A pension plan for today.

While the profile of today's work force goes through some dramatic changes, RJR Nabisco tries to stay one step ahead by designing its pension plan for all people.

Eighteen months ago, RJR Nabisco began a review of its pension programs. The company wondered if its plans were properly positioned to attract and retain the right employees in the work force of the 1990s and beyond. RJRN concluded that it needed a new type of pension plan to meet the benefit needs of an increasingly diverse work force. That new plan, named the Pension Equity Plan, was introduced to RJRN employees on January 1, 1993.

A TWO-TIERED WORK FORCE

Pension plan design has generally followed work force demographics. Prior to the 1980s, it encouraged and rewarded career employment through final average defined benefit plans that typically provided retirement income at age 65 equal to a percentage of an employee's average compensation over his last five years of service. By the mid-1980s, plan designs adjusted to younger and more mobile workers through cash balance plans that made annual additions (for example, 5 percent of this year's pay) to individual accounts, which grew at an interest credit rate stated in the pension plan document.

Today, the demographics of the work force are changing again. The eldest members of the baby boom generation are now 47 years old. Over the next 10 years, as this generation passes age 50, their thoughts are sure to turn to retirement security. The boomers won't be nearly as mobile in their 50s as they were in their 30s and 40s. However, the baby boomers will still be needed in the work force, because there's a "baby bust" generation right behind them, The baby busters will be young and, because there are so few of them, they'll be in a seller's market, perhaps giving them the ability to be even more mobile than the preceding generation.

In short, the United States is moving toward a fully employed, two-tiered work force. The emerging work force won't simply be older--it will also be very different. And with this bifurcated work force, America will need pension designs that are fair and equitable to all groups: short- and long-service employees, early- and mid-career hires, men and women who leave the workplace and return a few years later, high- and low-paid workers, and the young and old.

Traditional defined benefit plans provide very little buildup of value in the early years, while defined contribution or cash balance plans provide much less benefit buildup in the later years. RJRN knew it needed a design with the best features of both types of plans.

DOWN TO BUSINESS

RJRN has two domestic operating companies: R.J. Reynolds Tobacco and the Nabisco Foods Group. Until now, the two units had very different existing pension programs. The Tobacco Company had a traditional final average defined benefit plan, under which early retirement eligibility was age 55 with 20 years of service and unreduced benefits were available at age 55 with 30 years. An unusually high percentage of employees qualified for the latter condition. Tobacco primarily has a career work force, but it's much older than average, and regular retirements over the next decade will probably produce fundamental changes in its demographics.

At the other end of the pension spectrum, Nabisco had a cash balance plan for salaried workers, designed for a mobile employee base. However, Nabisco was making strategic mid-career hires (and these employees don't typically do as well under a cash balance plan). A significant portion of the population was not actually mobile. And the aging of the baby boomers was expected to create a need in the future for more traditional elements within Nabisco's pension program.

In short, both operating units had good reason to consider changing their pension plans. So RJRN created two working groups to review the programs, one at Tobacco and one at Nabisco. The company asked each group to separately determine an appropriate plan design for its own employees.

Important, though rather commonplace, objectives guided both groups: The company wanted to accommodate different career patterns and the need for staffing flexibility, provide competitive benefits, be easy to understand and communicate, provide portable benefits (like Nabisco's plan), provide benefits after retirement that are related to pay just before retirement (like Tobacco's plan), comply with complex IRS requirements and be cost neutral.

In reviewing the retirement plans at Tobacco and Nabisco, the working groups soon realized the plans were, in many ways, reverse images of each other. Often, an advantage of one was a disadvantage of the other. Here's what they found:

* Midcareer Hires--Tobacco secured a reasonable retirement income for midcareer hires with its final average defined benefit formula. After 20 years, a participant had a right to retirement income of 35 percent of final average pay at age 65 (less a small portion of his Social Security). In contrast, Nabisco's benefit was defined as a cash balance account that grew with interest. Interest couldn't accumulate sufficiently in a short period to provide a sizable retirement income to a midcareer Nabisco hire.

* Portability--Nabisco's terminating or retiring participants had portable benefits, with an ability to elect a lump sum (or an annuity) whenever they left. Tobacco's benefits were not at all portable, and participants who terminated had a right to retirement income only commencing at age 55 or 65, which was often far in the future.

