Loading...
HomeMy WebLinkAboutRESPONSE - RFP - 7135 ACTUARIAL SERVICES AND POLICY REVIEW GENERAL EMPLOYEES RETIREMENT PLANMilliman April 29, 2010 Mr. James B. O'Neill, II, CPPO, FNIGP City of Fort Collins Purchasing Division 215 North Mason Street, 2nd Floor Fort Collins, Colorado 80524 1099 Eighteenth Street Suite 3100 Denver, CO 80202-1931 USA Main +1 303 299 9400 Fax +1 303 299 9018 milliman.com RE: Request for Proposal, 7135 Actuarial Services and Policy Review General Employees' Retirement Plan Dear Jim: Enclosed are eight copies of Milliman's proposal to provide actuarial services for the City of Fort Collins General Employees' Retirement Plan. Thank you for considering Milliman. If you have any questions, please contact me (303-672- 9046) or Joel Stewart (303-672-9003). We look forward to hearing from you. Sincerely, Patricia Ann Kahle, FSA, MAAA, EA Principal and Consulting Actuary PAK:fm Enclosures ffices in Principal Cities Worldwide J:\DOCUMENTS\Proposals\10 CFC\10 COFC GE P cover.doc 1 U 1 I I The complete biographies of our proposed team are included as Appendix A. 3. A specific description of information, including electronic data formats, that the firm would need from the City to complete the study and services. Member data can be provided in a number of formats, preferably, Excel or Access files. Because we have previously performed GERP's valuations, we are familiar with the Excel files and supporting information provided in the past. If this is convenient for City staff, we will continue to use that format in the future. We will gladly work with you to revise the format if that is deemed necessary. Membership Data Required Generally, three types of membership data are needed for the actuarial valuation: data on active employees, data on retirees and beneficiaries and data on inactive members who retain benefits under the plan. The data should be provided for each individual member, and the data items needed for each member will vary depending on the category and plan specifics. Typically, the data items will include the following: 1. Data on active employees as of the valuation date. Data items needed for each member are as follows: • Name. • Employee identification number. a Information on any breaks in service during the prior plan year. • Hours credited during the prior plan year. • Actual earnings for the prior plan year (we will annualize). o Percentage full-time eligible (FTE), if less than 100%, as of December 31. © Compensation rate effective for the current plan year. We will adjust compensation by the FTE for projection purposes. 2. Additional data on active employees needed for preparation of the Pension Retirement Planning Statements (PRPS): Asset allocations of the 401(a) and 457 plan accounts by equities, fixed income, and cash. ® Contribution rates and end of year account balances for both the 401(a) and 457 plans, separately. o Long-term rate of return on equities and fixed income, to be used to determine future expected return for the 401(a) and 457 accounts for the PRPS. 3. Data on retirees, beneficiaries and disabled retirees and as of the valuation date. Data items needed for each member are as follows: o Copy of the City of Fort Collins Pension Payroll Register for December, listing all participants receiving a monthly pension. Proposal to Provide Actuarial and Consulting Services for ■ � M i I I i ma n the City of Fort Collins General Employees' Retirement Plan fl 1 I 1 B[E[ME [FQf S PERSPECTIVES NDCP benefits would spring the trap only if the payment was triggered by the occurrence of an event(s) connected to the restricted period (e.g., at -risk status, bankruptcy, etc.). The trap would not be sprung if the payment was made independently of these event(s). That is, the payment would be allowed during this period if it was made in accordance with the plan's provisions, due to the occurrence of a 409A permissible payment, such as a separation from service, death, or disability. All this uncertainty leaves open major questions, including whether NDCP accruals may or should continue during the restricted period and whether NDCP sponsors may or should pay benefits that were not fully funded prior to the restricted period. NDCPs with employee deferrals As difficult as the funding restrictions may be for NDCP sponsors in general, they pose even greater problems for sponsors of plans that include employee deferrals. Arrangements of this type invariably are set up with a rabbi trust so that the employee's deferrals can be directly deposited into this trust, much as 401(k) deferrals go into a qualified trust. Having the deferrals deposited in a rabbi, trust does not protect them against creditors of the sponsor in the event of insolvency. Doing so does protect them against a change in control (or change of heart of the plan sponsor) that could make it difficult to receive payment of NDCP benefits if the funds were not held by a third -party trustee. Many NDCP participants would not defer their compensation to the plan if not for the underlying rabbi trust. In general, 409A requires participants to make their deferral elec- tions prior to the beginning of the calendar year and for the elec- tions to remain in place until the beginning of the next calendar year. Consequently, NDCP sponsors should closely monitor their risk of entering a "restricted period" and be sure to communicate the risk and its consequences to executives as soon as possible before the restrictions apply. This proactive approach will provide participants the advance notice they will need to make timely elections to reduce or discontinue their future deferrals (if they are wary of continuing to defer at the same rate without the same level of protection for defer- rals). Similarly, a sponsor should promptly inform such participants if and when restrictions are expected to be lifted so they have suf- ficient time to increase or resume contributions once contributions can again be deposited directly into the rabbi trust. An action plan for NDCP sponsors Since NDCP funding restrictions depend on the DB plan's funding status, NDCP sponsors that may be subject to funding restrictions should discuss the DB plan's current funding status and future funding strategies with the qualified plan's actuary as soon as possible. "Applicable covered employees" will need to be identified for publicly traded companies. Private companies should either include their top executive officer or consult counsel for confirmation if they believe they are not affected by the restrictions. All NDCP documents and related trust agreements should be reviewed for any "springing provisions" that could result in full funding or distribution of benefits and that may be interpreted as occurring "in connection" with the restricted period. Last but certainly not least, until clarifying guidance is available, NDCP sponsors need to settle on an interim solution to the funding question prior to entering a restricted period. Some sponsors may take the position that given the lack of specificity in the statute's language, they would still be acting in good -faith compliance with the rules as long as they did not make any transfers or deposits to a trust or other arrangement. In other words, accruals under the plan could continue and payments could be made when due, provided they were triggered by a 409A permissible payment event stated in the NDCP and not by the restricted period or some other financial measure, such as some designated decline in the sponsor's stock price, gross revenues, or annual profits. The rationale here would be that even if later guidance tightened the wording to prohibit this interpretation, the later guidance would only apply prospectively. The more conservative approach would be to amend the NDCP to freeze NDCP accruals/contributions on the day before the restricted period begins and provide that no payments be made from the NDCP during the restricted period other than those attributable to amounts that were funded prior to the freeze date. Then, if a participant is due to receive a distribution in accordance with a 409A permissible payment event under the terms of the plan, only the portion of the benefit that is supported by funds that were set aside prior to the restricted period would be paid. If the NDCP sponsor had already set aside sufficient amounts to meet the benefit obligations as of the freeze date, it should be able to distribute 100% of the benefit if a 409A permissible payment event occurs during the restricted period. If any portion of such other- wise payable benefit is not supported by pre -restricted period funding, then that portion would be withheld until the restricted period ends, at which time the plan can also be amended to restart accruals/contribu- tions (and even restore lost accruals/contributions if desired). Plan sponsors should consult legal counsel before making a final decision, regardless of which approach is taken. For example, even if the conservative approach is adopted, the sponsor could still face a legal challenge if the plan benefits were not fully funded prior to the freeze date. The source of the challenge would not be the IRS but rather participants who are denied full or partial payments and argue that the sponsor is incorrectly withholding such payments due to its overzealous interpretation of the law (and thus reneging on its con- tractual obligation under the NDCP). In an attempt to ward off such challenges, sponsors who opt for this approach should communi- cate the changes to the affected participants in advance and obtain their consent to the amendment if possible. While all eagerly await further clues from the IRS, the one current certainty of the NDCP/ DB connection is that this is indeed a many-sided conundrum which must be examined from all actuarial and legal angles to keep puzzled participants and sponsors from becoming "at risk." Dominick Pizzano is a compliance consultant in Milliman's Woodland Park, N.1., office. This article was peer reviewed by Dawilla Madsen, who is also a compliance consultant in Woodland Park. 6 :: WINTER 2009/2010 1 ii I 1 1 [8[E[{(�\���� PERSPECTIVES �'Curre 11 esin Empl Y Benefi�a Considering a Healthcare Dependent Eligibility Audit? Heidi tenBroek, Employee Communications Consultant Employers across the country are auditing their healthcare plans to check the eligibility of employees' dependents. And they are saving significant money in the process. Employees are asked to verify (and often provide substantiating documentation) that their enrolled dependents meet the healthcare plan's definition of eligibility —otherwise, the dependents' coverage is dropped. For many employers, this is not just about trimming costs but also about doing the right thing. They see it as their fiduciary responsibility. When considering such an audit, be aware of your organization's culture and clear about your objectives. There are a number of ways these audits can be conducted and different ways to measure success. This article highlights: reasons to conduct a health plan dependent eligibility audit; • options for implementing audits; and • things to consider before, during, and after an audit. We don't have a problem... You may be surprised by how many ineli- gible dependents are covered under your healthcare plan. At the beginning of a recent audit, one Pacific Northwest employer pre- dicted that 3%-5% of the dependents cur- rently enrolled were possibly ineligible. They were surprised to learn that approximately 12% of the enrolled dependents were ineli- gible. Firms that perform eligibility audits generally predict 5%-10% of dependents are enrolled when they should not be. As you say to yourself, "But our employees are generally honest," consider some of the reasons an employee may have an ineligible dependent enrolled: • Bob enrolled in the plan years ago; his daughter Michelle is 25 now but still living at home while going to graduate school. He thinks she is eligible. • Candace covers her common-law spouse. However, her plan only covers same -sex domestic partners. • Steve is required to provide health cover- age for his ex-wife as part of their divorce settlement. He leaves her enrolled as his spouse under his employer coverage. • Diane is caring for her grandson. She does not have legal custody but the child is dependent on her for support. She just assumes he would be covered. Diane did not take the time to read and understand the definition of an eligible dependent. None of these employees has deliberately broken the rules of the plan —there were no checks in place to prevent these errors and little emphasis on plan eligibility rules. Look at your current practices Whether or not an eligibility audit will give you a robust return on your investment depends on how tightly you currently manage eligibility. Are you among the many employers who operate on the honor system? A typical scenario is one in which the employer publishes the definition of an eligible dependent in the enrollment materials and summary plan description but requires no documentation to substantiate relationships when a new hire enrolls. Many employers require documentation for mid- year status changes and an affidavit to cover domestic partners, but that is where it ends. Organizations with these practices often realize significant savings from an audit. Other employers already tightly manage eli- gibility. If you require documentation such as marriage or birth certificates, tax returns, and student status verification for all new enroll- ments and enrollment changes, you are likely to find fewer ineligible dependents through a full audit. Periodic random sampling or a more hands -off approach, such as requiring employees to attest to dependent eligibility (but not provide documents), may be all you need for maintenance. Laying the groundwork Before you ask employees to go through an audit, take a fresh look at your definition of eligibility. • Is it specific and detailed enough? • Is it clearly and simply written? • Have you actively communicated it? Your definition may declare that dependent children up to a particular age are eligible. But what does "dependent" mean exactly? And what do you mean by "children"? Are employees' grandchildren who live with them eligible? If the definition is buried in your Summary Plan Description, employees are unlikely to think about it no matter how well written. Emphasize the requirements through several "touch points." One point to emphasize is that it is the employee's responsibility to determine whether their dependent meets the eligibility requirements —not your organization's. The documents you choose to accept during an audit or on an ongoing basis can only tell part of the story. A marriage certificate shows that a person was the employee's spouse at some point, but what is their current relationship? A birth certificate for a 22-year old may show the employee is the parent but it does not confirm financial dependency or student status. You have to depend on the employee taking responsibil- ity for understanding the requirements and ensuring he or she meets them. You can do your part by making your definition clear, easy to understand, and visible. Read your company culture You may be surprised by how well employ- ees generally accept a dependent eligibility audit. Employees are concerned about rising healthcare costs too. The concept of fair - WINTER 2009/2010 :: 7 1 QE[M[E[FOfS PERSPECTIVES t „ < E,np.ior« B<n.e�s ness with regard to maintaining the plan for those who are eligible resonates as well. The keys are to communicate the goals of the audit, let employees know exactly what they need to do (and what happens if they do not take action), and give them ample time to gather their supporting documenta- tion. It also helps to follow up and report the results of the audit —people need to feel the trouble they went through was worthwhile. Do not forget about senior management. One of the first things you need to do is get their support. Leaders are employees too and if they grumble about the audit, it will affect the attitude of the entire organization. Also, leaders could be more interested than usual in some of the planning details, such as whether you will attempt to recoup: ineligible claim costs, timing of the audit, etc. Having them on board early will help smooth the entire process. ' What audit is right for you? The type of audit that is right for your organization depends on your culture and how stringent you have been in managing eligibility up to this point. There are a number of audit methods to consider: • Full audit of all employees who cover dependents. This is the most intensive type of audit. This approach requires all employees to submit written documenta- tion of an eligible relationship. Consider this option if you have not monitored dependents closely in the past. Outside firms can help with the administration of ' the process and employee questions. You will likely realize a significant return on your investment. • Full audit with affidavit only. Instead of requiring employees to submit documenta- tion, one option is to have all employees who cover dependents testify on a signed ' affidavit that their covered dependents meet the plan's eligibility requirements. This can be a paper process or online. Consider this if your organization is uncomfortable with asking employees to submit paperwork. • Random sampling. This approach requires less documentation to process; however, it is important to communicate about the audit to all employees for the message to be effective. If the issue is that employees are simply not paying enough attention, they will likely continue to ignore the issue unless their number comes up in the random sample. This may not be a good option if you want a fresh start before you implement new ongoing processes. Consider this approach if you already tightly manage eligibility. Things to consider Before the Audit: • Get leadership buy -in • Determine level of audit desired 1. Full documentation required? 2. Affidavit approach? 3. Random sample? • Decide how to handle displaced dependents 1. Offer continuation coverage? 2. Offer limited extension of coverage (e.g., three months)? 