Retirement Committee

Burlington Retirement Committee Meeting of the Committee Members Agenda February 25, 2014 5 – 8pm Fletcher Free Library Fletcher Room – 2nd Floor 235 College Street, Burlington, Vermont 05401 The Retirement Committee meeting will start promptly at 5:00pm Refreshments will be provided beginning at 4:45pm 5:00pm – 5:05pm Review and Approval of January 28, 2014 meeting minutes 5:05pm – 5:20pm Understanding the Key Retirement System Terms and Definitions City Councilor Karen Paul 5:20pm – 5:50pm Understanding Retirement System Actuarial Assumptions (What They Are, Options and Impact of Options, Why They Are Used , and How the Assumptions Used Can Impact a Retirement System’s Contributions and Funding Level) State Treasurer Beth Pearce 5:50pm – 6:20pm Understanding Burlington Retirement System: An Overview of Our System’s Different Benefit Plans Human Resource Director Susan Leonard 6:20pm – 6:30 pm Break 6:30pm – 6:40pm Questions on Buck Consultants Presentation to BERS on the FY13 Valuation Report Audio Recording of Presentation Sent Separately 6:40pm – 7:40pm Discussion of Goals for Burlington’s Retirement System Consistent with Principles Established at the First Meeting (January 28) All 7:40pm – 7:50pm Input for Next Meeting Agenda and Action Items All 7:50pm – 8:00pm Public Comment Next Meeting Time: Tuesday, March 11, 2014 5pm – 7pm Topics for Future Agendas:  Presentation by John Federico and Bob Rusten on the 20 year history of the system  VPIC Presentation on Asset Management (Karen Paul, Bob Hooper, or another VPIC representative)  Consultant Suggestions and Selection Consistent with Committee- Established Goals Burlington Retirement Committee Second Meeting on February 25, 2014 Meeting extended till 8pm to make up for canceled February 13 meeting (snowstorm) Meeting began at 5:10 Chief Administrator Bob Rusten facilitating State Treasurer Beth Pearce in attendance to present to the Burlington Retirement Committee All Committee members present except Jim Strouse and Bob Hooper (absences expected) Chip Mason and Eileen Blackwood – arrived minutes late Mayor Miro Weinberger – departed early Brief re-introduction of Committee members present for the benefit of public attending (~6 people) Times listed below are based on the planned agenda. 5:00pm – 5:05pm Review and Approval of January 28, 2014 meeting minutes Unanimous Approval of the minutes (motion made by Councilor Bushor) Bob Rusten: Comments on the agenda? (none) Mayor Weinberger: Brief remarks, thanking Committee members and State Treasurer Beth Pearce for joining the meeting 5:05pm – 5:20pm Understanding the Key Retirement System Terms and Definitions City Councilor Karen Paul Councilor Paul: Overview of key terms and definitions related to the retirement system, including:  what a pension plan is (defined benefit, defined contribution, hybrid)  how vesting works in general  actuarial assumptions are focused on (i) economic (ie interest rate, discount rate, ect) and (ii) demographic factors (age, life expectancy, ect.)  the components of the annual required contribution: what is "normal service" cost and "past service" cost  the difference between the annual required contribution (ARC) and the past service or the "unfunded liability"  a brief description of BERS and VPIC Bob Rusten: Thank you, Karen, for the helpful overview. Additional questions, knowing that Treasurer Pearce will focus on actuarial assumptions? (none) 5:20pm – 5:50pm Understanding Retirement System Actuarial Assumptions (What They Are, Options and Impact of Options, Why They Are Used , and How the Assumptions Used Can Impact a Retirement System’s Contributions and Funding Level) State Treasurer Beth Pearce 1 Bob Rusten: Introduction and thanks to Treasurer Beth Pearce for attending the meeting. Treasurer Pearce: Thank you, Bob. Will shorten the presentation on terms and actuarial assumptions in light of Councilor Paul’s helpful overview Treasurer Pearce provides an outline of important actuarial assumptions and components of a retirement system in general, including:  Actuarial valuation (as opposed to market valuation) “smooths” returns to reduce volatility in a retirement system, though this will change with the implementation of new GASB rules (67, 68)  ARC – as Karen said, this is how you pay for unfunded liability  Unfunded liability occurs when assumptions are wrong or a benefit is provided without sufficient funding  GASB 68 will eliminate the “ARC,” a substantial change in how Burlington will carry out accounting related to the retirement system  Amortization