Making Sense of the 2012 Social Security Trustees’ Report

Filed Under (Retirement Policy, U.S. Fiscal Policy) by Jeffrey Brown on Apr 24, 2012

Yesterday, the Trustees of the Social Security and Medicare Trust Funds issued their annual report on the financial status of these entitlement programs.  These annual reports have been published for decades, and are generally recognized as the most credible, unbiased, and objective assessment of the long-run financial situation facing these programs.  I am going to focus on the Social Security program in this post.

Interest groups and policy analysts from across the political spectrum immediately issued press releases trying to spin the findings of the report.  Here are the first two that crossed my virtual desk yesterday:

The National Academy of Social Insurance (of which I was a member for many years before finally resigning over frustration at their defense of the status quo) issued a release spinning the report in the most positive light possible:  The 2012 Trustees Report shows that Social Security is 100 percent solvent until 2033, but faces a moderate long-term shortfall. In 2011, Social Security had a surplus – revenue plus interest income in excess of outgo – of $69 billion. Reserves are projected to grow to $3.1 trillion by the end of 2020 … While the trustees’ projections indicate that major changes are not needed, modest changes should be made in a timely manner and can bring Social Security into long-term balance.

In sharp contrast, the Committee for a Responsible Federal Budget issued a release stating:  “Today, the Social Security and Medicare Trustees released their 2012 report on the financial status of Social Security and Medicare, showing that reforms will be needed soon to make these programs sustainable … Social Security as a whole is on an unsustainable path … Social Security’s financial status has deteriorated significantly since last year’s report … Currently, Social Security is adding significantly to unified budget deficits. Not counting the payroll tax holiday this year and last year, the program is projected to run a $53 billion deficit in 2012 and $937 billion from 2013 through 2022.”

Both NASI and CRFB are highly respected organizations, yet the pictures they paint could not be more different.  So, who is right?  Is it possible to reconcile these two views?

Like last week’s post, in which I tried to cut through the rhetoric over the cost of the Affordable Care Act, this post tries to cut through the rhetoric over Social Security’s finances by using a fictitious debate.  And just like last week, the answer to “who is right?” is “It depends …”

Let’s focus on what appears to be a factual disagreement.  NASI says “In 2011, Social Security had a surplus.”  CRFB says “Social security is adding significantly to unified budget deficits.”

How can the program be running both a surplus and adding to the deficit?

The answer is that it depends on whether you think about interest on the Social Security trust funds as being income or not.  One’s views about the Trust Funds also help shed light on whether we should view Social Security as being in financial distress now (the CRFB view), or whether we still have two decades before we have any real problems (the NASI view).

How does the Trust Fund work?  (For this post, I am going to ignore the distinction between the retirement and disability trust funds – implicitly, I am assuming that Congress will simply re-allocate the payroll tax revenue across the two programs, as they have done in the past when needed).

Let’s go back a few years to the pre-financial crisis, say, 2007.  Suppose you earned $50,000 that year.  You and your employer each paid 6.2% of payroll into the system, for a total of 12.4%.  This was approximately $6,200 that the U.S. Treasury collected, and this money was designated for the Social Security Trust Fund.

Social Security took most of that $6,200 (just to keep that math easy, let’s say they took $5,200 of it), and paid it out to current retirees and other beneficiaries (such as disabled workers, widows, etc).  The remaining $1,000 was not needed in that year, so it was handed back to the U.S. Treasury.  In return, the U.S. Treasury issued a $1,000 special-issue U.S. Treasury bond to the Social Security trust funds.  Like other U.S. Treasuries, this one was backed by the full faith and credit of the U.S. government.

Now, back in 2007, like in most years in recent history, the U.S. government was running budget deficits.  Thus, the Treasury department basically took your $1,000 and used it to finance the government spending that we were doing in excess of the income tax revenue we were bringing in.  They did not actually invest the money in financial securities – rather, they spent it.  Of course, they still owe the $1,000 to the Social Security trust fund.

This has been going on for about three decades.  As a result, the Social Security trust fund now owns several trillion dollars’ worth of government bonds.  And the U.S. Treasury pays the trust funds interest on these bonds.

Today, to a first approximation, the entire $6,200 that a $50,000 per year worker and her employer pay into the system is all going to pay benefits.  So there are no more new deposits to the trust fund.  But the balance of the account is quite large, and is spinning off interest.

So here is the key question.  Should the interest that Treasury is paying to the Social Security trust funds be counted as income?  Here is how a discussion might go between NASI and CRFB representatives.  (Any misrepresentations of views are mine alone).

NASI: “Of course the interest should count as income.  The interest grows the trust funds, and the trust funds represent a legal claim by the trust funds that will be backed by the full faith and credit of the U.S. government.”

CRFB: “Yes, but while these bonds – and their interest – represent an asset to Social Security, they are a liability to the U.S. Treasury.  And because the Treasury spent that money rather than saving it, it is crazy to think that we should count this as income.  The interest payments are just an accounting fiction, not a real flow of money into the government as a whole.”

NASI:  “Ah, but the trust funds do represent real savings.  If the Treasury had not issued this debt to Social Security, they would have had to increase public borrowing.  So the Trust Fund balance represents money that the U.S. did not have to borrow – and that is a form of saving.”

CRFB: “But for decades, Congress used the Social Security surpluses to hide the deficits in the rest of the government.  As a result, Congress spent more money over the past few decades than they would have if they had not been able to hide the true cost of their profligacy behind a unified budget framework.”

NASI: “There is no way to know for sure that the Social Security surpluses led to increased spending by Congress.”

CRFB: “Ah, but there is – at least two academic studies (here and here) have shown that this is exactly what happened.”

NASI: “Academic studies aside, there is no question that we should count this interest.  And if we do count it, it is clear that Social Security is running a surplus.  It is also clear that the program can pay 100% of promised benefits at least until 2033.”

CRFB:  “But that is a narrow perspective.  We care about the government budget as a whole – not just the narrow question of the Trust Funds.  From that perspective, what we know is that the amount of money we are collecting in payroll taxes today is no longer enough to cover the payments to beneficiaries.  The days of cash flow surpluses are gone.  And because interest on the trust fund is just one arm of government (Treasury) making a paper transfer to another arm of government (the Trust Funds), this does not represent real income to the government as a whole.  As such, the program is in dire straits, and needs to be fixed now.”

