to put the word "bankrupt" in quotes, then he knows he is stretching his point that there has been an exaggeration or a factual error. Informally, I certainly think of myself as one of the program's creditors--I'm paying FICA taxes now and being told that I'll get benefits later. Sounds a lot like a creditor/debtor relationship to me. We could look up the definition of bankrupt and find that it includes several usages, ranging from "legally declared financially insolvent" to "impoverished." We could then look up "insolvent" and find that it can mean "Insufficient to meet all debts, as an estate or fund." Note the word "all" in that definition.

The President used the term "bankrupt" in accordance with its common usage. The common usage just permits more than one interpretation. The system will not be "impoverished" and I don't think that "flat bust" is helpful language, but this does not strike me as a case of a factual error. Nutting accurately describes what the system's shortfalls are projected to be in 2042 and 2078.

Bush: "Most younger people in America think they'll never see a dime."

The facts: Social Security says younger people will see a lot more than a dime. Their retirement benefits - even under a "flat-bust" system -- will be significantly higher than today's benefits in real terms.

For low-income Americans, currently scheduled benefits for those who retire in 2080 are $19,906 per year in 2004 dollars. If Social Security can pay only 68 percent of those benefits, that would be $13,536 per year, compared with benefits of $8,804 for low-income retirees who retired last year.

For the highest earners, Social Security is currently promising $53,411 per year for those who retire in 2080 (or $36,319 per year if Social Security can pay only 68 percent). Current maximum benefits are $21,891 per year for those who retired last year.

The appropriate fact-check isn't whether the people will see more than a dime, it is whether that's what "most younger people in America think." It turns out that there a lot of people who, quite incorrectly, believe that the system goes away when the Trust Fund runs out of money. (I don't know if the polls would make this "most younger Americans.") It may in fact be true that, because of the imprecision of the language that is used in describing trust funds and bankruptcies, the popular discussion of the issue has led to that misconception. Nutting gets his discussion of what actually happens in the "flat bust" scenario right, and I do think it is incumbent on the President to explain what would happen in this scenario.

Bush: "In the year 2018, in order to take care of baby boomers like me and -- (laughter) -- some others I see out there -- (laughter) -- the money going out is going to exceed the money coming in."

The facts: According to the SSA, costs are projected to exceed income, including tax revenues and interest income from the trust funds' bonds, starting in 2028, not 2018. The 2018 date is when tax revenues alone no longer meet costs; workers have been paying extra taxes since 1983 to build up the trust funds' assets for just this
eventuality.

It seems to be a safe bet that the President was paraphrasing this statement in Section II.E of the Trustees Report:

Based on the Trustees' best estimate, program cost will exceed tax revenues starting in 2018 and throughout the remainder of the 75-year projection period.

The President said "money coming in" when he should have said "tax revenues." This one is definitely minor.

Bush: "The problem is, is that times have changed since 1935. Then, most women did not work outside the house, and the average life expectancy was about 60 years old -- which for a guy 58 years old, must have been a little discouraging. Today, Americans, fortunately, are living longer and longer. I mean, we're living way beyond 60 years old, and most women are working outside the house. Things have shifted."

The facts: According to the SSA, the life expectancy for a 65-year-old man in 1940 was 76.9 years. Today, a man aged 65 can be expected to live to 81. Most of the increase in life expectancy in the past half century has been for infants, not for the elderly.

The increase in the percentage of women working outside the home has boosted Social Security's resources, rather than depleted them. Today, many women who worked receive a widow's pension rather than their own earned benefits. All the payroll taxes they paid are funding someone else's retirement.

Life expectancy both at birth and at age 65 have improved over the past half century, and Nutting has cited the figures from this table correctly (assuming half century refers to 1940-2003). The distinction may matter because there is a difference in the implications for Social Security financing between improvements in life expectancy at birth and life expectancy at 65. If the improvement is in infant mortality, then the change is equivalent to an increase in the population growth rate, and this has a positive effect on Social Security's finances. The later in life the improvements occur, the less positive the effect, leading ultimately to a negative effect if the improvements occur only after retirement. (Imagine the effect if the only improvement was that no one died between ages 89 and 90.)

