SENATE FINANCE COMMITTEE April 15, 2014 1:40 p.m. 1:40:58 PM CALL TO ORDER Co-Chair Kelly called the Senate Finance Committee meeting to order at 1:40 p.m. MEMBERS PRESENT Senator Pete Kelly, Co-Chair Senator Kevin Meyer, Co-Chair Senator Anna Fairclough, Vice-Chair Senator Click Bishop Senator Mike Dunleavy Senator Lyman Hoffman Senator Donny Olson MEMBERS ABSENT None ALSO PRESENT David Teal, Director, Legislative Finance Division; Deven Mitchell, Executive Director, Alaska Municipal Bond Bank Authority, Department of Revenue; Gary Bader, Chief Investment Officer, Treasury Division, Department of Revenue; Michael Barnhill, Deputy Commissioner, Department of Administration; John Boucher, Senior Economist, Office of Management and Budget, Office of the Governor. SUMMARY SB 220 PERS/TRS STATE CONTRIBUTIONS SB 220 was HEARD and HELD in committee for further consideration. SENATE BILL NO. 220 "An Act relating to additional state contributions to the teachers' defined benefit retirement plan and the public employees' defined benefit retirement plan; and providing for an effective date." 1:41:46 PM DAVID TEAL, DIRECTOR, LEGISLATIVE FINANCE DIVISION, communicated his intent to provide fiscal context to the governor's proposal on the Public Employees' Retirement System (PERS) and Teachers' Retirement System (TRS) unfunded liability. He addressed a PERS and TRS "Comparing Payment Options" spreadsheet (copy on file). The spreadsheet included a budget with zero growth, education, and a growth rate for Medicaid, which provided a conservative look at spending projections. Under the base status quo scenario the annual payments of approximately $700 million were expected to increase to roughly $1 billion by 2024. He discussed a flat $550 [million] figure related to the capital budget, which was currently undetermined. He relayed that the projected deficit was substantial. He stated that the projected deficit of $2 billion fell to under $1.5 billion and reached closer to $3 billion by the end of the forecast period. 1:44:03 PM Mr. Teal relayed that under the projected expenditures and revenue the state's $15 billion in reserves would vanish by 2024. He detailed that the specific situation had driven the concern about controlling retirement costs. He communicated that the governor's plan was a proposal to fix the retirement systems, which included a $3 billion infusion in FY 15; the infusion would reduce annual payments. He elaborated that annual payments were lower, but at the end of the payment period the reserve position had not been significantly altered. He stated that reserves were lower in every year until further out where under the governor's plan they would be in a better position than under the base scenario; however, unfortunately they were still negative. Mr. Teal addressed that the pay-as-you-go method was designed to increase the life of reserves and the flexibility of their use; the method made a difference of approximately one year or $2 billion in the state's reserves in 2024. The savings were sufficient to satisfy some who were very concerned about the flexibility and life of the state's reserves. However, he surmised that the $2 billion was insufficient to invest in the gas line, which had been the purpose of saving the reserves. The strategy did not work as well as the state had hoped, it moved cost to the future, therefore costs were higher long-term. Additionally, it was a non-standard actuarial and credit rater's approach. He noted that the method did what it was designed to do; however, no one had liked it. Mr. Teal communicated that the methods acceptable to actuaries and credit raters could not significantly improve the state's reserve health in the near-term. The reserve health could be improved, but not with changes to the retirement system. Additionally, the governor's plan focused on the long-term health of the retirement system. He relayed that neither the governor's plan nor the pay-as- you-go method significantly increased reserves. 1:48:26 PM Mr. Teal shared that improvement to reserve health would require revenue increases or expenditure reductions. For example, an annual reduction to the capital budget of $400 million had a large impact on long-term reserve funds. Co-Chair Kelly clarified that Mr. Teal was only providing the capital budget as an example and was not proposing the cuts. Mr. Teal responded in the affirmative. He elaborated that making a reduction anywhere in the budget would work in the same way. He believed a $150 million capital budget would be less than required to make the state's federal match. He surmised that it was not something the legislature would want to do. Mr. Teal continued that if the state became convinced that the retirement system would use much of its reserves it may want to shift its focus to a long-term fix instead of preserving reserves in the short-term. For example, the governor's plan used a $3 billion cash infusion plus an annual payment of $500 million for nine years; the total expenditures on retirement under the governor's plan were $7.5 billion, which equated to half of the state's reserves. He elaborated that in another ten years the state's reserves would be depleted by another $5 billion if it continued with the same plan. He stressed that the retirement plan was capable of consuming the entire Constitutional Budget Reserve (CBR) over the upcoming 20 years. He stated that if spending could be reduced in other areas, the governor's plan would still leave the state with zero reserves in the future. He discussed a long-term look that constituted the point in time the retirement system would be fully funded (the actual end of the retirement system would occur in the 2070s or later). He stated that under any acceptable actuarial method the funding ratio was 100 percent; once at 100 percent, contributions went down to zero and the system would coast on its earnings. He acknowledged that the scenario had its risks, but future legislatures could address the issue in the 2030s and beyond. He stated that there was nothing the state could do about it at present because the models were projections and not predictions. He stated that the models were not accurate