That fictitious debate roughly captures the economic disagreement underlying these two very different assessments of the latest Trustees’ Report.

I happen to support the CRFB view that the problem is serious, that we need to address it sooner rather than later, and that there is no pain-free solution.  But at the end of the day, it is impossible to fully refute the NASI view because we cannot go back in time and re-run an alternate history to know how spending would have responded in the absence of past Social Security surpluses.

Regardless of which view one holds, it is becoming increasingly difficult to deny the existence of a financing problem.  Even if you take the NASI view that we do not have a problem until the trust funds run dry in 2033, it is worth noting that this date is quite a bit earlier than what has been previously estimated.  Furthermore, 21 years is not a very long time when we are talking about a retirement program.  After all, nearly half of today’s 65-year olds will still be alive in 2033 and relying on Social Security benefits.  Today’s 46-year olds will reach their normal retirement age in 2033.  And today’s college students will be nearly half-way to their own retirement age.  We need to make changes now – so that we have time to phase-in the changes gradually and to allow individuals to adjust.

So, regardless of one’s views about the trust funds, it seems obvious to me that the real story behind the release of the Trustees’ Report is that the problem is real, it may be larger than we previously thought, and that it is not going to go away on its own.

Making Sense of the War of Words over the Cost of Obamacare

Filed Under (Health Care, Retirement Policy, U.S. Fiscal Policy) by Jeffrey Brown on Apr 18, 2012

A war of words (and numbers) has broken out in the policy wonk world over the effect of Obamacare on the deficit.  It is important, entertaining, and confusing.  This blog attempts to bring a bit of clarity to the debate.    

 It began last week with an article, written by Charles Blahous and issued by the Mercatus Center, that argued that Obamacare increased the deficit.  The piece was discussed in the Washington Post (and on my blog) on the day it was issued.

It took almost no time at all for Paul Krugman to denounce the study.  He first began, in typically unfortunate fashion, by attacking the credibility of the author through a suggestion that Blahous was just another Koch-funded crazy who should not be believed.  He then went on to make a slightly more substantive argument about the fact that Blahous’ result rested upon a view (that Krugman called “bogus”) about what Obamacare spending should be compared with.

Blahous publicly responded, defending his position.  A few days later, former CBO Director and former OMB Director Peter Orszag joined the broadside attack against Blahous.  Peter also joined in the credibility attack and went on to also attack Blahous’ choice of baseline. 

So who is right?   The point of this post is to try to provide a bit of clarity on the issue. 

Before proceeding, I should disclose my own personal biases.  First, I consider both Chuck Blahous and Peter Orszag to be personal friends – and I believe both would agree with that assessment.  I have known and worked with both of them for over a decade.  I have an incredibly high level of respect and admiration for both Chuck and Peter as public servants, as intellectuals, and as individuals.  This is not the first time they have publicly tangled (they did so frequently over Social Security reform).  Ideologically, I almost always find myself on the same side of issues as Chuck.  But Peter is an outstanding economist, and when his views are also echoed by other highly respected economists like David Cutler of Harvard (one of the most highly respected health economists in the world, who engaged in a debate with Chuck on my Facebook page), I often find myself temporarily in a state of cognitive dissonance.  When this happens, I try to figure out the core reason for the disagreement.  Is it different values (e.g., perhaps one cares more about redistribution and the other more about economic efficiency)?  Is it different assumptions (e.g., fundamentally different views about how the politics will play out or on how future health costs will evolve?)  In such cases, two very smart people can disagree on policy, without either being “wrong.”

But this debate seems different.  This is – or at least should not be – an ideological debate.  The question here is deceptively simple.  It is a debate over a “fact.”  Either Obamacare increases the deficit, or it does not. 

So who is right?

The correct answer is “it depends.”

To understand the long-term effect of any public policy change, one must first ask the question “compared to what?”  And this is where Blahous and Krugman/Orszag differ.

The following is a FICTITIOUS conversation between Blahous and his critics.  I am trying to be clear on their views.  The material in “quotes” is taken from their writing.  The rest is my own attempt to explain their views, and I alone am responsible for any misattributions.  The Orszag quotes can be found hereThe Krugman quotes are here.  Blahous’ views can be found in his original paper, his follow-up post on Forbes, and a new post at E21.  The use of the term “Obamacare” is mine.    

Me:  “If I look at the new spending programs under Obamacare, and compare that to any spending reductions or tax increases under Obamacare, does the program increase or decrease the deficit?”

Blahous:  Over the next ten years, the increases in spending from Obamacare – Medicaid/CHIP, new exchange subsidies, making full Medicare benefit payments for an additional eight years, etc. – exceed the ways that it reduces spending or raises taxes by $346 billion through 2021.  (This is based on a CBO projection of $352 billion adjusted slightly by Chuck.)

Krugman:  This is just “another bogus attack on health reform.”

Orszag:  Indeed.  The cost savings exceed the new costs by $123 billion through 2021.   

Blahous:  But you are both ignoring the cost of extending the solvency of Medicare!  One of the effects of Obamacare is to extend our full financing commitment to Medicare through 2024.  This costs money.  Add up all the things the legislation does, and it is $346 billion more than the legislation’s cost-savings.

Orszag:  This is a “trick.”  The Blahous analysis “begins with the observation that Medicare Part A, which covers hospital inpatient care, is prohibited from making benefit payments in excess of incoming revenue once its trust fund is exhausted. He therefore argues that the health reform act is best compared to a world in which any benefit costs above incoming revenue are simply cut off after the trust-fund exhaustion date. Then, he argues that since the health-care reform act extends the life of the trust fund, it allows more Medicare benefits to be paid in the future. Presto, the law increases the deficit by raising Medicare benefits.” 

Blahous:  Look guys, this is really simple.  Without the ACA, Medicare would have been insolvent in 2016.  Under the new legislation, we are making a binding commitment to make full benefit payments through 2024.  These are real payments to real people.  How can you ignore the extra commitments through 2024?  After all, you claim the Medicare solvency extension as one of the achievements of the ACA.

Krugman:  “OK, this is crazy. Nobody, and I mean nobody, tries to assess legislation against a baseline that assumes that Medicare will just cut off millions of seniors when the current trust fund is exhausted.”

