Printer Friendly

Pension profits become corporate costs.

In mid-June of this year, Standard & Poor's fretted that the S&P 500 companies' underfunding of their defined benefit pension funds had jumped to $226 billion from $212 billion at the end of 2002. In mid-August, however, after substantial increases in stock prices and bond yields, S&P forecast that the aggregate pension fund deficit will be $182 billion at the end of the year, much lower but still a big number. And the Pension Benefit Guaranty Corp. (PBGC), the federal agency that takes over busted defined benefit plans, estimates that private plans are underfunded by $400 billion. These negative numbers are disturbing, especially since there was a $5 billion pension surplus in 2001 and $235 billion overfunding in 2000. That volatility also adds to the confusion and misconceptions shared by many about this complicated, highly technical, and regulation-ridden sector.

Too Big To Ignore

Still, pension plans are too big and too important to beneficiaries, corporations, Washington, and the economy to ignore. Their assets have leaped in the last two decades, and even after the big bear market they remain well over 100 percent of disposable income, as shown in Figure 1. As most are aware, pension plans fall into two general categories, defined benefit and defined contribution, the latter including 401(k) plans. As the defined contribution label implies, the contributions, usually with an employer match of a set percentage of employee contributions, are predetermined or defined. The investment results, however, are not guaranteed, regardless of whether they are directed by the employer or employee. So, the employer has no obligation for retirement benefits beyond its contributions, and there are generally no effects on corporate earnings beyond the deduction of those contributions in the years they are made.


In contrast, defined benefit plans--the traditional pension plans--guarantee the employee a set monthly income upon retirement or a lump payment and are normally funded entirely by the corporation. Thus, the company needs to contribute adequate assets to the pension fund to ensure that those retirement benefits will be paid on schedule. As a result of these ongoing obligations, defined benefit plans affect future as well as current corporate earnings.

In recent years, corporations have favored defined contribution over defined benefit plans, largely due to their zeal to convert variable into fixed costs in a world of nonexistent pricing power. A 401(k) contribution is normally tied to this year's employee compensation, but a defined benefit obligation is determined by past years of service and pay as well as future employee tenure and compensation. Also, the corporation is responsible for the pension benefits regardless of the good or bad performance of the pension fund investments.

Not surprisingly, then, the number of defined benefit plans has been shrinking in favor of defined contribution plans. In 1983, at their peak, there were 175,143 defined benefit plans but in 1998, the latest year for data, the number has dropped to 56,405. Meanwhile, corporate 401(k) plans leaped to 300,593, and defined benefit plans are primary for six of ten workers covered by pension plans. As a result, as shown in Figure 2, assets in defined contribution plans have been greater than those in defined benefit plans since the mid-1990s.


Nevertheless, Figure 3 shows that defined benefit plan assets are still over $1.5 trillion and continued to rise until stocks swooned. Given their long lives, they will continue to be important to corporate costs and profits for decades.



It's not surprising that there are many misconceptions about the effects of defined benefit pension plans on corporate earnings. The complexity of the calculations involved in determining pension fund net income is immense. Then there are the differences between regulatory and stockholder accounting. Regulatory accounting is geared to insure that defined benefit plans meet their obligations to retirees. It is concerned with underfunding and such important details as the interest rate used to discount future liabilities. This rate was once the yield on 30-year Treasury bonds for underfunded plans; but for now plans can use higher rates, which reduce the present value of their liabilities. Regardless, when assets drop below an average of 90 percent of the level necessary for adequate funding for three consecutive years or 80 percent in one year, sponsoring companies must contribute additional assets. These assets, however, aren't always cash. Northwest Airlines recently received permission from the Department of Labor to contribute stock of its majority-owned but non-publicly traded Pinnacle Airlines. General Motors and PanAm got similar dispensations in earlier years. United is entertaining similar thoughts to deal with pension shortfalls that equaled $6.4 billion at the end of 2002.