However, there is no discussion in the Trustees Report that validates the assertion, "Most of the increase in life expectancy in the past half century has been for infants, not for the elderly." In earlier posts, I've cited the exact opposite statement from an article in the New York Times. Where is Nutting's evidence on this point? In fact, one could hardly blame the President if he were again paraphrasing Section II.E of the Trustees Report:

Under current law the cost of Social Security will increase faster than the program's income, because of the aging of the baby-boom generation, expected continuing low fertility, and increasing life expectancy.

The growth of costs relative to income is, in fact, "the problem," and the Trustees are the ones who cite increased life expectancy as a driving factor. One could quibble with the President's statement of the problem as being in the present rather than the future (if Nutting's assertion is accurate), but even that seems like a stretch if it is to claim that he made a factual error or an exaggeration.

It's less obvious to me what the President's reference to women working outside the home means in this context. Nutting has explained the impact of women working outside the home on the system's current finances accurately. I can see two possibilities. The first, which is a big stretch, is that this was a reference to the decline in fertility rates, which is in part caused by women working outside the home to a greater extent than in the past. The second is that the President has another talking point in his head about how the current system can be unfair to women working outside the home (for exactly the reason that Nutting discusses) and could be made more fair through personal accounts. My guess is that the President just confounded the two points that he wanted to make here.

Okay, so what do we learn from this? If Nutting stripped out all of the "gotcha" language, it would be a pretty good article, like the ones I often read from FactCheck.org. It would fill in some of the gaps in what the President said and help clarify the issues that are relevant.

It's worth pointing out, as well, that the President isn't a specialist in the area, and non-specialists in any area are prone to making imprecise statements. The President does have several specialists working for him, including Chuck Blahous, a Special Assistant to the President on the National Economic Council. He was featured yesterday in the "Ask the White House" forum, where he took questions on Social Security reform. That discussion may provide some clarification.

Other blogs commenting on this post

Senator Kerry's Statements about Social Security

This post follows on the reference in an earlier one to what FactCheck.Org reported about the Kerry campaign's criticism of the President's plans for Social Security. My goal here is to clear up the misstatements that Senator Kerry has made about executive and legislative branch analyses of the Commission to Strengthen Social Security's report from 2001.

While in Washington at the CEA, I had the opportunity to be the primary author on a chapter in the Economic Report of the President (ERP) about Social Security. You can download the whole document and read Chapter 6, or you can use that website search it for the relevant passages that I'll quote below.

On the campaign as of late (see Senator Kerry's speech in Wilkes-Barre on Tuesday and a press release and its background material from the same day), there are several examples of what I will politely refer to as "poor form." I'll take most of these quotes from the press release, just to keep things simple, but in an order that makes sense for the purpose of the post.

As the press release starts enumerating "facts," it states:

George Bush’s Plan Costs $2 Trillion – According to His Own Advisers

The plan in question is a version of Model 2 presented by the President's Commission to Strengthen Social Security. You can find the Commission's final report here. You can find the Social Security Office of the Chief Actuary's scoring of the plan here. It is not a plan that the President has endorsed. Calling it "George Bush's plan" is inaccurate. Why is there any confusion? On page 142 of the ERP, in the third paragraph of a section on "Can We Afford to Reform Entitlements?" the report states:

As an illustration, consider the recent President’s Commission’s Model 2, under the assumption that all eligible workers will voluntarily choose to establish a personal retirement account (thereby maximizing the transition costs to be discussed below).

The press release misrepresents an illustration as an endorsement. Were we really supposed to write about the economics of reforming Social Security through personal accounts without providing an example of such a plan? The press release gives something of a clue in an earlier sentence:

George Bush’s economic advisers analyzed one and only one Social Security plan: Plan 2 put forward by his Social Security. [sic--where's the proofreader for this campaign?]

That would be a tough one for them to verify. I know that they didn't call me to ask me how many plans I analyzed. To clear up any of their confusion, I can attest personally that I analyzed numerous Social Security plans during my time at CEA. We only used one for the purpose of illustration. There must be some rule somewhere that stipulates that if we use one plan for an illustration, then we have to use many plans for illustration, even if doing so would not help answer the question, "Can We Afford to Reform Entitlements?" and would needlessly add length to the 240+ pages of text already in the ERP chapters. I will be curious to read an Economic Report from a hypothetical Kerry administration. Will they opt for no illustrations or needless illustrations?

