Warning: Declaration of action_plugin_safefnrecode::register(Doku_Event_Handler &$controller) should be compatible with DokuWiki_Action_Plugin::register($controller) in /home/public/dw/lib/plugins/safefnrecode/action.php on line 14

Warning: Declaration of action_plugin_blog::register(&$contr) should be compatible with DokuWiki_Action_Plugin::register($controller) in /home/public/dw/lib/plugins/blog/action.php on line 0

Warning: Declaration of action_plugin_blockquote::register(&$controller) should be compatible with DokuWiki_Action_Plugin::register($controller) in /home/public/dw/lib/plugins/blockquote/action.php on line 0

Warning: Declaration of action_plugin_include::register(&$controller) should be compatible with DokuWiki_Action_Plugin::register($controller) in /home/public/dw/lib/plugins/include/action.php on line 0

Warning: Declaration of action_plugin_popularity::register(&$controller) should be compatible with DokuWiki_Action_Plugin::register($controller) in /home/public/dw/lib/plugins/popularity/action.php on line 0

Warning: Declaration of action_plugin_redirect::register(&$controller) should be compatible with DokuWiki_Action_Plugin::register($controller) in /home/public/dw/lib/plugins/redirect/action.php on line 0

Warning: Declaration of action_plugin_captcha::register(&$controller) should be compatible with DokuWiki_Action_Plugin::register($controller) in /home/public/dw/lib/plugins/captcha/action.php on line 0

Warning: preg_replace(): The /e modifier is no longer supported, use preg_replace_callback instead in /home/public/dw/inc/auth.php on line 656

Warning: preg_replace(): The /e modifier is no longer supported, use preg_replace_callback instead in /home/public/dw/inc/auth.php on line 656

Warning: preg_replace(): The /e modifier is no longer supported, use preg_replace_callback instead in /home/public/dw/inc/auth.php on line 656

Warning: preg_replace(): The /e modifier is no longer supported, use preg_replace_callback instead in /home/public/dw/inc/auth.php on line 656

Warning: preg_replace(): The /e modifier is no longer supported, use preg_replace_callback instead in /home/public/dw/inc/auth.php on line 656

Warning: preg_replace(): The /e modifier is no longer supported, use preg_replace_callback instead in /home/public/dw/inc/auth.php on line 656

Warning: preg_replace(): The /e modifier is no longer supported, use preg_replace_callback instead in /home/public/dw/inc/auth.php on line 656

Warning: preg_replace(): The /e modifier is no longer supported, use preg_replace_callback instead in /home/public/dw/inc/auth.php on line 656

Warning: preg_replace(): The /e modifier is no longer supported, use preg_replace_callback instead in /home/public/dw/inc/auth.php on line 656

Warning: preg_replace(): The /e modifier is no longer supported, use preg_replace_callback instead in /home/public/dw/inc/auth.php on line 656

Warning: preg_replace(): The /e modifier is no longer supported, use preg_replace_callback instead in /home/public/dw/inc/auth.php on line 656

Warning: preg_replace(): The /e modifier is no longer supported, use preg_replace_callback instead in /home/public/dw/inc/auth.php on line 656

Warning: preg_replace(): The /e modifier is no longer supported, use preg_replace_callback instead in /home/public/dw/inc/auth.php on line 656

Warning: preg_replace(): The /e modifier is no longer supported, use preg_replace_callback instead in /home/public/dw/inc/auth.php on line 656

Warning: preg_replace(): The /e modifier is no longer supported, use preg_replace_callback instead in /home/public/dw/inc/auth.php on line 656

Warning: preg_replace(): The /e modifier is no longer supported, use preg_replace_callback instead in /home/public/dw/inc/auth.php on line 656

Warning: preg_replace(): The /e modifier is no longer supported, use preg_replace_callback instead in /home/public/dw/inc/auth.php on line 656

Warning: preg_replace(): The /e modifier is no longer supported, use preg_replace_callback instead in /home/public/dw/inc/auth.php on line 656

Warning: preg_replace(): The /e modifier is no longer supported, use preg_replace_callback instead in /home/public/dw/inc/auth.php on line 656