Blahous:  But under a literal interpretation of current law – which is how most budget scoring is done in Washington – a law that extends Medicare for additional years would be scored as a cost.  Do you acknowledge that under a literal change in law, this legislation puts us $346 billion deeper in the hole? 

Krugman:  The literal law does not matter.  Everyone knows that Congress is not going to allow Medicare benefits to be slashed in 2016.  To suggest these costs are a cost of Obamacare is misleading.  “In general, you almost always want to assess legislation against ‘current policy’, not ‘current law’; there are lots of things that legally are supposed to happen, but that everyone knows won’t, because new legislation will be passed to maintain popular tax cuts, sustain popular programs, and so on.

Blahous: But we have to abide by these budget rules in other contexts.  For example, let’s look at the Alternative Minimum Tax. The Congressional Budget Office counts the revenue from the AMT in its baseline budget projections, even though it knows full well that Congress is likely to continue to provide AMT relief before that revenue is collected.  Similarly with the “doc fix” in Medicare!

Orszag:  Yes, but by your logic, if we just assume that Medicare benefits are cut when the trust fund runs dry, or that Social Security benefits are cut when its trust fund runs dry a few decades later, then we do not have a long term budget problem!  Indeed, Chuck, you are “far too modest. The government is not legally allowed to issue any debt above the statutory limit, so (you) should have assumed the deficit would disappear when we reach that limit at or around the beginning of next year.”

Blahous:  Look, when you make Medicare benefit payments, real money leaves the US Treasury.   We can’t send the same check to Medicare and to Medicaid.  If you want to take credit for all the benefits of the ACA – one of which was to extend Medicare – then you have to account for the Medicare commitments as well as the Medicaid ones.  Even if you don’t think we would have allowed benefits to be suddenly cut, historically Congress has always enacted other savings to avert Medicare insolvency.  And, now that Medicare solvency is extended through 2024, the pressure on Congress to enact further savings is reduced.  So it’s not only as a matter of literal law but as a matter of practical budgetary behavior that the ACA worsens the outlook.  No matter how exactly you think things would have played out under prior law, this legislation still worsens deficits by $346 billion relative to prior law.

Krugman:  Don’t believe any of this.  The Mercatus Center is funded by the Koch brothers.  The Koch brothers, by golly!!

Blahous:  Look guys, I am trying to make a real point here, not engage in character assassination.  If carried to its logical conclusion, this is not only a departure from interpreting actual law, it is also fiscally dangerous.  You guys are basically saying that there are no prior law restraints on Medicare spending.  So every time we extend the program’s solvency, it does not cost anything!  

Me:  Okay, guys, thanks for clearing that up.  I understand it all so much better now. 


So there you have it.  A knock-down, drag-out battle over budget baselines.  The debate is not over the cost of things like the coverage mandate.  It is a debate over the proper way to account for an extension of Medicare’s solvency. 

To summarize:

Relative to a world where Medicare expenditures are brought into balance with revenues within the next few years (which does appear to be required under the literal reading of current law), ACA increases Medicare expenditure and the deficit.  This is the Blahous view.   

Relative to a world in which we project current practice forward, ACA reduces Medicare expenditure and the deficit.  This is the Krugman and Orszag view. 

I think most reasonable people can understand both points.  And I don’t think this really calls for name-calling and credibility-questioning.  But in Washington, that is what passes for debate.

Most ordinary people probably think that what we should be doing is making some cuts, but not cut so deeply as to eliminate the entire Medicare shortfall.  If so, the effect on the deficit is better than if we did nothing, but worse than if we solved the problem. 

So most people probably think the “truth” (whatever that means in this context) lies somewhere in the middle.

How the Supreme Court can Reduce the Deficit: The Fiscal Impact of Ending Obamacare

Filed Under (Health Care, U.S. Fiscal Policy) by Jeffrey Brown on Apr 10, 2012

West face of the United States Supreme Court b...
(Photo credit: Wikipedia)

Last week’s U.S. news was dominated by the oral arguments before the Supreme Court of the United States (SCOTUS) on the constitutionality of the Patient Protection and Affordable Care Act (PPACA), also known more succinctly as the Affordable Care Act (ACA), or, simply, “Obamacare.”  Most of the news coverage revolved around legal issues, such as how to define a “limiting principle” that would distinguish health insurance from other goods and services.  A few of those analyses, including one by my colleague Nolan Miller at the University of Illinois, provided useful economic insights on these legal questions.

But what I have not seen much of – until now – is a careful analysis of the impact of repeal on the federal budget.  Yes, there is plenty of rhetoric around this topic, with Democrats arguing that PPACA saved money and Republicans arguing that it created a huge new entitlement.  But there has been very little careful analysis.

That changed today, when the Mercatus Center at George Mason University released a meaty new report written by Charles (“Chuck”) Blahous.  His analysis shows quite clearly that the Supreme Court now finds itself in the position of having an enormous impact on the long-run fiscal situation in the U.S.

As background, Chuck Blahous is one of two public trustees of the Social Security and Medicare trust funds, having been appointed to this post by President Barack Obama and confirmed by the U.S. Senate.  Previously, Chuck served all eight years of the G. W. Bush administration at the National Economic Council.  After spending over two decades in both the legislative and executive branches of the U.S. government, Chuck knows the ins and outs of federal budgets.  He is also widely respected on both sides of the aisle as a serious policy analyst.

In a nutshell, here is what Chuck’s careful analysis finds:

  1. PPACA is expected to increase net federal spending by more than $1.15 trillion over the next decade.
  2. PPACA is likely to add more than $340 billion, and perhaps as much as $530 billion, to federal deficits over the next decade.
  3. Despite these realities, government scorekeeping rules lead to deep confusion over the fiscal impact, and have the effect of making PPACA appear less expensive than it really is.

How can this be?  In part, the law “relies upon substantial savings already required under previous law to maintain the solvency of the Medicare Hospital Insurance (HI) Trust Fund.  These do not represent new net savings … but substitutions for spending reductions that would have occurred by law in the absence” of this act.  There are other issues at play as well.

All of this is “public,” in the sense that it has been disclosed in scoring documents by the Congressional Budget Office (CBO). But the CBO is constrained to report the effect of government tax and spending programs according to various scoring rules – even when those rules deviate substantially from the likely political or economic reality.  Skilled politicians have learned to use these scoring rules to their advantage.

