A Different View of the ASB

A Different View of the ASB

This article is written in response to the article that appeared in the February/March issue of The Actuary entitled Better Standards for Setting Standards.
Larry Sher

I commend Dave Kass1 for his thoughtful and timely article on how Actuarial Standards of Practice (ASOPs) should be set in the United States and I join him in encouraging all actuaries to participate in this discussion. I am a member of the ASB; however, what follows are my views and opinions and should not be construed as representing the views and opinions of the ASB.

Dave begins by describing the current process for developing standards as "cataloguing" what most qualified actuaries are doing. He asserts that the current "flexible" system allows bad practices to continue and does not embrace new developments in actuarial science or in other fields such as economics. He expresses concern that if the current approach continues, our profession will risk losing our ability to self-regulate–"We must lead or be lead."

Like Dave, I too am very concerned about: (1) the prospects of losing our ability to set our own standards, like what recently occurred in the United Kingdom, (2) the time it takes for ASOPs to be developed or updated, and (3) the impact of results-oriented pressures from our principals. However, Dave and I differ on (1) where we currently are and (2) how to best improve the situation.

Current Standard Setting Process

I disagree with Dave's characterization of the current process as a "consensus" and "hands-off" approach. Dave refers to The Introduction to the Actuarial Standards of Practice to support his view. While I think that, when read in its entirety, the Introduction does not support that view, some clarification–or perhaps a change in emphasis–in that document could be helpful.2

In my experience, the ASB strives to articulate standards that embody "appropriate" practice, not minimal or poor practice. The process typically begins by identifying the range of practices in the given area because often the best indicator of appropriate practice is "generally accepted practice." But just because a practice is used by a number of actuaries does not necessarily mean that the ASOP will embrace it.

The standards are written to allow practitioners to use what they consider to be the best practices, but the standards do not define nor mandate any particular practices as being "best." That is, they typically set forth general procedures that the actuary should follow in assessing what is appropriate in a given situation and call on the actuary to exercise professional judgment.

Dave's View of ASB History

Dave provides us with some interesting history of our professional organizations, including a discussion of the developments that led to the formation of the ASB in the mid-1980s. Dave suggests that the current approach (as he sees it) might have been fine many years ago when the profession was smaller, and actuaries knew what the "right thing" was and had the integrity to do it. But, according to Dave, the world we now live in demands a different approach, one where the ASB should "lead" rather than "follow."

Dave asserted that over the 21-year history of the ASB that its process has been the same–"simply compiling a catalogue."

Actions Speak Louder Than Words

I have been involved with standard setting matters for almost half of the 21 years the ASB has existed (over five years on the Pension Committee in the mid-late 1990s and during the most recent four years on the Board). My involvement with two key pension standards–ASOP 27 Selection of Economic Assumptions for Measuring Pension Obligations, and ASOP 4 Measuring Pension Obligations and Determining Pension Plan Costs or Contributions–illustrates that Dave's "cataloguing" characterization is unfair.3

ASOP 27

ASOP 27 was adopted by the ASB in December 1996 after about seven years of deliberations and three exposure drafts. Two areas where ASOP 27 broke new ground were: (1) the requirement to use explicit assumptions (i.e., each assumption must be reasonable on its own), and (2) the introduction of the concept of "best estimate range" for selecting economic assumptions.

While many pension actuaries were already using explicit assumptions in most or all circumstances, many others were not. In fact, the use of explicit assumptions went beyond what was required in ERISA. The Pension Committee and the ASB decided to "raise the bar" in this respect because it was felt that it was better practice (i.e., more transparent) for actuaries to select assumptions, each of which is reasonable.

At the time, the Pension Committee concluded that the best-estimate range concept, in conjunction with measurement specific factors, was an important development. It was intended to provide a corridor, at least conceptually, around the selection of economic assumptions, and to emphasize to actuaries and users of our work products that assumptions are not predictions and that variations in experience are inevitable.

I realize that ASOP 27, including the best-estimate range concept, continues to be controversial and, as discussed later, may be a source of Dave's irritation with the standards.

ASOP 4

The development of an update to ASOP 4, which began around 2000 and is now in its third exposure draft, also illustrates how the ASB process really works. There are many elements included in the third exposure draft that can hardly be described as merely cataloguing practices and that clearly address one of the issues that concerns Dave–needing more ammunition to deal with pressure from our principals. Here are some examples.

