What’s the Debate About?
What's The Debate About?
Should public pension plans disclose a market value of liability? Are current disclosures adequate? Turn the page for an insightful discussion.
Moderated by Andy Peterson
Over the past few years, the topic of how to measure public pension plan liabilities appropriately has been discussed widely by actuaries and others working with public pension plans. The crux of the debate is whether public pension plans should disclose a market value of liability (MVL) or whether current disclosures are adequate. While actuaries in this arena are keenly aware of the debate, actuaries not working with public plans–even those in the pension arena–are likely not aware of the discussion.
The Actuary invited several professionals with a keen interest in this debate to help the SOA's membership on this issue. The purpose of this roundtable is to provide an overview of the issue and discuss some of the implications for public plans and stakeholders in the actuarial community. Further academic research on this topic is needed and is currently underway through a broad Public Pension Plan Finance Call for Papers issued by the Pension Section Council in early August 2008.
The professionals we asked to participate in the roundtable include: Keith Brainard, the research director of the National Association of State Retirement Administrators (NASRA); Richard Ennis, an investment consultant, the Chairman of the Board of Directors for Ennis Knupp + Associates and an editor for the Financial Analysts Journal; David Kausch, FSA, EA, MAAA, MSPA, an actuary with Gabriel, Roeder, Smith & Company that consults with public pension plans; and Bob North, FSA, EA, MAAA, FCA, FSPA, the chief actuary of the New York City Office of The Actuary.
Andy Peterson, FSA, EA, MAAA, FCA, SOA staff fellow, Retirement Systems, moderated the discussion, held Aug. 7, 2008.
The opinions and views expressed by the participants of this discussion are solely those of the individual participants and not those of the participants' employers.
ANDY: As we start the discussion today, please provide a brief summary of your professional background and why you're interested in this public pension plan issue.
KEITH: My background is in public administration and public policy. I provided fiscal analysis to two state legislatures and served as manager of budget and planning for a large statewide retirement system. I currently serve as research director for NASRA, whose members are the directors of 82 statewide public retirement systems in the United States. Due to my position and to my interest in issues pertaining to statewide public retirement systems, I have a keen interest in the ultimate outcome of this debate.
RICHARD: I am an investment advisor to institutional investors of all types. Investment policy analysis for pension plans requires integrating asset value, funding and liability values. Finance models, such as so-called pension asset-liability models, require that liabilities as well as assets be carried at their fair market value rather than arbitrarily established values. The whole field of finance revolves around fair market values, so my interest is seeing the public pension system collectively work toward more relevant investment policy solutions.
DAVID: I'm a fellow of the SOA, and I'm currently practicing as a public sector pension actuary. I spent nine years as a private sector pension actuary where I witnessed the increased use of market value of liability. I also spent three years teaching actuarial math and financial economics at the university level before I landed in my current position. I think financial economics and market value of liability add some very interesting perspectives. They are often counterintuitive to actuaries and others, and I feel this is a subject that needs very careful introduction to the public sector where it is extremely foreign.
BOB: I've been the chief actuary for the New York City retirement systems for 18 years. Before that I worked in the private sector as a consultant overseeing asset liability forecasts and also had a stint in a merchant banking firm helping to buy and sell companies. My interest in this developed in the late 1990s as the topic of financial economics began to be applied to pension plans–I followed the debates and found it very interesting. Since 2003, I have prepared and included in the Comprehensive Annual Financial Reports (CAFRs) for the New York City Retirement Systems market values of liabilities and their comparisons to market values of assets. I started doing this because I think there are limitations in the existing disclosures of information by actuaries. I'm interested in seeing this topic pursued further because I think it can add value to the discussions that everybody has around pension finance.
ANDY: Keith, given your background with NASRA, can you provide an overview of the public pension plan arena so that our readers have an understanding of the scope of these systems, particularly the number of participants covered, size of assets, etc.KEITH: State and local governments employ 12 percent of the nation's workforce, and the public pension community includes more than 2,000 retirement systems providing pension and other benefits for some 15 million working and seven million retired public employees who include school teachers, firefighters, police officers, nurses and other occupations. Public retirement systems have assets of around $3 trillion and distribute roughly $160 billion annually in pension benefits. It's worth noting that although there are more than 2,000 public retirement systems in the United States, the largest 75 of these account for some 80 percent of all of the community's assets and participants. I'd also like to point out that as a group, the public pension community is in reasonably good funding condition, a fact verified by recent studies published by the United States Government Accountability Office and the Pew Center for the States.