It is also misleading to label this as "Plan 2 put forward by his Social Security" Commission. As the ERP clearly states, we have assumed 100 percent participation in the accounts for the purpose of illustration, so that no one would be able to say that we had understated the changes in deficit or debt that could result. Taking up the accounts is voluntary, so an assumption about take-up has to be made. Here's what the actuaries said (quoting from the actuarial memorandum):

For Model 2, participation would be expected to be higher. If the benefit offset yield rate is computed as 2 percent above the realized or expected inflation rate, actual net yields on personal accounts would generally, but not always, exceed the benefit offset yield rate. Due to this uncertainty, the 67-percent participation assumption is likely to be the most appropriate of the three assumptions in this case.

Plan 2 put forward by the Commission would more accurately use the 2/3 participation rate. The Kerry campaign can have its $2 trillion in nominal costs over 10 years, but it also has to then agree that everyone would opt for personal accounts. Not a very good assumption for them to maintain while saying personal accounts are a bad idea. Which would they like?

Back to the press release. This next quote is particularly frustrating:

The Economic Report of the President 2004 says that “personal retirement accounts widen the deficit by design.”

How about an ellipsis folks, or better yet, the whole sentence? On page 144, the ERP states:

Personal retirement accounts widen the deficit by design—they refund payroll tax revenues to workers in the near term while lowering benefit payments from the pay-as-you-go system in later years.

The background file on the Kerry website has the whole paragraph with some red underlining of the part of the sentence that is quoted. Are we to understand that the Kerry campaign thinks that the second part of the sentence is irrelevant? It is most of the reason why anyone would add personal accounts to the system (the remainder being the opportunity to invest their contributions in financial assets other than Treasury bonds).

The press release then states (continuing its reference to the ERP):

Chart 6-4 shows the “change in the deficit” as a share of nominal GDP, these numbers correspond to $2 trillion in nominal dollars. The precise numbers are available in a Memorandum from the Social Security actuaries, they show that the current dollar cost is $2.004 trillion from 2005-14. [sic--can we get somebody over to the Kerry campaign who can edit a run-on sentence?]

Well, yes, that is one thing that is shown in (the data underlying) Chart 6-4, but that's not the most important thing that Chart 6-4 shows. To get an idea of what that might be, just read the caption to the chart:

Relative to GDP, reform initially increases then reduces the deficit and debt.

I think that this is one of those things that a campaign staffer might miss by only underlining the first parts of sentences. The chart shows that all of the debt that would be accumulated during a transition is repaid, with interest, out of reduced program expenditures later on. This is what it means when the dashed curve crosses the horizontal axis and goes into negative territory. Looking at only the first 10 years of a reform that would affect cash flows for many decades is obviously not the right way to discuss policy. (In other areas, the Bush administration has made this mistake as well.)

The press release then quotes three newspaper stories that make similar errors. It finishes with some excerpts of a CBO report. The first one is:

CBO estimates that Bush’s plan will force benefit cuts (even when you include the value of the individual accounts) that grow to up to 45 percent. According to CBO, the President’s plan “would reduce expected retirement benefits relative to scheduled benefits, even when the benefits paid from IAs [individual accounts] under CSSS Plan 2 are included… For example, benefits for the 1980s birth cohort would be 30 percent lower, and benefits for the 2000s cohort would be 45 percent lower.”

Note first that the CBO doesn't refer to this as "Bush's plan." More importantly, I suppose that the ellipsis in this paragraph includes the following footnote from the CBO's report:

13 Since the medians are presented here as point estimates, IA payouts are computed assuming risk adjusted returns equal to the Treasury bond rate.