3. Assist in finding individual coverage? Involve the field human resources (HR) representatives (the eyes and ears that can provide support for the audit) Specify the date when coverage ends • Determine desire for claims recovery Take advantage of open enrollment 1. Heavily communicate dependent defi- nition and announce impending audit (quasi -forgiveness period) 2. Mull full dependent reenrollment During the Audit: • Decide on the level of HR's involvement 1. Blast e-mail reminders? 2. Personal reminders? Review questionable documentation Benefits Perspectives is published by Milliman's Employee Benefits Editorial Committee as a service to our clients. Additional copies are available through any of our offices. Articles or excerpts from this publication may be reproduced with permission when proper credit is attributed to the firm and the author. ' Editorial Committee Rich Voorhees, Carl Hansen Troy Pritchett Editor -in -Chief Janet Kalwat Marjorie N. Taylor ' Eddy Akwenuke Jeffrey R. Kamenir Katherine A. Warren Patricia Dodd Martha Moeller 8 :: WINTER 2009/2010 • Handle documents sent to HR in error Determine how often to receive updates and by what means • Do a final review of unverified dependents before final notification After the Audit: Communicate results • Decide how to respond to appeals • Implement ongoing verification process Continue communication • Schedule recurring audits 1. Random sampling? 2. Full audits? Taking the next step In addition to potentially achieving cost savings, verifying dependent eligibility helps you fulfill your fiduciary responsibility to manage the plans for the exclusive benefit of eligible participants and beneficiaries. It shows employees that you are taking action to keep not only the organization's expenses, but their costs, as reasonable as possible. If you've decided an eligibility audit is right for you, lay the groundwork by: • having initial conversations with management to gauge its support; • reviewing your definition of eligible depen- dents and how it was communicated; • taking an objective look at your organiza- tion's culture and its tolerance for this level of inquiry; and • emphasizing dependent eligibility during your next open enrollment and requiring employees with ineligible dependents to take them off the plan. Heidi tenBroek is an employee communications consultant with Milliman in Seattle. This article was peer reviewed by Penny Plante, a benefits consultant in Seattle. Because the articles and commentary prepared by the professionals of our firm are often general in nature, we recommend that our readers seek the counsel of their attorney and actuary before taking action. The opinions expressed are those of the authors. Inquiries may be directed to: Perspectives Editor 1301 Fifth Avenue, Suite 3800 Seattle, WA 98101-2605 (206) 624-7940 perspectives.editor@milliman.com 02010 Milliman, Inc. All Rights Reserved 1 I March 30, 2010 CAB 10-05S K7 M I I I I e'1'1a n • • Employee°Ber Employer -Sponsored Health Plans under Healthcare System Reforms SUMMARY Employers that sponsor health benefit plans for their employees face significant new challenges under the sweeping changes made by the "Patient Protection and Affordable Care Act" (P.L.111-148, signed on March 23), as amended by the "Health Care and Education Reconciliation Act" (H.R.4872) that the President signed into law on March 30. This Client Action Bulletin provides a general chronological overview of the law's key provisions affecting employer -sponsored plans. Some provisions take effect in 2010, thereby applying several insurance market reforms sooner for sponsors of noncalendar-year plans. The first major changes for calendar -year plans will start in January 2011. Plans subject to a collective bargaining agreement may have later effective dates. Additional details about and analyses of the final healthcare system reforms will be covered in future Milliman publications. DISCUSSION 2010 Effective in the accounting period that includes March 23, 2010, plan sponsors that receive the federal subsidy for their Medicare Part D prescription drug coverage must recognize the tax liability of the effects of the 2013 deduction disallowance (see 2013, below) on company financial statements. Effective June 21, a temporary reinsurance program will reimburse plan sponsors for 80% of the claims between $15,000 and $90,000 for pre -Medicare -eligible retirees (aged 55 through 64). The program will expire at the end of 2013 or, if earlier, when the $5 billion in funding is depleted. Part D enrollees who reach the coverage gap (the "donut hole") will receive a $250 rebate, and the donut hole will start to phase out so that beneficiaries will pay the standard 25% cost sharing for prescription drugs by 2020. 2011 o Employers with fewer than 25 employees and average annual wages of less than $50,000 will qualify for a sliding scale tax credit if they purchase health insurance for their employees and contribute at least 50% of the insurance costs. ■ A new voluntary long-term care program ("CLASS Act") will allow an employer to automatically enroll ' individuals and collect premiums if they do not opt out. ■ Calendar -year group health plans may not impose lifetime dollar limits; may only impose "restricted" annual dollar limits (to be defined); and must extend coverage to an employee's child up to age 26 if the adult child ' is not eligible to enroll in an employer -sponsored plan. ■ Employers must include the "value" of employees' health coverage on Form W-2. ■ Funds from health flexible spending accounts (FSAs), health savings accounts (HSAs), and health reimburse- ment accounts (HRAs) may not be used to reimburse the costs of nonprescribed over-the-counter drugs. 2013 ■ Contributions to health FSAs will be capped at $2,500 annually (to be indexed). ■ Employees (but not their employers) will be subject to an additional 0.9% Medicare Hospital Insurance ' payroll tax (i.e., a total of 2.35%) if their wages exceed $200,000 for a single taxpayer (or $250,000 for joint filers). ■ Sponsors that receive the federal subsidy for their Medicare Part D prescription drug coverage will no longer be able to deduct the subsidy. A member of Abelica Global milliman.com I � ,; iF�iployee Banefit� 2014 ■ An employer will have to give certain workers the option of enrolling in the group health plan or receiving a tax-free voucher with which to purchase insurance from the new state -based Exchanges. Eligible workers generally are those for whom the group plan premiums represent between 8.0% and 9.8% of the workers' household incomes or whose household incomes are up to 400% of the federal poverty level. The voucher must be equal to the employer's largest cost -sharing contribution to any of the employer's plans and may be adjusted for age or family. Coincident with this employer responsibility, most individuals must obtain qualified health coverage or pay an annual penalty. ■ The insurance offered to individuals and small employer groups through an Exchange will be categorized into four tiers based on their value as a percentage of costs paid. The different levels will require an actuarial certification: the bronze tier pays 60% of expected coverage costs; the silver, 70%; the gold, 80%; and the platinum, 90%. For employer -sponsored plans that are not part of the Exchanges, actuarial certification will be required to ensure that the plans satisfy the "qualified coverage" standards. ■ An employer with 50 or more full-time equivalent (FTE) workers will be subject to a $2,000 penalty per FTE worker if it does not offer "qualifying" coverage and it has at least one FTE worker who receives a federal premium tax credit or cost -sharing subsidy. (FTEs work at least 30 hours per week.) If the employer offers coverage but has at least one FTE worker receiving the federal assistance, the employer penalty will be the lesser of: a) $3,000 per subsidized worker, or b) $2,000 per FTE worker. The first 30 FTE workers are excluded from the penalty calculation, and no penalty will be assessed for employees who have not completed a plan's applicable 90-day waiting period. ■ All group health plans will have to satisfy certain requirements to be considered a qualified health plan. Among the requirements is the offering of an "essential" benefit package that the Secretary of Health and Human Services will define and that shall include benefits in several broad categories, such as ambulatory patient services, emergency services, hospitalization, prescription drugs, rehabilitative services and devices, and laboratory services. Qualified plans also must have specific cost -sharing limits. ■ Employers must file annual reports to the federal government providing information about the health plan. ■ A plan must eliminate all annual limits and preexisting condition exclusions; extend coverage to nondepen- dent children up to age 26 regardless of whether coverage under another employer plan is available; and have no waiting periods exceeding 90 days. In addition, all individual and small -group insurance offered through the Exchanges will be guaranteed issue, guaranteed renewable, and with no health underwriting, no preexisting condition exclusions, and limits on premiums according to specified rating bands. 2018 ■ High -cost ("Cadillac") plans with total health coverage costs exceeding specified dollar thresholds will be subject to a 40% excise tax, collected from insurance companies and self -insured employers. These costs exclude stand-alone vision and dental coverage, but take into account health plan premiums and ' any health FSAs. The dollar thresholds are $10,200 for single cove rage/$27,500 for family coverage, with higher amounts for retirees and employees in "high risk" occupations. The thresholds will be indexed to the Consumer Price Index (CPI) plus 1% in 2018 and 2019, then only to CPI after 2019. Adjustments for age and gender in determining the coverage costs are permissible if the employer's demographics are not ' representative for a national risk pool. A special rule for multiemployer plans allows for the family coverage threshold to apply for single coverage. ACTION The effects of the new law on group health plan sponsors will vary according to many factors, including, for example, whether the plan sponsor is large or small, whether the employer purchases insurance or self -insures, whether plan participants qualify for federal assistance, and whether the benefits offered can satisfy the "qualifying" standards. But nearly all plans will face numerous challenges in administrative systems and processes, employee communications, and benefit program design changes. There are also new requirements that will have indirect implications, such as taxes and fees imposed on the health provider markets that can be expected to be passed on to group health plans. In addition, until regulations are issued by the various agencies, many of the details about the implementation of the new requirements will not be known. For additional information about the new law or for assistance with the new health reform requirements, please contact your Milliman consultant. Milliman Client Action Bulletins contain general information that is not intended to constitute the rendering of legal, tax, investment, or accounting advice. Application to specific circumstances should rely on further professional guidance. 2 :: March 30, 2010 ©2010 Milliman, Inc. All Rights Reserved I r] Milliman APPENDIX E SAMPLE SIMULATION MODEL Proposal to Provide Actuarial and Consulting Services for the City of Fort Collins General Employees' Retirement Plan 11 Sample Employee Contribution Study Projections Based on January 1, 2010 Actuarial Valuation I Benefit Multiplier for New Hires: 2.00% Current Plan is 2.0% I 2010 2011 2012 2013 2014 2015 2016 20171 2018 2019 2020 Actual Return 8.00% 8.00% 8.00% 8.00% 8.00% 8.00% 8.00% 8.00% 8.00% 8.00% 8.00% Amortization Period 20 20 15 15 15 15 15 15 15 15 15 DB Employee Contribution Rate 0.00% 0.00% 0.00% 0.00% 0.00% 0.00% 0.00% 0.00% 0.00% 0.00% 0.00% DB Employee Contribution 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 Budget Contribution millions) (May 1 ARC 3.709 3.787 4.486 4.890 5.280 5.306 5.323 5.312 5.323 5.333 5.315 May 1 Annual Re 'd Contribution ARC mils 3.709 3.787 4.486 4.890 5.280 5.306 5.323 5.312 5.323 5.333 5.315 May 1 ARC + Employee Contribution millions 3.709 3.787 4.486 4.890 5.280 5.306 5.323 5.312 5.323 5.333 5.315 Employer Contribution = Max (Budget or ARC 3.709 3.787 4.486 4.890 5.280 5.306 5.323 5.312 5.323 5.333 5.315 Total Employee + Employer Contribution mils 3.709 3.787 1 4.486 4.890 5.280 1 5.306 1 5.323 5.312 5.323 5.333 5.315 Funded Ratio 67.2%1 68.3%1 67.3% 64.2%1 61.7%1 63.8% 65.7% 67.5% 69.0% 70.3% 71.5% Funded Ratio - 80% -- 70% 60% 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 These are for projection purposes only, and are based on assumption listed above. Actual results will vary. Annual Required Contribution $5.5 $5.0 $4.5 24.0 o_ i.5 $3.0 $2.5 $2.0 - 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 1 I Benefit Multiplier for New Hires: Sample Employee Contribution Study Projections Based on January 1, 2010 Actuarial Valuation 2.00% Current Plan is 2.0% 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 Return 8.00% 8.00% 8.00% 8.00% 8.00% 8.00% 8.00% 8.00% 8.00% 8.00% 8.00% iActual Amortization Period 20 20 15 15 15 15 15 15 15 15 15 DB Employee Contribution Rate 2.00% 3.00% 3.00% 4.00% 4.00% 5.00% 5.00% 5.00% 5.00% 5.00% 5.00% DB Employee Contribution 0.415 0.637 0.661 0.878 0.908 1.130 1.161 1.183 1.210 1.236 1.260 ' Budget Contribution millions) (May 1 ARC 3.284 3.134 3.810 3.993 4.353 4.154 4.143 4.112 4.097 4.082 4.042 May 1 Annual Re 'd Contribution ARC mils 3.284 3.134 3.810 3.993 4.353 4.154 4.143 4.112 4.097 4.082 4.042 May 1 ARC + Employee Contribution millions 3.698 3.771 4.471 4.871 5.262 5.284 5.303 5.294 5.307 5.319 5.302 Employer Contribution = Max (Budget or ARC Total Employee + Employer Contribution mils 3.284 3.698 3.134 3.771 3.810 4.471 3.993 4.871 4.353 5.262 4.154 5.284 4.143 5.303 4.112 5.294 1 4.097 5.307 1 4.082 1 5.319 4.042 5.302 Funded Ratio 67.2% 68.3% 67.3% 64.2% 61.7% 63.7% 65.6% 67.3% 68.8%1 70.2%1 71.3% Funded Ratio 80% 70% - ' 60% - 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 IThese are for projection purposes only, and are based on assumption listed above. Actual results will vary. 1 Annual Required Contribution $4.5 - - - ----- $4.0 - W.5 - 0 F3.0 - $2.5 $2.0 - 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 n �J t I o List of participants who retired or died during the year, including termination date, form of payment, monthly benefit or lump sum amount (depending on election), and beneficiary's date of birth if applicable. 4. Data on vested inactive members as of the valuation date. Data items needed for each member are as follows: ® List of participants who terminated without receiving any benefits during the plan year. This list should include termination date and monthly benefit amount. If some of the information requested above is not readily available, we are willing to discuss alternatives to simplify the data requirements. The list shown includes that information we feel would be necessary to perform a thorough actuarial valuation of the system based on our knowledge of the current plan provisions. Financial Data Required The financial data needed for the valuation follows: 1. A cash flow statement for the year prior to the valuation date showing: • Market value of assets at the last valuation date o Benefit payments during the year • Contributions o Expenses • Earnings o Market value of assets on the current valuation date. 2. Detail on benefit payments. Monthly benefit payment ledger Lump sum ledger 3. Statement of the assets by investment type on December 31. 4. A "not to exceed" cost estimate of the Actuarial Reports, and an approximate time frame to complete it after information is provided to you by January 31, 2011. Note: The Committee needs a draft report by March 31, 2011 and a final report by May 15, 2011. We will be able to complete the valuation by March 31 if we are given the data by January 31. Specifically we will compare the data to the previous year's data and look to see if it is reasonable and if there are any inconsistencies during February. Also during February, we will review the plan population experience and compare it to our expectations. At the March GERC meeting we will review the actual asset and population experience with the Committee, and we will recommend assumption changes to the Committee for the valuation as needed. Once the assumptions are set in early March, we will run the valuation ■ Proposal to Provide Actuarial and Consulting Services for M i I I i ma n the City of Fort Collins General Employees' Retirement Plan I I f� - ------------- through our pension system and compile results in a draft actuarial valuation report. This draft report will be presented to the committee at the April board meeting, and the report will be finalized by May 15cn The "not to exceed" cost of the actuarial valuation for the January 1, 2011 plan year is $9,600. Future annual fees would increase by the Denver Boulder Greeley CPIU as published by the Colorado State Planning and Budget Office. 5. If possible, a "not to exceed" cost estimate of the Personal Retirement Planning Statements should be provided. Note Statements must be distributed to Plan Participants by March 31. We will be able to complete the Personal Retirement Planning Statements by March 31 if we are given the employee data by January 31 and any outside data such as account balance information by February 15. Processing the benefit statement data will be done in conjunction with the valuation data. The "not to exceed" cost of the Personal Retirement Planning Statements for the January 1, 2011 plan year is $3,200. Future annual fees would increase by the Denver Boulder Greeley CPIU as published by the Colorado State Planning and Budget Office. 