that far in the future. Mr. Teal addressed a line graph titled "Cumulative Costs of Options to Eliminate TRS Unfunded liability ($ millions)." He relayed that under the base scenario the state would make annual state assistance payments; the costs were cumulative. He remarked that the graph for PERS looked similar. He shared that under the base scenario the cost of paying off the unfunded TRS liability was approximately $8.5 billion. Under the governor's plan the slope was lower (as indicated on the graph in blue); at some point around 2023 the base and governor's scenarios crossed where the scenarios had spent approximately the same amount of money. Subsequent to the plans' crossover the governor's proposal became less expensive in the long-term and saved approximately $500 million. He discussed that cheaper retirement costs meant lower deficits in the future. He provided further remarks specific to the graph. He noted the graph's focus on long-term cost. 1:55:14 PM Mr. Teal continued to address the graph. He asked whether flexibility was needed in state reserves if it was committed to paying retirement and to a healthy retirement system. He noted that holding money in reserves such as the CBR did not really help. He stressed that it did not matter whether the reserves had been used for a cash infusion or annual payments; the same amount would be used by 2024 in either scenario. He remarked that his statements were applicable to any actuarially acceptable plans. The graph included the pay-as-you-go scenario. The option achieved its purpose of spending less in the early years through 2024. The state would pay for using the method because in the long-term it would keep paying; in the very long-term the plan switched from being the cheapest to the most expensive. He reiterated that the plan's purpose was to preserve reserves; it did not really work to use the method unless the state considered $2 billion as working. He reminded the committee that the $2 billion in savings would have been consumed faster under the pay-as-you-go plan. Mr. Teal noted that every other actuarial acceptable plans had things in common. Each of the plans crossed the baseline scenario around 2023 or 2024; under any of the scenarios the total annual cost plus a cash infusion was about the same. He continued that there was an area that every plan would "rotate around." He elaborated that a plan that was higher in the early years would be lower in the later years. He referenced the actuary Buck Consultants that had relayed there was always a choice to pay upfront or to pay more at a later time. He stated that paying more upfront could significantly reduce the state's total costs. 1:58:21 PM Mr. Teal calculated changes in the graph for the committee. He discussed undiscounted dollars. He stated that with a discount of 8 percent, which is what the state claimed it would earn, all of the plans would cost approximately $4.5 billion in the long-term. He noted that the cost should not be a surprise because it was what the state owed in unfunded liability. He observed that every plan had to pay off the benefits; it was just a matter of when the state chose to pay them. Mr. Teal altered the graph's discount rate to match the rate of inflation and noted that it did not look significantly different than a non-discounted rate. Under a zero rate the graph lines were further apart and easier to read. He explained that all of the plans were within about $500 million at the 2023/2024 crossover point. He addressed a level percent of pay method. He shared that the method was slightly cheaper in the early years than other options using a $2 billion cash infusion; however, in later years the cost was identical to the governor's plan. Under a 30- year amortization the payments would be extended significantly. Payments went flat in the early 2030s under the base scenario, the mid-2030s under the governor's plan; the 30-year amortization would extend payments out beyond 2040. The level-dollar method added a little more money in early years and reduced total costs. The dotted red line showed a $3 billion cash infusion with an annual fixed payment of $151 million; the method would cost slightly less than other $3 billion cash infusion plans. The level percent method would cost slightly less; level dollar would cost the state $5.5 billion at present instead of $8.5 billion to pay off the unfunded liability. 2:02:32 PM Mr. Teal addressed the state's commitment to funding the retirement system. He questioned why the state would choose any method but the cheapest one if it was committed to paying the benefits. He addressed reasons the state may not pick the cheapest option. First, the state did not want to pay off the entire unfunded liability; it could pay $4 billion, which would be a flat line. However, the state may end up with more money than it needed if it earned over the required 8 percent. He stated that there was no reason to immediately achieve a system that was 100 percent healthy. Second, reserves outside the retirement system had a clear and definite purpose, which allowed the state to reduce spending gradually. He detailed that the funds would not be retrievable once they were infused into the retirement account. He elaborated, that under the scenario, in the event of a lack of reserves, the legislature may be forced to take actions in the operating or capital budgets, the Permanent Fund Dividend, or to enact taxes. Third, raters looked at unfunded liability and reserves. He did not know which was more important; according to raters each item was worth approximately 10 percent of the state's credit rating. Mr. Teal discussed the governor's fixed payment proposal that was "a bit nonstandard" (not in the sense that it was actuarial unsound; it did pay the unfunded liability and was a method used by California). However, bond raters had their own models; some explanation would be needed about the state's use of the non-standard method. He discussed varying cash infusion amounts and relayed that the more cash put in upfront meant costs would be cheaper in the long run. However, the issue needed to be balanced against the state's need for reserves. He stated that the existing law did not need to be changed, it currently worked well; however, he suggested that the legislature change the law. He detailed that the change followed the advice of National Conference of State Legislatures (NCSL) experts and other national experts that advised putting the actuarial method and plan into statute because in the absence of Governmental Accounting Standards Board (GASB) funding guidance there was no plan to follow. He elaborated that putting the plan in statute would allow the state to determine whether it was following the plan and paying off the unfunded liability in the expected manner. He explained that the actuarial method (level dollar or level percent) could be put in the amortization term; the real decisions would come down to fiscal notes, and the method and term of amortization. 