This may be throwing good money after bad. Because of its absorption of the pension funds of bankrupts like LTV Steel, the PBGC swung from a $7.7 billion surplus in 2001 to a deficit of $5.7 billion on July 31 of this year. Also contributing to the deficit: when the PBGC takes over troubled funds, it often finds that due to lax reporting rules, they are much more underfunded than their corporate sponsors believed. Bethlehem Steel reported that its pension fund was 84 percent funded, but the PBGC found it to be 45 percent, a $4.3 billion total shortfall. USAirways pilots' pension plan was supposed to be 94 percent funded, but turned out to be only 33 percent, pushing the total underfunding to $2.5 billion.

Many municipal defined benefit pension funds also are substantially underfunded due to big stock losses. As a result, a number are considering switching to defined contribution plans.

Contrary to the understanding of many, money flows between corporations and their defined benefit pension plans are a one-way street. It's essentially impossible for corporations with overfunded plans to pull money out. That was feasible years ago before corporate raiders took over firms with fat pension funds and replaced the fund investments with junk bonds. Even though many of those bonds were destined for default, their high interest rates allowed the takeover artists to loot the pension funds of seemingly excess assets. The ensuing national uproar led to punitive taxes on repatriated assets that took the fun out of that maneuver.

GAAP Accounting

Nevertheless, under GAAP accounting used for shareholder reporting, the year-by-year calculation of required pension fund expense for a sponsoring company can result in a negative number (i.e., an addition to corporate profits). Interestingly, that calculation is not a matter of subtracting liabilities from assets year-by-year, but of comparing the changes in various asset and liabilities categories. I am indebted to Lee Trad, actuary at Pricewaterhouse Coopers, for breaking this intricate process into five steps and helping us understand it. Needless to say, however, any misinterpretations and errors are mine. These five steps are given below.

One. Determine this year's service costs (i.e., the current price of the liability stream of future benefits attributed to employee activities this year). This cost results because employees have earned another year of benefits. The calculation involves estimates of future benefits based on employee longevity, early retirement, mortality, and so forth. These future benefits for the year are discounted back to the present, at the settlement rate, usually the yield on high quality bonds of the same duration as the average employee's working life. Settlement rates averaged 6.7 percent for S&P 500 companies last year.

As noted earlier, the higher the bond yield, the smaller the current value of those future benefits. It is estimated that a decline in long-term interest rates by one percent reduces the present value of liabilities by ten percent. Corporations have some flexibility in setting this discounting rate, which has been declining since the early 1980s along with overall interest rates. So, there has been an incentive to drag corporate heels in reducing the rate used in order to avoid bigger annual service costs and, hence, bigger hits to corporate profits. Fed Ex, for example, used 7.1 percent last year compared to the 6.7 percent average.

Two. Calculate the interest costs on past service liabilities, technically known as projected benefit obligations. Past service liabilities are the present value of retirement benefits earned in all earlier years, discounted at the same high-grade bond yield used in Step One. Note that full vesting is assumed. These past service liabilities are then multiplied by the same discounting rate since the current discounted value of those liabilities is greater this year than last year by the resulting amount.

Notice that the discounting rate to determine the present value of projected benefit obligations is the same as the rate by which they are multiplied to calculate this year's cost. So, in essence, the interest rate effect washes out. In other words, the assumed discounting rate does not materially affect the pension costs calculated in Step Two.

Three. Determine the expected annual increase in fund assets by multiplying the market value of the fund's assets by the assumed long-term rate of return on those assets. This sounds straightforward, but there is plenty of room for corporate discretion.

A number of companies, including IBM until this year, assumed returns of ten percent or even more. Portfolio gains of that magnitude were achievable in the late 1990s boom days but not since then nor, in all likelihood, in the years ahead. In fact, many firms are cutting their assumptions for long-run returns, and the average for S&P 500 corporations is now nine percent. Fed Ex has cut its assumed return from 10.9 percent, but still is high at 10.1 percent. Warren Buffett's Berkshire Hathaway uses 6.5 percent, down from 8.3 percent in 2000. If my forecasts for much-lower stock appreciation and bond yields are realized, many others will be forced to cut their assumptions even below 6.5 percent in the next several years.

