LEGAL AND INVESTING IMPLICATIONS OF LDI SAFEGUARDS FOR PENSION RISK
Among the most significant of the damages to result from the recent market turmoil may be the losses suffered by US pension plans. In 2008, DB (defined benefit) plans plummeted from fully funded or even over-funded status to historically large deficit levels. Pension health, as measured by the ratio of assets to liabilities, dropped precipitously during the same period, from 104% to 75%, leaving a $400 billion funding gap (Mercer 2009). Many of these now underfunded plans are sponsored by companies facing drastic deterioration in financial health or, in some cases, potential bankruptcy. The PBGC (Pension Benefit Guaranty Corporation), the agency responsible for insuring DB plans, is itself experiencing significant and rising deficits (Elliott 2009). The result may be escalating uncertainty as to the status and future of benefits for a large number of pension plan participants.
Could the recent losses and volatility of DB plans in the US have been avoided or, at the least, mitigated? A review of industry practices reveals that investment strategies called LDI (liability-driven investment), which are based on risk hedging and which are growing in prevalence within institutional asset management practice, do offer DB plans a means of reducing the forms of risk realized during the past year. What’s more, applicable law suggests it is possible that fiduciary liability could arise from a failure to adopt such safeguards, particularly in the case of a decision-making process that is found to have been inadequate.
DURATION MISMATCH AND REDUCTION OF RISK
Traditionally, the overall investment performance of a DB plan has been measured through the use of “asset-based” benchmarks—benchmarks based on the performance of comparable assets in the market. Benchmarks are generally comprised of indexes such as the S&P 500 for equities and the Lehman Aggregate Bond Index for fixed income. Performance is measured according to the respective returns and volatilities of the plan’s asset classes relative to each of their appropriate asset benchmarks. However, as demonstrated in the recent market downturn, a plan’s liabilities can also experience significant volatility, and often in a direction different than the plan’s assets. In risk terms, this deviation is referred to as “tracking error to liabilities,” “surplus volatility,” or volatility in a plan’s “funded status” (the last of these refers to the fair value of plan assets minus the present value of projected pension obligations, based on actuarial assumptions). Such volatility is largely driven by changes in interest rates, which result in changes in the present value of plan liabilities and of asset values of plan-held fixed income investments. This volatility often occurs in connection with increased volatility in other asset types, such as equities.
A plan’s liabilities may be thought of as having a risk profile similar to a fixed income investment, but in reverse (or from the viewpoint of the issuer). At different times, various payment requirements are expected, all of which are valued on a present value basis. Similar to bonds, the interest rate sensitivity of liabilities can be measured in large part by their duration. In this context, duration can be thought of as the present-value-weighted average of the deadlines for benefit payments. Studies indicate the portfolio duration of liabilities for a typical DB plan to be between 12 and 14 years, whereas the duration of the DB plan’s assets (based on the fixed income portion of the portfolio) is generally much shorter, 4.5 to 5 years (Moore 2007).
The resulting “duration mismatch” may cause multiple risks to a plan. A decrease in interest rates will cause an increase in the plan’s liabilities in excess of the gain from its assets, creating investment losses and added pension expense. Additionally, an increase in the plan’s liquidity risk may be argued to arise from decreased certainty that the plan’s cash inflows (from the assets with shorter durations) will match those of the plan’s liabilities (the benefit payments that are due), in terms of the amounts and timing.
Although ERISA prudence may be reviewed by attorneys and in the context of case-law precedent, it is ultimately determined by the rapidly evolving practices of the investment management industry and its experts.
LDI strategies are gaining recognition within the asset management community for their ability to cost-effectively reduce these risks (Appell 2009): first by building a benchmark based on timing and on the amount of plan liabilities (rather than on the targeted return of assets); and then by constructing a fixed income portfolio designed to track the benchmark. This entails “extending” the duration of the plan’s fixed income assets to more closely match the much longer duration of the liabilities. In order to obtain maximum risk reduction benefits, a combination of leverage and derivatives such as interest rate swaps may also be used. Furthermore, in order to generate sufficiently high returns, a remaining portion of the portfolio is allocated to riskier assets such as equities, but in a much smaller portion than the traditional allocation.