2:07:15 PM Mr. Teal addressed fiscal notes. He discussed that TRS was in worse shape than PERS; its funding ratio was approximately 53 percent, which was in the "danger zone" in terms of anyone comparing the health of retirement systems. For each $1 billion contributed to TRS, the funding ratio would increase by approximately 10 points. He detailed that a $1 billion cash infusion would increase the number to 63 percent, whereas $2 billion would increase the figure to 73 percent. A $3 billion infusion would bring the figure up to 80 percent, which was out of the danger zone; however, 70 percent was not bad. He communicated that because TRS was a single-payer it was a better investment than PERS. He explained that the state would pay for TRS either through the school districts or via state assistance; no one else made TRS contributions. Mr. Teal relayed that PERS was currently at a 63 percent funding ratio; however, it was a bigger system and had a larger unfunded liability. He elaborated that an infusion of $1 billion would increase the ratio to 68 percent; a $2 billion infusion would be required to obtain a ratio above 70 percent. He addressed that a portion of the cash essentially constituted revenue sharing or municipal assistance. He advised the committee to consider that municipalities would have balance sheet issues because of GASB rules; municipalities were required to report their unfunded liability on their balance sheets, which would make some municipalities look "pretty horrible" in terms of their financial standing. In order to pay off enough of the PERS liability the state would need to contribute several billion dollars. He was surprised by data he had received that would turn off state assistance sooner than he had thought; however, the unfunded liability would not be reduced any sooner (the rate would drop to 22 percent due to cash infusions). He expounded that the state would not have to pay, but the unfunded liability would still exist for the municipalities and the state. Vice-Chair Fairclough asked about the necessary cash infusion for PERS and TRS to reach the 22 percent figure. Mr. Teal responded that a $3 billion cash infusion would cut state assistance to zero after 2018. Vice-Chair Fairclough asked if the infusion would be directly into PERS. Mr. Teal replied in the affirmative. He detailed that if the state made a cash payment of $2 billion, state assistance would continue through 2038, which was more what he would expect. Based on the information he believed a $3 billion infusion into PERS would be more than the state would want to contribute; $2 billion would be fine, but more than $1.5 billion may not be necessary, which would achieve a funding ratio in the mid-70 percent range. Co-Chair Kelly asked for verification that an infusion of $1.5 billion to PERS and $2 billion into TRS would put the retirement system in the healthy zone from an outside investor perspective. Mr. Teal replied in the affirmative. He elaborated that a $3 billion cash infusion split slightly heavier towards TRS would work fine. Exactly how much the state wanted to infuse depended on its view of the tradeoff; it could pay more upfront or pay more later. He stated that the same tradeoff was achieved on the state's choice of assumptions (i.e. level dollar, fixed payment, or level percent of pay); the tradeoff was not as critical, but it was the payoff between paying upfront versus paying later. He stated that it was not possible to wish the debt away. There were only two ways to eliminate the unfunded liability: 1) to pay it off; or 2) through a good investment climate that would allow the state to partially earn its way out of the system. He recommended not going above 80 percent with a cash infusion because the amount would provide the state with enough assets to help earn its way out without danger of going too high or the need for a reserve account on the side. 2:13:57 PM Vice-Chair Fairclough agreed with most of Mr. Teal's presentation. She asked for verification that the state would receive the best bang the state would get for its buck was about 10 percent. Co-Chair Kelly asked for clarification on the ratio under discussion. Mr. Teal replied that he was referring to the credit raters' criteria and not a ratio. Earlier in the year the Department of Revenue (DOR) had provided data to the committee listing the criteria. He explained that the health of the state's retirement systems accounted for approximately 10 percent, which was the most significant black mark on its credit rating. Other items, including reserves (e.g. CBR) accounted for another 10 percent of the state's rating. He noted that the state had a "huge" CBR balance and guessed that the raters would view the state as being in great shape if it was not facing the large deficits. He detailed that credit raters looked at a number of things. He could not say whether it was better to switch reserves from the CBR to the retirement system; the items were given approximately the same weight by credit raters. He added that there were some rough rules of thumb for retirement systems; a system was in the danger zone if it was under 60 percent funded, whereas a system that was over 80 percent funded was not considered in danger of going broke. Raters liked to see a ratio of 100 percent, but understood that the ratio depended on economic conditions. There were not many systems funded close to 100 percent due to market losses. 