  1. Section 3.2 (Prescribed Assumption or Method Selected by Plan Sponsor). The actuary is to evaluate whether a "prescribed" assumption or method selected by a plan sponsor is reasonable for the purpose of a measurement.4 For example, if a plan sponsor instructs the actuary to use a 10 percent expected return assumption under SFAS No. 87 and the actuary considers the assumption to be unreasonable, the actuary must generally disclose that conclusion.5
  2. Section 3.9 (Interrelationship Among Actuarial Assumptions, Procedures and Plan Provisions). Where a plan's provisions are difficult to measure using deterministic assumptions or procedures (e.g., a floor-offset arrangement), the actuary is to consider using alternative assumptions or procedures such as stochastic modeling or option pricing techniques. If these provisions are significant and the contingent nature of the obligations is not reflected in the valuation, the actuary is to so disclose.
  3. Section 3.13 (Ability to Pay Benefits When Due). In some cases a contribution allocation procedure that otherwise meets ASOP 4 (and other ASOPs) may not be expected to produce adequate assets to make benefit payments when due. The actuary would have to select an actuarial cost method and an amortization method to minimize that possibility. When the actuary is not responsible for selecting the cost or amortization method, a disclosure would be required if, in his or her professional judgment, the asset adequacy provision is in jeopardy.
  4. Section 4.1(m) (Communications Requirements). The actuary is to include in the communication of the work product a statement indicating that future measurements may differ significantly from the current measurement and that either (1) the scope of the assignment included an analysis of range of such future measurements, or (2) due to the limited scope of the assignment no analysis was performed of the potential range of future measurements.

Kass Recommendation

Dave recommends that ASB members "use their skill, experience, intelligence and judgment to evaluate all practices that warrant consideration, evaluate them in light of the current realities and specify which practice(s) are to be designated generally accepted."

On its face, this recommendation seems reasonable and may not be too much different from what the ASB currently does (although I believe the ASB would prefer to designate such practices "appropriate"). But, if what Dave has in mind is to severely limit those practices that are "designated generally accepted," I think that would be problematic for a few reasons. First, for the ASB to severely limit practice could be a violation of the law, as discussed by Lauren Bloom, former ASB legal counsel, in a recent article co-authored with Michael LaMonica (former ASB chair).6

Second, even if the legal obstacles could be overcome, severely restricting practice would almost certainly involve a move away from principles-based standards in favor of rules-based standards (i.e., with more prescriptions). That is what Dave seems to have in mind since he points to the Financial Accounting Standards Board (FASB) as a model for how the ASB should operate. Putting aside the cost of implementing such an approach,7 FASB has the legal authority (through the SEC) of the U.S. government and therefore can impose "best practice" standards without having to be concerned with anti-competitive issues that the ASB must consider. Moreover, while FASB has stated a goal to move toward a principle-based approach, its primary pension standard, SFAS No. 87, is hardly a model that I would think most actuaries would want to emulate given that it is riddled with prescriptive requirements.

Given the diversity of assignments that actuaries undertake, any move to severely restrict practices, whether through rules-based standards or through quasi-principles-based standards, would effectively be substituting the Board's judgment for that of each individual actuary. Effectively, this approach would be tantamount to issuing "best practice" standards even if that is not what Dave had in mind.

Reading Between Dave's Lines

In addition to ASOPs 27 and 4, I have also been involved in the development of a new standard–Selection and Use of Asset Valuation Methods for Pension Valuations. In response to the first exposure draft, Dave and 10 other actuaries submitted comments in a May 15, 2002 letter asking the ASB to "make a positive professional statement by endorsing the use of market value as the single best measure of pension assets." The letter went on to suggest that market value be designated as a "best practice within a range of acceptable but notably less scientific measures." Even assuming that the ASB had the ability to follow this suggestion under the law, such an endorsement surely would force actuaries to restrict their practices and to use market value of pension assets in most or all of their assignments. The Pension Committee considered this suggestion and decided not to adopt it because it would be contrary to the standard setting process and because many actuaries (including other commentators) did not agree that market value of assets should be virtually mandated.

Those advocating financial economics views on pension finance have expressed dismay that their market oriented approach for valuing pension liabilities is not explicitly endorsed in ASOP No. 27. I believe Dave is in this camp. But, the fact that ASOP 27 does not explicitly mention financial economics does not mean that one cannot use those principles in selecting a discount rate or other economic assumption. I interpret ASOP 27 to allow the use of a reasonable application of financial economics principles without having to resort to the deviation clause.

