aftermath

we dream, we create, we change, we love

Archive for the ‘άctuary’ Category

8-Fold Pension Cost Hike . . . Not!

leave a comment »

[T]he loss in [pension] funded status in 2008 is projected to produce an increase in pension expense for 2009 (and a charge to corporate earnings) in excess of $70 billion [i.e., about 8 times the 2008 pension cost of $10.4 billion].

Milliman 2009 Pension Funding Study

As early as last November, even before we knew where the dust of 2008’s maelstrom would settle, I was anticipating aggregate 2009 costs for global pension plans of the S&P 500 to be in the neighborhood of triple 2008 pension costs, at most. That estimate came from: an expectation that even continued wild volatility in equity markets and interest rates would be within certain ranges; an understanding of how current accounting under SFAS 87 works; and an awareness – so far, as best as I can tell, a unique awareness not possessed by any other observer – of how high pension cost would be if the deferral-and-spreading methodologies of SFAS 87 were not in place (e.g., as discussed in one of several instances in Not Your Daddy’s Pension Cost). Leading me to find guesses such as Milliman’s expectation of an 8-fold cost (actually, one of the more conservative of speculations, most other published projections being even wilder) to be somewhat absurd, to characterize it euphemistically.

As material as elimination of SFAS 87’s methodologies would be, the effect on 2008’s pension cost of such a drastic change in GAAP would have barely been the same quantum level as those wild guesses were bestowing on costs under existing GAAP. So with many companies using smoothed asset values, with even the investment losses from those smoothed values only facing recognition outside a corridor, with even those amounts spread over expected service lives of employees, and with everything else SFAS 87 offers for minimizing cost volatility (e.g., few if any companies reducing expected rates of return on assets, contrary to another speculation given in the Milliman report, unsubstantiated by the 2009 rates already given in many companies’ 2008 annual reports), it’s difficult to see how 2009 pension cost would be within the same galaxy as was being speculated.

And with today’s publication of IBM’s interim fiscal report for the first quarter of 2009 (along with similar interim reports previously published by other companies), we now have enough hard data to act like an election-night TV analyst who can declare the winner even before the final votes have hit the bottom of the ballot box. In the aggregate, for any broad group of companies, pension cost for 2009 will be about 2-3 times pension cost of 2008. A significant jump, to be sure, but nowhere near the 8-fold level. Indeed, about the only individual companies that might see an 8-fold cost level or higher are those that had a 2008 pension cost near zero, from which point just about any change at all – whether increase or decrease – appears artificially high.

And here, the difference between a 3-fold cost level versus an 8-fold cost level is not a matter of actuarial assumptions or expected future economic conditions or any such opinion-based distinction. It’s merely a matter of correct application of SFAS 87 technique versus inappropriate approximations. Or perhaps even worse: delegation of the task for developing the number to lower-level staff who do not understand how SFAS 87 works, without an expert’s follow-up to check the results.

Whatever. This is where Milliman gets to be like S&P and so many others, who have published bad pension reports with faulty results based on weak methodology and incorrect data, but then have never gone back to correct their mistakes, indeed in some instances have perpetuated the blunders year after year after year. Thing is, although the $70 billion increase speculation may have turned a head or two when the Milliman report was published last month, by now nobody remembers or cares. And since to the best of my knowledge, I remain the only one who bothers to look at first-quarter financial statements, Milliman’s mistake will be completely overlooked by next spring, when most of 2009’s annual financial reports confirm what I’m stating here, that Milliman’s 8-fold level is significantly overstated.

And still, it wasn’t just a bad estimate. It was simply wrong.

(As I’ve previously disclaimed, posts such as this represent efforts of my favorite pastime. My formal work does not involve any of this, and none of it represents any position or comment that should in any way be attributed to my employer. Likewise, as always, it represents general personal impressions and should not be treated or used as formal professional advice.)

actuary cat feed subscribe to aftermath’s άctuary category

bumper sticker [www.internetbumperstickers.com] - actuary

Written by macheide

28 April 2009 at 1:31 pm

Posted in άctuary

Pension’s Target Equity Allocations Shifted Down

leave a comment »

While stated equity targets have not declined significantly, actual equity allocations at year end were much lower. For the 83 companies that provided data, the average target equity allocation is 55 percent for 2009, compared with 58 percent for 2008. However, actual equity allocations fell over the year, to 48 percent at the end of 2008 from 59 percent the year before.