As Chuck points out in his paper:

“A full understanding of the ACA’s budget effects requires appreciation of the distinction between two important points:

  1. CBO found that the ACAD would reduce federal deficits when a specific scoring convention was applied;
  2. The same analysis shows implicitly that the ACA would substantially increase federal deficits relative to previous law.

The paper is over 50 pages in length (including the helpful Q&A in the appendix), but is well worth a read if you want to know the details behind the calculations.

But if you don’t have time to read it, here is the bottom-line: “Taken as a whole, the enactment of the ACA has substantially worsened a dire federal fiscal outlook.  The ACA both increases a federal commitment to health care spending that was already unsustainable under prior law and would exacerbate projected federal deficits relative to prior law.  This is an unambiguous conclusion …”

Were the Supreme Court to strike down all or part of this Act, we should view it as an opportunity to revisit health care reform in a way that reduces, not increases, public spending.

Is Obama the “Damn Politician” that FDR Warned About?

Filed Under (Retirement Policy, U.S. Fiscal Policy) by Jeffrey Brown on Feb 17, 2012

As posted on Forbes yesterday …

In 1941, President FDR explained why he chose to fund Social Security through a payroll tax in as follows:

“We put those payroll contributions there so as to give the contributors a legal, moral, and political right to collect their pensions and their unemployment benefits. With those taxes in there, no damn politician can ever scrap my social security program”

For more than seven decades, FDR’s strategy has proven effective.  Talk to someone in or near retirement – even people who consider themselves small government conservatives – and you will often hear them state that they have a right to their Social Security benefit because they paid for it over their working life.

President Roosevelt knew that the key to the political sustainability of Social Security was the establishment of an entitlement mentality, and the key to establishing an entitlement mentality was the linkage between payroll contributions and benefits.  If Social Security were structured as a means-tested welfare-style program – that is, it if were financed by a progressive income tax rather than through payroll contributions – it might have never lasted this long.

Given this, it is important that President Obama and Congress have just agreed to extend the payroll tax cut and to continue to use budget gimmickry to turn Social Security into a partly general-revenue-financed program.

Here is how it works.  The 2% payroll tax cut reduces revenue to Social Security by about 15 percent.  But Social Security does not have a spare 15 percent of revenue lying around: rather, it is currently running quite close to break-even on a cash flow basis, and faces enormous long-run deficits.  To get around this, President Obama and Congress have decided to replace the lost payroll tax revenue by transferring money from general revenue (which derives primarily from the income tax) into the Social Security trust funds.

This budget gimmick has the short-term political benefit of making the Social Security trust funds seem unaffected by this tax cut.  But it also means that we are deviating substantially from FDR’s vision of a retirement program being paid for (on a pay-as-you-go basis) by participant contributions.  By moving down the path of general revenue financing of Social Security, we achieve the short-term “progressive” aim of increasing the degree of income-based redistribution (because income tax rates rise with income, whereas payroll tax rates do not).

But in the long-run, this has the potential to erode political support for the program.  By shifting the funding burden onto the income tax, the program starts to look more like a welfare program than a contributory social insurance program.

I am not the first to notice the irony of this.  My very good friend Chuck Blahous, who served eight years in the National Economic Council for President George W. Bush, and who was appointed by President Obama as one of two Public Trustees for Social Security, just released a paper explaining why this payroll tax cut is bad policy.  Among the seven reasons he provides is that doing so destroys the “historical Social Security compact.”  In a Washington Post article back in December, Dr. Blahous stated that these budget gimmicks are “a grave step for Social Security.”

This view is not limited to experts on the Republican side: the other Public Trustee of Social Security (a Democrat) – Robert Reischauer, the highly respected president of the Urban Institute — agrees with Dr. Blahous.  While Reischauer was more sympathetic to the tax cut, he also noted that it “could, if it continues for a substantial period of time, undermine one of the foundational arguments that makes the Social Security program inviolate.”

Perhaps the most succinct summary of the irony comes from Jason Fichtner, a Senior Research Fellow at the Mercatus Center and former Chief Economist and (acting) Deputy Commissioner for the Social Security Administration.  He summed it up the situation quite succinctly in an email to me by noting that “in 15 years we might look back on this time in history and discuss how President Obama, as a Democrat, was the president that started the path to killing Social Security.”

So, maybe President Obama really is the Damn Politician that FDR was worried about?


Fiscal Sustainability AND Retirement Security: A Reform Proposal for the Illinois State Universities Retirement System (SURS)

Filed Under (Retirement Policy, U.S. Fiscal Policy) by Jeffrey Brown on Feb 9, 2012

I have released a paper today that proposes a new plan for the State Universities Retirement System.  Co-authored with Robert Rich, the Director of IGPA, the paper proposes a hybrid system that would be partially funded by both workers and universities. It contains several components that reflect some of the ideas that have been publicly discussed by state leaders in recent weeks.

 The proposal has four basic components: 

1) Create a new hybrid retirement system for new employees that would combine a scaled-down version of the existing SURS defined benefit plan with a new defined contribution plan that would include contributions from both employee and employer; 

2) Peg the SURS “Effective Rate of Interest” to market rates; 

3) Redistribute the SURS funding burden to include a modest increase in employee contributions and new direct contributions from universities, thereby reducing state government’s burden on state government; and

4) Align pension vesting rules with the private sector, which would decrease the years new employees hired after January 1, 2011 would need to work for their pension benefit to be vested.

The plan is intended to substantially reduce state expenditures on public pensions, while still providing a reasonable source of secure retirement income to university employees. 

Click here to read the full paper.

Reducing Regulatory Obstacles to Retirement Income Security

Filed Under (Retirement Policy, U.S. Fiscal Policy) by Jeffrey Brown on Feb 7, 2012

With nearly 80 million baby boomers starting their march into retirement, many policy-makers have begun to focus on how to provide secure retirement income in a fiscally sustainable way.  This is no small challenge in an era of enormous deficits.

Although Social Security plays an important role in providing income that retirees cannot outlive, the benefits provided by Social Security are insufficient to ensure that most retirees can maintain their pre-retirement living standards.  However, increasing these benefits would be horrible fiscal policy: because the pay-as-you-go nature of Social Security has collided with an aging population, this program faces enormous fiscal problems that are going to require reductions – not increases – in the rate at which benefits grow.