ANDY: For those not familiar with the discussions, what is meant when we talk about the market value of liability disclosures, and how does this vary from current public plan disclosures? Also, please address how this differs from the current calculations of the annual required contribution (ARC)?
BOB: The market value of liability refers to the present value of benefits earned to date and discounted at a rate consistent with the likelihood of payment. For public pension plans that means discounting the benefits earned to date that are expected to be paid in the future, generally at a risk-free interest rate because in the public sector there's almost certainty that the benefits will be paid either through the constitutional protection rights of the various states or contractual guarantees as well as the assets that are already set aside. It's not a termination liability, it's an economic value. It's intended to provide a consistent number across all plans. To date, I've personally only used it for disclosing the value of benefits earned, not to calculate annual contributions. There may still be great value in traditional actuarial practice and using traditional methods to develop funding schemes and budgeting targets.
DAVID: Bob described the way market value of liability is calculated and that is a different liability, as he said, from what is currently required for the annual required contribution and current required disclosures. In the current disclosures, the liability that is calculated is referred to as actuarial accrued liability and that is determined in a different way. The main difference is rather than using risk-free rates to discount the payments, the long-term expected return on assets is used to discount and this can be a materially different result. Secondly, market value of liability doesn't take into account future salary growth since it's just looking at accrued benefits, but future salary growth does affect both the ultimate benefits paid and the financing of those benefits–that is taken into account with actuarial accrued liability. The last difference is that the actuarial accrued liability can be determined under several different cost methods that provide flexibility in spreading the costs over time. Market value of liability allocates cost over time as basically the present value of benefits as they accrue, as Bob said. The common alternate that's used frequently in the public sector is the entry age normal cost method that allocates contribution rates as a level percentage of payroll over time. This last difference is actually important to the discussion since the common use of actuarial services and disclosures in the public sector focuses on budgeting contributions. The current accounting standards require disclosure of just the actuarial accrued liability as well as the history of both the actual contributions made by the government entity and a comparison to the annual required contribution as determined by the actuary. The theory is that accountability in the public sector is just as important as accounting. Market value of liability is not required for disclosures at this point and to my knowledge, it's not used by any retirement system for determining the annual required contribution. I think New York City might be one of the very few, if not the only system, that actually voluntarily discloses it at this point.
ANDY: That is a good segue into the next question I'd like to ask Keith and Richard. What do you feel is the key purpose of government disclosure in funding requirements?
KEITH: The key purpose for disclosure is to inform stakeholders about the plan; its stewardship; required contributions by participants, plan sponsors and other contributors; and the plan's compliance with finance-related statutory, regulatory and contractual provisions. The key purpose for funding requirements is to provide sufficient funding to permit the plan to pay promised benefits when due with as much predictability and as little volatility as possible.
RICHARD: One purpose of public disclosure is ensuring that all the stakeholders–and I would include taxpayers, elected officials, system members and bond holders–have a valid understanding of what is owed to plan participants. Currently they receive extensive actuarial information in the CAFRs and elsewhere that include any number of different calculations of pension liability, but they do not have access, except for the New York City systems, to the single figure of how much is owed. Disclosure is, I think, the primary issue here and advocates of disclosing market value of liability to my knowledge rarely find themselves becoming deeply involved in the computations that are made for funding purposes. It's important to recognize that the process for coming up with a budget of contributions for a particular plan is a different one from the disclosure of what is owed.
ANDY: You both made reference to stakeholders and listed a few of them. Let's pursue that a bit further. Who are the key stakeholders and how do they use these calculations? I'm thinking not just in the context of the market value of liability, but in this broad sense of government disclosure.
BOB: Being public plans, the stakeholders are far and wide, including employees, employers and taxpayers. There are fiscal monitors and there are policy makers at the legislative level. Insurance departments regulate certain state funds and there are bond holders. All are entitled to the most useful information that can be provided. For example, with respect to trustees, employers and decision makers such as the legislature, I think additional information and understanding of the risks inherent in their decisions are important. In order to get to the risk, one has to begin with the fundamental value of what is owed. I mention fiscal monitors who often criticize public plans because they're not sure what they're looking at in terms of the values and the amounts that have been committed by taxpayers to the systems. I believe market value of liability not only provides that information, but over time, allows people to better appreciate the risks that are involved when one takes the benefits of the equity risk premium in advance in the funding process and leaves future generations of taxpayers to bear the risk.