6. Costs should also be provided for attendance of monthly meetings and hourly rates for special projects. Our fees to attend the monthly Committee meetings will be billed at time and expense with a "not to exceed" fee of $750 starting January 1, 2011. Our fees to attend a monthly meeting via phone, will be billed on time and expense with a "not to exceed" fee of $400. Future annual fees would increase by the Denver Boulder Greeley CPIU as published by the Colorado State Planning and Budget Office. In preparing this proposal, we have taken particular care in addressing the fee basis because we recognize that budget limitations may be a primary concern. The actuarial services described in this proposal are quoted on a not to exceed basis for the services listed below: O Annual actuarial valuation of the benefits provided, including a review of the Plan's funding level, and GASB disclosure information. Preparation of Personal Retirement Planning Statements. O Attendance at monthly meetings. • Proposal to Provide Actuarial and Consulting Services for M i I I i ma n the City of Fort Collins General Employees' Retirement Plan 7 u 11 I LI 1 The following table summarizes the annual "not to exceed" fee to be charged for the services listed under bullets 4, 5, and 6: "Not To Exceed" Actuarial Report $ 9,600 Personal Retirement Planning Statements 3,200 Total of Annual Required Work $ 12,800 Attendance at 12 Monthly Meetings ($5,500 - 10 meetings via phone and 2 in person) $ 5,500-9,000 Total $ 18,300-21,800 If the work effort required to complete the project generates a lower fee based on our hourly rates, we will bill the lower amount. Benefit calculations are not included in the fees specified above. Benefit calculations are expected to require one to two hours to complete per retiring or terminating participant, with an expected cost of less than $300 per calculation. On occasion, benefit calculations involving breaks -in-service, Section 415 limitations, deaths, QDROs or other special circumstances may require additional time and expense. We propose to bill separately, on a time -and -expense basis, for consulting services associated with special projects. The allocation of staff on any special project will be designed to balance the skills of the Milliman staff with the need for the lowest possible fee. Some projects may, by their nature, require that substantially all of the time be provided by the consulting actuaries, but others may be heavily weighted by lower level actuarial staff and clerical staff. When special projects arise we would be happy to provide an estimate of the additional fee involved and if the project is well-defined, to provide a not to exceed fee. The other services listed in the RFP will be billed at our regular billing rates on a time and expense basis. We will use the staff member with the lowest billing rate that best fits the project. The 2010 billing rates follow. We anticipate that billing rates will increase at 3% per year to adjust for cost of living: Staff Designation 2010 Rate Pat Kahle $340 Joel Stewart $245 Francine Moyer $195 Pat Beckham (special consultant) $350 7. Sample actuarial report completed within the last year covering similar services to those requested in this proposal. A sample actuarial report is enclosed under Appendix B. Please note that we are flexible in our report format and will change or add content depending on the client's needs. Proposal to Provide Actuarial and Consulting Services for M i I I i ma n the City of Fort Collins General Employees' Retirement Plan 1 11 SECTION 3 STANDARD FOR SUPERVISING ACTUARIES Specific responses to the Committee's requirements follow: 1. Consultant credentials The following chart summarizes the specific consultant credentials for each professional who will be assigned to the City's account. Professional Years of Years With Name and Title Role Qualifications Experience Milliman Patricia Kahle, Principal and Lead Consultant FSA, MAAA, 25 25 Consulting Actuary EA Joel Stewart, Consulting Project Manager EA, MAAA - 13 10 Actuary and Co -Lead Francine Moyer, Associate Actuarial ASA, MAAA 16 13 Actuary Support Patrice Beckham, Principal Senior Resource FSA, MAAA, 32 23 and Consulting Actuary Actuary EA *Key to actuarial credentials FSA - Fellow of the Society of Actuaries ASA - Associate of the Society of Actuaries EA - Enrolled Actuary MAAA - Member of the American Academy of Actuaries Please see the resumes for each team member in Appendix A for more detail on each team member's experience. ' 2. Experience and involvement in public retirement systems Prior to transferring to the Denver office, Pat Kahle served as lead consultant for over twelve Public Sector retirement plans in Oregon, as well as assisting with several other plans, including the Oregon PERS. Pat is currently the lead consultant on all public sector and corporate plans in the Denver office. In that capacity, she provides actuarial valuation, cost ' studies, experience studies, and solvency and funding projections. She also oversees all administrative services required and presents valuations to a number of different retirement committees and/or boards. Both Joel and France also provide services to all of the Denver Public Sector clients. Milliman provides actuarial services to many municipal public sector plans that are similar to the GERP in size and complexity. A listing of the public plans serviced by Milliman is shown in Appendix C. Please note the variety of plans we serve such as state systems and the uniformed systems clients. ■ Proposal to Provide Actuarial and Consulting Services for p 9 9 Milliman the City of Fort Collins General Employees' Retirement Plan i 3. Testifying Pat Kahle, the primary actuary on our proposed project team, has had many years of experience with city, district and municipal public retirement systems in both Oregon and Colorado. She has worked with City Managers, Finance Directors and Human Resource professionals. When asked, Pat has presented benefit or actuarial information to city councils, board of commissioners and state retirement systems. These meetings ranged from education sessions to meetings where she had to explain undesirable results. Pat worked as co -lead actuary for a statewide system and as such has presented to the State retirement board. She has also competed audits for state systems in Wyoming and New Mexico (Retiree Health) and has presented to each board of the statewide system. ' The senior resource actuary, Pat Beckham serves as Lead Consultant for numerous statewide systems. She has extensive experience testifying before statewide legislative and administrative bodies. 1 11 4. Communication skills Actuarial concepts are abstract and not easily understood. Milliman consultants pride themselves in being able to explain these abstract concepts in everyday language. We use graphs and charts to help relay these concepts as illustrated in our valuation presentation made to the City in April. We often find that simulation models and graphs provide our clients with clear pictures of the effects of changes in assumptions, market return, or plan changes. Samples of simulation models are shown in Appendix E. In this particular model, the client was exploring the impact of adding required employee contributions. Please review the Annual Required Contribution graph to see how adding employee contributions lowers the employer contribution levels. We also find that if we listen carefully to our clients we can identify ideas or concepts that are not clear. We work carefully with clients to create presentations for committees or boards that contain the correct level of detail for the audience. Proposal to Provide Actuarial and Consulting Services for M i I I i ma n the City of Fort Collins General Employees' Retirement Plan 10 I k SECTION 4 PROFESSIONAL SERVICES AGREEMENT REVIEW We have reviewed the Professional Services Agreement included in the RFP. We are able to accept the Agreement with the Addendum on the following pages. This addendum is consistent with contract and Addendum that we currently have with the City. Our legal department updated the addendum to conform with the language used in the Fort Collins Professional Services Agreement. For example the word "Actuary" in the previous addendum has been changed to "Professional". The Addendum is shown on the following page. • Proposal to Provide Actuarial and Consulting Services for M i I I i man the City of Fort Collins General Employees' Retirement Plan 11 Addendum to the Professional Services Agreement For the City of Fort Collins, Colorado and Milliman, Inc. This addendum, executed on [INSERT APPLICABLE DATE] applies to the Professional Services Agreement dated [INSERT APPLICABLE DATE] ("Agreement") entered into by the City of Fort Collins, Colorado ("City") and Milliman, Inc. (the "Professional"). The parties agree to the following modification to Section 6 (Design, Project Insurance and Insurance Responsibility), sentence number two: "Subject to the Limitation of Liability set forth in Section 16, the Professional shall defend, indemnify, save and hold harmless the City its officers and employees, in accordance with Colorado law, from all damages whatsoever claimed by third parties against the City and for the City's costs and reasonable attorneys fees arising directly or indirectly out of the Professional's negligent performance of any of the services furnished under this Agreement. The parties agree to add the following provisions: 16. Limitation of Liability. The Professional will perform all services in accordance with applicable Professional standards and shall correct or re -perform any services found to be defective. The parties agree that the Professional shall not be liable to City for damages in any tort or contract claim, including, but not limited to, claims for Professional negligence, malpractice or breach of contract, where such damages exceed $1,000,000. In no event shall the Professional be liable for any incidental or consequential damages. The foregoing limitations shall not apply in the event of the intentional fraud or willful misconduct of the Professional. The provisions of this Section will survive the expiration or termination of the Agreement. 17. No Third Party Distribution Third Party Distribution. The Professional's work product is prepared for the use and benefit ' of the City in accordance with its statutory and regulatory requirements. The Professional recognizes that materials it delivers to the City may be public records, subject to disclosure to third parties. However, the Professional does not intend to benefit and assumes no duty or ' liability to any third parties. The Professional intends to include a footer on each page of its work product as follows: "This work product was prepared solely to provide assistance to the City of Fort Collins. It may not be appropriate to use for other purposes. Milliman does not intend to benefit and assumes no duty or liability to other parties who receive this work." The City agrees to provide full page copies, including such footer, to any third parties who are provided a copy of the Professional's work product, or when posting the Professional's work product to an intranet site. 11 18. Dispute Resolution ■ Proposal to Provide Actuarial and Consulting Services for Milliman the City of Fort Collins General Employees' Retirement Plan 12 Mediation. In the event of any dispute arising out of or relating to the engagement of the ' Professional by the City, the parties agree first to try in good faith to settle the dispute voluntarily with the aid of an impartial mediator who will attempt to facilitate negotiations. A dispute will be submitted to mediation by written notice to the other party or parties. The mediator will be selected by agreement by the parties. If the parties cannot agree on a mediator, a mediator will be designated by the American Arbitration Association at the request of a party. The mediation will be treated as a settlement discussion and therefore will be confidential. Any applicable statute of limitations will be tolled during the pendency of the mediation. Each party will bear its own costs in the mediation. The fees and expenses of the mediator will be shared equally by the parties. Arbitration. If the dispute has not been resolved within 60 days after the written notice beginning the mediation process (or a longer period, if the parties agree to extend the mediation), the mediation will terminate, and the dispute will be resolved by final and binding arbitration under the Commercial Arbitration Rules of the American Arbitration Association. The arbitration will take place in Denver, Colorado, before a panel of three arbitrators. Within 30 days of the commencement of the arbitration, each party will designate in writing a single neutral and independent arbitrator. The two arbitrators designated by the parties will then select a third arbitrator. The arbitrators will have a sufficient background in either employee benefits, actuarial science, or law to reasonably prepare them to decide a dispute. The arbitrators will have the authority to permit limited discovery, including depositions, prior to the arbitration hearing, and such discovery will be conducted consistent with the Federal Rules of Civil Procedure. The arbitrators will have no power or authority to award punitive or exemplary damages. The arbitrators may, in their discretion, award the cost of the arbitration, including reasonable attorney fees, to the prevailing party. Any award made may be confirmed in any court having jurisdiction. Any arbitration shall be confidential, and except as required by law, neither party may disclose the content or results of any arbitration hereunder without the prior written consent of the other parties, except that disclosure is permitted to a party's auditors and ' legal advisors. THE CITY OF FORT COLLINS, COLORADO By: James B. O'Neill, CPPO Director of Purchasing & Risk Management I Date: IMILLIMAN, INC. M Patricia Ann Kahle Principal ■ Proposal to Provide Actuarial and Consulting Services for M i I I i ma n the City of Fort Collins General Employees' Retirement Plan 13 SECTION 5 GENERAL INFORMATION ABOUT MILLIMAN Although there are many qualified actuarial firms, we feel that no firm can match the combination of Milliman's experience, high technical ability and client service. In a world I overloaded with information, our clients rely on us to help find the right information and turn it into actionable knowledge to solve their toughest problems with confidence. We are committed to bringing depth, clarity and context to the issues and challenges that our clients face every day. Our clients rely on us to be industry experts, trusted advisors, and creative problem - solvers. Here's why: o INTELLIGENCE. Milliman attracts — and retains — the best and brightest, talented professionals who are highly skilled, savvy business consultants and actuaries. We thoroughly understand the complex industries in which we work. We help clients make business sense of complicated technical situations, with practical intelligence that yields practical results. ' INDEPENDENCE. Milliman consultants are known for objective analysis and independent, unbiased advice. Our firm is independent — owned by our employees, not by a parent organization. We adhere to the highest professional standards, with integrity and no hidden agendas. The Milliman culture also allows for true market -driven growth, with consultants free to focus on areas that are important to our clients. 1 LI INNOVATION. At the forefront of the industries we serve, we are continually innovating to help clients prepare for changes in the business landscape. We combine established approaches and solid thinking with fresh ideas and creative applications that will help your business today and tomorrow. And with our industry -leading, innovative tools and exte. ive research, your business will have the support it needs to succeed. C IN THE KNOW. Our consultants are industry experts, supported by Milliman's wide- ranging experience, large knowledge base, significant research resources and active involvement in multiple markets. With a window into your business, we can help you anticipate changes, recognize opportunities and develop proactive strategies. IN DEPTH. One of our key strengths is bringing clarity to complex situations. By listening closely and learning continually, we gain an increasing depth of understanding that can help you see the big picture and approach problems from all sides. Milliman answers are not simplistic — we use the best data and information, skilled analysis and broad business thinking to move from challenge to solution. IN FOCUS. We're a client -focused firm and our approach is tailored to each client and each project. While we have many Milliman-developed tools to offer, we don't sell cookie -cutter, pre -packaged answers to your business problems. With your specific needs in mind, Milliman consultants deliver customized, thoughtful and flexible solutions targeted to your environment. C IN TOUCH. Personal attention and long-term relationships are the cornerstones of our deep commitment to service. We work with our clients as partners and genuinely care about your business. Whether it's helping you to reduce risk, avoid costly mistakes or model a range of solutions to a given problem, Milliman consultants are responsive to your needs and help you succeed. ■ Proposal to Provide Actuarial and Consulting Services for Milliman the City of Fort Collins General Employees' Retirement Plan 14 Milliman Proposal �..��. V^[ �'�i� i.w+�t � 'S s+ me-,µ A.