Adoption, or even partial implementation, of such a strategy can result in significant reduction in the risk associated with a DB plan’s funded status. (The relative merits in the timing and degree of implementing an LDI strategy are beyond the scope of this article. For a recent comparative study of traditional and LDI portfolio performance, see Watson Wyatt 2009.) The reduction in risk is largely due to the fact that investments are structured to provide returns and income that match the timing and amounts of the plan’s liabilities, so that the plan is essentially “hedged.”
It should be noted that the enterprise-wide risks of a plan sponsor, which impact its solvency or its ability to provide plan funding, may also be managed under an LDI strategy by reducing the correlation between firm operational performance and plan performance (Peskin and Hueffmeier 2008). Enterprise-wide risks comprise a number of factors, including firm leverage, cyclicality, and exposure to foreign currencies and other capital markets.
Although an LDI strategy may involve the use of derivatives, such usage is not required and may be limited in accordance with the needs, expertise, and risk tolerances of a plan. For example, interest rate derivative positions often result in interim cash flow requirements to the portfolio, which fluctuate with changes in interest rates. Such impacts could, and should, be modeled under different interest rate scenarios (see the US Department of Labor’s letter to Eugene A. Ludwig, 1996). In addition, plan fiduciaries lacking expertise may require outside advice to understand how derivative investments operate and would impact the plan (Mangiero 2008).
THE LIABILITY SCENARIOA hypothetical situation can illustrate how a plan might be faced with the type of investment decision-making issues at stake here, and can highlight several of the relevant fiduciary liability issues.
You sit on the investment committee for your company’s multibillion-dollar defined benefit pension plan. It is spring of 2007 and the plan is “fully funded” (pursuant to the funding requirements of ERISA, the Employee Retirement Income Security Act of 1974) and in excellent financial health. With a portfolio allocation of 65% equities, the plan’s strong investment performance can be attributed in large part to that of the S&P 500, which has earned double-digit annualized returns for the previous five years. Despite the success of the current portfolio’s performance, the committee is being presented with a portfolio allocation strategy called “liability-driven investing,” which would represent a radical change.
In a presentation to the committee, the new investment advisor candidates argue that the plan funding level is exposed to a series of substantial risks. First, the plan is exposed to significant asset-based volatility associated with the plan’s equities and other “risky” assets. Another primary risk arises from the volatility in interest rates, which causes the present value of the plan’s liabilities to fluctuate. The candidates explain that these risks could be greatly reduced, if not largely eliminated, through the use of a “hedging” strategy.
A new performance benchmark for the plan would be created, based not on investment indexes such as the S&P 500 or the Lehman Aggregate Bond Index, but on the plan’s liabilities. This would be accomplished by constructing a model of the timing and the amount of the plan’s forecasted future benefit payments.
The investment advisor would build a portfolio of fixed income investments, with incoming interest and cash flows designed to “match” the anticipated outgoing benefit payments of the plan. Portfolio performance would be measured against the new liability-based benchmark.
A portion of the plan assets would remain in equities and other asset classes intended to yield higher or excess returns, but this percentage would be far less than the plan’s traditional allocation. The end result would be a portfolio with overall lower returns than the current plan, which would be likely to exhibit less volatility as measured against the plan’s liabilities.
Aspects of this proposal appeal to the committee, but you ultimately reject it, at least for the time being. A primary concern is the strategy’s complexity. You and the other committee members are uncertain that you possess the expertise necessary to approve and monitor the strategy implementation. There is also discussion of the potential use of interest rate swaps in the fixed income portion of the portfolio. The investment advisor candidates have recommended derivative usage, with the caveat that it would not be strictly necessary; and a number of committee members are uncomfortable with derivatives generally (again, largely due to a lack of committee expertise and understanding of the risks involved).One other concern has weighed on you. In order to implement this strategy, you would be required, pursuant to financial accounting standards, to make a significant downward adjustment to your expected return assumption for the plan’s investments. You are certain the immediate negative impact on company reported earnings would please no one in upper management.
Two years later, in spring of 2009, after an unexpected credit crunch and subsequent market downturn, your company’s defined benefit plan is now seriously underfunded. Not only has your plan portfolio lost in excess of 50% of its value, but plan liabilities, calculated on a present value basis, have actually risen (in large part due to current interest rates at historically low levels). There is now significant uncertainty that the plan or the company has the ability to pay promised pension benefits when due. There is even informal discussion about termination of the plan. You ask yourself whether the committee was too hasty when it rejected the LDI approach two years ago. Of course, hindsight is always 20–20, and your plan has fared no worse than a large number of other US pension plans. Notwithstanding, you call your attorney and ask, “As pension plan ‘fiduciaries’ under ERISA, could the investment committee members be liable for failing to adequately consider a hedging strategy before rejecting it?”