Watson Wyatt Worldwide

So although the 2008 market meltdown brought actual equity allocation in pension funds down below 50%, employers have not materially altered the target equity allocations under their pension plans’ investment policies?

If that were true, it would not be a trivial observation. Not very trivial at all. For decades, the typical pension fund held its equity allocation target steady near something like 65%, through bear and bull. Which then automatically drove an ongoing buy-low-sell-high strategy: if a bull market drove the equity allocation toward 70%, then the fund would sell stocks (at relatively high prices) to bring the equity allocation back down toward the target; conversely, if a bear market drove the equity allocation down toward 60%, then the fund would buy stocks (at relatively low prices) to bring the equity allocation back up toward the target. So if pension funds were truly holding their targets in place in the face of the meltdown, then we could expect to see some rather heavy duty cash moving into the market. Which was largely behind my bloody spring followthrough hopes, thinking that maybe just maybe I might consider heading back into the market myself as early as Memorial Day, to join the parade of all that new money coming in, instead of waiting until late 2009 or early 2010.

Except, Wyatt is wrong. And not just on the numbers, but on exactly what those numbers mean regarding the commitment of pension funds to their stocks.

Read the rest of this entry »

Written by macheide

26 March 2009 at 4:55 pm

Posted in άctuary

Coincidence? Doubtful

with one comment

Watson Wyatt Worldwide, a consulting firm, estimated pension assets declined 26 percent in 2008. The firm also reported the 100 largest US pensions were just 79 percent funded in 2008, compared with 109 percent funded at the end of 2007.

Boston Globe

For the group [of 100 companies with largest pension plans], funded status decreased to an average of 79% from the prior year’s 105.2% . . . .

Milliman

Just a coincidence, that a newspaper carries a story citing Wyatt for a figure based on the “100 largest US pensions” the day after Milliman produces a full report on what seems to be the same set?

I don’t believe in that kind of coincidence.

For the past 9 years, Milliman has produced the best annual report of pension funded status – by far the earliest (usually at least two months before the next competitor in line, and typically about six months before most others), by far the most complete (covering every main element including cost and contributions, measurement assumptions, asset allocation, and even OPEB funding, whereas many of the competitors venture little beyond reporting on pension funded status), the most accurate (definitely more credible than what sometimes seems the most cited competitor’s report, one from a credit rating agency, a report typically containing crucial data and analysis errors in as many as a third of the study’s companies’ numbers), and by far the one with the best insights (several of the competitors’ studies suitably illustrating the value of actuarial advice via the weakness of their analysis). At best, competitors have only found room to chase Milliman in the selection of the study’s members – Milliman increased its study size, from 25 the first year and 50 the second year, to settle on 100 companies for all remaining studies; competitors pretend to have more complete data by reaching for the S&P 500 or the Fortune 1000, or even sometimes pretending to show results for the S&P Composite 1500. But just as the Dow Jones Industrial 30 can serve as proxy for the S&P 500, or as one need not know where the Russell 5000 might be valued if one has the S&P 500 in hand, so too the Milliman 100 has slowly but surely become an acceptable standard that adequately reflects broad trends that are only echoed when results for the “broader” sets are eventually published.

So then, has Wyatt finally realized that stretching to the Fortune 1000 or faking the S&P Composite 1500 wastes precious resources without adding sufficient insight to their investigations? Even if that is so, why not then come out with their 100 last week or several weeks ago, when I myself had already started poking similar numbers out into this blog? Or why not next week, when the Enrolled Actuaries Meeting could have given them a springboard for further discussion? Why the very morning after the Milliman report? Again, I really don’t believe in such coincidence.

