Social Security: OASI and DI

A brief response to a comment from yesterday’s post. When CBO projects the point at which the Social Security trust fund will be “exhausted” that means there are no more securities to redeem to pay benefits that are now greater than payroll taxes flowing into Social Security, and under current law when this occurs benefits must equal taxes flowing in. Thus,  if we do absolutely nothing, there will be an ~ 25% benefit cut in about 20 years (shown as 2033 in table below, 3rd column OASDI, last row). As an aside, while this would be a big cut, it also means that people saying “Social Security won’t be there at all when you retire” don’t understand the program’s finances. I don’t suggest doing nothing and allowing such a cut, but ~75% of benefits is a lot greater than 0% of benefits, obviously.

However, something must be done to Social Security (2012 Trustees report from which above table taken) within the next few years, because the trust fund that back stops the Disability Insurance portion of the program will be exhausted in 2016 (DI column above, last row). Again, this doesn’t mean there will be nothing, but there will occur an automatic benefit cut within the next few years sans action. When CBO provides a date of 2033 for OASDI trust fund exhaustion, they are assuming that Congress and the President will pass a law that will allow the co-mingling of the OASI (old age, survivor) with the DI (disability) trust funds, thus allowing Disability benefits to not be cut around 2016. It has long been assumed they would act in this way, but it is my understanding that it will take legislative action for this to occur.

cross posted at samefacts

Social Security First?

Bob Pozen writes that reform of Social Security is the route to a deal to avoid the looming ‘fiscal cliff.’ I wrote something similar in March, 2011, and followed up with more on my view of the benefits of moving sooner rather than later on Social Security reform, a theme that is echoed in my book. Several quick points about Social Security and the myriad policy problems we face.

  • Social Security is not as big a problem as is Medicare (nowhere close), for example, but there is a fiscal shortfall that would result in a 20-25% benefit cut in about 25 years with no action, so something needs to be done. And because Social Security is a simple program (it mails people checks), any fix that was agreed to would work roughly as planned. It truly could be taken off the table by fixing it, allowing for focus on health reform (that will never be finished).
  • There are only so many ways to close the fiscal shortfall in Social Security (increase taxes and/or reduce benefits; CBO w options), and none of these options are made better by delay. One reason to go ahead with Social Security is that it can actually be done.
  • Health reform is many times more difficult than is Social Security reform. Whether it is implementing the ACA, or the mysterious Republican “step by step” approach that will be unveiled, well someday, there will be many mid course corrections and things that don’t work well under any health reform approach. That is because it is much more complicated than simply mailing checks.
  • Progressive reforms such as increasing the minimum benefit, and raising the payroll tax cap could be argued for more forcefully in the context of moving ahead to address Social Security’s fiscal shortfall. Any changes and offsetting benefit reductions for upper income retirees can also be done more gradually the sooner we start.

There is not likely to be much deficit reduction gotten from Social Security over the long run, nor should there be; most of that should come from an increase in taxes collected as a percent of GDP over time, and health care reform efforts that attempt to slow the rate of health care cost inflation. However, a deal on Social Security could be the beginning steps to a more comprehensive deal that paves the way for a long term sustainable budget. And I continue to believe that taking credible steps to address some of our long term problems would help make the case for more short term efforts to boost the economy.We need to walk and chew gum at the same time, but now appear unable to do either.

For me anyway, there would also be a large intangible benefit that would come from seeing our political system practically address some of the problems we face.

The Subsidy of Marriage by Social Security

(cross posted at Same Facts)

Josh Barro notes that marriage cannot be an entirely state issue because of the way in which it impacts things like Social Security and Medicare eligibility. Below is the majority of a post I wrote in September, 2011 on the subsidy of marriage provided by Social Security…..

Gene Steurle and Stephanie Rennane have a nice policy brief put out by the National Institute for Health Care Management on the lifetime contributions and benefits of Social Security and Medicare. This is mostly familiar stuff, with lifetime Medicare benefits consistently being several times larger than contributions to the pay-as-you-go program, and Social Security lifetime contributions and benefits being more similar for singles and two-earner couples. However, one thing jumped out from this figure (circled):

The profound subsidy of marriage that is inherent in how Social Security determines benefits for spouses who do not pay Social Security taxes in earlier life. As Steurle notes:

While a single woman who worked a full career at the average wage can expect to receive Social Security benefits roughly in line with her payroll contributions, a married woman who never worked but whose husband paid the same taxes as the single woman can expect to receive about $180,000 in spousal and survivor benefits. Unlike private pensions, these additional benefits are essentially free but only to those who are married, regardless of need, contributions or any child rearing.