LEGAL RESPONSIBILITIES AND ERISA
The scenario above raises a legal issue: can a pension plan fiduciary incur liability simply by maintaining a “tried and true” portfolio strategy? Surprisingly, applicable law suggests that the answer is yes, largely because of one of the most stringent but fluid legal requirements: the prudent investment process.
Like the CFA Institute Standards of Professional Conduct, ERISA establishes duties of care and loyalty governing all plan fiduciary conduct, including investment strategy decision making. Under the ERISA “prudent man” standard of care, fiduciaries are required to act “with the care, skill, prudence, and diligence under the circumstances then prevailing that a prudent man acting in a like capacity and familiar with such matters would use in the conduct of an enterprise of a like character and with like aims” (ERISA § 404(a)(1)(B), 29 USC § 1104(a)(1)(B)).
What is perhaps most legally significant about “prudence” is its emphasis on process. Fiduciaries exercise great freedom in their investment decisions, but they must do so prudently. Accordingly, in reviewing an investment or course of action, a court disregards the outcome yielded and focuses only on the procedure employed. Under DOL (Department of Labor) regulations, the meaning of prudent process is further refined. An “investment” or “investment course of action” is deemed “prudent” if “the fiduciary ... has given appropriate consideration to those facts and circumstances that … the fiduciary knows or should know are relevant … and … has acted accordingly” (emphasis added) (US DOL Reg. 2550.404a-1).
The regulations define “appropriate consideration,” regarding an investment or investment course of action, as including the following two components:
- A determination that it is reasonably designed, as part of the portfolio, to further the purposes of the plan; and
- Consideration of factors (as they relate to the given portion of the portfolio), including:
- Portfolio diversification,
- Portfolio liquidity and current return rates relative to the plan’s anticipated cash flow requirements, and
- Portfolio projected return relative to plan funding objectives (US DOL Reg. 2550.404a-1).
The directive that an investment or investment course of action be “reasonably designed, as part of the portfolio,” has led courts to view this as the DOL’s adoption of modern portfolio theory. Notably, the qualitative and quantitative factors to be considered under the regulations include plan liability cash flows and plan funding requirements, which are critical components of an LDI strategy. It should be noted that the regulation is intended as a “safe harbor,” rather than as the exclusive manner of compliance with prudence requirements. Regardless, courts have looked to its provisions for guidance in determining the prudence of investment actions.
Applicable law generally holds that the prudent investment process is reviewed from the standpoint of experts and in light of available industry standards. Courts hold that the ERISA prudent person standard is applied from the perspective of “a prudent fiduciary with experience dealing with a similar enterprise” (US v. Mason Tenders District Council of Greater New York, 909 F. Supp. 882, 886 (SDNY 1995)). Fiduciaries who lack expertise in a given area have been held to have an affirmative duty to seek out such expertise. Courts also generally look at prevailing industry methods and standards when evaluating whether a fiduciary’s methods and process satisfy ERISA prudence requirements (Ironworkers v. Loomis Sayles & Co., 259 F.3d 1036, 1044) (in analyzing “appropriate consideration” under DOL regulations, the court examined the industry prevalence of option-adjusted spread analysis versus the Bloomberg system). This approach essentially means that, although ERISA prudence may be reviewed by attorneys and in the context of case-law precedent, it is ultimately determined by the rapidly evolving practices of the investment management industry and its experts.
Accordingly, to the extent that a court finds that prevailing industry standards include an understanding of risk related to pension liability (interest rate risk, portfolio tracking error, and surplus volatility) and a knowledge of risk-hedging solutions (LDI and liability-based benchmarking), that court could hold that a wholesale rejection of these concepts constitutes an imprudent investment decision.
In addition to the prudence requirements and duty of care, the fiduciary investment decision-making process is subject to an ERISA duty of loyalty, which requires that a fiduciary act be “solely in the interest” of plan participants, for the “exclusive purpose” of providing benefits and “defraying reasonable expenses” of plan administration (ERISA § 404(a)(1)(A), 29 USC § 1104(a)(1)(A)). This means that when fiduciaries consider DB plan investment strategies, they may not take into account factors other than those involved in the provision of benefits under the plan. Such impermissible factors would include the impact of certain investment strategies on the financial statement reporting of the plan sponsor.
