Call to Order and Roll Call
The 11th meeting of the Public Pension Oversight Board was held on Monday, November 23, 2015, at 12:00 PM, in Room 169 of the Capitol Annex. Senator Joe Bowen, Chair, called the meeting to order, and the secretary called the roll.
Present were:
Members: Senator Joe Bowen, Co-Chair; Representative Brent Yonts, Co-Chair; Senator Jimmy Higdon; Representatives Brian Linder and Tommy Thompson; Robyn Bender, Tom Bennett, Jane Driskell, James M. "Mac" Jefferson, Sharon Mattingly, and Alison Stemler.
Guests: Representatives Arnold Simpson, Derrick Graham, Brad Montell, and James Tipton; Mike Harmon, Representative and Auditor of Public Accounts-Elect; William B. Fornia, Actuary, Pension Trustee Advisors; Rogier Slingerland, CEM Benchmarking Inc.; Kim Nicholl, Senior Vice President and Actuary, Segal Consulting; Bill Thielen, Executive Director, Kentucky Retirement Systems; David Peden, Chief Investment Officer, Kentucky Retirement Systems; Beau Barnes, Deputy Executive Secretary of Operations and General Counsel, Kentucky Teachers’ Retirement System; Donna Early, Executive Director, Judicial Form Retirement System; Damien Stanton, SaveCERS.org.
LRC Staff: Brad Gross, Committee Staff Administrator; Greg Woosley, Analyst; Bo Cracraft, Analyst; Terrance Sullivan, Analyst; and Maurya Allen, Committee Assistant.
Approval of Minutes
Representative Thompson moved that the minutes of the October 26, 2015, meeting be approved. Representative Yonts seconded the motion, and the minutes were approved without objection.
Kentucky Teachers’ Retirement System Funding Work Group Update
Mr. William B. Fornia, President, Pension Trustee Advisors was present to provide a summary of the Kentucky Teachers’ Retirement System (KTRS) Funding Work Group’s meetings. Mr. Fornia stated that meeting presentations were available online for further study if the board members were interested. Additionally, a workbook has been created to summarize various funding solution projections and is also available.
The first meeting of the KTRS Funding Work Group (Work Group) was on July 17, 2015, where the most important issues facing the system were discussed. Primary among these issues were benefits and their importance to the Commonwealth and to the education system. A key point discussed was that the cost for the benefits for teachers hired after 2008 is 15.68 percent of pay, of which the teachers pay 9 percent on average, leaving approximately 6.6 percent of pay as a state contribution. This contrasts with Social Security where it would be 6.2 percent of pay. This means that for future benefits, the KTRS benefit is actually better than it would be for Social Security, in which teachers do not participate. KTRS investment history was also reviewed, with a comparison to peer groups where possible. This review and peer group comparison indicated that the underfunded nature of the system is in part due to investment return, but is not due specifically to KTRS’s investment return being less than average. The 2008 changes to the plan were overviewed, but retiree healthcare was not included as a significant point of discussion by the work group. Mr. Fornia also reviewed a chart illustrating that contributions made by other states to their pension systems has been increasing, whereas the contributions made by Kentucky has remained roughly the same over the last fifteen years. The statutorily required contribution rate is below the actuarially required contribution (ARC) rate causing the system to be underfunded. Currently, the statutory contribution rate is approximately 14 percent of pay, or $500 million, below the ARC. Pension Obligation Bonds (POBs) were also briefly discussed at the first Work Group meeting, along with the potential for POBs to resolve the problem, but it was noted that POBs would mostly serve to ease the transition into a higher state contribution level.
The second Work Group meeting was on July 31, 2015, where the long-term consequences of the unfunded liability were discussed. It was noted that if the contribution is not increased, the system is projected to be insolvent within 20-25 years. The broad options to address this are either putting more money into the system, in the form of increased contributions, or taking less money from the system, in the form of reduced benefits. The Work Group agreed that any changes made should be legal and should not negatively impact the education system. The primary consideration regarding legality that was discussed is the inviolable contract currently in statute between the system and retirees. There are a few exceptions to this that can be modified, such as when a teacher is 55 years-old with 27 years of service their final average salary is based on a 3-year average as opposed to a 5-year average for determination of benefits. This calculation could be changed, and the savings were quantified by KTRS, in order to improve the funding of the system. Mr. Fornia then showed a chart that illustrated several methods to go from current levels of funding to 100 percent funding. In the chart, POBs show an initial jump in funding levels, whereas payment of the ARC results in a more gradual achievement of full funding. However, the proposal of POBs increases risk to the system and would not be a complete solution. It was considered most favorable if the issuance of bonds was part of a structured long-term plan, not as a quick fix.
The third meeting of the work group was held August 28, 2015. Mr. Fornia said that this meeting was comprised predominately of benefit comparisons between Kentucky and other states. He noted that, on average, future teachers have significantly higher benefits if they are hired at a young age and work at least 30 years relative to their peers. Older hires have benefits that are slightly below average, but benefits for average aged hires are slightly above average, relative to their peers. Teacher compensation was also compared and found to be lower in Kentucky than in many states, even after adjustments for cost of living are made. Studies looking at solutions used by other states were also reviewed. Mr. Fornia stated that it was important to note that some reform had already been implemented to reduce the cost for future teachers by approximately 1.25 percent. He also noted that teachers that are hired into KTRS at a later age are likely to have Social Security benefits from previous employment and that there really was no way for the models to take that into account. Teacher contribution rates were also looked at, but he noted it was difficult to compare given that different states, and sometimes even different school districts within the same state, vary whether they pay the teacher contribution or whether that is the responsibility of the teacher.