The subsidy would be the same if the gender roles were switched.

Determining what type of contracts (marriage, civil union, etc.) provide the benefit described here–allowing one member of a couple to obtain eligibility for Medicare and a Social Security pension even if they have no market wages that were subject to payroll taxes over their younger years– and whether all American citizens have access to this subsidy, is a key public policy detail that will have to be worked out if marriage equality is to have its fullest meaning.

Social Security disability, rewind

The 2012 Social Security Trustees report estimates the year at which benefits will outpace Social Security (OASDI) payroll tax receipts plus spending authority granted by IOUs (from when more payroll taxes flowed in than benefits were paid out), and the program will no longer able to pay full benefits as 2033.

However, under current law, the disability portion (DI) of Social Security and the old-age retirement portions (OASI) are separate and always have been. The date of 2033 reported in the media and shown below under OASDI, assumes that Congress will pass a law allowing the mingling of funds from the old age (OASI) and disability (DI) portions of Social Security to pay for the disability shortfall, that will come about in 2016.

As a stand alone program, the old age portion of Social Security can pay full benefits through 2035. Assuming the parts of the program are going to be co-mingled (which has always been assumed) puts the year when the combined program could no longer pay full benefits without changes at 2033. The shortfall facing the disability portion of Social Security is much smaller (stated as percent of taxable payroll [and not GDP], 0.37 v. 2.30) than is the one facing the old age retirement portion.

At some point, Congress will actually have to do some of these things that have always been assumed to be straightforward. I wrote nearly the same post about 11 months ago. Also, keep in mind that the long range financing of Medicare is set to move away from payroll taxes, and therefore trust fund accounting, and toward income taxes, by default.

PS: I apologize for the quality of tables. Taken from pdfs in a manner that usually has higher quality.

What about universal burial insurance?

What about burial insurance was one of the counterfactuals–is this ok too?– that was batted about yesterday as the Justices heard arguments about the individual mandate and Conservatives were asking where the limit of federal power might be. Doesn’t Social Security already provide something like this? Sort of, maybe.

Social Security today provides a one time payment of $255 that is payable upon death to a surviving spouse or children. However, the history of this one time lump sum death benefit (that has existed since the program inception in 1935) is a bit more ambiguous, and it was initially meant to recompense the estate of someone who paid in but who would otherwise get no benefits due to an early death. When survivor benefits were added in 1939, the lump sum death benefit was retained and became viewed as ‘burial insurance’. Says the SSA website:

The major Amendments of 1939 introduced survivors benefits into the program and began regular monthly benefit payments in 1940. Due to the addition of survivors benefits, the original LSDB was discontinued. In its place, the current LSDB was introduced, with the intention that this would assist surviving family members when regular survivors benefits were not otherwise payable. If there were no surviving family members, the LSDB could be paid to an individual who assisted with the burial expenses of the worker. So the LSDB was not strictly a burial benefit, although it evolved over the years to be considered as such.

The amount of the LSDB was defined as 6 times the Primary Insurance Amount (PIA). The PIA is basically the monthly benefit amount for the worker at full retirement age.

The average LSDB payment in 1940 was $145.79. The minimum payment ever made under this 6X computation rule was $63.75 and the maximum payment $273.60.

The purchasing power of the lump sum death benefit has obviously been greatly reduced (average pay out was $145.79 in 1940; today it is a flat $255); that amount may have paid for a burial 70 years ago, but certainly would not today. As survivor benefits expanded, the ‘burial insurance’ component of Social Security was greatly reduced in real terms (not updated for inflation, etc. in the same way that other benefits have been).