The CBO shares some of the blame for this one. CBO inappropriately uses the phrase "expected retirement benefits" in the text but then clarifies that it is using the Treasury bond rate to accumulate the accounts. The appropriate terminology for CBO would be to say that it is assuming that the portfolios were invested entirely in Treasury bonds. This is the most conservative investment approach, since it chooses to take on zero equity risk. As long as the equity premium is positive, this assumption serves to overstate the reductions in expected benefits that would occur. (As implied in the footnote, we would really want to see the whole distribution of possible benefit levels, not just point estimates.)

This is a very odd assumption to make as a baseline. As far as I know, there is no presumption among people who propose adding personal accounts to Social Security that people who opt for them would then choose to invest them in such a way that eliminated one of the key advantages of having a personal account--the opportunity to obtain higher expected returns in exchange for taking on some financial risk.

Another issue with this quote is that the CBO is making a comparison to "scheduled" benefits, but we know that there are not enough projected revenues to pay scheduled benefits. Comparing an infeasible current law baseline to a reform that is feasible is obviously not a sensible thing to do. The last part of the press release seems to be designed to address that:

Even compared to a scenario with benefit cuts to extend solvency, Bush’s plan would still cut benefits. CBO analyzed a hypothetical scenario with benefits to ensure Social Security is solvent. Specifically, CBO assumes that benefits are paid by payroll taxes after Social Security’s projected insolvency in 2052. Bush’s plan has lower benefits than even this scenario.

But, of course, the 45 percent number is the one that will be quoted--for example, in the speech:

According to the nonpartisan Congressional Budget Office, the Bush privatization plan would cut Social Security benefits. It will cut them by 23 to 45 percent.

Having the last paragraph in the press release does not undo the critique that the Kerry campaign has made an inappropriate comparison in the speech. Note as well that the speech does not clarify that the reductions in benefits are explicitly not going to affect anyone in or very near retirement today.

The press release also does not refer to the benefit simulations that were done in the actuarial memorandum it cited earlier. The Social Security actuaries are also nonpartisan--this isn't an issue of credibility. The simulations that appear to be least favorable to personal accounts are presented in this table. This table includes projections for workers with low, medium, high, and maximum earnings, for different assumptions about the portfolio investment, and for different years of eligibility.

Focusing again on the worst case scenario for the personal accounts (the 2075 retiree), the table shows two useful things. First, even with a portfolio entirely in Treasury bonds, a low-earning retiree would obtain more than 100 percent of the benefits payable (the concept being used in the last paragraph of the press release). This is due to the redistributive elements of Model 2 that were included for this purpose. This is not true of the other earnings levels--an important refinement of the point made in the CBO report.

Second, with a portfolio invested half in equity and half in a mixture of corporate and Treasury bonds, workers with low, medium, and high earnings are expected to get over 100 percent of payable benefits, though not as much in benefits as is scheduled under current law. And again, this is an expected benefit level, so it includes some financial risk, and whether you believe this is good policy will depend in part on whether you believe first that the expected return assumptions are valid and second that it is appropriate to let workers decide for themselves if the return justifies the risk. The Kerry campaign has ignored an important section of the actuarial memorandum that it cites in other places.

That's my reading of the Kerry campaign's criticism of the President's Social Security policy. Every quote in that press release is selectively excerpted to confuse the reader about the key issues with the Commission's plan, the President's association with it, the ERP's discussion of it, and CBO's analysis of it. There is also nothing in the Kerry speech that provides anything more than boilerplate about what the Senator would do to reform the system. (Start reading after the criticisms of the President--not all of which are inaccurate--with the statement, "John Edwards and I have a real plan to cut the budget deficit in half and protect Social Security." Tell me if you see a plan.) There is also absolutely no recognition of the magnitude of the hole in Social Security's finances.

I'll state the following for the record (with this whole post as the context): I am disappointed that this issue hasn't moved more quickly under President Bush and will take some responsibility for not doing more to help it along while working in Washington. I believe that the President should submit some plan to Congress that restores the system to solvency--whether the Commission's Model 2, another plan that has been scored by the actuaries, or something better--to restart the bipartisan reform process. It would be reasonable to hold him to account for the failure to do that, but only someone who actually had a plan could credibly do that.

In an upcoming post, I'll discuss what I think might be better and what Democrats who do not favor personal accounts should be saying about how they would reform Social Security.