But I must say, that “109 percent” cited in the Globe article is either a misprint (did the fax blur Milliman’s 105 into a 109?) or an outright blooper. I hereby give myself this exercise, but from extensive experience in the numbers I suspect it to be a failed exercise: find any 100 companies of any size from any set that would average a 2007 funded ratio as high as 109. For smaller sets of a few dozen companies at most, sure, throw in the very small handful that are over 109, and you can pull that average; but not for 100 companies. Heck, even for 2007, when funded ratios reached the highest peak for the millennium, only 123 companies among the S&P 500 had assets in excess of liabilities; and by far, most of those companies were funded closer to 100 than as high as 109. Not that anyone much cares: bad pension numbers like the Globe’s 109 get passed along all the time, are left uncorrected, and pretty much pass into practice as if they are truth, without so much as a shrug, just because it has been supposedly credited to an “expert.”

Whatever. OK, so the Boston Globe ain’t the Financial Times or the Wall Street Journal or Pensions & Investment Age or Bloomberg or any source that many business people might go for their pension information, so does it much matter that this blurb credited to Wyatt should have given the nod to Milliman? In the greater scheme of things, probably not. It’s still irksome.

(As I’ve previously disclaimed, posts such as this represent efforts of my favorite pastime. My formal work does not involve any of this, and none of it represents any position or comment that should in any way be attributed to my employer. Likewise, as always, it represents general personal impressions and should not be treated or used as formal professional advice.)

actuary cat feed subscribe to aftermath’s άctuary category

bumper sticker [www.internetbumperstickers.com] - actuary

Written by macheide

25 March 2009 at 8:43 am

Posted in άctuary

Pension Funding Domestic v Foreign

leave a comment »

domfor2008As we always point out, and as is at least footnoted in the better among the published reports, pension numbers reported in corporate financial statements of publicly traded U.S. companies include both domestic and foreign pension plans. And as we’ve always observed, but is inadequately examined by any of the published reports, foreign pension plans are typically significantly less well funded than are domestic pension plans (even when unfunded supplemental executive retirement programs are included with the domestic numbers).

One important trend which so far has been overlooked by published reports and the business press: although foreign pension plans shared the hangover from the 2008 market meltdown, relatively they lost less in funded status than did domestic pension plans. Credit four primary reasons:

  • Underfunding Shifted Emphasis to Liability Changes – Although not universally constant across the board, companies generally increased discount rates used to measure pension obligations, reflecting higher interest rates in corporate bond markets. Higher discount rates tempered pension liability growth or even led to declines in pension liability. For all pension plans, the effect of those pension liability changes tended to move counter to losses suffered by pension assets. But since foreign pension plans tend to hold less assets, the pension liability changes were more pronounced for 2008. Particularly representative of this effect were pension plans for subsidiaries in Germany, where pension plans are completely unfunded in the traditional U.S. sense – there, only the effect of liability changes prevail.
  • Lower Equity Investment for Funded Plans – Particularly for pension plans in subsidiaries in England, but elsewhere as well, foreign pension plans that are backed by pension assets overall have lower equity investment allocations than remains present for U.S. pension funds. So although foreign equity markets overall dropped more during 2008 than did U.S. equity markets, foreign pension funds generally saw lower investment losses than did their U.S. counterparts.
  • Dollar Muscle – Thank the strength of the U.S. dollar against other currencies for a healthy share of the relatively lower loss of funded status experienced by foreign pension plans during 2008. All things considered, a strong U.S. dollar helps the funded status of underfunded foreign pension plans that are then converted back into dollars for U.S. reporting. (I’ll come back to this with more detail in a future post.)
  • Persistent Employer Contributions – This isn’t a distinction for each and every company, but it remains so for enough of the elephants in the bath to be worth note. For some companies, whereas employer contributions for qualified pension plans for U.S. employees remain low or zero due to credit balances accumulated several years ago (then largely in response to an accounting rule that has since been eliminated), employer contributions for foreign pension plans generally increased during 2008. So relatively speaking, asset losses were offset by employer contributions more so for foreign pension plans than for domestic pension plans.