Look back at CBOs 2000 long term budget outlook

Len Burman reminds us that the reality of the merging of the retirement of the baby boomers with the overall health care cost problem in the U.S. is not exactly breaking news. In the 2000 CBO long term budget outlook, CBO was urging caution about what to do with burdget surpluses (remember those!):

The aging of the large baby-boom generation and growth in the cost of health care will dramatically increase spending for federal health and retirement programs under current law. If policymakers act to ensure that the budget remains in surplus over the near term, the resulting drop in debt held by the public and the lower interest costs that follow will help offset some portion of that increase. Preserving the full amount of the projected surpluses could substantially delay the onset of fiscal problems and help boost GDP, providing a larger base of resources from which to meet the increased demand for spending. But even if policymakers preserved all projected surpluses, spending and revenues would be unlikely to balance over the next 75 years. {emphasis mine}

Even if all of the projected surpluses in year 2000 were reserved to offset these long term trends, an increase in taxes, decrease in benefits, or both would have been needed to totally pay for the retirement and health care needs of the baby boomers. That warning today, seems quaint.

Gleckman on Social Security

Howard Gleckman puts forward six ideas for Social Security reform.

  1. Create a respectable minimum benefit for low-income workers, increase some widows’ benefits, and create an additional benefit for the very old (say, 85 or older).
  2. Raise the retirement age, including the minimum benefit age of 62. An extra year of work would solve about one-third of the program’s funding problems. More and more of us can work into our 70s and a modern Social Security system should reflect that.  It makes no sense for government to signal that we should stop working at 62 when we are likely to live for two more decades.
  3. Protect those who work physically demanding jobs. While the percentage of older Americans who do manual labor is shrinking, those who do this work need to be protected. Long overdue reforms in Social Security’s badly broken disability system would help.
  4. Increase contributions and reduce benefits for high-earners. Everybody would still get some benefit—Social Security is not welfare and must retain its status as social insurance. But there is no reason why it can’t be made more progressive.
  5. Preserve the defined benefit nature of Social Security. Adding an additional savings component is a good idea. But the public is not interested in taking on additional risk with their retirement.
  6. Be absolutely transparent about benefits and structural changes. Whatever Congress does, there should be no surprises. As it is, many young people have no confidence in Social Security. Reforms should restore their faith in this key piece of the old-age safety net. But government should also be clear that in the future Social Security will only supplement—and not replace– other retirement savings for middle- and upper-income retirees.

I believe there are two main benefits of acting sooner rather than later to address Social Security.

  • Doing so now will expand the options available, making an increase in the minimum benefit (option 1) and protecting those workers in physically demanding jobs (Option 3) more likely. Such tweaks would at least partly address differences in lifespan gains across different income groups that are an issue with eligibility age increases.
  • Since Social Security pays cash benefits that are indexed in one way or another to inflation, any fix agreed to should work as expected, and Social Security could truly be ‘off the table’. In this way it is very different from any health reform plan (Affordable Care Act or any other), whose savings will be more difficult to predict, and will require mid-course corrections and endless tinkering, simply because purchasing health care is more complicated than mailing checks.

Social Security disability woes

Dylan Matthews underlines the need for reform of the Disability Insurance (DI) portion of Social Security, playing off of a story in today’s WSJ about high rates of approving disability claims. The summary of the 2011 Annual Social Security Trustees’ report provides a clear statement of the fact that the DI portion of Social Security faces a problem sooner than does the Old Age Survivor Income (OASI) portion of Social Security (p. 4).

The table above shows that the DI trust fund will be exhausted in 2018, as compared to 2038 for the OASI (retirement portion) portion of Social Security. Exhaustion of the trust fund means that the benefits paid out must equal the taxes flowing into that portion of Social Security, beginning in that year, if no changes are made. Starting in 2005, the disability claims paid by the DI portion of Social Security were greater than the taxes flowing into DI. Full disability benefits have been paid since 2005, and will continue to be paid so long as there are special issue Treasury bonds to be redeemed (that were purchased with surplus taxes flowing into the DI portion of Social Security since the 1980s). Once they are gone, either benefits will be cut, or Congress will have to provide another source of revenue.

Several points to highlight:

  • DI faces this ‘something must happen or benefits will drop’ date in 2018 as compared to 2038 for the OASI
  • OASI cannot cross subsidize DI under current law, but Congress could change this
  • The Trustees seem to assume that Congress will allow this cross-subsidization, because the OASDI trust fund exhaustion date (for both trust funds combined) is given as 2036

update: fixed a typo

Interview with Charles Blahous-part 3

Part 3 of my interview with Charles Blahous focusing on Medicare. Here is part 1 (General); and part 2 (Social Security).