Chairman Bowen commented that, although not shown in the charts presented, the state had been making a 13.1 percent contribution since 2007. Senator Higdon added that the normal cost provided to the General Assembly looks good, but he noted that many factors come into play after teachers are actually hired and that effect the normal cost. Mr. Fornia agreed that when investment returns did not go as expected, as was seen since 2000 and especially in 2008, the system’s funding was negatively impacted and it was unable to catch back up to the predictions made at the outset. Representative Harmon asked for the different rates of return for the systems to be applied, and for them to be compared to illustrate the systems actual returns versus the projected returns. Mr. Fornia stated that would be a very informative comparison, but he noted that the data needed is not information he has access to. Instead, he noted that perhaps the question could be forwarded to KTRS and Kentucky Retirement System (KRS) staff for them to answer.
Mr. Fornia continued to present graphics illustrating the comparison between Kentucky and other states in terms of wages and pension reform strategies. Senator Higdon asked specifically about the strategies of Missouri and Nevada regarding shared risk plans and how those work. Mr. Fornia answered that those have been in place for some time and they work by allowing for the contributions made by members to also change over time. In most states, it is like Kentucky where the member contribution is fixed. In Nevada and Missouri, as costs go up or down, the member contribution also increases or decreases. It appears that the change is not simple, but that it does appear to solve many problems in the way current systems are funded. Representative Graham stated that in the Missouri plan, as the contribution level of the teachers has risen, so has the state’s contribution. As a result, he asked if the plan had greater stability than other plans. Mr. Fornia said that both the Missouri and Nevada plans are better funded than Kentucky’s plan, but that it does not fix all of the problems and that the primary goal is still full funding for the plans. He referenced Wisconsin as an example of what he believed was a better solution, which adjusts the cost of living adjustment (COLA) every year relative to plan funding.
The fourth meeting of the Work Group was held on September 11, 2015, and featured testimony from impacted constituent groups, especially teachers, focusing on the importance of the system. The fifth meeting on September 25, 2015, looked at broad solution suggestions to strengthen solvency, including increased contributions, or reduction of benefits, or some combination of the two. It was agreed by the Work Group members that each option set were individually only partial solutions. In the sixth meeting on October 16, 2015, the group modeled different strategies to see which ones projected the most feasible solution to reach full funding within 30 years. The first projection was of a 5 percent benefit reduction for future teachers, and this projection resulted in the system becoming insolvent over time regardless of the benefit reduction. The second projection included an increased contribution of 5 percent of pay with no benefit reduction, and this projection resulted in solvency for over a decade, but it did not reach full funding within a generation as desired. The third projection was a combination of increased contributions of 5 percent and benefit reductions of 3 percent for future teachers. This model was the first to show an upward trend, but it also would not reach the target within 30 years. Reducing benefits for current teachers by 1 percent of pay required 40 years to reach full funding. However, it was noted that if returns are poor, with an example of a 6 percent return as opposed to the assumed 7.5 percent return, the solution does not work at all. One solution that reached actuarial soundness included an 8 percent contribution increase, a 3 percent reduction in benefits for future teachers, a 1 percent reduction in benefits for current teachers, and an extension of a 2.7 percent special assessment. Mr. Fornia stressed that this was not necessarily the solution the Work Group was endorsing, but it was merely a scenario to help demonstrate the kinds of changes that might be necessary to reach full funding. He also stated that there was greater flexibility in changing benefits for future hires because of the inviolable contract and that had to be taken into consideration.
Another possible scenario involved major changes, such as moving teachers into Social Security. That scenario, as well as ones involving creation of a defined contribution plan or a hybrid plan, were found to not help with plan costs. Lastly, he noted that the use of POBs was best understood to be part of a solution when used in tandem with an increase in the contribution rate. At the November 16, 2015 meeting, another scenario was considered that included no benefit reductions or POBs, but instead centered on phasing in a 12 percent contribution increase over the next 12 years. This scenario also showed full funding by 2044, and it generated a great deal of discussion because it included not paying the full ARC, which had previously been estimated at 14 percent of pay. Mr. Fornia explained that this discrepancy is because when the ARC is calculated, it doesn’t include any special appropriations that will end or any payments toward POBs that may eventually be directed into the system as the prior issued POBs are paid off. Additionally, teachers hired since 2008 are already projected to have 1.25 percent lower pension benefits and that is also not taken into account in ARC estimations. Mr. Fornia also noted that some problems with phasing into a 12 percent contribution include the fact that credit rating agencies may not view it as a credible solution, and that because KTRS is currently only 60 percent funded, it is difficult for the system to invest at a level necessary to achieve the 7.5 percent investment return on which the projections rely. Finally, he noted that the next and last meeting of the Work Group would be held on December 1, 2015, when the group would work on finalizing guiding principles and concluding recommendations.