In addition, the assets that are multiplied by the assumed long-term appreciation rate in this step can be smoothed or averaged over as much as five years. Thus, the good investment years in the late 1990s offset the bad results since then in these smoothed averages which are then multiplied by the nine or ten percent assumed rates of return. As a result, the investment results in this calculation for many firms were much better than the actual performance in 2000-2002.

Note that the combination of Steps One and Three can be powerful in creating profits for the pension plan's sponsor. If the plan's smoothed assets exceed the present value of liabilities and the assumed long-run asset appreciation rate is much greater than the settlement rate, it is a favorable double whammy. Smaller liabilities multiplied by a lower rate are subtracted from greater assets multiplied by a higher rate. It works both ways, however, as I will discuss later.

Four. Adjust for new or amended plans. For example, if an old plan is closed and a new one established and the initial assets exceed the liabilities, the difference would result in negative amortization of the overfunded amount (a plus for profits) over the next 12 to 15 years. This was the case for many firms in 1987 when the current pension accounting rules were enacted, but those corporate benefits are largely history now, 15 years later. Instead, underfunded new and revised plans will cause positive amortization, which will negatively affect profits.

Five. Calculate the ten percent corridor, which is the way the significantly overfunded or underfunded plans are forced to recognize reality in their accounting. First, calculate ten percent of past service liabilities and ten percent of market-related assets--smoothed or current market value, whichever the corporate sponsor chooses--and take the greater of the two. That is the ten percent corridor.

Next, calculate the cumulative difference since the plan's inception between the assets based on the assumed appreciation rate and the market-related assets. Add in the cumulative amount by which liabilities differ from the long-run plan due to greater-or smaller-than-projected pay increases, longer or shorter life spans of retirees, bigger or smaller employee turnover, change in the discounting rate, and so on. Then compare the resulting net amount with the ten percent corridor. If it is less than the corridor, nothing results. But if it is more, the difference is amortized over the 12 to 15 year working life of the average participant. So, if investment performance has been superb over the plan's entire life and lots of people quit before retirement while retirees died young, negative amortization would result and aid corporate profits. But poor investment performance and the declining discounting rate can create positive amortization and a negative hit to the sponsor's earnings. Moving from inside the corridor, with no effect on corporate earnings, to outside can have a decided impact on profits. It can be like stepping off a cliff.

After all five of these steps are completed, the net result determines the corporation's accounting contribution to the defined benefit pension plan, which, as noted earlier, can be positive or negative, resulting in a loss or profit for GAAP corporate earnings.

Corporate Help

It is clear from corporate reports that the corporate earnings of many companies have been helped by their defined benefit pension funds in recent years, but investors now know that those profits were of dubious quality. Many firms took full advantage of such assumptions as the return on assets, the discounting rate, and smoothing of asset returns to aid their bottom lines. They made other favorable assumptions like forecasting a continuation of the high employee turnover rates of the late 1990s when jobs were plentiful. In 2001, Verizon lost $3.1 billion in its pension fund but reported $1.85 billion in pension income. With its assumed 9.25 percent return on assets, Verizon included a $4.8 billion portfolio credit in its Step 3 calculations. The salutary effect on the firm's bottom line boosted its pretax profit to $2.77 billion and helped substantiate incentive bonuses for company executives. Similarly, IBM lost $2.4 billion in pension assets but booked a $4.2 billion gain with its ten percent assumed long-term asset gain. After benefit and service expenses, net pension income was $1 billion, or almost 10 percent of IBM's $10.95 billion in 2001 pretax earnings. In contrast to these firms with large defined benefit plans, corporations that rely on defined contribution plans, often newer and tech-oriented firms, have trivial pension fund income and, therefore, little effect on their reported earnings.

The Game Is Over

Even though assumptions about pension asset appreciation, smoothing, discounting rates, employee longevity, and so forth. can produce corporate accounting profits for a while in the face of adverse conditions, accounting does eventually catch up with reality. The corridor in Step Five forces it to. The market value of defined benefit plan assets soared with stocks in the late 1990s but then declined with the bear market. Meanwhile, the big decline in bond yields boosted liabilities. The net result is that assets are now below liabilities; and balances have swung from huge surpluses to big deficits, as mentioned at the outset.