Thus we have two opposing forces: a need for more retirement income, and a need to cut government spending on entitlement programs like Social Security.  What can be done?

Fortunately, the private sector can play an important role here, but only if the regulatory environment allows for it.  Presently, the regulatory landscape surrounding employer sponsorship of retirement plans is burdensome and enormously complex.  In many cases, the best thing the government can do to promote a greater role for the private sector in providing guaranteed retirement income is to “get out of the way.”  Ironically, however, there are other instances in which the best way the government can promote private sector solutions is to get more involved – if only by providing guidance on how plan sponsors can improve their plans without running afoul of existing regulations.  Getting guaranteed income options into 401(k)’s and other retirement plans is one such case.

In recent years, the financial services industry has increasingly focused on how to provide plan sponsors and plan participants with products that help to provide guaranteed lifetime income.  The resulting innovation over the past decade has been impressive, as companies have introduced a wide range of insurance and investment products that provide individuals with lifetime income.

However, employers that sponsor 401(k) plans have been slow to adopt.  As a result, most 401(k) participants in the U.S. still do not have access to annuities or other income products in their plans.  Although there are many reasons for this, there is little question that part of the reluctance of plan sponsors to provide annuities is that they have been scared off by regulatory and fiduciary concerns.

Last week, the Treasury Department proposed guidance to help address a few of the many issues that stand in the way of better private sector retirement plans.

In a nutshell, the proposed guidance does three things:

First, it makes it easier for plan sponsors to allow retirees to have a mix of lump-sum and annuity choices.  Put simply, it makes very little sense for most retirees to annuitize either 0% or 100% of their retirement assets.  Annuities provide guaranteed income, help to protect against out-living one’s assets, and help to guard against market volatility.  On the other hand, having some non-annuitized wealth available is extremely valuable when faced with uncertain expenses such as for long-term care.  Given that the optimal financial plan for most individuals would be to have some of both (e.g., annuities and a lump-sum), it only makes sense for our regulatory infrastructure to encourage this.

Second, a number of academic papers have established the potential value of annuity products that have a deferred payoff structure.  That is, with a small fraction of one’s wealth at, say, age 65, one can buy a product that will start paying income at age 85.  In the industry, these are sometimes called “longevity insurance” (although the name is very unfortunate, because all life annuities – whether they are deferred or not – are providing insurance against the financial costs of longevity).  The proposed regulatory guidance would help ensure that these products are more easily available.

Third, Treasury issued two “revenue rulings” that clarify how rules designed to protect employees and their spouses apply when a plan offers an income option.

These rules are useful, but far from sufficient.  Looking ahead, plan sponsors and participants would be better off if policymakers also took at least three additional steps.

First, the Department of Labor needs to provide much greater clarity about how plan sponsors who wish to provide lifetime income options can do so while protecting themselves from fiduciary risk.  This could include providing a “safe harbor” rule for the selection of the annuity provider.  Too many plan sponsors continue to be spooked off by the specter of fiduciary liability if they choose an annuity provider that runs into financial distress in the future.

Second, Congress should reform the Required Minimum Distribution rules to eliminate the various implicit and explicit barriers to lifetime income.  These rules were written by tax lawyers to ensure that the IRS could eventually get its hands on tax-deferred savings.  If these rules were instead written with an eye towards retirement income security, they would look quite different.

Third, we should encourage plan sponsors to report 401(k) and other defined contribution (DC) balances in terms of the monthly income the plan will provide, rather than simply as an account balance.  The Lifetime Income Disclosure Act that received bipartisan sponsorship in the U.S. Senate last year would be a positive step in this direction.  (My Senate testimony on this Act can be found by clicking here).

This need not be a partisan issue.  Republicans should recognize that strengthening retirement income security in our private pension system will give us more freedom to address our burgeoning Social Security deficits.  Democrats should view this as an opportunity to ensure that employers “do the right thing” by providing retirement plans to employees that actually succeed in providing a secure retirement.

Warren Buffett is not the Oracle of Public Finance

Filed Under (Finance, U.S. Fiscal Policy) by Jeffrey Brown on Feb 1, 2012

It is being reported today that Senator Sheldon Whitehouse (D-R.I.) is introducing a bill that would impose a minimum 30% tax on individuals earning more than $1 million per year.  This type of tax policy – which is essentially a new version of Alternative Minimum Tax – has been dubbed the “Buffett Rule” due to the news last year that Warren Buffet had a lower tax rate than his secretary.

Warren Buffett claims to have a tax rate of 17.4 percent.  His claim, however, is only true if one ignores one of the most basic economic principles of tax analysis: that the person who writes the check is not necessarily the same as the person that bears the economic burden of a tax.  In economics, this distinction is known as the difference between “legal incidence” (i.e., the entity with legal responsibility for paying taxes) and “economic incidence” (i.e., a measure of who really bears the economic burden of the tax).

In almost any undergraduate public finance textbook, one can find simple examples of how these concepts diverge.  For example, politicians often make a big deal of the fact that the FICA payroll taxes used to support Social Security and Medicare are split evenly between employers and employees.  But economists tend to believe that nearly all of the economic burden of the payroll tax falls on workers.  In other words, even though employers pay their share of the FICA tax, in the long-run the result is that workers are paid less than they would be paid in the absence of the tax.  Thus, it is the workers and not the firms who are truly paying the tax, in spite of how it appears.

The discussion around Mr. Buffet’s taxes – as well as the more recent discussion around the release of Governor Romney’s tax returns – has completely missed this point.  Those discussions have focused solely on the legal incidence of the personal income tax system, and have failed to think through the economic incidence of the overall tax system.

How so?  It is not uncommon for wealthy individuals like Mr. Buffett to receive much of their income in the form of dividends and capital gains.  This type of income may appear as if it is receiving “preferential” tax treatment, but the reality is that it is taxed heavily.  This is driven by the fact that corporate income is taxed at the corporate level before it is available to be paid out as dividends (or used to repurchase shares, which can lead to capital gains for investors who retain their shares).  The U.S. imposes a very high – 35% – marginal tax rate on corporate income.  Thus, if a firm earns another $1000, it pays $350 in taxes, leaving only $650 to go to shareholders.  If those shareholders are then taxed at a 15% rate, that is another $97.50 that goes to the government.  This leaves only $552.50 in the pockets of shareholders for every $1000 of pre-tax earnings that are paid as dividends.  Thus, the effective marginal tax rate on this income is more like 47.5% than it is 15%.