The result: As seen on this post’s chart of the past 10 years for S&P 500 companies that separately disclosed their domestic versus foreign pension plans, here showing pension funded ratios for domestic pension plans (higher blue circles) versus pension funded ratios for foreign pension plans (lower purple squares), the funded status for domestic pension plans remained above that of the foreign pension plans. However, as also is readily apparent from this chart, foreign pension plans did not suffer as badly during 2008 as did domestic pension plans, with the spread between the domestic pension funded ratio versus the foreign pension funded ratio squeezing down to a margin smaller than any experienced through the past decade.

But don’t expect the two lines on this chart to cross. Even if all four trends were to persist through 2009 – an unlikely scenario – at most we would see the spread narrow further, without merging or flipping.

(Remember, as I’ve previously disclaimed, posts such as this represent efforts of my favorite pastime. My formal work does not involve any of this, and none of it represents any position or comment that should in any way be attributed to my employer. Likewise, as always, it represents general personal impressions and should not be treated or used as formal professional advice.)

actuary cat feed subscribe to aftermath’s άctuary category

bumper sticker [www.internetbumperstickers.com] - actuary

Written by macheide

24 March 2009 at 4:16 pm

Posted in άctuary

Elephants’ Pensions Fatter II

leave a comment »

frranked2008As I observed in yesterday’s aftermath post, the pension plans of the largest corporations with the largest pension funds are generally better funded than pension plans of smaller corporations with smaller pension plans. Whereas yesterday’s post illustrated that trend by grouping large companies versus smaller companies and charting them against aggregate assets, today’s chart looks at the funded ratio of the global pension plans of each and every S&P 500 pension sponsor, then simply plots those against the rank of the company in terms of pension assets as of the close of the 2008 fiscal year, from the largest pension sponsors at the far left of this chart to the smallest at the far right.

Were we expecting to see a more obvious decline, left to right? That’s not there, per se. But closer inspection of this chart does show the same trend as yesterday’s chart.

Let’s start by observing one somewhat remarkable thing that shows through on this chart – how very many of the companies have pension plans with a funded ratio between 70% and 75%, through the middle of the yellow zone on this chart. From nearly the largest through the smallest, the large number of companies at that level almost draw their own line across the chart, without necessitating having me do so. See that line, and the trend discussed in yesterday’s post emerges via the divergences from that line.

First, with the exception of Exxon and Caterpillar – the two “elephants” actually below the yellow zone – that yellow zone is empty at the far left edge of the chart, within the region enclosed by a red rectangle. And the thing is, since the companies within that red rectangle in the aggregate hold well over half of the pension assets for the entire S&P 500, those elephants drive the calculation of the aggregate funded ratio for the entire universe, since the aggregate funded ratio is essentially a weighted average, as if the entire universe of companies pooled everything into one gigantic single pension fund. That’s where yesterday’s graph – charted on the basis of aggregate assets – versus today’s – charted on simple ranking – better illustrates how the size of the pension fund, together with the higher funded ratios typical among the elephants, drives most of the results we were seeing yesterday.

But it’s not just the weight the elephants are throwing around. Look at the green band of the chart, left to right – the number of companies in that band tends to decrease as we reach the right side, the companies with smaller pension plans. In other words, even when we’ve left the elephants behind, we’re still seeing results that exhibit the trend we presented yesterday: lower funded ratios for the smaller pension plans.

And I’m still aiming toward getting around to an examination of why we’re seeing these results. Later.

(Remember, as I’ve previously disclaimed, posts such as this represent efforts of my favorite pastime. My formal work does not involve any of this, and none of it represents any position or comment that should in any way be attributed to my employer. Likewise, as always, it represents general personal impressions and should not be treated or used as formal professional advice.)

actuary cat feed subscribe to aftermath’s άctuary category

bumper sticker [www.internetbumperstickers.com] - actuary

Written by macheide

20 March 2009 at 9:54 am

Posted in άctuary

Elephants’ Pensions Fatter

leave a comment »

frsized2008The pension plans of the largest corporations with the largest pension funds are generally better funded than pension plans of smaller corporations with smaller pension plans. As illustrated by this chart of funded ratios for the S&P 500 as of the close of fiscal years ending in 2008.