Chairman Bowen commended Mr. Fornia for providing such a thorough overview in such a short time and asked for specific questions regarding the presentation. Representative Yonts asked if sick leave payments, POB debt service, and COLAs factored into the scenarios. Mr. Fornia answered that there was an assumption that those amounts would be used to contribute to KTRS funding, with the understanding that those funds are currently spoken for elsewhere and it would take General Assembly action to dedicate those funds to KTRS. Representative Yonts also asked how credit rating evaluations might be impacted if contribution rates are phased in over a 12 or 20 year period. Mr. Fornia felt that a long phase-in would not be favorable, but he noted that he was not an expert on the subject matter. Representative Yonts finally asked if it would be helpful for COLAs and sick leave to be rolled into the overall contribution rate as opposed to be funded with a separate appropriation, at least as far as budgeting and certainty were concerned. Mr. Fornia stated that the concept was essentially considered in the projection where, under one scenario, the contribution would go up briefly and then slide back down. The 12 percent proposal assumes that the percentage of pay commitment, as it currently stands, remains the same over time with a gradual increase of 1 percent per year over 12 years and with all the current benefits going out remaining in place, including the sick pay appropriations.
Kentucky Retirement Systems (KRS) Investment Cost Analysis
Mr. Rogier Slingerland, Partner, CEM Benchmarking Services, was present with Mr. Bill Thielen, Executive Director, Kentucky Retirement Systems, and Mr. David Peden, Chief Investment Officer, Kentucky Retirement Systems to present the KRS Investment Cost Analysis as recommended by the Public Pension Oversight Board last year. Mr. Slingerland explained that CEM Benchmarking specializes in benchmarking pension systems and works with approximately 400 systems worldwide, but does not offer consulting or advice on what systems should do. Their primary focus is on data and comparisons with other systems.
He stated that they begin by establishing a peer group and a universe for the specific system being analyzed. The peer group emphasis on a cost story minimizes the impact of economies of scale, where larger funds get better fees and could present an unfair advantage when comparing costs. There were 18 United States public funds that represented KRS’s peer group, ranging in assets under management from $8.9 billion to $25.5 billion. The median size of the group was $15 billion, which compares to about $15.5 billion for KRS. The comparisons in the report also show a universe of all United States public pension funds that CEM works with, referred to as U.S. Public throughout the presentation. CEM also uses a 5-year horizon for investment returns, beginning with a net total return, or the total return after costs, which for KRS is 8.2 percent. This compares to a peer group median of 9.7 percent and 9.8 percent for the universe. Mr. Slingerland drew members’ attention to graphs illustrating individual years as well as the 5-year review, which include the median returns, percentile rankings, and KRS’s position in relation to that information. Mr. Peden noted that all of the information being presented is based on a calendar year at the end of 2014, not the fiscal year, in order to provide the most accurate comparison to other funds.
Looking at the 5-year policy return, or benchmark return, it was 9.2 percent, which was very close to the peer median of 9.3 percent and just below the U.S. Public median of 9.7 percent. Some reasons for that, according to Mr. Slingerland, were caused by a lower allocation of stock, higher allocation to inflation indexed bonds, and higher allocation to hedge funds. These are partially offset by a higher allocation to private equity, which is one of the better performing asset classes. The 5-year net value added, or the net total return minus the return over the policy benchmark, was a negative 1 percent, which compares to a peer median of zero percent and a U.S. Public median of 0.1 percent over the five year period. The net returns were higher in U.S. stock, emerging market stock, and private equity relative to the U.S. Public average. KRS was lower in ACWI ex U.S., fixed income, real estate, and hedge funds. Mr. Slingerland noted that the graphs presented were highly aggregated and that a more in-depth analysis was in the full report.
The long-term, or 20-year horizon, comparison was utilized to accommodate long-term investments. Over that period, the net return for the system was 8.3 percent and was equal to the peer return. For the U.S. Public it was 8.4 percent, showing that KRS is comparable to the universe of funds over the 20-year period. The benchmark return was above the peer median return and the U.S. Public median at 8.7 percent. This results in a net-value added being below both the U.S. Public median and peer median at negative 0.4 percent. Chairman Bowen asked how KRS would rate. Mr. Slingerland said that because his role was only to summarize what the data shows, he could not rate the system or advise the KRS board as to how to invest the systems’ funds. In summary, the net returns as well as the benchmark returns were below that of the U.S. Public funds over the 5-year term. Chairman Bowen asked for how many other funds CEM performs this kind of analysis. Mr. Slingerland stated that there are about 130 world-wide and 70-80 in the United States that they work with, but they collect data from approximately 125 U.S. Public funds. Senator Higdon asked for clarification about the net value added data. Mr. Slingerland said that if all paths of investments were followed, there was a slightly negative net value added because the value added was zero and some fees were paid resulting in a slight loss. So the active management over that period has not rewarded above and beyond the benchmark total level. Chairman Bowen then asked if the system was over-performing. Mr. Slingerland said no, KRS is not over-performing.