2:16:49 PM Vice-Chair Fairclough referred to a presentation by the administration earlier in the week. She thought the administration had relayed that debt counted for 20 percent of the rating agencies calculations and that 30 percent of the state's AAA rating was related to the state's reserves and no tax. She was trying to reconcile the figures with Mr. Teal's information. She was interested in the impact. 2:18:36 PM AT EASE 2:22:41 PM RECONVENED Vice-Chair Fairclough reiterated her prior question related to criteria used to review the state's bond rating at a national level. DEVEN MITCHELL, EXECUTIVE DIRECTOR, ALASKA MUNICIPAL BOND BANK AUTHORITY, DEPARTMENT OF REVENUE, responded that the three independent rating agencies that reviewed the state's credit each had different evaluation criteria. He did not know all of the criteria. He referred to a prior presentation he had provided to the committee where he had cited Standard and Poor's; 20 percent of its weighting applied to debt including both pension liability and other government obligations. He believed the criteria provided to the committee by DOR Commissioner Angela Rodell applied to a different rating agency (potentially Moody's). Measurement of financial strength was made up by a combination of a number of factors including unfunded liability, debt, and fiscal practices. Vice-Chair Fairclough asked for verification that the state was taking a conservative approach in its financial analysis related to the 20 percent. She surmised the information provided by DOR earlier in the week provided a range between rating agencies (i.e. 20 and 30 percent) and did not include the numbers together. Mr. Mitchell answered in the affirmative; 20 percent fell under one scenario and 30 percent fell under another [Note: the answer to Vice-Chair Fairclough's question was modified during the meeting at 2:31:53 pm]. 2:26:21 PM Senator Bishop thanked Mr. Mitchell and remarked that his testimony matched prior testimony from DOR. Vice-Chair Fairclough spoke to the state's consideration of the unfunded liability and balancing a cash infusion and reserves. She asked Mr. Teal and the administration to address the Alaska Retirement Management Board (ARMB) returns and PERS/TRS accounts compared to the CBR. She wondered where the money would serve Alaskans the best. GARY BADER, CHIEF INVESTMENT OFFICER, TREASURY DIVISION DEPARTMENT OF REVENUE, stated that no one knew for certain what the future holds, but historical analysis had been done on a blended rate of the CBR (rates of return of the main and sub accounts). Analysis showed that over long periods of time, PERS and TRS along with other long-term funds such as the Alaska Permanent Fund outperformed the blended rate of return; therefore, the department believed the money was better placed in the long-term PERS/TRS accounts than in the CBR. He relayed that the CBR subaccount had a better rate of return than the blended rate; however, because of the necessary liquidity of CBR investments, the department still believed that over the long-term the money was better placed in PERS/TRS. Mr. Teal agreed with Mr. Bader's explanation. 2:29:18 PM AT EASE 2:31:53 PM RECONVENED Co-Chair Kelly asked Vice-Chair Fairclough to discuss her question and answer that had occurred during the recent at ease. Vice-Chair Fairclough restated her prior question related to criteria used to evaluate the state's credit rating. She asked if the impact was 20 percent or if it went up to 50 percent. She stated that 50 percent of the health of Alaska's credit rating was in play related to examples provided during the meeting. She stated that the figures were based 10 percent on revenues, 10 percent on balance reserves, 10 percent on liquidity, 10 percent on bond debt, and 10 percent on adjusted net pension liabilities, which totaled 50 percent. She observed that movement inside the 50 percent was a balancing issue. She clarified that individual factors could be up to 50 percent of the state's credit rating. Co-Chair Kelly observed that each one of the components were also individually weighted. Vice-Chair Fairclough agreed. Mr. Mitchell agreed. He had misunderstood the previous question and thought the conversation had been about comparing one rating agency's analytic process to another rating agency process. He believed that the amount was at least [50 percent]. He stated that Alaska had a somewhat unique credit that did not always fit in the box used for cross comparison of credits. Subsequently, some of the state's strongest criteria overbalanced slightly. 2:34:50 PM Co-Chair Kelly asked the administration to provide any further remarks on the discussion. MICHAEL BARNHILL, DEPUTY COMMISSIONER, DEPARTMENT OF ADMINISTRATION, appreciated Mr. Teal's presentation. He observed that the presentation had highlighted a number of factors that the administration spent significant time on during the crafting of the governor's proposal. He believed one of the most pertinent factors was the affordability to the state's General Fund in the near, mid, and long-term. He communicated that some proposals were more affordable in the long-term while others were more affordable in the short-term. He remarked that the curves tended to converge in later years. He stated that it was important to balance and look separately at the short, mid, and long-term. He elaborated that the state needed to focus on the all-in costs to the state. Mr. Barnhill spoke to budget certainty and relayed that a switch to a fixed contribution system under the governor's proposal created at least near-term budget certainty in the fixed payment, which was important. The issue was particularly important as the state looked at using the next five years as a transition time towards a potential gas pipeline investment. Additionally, the plan acceptability to stakeholders was also important. There was a significant array of stakeholders (almost every Alaskan was a stakeholder in some way). He discussed that the relative acceptability of the various proposals fell far short for some stakeholders. The state was looking for a proposal that was more than marginally acceptable to most stakeholders; the administration believed the governor's plan fell under this category. Finally, the state was looking at the various risks to beneficiaries that the various proposals presented. The administration believed the governor's plan decreased risks to beneficiaries in some fashion, which was important to beneficiaries and balancing stakeholder interests. He acknowledged that all of the plan proposals had pros and cons; ultimately the administration was looking for a fair and acceptable balance. 