The 22 elephants with pension assets in excess of $10 billion per corporation stake out the leftmost green square, comprising more than half of the S&P 500’s pension assets, measured along the chart’s x axis. In the aggregate, the funded ratio of global pension plans for those elephants was 81.2% at the end of 2008, above the overall S&P 500’s aggregate pension ratio of 79.1%, indicated on this chart by the higher heavy dotted line.

Almost three times as many beasts staked out the blue region, indicating corporations with pension assets between $3 billion and $10 billion per corporation. For those companies, the aggregate funded ratio of global pension plans was 79.2%, almost exactly equal to the S&P 500’s overall pension funded ratio.

The purple region then indicates the 87 corporations with pension funds of $1-3 billion each, with an aggregate funded ratio of 76.0%. The brown region indicates 53 companies with pension funds of $0.5-1 billion each, with an aggregate funded ratio of 68.9%. And the red region indicates 111 companies with pension funds less than $0.5 billion, with an aggregate funded ratio of 67.2%.

Several off-the-cuff observations –

  • Research Universe Will Affect the Results . . . Slightly. Various studies of these results are typically published by actuarial consulting firms, banks and rating agencies, and others. Each of those outfits typically uses a different research universe for its study – one firm using only the 100 largest pension sponsors; several using the S&P 500; and at least one purporting to use the S&P Composite 1500. As this chart indicates, research based on the 100 largest “elephants” can be expected to report higher pension funded status figures than the reports that use the S&P 500 set, i.e., several ticks above the 79.1% aggregate funded ratio indicated by the heavy dotted line. Even so, while stretching out to the S&P Composite 1500 or beyond will drag the pension funded ratio lower than the S&P 500 results, the pension funds for the additional companies are so small that the effect is diminished relative to the overall aggregate, yielding a funded ratio for the larger set that should be only slightly below the 79.1% S&P 500 level.
  • Median Results Mean Something! For the full S&P 500, in contrast to the aggregate funded ratio of 79.1%, the median funded ratio corporation by corporation was 72.3%, indicated on this chart by the lower, lighter dotted line. While the 79.1% number tends to get most of the press, that number essentially acts as though the companies with well funded plans could easily and would willingly share their pension surpluses (or lesser deficits) with the companies that have less well-funded plans. Which of course is not the case. Note that here, the research universe can make a huge difference: the median pension funded ratio for the top 100 pension sponsors is around 80%, essentially ignoring that well over a third of the S&P 500 plan sponsors have pension funds that may be facing legal restrictions and significantly higher pension costs and contributions due to severe underfunding.

And why are the elephants’ pension funds fatter than those of the smaller beasts? I’ll get around to that interesting “why” in a subsequent post.

(Remember, as I’ve previously disclaimed, posts such as this represent efforts of my favorite pastime. My formal work does not involve any of this, and none of it represents any position or comment that should in any way be attributed to my employer. Likewise, as always, it represents general personal impressions and should not be treated or used as formal professional advice.)

actuary cat feed subscribe to aftermath’s άctuary category

bumper sticker [www.internetbumperstickers.com] - actuary

Written by macheide

19 March 2009 at 5:20 pm

Posted in άctuary

2008 PBO FR 79% for S&P 500

leave a comment »

frpbosp5001999-20081For the S&P 500 as of 12/31/2008, the aggregate funded ratio of global defined benefit pension plans as of the end of fiscal years ending during 2008 was 79.0%, down drastically from 104.0% as of the close of the 2007 fiscal years, as measured on the basis of the ratio of market value of pension plan assets to projected benefit obligations.

We have yet to see whether 2009 will make this recent decline as severe as the 2001-2002 decline, although January and February certainly did nothing to stop the bleeding.

More details on funded status and other observations gleaned from corporate financial reports forthcoming soon.

(Remember, as I’ve previously disclaimed, posts such as this represent efforts of my favorite pastime. My formal work does not involve any of this, and none of it represents any position or comment that should in any way be attributed to my employer. Likewise, as always, it represents general personal impressions and should not be treated or used as formal professional advice.)

actuary cat feed subscribe to aftermath’s άctuary category

bumper sticker [www.internetbumperstickers.com] - actuary

Written by macheide

16 March 2009 at 5:31 pm

Posted in άctuary