o The period to pay down your unfunded liability o By Federal standard, cannot go beyond 30 years o Closed v. open amortization – the State has a closed amortization system, meaning the unfunded liability must be paid off in 30 years time certain from the date started vs Burlington which uses the "open" approach which means that each year the valuation is done and the amortization is recalculated  Treasurer Pearce recommends the Committee examine that difference – closed v. open amortization system – in the Burlington Retirement Committee discussions. The open system pushes off payments further in time, thus increasing total interest payments and the burden on taxpayers. The trade-off is more payments in the early years. The “interest rate cost” is an opportunity cost of lost interest generation from possible appreciation (paying more today is expected to generate greater investment returns)  Projected unit credit cost o This is the method Buck Consultants uses with Burlington o Like an open amortization, it can create more substantial payments later in time, and Treasurer Pearce recommends the Committee examine this assumption and why Buck Consultants (the City’s actuary) employs the method in Burlington; she recommended that the combination of both an open amortization and projected unit credit could be potentially a concern.  Actual gains and losses o Actuaries make assumptions – i.e., likelihood of retirement or COLA increases or anticipated inflation in a given year – in order to make projections about the future health of the retirement system. o The assumptions are long term – i.e., currently we have a low interest rate environment, but that shouldn’t necessarily change the assumptions made to project the health of a retirement system, which are a long-term view of a short-term rate o Importantly, the assumptions are based on experience studies. Burlington does this every five years. The State moved their most recent study up in time following the Great Recession o Buck has a model recognized as good in the industry for using these experience studies and other factors to develop long term, forward looking assumptions  Valuation: o How do you value a system based on these assumptions? 2   For example, take an assumption of investment return – a rate of return that is assumed to be high relative to the actual return will generate increased unfunded liability if the true returns do not meet the goal  Similarly, take an assumption of inflation – if inflation is higher than assumed inflation rate, this can generate losses in the system Councilor Knodell: Who decides and sets these assumptions? Treasurer Pearce: Not sure about Burlington specifically. In the State it is the board, VPIC, and an actuary. Strongly recommends not overruling an actuary, because it will complicate the audit process. [In Burlington, the BERS Board decides on these assumptions] Councilor Knodell: The rate of return assumptions, if the product of an experience study, is backward looking. Treasurer Pearce: Actuary uses a capital model process to project out over 30 years, using the experience study as a component of that analysis Bob Rusten: BERS discussed doing an experience study soon. Another factor is risk tolerance. Some of this is being discussed at the BERS Board Councilor Paul: For clarification, the BERS Board is the body that determines the assumptions underlying the City’s retirement system. Bob Rusten is a voting member. John Federico: [Also a member of the BERS Board] The last experience study was 2012. [Jim Strouse is the Chair of the BERS Board, Bob Hooper is the Vice Chair – both are also members of the Burlington Retirement Committee, but are not present at this meeting] Councilor Paul: In Buck’s model for Burlington, in the early years Buck does not expect 8 percent return, but does project those terms out over the longer-term horizon. John Federico: In the experience study, Buck notes the projected rate of return for 10, 20, 30 years. Bob Rusten: With so many things to cover – we should acknowledge that (i) we need to examine Burlington’s experience study and (ii) that changes we make in one component of the system will have consequences across the rest of the system. Treasurer Pearce: Will skip ahead on some of the definitions, but one major point:  Projected unit credit is a backloaded policy, and in combination with an open-ended amortization system – does this make sense? Will send note on critique of this combination. Also, with respect to VMERS, there are 447 – 449 entities in VMERS, and there is a question with the