DAVID: One stakeholder that I didn't hear on Bob's list is the retirement system itself. These systems are separate legal entities and typically have a board of elected or appointed trustees. The board has a duty to see that the benefits that are promised are secured and the actuary will calculate the annual required contribution for the system. The system has the responsibility of collecting this amount from the government entity that promised the benefit. Now, from a financial-economics perspective, the retirement system and its board are often described as a pass-through entity between the taxpayers and the beneficiaries, when you're looking at it as a financial transaction. Unfortunately, this terminology really undermines the case for financial economics because it carries the connotation that the boards and retirement systems themselves are irrelevant. The fact is that the retirement systems are the ones that are responsible for managing trillions of dollars in assets and are the ones that currently hold the governments accountable for contributions with the help of the accounting standards and they are the primary users of the actuarial information we provide. In my opinion, if you want someone, such as these boards, to embrace a new paradigm, you need to convince them how it will help them, not tell them that they're irrelevant to the new paradigm. As actuaries, I feel we have a lot of work to do in communicating the pros and cons of financial economics to those outside of the profession and this needs to happen before we can address the question of possible new disclosures.
KEITH: Another stakeholder group that hasn't been mentioned is those who rely on public services, which essentially is all of us. Users of disclosures, not just those who rely on public services, but all other stakeholders, have expectations that these disclosures fairly reflect the plan's operating environment, which is something that, in my view, market value of liabilities does not fairly reflect.
RICHARD: I'd like to focus, just to keep it interesting, on a different set of stakeholders, namely bond holders of the public debt of cities and states and counties. Chairman Christopher Cox of the Securities and Exchange Commission has indicated and reiterated to me recently a high priority is cleaning up the muni market–referring to all elements of the public-issuer bond market. The muni market is regulated by the SEC. For a number of reasons I won't get into, that market is not viewed as being anywhere near as well run or as transparent as the public markets are for private issuers of securities. So the SEC's goal is to, in Cox's terms, clean up the muni market. An important part of what he is referring to is disclosure and reporting and the integrity of reporting of public entities. That's not to single out pensions as an issue, but overseers such as the SEC and other accounting overseers are interested in improving the lot of one set of stakeholders, namely the users of financial statements who are considering purchasing public securities.
ANDY: Some of you have already addressed this question a bit in your previous responses, but what are the key reasons why you believe that market value of liability calculations are or are not valuable to stakeholders in the public pension system?
KEITH: Market value of liability calculations don't reflect the public pension operating and legal environments and could cause confusion about a plan's true funding condition. Market value of liabilities is based on the theoretical and unlikely event of a plan termination, a scenario few, if any, public pension plans are likely to experience. Unlike corporations which can, and do, go out of business and be acquired, governmental entities like cities and states are going concerns. In addition, pension benefits generally are guaranteed by state constitutions and statutes, and in many cases plan participants are legally entitled to continue to accrue benefits once they have begun. Requiring public pensions to report a liability on the basis of their termination is inconsistent with their operating and legal environment. Mandating that public pensions disclose an MVL could mislead stakeholders and result in confusion about a plan's funding condition. It could also permit policy makers and others to pick and choose the measure that best suits their purpose. MVL also could lead to required costs that are higher than necessary and to needlessly limit asset allocations and reduce investment returns. Investment earnings make up nearly two-thirds of public pension revenues, a proportion that would be lower if public pensions were required to invest on the basis of principles that drive MVL disclosure. The imposition on corporate pensions of MVL type disclosures has resulted in significant volatility of corporate plans' funding levels and required costs. This volatility is in sharp contrast to the very steady and predictable changes in funding levels and costs experienced by public pensions. The uncertainty and volatility experienced by corporate pensions have been a major factor in their near-demise, and I do not wish for that fate to visit public pensions as the result of an accounting or actuarial criterion that is inconsistent with the nature of the public pension operating environment.
RICHARD: As far as the participants are concerned, it would answer their question, "Is my pension plan fully funded?" As far as taxpayers are concerned, it would help them answer the question, "How much do we owe for pensions that have been earned?" For elected officials (legislators and governors) who are the ones who in fact make decisions of determinations about benefit levels and funding, it would help them understand whether the pension plan is fully funded by an objective standard; i.e., can we afford benefit increases, should we increase appropriations, etc. Bond holders and others also want to have an answer to the question, "How much is owed for pensions?"
