Pension Funding question?

[Guest Hickory6]

Hello, everyone...

A recent study from Merrill Lynch research called "Decomposing Pensions" examines the funding status of DB plans in the S&P 500, looking at both the funding status of ERISA plans alone as well as the funding status of all future obligations, including those not covered by ERISA such as medical benefits.

According to the study, pension and post-retirement benefits are underfunded across the S&P 500 by 23%, expressed as a percentage of fair value of current pension assets.

Some companies are using expected returns on pension plan assets as high as 10% per year (i.e., General Motors), a figure which John C. Bogle, founder of the Vanguard Group, and Big League Bond fund manager Bill Gross dismiss as 'pie in the sky.'

But somewhere, sitting around some conference table, some people are still arguing the case for a 10% expected return on pension assets. Is it a matter of CEOs and Boards of Directors not listening to the arguments of pension trustees? Have ERISA fiduciaries sounded the alarm bell in the past? Have your concerns been dismissed by current-earnings-focused management?

What are the plans at your companies to bring pensions back close to funded status? What are you expecting in terms of future returns on pension plan assets?

Or is the media making too big a deal out of the pension issue?

Your thoughts, comments, concerns, suggestions?

Jason Van Steenwyk

Reporter,

Mutual Funds Magazine

954-229-6907

Jason_Van_Steenwyk@timeinc.com

[MGB]

You hit the nail on the head. The media is making a big deal out of a nonissue.

[Guest Hickory6]

Ok...what is the media not getting?

[MGB]

These companies have been using the same assumption for the past decade or more. When the plan was earning 15-25% return each year in the mid-90s, no one was thumping their chest on a soapbox claiming that 9 or 10% was unreasonable and that they should switch to something higher.

The additional actual return those years created gains that are being spread over the future (typically 15 to 20 years). Now that they are receiving returns lower than 9 or 10%, they are producing losses offsetting those gains that are still being carried over. The net result is exactly what the methodology expects: you have gains during an upcycle and losses during a downcycle and they even each other out by being spread over a long period of time.

The expected return assumption is a long-term expectation of how these gains and losses will work themselves out. It is not intended to match any one-year expectation or actual return.

Has there been a fundamental change in the investment environment that would indicate the next 30, 40 or more years (the time frame of pension projections) are going to be very different than what was expected in the future just a few years ago? Obviously, there are some that claim this. Others just see the current situation as a down market that will recover (albeit maybe later than sooner). If you believe the first contention, you should lower your expected return. If you believe the second contention, you should INCREASE your expected return...not only will there be the same long-term expectations, but there will be additional return to account for the rebounding of the market. Neither side is "right", nor can anyone claim with certainty that one side is clearly superior to the other. Sitting in the middle, one should stay with whatever they were expecting before (unless, of course, they were expecting too much to begin with...but then, there was that 20 to 25% return for a number of years).

Is 10% too high? I would not be comfortable with it. But then, there are very few firms going this high. Most are in the 8 to 9% range.

The funded status of plans is a completely different issue and is unrelated to these assumptions of the expected return on assets. The funded status is based on the discounted present value of the liabilities compared to the actual assets. These are not discounted at the expected rate of return (the 9 or 10% alluded to above); they instead are discounted at the rate found in high-grade bonds which is currently around 7%. If interest rates edge up in the future, the underfunding could go away just from a change in this discount rate.

[mbozek]

Jason:

If you had taken the time to research the issue you would have found that this type of report is issued by investment firms every time there is a bear market and is discredited when the bull market returns.

The 23% figure that you cite assumes that the plans would be terminated today-- if the plans continue then the underfunding is amortized over a specified period of time ( e.g., 10 years for pension plans). If the value of the pension assets increase in futaure years then the underfunding will decrease or disappear and be replaced by a surplus amount. Also most retiree health benefits are not guaranteed- Employers reserve the right to reduce or eliminate such benefits in the future. Employers can also terminate pension plans and cancel liabilities for future benefit accruals. Neither John Bogle or bill Gross are fiduciaries for retirement plans- Bogle believes in index funds and Gross is a Bond Fund manager. If the plan assets consists of 60% in stocks and 40% in bonds the historical average rate of return should be about 8.4% (10% stocks x .6 and 6% bonds x .4). Individual plans may be able to raise the return above the historical average the same way that some investment managers can produce long term returns in excess of the average return for stocks and bonds.

[Guest Hickory6]

Mbozek

<>

Well, that IS why I'm here, isn't it?

At any rate, I don't mean to present Bogle and Gross as ERISA fiduciaries...they are not. However, they are capable market forecasters.