Warning: preg_replace(): The /e modifier is no longer supported, use preg_replace_callback instead in /home/public/dw/inc/auth.php on line 656

Warning: preg_replace(): The /e modifier is no longer supported, use preg_replace_callback instead in /home/public/dw/inc/auth.php on line 656
A pragmatic assessment of the state retirement fund crisis [ClearOnMoney]
Top

Site

About
Contact

Warning: preg_replace(): The /e modifier is no longer supported, use preg_replace_callback instead in /home/public/dw/inc/auth.php on line 656
Sitemap

Profile

Warning: preg_replace(): The /e modifier is no longer supported, use preg_replace_callback instead in /home/public/dw/inc/auth.php on line 656
Trace:
A pragmatic assessment of the state retirement fund crisis

Reference

Main
Updates
Schedule

Warning: preg_replace(): The /e modifier is no longer supported, use preg_replace_callback instead in /home/public/dw/inc/auth.php on line 656

Warning: preg_replace(): The /e modifier is no longer supported, use preg_replace_callback instead in /home/public/dw/inc/auth.php on line 656
Commentary

A pragmatic assessment of the state retirement fund crisis

23 Oct 2010 by Jim Fickett.

Unfunded obligations of state retirement funds, including both pensions and retiree health care, are in the neighborhood of $3 trillion. This is something like two years' worth of aggregate state income. On average, then, this is a heavy, but not crushing, debt load. Because politicians have delayed facing the issue, annual expenditures will rise. One reasonable projection puts pension plan contributions needing to increase by about 5% of aggregate budgets. Health plan expenditures are less well characterized, but might add another 2%, bringing the total adjustment to 7% of state budgets. The difficulties in both pension and retiree health plans are spread very unevenly. Some states face minor adjustment, and some will likely face adjustments of perhaps twice the average. For comparison, the drop in state tax revenues due to the recent crisis was about 10%.

Focus on what investors need to know

Much has been written about how much states have promised to retirees and how little they have saved to make good on those promises. In this post I would like to move beyond the scary numbers, to find parallels that provide some sense of scale, and then to look at some of the pragmatic issues investors must face. How likely is it that Illinois or New Jersey will require a federal bailout? Is the situation so bad that one should check into the condition of the state retirement funds before moving and becoming a taxpayer in a new location? Should one worry about municipal bonds in general, only in a few specific cases, or not at all?

In this post I will try to scale the problem from several perspectives, looking at both pension funds and retiree health care funds, and at both total obligations and cash flows.

There is much more data on state plans than on local government plans. The main reason is that there are thousands of local plans and, so, fewer researchers willing to aggregate data on them. But looking at state plans does give one a good start on understanding the situation since, according to Census, in 2008 state plans had 13.1 million active members, while local plans had only 1.6 million; presumably the states similarly dominate retiree health care plans. Local plans are not unimportant, but will be taken up another time.

There is a new first-draft reference page, State retirement funds, with the details behind all the facts in this article. A few of the most important sources are the Bureau of Labor Statistics, the Census Bureau, the Boston College Center for Retirement Research, the Pew Center for the States, and the papers of economist Joshua Rauh.

Unfunded pension obligations

State pension plan promises exceed funding by about $2 trillion.

To be more precise about that statement requires a more careful definition.

The Actuarial Accrued Liability (AAL) is the sum of all retirement benefits that have already been earned by current employees and retirees. In the general spirit of accrual accounting, this ignores all the benefits that will be earned in the future as current employees continue to work and new employees are hired. The sum of all accrued benefits is reduced to a present value by a “discount rate” representative of investment returns. That is, the AAL recognizes that what the state needs to set aside today, in order to pay benefits in the future, is less than the nominal value of those benefits, because the money can be invested and will (hopefully) grow. The discount rate is the subject of much controversy, which I have written about previously. Most plans assume a discount rate of about 8%; for this article I will take it as given that (1) 8% is definitely too high, and (2) a reasonable rate is 4-6%.

The Unfunded AAL (UAAL) is the AAL less any money already set aside to pay the future benefits. A more informal term, the “unfunded obligation” usually means either the UAAL, calculated by the state government, or some variant on it, calculated by a researcher.