Of course, there are at least two important caveats to this stylized example.  First, the economics profession has simply not been able to come up with a definitive estimate of who really bears the burden of the corporate income tax.  One of the leading tax scholars of our day – Alan Auerbach of the University of California at Berkeley – wrote a terrific summary of what we know on this topic back in 2005 (the paper, which was ultimately published in the NBER Tax Policy and the Economy series, is available as an NBER working paper here.)  He notes that one of the major lessons is that “for a variety of reasons, shareholders may bear a certain portion of the corporate tax burden … thus, the distribution of share ownership remains empirically quite relevant to corporate tax incidence analysis.”  This is hardly a ringing endorsement that we should assume the entire incidence falls on Warren Buffet and other shareholders, but it is quite clear that we should not be ignoring corporate taxes when making policy statements about the fairness of the tax system.

A second caveat is that not all corporations face a 35% marginal effective tax rate.  Corporate income taxation is nothing if not a complex labyrinth of rules, exceptions, and exceptions to the exceptions.  Again, however, we know that for most corporate earnings, the rate of corporate taxation is well above zero, which is the rate it would need to be for us to feel as if we can ignore it when making statements of the kind Mr. Buffett makes.

A fellow Forbes contributor, Josh Barro, points out a number of problems with the Buffett Rule, the most important of which is that it would exacerbate the already-existing tax distortion that favors debt over equity.  If Congress wants to do this, that is their prerogative.  But we should not allow them to justify potentially bad tax policy on the basis of a naïve and misleading understanding of tax incidence.

How Teachers can be Both Undervalued and Overpaid

Filed Under (Other Topics, U.S. Fiscal Policy) by Jeffrey Brown on Jan 24, 2012

In recent months, there has been a spirited debate about the value of teachers and whether they are undervalued or overpaid.  The point of this post is to explain how both statements can be simultaneously true.    

 Before continuing, I should disclose that I come from a family of teachers.  My mother and father were both public high school teachers.  My niece, who I adore, is about to enter the profession after she graduates later this year.  If that were not enough to bias me in favor of teachers, I count myself among the millions who attribute much of my success in life to a handful of incredible educators who really made a difference in my life.  And I am impressed on a daily basis with the phenomenal teaching that my children receive. 

 Many of those in the teaching profession feel that they are under-valued by society.  And given some of the recent political rhetoric, their beliefs are not unfounded. 

 At the same time, my good friend Andrew Biggs of the American Enterprise Institute has written a number of highly interesting and provocative pieces arguing that public school teachers are over-paid.  And, as an economist, I find many of his points quite persuasive.        

 So how do I reconcile these views?  After giving this issue much thought, I have concluded that teachers are undervalued.  But this does not necessarily mean that they are underpaid. 

 Let me begin in a roundabout way by reminding readers of what economists call the “diamond water paradox.” 

 Water is critical for life.  Yet in the United States, we treat water as virtually free: when was the last time you were required to insert a few quarters to get a drink from a water fountain? 

 In contrast, diamonds serve very little practical purpose to individuals (I am ignoring industrial uses of diamonds as well as any romantic purposes).  They certainly do not help keep us alive.  And yet they are enormously expensive. 

What is going on here?  In a nutshell, water is plentiful, while diamonds are scarce.  Thus, while water creates enormous social value, it is relatively inexpensive “on the margin.”  This does not mean it is not valuable overall!  To put it in the language of economics, water generates enormous amounts of “consumer surplus.”  This means that, overall, the value that water creates for society far exceeds its market price. 

Yet few would argue that we should have to pay diamond-like prices for water just to prove to water that we understand how much more important it is to our lives than those silly, useless diamonds.

Like water, good teachers add enormous value to society.  This is not just a “feel good” statement written by the son of teachers:  there is strong empirical evidence to back it up.  Most recently, a new NBER working paper by three brilliant economists (Raj Chetty, John Friedman and Johah Rockoff) is able to link school district data on 2.5 million kids to the tax returns of these children many years later (after they have grown up and entered the labor force).  Their results show that good teachers create enormous social value.

The authors state:  “Replacing a teacher whose [value-added] is in the bottom 5% with an average teacher would increase students’ lifetime income by more than $250,000 for the average classroom in our sample.”

Think about that for a moment:  in a single year, an average quality teacher (relative to a low quality teacher) can create a quarter million dollars of economic value!  Needless to say, that is far above what we pay our good teachers.  Like water, teachers are generating enormous “consumer surplus” for the students and communities that they serve.

Those who advocate for teachers sometimes use data like the above to argue that teachers are underpaid.  Indeed, in a recent NYT piece, David Hambrick wrote “where I live, the average starting pay for a teacher is about $20 per hour.  A bartender can make double that.  Which job is more important?” 

While the NBER researchers referenced above did not measure the lifetime economic value created by good bartenders, I am going to simply assert that “teachers are more important.” 

But does that mean teachers are underpaid? 

Not necessarily. 

Just like we do not pay diamond-like prices for water even though water is more important, neither should we necessarily pay teachers more than plumbers, bartenders, or professional athletes. The reason, to be blunt, is that we do not have to, because we have plenty of smart, dedicated individuals that are willing to teach at current compensation levels.

My father – who dedicated 30 years of his life to teaching high school social studies – once reposted a quote on Facebook along the lines of “teachers are not in it for the income, we are in it for the outcome.”  

I believe that.  Indeed, many of the best teachers that I have had in my life were teachers because they loved the job.  They believed what they were doing made a difference in the lives of the students and their communities.  And they were right.

But, in the cold, hard logic of economics, these intangible benefits to teaching are precisely why we have so many people willing to teach, even though the compensation is far below the value-added to society.  The same is true for other professions that attract passionate, selfless, and altruistic individual, such as social work or those who work in the not-for-profit sector.  Fortunately for society, but perhaps unfortunately for those in these professions, there are plenty of people motivated to do these jobs in spite of the pay. 

Some will say this is “not fair.”  That may or may not be true, depending on one’s definition of fairness.  But is it rational?  Absolutely.  Almost nobody likes to pay more than necessary to obtain the goods and services they value, and that includes education.  And if the calculations of Biggs and Richwine even approximately correct, taxpayers may be paying more than necessary to attract and retain the teachers that we have.            