auditors and how the system can be audited without burdensome costs to the State [The slides discussed by Treasurer Pearce are on the State Treasurer website, including a chart on “popular misconceptions” related to the new GASB rules] Bob Rusten: Beth, can you speak more to the point about a closed v open system, vice closed v open amortization? With an open system, you can face a temptation to fudge the numbers by projecting different costs for the future to reduce the apparent cost 3 Beth Pearce: Yes. A closed system is better for comparative purposes, and does not account for differences in employee benefits over time Councilor Bushor: GASB 68? Treasurer Pearce: Government Accounting Standards Board Bob Rusten: Something we will discuss further at a future meeting. Councilor Paul: For the next meeting – p. 23 and 44 of the FY2013 Actuarial Valuation (on the City website, here) 5:50pm – 6:20pm Understanding Burlington Retirement System: An Overview of Our System’s Different Benefit Plans Human Resource Director Susan Leonard Susan Leonard’s presentation covers major components of benefit calculation, including:  Outline of the City’s defined benefit plan  Benefit calculation components (accrual rate, number of years of service, average final compensation)  The growth in types of plans in Burlington to make the system complex to administer Councilor Bushor: If you retire early, can you defer payment? (yes) If so, how do you make estimates with so many moving targets? John Federico: This is what the actuarial valuation report is intended to do – the actuary has details year by year to help make projections on who will take full benefits or less than full in a given year Councilor Mason: The plan design is for a city 10 to 20 times Burlington’s size. John Federico: Is it possible to break down the cost to the system of different benefit types, beyond the “Class A / Class B” distinction? Susan Leonard: People have asked, and we have asked the actuary, but they have not been able to provide those numbers because they are not statistically significant. Bob Rusten: An unfair question for Susan: There are a number of different benefit options available – is it possible that one option costs more to the system than another? Susan Leonard: The cost to the system is assumed to be comparable across the different options. Joe Keenan: People look at the choices they have, and it is difficult to predict what is the best option when weighting a choice like COLA or no COLA – it requires considering (i) discipline to invest and (ii) frankly the uncertainty of COLA growth continuing. Bob Rusten: The actuary, and we, are assuming these options all cost the same – that’s another assumption that we should consider. If the cost to the system of the different options is different, should the contribution for the different options also be different? John Federico: Susan, could you share the excel file showing the multiple different plans? 4 Susan Leonard: Would want to review some questions before making the document public. Councilor Knodell: I want to make sure I understand the benefit calculation: accrual rate * years of service * average final compensation? And is the accrual rate the rate when you retire or the rate when you enter service? Susan Leonard: Yes. Accrual rate is what was in effect when you were hired. Bargaining can create exceptions, though bargaining changes usually effect new employees. Councilor Paul: From an employee perspective, can you have a conversation with your peers if there are so many separate plans. Bill Rasch: You can have the conversation. I wouldn’t say it is easy because there is so much variation, but people understand their plan. Susan Leonard: Yes. And people do know what rules apply to themselves, but it is complicated administratively. Mike Flora: Forming this Committee is an indication to some that potential changes could be coming, and those changes affect those who are 10, 15 years in the most – those at the end of their careers are likely excluded from any change and appropriately so. Those who are new will know the new rules. Bob Rusten: Thanks all, short break followed by questions about the Buck presentation to BERS (February 20, on the FY13 valuation) 6:30pm – 6:40pm Questions on Buck Consultants Presentation to BERS on the FY13 Valuation Report Audio Recording of Presentation Sent Separately Bob Rusten: Questions? Jeffrey Wimette: Difference in evaluation in gain between Buck and Dahab (Dahab is an outside consultant that provides an external review)? Further, questions about VPIC, its responsibilities, and how its investments affect Burlington Bob Rusten: Difference was due to a cash issue – Dahab had the wrong figure, and Buck’s was correct. On VPIC, considering bringing Steve Jeffrey, a board member of VPIC, to speak to the committee about the investment allocation Councilor Knodell: What drives the downward trend? Bob Rusten: Several factors, but the growth in the unfunded liability was driven by a failure to hit the investment target Councilor Mason: 2013 was a historic year for the stock market, and yet we did not hit our investment benchmark. Why? Was this discussed by BERS, and should we consider changing the assumption? 