DB plans enable the crucial transfer and pooling of the burden of retirement income risk from many individual employees to their employer. This past year has demonstrated that, notwithstanding this transfer of risk, enormous risk exposure is still present within our defined benefit pension plan system, and the resulting volatility will most certainly impact retirement benefits. Whether this exposure is in fact unnecessary is a current question.
Plan fiduciaries may wish to consider carefully and thoroughly viable solutions for reducing DB plan risk that are readily available and gaining traction within the investment management industry. As to whether and in what contexts fiduciaries could be held liable for failing to do so, the issue remains uncertain. However, “consideration” lies at the heart of prudence under ERISA. Applicable law makes clear that consideration must be informed by relevant expertise and knowledge of industry practices, and not influenced by factors other than those involved in the provision of plan benefits. As the risks inherent in DB plan funding continue to be realized, and as the pension investment industry continues to develop solutions to those risks, it is reasonable to expect that appropriate consideration of those solutions will become a required component of the prudent investment process.
Appell, Douglas. January 2009. “Market Nightmare Could End in Dream Ending for LDI.” Pensions & Investments.Elliott, Douglas J. May 20, 2009. “A Guide to the Pension Benefit Guaranty Corporation.” Brookings Institution [Initiative on Business and Public Policy at Brookings] and Center on Federal Financial Institutions.
Mangiero, Susan. October 2008. “Pension Risk Management: Derivatives, Fiduciary Duty, and Process.” Society of Actuaries, and Pension Governance, LLC [Joint Pension Risk Research Project]. [The most frequent responses to the question “What reasons account for your decision not to use derivative instruments to manage the risk of your defined benefit plan(s)?” were: “lack of fiduciary understanding,” “perception of excess risk,” and “[DB] plan risk not considered significant.” In response to the question “What defined benefit plan risk areas concern you?” the number one risk concern among DB plan sponsors was “accounting impact.”]
Mercer. January 7, 2009. “Pension Plan Deficit Hits Record $409 Billion for S&P 1500 Companies; Pension Expense May Rise.”
Moore, James. May 2007. “Jim Moore Discusses Liability-Driven Investment Strategies and Concepts.” PIMCO.
Muralidhar, Arun S. 2001. Innovations in Pension Fund Management. Stanford, CA. Stanford University Press.
Peskin, Michael, and Chad Hueffmeier. August 2008. “Asset-Liability Management within a Corporate Finance Framework.” Morgan Stanley.
Waring, M. Barton, and Laurence B. Siegel. January–February 2007. “Don’t Kill the Golden Goose! Saving Pension Plans.” Financial Analysts Journal, vol. 63, no. 1. 31–45.
Watson Wyatt. May 31, 2009. “Funded Status Watch Report.”
Zion, David. 2003. “The Magic of Pension Accounting, Parts I to IV.” Credit Suisse.
ERISA § 404, Title 29 USC § 1104. August 17, 2006 [last revised].
US Department of Labor. March 21, 1996. Letter to the Honorable Eugene A. Ludwig, signed by Olena Berg.
US Department of Labor Regulation 2550.404a-1. CFR, Title 29, Chapter XXV, Part 2550.404a-1. July 1, 2005 [last revised].
Martin J. Rosenburgh, Esq., is an independent legal advisor specializing in securities and ERISA compliance, and a CFA Level II candidate. Andrew C. Spieler, PhD, CFA, FRM, is associate professor and director of the quantitative finance program at Hofstra University’s Frank G. Zarb School of Business, and chair of the New York Society of Security Analysts Derivatives Committee.
This text is based on the authors’ article “Twenty-First Century Pensions: The Risk, the Hedge, and the Duty to Consider,” which appeared in the Journal of International Business and Law (Spring 2009, vol. 8, no. 1), and which presents a more detailed analysis of the legal issues. The authors thank Michael Peskin, Jim Moore, Ron Ryan, Susan Mangiero, and participants at the New York Society of Security Analysts conference “Pensions at Risk” for their thoughtful contributions, and take sole responsibility for any errors in interpretation.