`✓ L �P y 0 P" 1 I 1 1 1 I a F1 I I I 1 0 I APPENDIX A PROPOSED PROJECT TEAM BIOGRAPHIES The resumes of the proposed team members follow. ■ Proposal to Provide Actuarial and Consulting Services for M i I I i ma n the City of Fort Collins General Employees' Retirement Plan I I I 0 Patricia Ann Kahle FSA, EA, MAAA Principal, Consulting Actuary Pat will be the lead actuary. CURRENT RESPONSIBILITY Pat is a principal and consulting actuary with the Denver office of Milliman. She joined the firm in 1985. EXPERIENCE Pat's area of expertise is retirement programs for both corporate and public -sector clients. Specifically, she works with defined benefit plans, defined contribution plans, retiree medical plans, retiree stipend plans, and nonqualified plans. She also assists clients with plan design, actuarial valuations, financial disclosure, plan audits, and retirement plan administration. Pat transferred to the Denver office in 2003. Prior to that, she was an actuary in the Portland, Ore., office, where she was the actuarial manager for a Fortune 500 company. In that capacity, Pat was responsible for the company's complex database, the valuation and FAS disclosures for the client's 15 pension plans, four retiree medical plans, and the supplemental retirement plan. In addition, she worked with the client's FAS disclosures for the Ireland subsidiary. Pat was also the lead consultant for more than a dozen public sector clients. She currently is the defined benefit practice leader in Denver. She provides actuarial and administrative services for both corporate and public -sector defined benefit plans and other postemployment benefits plans. This includes actuarial valuations as well as disclosures under FAS, GASB, and SSAP. She has been the co -lead consultant on a statewide system and has preformed audits on a statewide defined benefit retirement plan and a statewide retiree medical plan. PRESENTATIONS AND PUBLICATIONS Pat is a frequent speaker on actuarial topics at local benefit conferences, as well as Milliman's annual in-house technical meetings. She has presented sessions at the AICPA National Governmental Accounting and Auditing Update on basic actuarial concepts and GASB 43/45. PROFESSIONAL DESIGNATIONS o Fellow, Society of Actuaries o Member, American Academy of Actuaries o Enrolled Actuary, ERISA o Member, Denver Chapter of the Western Pension and Benefits Conference o Past president, Portland Actuarial Club EDUCATION BS, Oregon State University [�! Milliman 1099 18th Street, Suite 3100 Denver, CO 80202-1931 Tel +1 303 299 9400 Fax +1 303 299 9018 Email pat.kahle@milliman.com milliman.com I I 1 0 t I N � Joel E. Stewart EA, MAAA Consulting Actuary Joel will be co -lead actuary and actuarial manager. CURRENT RESPONSIBILITY PROFESSIONAL DESIGNATIONS Joel is an associate actuary in the Denver o Enrolled Actuary, ERISA office of Milliman. He joined the firm in 1999 o Member, American Academy of Actuaries with three years of prior experience. EDUCATION EXPERIENCE BS, Mathematics with Actuarial Option, Joel's area of expertise is in defined benefit Pennsylvania State University pension plans. He specializes in valuations and administration, including benefit calculations, special studies for benefit changes, funding valuations, and FASB and GASB valuations. Client projects have included ongoing plan valuation and administration, expense projections, asset/liability studies, development of simulation models for the redesign of benefit plans, and plan termination administration. Joel serves on Milliman's Pension System Enhancement Committee, the group responsible for overseeing software development. LA Milliman 0 1099 18th Street, Suite 3100 Denver, CO 80202-2431 Tel +1 303 299 9400 Fax +1 303 299 9018 Email joel.stewart@milliman.com milliman.com I 1 I 0 I L 0 I Francine L. Moyer ASA, MAAA Associate Actuary Francine will provide technical assistance. CURRENT RESPONSIBILITY PROFESSIONAL DESIGNATIONS Francine is an associate actuary in the Denver m Associate, Society of Actuaries office of Milliman. She joined the firm in 1996. o Member, American Academy of Actuaries EXPERIENCE EDUCATION Francine's area of expertise is in defined BS, Actuarial Science, University of Illinois at benefit pension plans. She specializes in Urbana -Champaign valuations and administration, including benefit calculations, special studies for benefit changes, funding valuations, and FASB and GASB valuations. Client projects have included ongoing plan valuation and administration, expense projections, asset -liability studies, development of simulation models for the redesign of benefit plans, and plan termination administration. 1099 18th Street, Suite 3100 Denver, CO 80202-2431 Tel +1 303 299 9400 Fax +1 303 299 9018 Email francine.moyer@milliman.com K! Milliman 0 milliman.com Patrice Beckham FSA, EA, MAAA Principal, Consulting Actuary �j Pat will serve as a special resource. CURRENT RESPONSIBILITY Patrice is a principal and consulting actuary with the Omaha office of Milliman. She joined the firm in 1987. EXPERIENCE Patrice has been in the employee benefits field since 1978, working with clients in the design, installation, and administration of all types of retirement plans, including defined benefit, defined contribution, deferred compensation, and nonqualified plans. Patrice has experience performing actuarial and consulting services for many public and private sector clients. She has assisted them in a broad range of consulting assignments, including plan design and comparability, mergers and spin-offs, pension expense allocation projection studies, and actuarial audits. Patrice is currently the consulting actuary to Iowa PERS and Kansas PERS. She has served as actuary and consultant to Omaha Public Schools, Iowa Municipal Fire and Police System, and numerous smaller public retirement systems. She has testified before the Iowa, Nebraska, and Kansas legislative committees on pension -related issues. a a a 0 Patrice has assisted clients with the communication of their plans, as well as supervised the administrative work for defined contribution plans, including reconciliation of the individual account records with the trust accounting, valuation of individual participant accounts, preparation of individual statements, calculation of termination payouts, and performance of nondiscrimination testing. 1120 South 101 st Street, Suite 400 Omaha, NE 68124-1088 Tel +1 402 393 9400 Fax +1 402 393 1037 Email patrice.beckham@milliman.com PROFESSIONAL DESIGNATIONS o Fellow, Society of Actuaries o Enrolled Actuary, ERISA o Member, American Academy of Actuaries EDUCATION BS, Mathematics/Actuarial Science, University of Nebraska, Lincoln [�-! Mllllman milliman.com I APPENDIX B SAMPLE ACTUARIAL VALUATION REPORT Proposal to Provide Actuarial and Consulting Services for M i I I i ma n the City of Fort Collins General Employees' Retirement Plan 1 1 1 1 1 1 1 1 1 JANUARY 1, 2009 ACTUARIAL VALUATION OF THE CITY OF ANYTOWN RETIREMENT PLAN Prepared by: Patricia Ann Kahle, FSA, MAAA, EA Principal and Consulting Actuary :f• Joel E. Stewart, EA, MAAA Consulting Actuary April 2009 January 1, 2009 Actuarial Valuation City of Anytown Retirement Plan ■ This work product was prepared solely for the City. It may not be appropriate to use for other purposes. Milliman M i I I i ma n does not intend to benefit and assumes no duty or liability to other parties that receive this work. April 28, 2010 Retirement Board City of Anytown 123 Street Denver, Colorado 80202 IDear Members of the Board In accordance with your request, we have completed an Actuarial Valuation of The City of ' Anytown Retirement Plan as of January 1, 2009 for determining contributions beginning January 1, 2009. The major findings of the valuation are contained in this report. This report reflects the benefit provision and contribution rates in effect as of January 1, 2009. In preparing this report, we relied, without audit, on information (some oral and some in writing) supplied by the City's staff. This information includes, but is not limited to, statutory provisions, employee data, and financial information. In our examination of these data, we have found them to be reasonably consistent and comparable with data used for other purposes. Since the valuation results are dependent on the integrity of the data supplied, the results can be ' expected to differ if the underlying data is incomplete or missing. It should be noted that if any data or other information is inaccurate or incomplete, our calculations may need to be revised. On the basis of the foregoing, we hereby certify that, to the best of our knowledge and belief, this report is complete and accurate and has been prepared in accordance with generally recognized and accepted actuarial principles and practices which are consistent with the Actuarial Standards of Practice promulgated by the Actuarial Standards Board and the ' applicable Guides to Professional Conduct, amplifying Opinions, and supporting Recommendations of the American Academy of Actuaries. ' We further certify that all costs, liabilities, rates of interest, and other factors for the System have been determined on the basis of actuarial assumptions and methods which are individually reasonable (taking into account the experience of the System and reasonable expectations); and which, in combination, offer our best estimate of anticipated experience affecting the City. We further certify that, in our opinion, each actuarial assumption used is reasonably related to the experience of the Plan and to reasonable expectations which, in combination, represent our best estimate of anticipated experience under the System. Nevertheless, the emerging costs will vary from those presented in this report to the extent actual experience differs from that projected by the actuarial assumptions. The City has the final decision regarding the appropriateness of the assumptions. Future actuarial measurements may differ significantly from the current measurements presented in this report due to such factors as the following: plan experience differing from that anticipated by the economic or demographic assumptions; changes in economic or JADOCUMENTS\Proposals\10 CFC\Sample Public Sector Val Report 2009.doc Retirement Board April 28, 2010 Page 2 demographic assumptions; increases or decreases expected as part of the natural operation of the methodology used for these measurements (such as the end of an amortization period or additional cost or contribution requirements based on the plan's funded status); and changes in plan provisions or applicable law. Due to the limited scope of our assignment, we did not perform an analysis of the potential range of future measurements. Actuarial computations presented in this report are for purposes of determining the recommended funding amounts for the City. Actuarial computations presented in this report under GASB Statements No. 25 and 27 are for purposes of fulfilling financial accounting requirements. The computations prepared for these two purposes may differ as disclosed in our report. The calculations in the enclosed report have been made on a basis consistent with our understanding of the City's funding requirements and goals. The calculations in this report have been made on a basis consistent with our understanding of the plan provisions described in Appendix B of this report, and of GASB Statements No. 25 and 27. Determinations for purposes other than meeting these requirements may be significantly different from the results contained in this report. Accordingly, additional determinations may be needed for other purposes. Milliman's work product was prepared exclusively for The City of Anytown for a specific and limited purpose. It is a complex, technical analysis that assumes a high level of knowledge concerning the City's operations, and uses City data, which Milliman has not audited. It is not for the use or benefit of any third party for any purpose. Any third party recipient of Milliman's work product who desires professional guidance should not rely upon Milliman's work product, but should engage qualified professionals for advice appropriate to its own specific needs. We would like to express our appreciation to City staff, who gave substantial assistance in supplying the data on which this report is based. I, Patricia Ann Kahle, am a consulting actuary for Milliman, Inc. I am a member of the American Academy of Actuaries and meet the Qualification Standards of the American Academy of Actuaries to render the actuarial opinion contained herein. We respectfully submit the following report, and we look forward to discussing it with you Sincerely, Patricia Ann Kahle, FSA, EA, MAAA Principal and Consulting Actuary PAK:fm J:\DOCUMENTS\Proposals\10 CFC\Sample Public Sector Val Report 2009.doc 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 TABLE OF CONTENTS SECTION PAGE 1 Executive Summary...............................................................................................1 2 Proposal Requirements.........................................................................................4 3 Standards for Supervising Actuaries......................................................................9 4 Professional Services Agreement Review...................................................11 5 General Information About Milliman.....................................................................14 APPENDICES A Proposed Project Team Biographies B Sample Actuarial Valuation Report C Milliman Public Sector Client Listing D Sample Research Publications E Sample Simulation Model Proposal to Provide Actuarial and Consulting Services for M i I I i ma n the City of Fort Collins General Employees' Retirement Plan THE CITY OF ANYTOWN RETIREMENT PLAN JANUARY 1, 2009 ACTUARIAL VALUATION TABLE OF CONTENTS SECTION PAGE 1 SUMMARY OF THE FINDINGS...................................................................................1 ' 2 SCOPE OF THE REPORT.........................................................................................6 3 PARTICIPANT DATA................................................................................................7 Table 1 Reconciliation of Participants...............................................................8 Table 2 Summary of Active Participants...........................................................9 Table 3 Summary of Deferred Vested Participants........................................11 Table 4 Summary of Retirees and Beneficiaries............................................12 Table 5 Impact of Inflation on Retired Participants.........................................13 4 FINANCIAL DATA..................................................................................................14 Table 6 Change in Market Value of Assets ................................................ 15 Table 7 Development of Actuarial Value of Assets ..................................... Table 8 Summary Market Value Assets 16 17 of of ............................................. 5 DEVELOPMENT OF ANNUAL REQUIRED CONTRIBUTION.........................................18 ' Table 9 Unfunded Actuarial Liability..............................................................20 Table 10 Development of Annual Required Contribution................................21 6 GASB No. 25 AND 27 DISCLOSURE INFORMATION...............................................22 Table 11 Calculation of Net Pension Obligation and Annual Pension Cost...............................................................24 Table 12 Schedule of Funding Progress..........................................................25 Table 13 Schedule of Employer Contributions.................................................26 7 HISTORY AND PROJECTIONS................................................................................27 Table 14 Historical Statistics.............................................................................28 Table 15 Twenty -Year Projection of Benefit Payments....................................29 APPENDICES A ACTUARIAL PROCEDURES AND ASSUMPTIONS......................................................30 BPLAN SUMMARY..................................................................................................33 January 1, 2009 Actuarial Valuation City of Anytown Retirement Plan ■ This work product was prepared solely for the City. It may not be appropriate to use for other purposes. Milliman M i I I i ma n does not intend to benefit and assumes no duty or liability to other parties that receive this work. I IJ F1 THE CITY OF ANYTOWN RETIREMENT PLAN SECTION 1 SUMMARY OF THE FINDINGS This is a report of our actuarial valuation of the Plan as of January 1, 2009. The purpose and findings of our valuation are as follows: 1. Highlights. The following table summarizes some of the key results of our valuation of the Plan, along with the comparable figures from the prior year valuation. January 1, 2009 January 1, 2008 Market Value of Assets (MV) $ 44,393,850 $ 59,559,469 Actuarial Value of Assets (AV) 53,272,620 56,899,727 Ratio of AV to MV 120.0% 95.5% Discount Rate for Liabilities 8.00% 8.00% Present Value of Future Benefits $ 96,491,448 $ 91,331,295 Present Value of Future Normal Costs 17,158,464 16,860,266 79,332,984 74,471,029 Entry Age Normal Actuarial Liability (EANAL) Funded Ratio (AV = EANAL) 67.2% 76.4% Funded Ratio (MV = EANAL) 56.0% 80.0% Unfunded Actuarial Liability 26,060,364 17,571,302 Annual Required Contribution (ARC)*: at Beginning of Year as of May 15t 3,200,392 3,283,556 3,000,717 3,078,692 as a Percent of Compensation 15.37% 14.75% * ARC is net of employee contributions. The Government Accounting Standards Board (GASB) measures the adequacy of plan funding by comparing the value of plan assets to the actuarial accrued liability at the valuation date. The actuarial value of plan assets covers 67.2% of the plan's actuarial liability under the Entry Age Normal Cost method, and the market value of plan assets covers 56.0% of the plan's actuarial liability. This is a decrease from the January 1, 2008 valuation when the actuarial value of plan assets covered 76.4% of the actuarial liability, due mainly to asset losses. The Annual Required Contribution (ARC) increased as a percentage of compensation from 14.75% to 15.37%. This increase is also due mainly to asset losses. The 2008 percent of compensation has been revised from the prior report to reflect compensation for those participants under the assumed retirement age. January 1, 2009 Actuarial Valuation City of Anytown Retirement Plan ■ This work product was prepared solely for the City. It may not be appropriate to use for other purposes. Milliman M i I I iman does not intend to benefit and assumes no duty or liability to other parties that receive this work. i� 2. Analyze recent plan experience and select appropriate