The GAAP In The Debt Debate
There's a fascinating (and enraging to both sides) argument going on over whether debt really matters. On one side are the Austrian-inspired hard money advocates who see excessive borrowing as even worse for governments than it is for individuals and families. On the other side are the Krugman/Bernanke Keynesians who think debt is no big deal if that's what it takes to sustain "aggregate demand". But of course you know all this.
An interesting and less well-understood offshoot of this argument involves the nature of debt. Is it strictly limited to contracts with specific repayment terms? Or should unfunded liabilities be seen as money owed? If the latter, then we owe a hell of a lot more than we think we do. Not surprisingly, the battle lines are identical to those of the greater debt debate, with followers of Hayek and von Mises (like Ron Paul and Peter Schiff) on one side and Keynes, Krugman and president Obama on the other.
Here's an argument in favor of viewing unfunded liabilities as debt:
Hiding the government's liabilities from the public makes it seem that we can tax our way out of mounting deficits. We can't.
A decade and a half ago, both of us served on President Clinton's Bipartisan Commission on Entitlement and Tax Reform, the forerunner to President Obama's recent National Commission on Fiscal Responsibility and Reform. In 1994 we predicted that, unless something was done to control runaway entitlement spending, Medicare and Social Security would eventually go bankrupt or confront severe benefit cuts.
Eighteen years later, nothing has been done. Why? The usual reason is that entitlement reform is the third rail of American politics. That explanation presupposes voter demand for entitlements at any cost, even if it means bankrupting the nation.
A better explanation is that the full extent of the problem has remained hidden from policy makers and the public because of less than transparent government financial statements. How else could responsible officials claim that Medicare and Social Security have the resources they need to fulfill their commitments for years to come?
As Washington wrestles with the roughly $600 billion "fiscal cliff" and the 2013 budget, the far greater fiscal challenge of the U.S. government's unfunded pension and health-care liabilities remains offstage. The truly important figures would appear on the federal balance sheet--if the government prepared an accurate one.
But it hasn't. For years, the government has gotten by without having to produce the kind of financial statements that are required of most significant for-profit and nonprofit enterprises. The U.S. Treasury "balance sheet" does list liabilities such as Treasury debt issued to the public, federal employee pensions, and post-retirement health benefits. But it does not include the unfunded liabilities of Medicare, Social Security and other outsized and very real obligations.
As a result, fiscal policy discussions generally focus on current-year budget deficits, the accumulated national debt, and the relationships between these two items and gross domestic product. We most often hear about the alarming $15.96 trillion national debt (more than 100% of GDP), and the 2012 budget deficit of $1.1 trillion (6.97% of GDP). As dangerous as those numbers are, they do not begin to tell the story of the federal government's true liabilities.
The actual liabilities of the federal government--including Social Security, Medicare, and federal employees' future retirement benefits--already exceed $86.8 trillion, or 550% of GDP. For the year ending Dec. 31, 2011, the annual accrued expense of Medicare and Social Security was $7 trillion. Nothing like that figure is used in calculating the deficit. In reality, the reported budget deficit is less than one-fifth of the more accurate figure.
These real-world impacts will be felt when currently unfunded liabilities need to be paid. In theory, the Medicare and Social Security trust funds have at least some money to pay a portion of the bills that are coming due. In actuality, the cupboard is bare: 100% of the payroll taxes for these programs were spent in the same year they were collected.
When combined with funding the general cash deficits, these multitrillion-dollar Treasury operations will dominate the capital markets in the years ahead, particularly given China's de-emphasis of new investment in U.S. Treasurys in favor of increasing foreign direct investment, and Japan's and Europe's own sovereign-debt challenges.
When the accrued expenses of the government's entitlement programs are counted, it becomes clear that to collect enough tax revenue just to avoid going deeper into debt would require over $8 trillion in tax collections annually. That is the total of the average annual accrued liabilities of just the two largest entitlement programs, plus the annual cash deficit.
Nothing like that $8 trillion amount is available for the IRS to target. According to the most recent tax data, all individuals filing tax returns in America and earning more than $66,193 per year have a total adjusted gross income of $5.1 trillion. In 2006, when corporate taxable income peaked before the recession, all corporations in the U.S. had total income for tax purposes of $1.6 trillion. That comes to $6.7 trillion available to tax from these individuals and corporations under existing tax laws.
In short, if the government confiscated the entire adjusted gross income of these American taxpayers, plus all of the corporate taxable income in the year before the recession, it wouldn't be nearly enough to fund the over $8 trillion per year in the growth of U.S. liabilities. Some public officials and pundits claim we can dig our way out through tax increases on upper-income earners, or even all taxpayers. In reality, that would amount to bailing out the Pacific Ocean with a teaspoon. Only by addressing these unsustainable spending commitments can the nation's debt and deficit problems be solved.
So we're irredeemably bankrupt...or not. Here's a direct rebuttal from The Atlantic that dismisses the "unfunded liabilities are debt" idea as irrelevant:
The problem with budgeting 75 years into the future is that you end up with a lot of numbers that are much more meaningful to actuaries than to other living people
Wanna scare somebody about America's debt on the eve of the Fiscal Cliff? I mean, really scare somebody? Here's a trick. Don't talk about the debt. Talk about "unfunded liabilities."