Social Security Reform and the Budget

Brad DeLong levels some accurate criticism of the Administration's unwillingness to put Social Security reform in the budget. The crux of the matter is summarized in Jonathan Weisman's story in the Washington Post:

Republican budget writers say they may have found a way to cut the federal deficit even if they borrow hundreds of billions more to overhaul the Social Security system: Don't count all that new borrowing.

The issue is that it does not appear that the Administration will raise new revenue to fund personal accounts that will eventually substitute for a portion of Social Security benefits as stipuated in current law. As I have discussed before, this does not mean that the reform is infeasible. But it does mean that the unified budget deficit--which includes the Social Security and Medicare surplus or deficit--will be wider over a period of decades as debt is issued to cover the personal account contributions. Eventually, the unified budget deficit and total federal debt will be smaller, as the other measures included in the reform (if it is the President's Commission's Model 2, for example) reduce future benefits by enough to restore solvency and repay the debt issued during the transition. (See Chapter 6 of the 2004 Economic Report of the President for a full discussion.)

The Washington Post article goes on to quote Senator Judd Gregg of New Hampshire, the incoming chairman of the Senate Budget Committee:

"You cannot look at Social Security in the context of a five-year budget," the window that current White House and congressional spending plans cover, Gregg said. "To do so is naive and foolish. . . . If this is simply scored as a five-year exercise, we're never going to solve the problem."

I think he's right, in a political if not economic sense. We could avoid these five-year budget horizon issues if we funded, rather than borrowed, the money to contribute to the personal accounts. But even with additional funding, for a program that spans all generations of workers and beneficiaries, an alternative perspective may be a more useful guide to long-term budget policy.

I have argued that the virtue of reform is to eliminate the unfuded obligations of $10.4 trillion in the Social Security system. It would seem simple enough for the Administration to say that the goal is long-term solvency, this plan restores long-term solvency, and the additional deficits and debt are just a temporary (i.e., only a few decades) phenomenon that disappears when they are repaid out of program saving. If it's so simple, why is the Administration now contemplating accounting acrobatics to try to move its policy forward?

The answer appears in (among other places) the Fiscal Year 2005 Budget released in February. The budget target announced in the President's budget message was to "cut the budget deficit in half in five years." The budget deficit here is the unified budget deficit--which includes the Social Security surplus--and that's the rub. Look at Table S-12 in the Summary Tables of the budget to see why. That table shows:

Deficit Measure	2004	2009
Unified Deficit	–521	–237
On-budget deficit	–675	–501
Off-budget surplus	154	263

The "cutting in half" is taking the $521 billion down to $237 billion. However, this reduction in the unified deficit involves a reduction of only $174 billion in the on-budget deficit, combined with an increase of $109 billion in the off-budget surplus. When the Administration announced its budget target, it fully incorporated not just the level of the off-budget surplus, but the projected growth in that surplus over the five-year period. (To me, this qualifies as "spending the Social Security surplus," but that's a topic for a later post.) The Administration did not announce, "we can cut the on-budget surplus by a quarter" (i.e., 174/675). That would not have appeared to be an aggressive enough position to take on fiscal discipline.

The Administration is relying on the Social Security surplus (and its growth) to reach its near-term budget target, but it does not want to acknowledge the impact of personal accounts on that unified budget deficit. Had it stated its budget target in terms of the on-budget deficit--to cut it by a quarter over five years--then it would have a much freer hand now in putting the reform of Social Security off-budget. Doing that would be an example of an "alternative perspective" that would be a reasonable guide for long-term budget policy. What is not reasonable is to shift between two budget perspectives in order to make stated targets easier to achieve.

In Praise of Gene Steuerle

Suppose we just got rid of this thorny democratic process of reforming Social Security and appointed one person to handle the job. My nominee for the position of "Social Security Czar" is Gene Steuerle of the Urban Institute. To see why, follow this link (via Arnold Kling) to his recent testimony before the House Ways and Means Committee. Can we exhume the Voxy for Congressional testimony? Let's just call it an instant classic.