2:38:17 PM AT EASE 2:40:02 PM RECONVENED JOHN BOUCHER, SENIOR ECONOMIST, OFFICE OF MANAGEMENT AND BUDGET, OFFICE OF THE GOVERNOR, thanked the committee and Mr. Teal for their work on the issue. He remarked that Mr. Teal had highlighted some of the challenges the state faced for the next ten years. He did not want to give the committee the impression that the administration was unaware of the problems and that it would happily spend its way out of reserves for the next ten years. He pointed out that the administration had demonstrated over the last ten years that it was serious about curbing the operating budget. The administration was working to address some of the large issues including the retirement system and Medicaid spending. The administration appreciated that the issue was a balancing act and looked forward to working in partnership with the legislature during deliberations. He thanked the committee for its work. Vice-Chair Fairclough remarked that the chairman had been good about having "kitchen cabinet" conversations on big issues. She noted that Mr. Teal had pointed to some considerations. She wondered what 20 to 30 years meant to local municipalities versus the State of Alaska. She discussed Mr. Teal's presentation to the committee on the level dollar method (to frontload the system). She addressed where the state would start if it could not fund $11.9 billion at present. She surmised that 70 percent was the first step. She was interested in hearing all perspectives. She remarked that when the unfunded liability was spread to local communities some would experience significant debt. Co-Chair Kelly asked Mr. Teal to address what the options looked like with level percent of pay versus level dollar and 25 versus 30 years. 2:44:50 PM Mr. Teal replied that the concepts were not easily grasped and addressed the line graph titled "Cumulative Costs of Options to Eliminate TRS Unfunded liability ($ millions). He spoke to a $2 billion cash infusion and level percent of pay (green dashed line). The line started lower and crossed over with all of the other options around 2023/2024; beyond that point level percent of pay meant that cost increased as payroll increased over time. He explained that because the option cost less upfront, it ended up costing more in the long-run. Amortizing over 30 years was slightly more expensive upfront than level percent of pay, but it extended the payments for a longer timeframe. He elaborated that the cost would continue to climb (similar to pay-as- you-go). He advised that it was a tradeoff between near and long-term. Mr. Teal used a home mortgage as an example and explained that a person preferring a 15-year mortgage would choose the level dollar method because it paid the debt off faster. Level dollar was more like a mortgage; a constant amount was paid, which was similar to the governor's plan that would pay a fixed $343 million per year. He pointed out that level dollar would have a 25-year amortization with a cash infusion of $2 billion and an annual cost of approximately $196 million, which would extend payments by two years and would vary slightly as opposed to the governor's fixed payment plan. Mr. Teal relayed that as with a home mortgage, the higher the down payment the lower the down payments; additionally, a 30-year mortgage would have lower payments, but would cost more in the end. He communicated that the three "levers" included how much cash would be put down upfront, what the desired annual payment would be, and the desired length of time for making payments. He discussed the balancing act and considering whether the preference was flexible reserves or having reserves tied up in retirement. He stated that tying them up in retirement was great because the system would be healthier and interest earnings were 8 percent instead of 4 percent. However, he asked how healthy the system had to be. He elaborated that credit raters would rate the health of the system based on the retirement system funding percentage; if the system was funded at 100 percent the ratings agencies may give the state 10 points, whereas, if the system was funded at 80 percent the agencies may give the state 9 points (towards their rating scale). He noted that the point system was not directly proportional. He communicated that if the retirement system was below 50 percent funded, the state's credit rating would be downgraded for an unhealthy system. Mr. Teal referred back to his presentation and relayed that level percent of pay started out with lower payments, but as payroll climbed, the option became more expensive than level dollar. He stated that it was a choice of methods and the methods made a difference. With level percent of pay and a $2 billion cash infusion the state would pay $8 billion. With the level dollar amortization method the state would pay $6.8 billion. 2:50:29 PM Co-Chair Kelly noted that the second option was fine if there were no other goals in mind; however, the state did have other goals. He observed that the state would not make substantial gain anywhere by 2024; however, it could hold on to some of the money longer within the window, which was the only reason he liked level percent of pay better than the level dollar method. He did not want the cash infusion to be "massive." He believed the state should bring the retirement system up into the healthy range (somewhere in the 70 percent range). Additionally, he liked weighting the funding heavier towards TRS because the state was responsible for all of the payments. He liked level percent of pay because it provided some flexibility between the present day and 2024 where the state was trying to get more revenue. Mr. Teal responded that regardless of the method used, the state was always allowed to put in more money if it chose. Senator Bishop referred to Mr. Teal's testimony that NCSL recommended putting the payment plan in statute. He wanted to keep the state's bond rating health. He discussed testimony from Mr. Teal, the administration, and others on reasons to keep the system healthy. He agreed that money applied to the retirement system would not be lost on Alaska's economy. He stressed that the money would be spent in-state, which would keep the state's economy turning. 