The U.S. national debt comes out to about $16 trillion today. That's something. But it's nothing compared to the extra $87 trillion in unfunded liabilities to Social Security, Medicare, and federal pensions. Here's how that works. If you add up all of the U.S. government's promises to pay retirement and health care benefits for the next 75 years and subtract the projected tax revenue dedicated to those programs over the next 75 years, there is a gap. A $87 trillion gap -- in addition to a $16 billion hole.
"Why haven't Americans heard about the titanic $86.8 trillion liability from these programs?" Chris Box and Bill Archer ask in the Wall Street Journal. The authors blame the U.S. government for using shoddy accounting and for misleading the American public on their finances. In fact, the most misleading thing about that $87 trillion is the way the figure is often used in the media.
(1) That's not our debt. Our $16 trillion in debt and our $87 trillion in "unfunded liabilities" represent two very different ideas: real past promises and projected future promises. Real past promises are, well, very real. We have to pay back our debt. Failing to do it would be an illegal and disastrous default. Unfunded liabilities are future promises, and, since they're not as real, we can change them whenever we want without destroying ourselves. For example, raising the taxable income ceiling and slowing the growth of benefits could reduce the Social Security gap to zero tomorrow.
(2) 75-year projections are scarier than they are informative. Seventy-five-year projections always sound gargantuan because, well, they're calculated over three-quarters of a century, which is an awfully long time to count anything. But here's the flip side: In 75 years, our economy will be massive. Growing slowly at a 2% annual average, our GDP would be $66 trillion in today's dollars in 2087. That's an incomprehensibly big number, too. Once you run out any number over 75 years, the mind starts to boggle. That's good for scaring people with mind-boggling numbers, but it's not so good for informing. When Republicans say unfunded liabilities come out to $520,000 per U.S. household, they're taking a figure from 2087 and dividing it over a 2012 population to exaggerate. Scary, to be sure, but not very informative.
(3) Projections can change fast. An unfunded liability is a projection, and projections shift all the time for two big reasons: (1) Circumstances change and (2) laws change. Let's take circumstances first: The shortfall in Medicare and Social Security is exquisitely sensitive to just about every demographic trend you can imagine, including longevity, immigration, income growth, and birth rates. Furthermore, it assumes that seven decades of innovation will do nothing to change the rate of health care inflation, which is a brave assumption. We know next to nothing about how medical inflation will change after this decade. The fact that actuaries pretend to know the future doesn't make them oracles. It just makes them dutiful actuaries.
Now, about our laws. Strictly speaking, the U.S. doesn't have an entitlement problem, or even a Medicare problem. Rather, we have a health care cost problem -- medical insurance and hospital costs and so on are getting expensive faster than our ability to pay for them. Medicare and Medicaid are part of this big expensive system. If we cut these programs without changing the system, we won't be "saving" money, so much as shifting costs to old folks, who will be forced to pay much more for their health care, or else see much worse coverage.
Closing this liability gap -- whether it's $87 trillion or $8 trillion -- will require patience. The bad news is that none of these measures to fix our real problem (health care costs) will be easy, few of them will be initially popular, and most of them might not work. The good news is we have 75 years.
It's interesting that people who don't like debt see everything as debt (the above articles don't mention derivatives, but the case can be made that they're also a dangerous form of debt) while people who like debt or consider it useful or irrelevant or whatever find reasons not to worry about any kind of obligation, whatever it's called. There's a lesson about human nature here.
But there's also a reason that generally accepted accounting principles (GAAP) require corporate and state/local pension plans to calculate their liabilities to the best of their ability -- and then fund the resulting guestimate: We know these plans will cost something, and it's crucial to fund the obligation that seems most likely, based on the work of actuaries trained to get as close as possible. Over time with any plan you make adjustments, but it's very rare for a pension or entitlement to announce that its costs are dramatically lower than expected. Almost without exception the opposite is true.
So sure, health care could cost a lot less when stem cells and gene therapy go mainstream. But in the meantime we've got this wave of boomers retiring, many of whom are going to cost a fortune over the next decade no matter what stem cells do. AND they're going to end the career of any politician who tries to scale back their coverage. So defaulting on retiree health benefits would be more politically dangerous than stiffing China on its bond interest. These liabilities won't, in short, be eliminated with any stroke of the pen.
Actually it's kind of disorienting trying to understand and give respectful hearing to the "debt doesn't matter and there's not that much of it anyhow" argument. This is one of those things (like school choice) where one side seems to get it so completely that it's hard to credit the other side with honest intentions. Whatever we call these things, they're obligations that we're dumping onto either our kids and grandkids or our own investment portfolios (which will be vaporized if we really do owe $2 million per family of four). So either our kids are impoverished in 2025 or boomers are bankrupted in 2015.
The author of the Atlantic piece doesn't completely get the concept of discounting. But that's a subject for another time. More interesting is this:
"The good news is we have 75 years....In 75 years, our economy will be massive." Well, an Austrian economist would say that once your national balance sheet becomes sufficiently fragile, your economy has to shrink, dramatically and for a long time, until the excess debt is wiped out. So this long-term future growth will begin from a base that's a lot smaller than today, and therefore won't reach the size that a simple linear projection would imply. But it is true that projecting anything out more than a few years is kind of silly, based both on history (compare the expectations of people in 1912 with the way the century actually turned out) and on technology. The singularity will shred our long-term plans.
But the coming decade, when current law is etched in stone and technology and demographics are virtually certain to raise rather than lower costs, looks mathematically pretty grim. We boomers are retiring now, and our benefits have to be paid now. So the big cost spike is coming fairly soon - when we're already accumulating a trillion dollars a year of real, official debt.