actuarial assumptions. ' To value the Plan, the actuary must predict future events such as investment return, mortality, and rates of termination by means of "actuarial assumptions." The validity of our valuation depends on how closely future Plan experience follows our assumptions. ' Experience different from that assumed gives rise to actuarial gains or losses, which affect future costs. The actuarial assumptions we used in this valuation are stated in Appendix A. The following comments address some of the more important assumptions. a. Rate of investment return i� 1 11 Beginning with the 2003 actuarial valuation, the assumed future investment return rate is 8.0%. A study of actual investment performance on the market value of assets produced the following results: Annual Rate of Investment Return For Period Ending For One -Year Period December 31, 2008 Ending Average December 31 Annual Rate* Period Annual Rate 2008 (25.7)% 1 year (25.7)% 2007 8.1 2 years (10.4) 2006 14.8 3 years (2.7) 2005 9.9 4 years 0.3 2004 11.7 5 years 2.5 2003 23.1 6 years 5.7 2002 (7.6) 7 years 3.7 2001 (4.7) 8 years 2.6 2000 (1.8) 9 years 2.1 1999 17.2 10 years 3.5 1998 20.0 11 years 4.9 1997 22.0 12 years 6.2 1996 13.9 13 years 6.8 1995 27.5 14 years 8.2 1994 (2.5) 15 years 7.4 1993 8.0 16 years 7.5 1992 8.2 17 years 7.5 1991 23.1 18 years 8.3 1990 (7.2) 19 years 7.4 1989 14.5 20 years 7.8 1988 15.0 21 years 8.1 1987 0.7 22 years 7.8 1986 19.0 23 years 8.2 Rates of return for 1997 and earlier as reported by the prior actuary and used without audit. Rates of return are computed net of all expenses. January 1, 2009 Actuarial Valuation City of Anytown Retirement Plan ■ This work product was prepared solely for the City. It may not be appropriate to use for other purposes. Milliman M i I I i ma n does not intend to benefit and assumes no duty or liability to other parties that receive this work. 2 fl These historical returns vary by the time period reviewed. The five-year time period averages are under the expected 8.0% return, as shown below: Annual Rate of Investment Return For Period Ending December 31, 2008 5 years 2.5% 10 years 3.5 15 years 7.4 20 years 7.8 However, historical returns are not indicative of future results. One of the most important assumptions in an actuarial valuation is the investment ' return assumption, which represents the expected long-term rate of return on plan assets. Actuarial Standard of Practice (ASOP) No. 27 provides guidance to actuaries on selecting assumptions for measuring obligations under defined benefit pension ' plans. ASOP No. 27 instructs actuaries to consider recent and long-term historical economic data, and also explicitly advises against giving undue weight to recent experience. Recognizing that there is not one "right answer", ASOP No. 27 calls for the actuary to develop a best estimate range for the valuation investment return assumption, and then to recommend a specific point within that range. Milliman's investment practice 1 has developed a model that provides a best estimate range of gross rates of return based on a plan's asset allocation and certain assumptions regarding future economic experience. 'Based on this model and a 30-year time horizon, we have developed the following estimate range: Expected ' Percentile Rate of Return 25th 6.22% 50th 7.55% 75th 8.90% In other words, 50% of the time the annual return will be higher than 7.55%, and the other 50% of the time the return will be under 7.55%. Also, please keep in mind that expenses for this plan decrease the expected rate of return shown in the table above by approximately 0.3%, so the 50th percentile in the table above would be adjusted to approximately 7.25% net of expenses, well below the 8.0% assumed. We believe that this assumption should be reviewed and possibly lowered for the 2010 actuarial valuation. b. Termination and retirement rates ' Members who are not eligible for Rule of 80 retirement are expected to retire at age 65. Graded retirement rates based on age are used for participants eligible for the Rule of 80 starting at age 55. ' Graded withdrawal rates prior to assumed retirement age are also used. 1 January 1, 2009 Actuarial Valuation City of Anytown Retirement Plan This work product was prepared solely for the City. It may not be appropriate to use for other purposes. Milliman LLA M i I I i ma n does not intend to benefit and assumes no duty or liability to other parties that receive this work. 3 L_ I Based on the assumptions used for the 2008 actuarial valuation, 12 of the active ' members on January 1, 2008 were projected to withdraw from covered employment during 2008 prior to their assumed retirement age, and a total of 23 active members actually terminated before reaching their assumed retirement age during the year. Seven active participants were expected to retire during 2008, and six participants actually retired from active status during 2008. Recent withdrawal experience reflects higher withdrawal rates than expected. We recommend an experience study of both the retirement and withdrawal assumptions prior to the 2010 valuation to determine if changes to these assumptions are needed. c. Salary increase assumption The assumed rate of future salary increases used for the valuation of the plan is 5.0%. Average salary increases for those participants continuing in covered employment and working full time was 4.6%. The median salary increase from 2007 to 2008 was 3.8%. Recent retirement experience reflects actual salary increases that vary significantly from those expected. This could be due to the underlying inflation assumption, productivity assumption, or merit assumption. We recommend an experience study of the salary increases prior to the 2010 valuation to determine if changes to this assumption are needed for the next valuation. 3. Review the financial effect of experience gains and losses and changes in plan ' benefits. The plan's unfunded actuarial liability increased from $17,571,302 on January 1, 2008 to $26,060,364 on January 1, 2009, an increase of $8,489,062. We expected the unfunded actuarial liability to decrease by $228,928, to $17,342,374, if all our assumptions were met and no changes to the plan or assumptions were made. The most significant increase in the unfunded actuarial liability is the recognition of investment losses from 2008. The unfunded actuarial liability is $8,717,990 larger than expected due to the following: Unfunded Actuarial Liability, January 1, 2008 $ 17,571,302 ( Expected Unfunded Actuarial Liability, January 1, 2009 $ 17,342,374 Changes During 2008 1 Investment (gains)/losses from prior periods $ 8,328,347 O Salary changes different than assumed (192,724) New members 70,895 © Lump sum payments (19,576) Withdrawal Experience (9,382) Retirement Experience 65,268 a Mortality Experience 232,533 o Miscellaneous demographic (gains)/losses 242,629 Total Changes $ 8,717,990 Unfunded Actuarial Liability, January 1, 2009 $ 26,060,364 January 1, 2009 Actuarial Valuation City of Anytown Retirement Plan ■ This work product was prepared solely for the City. It may not be appropriate to use for other purposes. Milliman M i I I i ma n does not intend to benefit and assumes no duty or liability to other parties that receive this work. 1� 4. Calculate the plan's Annual Required Contribution and provide the disclosure information required by Government Accounting Standards Board Statement (GASB) No. 25 and No. 27. Section 5 provides the detail on the calculation of the Annual Required Contribution to the plan for the 2008 plan year. Section 6 sets forth certain information required for the plan's disclosures under GASB No. 25 and No. 27. Highlights of the 2008 plan year • There were 619 members reported on January 1, 2009, 347 of who were active members who continue to accrue benefits under the plan. The remaining 272 were inactive members retaining or currently drawing benefits under the plan. This represents a small increase in the number of participants compared to 601 participants on January 1, 2008. o The plan assets earned (25.7)% during the 2008 plan year on a market value basis, well below the 8.0% assumed rate of return. o At the end of 2008, the market value of fund assets was $44,393,850, and the actuarial value of fund assets was $53,272,620. As of January 1, 2009, the actuarial value of assets covered 67.2% of the plan's actuarial liability and the market value of assets covered 56.0% of the actuarial liability. o Employee contributions for 2008 totaled $409,298. ' o The annual recommended contribution for 2009 is $3,200,392 as of January 1, 2009 and $3,283,556 adjusted for interest through May 1, 2009. 1 A 11 January 1, 2009 Actuarial Valuation City of Anytown 1 Retirement Plan ■ This work product was prepared solely for the City. It may not be appropriate to use for other purposes. Milliman M i I I iman does not intend to benefit and assumes no duty or liability to other parties that receive this work. ' 5 THE CITY OF ANYTOWN RETIREMENT PLAN SECTION 2 SCOPE OF THE REPORT Section 1 of this report presents a summary of the findings resulting from this valuation. Section 3 contains a summary of the data we received regarding participants of the plan. Section 4 describes the basis we use in assigning values to the Fund's assets, and contains various summaries of the assets in Tables 6 through 8. Section 5 expands upon Section 1 in various areas of our findings, and describes the actuarial 1 concepts and methods upon which the findings are based. Tables 9 and 10 present the related calculations. Section 6 provides the required GASB No. 25 and No. 27 disclosure information. 1 Section 7 provides historical statistics of the Plan and a 20-year projection of benefit payments. All of the calculations of the valuation were carried out using certain assumptions as to the future experience of the plan. Appendix A summarizes these assumptions. Appendix B outlines the benefit and contribution provisions of the plan as of January 1, 2009. January 1, 2009 Actuarial Valuation City of Anytown Retirement Plan ■ This work product was prepared solely for the City. It may not be appropriate to use for other purposes. Milliman M i I I i ma n does not intend to benefit and assumes no duty or liability to other parties that receive this work. THE CITY OF ANYTOWN RETIREMENT PLAN SECTION 3 PARTICIPANT DATA The actuarial valuation of the plan is based on the participant data provided to us by the City. In total, 619 participants were reported to us and included in this valuation. The data includes active participants (347), deferred vested participants who retain benefits under the plan (172), and retirees and beneficiaries receiving benefits as of January 1, 2009 (100). Table 1 includes a reconciliation of the participant data from January 1, 2008 to January 1, 2009. The age and service characteristics of the 347 active participants in the plan as of January 1, ' 2009 are shown in Table 2. As indicated in Table 2, the average age of the active participants on the valuation date was 47.5, slightly higher than the average age of 47.1 of the active participants on January 1, 2008. The average years of service of the active participants on January 1, 2009 was 11.8, a slight increase from the 11.7 average years of service of the active participants on January 1, 2008. I J In addition to the active members, there were 272 inactive participants with benefits under the plan, 172 of which are not yet in pay status, but retain rights to future benefits under the plan. Table 3 contains a summary of the inactive participants not yet in pay status and the amounts of their benefits. Table 4 summarizes the information provided on the 100 members and beneficiaries who are currently receiving monthly benefits. Table 5 illustrates the impact of inflation on the retiree benefits since the date of the last ad hoc increase, which was effective January 1, 2005. The number of active employees has remained relatively stable while the overall number of plan participants has grown slowly over the last several years, as indicated in the chart below: 70 60 50 40 30 20 10 Plumber of Plan Participants 1999 2001 2003 2005 2007 2009 ©Active Employees [DRetirees ffiVested Inactives January 1, 2009 Actuarial Valuation City of Anytown Retirement Plan ■ This work product was prepared solely for the City. It may not be appropriate to use for other purposes. Milliman M i I I i ma n does not intend to benefit and assumes no duty or liability to other parties that receive this work. 11 7 THE CITY OF ANYTOWN RETIREMENT PLAN TABLE 1 RECONCILIATION OF PARTICIPANT DATA (January 1, 2008 to January 1, 2009) Actives Deferred Vested Retired Total Included in January 1, 2008 valuation: 346 163 92 601 Change due to: New entrants 29 0 N/A 29 Rehired 1 (1) 0 0 Termination Nonvested (9) N/A N/A (9) Vested (14) 14 N/A 0 Retirement (6) 0 6 0 Death 0 (2)* 1** (1) Cash out 0 (2) 0 (2) Other 0 0 1 *** 1 Net change 1 9 8 18 Included in January 1, 2009 valuation: 347 172 100 619 * Includes one death benefit paid as a lump sum. ** Includes one new beneficiary from deferred vested, and one new beneficiary from retired status. ***Includes one QDRO who was paid a lump sum in January 2009. January 1, 2009 Actuarial Valuation City of Anytown Retirement Plan ■ This work product was prepared solely for the City. It may not be appropriate to use for other purposes. Milliman M i I I i man does not intend to benefit and assumes no duty or liability to other parties that receive this work. 8 �■I� I� � �r � � i� III � � � � I. � w � � � I� THE CITY OF ANYTOWN RETIREMENT PLAN TABLE 2 SUMMARY OF ACTIVE PARTICIPANTS AS OF JANUARY 1, 2009 Years of Service 1 to 5 5 to 10 10 to 15 15 to 20 20 to 25 25 to 30 30 & Up Total Age Number of Participants Under 25 6 0 0 0 0 0 0 6 25 to 29 7 2 0 0 0 0 0 9 30 to 34 10 8 2 0 0 0 0 20 35 to 39 15 16 6 3 0 0 0 40 40 to 44 11 15 10 7 7 0 0 50 45 to 49 11 11 8 10 18 2 0 60 50 to 54 19 14 10 9 13 5 3 73 55 to 59 9 14 11 5 8 5 1 53 60 to 64 2 4 2 3 6 2 10 29 65&Up 1 1 1 1 1 1 1 7 TOTALS 91 85 50 38 53 15 15 347 Total Compensation Under 25 194,870 0 0 0 0 0 0 194,870 25 to 29 280,485 72,314 0 0 0 0 0 352,799 30 to 34 417,183 493,194 123,839 0 0 0 0 1,034,216 35 to 39 786,730 940,546 371,793 193,679 0 0 0 2,292,748 40 to 44 438,012 865,407 645,784 524,882 522,684 0 0 2,996,769 45 to 49 517,704 591,804 509,342 668,945 1,330,364 167,309 0 3,785,468 50 to 54 813,205 685,550 676,513 666,897 943,429 496,335 192,947 4,474,876 55 to 59 454,820 826,863 628,645 384,424 669,423 406,357 66,285 3,436,817 60 to 64 92,288 178,805 162,313 213,837 450,827 136,091 789,021 2,023,182 65&Up 36,426 84,207 59,035 67,567 66,494 64,785 94,820 473,334 TOTALS 4,031,723 4,738,690 3,177,264 2,720,231 3,983,221 1,270,877 1,143,073 21,065,079 January 1, 2009 Actuarial Valuation City of Anytown Retirement Plan ■ This work product was prepared solely for the City. It may not be appropriate to use for other purposes. Milliman does not intend to benefit and assumes no duty or liability to other M i l l i ma n parties that receive this work. 9 A® 0 � j � l Millima 1099 18" Street, Suite 3100 Denv er, CO 80202 �n ��� USA ' Tel +1 303 299 9400 Fax +1 303 299 9018 milliman.com i April 29, 2010 ' General Employees' Retirement Committee City of Fort Collins GERP 215 North Mason Street, 2nd Floor Fort Collins, CO 80522-0580 Dear Retirement Committee: tIn response to your Request for Proposal, we are pleased to present this proposal to provide actuarial consulting services to City of Fort Collins General Employees' Retirement Plan (GERP). We have attempted to be succinct, while providing sufficient information for an accurate review of our capabilities. If any issues are unclear, we will be glad to elaborate. As a Principal of the firm, I am authorized to make presentations for Milliman and am particularly pleased to have this opportunity to commit the firm to perform the required services within the time frames provided under the conditions outlined in this proposal. I certify that Milliman, the Principal Actuary, and our staff are qualified to provide the services presented in the Request for Proposal. We also believe we have offered you a team of exceptionally well -qualified consulting actuaries with more than 75 years of experience working with public sector plans. All of our clients have taken the same thoughtful and careful consideration in choosing their actuarial services provider as you are now taking. We invite you to talk to any one of them to see if they would choose Milliman again. We believe that if you discuss our consulting services with any of our current clients, they will confirm our commitment to quality, the integrity of our work, and the responsiveness to our client's needs. We believe our references are our best marketing tool. We have attempted to provide you with the information you requested and at the same time to keep this proposal comparatively brief. We believe we provide a high quality service for a reasonable fee, not only for the fixed fee regularly scheduled work, but for all other special projects as well. Again, we suggest you check this statement with our current clients. tTo the best of my knowledge, all representations made in this proposal are true. Furthermore, Milliman does not have any conflict of interest with respect to this proposed contract that would cause, or appear to cause, impairment to our objectivity to perform the assignments requested. Milliman is committed to our public employee retirement system practice, and we hope you will consider our proposal to be responsive to the needs of the GERP's Members, Retirement LCommittee, and the public. JADOCUMENTS\Proposals\10 CFC\10 COFC GERP.doc THE CITY OF ANYTOWN RETIREMENT PLAN TABLE 2 (continued) SUMMARY OF ACTIVE PARTICIPANTS AS OF JANUARY 1, 2009 Years of Service 1 to 5 5 to 10 10 to 15 15 to 20 20 to 25 25 to 30 30 & Up Total Age Average Compensation Under 25 32,478 0 0 0 0 0 0 32,478 25 to 29 40,069 36,157 0 0 0 0 0 39,200 30 to 34 41,718 61,649 61,920 0 0 0 0 51,711 35 to 39 52,449 58,784 61,966 64,560 0 0 0 57,319 40 to 44 39,819 57,694 64,578 74,983 74,669 0 0 59,935 45 to 49 47,064 53,800 63,668 66,895 73,909 83,655 0 63,091 50 to 