Contact: Mercer Feedback
Written by: Jonathan R. Barry
Once again, a new year-end disclosure season is fast approaching. It seems like not so long ago that pension information was far down the list of concerns for the CFO or corporate controller at this time of year. But a combination of factors, including the global push towards accounting transparency, Sarbanes-Oxley legislation in the United States, and growing global concerns about retirement issues, continue to elevate the calculation and disclosure of pension and postretirement benefit numbers to a position of great importance.
With the move to make balance sheets more reflective of actual pension deficits, we anticipate that there will be yet more focus on the accuracy of postretirement liabilities. And this scrutiny is expanding beyond year-end: Many of our European clients that are already using mark-to-market balance sheets now ask for updated pension accruals at interim reporting dates (generally half-yearly in Europe). These are now seen as important disclosures to shareholders of the way deficits (or surpluses) are developing.
The directors of the company are responsible for approving the actuarial assumptions used in the accounting calculations. But how can they be sure that the policies they set are implemented consistently in their subsidiaries around the world? For example, the rate of return on US equity investments is clearly an important assumption in the United States, but it is also likely to be an element of the assumptions used in other countries. It would be unfortunate if the assumptions used in different countries were inconsistent.
It is difficult enough for an employer with one pension plan in one country to ensure that assumptions are appropriate and that disclosures meet all necessary requirements. For a multinational employer with plans in many countries, the task can be daunting, and the questions may seem endless. For instance, is the US discount rate comparable to the one used in the United Kingdom? Are Japanese mortality rates appropriate versus those used in Germany? What about the return on assets assumption in the Netherlands? How can the person responsible for these numbers rest easy knowing for certain that the pension financials are accurate and consistent?
This article takes a look at some of the trends we have seen over the past few years and gives some suggestions as to how to tame the maelstrom surrounding multinational pension disclosure.
Over the past few years, the drive towards accounting transparency has led to more disclosure about the assumptions used to value plan liabilities and more immediate recognition of the difference between plan assets and liabilities for pensions and other postretirement benefit plans. Immediate balance sheet recognition of pension deficits is a requirement under FRS 17, optional under IAS 19, and will be required at the end of the year under FAS 87.
Immediate recognition means that the effect of an actuarial gain or loss on the company balance sheets is often 10 to 15 times greater than it has been using the traditional approach. If the pension liability is large relative to the corporate balance sheets (as it is with many multinationals), this volatility can have significant ramifications, including reductions to shareholder equity, violation of debt covenants and changes to incentive plan payouts.
This increased volatility has led to greater investor and auditor scrutiny of pension financials and a need for corporate finance to better understand, manage and mitigate pension risks where possible.
In practice, arriving at pension liabilities has been a compromise, juggling accuracy, cost and reporting deadlines. There is often no “right answer” to what assumptions or methodologies to use and experts will vary in their good-faith best estimates. As such, we often see ranges of assumptions in practice as well as in interpretations of the accounting methodologies that apply to certain types of plan provisions. In addition, it can be difficult to ensure consistency in assumptions and methodologies between countries. The key to getting your arms around disclosure calculations is to first develop a well-documented process of what and how assumptions and methodologies will be used, and then to stick to it.
Over the past two years, we have seen more and more companies establish a well-defined disclosure process document that details all the pertinent assumptions and methodologies that will be included in the calculation of year-end pension and postretirement benefit obligations. These documents come in all shapes and sizes, but at a minimum, they will describe and justify:
The document should also include a brief description of how the assumption is developed, the actual assumption selected and commentary on any deviation from the standard policy. In addition, the process document can be expanded to include other pertinent disclosure data, such as:
Does the process need to be the same in each country? Not necessarily, and in some cases, that may not even be possible. For example, consider the choice of discount rate. A process gaining popularity in the United States is to match expected cash flows of a pension plan against a yield curve of high-quality corporate bond rates. This process follows the spirit of FAS 87 and SEC guidelines, is generally accepted by auditors and, in fact, is often deemed preferable to simply looking at an index. So why don’t we just do that in every country? Unfortunately, not every country has a well-developed fixed-income market; for example, practical yield curves of high-quality corporate bonds cannot be constructed in countries such as Korea or Mexico. However, Mercer has already developed or is in the process of developing yield curves for several geographies, including the United Kingdom, the European Union, Canada and Japan, so that a consistent process of discount rate selection can be constructed widely. But even in these cases, it should be noted that the method of developing a yield curve in each country cannot be completely uniform because of country-specific differences in the depth of the bond market and other factors.
One particular assumption that has been garnering more scrutiny than others over the past few years is life expectancy. Life expectancy continues to increase in most developed nations due to improved access to medical care, prescription drug development and breakthroughs in disease treatment. The general consensus is that continued improvements in these areas will carry over to expanded life expectancies into the future, but there is much less agreement as to how much and how quickly.
Pension liabilities in most countries have traditionally been valued with “static” mortality tables that either did not incorporate future life expectancy improvements or included some improvements indirectly through the use of margins. This trend has begun to change. Many companies have adopted more up-to-date mortality tables – some with variations for the type of workforce covered – and have either explicitly projected improvements to a specific future year or have introduced “generational” mortality improvements. A generational mortality table will assume, for instance, that a person who turns 65 in 2026 will have a longer life expectancy than someone who turns 65 today.
So why not use this type of assumption for every country where a company sponsors pension plans? Again, much like the discount rate, it may be desirable, but not practical. Information on potential generational improvements is very subjective and may not have been well-developed in some countries. Also, from a technical perspective, generational mortality can be difficult to calculate in some cases and may not be worth the effort if a plan’s pension liabilities are small relative to the total company obligation.
In response to the need for greater accuracy when disclosing pension liabilities, actuaries continue to develop more sophisticated tools for establishing assumptions, and these have been generally well-received by clients and auditors. However, the key to successful and supportable disclosure lies in a well-defined process document. A plan sponsor should decide on a policy, learn to justify and adhere to any resulting inconsistencies, and then review the process each year to look for areas to improve upon.