5 John Federico: VPIC has a very low risk tolerance, hence the low investment return even in a record year, and that has a negative effect on the unfunded liability. Bob Rusten: Agreed, and there was discussion about changing the 8 percent investment assumption for the reason Chip outlined 6:40pm – 7:40pm Discussion of Goals for Burlington’s Retirement System Consistent with Principles Established at the First Meeting (January 28) All Bob Rusten: Explanation of SMART Goals (copied below), and the possible goal areas based on discussions at the first meeting. At some point, we will have to decide if the goals as a group that we decide on are contradictory, and then use these to shape the scope of a consultant. But first, feedback on the possible goal areas listed below? Emphasizing that here we want to figure out what the goals are, NOT how to reach them. Goals (what we want to achieve and not how to do so): Specific Measurable Action-Oriented Realistic Time/resource Defined Possible Goal Areas: Contributions to Fund Unfunded Liability Benefit Complexity Public Confidence in and Support for Pension System Recruit and Retain Staff John Federico: The goals outlined are not an exhaustive list. The Committee needs to make sure to follow its charge and seek to address underlying issues Bob Rusten: Agreed not exhaustive, goals shape how we address underlying problems. The purpose here is to get consensus on what we should explore, not how to explore them. Councilor Bushor: I like these goals as outlined, and think they are consistent with the minutes, which I reviewed again prior to tonight’s meeting. “Contributions” should be a composite of City, employee, and taxpayer contributions; retirement, and unfunded liability are also longstanding goals to be addressed. 6 Mike Flora: Last meeting, I noted that addressing the unfunded liability and establishing a timeline for doing so the basic goal of the committee John Federico: A goal should be to have the system functioning well with the best funding arrangement. Bob Rusten: But using the “SMART” criteria above, what is best, and best to who? Mike Flora: A goal like 85 percent or 100 percent reduction of the unfunded liability could have substantially different consequences for City employees. Is it realistic to reduce the unfunded liability by 100 percent, and over what timeframe? Eileen Blackwood: The “how” piece isn’t what we should focus on – if 100 percent reduction is the goal, then something like a pension obligation bond might be on the table. We need to focus on the goal first, and the how later. John Federico: It also needs to be a sustainable fix – reducing 100 percent of the liability could be less helpful if you then create more liabilities the next year. The system needs to work. Councilor Knodell: 100 percent reduction sounds nice but is not feasible. A 10-15 year plan is the more appropriate approach– something significant in the plan. I don’t think 100 percent in 15 years is realistic Joe Keenan: 80 percent in 10 years. Bob Rusten: What do people think about that approach – again, focusing only on the goal, not on the how we get there at this point. And, Joe, do we need to do this year by year, or all in one year at year 10? Joe Keenan: Public perception is important. If the goal is 80 percent funded over 10 years, the system needs to show its stabilizing in the first couple years and improving thereafter. We need to assure the public we are addressing the problem. John Federico: The discount rate does not drive everything – we need to be clear that its more than one factor. I am not willing to agree to that goal unless an actuary is able to look at this Mike Flora: John, it’s not about the “how” yet – it is getting a clear goal. We’d then need to bring in an actuary to see if the goal is feasible. Eileen Blackwood: And I don’t know if we’re being ambitious enough – the Mayor is thinking about eliminating the unfunded liability in 3 years, and structuring the system so that there is no subsequent growth of unfunded liability. Councilor Paul: One way to do that is something like pension obligation bonds. I think it's important for everyone to understand pension obligation bonds and the downside to them. I think are a bad idea and can shift debt from one place to another. Eileen Blackwood: I agree that such a goal is ambitious, but we also don’t know how ambitious it is, and we’ll need to take it to external evaluation Mike Flora – Take options to a consultant – 3 years and 100 percent, the 10 year 80 percent or a 10 year 90 percent, for example. 