Obviously, the earlier pension fund aid to corporate bottom lines is about over unless you assume a return to the nonstop roaring bull market of the late 1990s and significantly higher interest rates. It is now clear that pension plans provided a portion of reported earnings in past years that was essentially artificial and, therefore, of low quality at a time that many corporate managements did almost anything to enhance earnings in order to satisfy investors. It is estimated that corporations added $50 billion annually to earnings from their pension fund net income in recent years. Many corporations have pumped the defined benefit pension fund well dry.

Indeed, the outlook is equally grim. If my forecast of mild deflation is correct, Treasury bonds will yield three percent while the total return on stocks will average four percent. So, the present value of pension liabilities will leap as assets grow slowly. As implied by my earlier discussion, the narrowing of the spread between the rate of asset appreciation and the discounting rate--applied to liabilities that exceed assets--will completely reverse the favorable double whammy of yesteryear. This new unpleasant world for defined pension funds also will make it difficult for a number of plans to stay within their corridor. The huge likely downside volatility is only adding to pension profits' low-quality image.

Accounting Profits Turn to Cash Payment

On balance, what earlier were accounting (but not cash) transfers from defined benefit pension plans to profits for many firms will turn to reductions in corporate earnings as well as cold cash pension fund contributions. The overall effect will probably be substantial because of the double blow--the elimination of a plus for profits and the advent of a negative for earnings and cash flow.

This will be especially painful for older, slower growing, heavily-unionized firms with large defined benefit plans and other heavy legacy costs, such as retiree medical outlays. In some steel company plans taken over by the PBGC, only one in eight participants is actively working. Adding further pressure, many of these corporations will probably be under intense heat by investors to increase their dividends to meaningful levels to insure adequate total returns, further straining cash flow.

Reactions to the new defined benefit pension reality will be significant. Look for the long-term trend of switching from bonds to stocks in pension portfolios (as shown in Figure 4) to be reversed as sponsors seek stability and predictability of returns and strive to match asset and liability growth. A likely narrowing of the spread between stock and bond total returns will reinforce that switch. To be sure, the bear market in stocks since early 2000, along with the high-quality bond rally, has provided the impetus. The once sacrosanct target of 70 percent stocks and 30 percent bonds in pension funds always struck me as the result of the long 1982-2000 bull market in stocks, not anything engraved on stone tablets. Those percentages may end up exactly reversed with 30 percent stocks and 70 percent bonds.


In addition, the shift from defined benefit to defined contribution pension plans will probably speed up. Corporate management will no longer see defined benefit plans as cash cows but as albatrosses around their necks and will close those plans even though union and other employees who lost bear market bundles in their 401(k)s will object. Some will compromise on cash balance plans--defined benefit plans that have individual employee accounts (which are portable) but that leave leaner pensions for long-time employees. Others will reduce benefits in other ways.

Watchdogs--Late As Usual

As usual, the watchdogs who police and regulate defined benefit pension plans did not show up to lock the barn door until the horse was stolen. Only late last year did Standard & Poor's redefine Core Earnings to include actual, not assumed, pension fund investment results. Even further behind is the Financial Accounting Standards Board that sets accounting standards. The Board is thinking--just thinking--about disallowing corporations from including pension fund profits in their earnings. It is expected to study the issue for three years before reaching a decision. By then, it probably won't make any difference.

A. Gary Schillingis president of A. Gary Schilling & Company.
COPYRIGHT 2003 The National Association for Business Economists
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2003 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:Forum On Emerging Issues
Author:Shilling, A. Gary
Publication:Business Economics
Geographic Code:1USA
Date:Oct 1, 2003
Previous Article:Forecasting with leading economic indicators--a neural network approach.
Next Article:The Los Angeles County Economic Development Corporation.

Terms of use | Privacy policy | Copyright © 2018 Farlex, Inc. | Feedback | For webmasters