Like water, good teachers are critical for meaningful human life.  They create tremendous value for society.  And we should respect and honor them for what they do.  But it does not necessarily mean that we should pay them like diamonds.    


Three Strikingly Different GOP Visions about Social Security Reform

Filed Under (Retirement Policy, U.S. Fiscal Policy) by Jeffrey Brown on Jan 17, 2012

At the Republic debate in Myrtle Beach, SC last night, in response to a question by Gerald Seib of Wall Street Journal, three of the candidates weighed in on Social Security reform.  Their responses revealed strikingly different approaches to economic policy.

Governor Romney took the practical approach.  After pointing out that he would protect everyone over the age of 55, he laid out two very specific changes to the benefit formula that would substantially reduce Social Security expenditures in the decades to come.  The first change would change the way that initial retiree benefits are calculated.  Under current law, benefits from one cohort of retirees to the next rise with average wages in the economy.  Governor Romney suggested, instead, a plan similar to what Social Security policy experts call “progressive price indexing.” This would continue to index starting benefits to wage growth for those at the bottom of the income distribution, but would index benefits at the top end of the income distribution to price inflation instead.  Because prices tend to rise less quickly than wages, this would reduce expenditures relative to current law.  The impact would be gradual – and thus the short-term cost savings would be limited, but over many decades can be quite substantial.  Second, Governor Romney indicated a willingness to increase the full retirement age by one or two years.  Importantly, increasing the full retirement age does not actually require that anyone work longer: rather, it simply moves the age at which one receives “full” benefits back by one to two years.  Variants of both of these reform proposals have been floating around Washington over the past decade.  In essence, this is a fiscally responsible approach that recognizes there is no pain-free way to fill in the fiscal gap.  While this is good fiscal policy, whether or not it is good politics remains to be seen.

In sharp contrast to Romney’s “eat your spinach” approach to reform, Speaker Gingrich suggested that we follow the “all dessert” approach to Social Security.  Rather than being upfront about the need for politically difficult changes to taxes or benefits, Speaker Gingrich suggested that the government can guarantee retirees that they can receive full promised benefits without paying a dollar more in taxes, despite the existence of a multi-trillion dollar shortfall.  How does he propose we do this?  By allowing workers to shift 100 percent of the employee payroll tax contribution (currently 6.2 percent of payroll) into personal accounts, leaving the 6.2 percent employer contribution going into the existing system.  Citing the examples of Chile and Galveston, Speaker Gingrich argues that people will not have to sacrifice any benefits.  As I discussed last week, however, he fails to acknowledge the huge implicit liability he is imposing on taxpayers by essentially guaranteeing that stocks will perform close to their average historical values.  They might, but to guarantee this without acknowledging the real economics cost is both fiscally reckless and intellectually dishonest.

Former Senator Rick Santorum used most of his response to correctly point out another fact about the Gingrich proposal: namely, that by diverting 6.2 percent of payroll into the personal accounts, we will have to borrow additional money to back-fill the missing payroll tax revenue, nearly every penny of which is now going to pay current retirees.  And the Speaker’s statement that we can somehow fill this gap by eliminating the overhead associated with consolidating anti-poverty programs is mathematically ridiculous.  Those numbers don’t even come close to adding up.

This is quite a different situation than we faced a decade ago when the President’s Commission to Strengthen Social Security (on whose staff I served) recommended personal accounts at a time when Social Security was projected to have another 15-plus years of surpluses.  One of the key rationales for personal accounts a decade ago was to ensure that those surpluses were saved, rather than redirected to underwrite other government spending.  The Commission plans also envisioned smaller accounts, further reducing the need to fund a transition investment.  Even so, the plan still had to come up with substantial short-term revenue to cover the transition, an aspect that contributed to the proposal’s demise.  Unfortunate, “carve-out” personal accounts – which I have supported in the past – is an idea whose time has come and gone.

Aside from criticizing Speaker Gingrich, Senator Santorum offered few specifics.  He did endorse means-testing, noting that we should reduce or eliminate benefits for the 60,000 retirees who earn over $1 million per year.  This is a perfectly reasonable suggestion, albeit with two problems.  First, if high earners receive no benefit whatsoever for paying into Social Security, then this converts the 12.4 Social Security payroll contribution into a pure tax, with all the associated efficiency losses.  Second, the money saved is a “drop in the bucket” compared to the size of the projected Social Security shortfalls.  Assuming that every one of those 60,000 millionaires gave up 100 percent of their benefits, this would save only a few billion dollars a year.  This is real money, but when one looks at the size of the expected annual Social Security shortfalls that we will face in another 20 years, we need dozens – if not a hundred – money saving ideas of this magnitude.

Thus, what we have witnessed are three fundamentally different approaches to Social Security reform.  One candidate who puts forward real meaningful solutions and is therefore criticized for not being sufficiently bold, one candidate who promises a free lunch at taxpayer expense, and one candidate who appears not to have put together a plan sufficient to the task ahead of us.  Only time will tell how voters respond to these three different narratives.

Disclosure: Over the past few weeks, I have begun to offer informal, unpaid advice to the Romney campaign’s policy staff on issues related to Social Security.  All opinions expressed in this blog, however, are mine alone.

This blog is cross-posted, with permission, at


The Economics of Jury Pool Representativeness

Filed Under (Other Topics) by Jeffrey Brown on Nov 14, 2011

The United States Constitution guarantees individuals the right to a trial by a jury.  To implement this right, however, one must find jurors.  Many argue that one characteristic of a good pool of jurors is “jury representativeness,” i.e., a set of jurors that is roughly representative of the community from which it is drawn.  As an academic, it is quite interesting to think about conditions under which “representativeness” is an appropriate goal (for example, we might not want a representative jury if the population of the local community has strong biases that would make it difficult to ensure a fair trial).  But, for purposes of this post, let us assume that representativeness is, indeed, a valuable characteristic of juries.

One way, in theory, to select representative juries is through random selection.  If every individual in a population has an equal chance of being selected as a juror, then the average jury will be representative of the population from which it is drawn.  This does not mean, of course, that each individual jury will be representative.  With only a small number of individuals (e.g., 12) serving on any given jury, most juries will not be representative.  But, importantly, because the average jury is representative, we would be able to say with a straight face that the jury-selection system is not biased against any defendant on the basis of being non-representative.