Mr. Slingerland continued to discuss cost benchmarking and placing it in context of risk and performance. KRS investment costs were $126.2 million or 81.6 basis points. This was the first year that CEM was also benchmarking the underlying performance fees and the top performance fees of hedge funds. Especially in fund-of-funds structures, several layers of fees are paid, including performance fees, and it is difficult to get an accurate picture of the total fees. Senator Higdon asked if the presented information represented actual fees or if there were still some undisclosed fees not shown. Mr. Peden stated that the information was an estimate made by CEM based on recorded fund-to-fund fees, but management fees embedded in that number are not recorded. However, it was noted the fees as presented are important to help understand implementation costs of fund-to-funds going forward. Mr. Peden added that the struggle to capture fees and report them is not unique to Kentucky, but is the same across the country. There has been a massive effort over the last two years to not only capture the fees, but to improve the accuracy of the data gathered about fees. Additionally, there is not any standardization at this time regarding fees reporting, but the process is underway to have some standards adopted in the next year.
Chairman Bowen asked if there is not even a definition of what a fee is, let alone a standardized method of reporting them. Mr. Slingerland said that there are currently efforts to develop a standard based on European requirements, where it is required to report all fees and underlying costs. That has driven CEM to look at the fees and work with organizations to develop a standardized way of reporting all fees. The problem is with the complicated structure of private assets and how to uncover those fees. For the KRS comparison, CEM tried to capture as much as possible in the best ways possible. Chairman Bowen asked, when the information is gathered, if it was a net figure of everything even without a standard definition of fees. Mr. Peden answered that the net asset value is net of all fees, and the return number is net of all fees, even when they are not broken into further categories. That has changed this year, and KRS continues to improve and streamline that process. Chairman Bowen asked is this was a standard exercise for all pensions, to net out fees. Mr. Peden stated that he was not sure if that was common practice for the rate of returns, but on these asset classes that are more difficult to disclose the fees, he said the practice was standard.
Mr. Slingerland clarified that it may not be possible to always know what the fees are, so CEM does not benchmark the performance fees for private assets. One reason is because there is not enough known about the funds or there is not high enough trust in the data. As the standards become more widespread, the data will be better and those kinds of fees can be included in the benchmarking process. It is possible to include fees from hedge funds, so those were included in the KRS report. Mr. Slingerland then showed that the costs for KRS are 6.9 basis points (bps) or $10,604,000 above the expected benchmark of 74.8 bps or roughly $115,000,000. This means that if the KRS portfolio was being managed by one of the peer group, it would cost less. There were two reasons found for why the costs were higher than the benchmark: 1) implementation style differences, which account for 5.1 bps above benchmark; and 2) paying more for similar services. Implementation style differences were driven primarily by KRS’s fund-to-fund hedge fund portfolio. Of the total assets KRS has invested in hedge funds, 95.3 percent is invested in a fund-of-funds structure, which has multiple layers of fees and is therefore a more expensive implementation style. Mr. Slingerland explained that the use of fund-of-fund hedge funds had a cost increasing effect of approximately $13 million dollars.
He went on to show how the costs of KRS compared to other pension systems in terms of asset classes and internal asset management. The cost effectiveness of KRS’s investments was then summarized in a cost effectiveness chart, a unique feature of CEM’s services. It compared the access cost, what KRS pays above benchmark, to the net value added, or value added after costs. This showed no correlation between paying more than benchmark and getting more value. The presentation concluded with a comparison of risk levels showing that KRS’s asset risk of 9.2 percent was below the peer median of 9.8 percent, and KRS’s asset-liability risk of 13.7 percent was also below the peer median of 14.2 percent. In an ideal world, the asset-liability risk would be zero because the assets would perfectly match the liabilities, but that is not the case for any fund. Therefore, this is a measure of what is paid, and KRS compares favorably with its peer group and is only slightly higher than U.S. Public funds.
Mr. Jefferson drew attention to the total cost and benchmark cost section of the full report, specifically the trend line of costs over the last five years, showing that costs have approximately doubled. He asked if that could be attributed to moving into a higher cost investment mix or if KRS is just better at capturing cost data now versus five years ago. Mr. Peden stated that it could be attributed to better capturing of fees as well as a move away from passive management, which was a big driver of costs. The introduction of hedge funds to the portfolio five years ago would also contribute to that trend. Mr. Jefferson then asked how KRS could be ranked regarding capturing fees compared to other systems. Mr. Slingerland stated that in his experience, systems that undergo the benchmarking process have a better understanding of fees and how they compare, which is a good step toward transparency. Additionally, many funds give less information and CEM works with them to get the detail that KRS provided initially, which allowed a very close approximation of true fees. That level of detail is only provided by about 60 percent of the systems benchmarked by CEM. Mr. Jefferson asked if fees aren’t reported by peer group funds if that metric might be biased downward in private assets. Mr. Slingerland said that all fees in private assets are down because there is no standardization and no pressure to have standardization. He added that there are funds that provide a great amount of detail, and CEM strives to bring any fund that reports below that average value up to that same level. Mr. Jefferson finally asked, based on experience, if there were firms that the Public Pension Oversight Board might work with to adopt a standard of reporting for all of the pension funds, including KTRS, within the state. Mr. Slingerland recommended Institutional Limited Partners Association (ILPA) as the best organization to work with, noting that this organization will be at the forefront of pushing for standardization. Specifically, ILPA has a template that funds can follow to achieve more standardization.