2:53:42 PM Co-Chair Kelly thought it was important to discuss the impact of GASB changes on municipalities. He asked Mr. Barnhill to address the subject. He discussed the committee discussion on a $2 billion cash infusion and paying more into TRS than PERS. He understood that the governor preferred another route. He asked about impacts to municipalities. Mr. Barnhill replied that GASB had adopted two GASB statements. He detailed that GASB 67 became effective in FY 14 and required all public pension plans with multiple employers to allocate the unfunded liability or net pension liability amongst the employers. In FY 15, all public employers who participate in a pension plan would be required to include the allocated net pension liability on their balance sheets. Co-Chair Kelly asked for verification that the municipalities would be required to carry the unfunded liability on their books even if the state was carrying a portion. He asked for verification that the state would also be required to carry the liability on its books. Mr. Barnhill responded that there was a special funding situation. He elaborated that in a special funding situation where a third party paid on behalf of a participating employer and the third party entity was "legally responsible" for the underlying obligation, then the third party was supposed to pick up the payments attributable to the net pension liability and include the liability on its books. He noted that there was an open question under GASB 68 whether the State of Alaska fell into the category of legally responsible. In FY 15 the employer contribution rate would be 44 percent; the employer contribution rate cap was 22 percent. Therefore, under the current methodolgy the state would pay the difference between 44 and 22 percent. He elaborated that if the state fell into the category of legally responsible it would pick up the net pension liability attributable to the 22 percent payment and would include it on its books. 2:56:39 PM Co-Chair Kelly asked if municipalities would be subsequently relieved of the obligation. Mr. Barnhill replied that the municipalities would be relieved of putting the liability on their books in the given year. He added that it was open question; in Alaska there was a prohibition against dedicated funds; therefore, the Department of Law (DOL) had informally advised that the statute providing for the funding was not enforceable (it was subject to appropriation). He equated the situation to a person's parents making a mortgage payment on their behalf for a five-year period (without signing the note). The question became whether the payment on the person's behalf made the parents obligors under the note. He stated that legally the parents would not be responsible. The administration had been informed that GASB intended for entities like the state to pick up the special funding on their books. He did not believe lawyers and accountants had been fully engaged on the drafting of the GASB statement. He believed there was a plausible argument that the state was not legally responsible for the payments. He concluded that it remained an item of discussion within the administration with DOL. Co-Chair Kelly asked about the impact on municipalities where nothing had changed, but they were required to carry the liability on their books. Mr. Barnhill clarified that GASB 67 and 68 only applied to pension liability, not to health liability. Of the $11.9 billion unfunded liability about $3.8 billion was health liability. Additionally, GASB rules did not allow actuarial smoothing; therefore, market value of assets would be used. Co-Chair Kelly asked for an explanation of actuarial smoothing. Mr. Barnhill replied that actuaries put 20 percent of any gains and losses into the actuarial value of assets and smoothed the gains and losses over a five-year time period. He added that under the GASB method all of the gains and losses were reflected in the year they occurred; therefore, it was a more volatile asset figure. 2:59:40 PM Co-Chair Kelly asked what happened to municipalities picking up the pension obligation. Mr. Barnhill used Anchorage as an example. Factoring in the amount the municipality contributed to PERS and the amount of its payroll (Anchorage's payroll divided by total payroll), the city was at approximately 8 percent. Therefore, 8 percent of the total PERS pension liability would be allocated to Anchorage for inclusion on its balance sheet. Co-Chair Kelly asked what would happen next. Mr. Barnhill answered that Anchorage had multiple balance sheets (i.e. utilities and various funds); he did not know how it would allocate the liability amongst its balance sheets. Co-Chair Kelly remarked that the GASB rule impact was the immediate consequence if the state was considering weighting a cash infusion to TRS. He wondered about the actual financial implications other than "wailing and gnashing of teeth." He asked what checks would be written as a result. Mr. Barnhill replied that there had been wailing and gnashing of teeth because the prospect of taking a meaningful percentage of the significant unfunded liability was not a welcome one for any of the municipalities. He addressed allocating between PERS and TRS. His understanding was that the requirement would apply with TRS as well; many municipalities consolidated their school district balance sheets on the municipality balance sheet. Presumably, those municipalities may be indifferent on whether funding was weighted to TRS or PERS. Co-Chair Kelly directed the question to Mr. Teal for further detail. He was interested to know the impact on