There is some uncertainty in any actuarial computation – you do not know exactly how long people will live, even on average, for example. But the main sources of variation in the unfunded obligation are (1) the discount rate, (2) the set of plans under discussion, and (3) the recent history of the stock market when the study is carried out. The earlier post pointed out how different savings today can be under different discount rate assumptions. As for the second point, it is less clear than you might think what is a state plan and what is a local plan. The well known Pew study defined its scope more carefully than most, including “plans for municipal workers or teachers when those plans are run by the state and the state maintains a financial interest”.

Estimates of the unfunded obligation range from around half a trillion dollars (from Pew, at the peak of the stock market and using the most optimistic discount rates) to over three trillion (from Rauh, using the most conservative possible discount rate and including some of the largest local government plans). As an investor, it is best to accept that this number is inherently imprecise, and not to try to pin it down much more than to say, “about $2 trillion”.

There is a great deal of variation between the states. For example, as of 2008, 11 states had funded more than 95% of the AAL, while 5 states had funded less than 65%.

Unfunded health plan obligations

The unfunded obligation for retiree health care is about $1 trillion.

There is more uncertainty, and much less information, about retirement health care plans than about pensions. (As an aside, the accounting category is “Other post-employment benefits”, but the expenses are primarily health care, and OPEB is taken in most studies as synonymous with retiree health care.) There is more uncertainty because a very important assumption built into the AAL calculation in this case is future health care cost inflation, and this is difficult to estimate. Most state calculations assume that health care inflation will soon decrease greatly; some independent estimates revise this assumption to continue current inflation rates. For health plans, as for pension plans, states often use too high a discount rate, while other estimates use a lower rate. Finally, reporting requirements were only recently implemented, and reporting is less uniform than for pension funds, so some studies admit that their methodology has missed some plans altogether.

While in the case of pensions at least some large fraction of the future obligation is funded, this is not the case for retirement health plan benefits. Pew reports that, in aggregate, only about 7% of future obligations had been funded as of 2008.

Official state estimates for the retirement health plan unfunded obligation, as of 2008, summed to $555 billion. A variety of other estimates, using more realistic assumptions, range between $1 trillion and $1.5 trillion. Again, one has to accept that there is no one right answer, and a very soft “about $1 trillion” is as precise as we can get.

Scaling the unfunded obligations

These unfunded obligations are essentially debt. Against what can we scale them? A recent study by Morgan Stanley (account required) played down state debt and unfunded liabilities by comparing them to GDP. But this is the wrong comparison; most of the tax revenue from the nation's production goes to the federal, not the state governments. It is better to compare to state revenues which were, in 2008, $1.3 trillion. So unfunded retirement obligations, including both pension and health care, are, very roughly, about two years' worth of state income.

There is no direct comparison to debts or obligations in other sectors. But just to give some idea of scale:

All this suggests that the unfunded retirement obligation, in and of itself, is a heavy burden that deserves the widespread attention it has been getting but does not, by itself, indicate state insolvency.

Pension plan cash flows

In an earlier post I showed that over the last 15 years or so cash flows on state pension funds were reasonably healthy. However expenditures have been growing at a worrying 9% compound annual rate.

A generally agreed principle is that each generation should pay for the services provided to it, so that retirement benefits earned today should be paid for today. The unfunded retirement obligations violate this principle, of course. Probably the generosity of benefits is peaking right about now, and this generosity will be paid for in the future, so both pension and retirement health care plan expenditures are likely to rise.

To discuss cash flows quantitatively, we need one additional definition. The Annual Required Contribution (ARC) is the amount that would cover new accruals in the given year, together with an amount that would pay down the UAAL over a thirty year period. In other words, if a state paid the ARC each year, the plan would be fully funded after thirty years. The ARC is only “required” in a mathematical sense; there is no legal obligation.

According to a projection by the Boston College Center for Retirement Research, if state and local governments were to calculate their unfunded pension liabilities using a realistic 5% discount rate, and then make the ARC, pension expenditures would rise from 3.8% of budgets currently to 9.1%, an increase of 5.3% of budget.

Many states will not, in fact, be so sensible, and many are in a much worse than average starting position. For a worst case scenario, if Illinois, which has the worst current funding ratio, were to spend down current plan savings, and then move to a pay-as-you-go model, expenditures would rise to 16% of the state budget.