Senator Higdon asked about administrative costs and whether the category internal asset management costs was the same metric. Mr. Thielen stated that it was and that this study was one of three being performed as a result of discussions with the Board over the last year. The first was the actuarial audit performed by Segal Consulting and reported last meeting. This was the second, and the third will be a benchmarking study on administrative costs also performed by CEM Benchmarking Services. He added that most pension funds do not handle health insurance, and that CEM even restricts its database to pensions, making it difficult to find a way to separate health insurance management costs from pension management costs in order to provide a more accurate comparison with other pension systems. The process of doing that has begun and the final report should be available in the spring.
Representative Thompson asked why asset-liability risk percentage presented was not higher than the peer median because KRS has such a low funding ratio. Mr. Slingerland explained that as a result of the way the model is built to allow for comparison to other funds, CEM does not look at the funded status in that regard. The number is less important than how KRS ranks compared to others. Something that could increase asset-liability risk would be a contractual COLA because it must be adjusted every year for inflation, whereas using an ad hoc COLA results in reduced risk, as does any change to membership that results generally in a younger, active base.
Kentucky Teachers’ Retirement System Actuarial Audit
Ms. Kim Nicholl, Senior Vice President and Actuary, Segal Consulting, presented the results of the full scope actuarial audit of the Kentucky Teachers’ Retirement System (KTRS). She first noted the project scope was different from the audit performed on KRS and presented at an earlier meeting. This audit was a full scope audit that replicates in full the pension, retiree medical, and life insurance valuations detail. They also evaluated the data used in the valuation and the reports, assumptions and methods used, as well as gave an assessment of conclusions. Segal also reviewed the GASB 67 and 68 reports and the 2005-2010 experience study.
The data was comprehensive, thus allowing for an accurate match of all three valuations within a tolerable range. Minor differences in calculations for death benefits for a small group were seen, based on a 5-year final average salary instead of a 3-year final average salary, as well as for disability benefits for pre-2002 hires, but these differences had a negligible impact on liabilities. Segal found that the valuation reports comply with Actuarial Standards of Practice, but recommended adding an executive summary that highlights results of valuations, modifying the description of additional pension contributions needed to amortize the unfunded liability over 30 years, and clarifying certain information and assumptions. However, overall the assumptions and methods were found to be reasonable. For the next experience study, Segal recommended a study of individual salary increases by netting out actual inflation experienced over the study period, and for the post-retirement mortality assumption to consider adding at least a 10 percent margin for future mortality improvements or to use a fully generational table.
Ms. Nicholl explained that Segal has a national audit practice that performs three to four audits each year. Based on that experience, they found this audit to be very clean, and in fact it was one of the cleanest they have done. The comments were very peripheral as far as liabilities were concerned. KTRS should feel confident that the results of the valuations are reasonable, reflecting the methods, assumptions, and benefit structure of KTRS accurately.
Chairman Bowen asked for the difference between the level two audit performed on KRS, which was claimed to be adequate, and the level three audit performed on KTRS. He specifically asked what additional information was garnered from a level three audit. Ms. Nicholl stated that a level two audit requests sample life data which covers all the different benefit divisions, eligibility rules, and assumptions that will be applied to participants in the valuation. For KRS, Segal was able to match the sample people and determine that the data covered all the different benefit structures and the assumptions. The additional step performed for KTRS was to calculate the liabilities for every single member in the plan, not just the sample test lives, which determines the total liability of the plan. Chairman Bowen asked what the additional cost was for a level three audit over a level two. Ms. Nicholl answered that for both types of audits, the information is programmed into Segal’s systems, making the additional cost approximately $10,000 to $15,000. Chairman Bowen asked how much the total cost was for the KTRS audit. Ms. Nicholl answered that it was between $85,000 and $95,000. She added that the complexity of the plans also impacts the cost, and that KTRS is a more straightforward system than KRS, which makes it difficult to compare KTRS to KRS.
Senator Higdon asked what a sample life would look like, from beginning to end. Ms. Nicholl said that for a sample life, Segal takes an individual’s information relating to their pension benefit, such as retiree healthcare, life insurance, age, date of hire, and salary. They then apply any assumptions, which for an active member might be their expected final average salary at every year in the future, and discount any expected benefit payments based on the assumptions to determine total liability for that individual. For an active member it is not possible to state the exact benefit because it is unknown when they will retire; however, it is possible to know the value of the benefit based on the assumptions that are applied. For example, in a group of one thousand 60-year-old members, each one has a certain probability that they could retire. Ten percent of them could retire, but it wouldn’t be possible to know which 100 individuals that is; instead, the model assumes that 10 percent of each person retires. For retirees, it is known exactly what their benefit is, and an estimation of how long it will be paid can be made based upon mortality assumptions. Senator Higdon asked if Segal looked at how much was paid in employee and employer contributions and the investment returns those funds earned over the years of service, and specifically if that money was adequate to pay the benefit for a particular employee or if a liability was created above the actuarial estimation. Ms. Nicholl stated that it was not standard to do that type of calculation per individual, but instead the unfunded liability is calculated for the total plan.