municipalities required to carry the liability on their balance sheets. He stated that no one was anticipating that the state would not pay the check. He wondered what would happen to the municipality in actual "check writing consequences." Mr. Teal answered that there would be very little impact in terms of check writing costs. However, the change would impact the municipalities' credit rating, which could be a big issue. Co-Chair Kelly observed that the issue would result in check writing costs later on. Mr. Teal agreed. He added that each employer could be allocated a share of the liability. For example, if Anchorage's share was 8 percent, it would be allocated $657 million of the liability; it would carry approximately 60 percent of the figure on its balance sheet because nearly 40 percent would be healthcare cost. He communicated that the state's law allowed any employer (e.g. the Municipality of Anchorage) to make and credit an extra payment to its own account. He elaborated that the municipality could pay down its unfunded liability to reduce its rate in the future. However, he could not imagine that occurring given the state's payments on behalf of municipalities and the cash flow of municipalities. 3:04:14 PM Mr. Teal continued to address Co-Chair Kelly's question. He stated that there were few (if any) municipalities that could come up with the funds to pay off their unfunded liability. For example, Juneau could not come up with $132 million to pay off its unfunded liability any more than Anchorage could come up with $650 million. He detailed that the state's share of the $8 billion liability was about $4 billion and the University of Alaska's share was approximately $500 million. He stated that PERS was also big in school districts; since the state paid school district PERS as part of the state formula it was actually responsible for an additional 15 percent of the liability. He relayed that only 22 percent of the unfunded liability would fall to municipalities; however, those numbers were "crushing." He noted that carrying the money on their books did not mean municipalities would have to write larger payroll checks or pay more money in contributions to the state, but if they borrowed money, their credit rating would be downgraded and borrowing at low costs would not be possible unless they used the Alaska Municipal Bond Bank, which used different credit. Senator Bishop referred to GASB rules 67 and 68. He discussed the 22 and 44 percent responsibilities. He wondered if the state should wait to take action until the GASB rules were implemented. He surmised that the state could end up having a larger PERS liability than the municipalities. Co-Chair Kelly asked for verification that the state already carried the liability on its balance sheet. He asked for confirmation that the state's balance sheet could not get worse under the scenario described by Senator Bishop. Mr. Barnhill did not believe that anyone had carried the actual unfunded liability on their balance sheet; no one wanted to do it. He stated that any deposit to the PERS Trust Fund would serve to reduce the total unfunded liability, which would reduce the amount allocated to municipal balance sheets regardless of any special funding. 3:07:45 PM Senator Bishop focused on the ratings agencies and the state's credit rating. He wanted to ensure that in two years' time the state would not determine it should have made a larger cash infusion due to GASB 67 or 68. Mr. Teal addressed the possibility of GASB impacting the state in unexpected ways. He noted if the state paid $2 billion towards PERS (one quarter of the total liability), it would reduce Anchorage's share by one quarter (approximately $150 million, except that health was included). He stated that if GASB made a decision that made municipalities look bad the state could make another cash infusion of $1 billion in the following year or two. He believed that the current proposals did not represent a complete payoff of the unfunded liability. He did not believe a complete payoff was a good approach. He referred to the administration's testimony that the goal was finding a "sweet spot"; a balance between paying the liability off upfront, helping municipalities, maintaining reserves, and a healthy retirement system. There was also a balance between paying more upfront or more later on. He observed that all of the decisions meant the legislature would have a difficult fiscal note. He referred to a bill a few years earlier that would have addressed the liability; some had recommended paying down the liability upfront in order to avoid dealing with it during a time of fiscal deficit. 3:10:53 PM Mr. Teal relayed that the bill had not moved forward. Subsequently, the issue had not gone away and had become increasingly difficult due to deficits. However, the responsibility of the debt was not relieved by the situation and it did not change the balance point. Vice-Chair Fairclough referred to Mr. Teal's testimony that a 50 percent funded retirement plan would be downgraded and considered unhealthy. She stated that TRS was currently funded at 52.1 percent. She agreed that the plan absolutely needed a cash infusion. She addressed the level dollar method versus a level percent method. She believed Mr. Teal had communicated that level dollar would frontload the system and let the cash work over time, whereas level percent would meet the state's actuarial obligations. She asked if her remarks were accurate. Mr. Teal agreed. Vice-Chair Fairclough spoke to the difference between level dollar and level percent methods. She used an example of paying interest on a personal credit card as level percent and making payment on principal as level dollar. She reasoned that credit card debt would be paid off faster if payments were not limited to the interest owed. Mr. Teal agreed. He elaborated that even under the level percent method the debt would be paid off. He referred to a spreadsheet titled "TRS Options: Undiscounted Annual State Assistance"; the spreadsheet included a $2 billion cash infusion and columns for level percent of pay and level dollar. The state would pay less with level percent in the early years ($170 million versus $220 million), but in the long-term more money would be paid for a longer period of time. He stated that the level dollar method would cost the state $6.8 billion whereas, level percent would cost $7.9 billion. He added that the different method would not make the payments easier. 