54 42,800 48,968 67,651 74,100 72,571 99,267 64,316 61,300 55 to 59 50,536 59,062 57,150 76,885 83,678 81,271 66,285 64,846 60 to 64 46,144 44,701 81,157 71,279 75,138 68,046 78,902 69,765 65&Up 36,426 84,207 59,035 67,567 66,494 64,785 94,820 67,619 TOTALS 44,305 55,749 63,545 71,585 75,155 84,725 76,205 60,706 HISTORICAL SUMMARY 2004 2005 2006 2007 2008 2009 Number 325 319 325 329 346 347 Total Compensation $17,670,232 $17,571,907 $18,682,569 $19,727,497 $20,395,146 $21,065,079 Average Rate of Pay: $54,370 $55,084 $57,485 $59,962 $58,946 $60,706 Average Monthly Accrued Benefit: $1,245 $1,250 $1,298 $1,344 $1,261 $1,332 Average Service: 13.0 12.7 12.9 12.8 11.7 11.8 Average Age: 46.5 46.7 47.4 47.4 47.1 47.5 January 1, 2009 Actuarial Valuation City of Anytown Retirement Plan ■ This work product was prepared solely for the City. It may not be appropriate to use for other purposes. Milliman does not intend to benefit and assumes no duty or liability to other M i l l i ma n parties that receive this work. 10 THE CITY OF ANYTOWN RETIREMENT PLAN TABLE 3 SUMMARY OF DEFERRED VESTED PARTICIPANTS AS OF JANUARY 1, 2009 Age Number Total Annual Benefit Average Monthly Benefit Under 30 0 $ 0 $ 0 30-34 3 15,072 419 35-39 13 58,356 374 40-44 21 138,828 551 45-49 43 235,428 456 50-54: 43 265,428 515 55-59 28 184,044 548 60-64 20 105,960 442 65 & Up 1 2,520 210 Total 172 $1,005,696 $ 487 HISTORICAL SUMMARY 2004 2005 2006 2007 2008 2009 Number 151 155 156 156 163 172 Total Annual Benefit $661,649 $749,396 $765,444 $782,688 $985,632 $1,005,696 Average Monthly Benefit: $360 $403 $409 $418 $504 $487 Average Age: 47.8 48.6 49.3 49.6 50.2 50.6 January 1, 2009 Actuarial Valuation City of Anytown Retirement Plan ■ This work product was prepared solely for the City. It may not be appropriate to use for other purposes. Milliman M i I I i ma n does not intend to benefit and assumes no duty or liability to other parties that receive this work. 11 i 1 1 1 1 1 1 1 1 1 1 THE CITY OF ANYTOWN RETIREMENT PLAN TABLE 4 SUMMARY OF RETIREES AND BENEFICIARIES AS OF JANUARY 1, 2009 Age Number of Inactives Annual Amount of Benefit Average Monthly Benefit Under 50 0 $ 0 $ 0 50-54 2 48,396 2,017 55-59 11 404,340 3,063 60-64 20 678,132 2,826 65-69 33 1,552,500 3,920 70-74 20 349,020 1,454 75 & Up 14 142,992 851 Total 100 $ 3,175,380 $ 2,646 HISTORICAL SUMMARY 2004 2005 2006 2007* 2008 2009 Number 57 70 75 83 92 100 Total Annual Benefit $1,134,388 $1,657,037 $1,910,223 $2,078,892 $2,479,512 $3,175,380 Average Monthly Benefit: $1,658 $1,973 $2,122 $2,087 $2,246 $2,646 Average Age: 67.3 66.9 67.3 67.4 67.8 67.9 * Excludes 2 participants paid lump sums during January 2007. January 1, 2009 Actuarial Valuation City of Anytown Retirement Plan ■ This work product was prepared solely for the City. It may not be appropriate to use for other purposes. Milliman M i I I i ma n does not intend to benefit and assumes no duty or liability to other parties that receive this work. 12 THE CITY OF ANYTOWN RETIREMENT PLAN TABLE 5 IMPACT OF INFLATION ON RETIRED PARTICIPANTS HISTORY OF CPI SINCE JANUARY 1, 2005 Year Annual Increase Cumulative Increase 2005 3.4% 3.4% 2006 2.5% 6.0% 2007 4.1 % 10.3% 2008 0.1 % 10.4% COST OF INCREASING BENEFITS TO RETIREES AND BENEFICIARIES ON JANUARY 1, 2009 TO ADJUST FOR CPI INCREASE Liabilities Increase Current Plan 26,789,743 One-time benefit increase to reflect CPI increases since January 1, 2005: 28,693,864 1,904,121 AMORTIZATION OF BENEFIT INCREASE This shows the estimated increase in the payments to the Annual Required Contribution (ARC) for 2009 to 2013 if the benefits to retirees and beneficiaries were increased to adjust for CPI increases. The ARC payments will continue until the plan is 100% funded. Year Increase in Unfunded Actuarial Liability due to CPI Increase Amortization Period Additional Payment to ARC for 2009 CPI Increase 2009 $1,904,121 20 $122,794 2010 1,923,833 20 124,065 2011 1,943,749 15 156,667 2012 1,930,049 15 155,563 2013 1,916,445 15 154,467 January 1, 2009 Actuarial Valuation City of Anytown Retirement Plan ■ This work product was prepared solely for the City. It may not be appropriate to use for other purposes. Milliman M i I I i ma n does not intend to benefit and assumes no duty or liability to other parties that receive this work. 13 1 1 1 THE CITY OF ANYTOWN RETIREMENT PLAN SECTION 4 FINANCIAL DATA Table 6 presents a summary of the changes in the market value of assets. During 2008, the market value of the plan assets decreased by $15,165,619. Benefit payments for 2008 were $3,344,469. The amount of benefits paid in 2008 represents a decrease $725,509 from the amount of benefits paid in 2007. For the purpose of the actuarial valuation, the investment income of the Fund consists of interest and dividends as well as realized and unrealized gains and losses on investments. During 2008, the fund earned (25.7)% on a market value basis, net of expenses, and earned (6.6)% on an actuarial value basis. The fund returned 8.1 % on a market value basis for 2007. Historical rates of return on assets for the last 23 years are displayed in the chart below: For funding purposes, investment returns that differ from our 8.0% assumed rate of return are Return on Assets (Net of Expenses) 30% 20% 10% 0% -10% 4P 4% 44 P`1• R� °.1�° 44 44 4ti 4�` 4ZP 44 ►q Nq Noi N°5 Noi Noi NC ti4 11P 11P ti4 tip —Assumed Return ®Actual Return recognized over a five-year period. The development of the actuarial value of assets is shown in Table 7. A summary of the assets by fund is shown in Table 8. January 1, 2009 Actuarial Valuation City of Anytown Retirement Plan G■ This work product was prepared solely for the City. It may not be appropriate to use for other purposes. Milliman M i I I i ma n does not intend to benefit and assumes no duty or liability to other parties that receive this work. 14 THE CITY OF ANYTOWN RETIREMENT PLAN TABLE 6 CHANGE IN MARKET VALUE OF ASSETS FOR THE PLAN YEARS 2008 AND 2007 2008 2007 Market value at beginning of year $ 59,559,469 $56,242,902 Income: Employer Contributions $ 3,078,692 $ 2,802,715 Employee Contributions 409,298 96,480 Investment income (15,105,751) 4,725,008 Total $(11,617,761) $ 7,624,203 Disbursements: Benefit payments $ 3,344,469 $ 4,069,978 Investment related expenses 140,549 180,198 Operating expenses 62,840 57,460 Total $ 3,547,858 $ 4,307,636 Net change: $(15,165,619) $ 3,316,567 Market value at end of year $ 44,393,850 $59,559,469 January 1, 2009 Actuarial Valuation City of Anytown Retirement Plan This work product was prepared solely for the City. It may not be appropriate to use for other purposes. Milliman M i I I i ma n does not intend to benefit and assumes no duty or liability to other parties that receive this work. 15 THE CITY OF ANYTOWN RETIREMENT PLAN TABLE 7 DEVELOPMENT OF ACTUARIAL VALUE OF ASSETS FOR THE 2009 PLAN YEAR GAIN/(LOSS FOR 2008 AND 2007 2008 Market value at beginning of year Contributions Benefit payments Expected Interest: Expected market value at end of year: Actual market value at end of year: Gain/(Loss) for Year: $ 59,559,469 3,487,990 (3,344,469) 4,770,498 $ 64,473,488 $ 44,393,850 $ (20,079,638) ASSET GAIN/(LOSS) RECOGNIZED 2007 $ 56,242,902 2,899,195 (4,069,978) 4,452,601 $ 59,524,720 $ 59,559,469 $ 34,749 To Be Original Recognized Recognized in Recognized in Amount This Year Prior Years the Future 2004 1,470,390 294,078 1,176,312 0 2005 827,093 165,418 496,257 165,418 2006 3,345,047 669,009 1,338,018 1,338,020 2007 34,749 6,950 6,950 20,849 2008 (20,079,638) (4,015,928) 0 (16,063,710) $(14,402,359) $ (2,880,473) $ (14,539,423) Total $ 3,017,537 ACTUARIAL VALUE OF ASSETS 1. Market value as of January 1, 2009 2. Prior gains/(losses) to be recognized in the future 3. Initial Actuarial Value of Assets (1. - 2.) 4. 80% of Market Value 5. 120% of Market Value 6. Actuarial Value of Assets as of January 1, 2009 (3., but not less than 4. or greater than 5.) $44,393,850 (14,539,423) $58,933,273 $ 35, 515, 080 $53,272,620 $53,272,620 January 1, 2009 Actuarial Valuation City of Anytown Retirement Plan ■ This work product was prepared solely for the City. It may not be appropriate to use for other purposes. Milliman M i I I iman does not intend to benefit and assumes no duty or liability to other parties that receive this work. 16 THE CITY OF ANYTOWN RETIREMENT PLAN TABLE 8 SUMMARY OF MARKET VALUE OF ASSETS December 31. 2008 December 31. 2007 Invested assets at market value PIMCO Funds $ 12,261,017 Artio Global Funds 6,581,416 Julius Baer Funds 0 T Rowe Price 11,080,888 Vanguard Funds 6,550,677 Ironbridge Funds 1,753,144 Principal (Real Estate) 5,085,554 Money Market 1,081,154 Other 0 Total $ 44,393,850 Receivables (Accrued Income and $ 0 Benefit Corrections) Liabilities (Benefits Payable): $ 0 Total market value of assets $ 44,393,850 $ 13,831,833 0 10,779,177 14, 952, 898 9,930,806 2,439,116 6,096,258 1,527,531 0 $ 59,557,619 $ 1,850 $ 0 $ 59,559,469 January 1, 2009 Actuarial Valuation City of Anytown Retirement Plan ■ This work product was prepared solely for the City. It may not be appropriate to use for other purposes. Milliman M i I I i man does not intend to benefit and assumes no duty or liability to other parties that receive this work. 17 1 THE CITY OF ANYTOWN ' RETIREMENT PLAN SECTION 5 ' DEVELOPMENT OF ANNUAL REQUIRED CONTRIBUTION A fundamental principle in financing a benefit program is that the cost of its benefits should be related to when those benefits are earned, rather than to when they are paid. Various methods i are used by actuaries to determine costs that satisfy this principle. ■ The method used to determine the incidence of the contributions in various years is called the actuarial cost method. This actuarial valuation is prepared using the entry age normal cost method. Under this method, the contributions required to meet the difference between current assets and current actuarial liabilities are allocated each year between two elements: o A normal cost amount, which ideally is relatively stable as a percentage of salary over the ' years, and • An additional amount that is used to amortize what is called the unfunded actuarial liability. The normal cost is that portion of the cost allocated to the current period. Under the entry age normal method the normal cost is the theoretical contribution rate that will meet the ongoing ' costs of the active participants as a level percent of pay to create a fund which would exactly cover the benefits required to be paid. If experience follows the actuarial assumptions exactly and the normal cost is contributed each year, the fund would be completely liquidated with the last payment to the last surviving participant. If the normal costs had always been paid at this level in the past, a fund would have been built up that is called the actuarial liability. When a plan is first set up, it will generally have no assets, so that an actuarial liability at that time is completely unfunded. As contributions are made, the unfunded actuarial liability is gradually paid off. The amount of the unfunded actuarial liability at any point will be affected by experience that does not follow the actuarial assumptions exactly, ' benefit improvements, and the amount of the contributions made. When actuarial experience is favorable, as it has been for many years in this fund, the assets may exceed the value of the actuarial liability creating a negative UAL, or a funding excess. The funding excess functions as ' a reserve for future adverse experience and future normal costs. For our calculations, we calculate the actuarial liability of the plan in a little different method ' than described above. Instead of accumulating prior normal costs retrospectively, we calculate the present value of all benefits that are projected to become payable to participants and subtract the present value of future normal costs. The retrospective concept is probably more readily understood, but the prospective method is a better mathematical tool for deriving the actual results and leads to equivalent values. The actuarial liability of the plan as of January 1, 2009 consists of liabilities for retirees and beneficiaries, vested inactive members who retain rights to future benefits, and active members. As shown in the following chart, the majority of the actuarial liability is attributable to current active members. ITable 9 shows how the unfunded actuarial liability was derived for the plan. January 1, 2009 Actuarial Valuation City of Anytown Retirement Plan ■ This work product was prepared solely for the City. It may not be appropriate to use for other purposes. Milliman M i I I i man does not intend to benefit and assumes no duty or liability to other parties that receive this work. ' 18 Actuarial Liability Annual Required Contribution ' The plan's annual required contribution (ARC) is actuarially determined in accordance with Statement No. 25 of the Governmental Accounting Standards Board (GASB No. 25). The GASB No. 25 reporting requirements became effective for fiscal years beginning after June 15, 1996. ( Under the entry age normal cost method, GASB No. 25 requires the unfunded actuarial liability or funding excess to be amortized over a period not longer than 30 years. The ARC consists of this amortization component plus the normal cost determined under the cost method. For this valuation, we utilized a 20-year level -percent -of -pay amortization of the unfunded actuarial ' liability or funding excess to determine the amortization component of the ARC. The amortization period will decrease to 15 starting in 2011. Table 10 shows the derivation of the ARC for this valuation. I January 1, 2009 Actuarial Valuation City of Anytown Retirement Plan ■ This work product was prepared solely for the City. It may not be appropriate to use for other purposes. Milliman M i I I i man does not intend to benefit and assumes no duty or liability to other parties that receive this work. 19 1 1 K! Milkman General Employees' Retirement Board April 29, 2010 Page 2 If you have any questions, please contact me. We greatly appreciate the opportunity to present this proposal and look forward to hearing from you. Respectfully submitted, Patricia Ann Kahle, FSA, MAAA, EA Principal and Consulting Actuary Milliman, Inc. 1099 18t" Street, Suite 3100 Denver, Colorado 80202 Phone: 303.299.9400 Facsimile: 303.299.9018 Email: pat. kahle(a)-milliman.com Website: www.milliman.com PAK:wp jAdocuments\proposa1s\10 cfc\10 cofc gerp.doc Offices in Principal Cities Worldwide THE CITY OF ANYTOWN ' RETIREMENT PLAN TABLE 9 ' UNFUNDED ACTUARIAL LIABILITY Actuarial Liability as of January 1, 2009 Retired Participants $ 26,789,743 Vested Inactive Participants 4,196,505 Active Participants 48,346,736 Total $ 79,332,984 Actuarial Value of Assets $ 53,272,620 Unfunded Actuarial Liability as of January 1, 2009 $ 26,060,364 Liability January 1, 2009 Expected Unfunded Actuarial on Unfunded Actuarial Liability as of January 1, 2008 $ 17,571,302 Normal Cost 1,881,016 Employer Contributions (3,078,692) Employee Contributions (409,298) ' Interest 1,378,046 Expected, January 1, 2009 $ 17,342,374 Changes Assumption Change 0 ' Plan Change 0 Experience (Gain)/Loss Asset (Gain)/Loss 8,328,347 Salary (Gain)/Loss (192,724) New Member Loss 70,895 ' (Gain)/Loss from Lump Sum Payments (19,576) Withdrawal (Gain)/Loss (9,382) Retirement (Gain)/Loss 65,268 Mortality (Gain)/Loss 232,533 Other Demographic 242,629 Total 8,717,990 Unfunded Actuarial Liability on January 1, 2009 $ 26,060,364 January 1, 2009 Actuarial Valuation City of Anytown Retirement Plan ■ This work product was prepared solely for the City. It may not be appropriate to use for other purposes. Milliman M i I I i man does not intend to benefit and assumes no duty or liability to other parties that receive this work. 20 THE CITY OF AINYTOWN RETIREMENT PLAN TABLE 10 DEVELOPMENT OF ANNUAL REQUIRED CONTRIBUTION AS OF JANUARY 1, 2009 AND JANUARY 1, 2008 1. Present Value of Projected Benefits a. For retirees and beneficiaries b. For deferred vested participants c. For active participants d. Total [(a) + (b) + (c)] 2. Present Value of Future Normal Costs 3. Entry Age Normal Accrued Liability [(1) — (2)] 4. Actuarial Value of Assets 5. Unfunded Accrued Liability [(3) — (4)] 6. Entry Age Normal Cost* 7. 