* Relation of Benefit to Salary Prior to Retirement--Since it had a final average plan, Tobacco's benefits were directly related to final average salary. At Nabisco, benefits were based on salary over one's entire career plus interest credits, and the result might or might not relate well to one's pay just before retirement. As a result, benefit adequacy was erratic; some participants did very well, and some didn't.

* Tangibility and Relevance--Almost all employees understand and appreciate lump sums, as contained in the Nabisco cash balance plan. On the other hand, the promise in a final average defined benefit plan for a deferred income starting at age 55 (or 65) is remote to young employees and of little value to mobile employees.

* Administration--Traditional defined benefit plans have PBGC (Pension Benefit Guaranty Corporation) premiums, contingent employer liability, complex funding and nondiscrimination rules and additional actuarial work. Defined contribution plans have none of this, but they do have monthly individual recordkeeping. A cash balance plan has the worst of both worlds, with the full array of defined benefit requirements plus individual recordkeeping.

* "Take-Aways"--Unless a retiree had at least 30 years of service, the early retirement benefit in the Tobacco plan was presented as a reduction of the age-65 benefit. This is a typical design element, but employees commonly view it as a take-away.

In contrast, a participant's account in Nabisco's cash balance plan would grow over time. At retirement, a participant would take his account as a lump sum or convert it into an annuity. There was no apparent takeaway. Ironically, the reduced early retirement pension in a traditional defined benefit plan is generally worth more than the account balance in an equal cost cash balance plan.

* Compliance--Cash balance plans are defined benefit plans designed to appear to participants like defined contribution plans. The Internal Revenue Code didn't anticipate this approach and, although these plans have existed for eight years, gray areas remain. Some recent reactions from the IRS signal a continuing potential for significant compliance problems in many cash balance plans. In contrast, the compliance standards for final average defined benefit plans are much more concrete.

* Benefit Buildup Pattern--With its unreduced early retirement benefit at age 55, the Tobacco plan had a marked benefit discontinuity. Benefits were quite a bit better if an employee left at age 55 than at 53 or 54. As a result, virtually nobody left voluntarily between ages 40 and 54 (or after age 60), distorting work force demographics. The Nabisco cash balance plan didn't have similar discontinuities.

* Fast Trackers--In a cash balance plan, a fast tracker, with higher than average salary increases over a career, will receive a lower retirement income, when expressed as a percentage of final pay, than an average employee. The fast tracker's salary increases keep pace with the plan's interest credits, negating the growth of the account as a percentage of pay.

On the other hand, interest credits compound at a faster rate than slow-track salary increases, giving the slow tracker better benefits as a percentage of pay. For example, a 65-year-old slow tracker (earning 3-percent raises) with a 30-year career in a hypothetical cash balance plan might be entitled to retirement income (if he chooses an annuity) of 39 percent of pay, while the 65-year-old average employee (earning 5-percent raises) would receive 30 percent and the 65-year-old fast tracker (earning 7-percent raises) would get only 22 percent. In comparison, a final average defined benefit plan might deliver a 30-percent benefit to everyone. Simply put, cash balance plans often reward poor employee performance and penalize exceptional performance.

Not surprisingly, both working groups concluded that neither current plan design would be appropriate for the emerging work force. RJRN needed a new plan to take the company into the 21st century. So the firm established the criteria for a new plan design that would benefit all groups, instead of addressing the needs of career employees at the expense of young and mobile workers, or vice versa.

Both working groups concluded that the plan's benefits should be expressed in terms of lump sums, for portability and understandability. However, to provide sufficient retirement income, benefits needed to be tied to final average pay. And, although lump sums were desirable, RJRN wanted to avoid individual accounts with their dual administration. Also, after a three-year process to qualify its cash balance plan with the IRS (which was ultimately successful), the company had no appetite for another intricate compliance challenge. The working groups wanted a smooth curve for the benefit buildup pattern, avoiding discontinuities that distort termination and retirement patterns. They also decided that those close to retirement should be grandfathered in the prior plan benefits. Finally, the package needed to be cost neutral.