3:14:58 PM Vice-Chair Fairclough wondered if the time period was 20, 25, or 30-years. She noted that financing a car over a longer period would make payments easier to make, but more would be paid in the long-term. Mr. Teal replied that the spreadsheet included two columns with a 25-year amortization. The column on the right used a 30-year amortization; the payments were lower, but payments ended in 2042 instead of 2036. The cost was $7.2 billion, which fell roughly in between the other two methods. Vice-Chair Fairclough wondered whether the current payment plan had been started on a 20, 25, or 30-year period. Mr. Teal answered that the current plan used a 25-year period. He added that the law allowed for a period of up to 30 years. Vice-Chair Fairclough compared the possible solutions to refinancing debt, which would be carried for a longer period of time or shortening debt to the 20-year time period to pay it off in a reasonable timeframe. She was concerned about how paying the debt off in a longer timeframe would negatively impact municipalities. She reasoned an extension could cost municipalities hundreds of millions of dollars. She asked if her analysis was fair. Mr. Teal responded in the affirmative. 3:17:35 PM Co-Chair Kelly agreed with remarks made by Vice-Chair Fairclough. However, he noted that the state did not currently have reliable revenue or income. The state did not know what the future would hold, which was vastly different than the average household. He reasoned that the decision to make longer or shorter credit card payments would be balanced with the need to save reserves for items like a new car, college education, healthcare, and other. He did not know the state had the luxury to save money for the distant long-term when it had goals of building a gas line to access new revenues. He looked at the 20-year versus 30-year payments. Mr. Teal replied that in 2021 the payment would be $198 million for a 20-year plan versus $182 million for the 30- year plan. The cost continued at $196 million under the 25- year plan [beginning in 2024] and $180 million under the 30-year plan [beginning in 2024]. He reiterated that annual payments were reduced under the longer-term plan, but ultimately the longer plan would be more expensive. Co-Chair Kelly referred to the 25-year versus 30-year comparisons. He was interested in the level dollar versus the level percent methods shown on the spreadsheet. Mr. Teal addressed the level percent of pay column on the spreadsheet, which started out paying $170 million versus the $230 million for level dollar; payments changed to $198 million and $196 million respectively by 2024. In later years level percent of pay increased to $250 million, while level dollar remained at $195 million. He detailed that in the future level percent would cost the state $100 million more annually and payments would be made two years longer. 3:21:23 PM Senator Hoffman observed that the spreadsheet Mr. Teal was speaking to was limited to TRS. Mr. Teal replied in the affirmative. Senator Hoffman referred to $1 billion payments under level dollar for 2018. He observed that under level percent of pay the payments were $800 million. He noted the $200 million difference. He observed that the graphs provided by LFD all differed depending on the scenario. He wondered where the state would be financially and its ability to build the gas line. He believed the decision to choose level dollar versus level percent needed to take into account the state's ability to build the gas line with its cash calls in 2024. He reasoned the state would need the gas line revenue out past 2024 in order to meet the obligations between 2024 and 2032. He concluded that somewhere the state had to put the needed cash calls into a schedule to determine where the state was headed. Co-Chair Kelly agreed. However, he observed that it was difficult to make reserve solutions in dealing with the PERS/TRS model. He reasoned it had to be spending or revenue. He surmised that none of the unfunded liability scenarios were positive. Senator Dunleavy discussed the level percent versus level dollars methods. He wondered which scenario worked better under an increasing inflation rate over 10 to 20 years. He asked when inflation was an ally. Mr. Barnhill replied that it depended on an estimation of how the investment markets were correlated to inflation. For example, he would double down on level dollar if a hyperinflation environment was resulting in booming equity markets. He stated that the situation was sometimes, but not always, the case. 3:25:57 PM Mr. Teal agreed with Mr. Barnhill. He added that pensions were essentially fixed cost; with inflation the pensions eroded. He remarked on the difficulty of the question. He explained that benefits increased with salary increases, which meant pension liabilities would climb with inflation. However, the scenario was offset if market returns also increased with inflation. He reasoned that the numbers could be included in a model; however, the outcome was determined completely by assumptions. He could not determine whether high inflation would be a help or hindrance; it would have an impact in many ways. Vice-Chair Fairclough believed the level dollar method was the best for the people of Alaska in the [retirement] system. She did not believe making a decision based on a large project was the right focus. She observed that the market would do what it did; if the state had a good project, it would have the ability to borrow at a low rate. She believed many investors would be at the table to purchase the bonds. She emphasized that the state had $1 billion on the table in the decision. 3:27:54 PM AT EASE 3:29:17 PM RECONVENED Co-Chair Meyer discussed the agenda for the following day. SB 220 was HEARD and HELD in committee for further consideration. ADJOURNMENT 3:30:34 PM The meeting was adjourned at 3:30 p.m.