20-Year Amortization of Unfunded Accrued Liability 8. Expected Employee Contributions 9. Annual Required Contribution at Beginning of Year [(6)+(7)+(8)] 10. Annual Required Contribution as of May 1st January 1, 2009 January 1, 2008 $ 26,789,743 4,196,505 65,505,200 $ 96,491,448 17,158,464 $ 79,332,984 53,272,620 $ 26,060,364 $ 1,934,687 1,680,595 (414, 890) $ 3,200,392 $ 3,283,556 $ 24,875,880 4,273,160 62,182,255 $ 91,331,295 16, 860, 266 $ 74,471,029 56, 899, 727 $ 17, 571, 302 $ 1,867,569 1,133,148 N/A $ 3,000,717 $ 3,078,692* * The 2008 ARC of $3,078,692 was shown in the January 1, 2008 actuarial valuation. The 2008 ARC was recalculated in our April 13, 2009 letter to KPMG to be $2,677,410. At the auditor's direction we will not revise this number with this valuation, and we will make the adjustment in the January 1, 2010 actuarial valuation. January 1, 2009 Actuarial Valuation City of Anytown Retirement Plan ■ This work product was prepared solely for the City. It may not be appropriate to use for other purposes. Milliman M i I I i man does not intend to benefit and assumes no duty or liability to other parties that receive this work. 21 THE CITY OF ANYTOWN ' RETIREMENT PLAN SECTION 6 ' DISCLOSURE INFORMATION REQUIRED BY GOVERNMENT ACCOUNTING STANDARDS BOARD STATEMENT No. 25 AND NO. 27 ' Tables 11 and 12 provide the disclosure information required by Government Accounting Standards Board (GASB) Statement No. 25, Financial Reporting for Defined Benefit Pension Plans and Note Disclosures for Defined Contribution Plans, and Statement No. 27, Accounting for Pensions by State and Local Governmental Employees. The assumptions and methods used for these funding disclosures meet the parameters set forth in GASB No. 25 and No. 27, and, other than those described below, are outlined in Appendix A of this report. ' GASB 25 generally requires two schedules of historical trend information for the plan: a schedule of funding progress and a schedule of employer contributions as they relate to the plan's Annual Required Contribution (ARC). The chart below provides historical information on the plan's ARC. The amount of the ARC over this period has been affected by a change in the cost method effective for the 2000 plan year, a benefit improvement adopted July 1, 2000, a change in the interest rate assumption for the 2003 plan year, a change in the expected retirement rates for the 2004 plan year, a change in mortality and lump sum interest for the 2005 plan year, and ' cost of living adjustments for retirees effective January 1, 2000, 2001, 2003 and 2005, and the plan change requiring employee contributions effective July 8, 2007 and first reflected with the January 1, 2008 valuation. In addition, the method for calculating the ARC was changed ' effective for the 2009 plan year. The ARC calculation is now the total normal cost, plus amortization of the entire unfunded actuarial liability, minus expected employee contributions. At the direction of the City, we did not revise the 2008 ARC. ' Annual Required Contribution $3,500 $3,000 N $2,500 N $2,000 o $1,500 $1,000 $500 $0 II11.7%11, =, F II Ili -' 1 1 11 1 1 1 1 1 ,= , D-, 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 The schedule of funding progress is included as Table 12 while the schedule of employer contributions is included as Table 13. The pension cost is the ARC plus an amortization of the net pension obligation. The net pension obligation is the cumulative difference between the ARC and the employer contribution to the plan. In this context, the net pension obligation will decrease if contributions exceed the ARC and January 1, 2009 Actuarial Valuation City of Anytown Retirement Plan This work product was prepared solely for the City. It may not be appropriate to use for other purposes. Milliman M i I I i ma n does not intend to benefit and assumes no duty or liability to other parties that receive this work. ' 22 U will increase if the entire ARC is not contributed each year. Table 11 shows the calculation of the net pension obligation and annual pension cost. January 1, 2009 Actuarial Valuation City of Anytown Retirement Plan This work product was prepared solely for the City. It may not be appropriate to use for other purposes. Milliman M i I I iman does not intend to benefit and assumes no duty or liability to other parties that receive this work. 23 THE CITY OF ANYTOWN RETIREMENT PLAN TABLE 11 UALCULATION OF NET PENSION VBLIGATION AND PENSION UOST Fiscal Year Annual Required Contribution (ARC) Net Pension Obligation (NPO) as of January 1 Interest on NPO to End of Year Amortization Factor ARC Adjustment Annual Pension Cost (APC)(') Actual Employer Contribution Net Pension Obligation at End of Year 1991 $ 548,347 $ 0 $ 0 N/A $ 0 $ 548,347 $ 468,788 $ 79,599 1992 577,193 79,599 6,763 9.9483 7,997 575,958 593,104 62,414 1993 668,349 62,414 5,305 9.8169 6,358 667,296 686,773 42,937 1994 678,433 42,937 3,650 9.8168 4,374 677,709 686,773 33,873 1995 726,436 33,873 2,879 9.6768 3,500 725,815 727,668 32,020 1996 735,464 32,020 2,722 9.5636 3,348 734,838 735,464 31,394 1997 733,629 31,394 2,668 8.6972 3,610 732,687 735,464 28,617 1998 626,891 28,617 2,432 8.1452 3,513 625,810 735,464 (81,037) 1999 377,640 (81,037) (6,888) 8.2435 (9,830) 380,582 626,891 (327,346) 2000 797,389 (2) (327,346) (27,824) 19.4083 (16,866) 786,431 797,389 (338,304) 2001 1,177, 716 (338,304) (28,756) 19.4083 (17,431) 1,166,391 1,210,181 (382,094) 2002 1,279,611 (382,094) (32,478) 19.4083 (19,687) 1,266,820 1,314,885 (430,159) 2003 2,004,684 (430,159) (34,413) 20.5379 (20,945) 1,991,216 2,056,777 (495,720) 2004 2,035,549 (495,720) (39,658) 20.5379 (24,137) 2,020,028 2,088,444 (564,136) 2005 2,267,640 (564,136) (45,131) 20.5379 (27,468) 2,249,977 2,326,566 (640,725) 2006 2,571,953 (640,725) (51,258) 20.5379 (31,197) 2,551,892 2,638,787 (727,620) 2007 2,731,729 (727,620) (58,210) 18.6936 (38,923) 2,712,442 2,802,715 (817,893) 2008 (3) 3,078,692 (817,893) (65,431) 15.5066 (52,745) (4) 3,066,006 3,078,692 (830,579) (4) (1) The annual pension cost is sum of (1) the annual required contribution computed under the aggregate cost method for 1999 and prior and under the entry age normal cost method for 2000 and later, plus (2) interest on the net pension obligation as of the beginning of the year, less (3) an adjustment to the annual required contribution. The adjustment to the annual required contribution is an amortization of the net pension obligation as of the beginning of the year. For 1999 and prior, the amortization is calculated as level percent of the current members' future expected salaries. For 2000 and later, the amortization uses the same amortization methodology used in determining the ARC. Beginning January 1, 2003, the interest rate used changed from 8.5% to 8.0%. (2) Revised from January 1, 2000 actuarial valuation to reflect July 1, 2000 plan changes. (3) Beginning with the 2008 Fiscal Year, the ARC is calculated as of May 1, 2008, which is the date the contributions are actually made to the plan. (4) The 2008 ARC and December 31, 2008 NPO will be revised to $2,677,410 and ($1,231,861) respectively with the 2010 valuation report. Details of the revisions were provided in Milliman's letter to KPMG dated April 13, 2009. January 1, 2009 Actuarial Valuation City of Anytown Retirement Plan ■ This work product was prepared solely for the City. It may not be appropriate to use for other purposes. Milliman does not intend to benefit and assumes no duty or liability to other M l l l i ma n parties that receive this work. 24 = M M = M M = M = M I=1 = M = = = THE CITY OF ANYTOWN RETIREMENT PLAN TABLE 12 SCHEDULE OF FUNDING PROGRESS Actuarial Valuation Date Actuarial Value of Assets (AVA) Actuarial Accrued Liability (AAL) Unfunded AAL (UAAL) Funded Percentage (AVA - AAL) Covered Payroll UAAL as a Percent of Payroll January 1, 1995 $14,377,014 $14,377,014 $ 0 100% 13,052,052 0.0% January 1, 1996 17,273,001 17,273,001 0 100 13,059,353 0.0 January 1, 1997 19,675,518 19,675,518 0 100 13,077,860 0.0 January 1, 1998 24,110,767 24,110,767 0 100 13,082,039 0.0 January 1, 1999 26,763,400 26,763,400 0 100 13,138,566 0.0 January 1, 2000 31,591,829 26,508,863 (5,082,966) 119 13,821,820 (36.8) January 1, 2001 35,290,307 38,073,318 2,783,011 93 14,521,027 19.2 January 1, 2002 38,636,901 42,064,688 3,427,787 92 15,183,008 22.6 January 1, 2003 38,197,080 50,659,141 12,462,061 75 17,186,164 72.5 January 1, 2004 41,937,007 53,863,674 11,926,667 78 17,670,232 67.5 January 1, 2005 43,405,751 58,884,036 15,478,285 74 17,571,907 88.1 January 1, 2006 46,634,331 65,365,669 18,731,338 71 18,682,569 100.3 January 1, 2007 51,513,297 69,889,905 18,376,608 74 19,727,497 93.2 January 1, 2008 56,899,727 74,471,029 17,571,302 76 20,395,146 86.2 January 1, 2009 53,272,620 79,332,984 26,060,364 67 20,827,017 125.1 (1) The funding method for 1999 and prior was the aggregate cost method, a cost method for which there is no actuarial accrued liability. The values shown for 1999 and prior in this column represent the actuarial value of assets. The actuarial accrued liability as of January 1, 2000 and later is calculated under the entry age normal cost method. The interest rate assumption was changed to 8.0% from 8.5% as of January 1, 2003. January 1, 2009 Actuarial Valuation City of Anytown Retirement Plan This work product was prepared solely for the City. It may not be appropriate to use for other purposes. Milliman does not intend to benefit and assumes no duty or liability to other M M i I I iman parties that receive this work. 25 THE CITY OF ANYTOWN RETIREMENT PLAN TABLE 13 SCHEDULE OF EMPLOYER CONTRIBUTIONS Year Ended Annual Required Contribution Employer Contribution Percentage Contributed December 31, 1991 $ 548,347 $ 468,788 85.5% December 31, 1992 577,193 593,104 102.9% December 31, 1993 668,349 686,773 102.9% December 31, 1994 678,433 686,773 101.3% December 31, 1995 726,436 727,668 100.3% December 31, 1996 735,464 735,464 100.0% December 31, 1997 733,629 735,464 100.4% December 31, 1998 626,891 735,464 117.5% December 31, 1999 377,640 626,891 166.0% December 31, 2000 797,389 797,389 100.0% December 31, 2001 1,177,716 1,210,181 * 102.8% December 31, 2002 1,279,611 1,314,885* 102.8% December 31, 2003 2,004,684 2,056,777* 102.6% December 31, 2004 2,035,549 2,088,444* 102.6% December 31, 2005 2,267,640 2,326,566* 102.6% December 31, 2006 2,571,953 2,638,787* 102.6% December 31, 2007 2,731,729 2,802,715* 102.6% December 31, 2008 1 3,078,692** 3,078,692* 100.0% * The City contributes an amount equal to the Annual Required contribution increased with interest to May 1 of each year. ** Beginning with the 2008 fiscal Year, the ARC is calculated as of May 1, 2008. The 2008 ARC will be revised to $2,677,410 and the Percentage Contributed will be 115.0% in the 2010 valuation. Details of the change are presented in Milliman's letter to KPMG dated April 13, 2009. January 1, 2009 Actuarial Valuation City of Anytown Retirement Plan o This work product was prepared solely for the City. It may not be appropriate to use for other purposes. Milliman M i I I iman does not intend to benefit and assumes no duty or liability to other parties that receive this work. 26 ' THE CITY OF ANYTOWN RETIREMENT PLAN SECTION 7 HISTORY AND PROJECTIONS Table 14 shows five years of the more important Plan statistics. ' o Assets: Comparison of the Market Value of Assets to the Actuarial Value of Assets, including the rate of return on the Market Value of Assets and Actuarial Value of Assets Investment return often represents the largest source of actuarial gain or loss. tC Contribution: Summary of the annual required contribution, calculated pursuant to GASB No. 25. F1, I 1 • Present Value of Projected Benefits: Present value of projected plan benefits, by participant status. Also included is the actuarial liability, calculated under the entry age normal cost method. • Participant Statistics: Changes, if any, in the active and inactive participants' characteristics over time can cause significant changes in costs. Assumptions: Summary of the interest and mortality assumptions used. Table 15 provides a projection of benefit payments over the next twenty years. This can be useful for the investment manager in planning future liquidity requirements, although this exhibit does not attempt to predict future lump sum payments to terminating employees. January 1, 2009 Actuarial Valuation City of Anytown Retirement Plan This work product was prepared solely for the City. It may not be appropriate to use for other purposes. Milliman M i I I iman does not intend to benefit and assumes no duty or liability to other parties that receive this work. 27 11 n THE CITY OF ANYTOWN RETIREMENT PLAN TABLE 14 HISTORICAL STATISTICS 2009 2008 2007 2006 2005 Assets Market Value of Assets $ 44,393,850 $ 59,559,469 $ 56,242,902 $ 49,012,327 $ 44,332,679 Actuarial Value of Assets $ 53,272,620 $ 56,899,727 $ 51,513,297 $ 46,634,331 $ 43,405,751 Ratio, Actuarial to Market 120.0% 95.5% 91.6% 95.1 % 97.9% Market Value Return (25.7)% 8.1 % 14.8% 9.9% 11.7% Actuarial Value Return (6.6)% 12.9% 10.5% 6.7% 3.9% Annual Required Contribution As of the beginning of the year $ 3,200,392 $ 3,000,717* $ 2,731,729 $ 2,571,953 $ 2,267,640 As a percent of payroll 15.37% 14.75% 13.85% 13.77% 12.90% Contributions Made Employer Unknown $ 3,078,692 $ 2,802,715 $ 2,638,787 $ 2,326,566 Employee Unknown $ 409,298 $ 96,480 N/A N/A Present Value of Projected Benefits For retirees and beneficiaries $ 26,789,743 $ 24,875,880 $ 21,303,890 $ 19,168,052 $ 16,743,690 For deferred vested participants 4,196,505 4,273,160 2,678,868 2,633,489 2,540,669 For active participants 65,505,200 62,182,255 61,705,763 58,642,446 53,540,073 $ 96,491,448 $ 91,331,295 $ 85,688,521 $ 80,443,987 $ 72,824,432 Total Actuarial Liability $ 79,332,984 $ 74,471,029 $ 69,889,905 $ 65,365,669 $ 58,884,036 Participant Statistics Retired Participants Number 100 92 85 75 70 Average Monthly Benefits $ 2,646 $ 2,246 $ 2,087 $ 2,122 $ 1,973 Vested Inactive Participants Number 172 163 156 156 155 Average Monthly Benefits $ 487 $ 504 $ 418 $ 409 $ 403 Active Participants Number of Participants 347 346 329 325 319 Average Age 47.5 47.1 47.4 47.4 46.7 Average Years of Service 11.8 11.7 12.8 12.9 12.7 Average Compensation $ 60,706 $ 58,946 $ 59,962 $ 57,485 $ 55,084 Total Compensation $ 21,065,079 $ 20,395,146 $ 19,727,497 $ 18,682,569 $ 17,571,907 Actuarial Assumptions Interest 8.00% 8.00% 8.00% 8.00% 8.00% Salary Growth 5.00% 5.00% 5.00% 5.00% 5.00% Taxable Wage Base Growth 4.00% 4.00% 4.00% 4.00% 4.00% Mortality Table Utilized 94GAM 94GAM 94GAM 94GAM 94GAM * Will be revised to $2,609,598 in the January 1, 2010 actuarial valuation. January 1, 2009 Actuarial Valuation City of Anytown Retirement Plan r This work product was prepared solely for the City. It may not be appropriate to use for other purposes. Milliman M i I I i ma n does not intend to benefit and assumes no duty or liability to other parties that receive this work. ' 28 1 1 1 1 1 1 1 1 1 1 1 1 THE CITY OF ANYTOWN RETIREMENT PLAN TABLE 15 TWENTY-YEAR PROJECTION OF BENEFIT PAYMENTS Plan Year Estimated Benefit Payments 2009 4,086,000 2010 3,370,000 2011 4,222,000 2012 4,685,000 2013 4,842,000 2014 4,972,000 2015 5,564,000 2016 6,244,000 2017 6,735,000 2018 7,425,000 2019 8,135, 000 2020 8,633,000 2021 9,491,000 2022 10,020,000 2023 11, 071, 000 2024 10, 844, 000 2025 11,367,000 2026 11, 824, 000 2027 12,428,000 2028 12,346,000 January 1, 2009 Actuarial Valuation City of Anytown Retirement Plan This work product was prepared solely for the City. It may not be appropriate to use for other purposes. Milliman M i I I i man does not intend to benefit and assumes no duty or liability to other parties that receive this work. 29 SECTION 1 EXECUTIVE SUMMARY Milliman, Inc. is pleased to present this proposal to provide actuarial services for the City of Fort Collins General Employees' Retirement Plan (GERP). This section sets forth our understanding of the project, our professional qualifications and our approach to actuarial services. REQUESTED SERVICES The General Employees' Retirement Committee (GERC) is searching for an actuarial firm to conduct annual actuarial valuations of the GERP and to provide actuarial consulting services. The scope of requested services is listed below: 1. Perform annual actuarial valuations of the GERP. The valuations will be completed by May 15 and will include a summary of the results, and information about all significant assumptions and findings noted. ' 2. Meet with the GERC prior to beginning the valuation to discuss the process, report format, scope, and significant assumptions. Discussions will include, but will not be limited to: 0 Evaluation of the actuarial cost method © Valuation procedures Economic assumptions C Demographic assumptions a Actuarial value of assets 3. Perform on -going pension consulting services for the GERC. 4. Provide required GASB disclosures and assist as needed in drafting the financial statement, notes and disclosures. 5. Provide annual Personal Retirement Planning Statements to plan members by March 31. ' 6. Attend monthly GERC meetings as requested. Membership data and assumptions will be discussed at the March meeting and the draft valuation will be presented at the April meeting. Additionally the actuarial valuation results and projections will be presented to the Council Finance Committee. Attendance at additional Council Finance Committee meetings 1 and City Council may also be required. 7. Update the City's benefit calculation software as needed. 8. Be completely familiar with the Plan documents, and if needed, assist the Committee in drafting changes to the Plan documents. 9. Prepare special reports, research, and analysis of topics as requested by the Committee, to ' be discussed at a regular monthly meeting. Reports will be made available to Committee members at least one week prior to meeting at which the report will be discussed. 