BOB: All stakeholders, I believe, benefit by having additional knowledge from which they, if they are in decision-making positions, can make decisions and others, if they're looking for information, just have better information. To give an example, some of the users are the unions and the participants of the plans. I have observed that as stakeholders and individuals, they already mostly understand the MVL concept. They recognize the value of what they have earned to date and that there is a contingent value for what they will earn in the future if they remain with the system. They also understand that their benefits are worth more, the lower the interest rate in the economy. Lower interest rates mean they can't earn as much on their own investments and so it's harder to retire, and they depend more on their pension. Having this value available to all the stakeholders provides information that is useful. For many people, such as my example of the typical employee, they already understand in a very broad way that MVL is an economically sound representation of value. I would note that I disagree with the comment that it is a termination liability. It is not intended to be that. It is intended to be a value of benefits earned to date. To date, I'm not sure that many proponents of its disclosure would mandate that it be used for all purposes. There are certain concerns about the implications of MVL disclosure, but I think that such disclosure just leads to a richer discussion about goals and objectives such as what is the best investment policy or is the funding policy accomplishing what it should. Absent that basic measure of value, everything is wound up in the actuarial complexity and there is no comparability among systems.
DAVID: I think the market value of liability may end up being a very useful tool in the toolbox for public sector plans, but I would actually venture to say that it's not particularly well understood at the board level. For example, for corporate pension plans, market value of liabilities is used frequently in liability-driven investing. Liability-driven investing essentially is considering not just the asset side of investing, but an asset/liability context, and it considers the liability as part of the portfolio as a negative asset, if you will. Market value of liability works very well in this context for liability-driven investing but this is, I think, a foreign concept for these retirement boards. These retirement boards spend an enormous amount of time analyzing investments from an asset-only perspective and they have hired professionals who help them with a lot of that information. Personally I think one of the shortcomings in market value of liabilities is ignoring salaries and inflation. In an asset liability context, you're talking about hedging, not just accounting. Hedging provides many more opportunities when you are looking at all the economic variables, such as salaries and inflation. But I think this part of the discussion is about the possible uses of market value versus the required disclosure of market value, and those are two very different questions. I think as actuaries we have a duty to get these people up to speed. For proponents of market value of liability, the case still needs to be made that this has utility to their retirement boards for whatever decisions they need to make before we can insist on requiring these as disclosures.
ANDY: How do you believe a disclosure of market value of liability would affect the key stakeholders we've been talking about and would it be a positive impact or a negative impact?
RICHARD: In the short run, it's going to have the effect of revealing that the value of pension liabilities in the sense of how much is owed is a much bigger number. It would have the effect of increasing unfunded pension liabilities and that is going to be an important point of readjustment for many, many plans. In the long run, it should contribute to the stability and viability of the system if all the parties realize they're using a common objective way of valuing pension liabilities. I think a lot of the stress and strain and pressure that these systems are feeling will diminish. I would disagree with the earlier comment that the systems cannot be terminated or that they're perpetual. A great deal of the angst in the public pension sector is over the prospect of their termination. So I think these systems need to move more toward being quasi-autonomous financial institutions, institutions that value their assets and liabilities at market and operate in the black.
KEITH: As I mentioned previously, I think that MVL disclosure will provoke confusion and open the door to those with ulterior motives to pick and choose a measure of their liking. I'm also concerned that it could needlessly lead to the abandonment of pension benefits. Richard commented he thought the application of MVL would in the short term lead to lower funding levels, which he believes is a more realistic reflection of funding levels, but they would be lower based on the use of a lower investment return assumption. The investment return assumptions used by most public pension plans are based on reasonable expectations of future returns that also are consistent with the historic returns that public pension plans have experienced. While there may indeed be a lower figure as a result of applying a risk-free rate of investment return, that does not necessarily produce a more realistic reading of the plan's funding condition.
DAVID: Thinking about Richard's first point, I think he's right on. This change would show higher liabilities and higher unfundeds, but if you think about it, we live in an era of sound bites, and we also live in an era where it's very popular to blame the government for just about everything. So, imagine a scenario where a government entity discloses a funded status of 85 percent based on current practices and 65 percent with a market value of liability disclosure. And imagine the great length that this government could go through to describe the difference in methodology in what the two numbers represent. What's going to end up in the headline? The headline is going to focus on the 65 percent and it's probably not going to be very positive. It's the same benefits, the same assets, the same entity and the same beneficiaries. The only difference is the way the liability is calculated and reported. A very likely, knee-jerk reaction in the short term to this news would be a demand for a benefit cut or a tax increase or possibly both. But what's changed with the new information? As actuaries, we have a duty to the public. We have a duty to make sure that our product is not misused. Proponents of disclosing market value view this as providing transparency. Well, is it really? The word transparency implies that it will be obvious with no chance of being misunderstood and that's not entirely clear here. I don't think that's as cut and dry as some MVL proponents would like to believe. In my opinion, changing the disclosures this way invites misuse, possibly more misuse than not disclosing it in a current political environment.