Chairman Bowen stated that it was unclear why a level two audit was adequate for KRS, whereas a level three audit was necessary for KTRS. Ms. Nicholl answered that some pension plans request a full scope audit, while other plans, such as KRS, request a level two audit. If KRS had requested a level three audit, Segal would have performed that audit. It was not Segal’s decision to do one type of audit over the other for either plan. Chairman Bowen said he understood that, but he recalled asking what type of audit was necessary for KRS, and the level two audit was defended as adequate at the time. Ms. Nicholl stated that it was appropriate and that a level two audit would have probably been adequate for KTRS as well, but that KTRS requested a higher level of detail.
Kentucky Retirement Systems Investment Update
Mr. Bill Thielen, Executive Director, Kentucky Retirement Systems, and Mr. David Peden, Chief Investment Officer, Kentucky Retirement Systems, were present to speak regarding the investment status of the system. Mr. Peden stated that it had been a challenging start to the fiscal year, with the aggregate portfolio down 4.8 percent versus the benchmark that was down 4.45 percent. The only positive performers were private equity, real estate, and interest rate sensitive fixed income. All other asset classes were down, especially U.S. equity, non-U.S. equity, and emerging market equity. As an example of how volatile the markets had been, he stated that over 60 percent of the S&P 500 had either a 10 percent loss or a 10 percent gain for the calendar year to date. In August of 2014, for the prior 3-year period, the U.S. stock market had produced three times its long-run average and half its normal volatility, and he told the committee that it would at some point revert to the mean, which appeared to be beginning during the second half of this year. The only recourse in this market is sticking to asset allocation knowing that the long-run averages will come full circle. He concluded by adding that there were recent changes to asset allocation that will go into effect January 1, 2016, which were approved by the investment committee at their November meeting, and he said he would discuss those asset allocation changes at a future meeting.
Senator Higdon asked if reports that the unfunded liability was increasing by $7 million a day or $2 billion a year were accurate. Mr. Thielen said that was not correct. The unfunded liability is currently approximately $17.5 billion, and the system has on hand approximately $16.1 billion in assets. The unfunded liability has increased, but not at the rate stated. Senator Higdon then asked if there was any projection of what the ARC request would be this year. Mr. Thielen said that, yes, the final valuation had been complete and the final copy would be presented on December 3, 2015 to the KRS Board of Trustees to adopt contribution rates. For the KERS non-hazardous plan, the ARC is 47.28 percent of payroll, which is higher than the projected rate from several months ago before the final investment numbers were available. He estimated that the figure represented $157.8 million more than was paid this year to meet the ARC. So the total for the state, including all three state-funded plans, would be $880 million to fully fund the ARC. This represents a $160 million increase from the last year of the current biennium. Much of the increase is driven by the assumption changes made following the experience study performed over the 2008-2013 period.
Representative Yonts asked if there would be a recommendation made by KRS before the upcoming legislative session. Mr. Thielen stated that the KRS board is meeting to discuss the assumed rate of return, based on the asset allocation study. In light of that, they plan to make a recommendation regarding previously enacted opt-out provisions that will match the assumed rate of return. Representative Yonts asked if the system has regulatory authority to require contractors to pay into the system. Mr. Thielen said that they do not, but that the issue is something that should be addressed. Finally, Representative Yonts asked if the KRS board would send three recommendations for concrete changes that need to be accomplished in the upcoming legislative session. Chairman Bowen echoed Representative Yonts’ request for concrete changes that are creative and aggressive. Representative Thompson asked if the $160 million increase was required annually or if that was a total figure. Mr. Thielen stated that it was per year of the biennium, and that the calculation was based on a payroll growth assumption of 2 percent.
Senator Higdon made a statement regarding the recent pay raise given to the KRS executive director, which was discussed at length during the previous meeting. He stated that he was troubled by the fact that the KRS board did not conduct a comprehensive and good faith search for a new executive director. It was especially disappointing to him that not one of the candidates for that position was even interviewed by the KRS board. Because Mr. Thielen’s hiring was accompanied by a large salary increase following an executive search that was truncated at best, Senator Higdon requested that Mr. Thielen retire at the completion of the 2016 Regular Session as originally planned. He further requested that Mr. Thielen continue under his current contract through the session and cancel the higher salary contract that will become effective December 1, 2015. In past Public Pension Oversight Board meetings, Mr. Thielen cited his experience working with the legislature as a principal skill he brought to the position. However, Senator Higdon stated that the circumstances of Mr. Thielen’s rehiring had damaged his credibility with the General Assembly, and while it was ultimately the Board of Trustees that was responsible for conducting the executive director search, he would have expected that Mr. Thielen’s experience dealing with the legislature would prompt him to advise the KRS board to not rehire him with a large salary increase without a meaningful executive search. Under the circumstances, and in view of all the current challenges that face KRS and the Commonwealth, Senator Higdon believed that new leadership for KRS was needed. Consequently, he requested Mr. Thielen’s retirement following the conclusion of the 2016 session as was originally represented to the stakeholders. Senator Higdon stated his appreciation for Mr. Thielen’s past service, and that he took no pleasure in making this request, but that he believed it was in the best interest of the Commonwealth.