I suppose the problem is ONLY a problem if markets do not match assumptions. If GM is assuming a 10% return on pension assets, for example (which it is, according to the Merrill Lynch analysts), and their 60/40 split returns its historic average say,8.4% over the next 10 years, do we not still have a problem which will either eventually force GM to restate earnings downward, or require them to make additional cash infusions?

As for the assertion that such criticisms are levied every time there's a sustained Bear Market, isn't this the first sustained bear market since ERISA became law?

[MGB]

GM is not assuming 10% for funding calculations...the assumptions for funding is made by their enrolled actuary. That calculation is the only thing that will cause future additional contributions or not. The 10% is completely irrelevant to whether or not they will need to fund more. This 10% figure is used for financial accounting only. You are mixing up calculations for accounting purposes versus calculations for funding purposes. The two are completely different and unrelated. Each has different rules applying to them on who sets the assumptions and what criteria should be used for setting them.

In the accounting calculations, if they only earn 8.4%, then the earnings will automatically be adjusted downward due to the mechanics of the amortization of gains and losses...there will not be any "restated earnings downwards", the actual return works its way into the calculations without their doing anything different.

[mbozek]

I guess you are too young to remember the 22 % decline of the Dow on Oct 21, 1987. It took 16 months for the Dow to get back to the pre Oct 21 level. Also, there were similar articles on pension funding in 1994 and 1995 . As I previously said you need to do further research to learn the distinctions between the actuarial assumptions for funding under the IRC and accounting rules for balance sheet reporting of retirement plans.

What have you been reading that makes you think that Bogle is a capable market forcaseter? Bogle's thesis on investing is that since no one can predict the future, indexing is the only strategy that makes sense for non professional investors because it guarantees an average market rate of return. Bogle has no background in advising professional investors.

[Ron Snyder]

At the Enrolled Actuaries Conferencce last year we had a presentation relative to long-term rates of return and reasonable expectations. The consensus of the presenters was that 8% is too high for actuaries to assume for long-term rates of return. While it is arguable that certain plans have achieved higher rates of return than that, it is also the case that many have not achieved even 6%.

A review of the General Motors defined benefit pension plans discloses the following assumptions employed by the actuaries for the plan, Watson Wyatt:

Salaried Employees Plan (001) Interest Rates

RPA - 6%

OBRA 87 - 6%

Plan Liabilities - 9%

Actual investment return - 13.0%

Hourly Employees (003) Interest Rates

RPA - 6.29%

OBRA 87- 6.59%

Plan Liabilities - 9%

Actual investment return - 11.9%

[david rigby]

Well, to be precise, the quoted items for General Motors are found in the Schedule B for the plan year beginning 10/01/1999. Thus, for example, the actual investment return would apply to the actuarial value of assets for the 12 months ending 9/30/1999. www.freeErisa.com (It is very possible that the actual yield on market value of assets was higher.)

But more importantly, as MGB points out, let's not get confused about actuarial assumptions for funding purposes (for example, those items quoted by vebaguru) and accounting purposes (which seemed to be the focus in the original post).

And another point, the assumed rate of return (for accounting purposes) can legitimately be affected by the asset mix. If the plan sponsor chooses a more conservative investment strategy such as 40% stocks and 60% bonds, then the expected yield (long-term average) would be lower than a 60/40 mix.

[mwyatt]

The popular (financial) press is indeed focused on the FAS-87 assumption and its impact on companies' earnings. As an actuary I know that FAS-87 has nothing to do with physical dollars required to go into a plan, and that most large plans utilize some form of smoothing method in the 412 valuation to dampen the swings in the market on contribution liabilities. On the other hand, when I read about major (say Dow index) companies that had over 30% of their stated earnings from penson income over the past few years, then I think that we as a profession should have an answer to these inquiries. Warren Buffett, for one, is not anyone to sneeze at (especially after all those .commers are back at McDonalds wearing funny little hats and handing out change again).

[David]

I agree with mwyatt. The only large plan work I ever did was many years ago as an entry level actuarial analyst so I am not really familar with exactly how these things work. That being said, I wonder, for a company such as GE that is recogonizing a large pension asset each year, what is the likelyhood of GE recapturing those funds and putting them to use to actually create revenue? And, would they be able to recapture all of the overfunding or only a portion of it? It seems to me that it is misleading to use the pension plan overfunding as such a signifiacnt source of income.