BOB: To follow-up on David's comments, yes, if you disclose these things, people notice them and it leads to a lot more discussion. Those who wish to pick up sound bites for their own interest at odds with the best interest of public plans will do so. But having information out there that is legitimate and valuable from an economics/fair value viewpoint of the world will, in the end, be helpful to both trustees and the public and all the other stakeholders. The public gets to see that maybe the measure matters and that you have to be careful to combine the percentage you hear with how it's measured. With respect to the trustees and those who monitor what goes on, there's an opportunity to get into the reason why the numbers are different and why actuaries do what they do. I'll reiterate the point that David's made on several occasions about the responsibility of actuaries. I think it's incumbent upon us to do the explanations, do the education, to provide not only a discussion of expected returns, expected contributions and values of benefits earned to date on an economic basis, but also to explain the differences in the underlying risks and the generations that are affected. I have found that most people who look at it say, "I better understand how public pension finance works and what the issues are, and therefore I'm in a position to make better decisions when I look at my goals and objectives relative to the decisions I have to make."
DAVID: I have one thing to add to Bob's comment. I agree with Bob, that as actuaries we need to educate our clients on the issues of financial economics. At the risk of being redundant, I think educating the boards and governments is different than requiring disclosures. The boards make use of tremendous amounts of information provided by experts both on the asset and on the liability side that's not ultimately disclosed. I think we can perform our duty to the public by providing this information in a forum that doesn't necessarily require disclosures.
ANDY: Finally, why do you believe it's important for actuaries, not just those working in the public plan arena to be aware of this issue?
DAVID: I think financial economics has become enormously important for our profession. The Morris Review in the United Kingdom really turned the actuarial profession on its head. The reaction in the United States has more or less been that as actuaries we must adopt financial economics as the only acceptable approach. Actuarial methods used by public pension plans are viewed by some actuaries as one of the last holdouts against this financial economics reform. I think it's important for all actuaries to understand and acknowledge that the value of financial economics is not universally agreed upon within the profession. When any profession is split on an issue like this, I believe the focus should be on reconciling the divide. For both sides to declare the other irrelevant and draw lines in the sand is a little bit embarrassing, and the current state of this debate among pension actuaries is closer to drawing lines in the sand than reconciliation. The Morris Review and CRUSAP have helped the whole actuarial profession by focusing on our image, and I think we can use this as an opportunity to improve our image by engaging in dialogue within the actuarial community and perhaps incorporating financial economics as an enhancement to our profession as opposed to a carte blanche replacement for current accepted practice.
KEITH: The future of public pensions, which is something in which all of us have an interest, may well rely on the outcome of the current debate over required disclosure of market value of liabilities. Actuaries should want their work product, and that of their professional colleagues, to reflect reality. Requiring disclosure of market value of liabilities does not meet that standard and threatens to diminish this venerable profession.
BOB: With respect to actuaries, I think we as a profession are facing a need to really review our terminology and our methodologies and the way we describe things to the outside world. The economics and financial communities use terminology that is different than ours, the principles of which are, one can argue, more sound. Discount rates used in the financial and economic worlds properly adjust for risk premium. Actuaries discount and take credit in advance for risk premium all the time. We should be modifying our terminology to make clear that for some applications we use discount rates that are market-based measures and that for other uses (actuarial, budgeting, etc.) we discount with expected returns. Further, I believe that in the end actuaries will adopt the concepts of financial economics in describing obligations measured on an economic and fair value basis. As the profession moves, I believe the accounting profession will move with us to fair valuation measurement. I think it's good for us to lead, and I don't believe that means we are required to throw out our traditional ways of budgeting. Our traditional tools can serve people well, and as Keith has pointed out very clearly, public plans with their defined benefit structures have been a great value to the public sector employees in this country and have served the taxpayers by delivering benefits at lesser costs than could be provided by alternatives. However, failing to provide true economic information is likely to bring more criticism and questions about whether the pricing was right. Being transparent from an economic viewpoint in educating the public and actuaries about this can help all parties to achieve some of our common goals which are well funded, well secured and well understood public pension plans.
RICHARD: What's at stake is credibility in one important arena in which the actuarial profession plays a critical role.
ANDY: This has been an interesting discussion–one that's going to be ongoing and one that we need to keep talking about as actuaries.
Andy Peterson, FSA, EA, MAAA, FCA, is staff fellow, Retirement Systems, at the Society of Actuaries. He can be reached at email@example.com. Keith Brainard can be reached at firstname.lastname@example.org. Richard Ennis can be reached at R.Ennis@ennisknupp.com. David Kausch, FSA, EA, MAAA, MSPA, can be reached at email@example.com. Bob North, FSA, EA, MAAA, FCA, FSPA, can be reached at firstname.lastname@example.org.