Health plan cash flows

The health plan problem is somewhat smaller than the pension problem. We have already seen that the unfunded obligation is probably about half as large. The same goes for current cash flows: states paid $15 billion on retirement health plans in 2008 (according to Pew), compared to $36 billion on pension plans (according to Census). And, since most retirement health plans are pay-as-you-go, that $15 billion is approximately the benefits paid, while benefits paid in pension plans were $175 billion (Census).

However health plan costs are lower partly because there has been almost no effort so far to fund the long-term costs. If states had made the health plan ARC, contributions would have been $43 billion, according to Pew. Other data are anecdotal, but worrying. For example, according to a study from the Empire Center, New York's health plan costs are projected to triple by 2026.

States vary greatly both in the generosity of their promises and in the state of their funding. The end result may be seen in the variation of the ARC. According to the Center for State and Local Government Excellence, the highest ARCs per capita, as of 2009, were Alaska at $553, Hawaii $556, and New Jersey $675. The lowest were Iowa at $8, Indiana $7, and North Dakota $6.

Raising the health care plan contributions from $15 billion to $43 would require approximately 2.1% of state budgets. We really need a careful projection like the Boston College study on pensions, but this 2.1% at least gives some idea of the adjustment that will eventually be needed for health plans. It may be too low given that health care inflation is understated and discount rates are too high. But it may be too high in that it is much easier, legally speaking, to reduce health plan benefits than pension plan benefits.

Scaling the cash flow problems

Very roughly, then, the adjustment facing states might be about 7% of total budget on average, and perhaps double that for the states that have managed the issue less well.

One comparison that can give some idea of scale is to look at other categories of budget expenditure. In California, the entire judicial system receives 2.9% of the state budget, state support for the ten campuses of the University of California is 2.4%, and environmental protection receives 1.2%. These examples show that major areas of critical function, sometimes created via long and bitter battles, can be in the range of 1-3% of the budget.

For another comparison, according to the latest report from the Rockefeller Institute of Government,

For the year ending in June 2010, the period corresponding to most states' fiscal years, total state tax collections declined by $19 billion or 2.7 percent from the previous year, and were down $84 billion or 10.8 percent compared to fiscal 2008.

A 10% decrease in tax revenues, even with the offset of special one-time help from the federal government, has resulted in dislocations that have made headlines for the last two years.

Overall, it is pretty clear that an additional 7% of the state budgets would cause considerable pain and not a little conflict. However most states would, as they have done this time, adjust. Some states will have a much larger adjustment and some states, perhaps many, will face difficulties at least as great as that in the recent crisis (though less sudden). Given the recent federal intervention, it would not be surprising to see the same again.

Putting the financial crisis and the retirement fund crisis together, it is a bit like a person who has had a negative amortization home loan for several years and, as the neg-am period ends, suddenly has to find room for mortgage payments in the budget. Not a small adjustment.

Takeaway lessons

A short summary of my expectations:

  • So far pensions have gotten most of the attention. The rapid growth in retirement health plan costs will probably get more attention before too long.
  • States will have to make large and painful budget adjustments. This will happen over a period of many years, and will be a major feature of state politics and budgeting for a long time.
  • At least some, perhaps many, states will have to face difficulties on the same scale as the recent crisis; federal intervention is not unlikely.
  • These issues will have a major influence on state tax policy and quality of life; I would, indeed, look into the condition of state and local retirement plans before deciding to move to a new location, and before deciding to buy property.
  • All this is not a reason to avoid the municipal bond market in toto. But it is no longer the case that most munis, or even most general obligation bonds, can be assumed to be safe. The difficulties are spread very unevenly; due diligence and careful selection will be important.

If you'd like to see how a particular state is doing

  • Pension plans: On page 43 of the Pew report there is a US map with the states colored to indicate the general state of their pension plans, and on page 56, in Appendix B, a table is given with the numbers underlying that map.
  • Health plans: In a report from the Center for State and Local Government Excellence, starting at the bottom of page 6, there is a table with, for each state, the unfunded obligation, the unfunded obligation per capita, the ARC, and the ARC per capita.