[mbozek]

The answer is that there is 0% chances of recapture becuase of the 50%/20% tax on pension reversions plus federal income and state income taxes. However, investors only focus on earnings of companies and under the accounting rules earnings from pension plans that are overfunded count as corporate earnings. E.g., IBM and GM stock are down this year because the earnings of the pension plans have declined. Also companies adopt cash balance plans to give a one shot boost to earning by creative financial adjustments. A few years ago Bank of America rolled out a cash balance plan with a self directed account feature. The purpose was to encourage employees to transfer 401(k) funds to the CB plan because the self directed funds were counted as plan assets with no offsetting liabilities under the CB plan because they were not regarded as benefit accruals. The quarterly earnings of B of A increased by several hundred million $ as a result of this transfer. The bottom line is that investors do not understand how companies can adjust the balance sheet for employee benefits (e.g., ignoring the value of stock options).

[MGB]

In order to recognize income, GE had to have recognized expense in the past, and/or would have recognized a certain amount of expense in the future. The posting of income is a recognition that they either overexpensed (and contributed - creating an overfunding) in the past and are now reversing that, or that their future expense has decreased due to experience in the past. The bottom line is that the plan is overfunded (which could have occurred because single digit interest rates were projected during the years before they actually realized 20%+ returns). If they had not overfunded it (and expensed all of that in the past), then they would not be now reversing it to recognize they have actually experienced the additional overfunding.

The focus on the current income statement of one year is inappropriate. That is not what financial accounting is meant to do (that is what the measure "cost accounting" does). Of course, the popular financial press has created this mythical overzealous focus on the current numbers. Financial accounting is supposed to be looking at the long-term nature of commitments and effects. In the aggregate, over the life of the plan, GE has not recognized net income -- that would be theoretically impossible. Any current income recognized is only a mechanism to reverse overexpensing in other years. (Now it perhaps would be more theoretically correct to "restate" prior earnings for decades to recognize the pension income as an offset to expenses in those years instead of recognizing it as current income, but that would create even more confusion.)

Yes, they put that income (even though there is no cash flow) to use to produce future revenue...through their employees. This is not some sideline operation totally set apart from the business. The pension plan has a major impact on their personnel.

The ability to recapture the excess assets is irrelevent to the nature of why it is accounted in the way that it is.

The only time that the above theoretical reasoning for financial accounting to recognize income is not appropriate is when the assets exceed the present value of total future benefits (with full projections). (There is no large company in this situation.) Income that is generated above that point could then be tied to the ability to recapture the funds. Under IAS 19 (international accounting for pensions) this is, in fact, what they accomplish through the "asset ceiling". They allow pension income, but only up to a point that is theoretically sound....i.e., if there is no future expense that is being offset because of current conditions and/or on top of that, the assets cannot ever be recovered (after the cost of termination including taxation, etc.), then income above that point cannot be recognized. Generally, no company in the US would hit this asset ceiling given the current amounts they are recognizing as income.

I became a member of the Academy's Pension Accounting Committee in the late 80s (and am a current member) and am constantly in conversations with the press about the nature of financial accounting and where pension calculations fit into the theory and basic principles of that style of accounting. One needs to understand the precepts, fundamentals, and objectives of financial accounting before they focus on the pension aspect. There are other styles of accounting that are based on other fundamentals and objectives under which pension income may not be appropriate.

[mwyatt]

Just as the tech companies had inflated earnings due to options, creative goodwill, etc. (for me the ephiphany was picking up the paper one morning in 1999 to find that Yahoo had bought Geocities for a higher price the same day that Ford bought Volvo - of course, Geocities' product was free websites for individuals while Volvo only sold expensive cars, but my stodgy mind at the time couldn't understand the "brilliant" logic that valued this deal) similiar manuevers were going on at the larger companies with pension accounting, etc.

I agree with MBozek's analysis that even best case with tax credits that they can only recoup about 50% of the overfunding. If they have to reflect reversion on taxes, amount is dimes on the dollar. How you can count on a trust for company earnings really doesn't make a tremendous amount of sense, but I'm just an actuary, not a CFO.

None of these manuevers seemed to bother people on the way up the curve (except maybe Abelson @ Barron's). Now that equity markets have tanked, people are looking for someone to blame (in some cases blame is easily assignable, witness Enron and Worldcom, sometimes the masses are looking for scapegoats). Maybe it is time to rethink the real usefulness of FAS-87; impact has strayed from the original intent of providing comparability of future pension liabilities to a tool to inflate company earnings.

[Ron Snyder]

Which brings us back to the original question posed by the reporter: Are pension plan liabilities undervalued significantly?