7 Bob Rusten: We also need to address the underlying problems that drive the growth of the unfunded liability Councilor Paul: As Beth noted, open amortization, credit unit adjustment, and how much money we pay out is interest on taxpayer money vice the actual money are all things to explore. Nothing is wrong with Joe’s suggestion of 80 percent funded in 10 years or the idea of reducing the unfunded liability in 3 years – but I would like us to examine Beth’s suggestions as well. Bob Rusten: So prior to setting a goal for unfunded liability, we look at some of the specific data points? John Federico: So you would ask Buck to run the numbers with those changes and see what the difference is (yes). I think we need to do that Mike Flora: And also would like to see what the adjustment is with GASB, if applicable Bob Rusten: My concern is that these suggestions amount to some minor tweaks to the system, and will not fix the issue. That’s why the goal discussion is important – if we want to actually address some of the unfunded liability. To do these tweaks, we also need to know how much would it cost, who could authorize it, and who would pay it. Eileen Blackwood: Why not also ask Buck about the 3 year, 100 percent reduction or the 10 years 80 percent funded approach? Bill Rasch: And is it Buck we want to do it? Others out there with better advice? I would like to have experts come in and speak to it. Councilor Bushor: Why haven’t we asked these questions before? Bob Rusten: We also need to know what questions to ask. We need a consultant to come in and help provide options to address the problem Karen Paul: Buck is a huge company. There are people there who do pension consulting who work at Buck. They are considered very good in the industry. John Federico: NCEPERS is another alternative – Bob your title may give you the ability to work or contract with them Bob Rusten: Would people trust if I could get someone in to the next meeting to function as the consultant? I could also bring back a list of names, though this will delay us. What is the group’s preference? Joe Keenan: If you can bring someone in, and we feel like we can trust them, let’s do it. If we don’t trust what is said, then we should explore other options. Chip Mason: I’m not concerned with delegation to Bob of this task. I am just not clear on what we are asking for. We need someone who can knowledgably address questions. I am not clear on what the options even are at this time. Bob Rusten: Clear that people want to reduce the trend of unfunded liability, the trajectory of City contribution, ablity to recruit and train employees – 8 John Federico - and people want to know that the retirement system will be functioning well 30+ years from now. Mike Flora: And there is a benchmarking opportunity with other municipalities, which are exploring sometimes fairly drastic measures. We should see what other comparable cities are doing as well Bob Rusten: So all are okay with me bringing in a consultant if I can to the next meeting? (all concur) 7:40pm – 7:50pm Input for Next Meeting Agenda and Action Items All Karen Paul: Include Steve Rauh from VPIC, potentially at the next meeting. Bob Rusten: And the discussion of the experience study next time? I will also try for a consultant. Chip Mason: If you have questions, we can also send those to Bob and he could get them to David Driscoll Bob: Agreed - If you have questions, send them to me by Monday on either the experience study or the valuation report and I will get them to David Driscoll preferably for response (including possibly a phone call) at the next meeting. 7:50pm – 8:00pm Public Comment Gene Bergmann Consider holding public comment at beginning or middle of event. 8pm too late. Rob Ruolff: You need to go out and do something, even if erroneous – you need to do something to address this problem Meeting ends at 7:48pm Next Meeting Time: Tuesday, March 11, 2014 5pm – 7pm Fletcher Free Library 9