Exact parity with UC pension fund not in the cards, Labs management says, but improvements will be pursued
Making the changes required to achieve exact parity between the Sandia pension plan for non-represented employees and the University of California pension plan (which covers Los Alamos and Lawrence Livermore national labs employees) "creates an unacceptable risk" to the soundness of the fund and the Laboratories' financial position.
"That's our view" based on months of exhaustive analysis, Sandia Executive Vice-President Joan Woodard told employees during several special meetings scheduled to update staff on pension-related issues.
Joan's comments came after Ralph Bonner, Director of Financial Systems and Pension Fund Management Center 10300, walked an audience of more than 200 employees at the Steve Schiff Auditorium through a nearly hour-long briefing on the processes used in the pension fund analysis. In all, six briefings were conducted last week for employees in New Mexico and California.
While acknowledging that achieving "exact parity" is not in the cards, Joan asked, "Does that mean we're not going to propose to do anything" to improve the Labs pension plan? "No!"
Labs management, Joan said, "is facing some tough decisions" on exactly how to change the pension plan in a way that keeps a balance among benefit improvements, financial security of the plan and potential impact on Sandia Corporation's financial position.
The next steps, Joan said, are to:
Joan opened by acknowledging that the pension fund -- particularly its comparability to the University of California (UC) fund -- is a "hot topic" around the Labs. She noted that "the issue has been worked for a long time," adding that senior management wants to be "as forthcoming as possible" about its efforts to improve the plan.
Joan said Labs President C. Paul Robinson had established a "wish list" of some "design principles" to be applied in the review and analysis of Sandia's pension plan. Briefly, they were:
Ralph said the analysis started with a benchmarking comparison of salaried employees prepared by Hewitt Associates, a highly regarded consulting firm. The Hewitt study looked at plans from peer companies with comprehensive benefit programs, including AT&T, Eastman Kodak, IBM, Xerox, GE, Motorola Lawrence Livermore National Laboratory, Pacific Northwest National Laboratory, and Lucent Technologies. The businesses and labs in the comparison group were selected by Sandia and approved by DOE. The benchmarking is required by a DOE order that requires a company's benefit values to be no more than 5 percent above the average for its peers In the Hewitt analysis, Sandia's pension plan, not including the 401(k) match, ranked between 2nd and 3rd -- behind the LLNL plan.
The companies in the Hewitt study all offer a so-called "defined benefit" pension plan. In a defined benefit plan, the company retiree is guaranteed a benefit; the company assumes all the risk associated with delivering on the guarantee. Many companies in the "new economy" don't offer a defined benefit pension; they offer only 401(k) plans, stock purchase plans, and other alternatives to traditional retirement programs. In this approach, the employee, not the company, assumes the investment risk and the company offers no guarantees. Sandia's benefits -- including its pension benefit - were compared to many new-economy companies' benefits in the Towers Perrin High Technology BENVAL study. In that study, Sandia's pension plan -- that is, its defined pension benefit -- ranked first. Its saving plan -- the 401(k) that passes investment risk along to the employee, ranked 22nd out of 32 reviewed companies. The Labs' pension and savings plans, when weighed together, ranked 6th among the 32 companies.
Ralph went through a comparative analysis between the Sandia plan and the UC plan. The analysis team developed an "example employee," a 30-year veteran MTS earning about $80,000 at retirement, with representative increases in salary over the years and participation in the 401(k) plan at a rate of 6%, which was adequate to earn the full company matching contribution of 4% of the employee's compensation. (The assumption was that the company match money of 4% in the Savings Plan grew at an annual rate of 8 percent.) Ralph spent a few minutes explaining the Labs' logic in including the 401(k) company match as a retirement benefit and then explained how his team converted that benefit to a pension formula equivalent.
Ralph's team ran the example numbers -- including the conversion of Sandia's high five-year salary average to a high-three year salary average, to be comparable to UC's plan, and adding in the Savings Plan company match as part of the retirement benefit -- through a comparative analysis. It asked, how does this employee's retirement benefit fare if he or she had worked at a UC lab? Then, how would the same individual fare in retirement based on the combination of Sandia's pension and 401(k) match?
The results: If the individual retired at age 65, the Sandia benefit would be about 78 percent of the UC benefit. Leaving at age 60 after 30 years, the benefit would be about 75 percent of the UC benefit. In contrast, because of the UC pension formula's age factor, a Sandian retiring at age 54 with 30 years of service would get about 109 percent of the benefit of a UC employee retiring at the same age. None of this analysis includes the value of the UC's automatic cost of living increase.
Ralph went through an analysis of what would be required to achieve parity with the UC plan, noting that certain federal laws constrain the Labs' options, since it is a private employer plan, originating with AT&T and managed by Sandia Corp. For example, the Employee Retirement Income Security Act of 1974 requires Sandia to make a contribution when dictated by its enrolled actuary. As a public entity, UC is not subject to the same federal laws.
The analysis showed that as of 1/1/2000, Sandia's actuaries estimated the Retirement Income Plan had a surplus - the difference between actuarial value of its assets and its actuarial accrued liabilities - of about$927 million.
Those assets and liabilities were plugged into a sophisticated, industry-standard computer model developed by former Harvard professor Irwin Tepper. The model makes 2,500 independent runs that generate expected portfolio returns and actuarial liabilities for the next 20 years. The data generated by the model can be used to determine a range of probabilities that Sandia will have to make a contribution to the pension fund sometime during the next 20 years. (See charts this page.) The Tepper model, Ralph said, is the best available tool for determining the amount of risk the Labs would be taking on by changing from the current pension plan to a UC plan.
Based on the model's results, the Labs has about a 14 percent chance of having to make a contribution to the pension fund sometime in the next 20 years to meet obligations -- using the current Sandia pension formula. However, using the UC pension formula, the Labs would have a more than 30 percent likelihood of having to make a contribution. The risk of a contribution jumps to 53 percent if an automatic cost-of-living adjustment, or COLA, formula is included with the UC formula. ("That's why you don't often find COLAs in private pension funds," Ralph said. "They cost too much.")
The Tepper model, in sum, demonstrates a significant likelihood that adopting the UC plan in its entirety would require the Labs to make a substantial contribution to the fund in the foreseeable future.
That high risk, Joan concluded after Ralph's briefing on the analysis, is more than would be prudent for the Labs to assume.
Back to top; Back to Lab News home page.
Last modified: April 3, 2001
View Sandia news releases and fact sheets
Questions and Comments || Acknowledgment and Disclaimer