Each of you has made good points. Interest rates are used for several purposes. Some are objective and others are subjective. Ultimately, all of them are tools to be used for a specific purpose, to give a partial picture. Is GM's plan underfunded or overfunded? It seems to me from the available data that their actuaries and financial analysts (including the Board of Directors) all have defensible positions.

Our writer friend was looking for a litmus test, a black-letter answer to the question posed. The problem is, such answer or test does not exist. Reasonablility of assumptions, like beauty, is in the eye of the beholder.

[Guest Hickory6]

Well, black-letter answers are nice. But I write for a monthly, and if something were that easy, then what do they need me for? They could just have a computer generate stuff from a press release, get two "just add water" quotes from "experts" selected for the predictable regularity with which they say inflammatory things, season to taste with trite maxims and bad puns, and serve lukewarm over a bed of advertising.

Of course, I write stuff that has to fit in these tiny little boxes, and I'm always dodging issues rather than open a can of worms that I can't close in 200 words or less. Maybe this is one of those stories.

But if the rest of the financial press, as you seem to suggest might be missing the point--using financial accounting for a purpose for which it is not well suited, then that's an interesting story in itself, I would think.

But it's not just the financial press--Merrill Lynch came out with a study recently called Decomposing Pensions (16 August). I'll post the fundamental conclusions here.

The methodology is as follows:

--The analysts manually collected over 60,000 data points from the individual 10k's of the S&P 500 companies.

(Was this a mistake? Because of Should they have instead used Form 5500s?)

As you state, since they're focusing on 10ks, they're probably also focused on FAS 87 accouting standards. Being a reporter, and not an accountant, I understand the distinction only through a glass, darkly, at this point, although it will get better.

Their findings are:

--From a quatlity of earnings point of view, accounting and disclosure rules are open to manipulation by aggressive company management. (but is this something that could be remedied to some extent by incorporating 5500 data into estimates and disregarding 10k data related to pension funding?)

--Complex and illogical in in some of its methodology, particularly where they lead to holding significant liability such as unfunded plans off balance sheet

--Likely to lead to "late" earnings estimate downgrades or at least uncertainty, as analysts understand the ramifications of declining asset values and unrealistic long term return assumptions on earnings. (But if I understand you correctly, the actuarial assumptions are more moderate than the 10ks reflect, and thanks in part to the practice of smoothing, are somewhat self-correcting, as companies are now recognizing revenue already in their pockets which they did not recognize before, when companies were earning 20% on a pension plan instead of 10%)

--The S&P 500 companies' pension funds, including post-retirement benefits) were 0.3% overfunded in 2000, as a % of fair value, and 23% underfunded in 2001. (If I understand you correctly, though, a chunk of that underfunding will be corrected as soon as interest rates rise again, just by virtue of a mathematical calculation).

--Net pension and post retirement charges reduced net income by 7.2 billion dollars, or by 4%, in 2001.

--For some reason, energy industry pensions were significantly more underfunded than any other industry.

--The study insists that for the past two years, GM has been using an expected rate of return of 10%., and breaks down the GM 10 k, in a blow up graphic, to illustrate their reasoning.

The next round of monthlys will probably pounce on this. I'm not in as much of a rush, if I don't feel I understand it.

Are they off base?

Jason Van Steenwyk

Reporter

Mutual Funds Magazine

954-229-6907

[mbozek]

Jason: arent you ignoring a more fundimental question : Should a serious/ informed/ astute investor even take into account a company's pension assets when making an investment decision? since pensions are long term liability with a pepetual maturity like some 100 year bonds issued by Boeing and Disney does it really matter whether plans are over or underfunded at any time prior to liquidation of the plan sponsor since the surplus cannot accrue to the shareholders and the liabilities are smoothed out over the long term? There are exceptions of course in certain industries such as steel where bankruptcies partially result from long ago negotiated pension obligations under union contracts, e.g. Bethelem Steel has 5 retirees for each worker, but does it really matter if IBM has a 3, 4 or 5 billion surplus in its DB plan? Or that IBM expects an 8, 10,or 12 % return on its plan assets? IBM cannot unlock the plan value to benefit its shareholders. GM may have pension and health benefit funding issues but they can be solved by reductions in force as was done in the 80's. Fewer workers are more productive workers. Maybe the joke is on the analysts and investors who see this pension funding issue as a significant indicator of the worth of a corporation in making investment decisions. Is it more important the Phillip Morris's has no funding liability for its pension plan or that it has singificant risk from cigarette litigation? Maybe you should ask why did

Merrill Lynch choose to release this report now?