Charles Blahous, one of the two public trustees for Social Security & Medicare, is a research fellow at the Hoover Institution, and was the Deputy Director for the National Economic council from 2007-09, and served as a Special Assistant to the President for economic issues from 2001-07. He is the author of Social Security: The Unfinished Work (2010: Hoover Institution).

Below are the questions I submitted to Dr. Blahous, and his answers.

  • What is the relative impact of demography as compared to cost inflation in explaining Medicares’ financing problems?

Pre-2035, demography is the bigger factor.  Ultimately health cost inflation would become the larger factor.  This in my view is hugely misunderstood.  Yes, health cost inflation is a huge issue, but it’s a harder factor to predict, and it doesn’t actually become Medicare’s biggest problem for some time.  We first have to come to grips with the number of years for which individuals receive support under Medicare.

  • What are the benefits and negatives of Medicare being financed by payroll taxes and general fund revenues?

The big negative is a lack of a transparency.  We just put out a report and the big headline was, “Medicare insolvent in 2024.” Well, that doesn’t tell much of the story, given that this describes Medicare Part A only, and Parts B and D never go insolvent by statutory construction.   The Medicare cost problem is manifested primarily through pressure on the general budget, an issue that is really obscured by part of its having a separate payroll tax structure and trust fund insolvency projection a la Social Security.  These different pieces of the puzzle don’t go together very well.  General revenue financing is what you adopt when you mostly want upper-income people to pay for something; a separate payroll tax is what you adopt if you place a higher priority on all program participants funding their own benefits.  These two are mixed together in current Medicare without much rhyme or reason.

  • What are your views of the cost saving approach represented by the Independent Payment Advisory Board (IPAB)?

As a Trustee, I really can’t have a view.  If the law states that the IPAB will produce a certain amount of savings and that it will occur unless Congress acts to override it, then we have to assume those savings will materialize. But as a long-serving Senate staffer, I’m much more skeptical.  I can’t count the times that we have turned to mechanisms like this to produce savings in Medicare, only to have Congress override the savings when they begin to bite.  If we have the political will to cut spending, then cut it.  If we don’t have the political will, then IPAB doesn’t have much of a chance. The fact that IPAB was a product of a bill supported by one party while strongly opposed by the other makes its long-term prospects even weaker.

  • Do you think Medicare costs can be slowed to sustainable levels without addressing cost inflation in private health insurance?

Too often, Washington policy makers tend to act as though federal health spending is the passive victim of runaway cost growth in the private sector.  But most of the academic literature suggests that the opposite is true; you have Finkelstein’s research showing that expansion of federal health insurance benefits was a major factor in driving national health inflation, and you have Kotlikoff’s showing that the growth in government-provided benefits has generally fueled per-capita cost growth not only in the US but throughout all advanced economies.  So step one here has to be to constrain the growth of uncapped federal support.

  • If you could bring about one health policy change during this Congress, what would it be?

Are we talking about the realm of the possible, or am I king for a day?   Our societal priorities seem to center around providing support for the typical health care expenses of those in need, and on providing protection from catastrophic health care costs for everyone, so that the lottery of life doesn’t randomly target some among us for suffering, despair, and the poorhouse.  If we scaled back the full smorgasbord of federal health programs to providing basic catastrophic insurance for everyone, and provided means-based support for routine expenses, then the rest of us could cover our own routine expenses and it would cost us a whole lot less to do so.  But this Congress obviously isn’t going to do that.  I suppose that as a trustee, the one thing I’d wish for is that we not proceed with plans to spend hard-won Medicare savings on this new health entitlement.  If we really want a new health entitlement, it should be able to finance itself; we shouldn’t finance that with the same money that we are claiming will extend the solvency of Medicare.

Interview with Charles Blahous-part 2

Part 2 of my interview with Charles Blahous focusing on Social Security. Here is Part 1.

Charles Blahous, one of the two public trustees for Social Security & Medicare, is a research fellow at the Hoover Institution, and was the Deputy Director for the National Economic council from 2007-09, and served as a Special Assistant to the President for economic issues from 2001-07. He is the author of Social Security: The Unfinished Work (2010: Hoover Institution).

Below are the questions I submitted to Dr. Blahous, along with his answers.

  • What is the problem(s) facing Social Security?