Mr. Thielen responded that he took issue with the assertion that the KRS board did not conduct a comprehensive search for a new director. Following the issuance of a request for proposals (RFP) for a consultant, there were only 5 or 6 responses and the KRS board hired a national firm with an extensive reputation for assisting public pension plans with executive searches across the country. The firm engaged in two months of soliciting applications, including over 800 direct contacts. The consultants advised the KRS board that this would be a difficult search in light of the changing administration, the complexity of the system, and the fact that the system has a funded status of 19 percent. The search firm concluded that it would be necessary for the minimum salary to be in the $200,000,000 to $230,000,000 range to attract any qualified candidates. There were 20 applications received after advertising for the position. One was withdrawn and six were eliminated based on a lack of qualifications. Five were classified as only minimally qualified, leaving only eight applicants, three of which were internal to KRS. One of the remaining five was interviewed four years ago, and that applicant recently left another pension plan under negative circumstances. This left only four applicants: a city comptroller, a state tax commissioner, an attorney with a research organization, and an operations manager for an economic development corporation. The search committee looked at those applicants and determined that interviews were not justified. The search committee then approached Mr. Thielen and asked if he would stay, which he had already assured them he would until a satisfactory replacement could be found. The salary and contract were suggested to him, and he agreed to the offer, with the understanding that if a suitable replacement could be found, he would leave immediately. His intention was always to retire, but he works for the KRS board and he will continue to do so until they are satisfied with the management and direction of the system.
Chairman Bowen stated that, of the twenty applicants, the fact that none were interviewed seemed inconceivable. He asked what excluded the internal applicants. Mr. Thielen stated that there are many challenges facing the system, and that he had vastly more experience than anyone on staff in working with the legislature. He also noted it would be possible to hire someone with much less experience, very likely with higher pay, but that person may not be able to help the KRS board. He reiterated that it was the board’s decision and that he would do whatever they asked of him. Representative Linder asked what the next step of the search would be, and whether the search continues or if it will be suspended until the end of the legislative session. Mr. Thielen said that he felt the executive search committee would continue to work, possibly working with the internal applicants in a mentoring type of program to train one of them to be a suitable candidate, but he did not want to speak for the board regarding specifics. Representative Linder asked if they might still be looking externally or raising the salary to increase the pool of applicants. Mr. Thielen stated that he did not feel they were doing so at this time. Representative Yonts requested clarification that the decision to rehire Mr. Thielen, with a salary increase, was not his, but had been made by the KRS board. Mr. Thielen stated that he could have declined the offer, but it was not his decision to not pursue any interviews during the search process, and he did not request the pay increase.
Kentucky Teachers’ Retirement System (KTRS) Investment Update
Mr. Beau Barnes, Deputy Executive Secretary of Operations and General Counsel, Kentucky Teachers’ Retirement System, discussed KTRS’s investments. He stated that the quarter ending September 30, 2015 was bad for equities, with the U.S. equity portfolio declining 8 percent and non-U.S. equity declining 13 percent. For the same quarter there were increases for real estate of 4.8 percent, for private equity of 3.6 percent, and for fixed income of 1.2 percent. Since then, the markets have rebounded and there should be improvement seen in the next quarter.
Mr. Jefferson asked why KTRS had shared their returns as gross of fees instead of net of fees as other systems had presented. Mr. Barnes stated that historically the fees KTRS had negotiated were extremely low, at approximately 25 basis points, which were among the lowest in the nation, and adding 25 basis points onto the reported fees would give a good approximation of what a net of fees performance number would be for the system. Mr. Barnes stated that LRC staff had requested reporting on a net of fees basis, which can be done, but because the fees had been so low, it was just historically not calculated that way. Mr. Jefferson asked if KTRS participated in fund-of-funds hedge funds or other investments where the fees were less transparent. Mr. Barnes answered that KTRS is underweight in regards to those asset classes, were not invested in hedge funds at all, and were only modestly invested in other alternative asset classes. Because of the negative cash flow position of the system, he did not foresee that changing in the near future. Additionally, in their annual financial report, KTRS discloses their fees specifically. He added that KTRS’s fees are widely recognized as being very low. Chairman Bowen stated that it would be nice for all of the pension plans to report in the same manner to the oversight board, if only to provide a measure of standardization in reporting among the state pension systems.
Representative Montell stated that 25 basis points did seem very low in light of the earlier reported benchmark costs of 74.8 basis points as an average for public pension plans. He asked what the current size of the systems assets were, and Mr. Barnes answered that it is approximately $18 billion currently being managed by KTRS. Representative Montell stated that indicated a huge savings over the average fees, and that there was a lot to be learned from how the system is being managed. Representative Simpson asked how the fees for other similar teacher retirement systems compared, and Mr. Barnes stated that KTRS has lower fees than most, which is largely attributable to asset allocation being predominately in stocks that have lower fees than other asset classes, like private equity.
Judicial Form Retirement System Investment Update
Ms. Donna Early, Executive Director, Judicial Form Retirement System, spoke regarding the Judicial Form Retirement System (JFRS) investments. She stated that the system handled the judicial and legislators retirement plans. Similarly to KTRS’s report, JFRS is invested 70 percent in equities and 30 percent in fixed income, meaning that the first quarter of the fiscal year showed a decline. However, she noted that trends were improving, with a gain of approximately 2 percent seen in the last month.