He starts as follows:

The Social Security debate could and should be part of a larger one in which we engage our fellow citizens in choosing the best direction for society as a whole as better things happen to us in the way of longer lives and new health care goods and services. How can we really take best advantage of these new opportunities? How can we spread the gains from this increased level of well-being and wealth to create a stronger nation with opportunity for all? And how should we share the costs?

Instead, the debate is upside down. Due to the ways we have designed our programs and our budgets, every year we spend greater shares of our national income in areas where needs have declined, and then claim we don’t have enough left over for areas—such as education, public safety, children, and anti-terrorism—where real needs remain and have often grown. I sometimes imagine sitting in the Ways and Means Committee room when someone from the National Institutes of Health comes in claiming to have found a cure, though expensive, for cancer. The members of committee, trapped in the logic of our current budget, find that instead of celebrating this advance, they commiserate among themselves about the increased cost for Social Security.

As a member of the baby boom generation, I remember youthful conversations among my cohort, regardless of political persuasion, that centered on what type of government we could help create to best serve society. As now scheduled, our legacy is to bequeath a government whose almost sole purpose is to finance our own consumption in retirement. Not only haven’t we come close to paying for the government transfers we are scheduled to receive, but we plan to pay for them by dwindling almost to oblivion the rest of government that would serve our children and grandchildren.

Yes, it does seem to be a question of priorities. He finishes his introductory remarks with:

Social Security is only part of this problem, but it is an important part for four reasons:

• It sets the standard for how long we should work and who covers the costs
  associated with our longer lives and the new medical care we receive;
• There are many inequities and inefficiencies in Social Security that are
  independent of its size;
• By default (in absence of new legislation), Social Security is designed to absorb ever-larger shares of our national income, thereby squeezing out other programs, particularly discretionary expenditures, that are not treated equally in the budget process.
• A number of related employee benefit reforms would likely increase private saving, enhance the well-being of low- and average-income workers in retirement, and improve the solvency of Social Security.

He addresses each of these in turn, and makes the following recommendations (with my numbering):

1. Increase the early and normal retirement ages so that at any given tax rate, the system provides fewer subsidies for middle-age retirement and increased revenues, higher annual benefits in retirement, higher lifetime benefits, and a greater portion of resources to those who are truly old.
2. Backload benefits more to older ages, such as the last 12 years of life expectancy, so as to progressively increase benefits in later ages when they are needed more and to increase labor force incentives for individuals still in late-middle age, as defined by life expectancy.
3. Provide a well-designed minimum benefit to help low-income households and groups with less education and lower life expectancies, while simultaneously reducing poverty rates (relative to living standards or wages) among the elderly.
4. Determine family benefits for middle- and upper-income individuals in an actuarially neutral manner by applying private pension standards, making sure that benefits are shared equitably, and reducing or removing significant discrimination against single heads of household, many abandoned spouses, two-earner couples, many divorced persons, those who marry others close to their own age, some who pay significant marriage penalties for remarrying, and those who bear children earlier in life.
5. Provide a minimum benefit that extends to spouses and divorced persons as well as workers to provide additional protections for groups that are particularly vulnerable, and as an alternative to free and poorly targeted transfers to higher-income households.
6. Count all years of work history, providing an additional work incentive and removing the discrimination against those who work longer.
7. Ensure responsible budgetary policy by changing the default rules to guarantee the system automatically moves toward balance—say, through adjustments in the retirement ages or the rate of growth of benefits for higher-income households—whenever the Social Security trustees repeatedly report a likely long-run deficit.
8. Reduce the tax gaming used with retirement plans when taxpayers simultaneously report interest deductions while deferring or excluding interest and other retirement plan income from taxation.
9. Provide additional incentive for plans that do a better job at providing a portable benefit for all workers, such as using the FICA tax exclusion to finance increased deposits to retirement accounts and guaranteeing all workers in a qualified plan a minimum level of portable benefits.
10. Make clearer in the law that employers can use opt-out, not just opt-in, methods of encouraging retirement plan participation—without threat of lawsuit.
11. Focus retirement plan incentives more on lower-wage workers, for instance, through an increase in a modified savers credit, which should be adjusted so that it is available for employer, as well as employee, contributions and so that the credit is deposited in retirement accounts.
12. Provide safe harbors from lawsuits for designated types of retirement and other benefit plans offered by employers who hire or retain older workers.
13. Restore the earnings base for Social Security by increasing the portion of cash wages subject to Social Security tax, capping the tax-free levels of health insurance that can be provided, and dealing with tax preferences for other employee benefits.