In a nutshell, it’s program costs that rise faster than the underlying tax base. And that in turn is driven by the increasing numbers of beneficiaries, the increasing number of years that they draw benefits, the decline in the ratio of workers to collectors, and the rise in per-capita benefit levels resulting from various program amendments over the years.  All of this is manifested in annual program cash deficits that began last year and which will rise dramatically under current law through the 2030s.  The other main problem I see in Social Security is the collapse of the previous bipartisan consensus as to what constitutes sound program financing.  The last time we fixed Social Security in 1983, there was a shared bipartisan understanding that the resources to pay benefits came from incoming worker tax contributions. There was indeed a Trust Fund then, but its assets weren’t counted when the trustees calculated the program’s actuarial balance, because no one seriously considered the Trust Fund as a viable means of covering for tax shortfalls for extended periods of time.  But now you have many people who point to the Trust Fund as the reason why we don’t need to fix anything until it technically runs out in 2036.  It would be nice if life were that simple, but it’s not.  If it were, all we’d need to do would be to issue more bonds to the Trust Fund and call it a day.  But the Trust Fund simply represents the program’s claim on future economic resources; it doesn’t answer the far more important question of what resources we are going to tap to pay those benefits.  I write a lot about this in my book, Social Security: The Unfinished Work.  We really have two problems in Social Security: we have a substantive problem — how do we finance the benefits of the baby boomers — which completely dwarfs the problem faced in the last Social Security crisis of 1983.  But we have an additional political problem, which is that our large Trust Fund on paper makes the problem appear, at least to some eyes, less urgent.  When you have a bigger problem to solve but a lesser sense of urgency about solving it, that’s a prescription for trouble.

  • What is the Social Security Trust Fund?

Mechanically, it consists of special-issue Treasury bonds.  Whenever Social Security runs a surplus, the surplus is used to finance current federal spending, and in return for the money the General Fund issues bonds to Social Security.  Those bonds earn interest at market Treasury bond rates.  They can be called in when needed to pay benefits.  In sum, the Trust Fund represents Social Security’s authority and financing source for benefit payments whenever incoming tax revenue falls short of benefit obligations.  That authority is limited by the size of the past surpluses Social Security has run, times interest.  Last year in 2010, Social Security tax revenue began to fall short of its benefit payment obligations.  But it could still send out the full benefit checks because there was a positive balance in the Trust Fund.  The Trust Fund really tells you just one thing; how long you can keep sending out the checks. It doesn’t tell you whether existing benefit schedules are equitable, whether existing tax schedules are equitable, where the resources will come from to finance benefit payments, nor when the projected shortfall has grown larger than the political system is still willing to solve.  There’s a tendency to read more information into the Trust Fund balance than is actually there.

  • It looks to me from the Social Security Trustees’ Report that the disability portion of the Social Security program faces an earlier date at which benefits must equal taxes received. Do I have this right, or can monies from OASI (Old Age Survivor Income) cross subsidize the DI (Disability) program?

You have it right.  Legislation could be enacted to reallocate revenues from OASI to DI, but that requires action by Congress.  DI would be insolvent by 2018 under current projections.

  • Are there any technical questions that need to be answered to enable Social Security reform, or are all the questions simply political in nature?

Per my answer on the Social Security Trust Fund, I believe there are a lot of important technical questions.  Sometimes there’s too much of an attitude of, “Hey, this is a relatively easy problem to solve,” without fully recognizing that some of the consensus has broken down on what solving the problem really means.  Do we want first of all to “solve the problem” for 50 years, for 75 years, for 100 years, or so that the central tendency of projections is indefinite sustainability?  After the 1983 reforms, we had a host of bipartisan technical panels and advisory boards come forward to say, in effect:  “It’s not enough just to have a non-zero trust fund if program operations are swinging wildly from large surpluses to even larger deficits, so make sure you’re ultimately on a path of annual balance.”  But increasingly you have some policy advocates just ignoring that standard because they don’t want to meet it.  The Simpson-Bowles commission, rightly in my view, produced a plan that met both standards of maintaining the trust fund and ultimately reaching annual cash balances.  If you don’t observe both, you allow for solutions that are really based on nothing more than budget gimmicks.  So, yes, we need to nail down agreement on those technical questions first.


The Interview concludes tomorrow with Medicare.