Chairman Bowen asked a question regarding transparency of the system. Ms. Early stated that the plan is governed by a board of trustees appointed by the legislature, the judiciary, and the Governor, and the board members take to heart all suggestions from the membership and the private sector because they are stewards of taxpayer money. In light of that and based on the number of statements of concern they have received, the board is looking at posting things to a website to improve transparency. Ms. Early stated that in her experience the system had always been very transparent and provided everything as requested, but she recognized there was always room to improve and the board was studying how to best address the issue. Chairman Bowen highlighted a sense of urgency regarding the matter, and he asked that Ms. Early communicate that urgency to the members of the board.
Representative Montell stated that the funding level of the legislative retirement plan had recently been brought to his attention. He asked what the contribution on the part of the state was to legislative retirement. Ms. Early answered that the actuarial valuation was only recently completed and will be presented, along with their actuarial audit, in a future meeting. She stated that the contribution rate was approximately 52 percent of payroll for legislative retirement and 43 percent of payroll for judicial retirement. She clarified that unlike other plans, it is an aggregate sum, taking into consideration all of the assumptions used by the system regarding salary and salary increases. Representative Montell stated he was shocked, and somewhat embarrassed, that the contribution was that high. He did not think it could be justified to pay a higher percent of payroll for legislative retirement than is paid for state employees. He had assumed that because so few legislators served 27 years and drew a full benefit, that it would be easier to keep the plan more fully funded than a typical plan. Ms. Early stated that the rates given were estimations for the upcoming biennium, and she noted that the most recent request was approximately 27 percent of payroll for the previous biennium, and that in some past years there was no contribution at all because the plan was adequately funded.
Auditor-elect Harmon asked why the assumed rate of return for JFRS was only 7 percent compared to an assumed rate of return of 7.5 percent for the other pension systems. Ms. Early answered that it was based on the recommendation of their actuary. She also stated that the system was on schedule for an audit by the Office of the Auditor of Public Accounts soon, and that she looked forward to working with the auditor-elect.
Representative Yonts reiterated Chairman Bowen’s sense of urgency regarding transparency of the system, meaning the need for a website, and noted it is a cause of concern among the public.
There were no further questions; however, Mr. Damien Stanton, a representative of SaveCERS.org and a constituent who would be retiring in 19 working days with 39 years of services in the County Employees Retirement System (CERS) Hazardous retirement plan requested an opportunity to address the board. He stated he worked for ten years as a police officer, with Senator Schickel, in the City of Florence, and that he currently serves as a deputy sheriff in Kenton County. He voiced concern about the KRS and referenced the testimony and questions regarding fees. Primarily, he was concerned that the issue had been addressed many times and yet the results remained the same. Mr. Stanton stated that in an effort to further investigate the matter, he had filed a lawsuit in Franklin Circuit Court, which he provided copies of to the members of the board. He explained that he was effectively antagonizing the system that was going to be providing his benefit in a short amount of time because of a desire to stop what he perceived as a bleeding of money from CERS. He noted that as a member of the system, he expected to receive the benefit he had worked for, and that he was concerned when the cost to secure benefits had driven down the funding level over the last fiscal year. He then referenced the KRS Comprehensive Annual Financial Report (CAFR), which previously featured the funding levels of the various pension systems. Based on the CAFR for the last fiscal year, the funding level for the CERS hazardous plan had declined from 59.8 percent to 58 percent and the KERS non-hazardous plan had declined from 21 percent to 19 percent. Mr. Stanton said that this has an impact on employees and employers. If businesses see this as a precursor to a reduction in the state’s bond rating, it could drive away industry. Lastly, he drew attention to a memorandum, found on KRS’s website, to Connie Davis, Director of Internal Audit, from Karen Roggenkamp, Chief Operations Officer, dated August 20, 2015, and with “FY 2015 Financial Highlights” as the subject. In this memorandum, a net loss of $363,219,320 is reported, as well as investment expenses of $80.41 million. He felt that fees that high should warrant looking into such high losses. Mr. Stanton concluded his remarks by asking that the systems please stop giving away pensions. Many individuals serve the city and counties of the Commonwealth and they deserve nothing less than they were promised. Chairman Bowen thanked Mr. Stanton for his testimony and said that the board has similar concerns.
Before adjourning, Senator Higdon commented on legislative retirement and House Bill 299 of the 2005 Regular Session, which added reciprocity measures to the statutes and he noted he filed a bill in 2010 to abolish those measures. That bill did not pass and every year new attempts are made to address reciprocity. He indicated that this coming session might be the appropriate time to finally pass measures to either abolish the reciprocity measures or allow for legislators to opt-out of the system.
Ms. Early returned to the table to clarify, because of the changes made by the hybrid cash balance plan, that there were two actuarial studies performed. The 52 percent contribution rate was correct, but it did not factor in whether it was total payroll of all legislators or the ones who only participate in either the hybrid cash balance plan or the defined benefit plan. She would further refine the information and bring it back to the board at a later meeting.
Chairman Bowen thanked her for the clarification and announced that the next Public Pension Oversight Board meeting would be December 17, 2015 at 12:00 p.m. He reminded the members of the board to be sure to submit anything for the final report to Brad Gross, Committee Staff Administrator, as soon as possible. With there being no further discussion, the meeting adjourned.