A SMILE ON EVERY FACE

The outcome of the discussions? A plan participant would accumulate lump-sum credits, as explained below, analogous to a traditional final average pay defined benefit plan. However, where the traditional plan would accumulate percentages of age-65 retirement income (for example, 2 percent of final average pay times service, payable as an annuity at age 65), the Pension Equity Plan would accumulate percentages of current lump sums (for example, 8 percent of final average pay times service, payable as a lump sum at termination). If a participant preferred an annuity, his lump sum would be divided by an appropriate annuity factor. As in a cash balance plan, either a lump sum or an annuity would be available upon termination or retirement.

This general approach to the plan design accommodated almost all of the desired characteristics of the new plan formula, combining the best traits of traditional final average and cash balance plans. All that remained was to find a formula with a benefit buildup pattern that was balanced among all groups--that is, more than the traditional final average but less than cash balance at younger ages (in cash terms), while more than cash balance but less than an equal cost traditional final average pay plan at the later ages (in terms of the annuity that can be obtained). If a formula could be designed with this benefit buildup pattern, the working groups were convinced it would have a good chance of also being cost neutral.

The solution was deceptively straightforward. Larger lump-sum credits at higher ages helped to maintain the adequacy of retirement income accrued late in a career, while lower lump-sum credits at younger ages avoided giving too large a lump sum to early leavers. RJRN's final formula is shown below.

To determine a participant's lump-sum entitlement, add up the percentages for each year of service and multiply that number by the final average pay. Say a participant retires at age 65 with 20 years of service at a final average salary of $50,000. Over a 20-year career, the employee earns lump-sum credits of 220 percent, which are applied toward the full $50,000 in the final average pay, plus an additional 55 percent in excess lump-sum credits, which are applied to the final average pay above a threshold ($20,000 in this example). All told, the lump sum equals $126,500. The lump sum could be converted into retirement income of 25 percent to 30 percent of final average pay, depending on annuity factors. This is well within competitive bounds.

RJRN's new Pension Equity Plan accomplishes the full set of stated benefit objectives. Also, since it's TABULAR DATA OMITTED translatable into traditional final average defined benefit terms for government testing purposes, compliance shouldn't be difficult. For example, although the formula may appear backloaded, it's frontloaded for testing purposes, since the cumulative annuity accrual would be the cash now divided by a deferred-to-65 annuity factor. These annuity factors are much higher at older ages and effectively remove any appearance of backloading in annuity terms.

Other important aspects of the Pension Equity Plan are these:

Grandfather Provisions--Grandfather provisions for those near retirement are straightforward. Tobacco employees who meet certain age and service criteria receive the better of the new and old formulas. Similarly, Nabisco had an existing grandfather group from a prior final average plan at the time that company converted to cash balance. This group continues its grandfather provision. Also, cash balance accounts are grandfathered for all Nabisco participants by converting them into lump-sum credits for prior service under the Pension Equity Plan.

Cost Neutrality---The plan changes are essentially cost neutral at both operating companies. The effect on costs depends on the changes in the benefits that are provided from the old plan to the new. At Nabisco, there will be a general benefit shift from slow trackers to fast trackers; in the new plan, fast and slow trackers will receive comparable benefits as a percentage of final average pay. This change has little effect on overall costs and coordinates well with RJR Nabisco's philosophy of tying total compensation to performance.

At Tobacco, benefits are significantly higher than in the prior plan at all ages up to 54. However, benefits are lower at 55 if a participant has 20 or more years of service. New plan benefits get closer again to the old plan benefit levels as a participant approaches age 65. Interestingly, although benefits are significantly higher in the new plan at, say, age 50, the correct comparison is not to the age 50 benefit in the old plan, since virtually no one left at that age. Someone who leaves at age 50 in the new plan might have left at 55 with 30 years of service in the old plan. If Tobacco's termination and retirement patterns change, the firm still can provide a higher level of benefits at most ages for the same cost as in the old plan.

Benefit Adequacy--Finally, like most large companies, RJRN also has a 401(k) plan. The benefits from the combination of plans are more than sufficient at appreciable service levels to provide retirement income in combination with Social Security that preserves the pre-retirement standard of living and more.

Mr. Angowitz is vice president of employee benefits at RJR Nabisco, Inc., and Mr. Lofgren is a senior consulting actuary with the Wyatt Company.