[Guest Hickory6]

mbozek,

Thanks for the reply. You do touch upon an interesting point. My original thesis was "does the underfunding of pension funds particularly affect old economy companies with slow-growing earnings but expanding numbers of retirees? I.e., the steel industry?

If so, then does that then mean that funds which invest in these kinds of companies --in effect, what we generally call 'value funds'--will be struggling against a headwind as they move to infuse scarce cash into their pension plans?

And so I'm back to the original post.

Should investors look at funding levels? Well, if pension plans are contributing a whopping 25%-30% of the companies earnings, and we have reason to suspect that the inflated pension earnings will not continue, then yes--with the huge assumption that p/e levels will remain constant, and who knows, right?

Me, personally, I don't worry about it. I'm primarily an indexer, and everything I buy right now is through a fund which I plan to hold for decades. And I expect that the smoothing will pretty much work itself out over time. (Actually, I kind of rely on that).

I recognize--and must write for--people with shorter term horizons and different approaches to investing.

[david rigby]

Some valid points from mbozek.

An issue that has gotten little attention in this discussion is cash flow. While plan X may be 105% funded (and there are many methods of measuring this), plan Y might be 75% funded. Of course, it is desirable to make these measurments in the same manner and with similar sets of assumptions (especially the discount rate used for the time value of money). Assuming that, then the lower funded plan will require greater contributions (that is, more cash from the plan sponsor). This might be useful to a prospective investor.

I don't mean to imply that cash flow is the only issue. However, it might be worth mentioning, particularly in a published article, that the total cost of any pension plan is not what goes in but what comes out in the form of benefits and expenses. The cash that is used to pay these benefits and expenses comes from investment income or cash contributions. If investment income is less than anticipated, then additional cash contributions will be required.

[Ron Snyder]

I would like to respond to your original thesis, whether the underfunding of pension funds particularly affects certain value stocks more than other companies.

IMHO as an actuary familiar with FAS 87 as well as 412, 404 and 411 valuations, the FAS disclosure rules sufficiently address the question of underfunding or overfunding of retirement plans on an ongoing basis. In my mind the more serious issues are: (i) Understated liabilities of DB plans likely to terminate due to changes within an industry, and (ii) understated liabilities for post-retirement health plans that use a 3-5% Medical COLA when actual increases are running from 10-14%.

Should investors look at funding levels? Absolutely! Pension plans should not be contributing anything to a company's earnings, let alone 25-30%. When a paper profit is created because of supposed or actual "gains" in retirement plan funding levels, it should be considered as an extraordinary gain and nothing else.

[mbozek]

Jason: Arent the arguments you raise against investing in value funds which own stock in old economy companies because of unfunded liabilities also a reason for investing in these co-- since there is a perception the retirement plans are underfunded and additional contributions will be required which will reduce the E & p of the sponsors, an investor can buy these companies at a discount. When the economy recovers and the plan assets increase in value the sponsor will be able to cease the additonal contributions and the E & p of such companies will rise along with the stock market. The share price of the value fund will rise to reflect such increased valuations. Also the impact of unfunded pension liabilities will be minimal in the future because over 200 large corporations have established cash balance plans which do not not have unfunded pension liabilities-- indeed CB plans are established to cancel unfunded future liabilities for those employees who are not grandfathered under the old final ave pay formula in order to fatten up the balance sheet. Investors do not take unfunded health care liabilities seriously because most retiree health care is not guaranteed and can be changed/reduced/eliminated by the employer.

The steel industry is not a good example of old economy cos with pension plans because it is not competitive in the global market place, not because of its pension plan liabilities. US steel cos cannot compete with foreign steel cos which have newer plants which produce steel cheaper, cheap foreign labor, govt subsidies and a lack of interest in controlling pollution. The US auto industry will survive to fund its pension plan liabilities but the airline industry will shrink and many more carriers will go the way of pan am, eastern and TWA regardless of their pension liabilities because high labor/capital costs and prolonged low cash flow are bad for any business. Pension liabilities are the tail of the dog.

[mwyatt]

Actually an interesting epiphany just struck me reading Mbozek's comment: one of the complaints about present stock prices is the high P/E ratios compared to historic levels. However, if we can equate a nonsignificant % of the earnings to pension/medical costs, which weren't present in financials prior to 1987 (think effective date of FAS-87 and 106), then maybe these current P/Es aren't so high after all (have to think optimistically, although my faith in most "real life" financial folks ain't so high at present - if they really knew what they were doing, why didn't they cash out before it all got wiped out?).