Numbers 1 - 7 & 13 deal explicitly with Social Security. Some of them are quite similar to ideas I have expressed in various forms in earlier posts. In brief: raise retirement ages, raise the maximum taxable earnings level, remove disincentives to current work, rationalize the spousal and minimum benefits, and put future solvency on autopilot through modest changes triggered by deterioration in the system's long-term health.

Numbers 8 - 12 deal with changes to private pension plans. With the ongoing growth in defined contribution plans relative to defined benefit plans, it is becoming less clear to me that employers need to be so fundamentally involved in pension plan design. So setting up a universal, lifetime savings account for each person (to which employers could contribute if they want to and with default options that encourage at least modest saving rates) may be the way to go.

Other blogs commenting on this post

Krugman on Obama on Social Security

Senator Obama's decision to make Social Security's long-term financial health an issue in last week's debate has generated some discussion among economists and pundits. As usual, Paul Krugman gets us going, claiming that Obama has been played for a "sucker:"

But the “everyone” who knows that Social Security is doomed doesn’t include anyone who actually understands the numbers. In fact, the whole Beltway obsession with the fiscal burden of an aging population is misguided.

As Peter Orszag, the director of the Congressional Budget Office, put it in a recent article co-authored with senior analyst Philip Ellis: “The long-term fiscal condition of the United States has been largely misdiagnosed. Despite all the attention paid to demographic challenges, such as the coming retirement of the baby-boom generation, our country’s financial health will in fact be determined primarily by the growth rate of per capita health care costs.”

How has conventional wisdom gotten this so wrong? Well, in large part it’s the result of decades of scare-mongering about Social Security’s future from conservative ideologues, whose ultimate goal is to undermine the program.

For the record, Obama has not said that our country's financial health won't be determined primarily by the growth rate of per capita health care costs. He is merely not using that as an excuse to ignore the challenges to Social Security's long-term financial health.

Greg Mankiw takes Krugman to task for ignoring a few people who have pointed out Social Security's long-term financial challenges who would hardly qualify as conservative ideologues. We might also point out that Peter Orszag, who is quoted so approvingly by Krugman, is the co-author of the Diamond-Orszag plan for "Saving Social Security." Conservatives like myself who do understand the numbers readily praise the plan as one that Democrats should have proposed years ago.

Krugman concludes:

But Social Security isn’t a big problem that demands a solution; it’s a small problem, way down the list of major issues facing America, that has nonetheless become an obsession of Beltway insiders. And on Social Security, as on many other issues, what Washington means by bipartisanship is mainly that everyone should come together to give conservatives what they want.

We all wish that American politics weren’t so bitter and partisan. But if you try to find common ground where none exists — which is the case for many issues today — you end up being played for a fool. And that’s what has just happened to Mr. Obama.

I will agree with Krugman that there have been some cases where that's what bipartisanship has meant, substituting "President Bush" for "conservatives." Medicare Part D is the leading example. However, I have not met the conservative who wants what Obama has proposed: raising the maximum taxable earnings level and doing nothing else. Conservatives want something like this, which makes all of the adjustments on the benefit side. Conservatives will settle for something like this, which includes Obama's proposal plus some other changes, like gradual reductions in future benefits, small increases to the payroll tax and retirement ages, and personal accounts to absorb any new revenues to make sure they are saved rather than spent.

What is frustrating about Krugman's column is that he's given up on finding common ground and mocks those like Obama who haven't. There may be no common ground at the moment on the issues Paul thinks are relevant, and the President is as much to blame for that as anyone, but that doesn't mean that there can be none after the next election. Krugman seems to be looking for his turn to be in the majority of 50.1%. I'm looking for something better and something different.

Read more commentary from PGL at Econospeak, Don Boudreaux at Cafe Hayek, and Tom Redburn in the New York Times, who quotes some non-conservative, non-ideologues about the size of Social Security's financing problems.