The ASB Pension Committee plans on taking up a thorough review of ASOP 27 in the near future and surely the issue of reflecting financial economics will be a major consideration. But, at least some financial economics advocates would like to see ASOP 27 written in a way that endorses financial economics concepts and marginalizes other more traditional actuarial views. I'm not sure Dave is in this camp, but if he is, that would help me better understand where he is coming from. As in the case of the asset valuation method standard, I believe that such an endorsement would be tantamount to a best-practice mandate from the ASB. Again, even if such an approach is lawful, I contend that most actuaries would view it negatively.

Beauty is in the Eye of the Beholder

While there is certainly room for improvement in the current ASB procedures, I am concerned that Dave's approach, assuming I understand it and that it would not violate the law, could lead to an elitist and unstable platform for setting standards. It could also greatly politicize the process.

Dave calls for leadership in setting standards–in other words, the ASB should narrow practice by either eliminating bad practices or "encouraging" (read, "mandate") good practices. I ask Dave to consider how he would react to an ASB that has a different view than he has on what a single approach to good practice is in a particular area? I can hear the cry already: "impeach those bums."

Under this paradigm, how might a Board react to a prior Board's decision that it disagrees with? Will Dave and others who agree with his positions push for certain people to join the Board to try and overturn the positions taken by previous Boards? We have seen how bad politics in Washington affects the public. We certainly do not need to politicize the standards setting process. Giving too much power to the ASB, no matter how smart we may think the Board members are, is a very dangerous proposition.

Conclusions

In principle, I agree with Dave Kass' concerns and his call for raising the level of practice among actuaries. But, we differ as to how the current standard setting process works. Dave feels that it is merely a cataloguing or codification approach of prevailing practices. I think that is an oversimplification since, in practice, the ASB (and its committees and task forces) often identify inappropriate or marginal practices and either ban or discourage them. The current exposure draft of ASOP 4 is a case in point.

I think Dave is right that it takes too much time to develop or change standards. We need to figure out ways to speed up the process, recognizing that the ASB and committee members are unpaid volunteers who usually have other paying jobs to deal with.

A related but somewhat different issue is Dave's view that the ASB is missing opportunities to improve practice by failing to embrace recent developments in actuarial science or from other fields such as economics. But, even if the ASB had more resources, I think it would continue to be prudent for the standard setting process to be thoughtful and deliberate as Dave agrees it should be. In my view, new ideas need time to be vetted before they are quickly embraced (or rejected) by the ASB. And the vetting process needs to be done in ways that engage broad cross-sections of affected actuaries.

Perhaps Dave summed up his position (and our different views) most clearly when he said: "Better new approaches–even unpopular ones–must REPLACE inferior older ones." In the real world, it is rarely evident when some new approach is introduced that it is truly superior to something that has been accepted as "right" for a long time. Even if all or most of the ASB members thought that new approach was superior to traditional actuarial concepts, in my view it would be imprudent (and possibly illegal) for the ASB to impose its opinion on the profession knowing that many or even most actuaries disagreed.

Finally, because the ASB and its committees and task forces tend to include people who care very much about good actuarial practice, there is a natural tendency to push the envelope to some degree. Whether there needs to be a concerted effort to push somewhat harder is subject to debate and I look forward to continuing this discussion with Dave and others.

Larry Sher is principal, Buck Consultants, an ACS Company.

Footnotes:

  1. I refer to Mr. Kass as Dave in this article–I have known him for several years and feel more comfortable referring to him in that way.
  2. A few actuaries, including Dave Kass, urged the ASB to delay publishing the Introduction document until the ASB changes it procedures to more closely reflect their views on how the ASB should operate. The ASB rejected that suggestion because it thought that it was important to have a document that articulated the current procedures with the hope that it would generate discussion on how to improve the procedures. Apparently that hope is now being realized!
  3. I apologize to our non-pension practice readers for emphasizing pension standards, but that is my practice area and it is also Dave's practice area (or at least one of them).
  4. A method or assumption is "prescribed" by the sponsor if a law or other binding authority (e.g., accounting standard) gives the plan sponsor the authority to select it.
  5. For this purpose the assumption would be unreasonable if, in the actuary's professional judgment, it conflicts significantly from what the actuary would consider to be reasonable. If the actuary is unable to make this assessment because he or she does not possess the necessary expertise or the plan sponsor is unwilling or unable to provide the information necessary to make the assessment, the actuary must disclose that.
  6. The article, "In the Eye of the Beholder: Why Best Practice Standards of Practice May Not Be Best," appeared in the March/April 2006 edition of Contingencies.
  7. The FASB members are paid full salaries and must disassociate with their prior employers. Given the much smaller size of our profession, membership dues would have to increase significantly to support an ASB that is compensated at market rates.