According to a recent news article, Champaign County, IL, picks jurors by randomly drawing names from various lists, including the list of registered voters, licensed drivers, those with state of Illinois ID cards, and those with certificates of disability.  Recently passed legislation, which will take effect on January 1, will add to the pool the names of individuals receiving unemployment benefits from the State.

For purposes of discussion, let’s assume that these various lists have done an effective job of providing a list of names that is roughly representative of the Champaign County population.  If so, then by randomly drawing names from these lists, the jury pool ought to be representative.  Right?

Wrong.   It is well-known that when a judge calls down to the “assembly room” for potential jurors, he or she is not selecting from a representative population.  Relative to the population, the people who appear for jury duty are disproportionately white, higher educated, and higher income than the population.  Today’s jury pool is an effective illustration.  I dutifully showed up at the Champaign County Courthouse at my designated 9:30 a.m. time for jury duty.  As I casually looked around the room, I could not help but notice how much the room was lacking in diversity.  While the lack of racial and ethnic diversity was immediately noticeable, I would also guess that, on average, this pool of jurors was drawn disproportionately from the upper three-quarters of the income and wealth distribution.

Why do we get non-representative jury pools even though we randomly choose names from the various lists?  It is because of what economists call “self-selection.”  Put in everyday language – not everyone shows up.  While I have been unable to confirm the statistic, I remember hearing on a local radio show the statistic that only about half of individuals who receive a summons for jury duty show up and/or have what we educators would call “an excused absence.”  To put it simply, “showing up” is not random.

There are many reasons that some groups are more likely to show up than others.  No doubt there are differing levels of trust and confidence in the judicial system.  Different levels of access to transportation.  Differing job and family situations that make it easier for some people to report for duty, and much harder for others.  Differing reactions to the bold print warning on the juror summons that warns recipients that a failure to report is a “criminal offense.”  And so on …

Aaron Aamons, chairman of the Citizens Advisory Committee on Jury Service, was quoted in the Champaign News-Gazette a few days ago stating that the solution is “education, education, education.”  As an economist, I have a different view.  I say it is all about “incentives, incentives, incentives.”  Or the lack thereof.

I suspect that nearly every individual who receives a jury summons goes through some sort of cost-benefit analysis.  They may not think of it in those terms, but they are almost surely asking themselves about factors that would go into such an analysis.  “Can I afford to miss that much work, or will I fall hopelessly behind?”  “Will my employer pay me, or do I have to take vacation time or uncompensated leave?”  “How much will it cost me to get the transportation I need to get to the Courthouse?”  “Who will watch my kids, and how much will I have to pay them?”  “What are the odds that the Sheriff will really show up and make me report for duty?”  (Answer: virtually zero).  “What are the reputational costs if I fail to appear?”

There are enormous social gains to having people willing to serve as jurors.  But the benefits to any individual juror are quite small.  Aside from a general sense of satisfaction associated with doing one’s civic duty, there are no direct benefits.  (Unless you count the pitiful $10 per day stipend, which is not enough to cover transportation and lunch for most people, let alone the cost of their time.)

So we have a classic “collective action problem” – it is in society’s best interest for people to serve, but it is in most individuals’ self-interest not to spend time on a jury.

One way to overcome such a problem is to mandate participation.  This is what we try to do, but the problem is that we do not enforce it.  As I once heard an official say, the County simply does not have the resources to enforce the mandate.

Another way to overcome the problem is to try to increase the non-financial benefits of service.  Labeling jury duty as a “civic duty,” appealing to patriotism, giving public kudos to those who serve, and other such appeals are meant to do this.  Undoubtedly, these are effective at getting some people to serve.  But it is obviously not sufficient to incentivize those who have not been showing up.

A related approach is to try to increase the costs of failing to show.  As an individual who tries to behave ethically and in accordance with the law, it is sufficient to deter me from skipping jury duty to simply tell me it would be a criminal offense.  But that is printed on every juror summons, and yet we still have highly unrepresentative jury pools.

One obvious answer is to pay jurors more that the pitiful $10 per day that they now pay.  If one is a single mother of three young kids who is holding down a minimum wage job at an employer that will not pay for time on jury duty, then the cost of appearing for jury service would be extraordinarily high – foregone wages, payment for childcare, and possibly a lost job.  As a matter of social policy, I find it hard to believe that we want to force the group least able to afford jury duty to show up for $10 per day.

But most state and local governments are severely strapped for funds.  Quite possibly the last thing they can afford is for each court to pay thousands of potential jurors hundreds of dollars per day for potential service.

Is there a solution?  Possibly – but one that only economists would love.  To implement it would be highly controversial, and not without reason.  But just for the sake of discussion, let me toss it out here.  The idea is simple – so simple, in fact, that the inspiration for it comes from a pre-school.

The pre-school to which I am referring offers parents a choice.  All parents are expected to volunteer 10 hours per academic year.  In lieu of volunteering, parents can donate a fixed amount (e.g., $75) per hour that they want “relieved” of duty.  So let’s say that a parent is expected to volunteer at a time in their life where the opportunity cost of their time is enormously high.  It may well be socially efficient for the parent to simply donate money to the school instead of volunteering.

You see where I am going with this … let’s suppose that the average jury pool has too many high income professionals (e.g., self-employed business people, highly paid executives, etc.) relative to the population, and not enough low-income individuals.  The jury pool would be more representative if we dropped a few of the high income folks and added some low income individuals.  So why not have the Court “auction off” a limited number of “excused absences,” and then use the money to provide a more meaningful level of juror pay to lower income jurors?

In a narrow, economic sense, it appears everyone wins.  Those released from duty are clearly better off, otherwise they would not have bid so high for release.  The low income individuals added to the pool are clearly better off if the additional compensation is sufficient to overcome the barriers to service.  And, if implemented effectively, the jury pool would be more representative than it is today.  Of course, one would need to avoid the temptation to auction off too many excused absences, or else one might end up with the opposite problem, i.e., too few high opportunity cost individuals on juries.

I am totally sympathetic to many alternative views – especially that this is not just an economic issue, but an issue that strikes at the core of how we define ourselves and our judicial system.  So I am not necessarily suggesting this as a serious policy proposal.  But it seems that the current system is so highly flawed that serious consideration ought to be given to alternatives, no matter how radical they might at first seem.