10. Respond to Committee and support staff questions on a timely basis. ■ Proposal to Provide Actuarial and Consulting Services for M i I I i ma n the City of Fort Collins General Employees' Retirement Plan THE CITY OF ANYTOWN RETIREMENT PLAN APPENDIX A ACTUARIAL PROCEDURES AND ASSUMPTIONS ' The actuarial assumptions used in the valuation are intended to estimate future experience affecting projected benefit flow and investment earnings. Any variations in future experience from that expected from these assumptions will result in corresponding changes in the ' estimated costs of the plan's benefits. The tables in this section give rates of decrement, referred to in actuarial notation by the ' general symbol "q'." The underlying theory is described more fully in Jordan, Life Contingencies, Society of Actuaries (Second Edition, 1967), page 277. Any age referred to in a table is always the age of the person at his or her nearest birthday. ' Actuarial Cost Method (Adopted January 1, 2000) The actuarial cost method we use to calculate the funding requirements of the Plan is called the entry age normal actuarial cost method. Under this cost method, the actuarial present value of the projected benefits of each individual ' included in the valuation is allocated on a level basis over the earnings of the individual between entry age and assumed exit age. The portion of this actuarial present value allocated to a valuation year is called the Normal Cost. The portion of the actuarial present value not provided ' for at the valuation date by future Normal Costs is called the Actuarial Liability. Actuarial Value of Assets The actuarial value of assets is determined by calculating an expected value equal to the prior year's market value of assets, plus cash flows of contributions and benefit payments for the year and assuming an 8.5% interest return, 8.0% beginning in 2003. The difference between this expected value and the actual market value is recognized over 5 years. The actuarial value of assets is then the actual market value minus the gains and losses for the past years that ' have not yet been recognized. The resulting value is limited to between 80% and 120% of the market value of assets. Annual Required Contribution (Adopted January 1, 2007) The Annual Required Contribution (ARC) is calculated using a 20-year amortization of the unfunded actuarial liability or funding excess to determine the amortization component of the ARC. The amortization period will decrease to 15 years on January 1, 2011. On each valuation date, the newly determined unfunded actuarial liability or funding excess is amortized over a full amortization period as a level percentage of pay. ' January 1, 2009 Actuarial Valuation City of Anytown Retirement Plan ■ This work product was prepared solely for the City. It may not be appropriate to use for other purposes. Milliman M i I I i man does not intend to benefit and assumes no duty or liability to other parties that receive this work. ' 30 Investment Earnings (Adopted January 1, 2003) 8.00% per annum, compounded annually, net of expenses. Earnings Progression Annual salary increases are assumed to be 5%. Expenses ' None assumed. Cost of Living Adjustments None assumed. Interest Credit for Employee Contributions 5.00% credited as if all contributions are made mid -year. tRetirement (Adopted January 1, 2004) Members not eligible for Rule of 80 retirement are expected to retire at age 65. The following table sets forth the probability of retirement from active employment for those members eligible for Rule of 80 retirement. Age Probability of Retirement 55 — 61 15% 62 35 63-64 15 65 65 66 — 69 25 70 & Over 100 Vested inactive members who are eligible for a Rule of 80 pension at termination are assumed to retire at age 55. All other vested inactive members are assumed to retire at age 65. Surviving spouses of vested inactive members are assumed to begin benefit payments at first eligibility. Disablement ' None assumed. Retiree Mortality (Adopted January 1, 2005) ' The 1994 Group Annuity Mortality Tables for males and females are used. ' January 1, 2009 Actuarial Valuation City of Anytown Retirement Plan This work product was prepared solely for the City. It may not be appropriate to use for other purposes. Milliman f M i I I i ma n does not intend to benefit and assumes no duty or liability to other parties that receive this work. 31 I Withdrawal Rates Graduated rates are used. Sample rates are as follows: Age at Termination Probability of Termination 25 12.3% 35 7.3 45 3.8 55 1.2 60 0.3 ' Marriage Rates All active and deferred vested participants not currently receiving benefits are assumed to be married. Male spouses are assumed to be three years older than their female spouses. ' Lump Sum Payments (Lump Sum Interest Rate Adopted January 1, 2006) 50% of members retiring from active status age 65 and over or under age 65 who are not eligible for a Rule of 80 benefit are expected to elect lump sum payments. These lump sum amounts are calculated using the 1983 Group Annuity Mortality blended 50%/50% for males and females and 6.0% interest. Amortization Period for ARC (Adopted January 1, 2007) ' The UAL is amortized as a level percentage of pay over a 20-year period. On January 1, 2011 the amortization period will decrease to 15 years. ' Changes in Actuarial Assumptions and Methods as of January 1, 2009 None. ' January 1, 2009 Actuarial Valuation City of Anytown Retirement Plan This work product was prepared solely for the City. It may not be appropriate to use for other purposes. Milliman M i I I i man does not intend to benefit and assumes no duty or liability to other parties that receive this work. ' 32 THE CITY OF ANYTOWN ' RETIREMENT PLAN APPENDIX B PLAN SUMMARY ' All actuarial calculations are based upon our understanding of the provisions of The City of Anytown Retirement Plan, as amended through December 31, 2008. The plan provisions reflected in this valuation include the changes in the restated plan effective January 1, 2008. ' This summary does not attempt to cover all of the detailed provisions. Plan Sponsor The City is both the Plan Sponsor and the Plan Administrator. ' Plan Year The Plan Year is the 12-month period beginning January 1 and ending December 31. Accrued Benefit The Accrued Benefit for each Participant is the Participants Normal Retirement Benefit ' calculated using Average Monthly Salary and Years of Service as of the calculation date. ' Average Monthly Salary (modified effective July 1, 2000) A Participant's Average Monthly Salary, as of a given date, is the average of the highest 3 consecutive calendar years of salary during the last 10 years of employment. ' Salary ' Salary is the gross compensation included as taxable income on Form W-2. Effective Date The original effective date of the plan is December 29, 1967. The plan was most recently restated effective July 8, 2007. ' Eligible Employee Classification All persons employed by the Employer. ' Participation Date ' Prior to July 8, 2007, an employee becomes a participant on the first of the month after completion of 12 months of employment and 1000 hours of service. Effective July 8, 2007, all employees become a participant on the first of the month following their hire date with the City. ' January 1, 2009 Actuarial Valuation City of Anytown Retirement Plan This work product was prepared solely for the City. It may not be appropriate to use for other purposes. Milliman M i I I iman does not intend to benefit and assumes no duty or liability to other parties that receive this work. 33 7 I Employee Contributions Effective July 8, 2007, all employees are required to make contributions to the plan based on the following schedule: 7/8/2007 — 12/31 /2007 1 /1 /2008 — 12/31 /2009 1 /1 /2010 —12/31 /2011 1 /1 /2012 —12/31 /2013 1/1/2014 & Later 1 % of Base Salary 2% of Base Salary 3% of Base Salary 4% of Base Salary 5% of Base Salary Employee contributions will be made by the City per payroll period. Normal Retirement Date A Participant's Normal Retirement Age is the first of the month coincident with or next following the attainment of age 65. Vested Accrued Benefit A Participant's Vested Accrued Benefit as of a given date is equal to the product of his Accrued Benefit multiplied by his Vested Percentage as of that same date. Vesting Schedule Prior to January 1, 1991, participants were 50% vested in their Accrued Benefit after completion of 5 Years of Service, increasing 10% per year to 100% vesting after 10 Years of Service. After January 1, 1991, participants are 100% vested after completion of 5 Years of Service. Participants are always 100% vested in their employee contributions. Year of Service For Benefit Purposes (Credited Service) Prior to December 1, 1978, Credited Service was a Participant's period of continuous employment after completion of 3 months of service. However, no Credited Service was given for any period for which an employee did not make contributions. After December 1, 1978, full years of Credited Service are granted for each Plan Year in which a Participant works at least 1,800 hours. Proportional credit is given for each year in which he works more than 1,000 hours, and in the year of hire and termination regardless of how many hours worked. The proportional credit is equal to the actual hours worked divided by 1,800. For Vesting Purposes A year of Vesting Service is granted for each Plan Year in which an employee completes 1,000 or more hours of service. January 1, 2009 Actuarial Valuation City of Anytown Retirement Plan ■ This work product was prepared solely for the City. It may not be appropriate to use for other purposes. Milliman M i I I i man does not intend to benefit and assumes no duty or liability to other parties that receive this work. ' 34 I Normal Retirement Benefits (modified effective July 1, 2000) ' Each Participant who becomes eligible for a Normal Retirement Benefit under the plan will be entitled to receive a monthly retirement pension benefit beginning at the Participant's Normal Retirement Date and payable in the Normal Benefit Form equal to: 2.0% of his Average Monthly Compensation multiplied by his Years of Credited Service. Normal Benefit Form Life Annuity with 120 Payments Guaranteed - Monthly pension benefit payable for the lifetime of the Participant, and if the Participant dies prior to receiving 120 monthly payments, the remainder of the 120 payments will be paid to the Participant's beneficiary. Early Retirement ' (a) Early Retirement Date A Participant's Early Retirement Date is the first day of the month so elected by the Participant which coincides with or next follows the date upon which the Participant attains age 55 and completes 10 Years of Service. (b) Early Retirement Benefit A Participant's Early Retirement Benefit is a monthly pension benefit equal to his Accrued Benefit determined as of his Early Retirement Date, reduced by 1/15th for the first 5 years and 1/30th for each of the next 5 years payments commence prior to age 65. Rule of 80 Retirement (modified effective January 1, 2000) ' Participants are eligible for Rule of 80 Retirement benefits when their age plus Years of Service total at least 80 at the time of termination. A participant's Rule of 80 Benefit is a monthly pension benefit equal to his Accrued Benefit determined as of termination and is payable upon attainment of age 55. Participants electing Rule of 80 retirement are not eligible to receive lump sum benefits. Optional Benefit Forms Optional Benefit Forms are available and equal to the Actuarial Equivalent of the Normal Benefit Form and may be in an amount more than or less than that provided by the Normal Benefit Form depending on the option selected. Such distribution may be as a Joint & 50%, 75%, or 100% Survivor Annuity, a Joint & 50%, 75% or 100% Survivor Annuity with Restoration Feature, a Life Annuity, or a Lump Sum. IPre -Retirement Death Benefit If a Participant dies, the Participant's beneficiary will receive a monthly benefit payable as a 10 Year Certain and Life Annuity in an amount equal to the actuarial equivalent of 90% of the Participant's vested Accrued Benefit. January 1, 2009 Actuarial Valuation City of Anytown Retirement Plan ■ This work product was prepared solely for the City. It may not be appropriate to use for other purposes. Milliman M i I I i man does not intend to benefit and assumes no duty or liability to other parties that receive this work. 1 35 Termination Benefit In the event of the termination of a Participant's employment for any reason other than death, disability or retirement after completing 5 Years of Service, the Participant will become entitled to receive a monthly pension benefit commencing on his Normal Retirement Date equal to his Vested Accrued Benefit. Deferred vested participants with 10 or more Years of Service are eligible to retire as early as age 55 with reduced benefits. A deferred vested participant whose age plus Years of Service at the time of termination total at least 80 are eligible for unreduced benefits beginning at age 55. Plan Changes None. January 1, 2009 Actuarial Valuation City of Anytown Retirement Plan This work product was prepared solely for the City. It may not be appropriate to use for other purposes. Milliman M i I I i man does not intend to benefit and assumes no duty or liability to other parties that receive this work. 36 it [i 11 APPENDIX C MILLIMAN PUBLIC SECTOR CLIENT LISTING ■ Proposal to Provide Actuarial and Consulting Services for M i I I i ma n the City of Fort Collins General Employees' Retirement Plan Ll n 11 1 Milliman's Public Sector Clients State Systems SYSTEM OFFICE DATE RETAINED Alaska, University of Philadelphia 2003 California State Teachers' Retirement System Portland, OR 1985 Guam, Government of San Francisco 2004 Idaho Department Of Labor Seattle 1979 Idaho Public Employees Retirement System Seattle 1965 Iowa Public Employees Retirement System Omaha 1954 Kansas Public Employees Retirement System Omaha 1994 Maryland Transit Administration Washington D.C. Missouri Prosecuting Attorneys and Circuit Attorneys Retirement Fund St. Louis 1994 Nevada Prepaid Tuition Philadelphia 1998 New Jersey Teachers' Pension and Annuity Fund Philadelphia 1996 New York State Energy Research Development Authority Albany 2006 Ohio Prepaid Tuition Philadelphia 1991 Oklahoma Public Employees Retirement System Omaha Puerto Rico Employees Philadelphia 2008 Puerto Rico Teachers Philadelphia 2008 Virginia College Savings Plan Philadelphia 2001 Washington Prepaid Tuition Philadelphia Westport, Connecticut 3 plans),Town of New Jersey 1982 Wisconsin - Edvest Philadelphia 1997 Proposal to Provide Actuarial and Consulting Services for i ill i I I i m a n the City of Fort Collins General Employees' Retirement Plan 1 1 Milliman's Public Sector Clients Municipal and County Systems SYSTEM OFFICE DATE RETAINED Avon, Connecticut Hartford 1995 Benton Harbor, City Of Chicago 2000 Caribou, Maine Hartford 1993 Central Lincoln People's Utility District, Newport Oregon Portland, OR 1980 Central Nebraska Public Power District Omaha 1967 City Of Long Beach Washington D.C. City of Newberg, Ore on Portland, OR 1990 Coos Bay North Bend Water Board Portland, OR 1983 Decatur,City Of Chicago 2000 Delaware Municipalities Washington D.C. 1995 Durham, Connecticut Hartford 1995 El Cerrito, California San Francisco 1995 Florence, Alabama Atlanta 1985 Fort Collins,City Of Denver 2000 Golden Gate Transit - Amalgamated Retirement Plan Portland, OR 1970 Guilford, Connecticut, School New York 1999 Guilford, Connecticut, Town of New York 1999 Harriman Utility Board Atlanta 1996 Howard County Washington D.C. 2004 Kern County Employees' Retirement Association Seattle 2004 Lane Transit District -Hourly Portland, OR 1977 Lane Transit District -Salaried Portland, OR 1977 Las Vegas Valley Water District San Francisco 1983 Laurel, City of Washin ton D.C. 1997 Liberty County Atlanta 1997 Long Island Rail Road Philadelphia 2001 Los Angeles County Employee Retirement Association Seattle 1998 Madison, Connecticut Hartford 1987 Manchester, Connecticut Hartford 1994 Manhattan & Bronx Surface Transit Operating Authority Philadelphia 2001 Metro Wastewater Reclamation District Denver 1998 Metro -North Commuter Railroad Philadelphia 2001 Metropolitan Transportation Authority Philadelphia 2001 Metropolitan Utilities District Omaha Proposal to Provide Actuarial and Consulting Services for E7 M i I I i ma n the City of Fort Collins General Employees' Retirement Plan OBSERVATIONS ' Closed Plan The GERP is a closed plan with continued benefit accruals for the active population. Closed plans face a decline in the number of active participants with decreasing future working lifetimes. This unique demographic leads to a number of issues such as: o Declining Contributions - Contributions are based on a percentage of pay. As the active population terminates employment, the contributions will decline. Contributions should cover annual benefit accruals as well as payoff a portion of the unfunded accrued actuarial liability. The current contributions are not large enough to pay for both accruals and the unfunded accrued actuarial liability. The City must explore new ways to increase contributions and remain within the current budget in order to pay for the unfunded liability. 1 1 1 Investment Policy - As the population moves from active to retired, the GERC may wish to consider modifying the investment policy. Thought should be given to risk tolerance when future contributions are declining and investment losses are more difficult to recoup. In addition, cash flow matching may be a helpful tool when most participants are receiving monthly benefits. Funding and Budget Constraints Like many other retirement plans, the GERP experienced poor investment returns during 2008. The impact of the asset losses and underfunding of the accruals have created a substantial unfunded accrued actuarial liability. Milliman will continue to assist the City in evaluating methods to manage the annual cost using asset liability models and revisions in the contribution pattern. Plan provisions should also be reviewed to see whether reasonable changes could improve the funding level of the plan. Assumptions Assumptions are classified into two categories, demographic and economic. Assumptions are not based solely on past experience, but rather a blend of past experience and expectations of future events. It is important to review assumptions annually to make sure that the assumptions and actual experience are reasonably related. Periodic comprehensive experience studies may be warranted if expectations are not being met. ■ Proposal to Provide Actuarial and Consulting Services for M i I I i ma n the City of Fort Collins General Employees' Retirement Plan SYSTEM OFFICE DATE RETAINED Morrow County, Oregon Portland, OR 1983 New Britain, Connecticut Hartford 1999 Newington Volunteer Fire Department, Connecticut Hartford 1991 Newington, Connecticut Hartford 1989 North Bend, Oregon Portland, OR 1983 North Branford, Connecticut Hartford 2001 Northern Colorado Water Conservancy District Denver 1997 Northern Virginia Regional Park Authority Washington D.C. 1985 Oakland Municipal Employees' Retirement System Portland, OR Ocean City Washington D.C. 2003 Oklahoma County Retirement System Denver 2005 Old Saybrook, Connecticut Hartford 1995 Omaha Metro Area Transit, Nebraska Omaha 1991 Omaha School Employees Retirement System Omaha 1954 Ornaha,City of Omaha PARS - City of Cerritos Washington D.C. 2004 PARS - Ventura Regional Sanitation District Washington D.C. Philadelphia Housing Authority Philadelphia 1996 Plymouth, Connecticut Hartford 2002 Pooler, City of Atlanta Port of Houston Authority Houston 1974 Portsmouth, Rhode Island Hartford 1993 Prince William County Supplemental Uniformed Plan Washington D.C. Ralston, City of Omaha 1995 Reading, City of Philadelphia 1998 Regional School District #13, Durham Connecticut Hartford 1995 Rivendale Park, Town of Washington D.C. 2000 Salem Area Mass Transit District Portland, OR Simsbury, Connecticut Hartford 1995 Smyrna, Town of Philadelphia 1998 Southeast Weld Fire Protection District Denver 1998 South Fire District, Middletown, Connecticut Hartford 1989 Southbury, Connecticut Hartford 2002 Springfield Retirement Plan, City Of Portland, OR 1987 Springfield School District #19 Portland, OR 1995 Staten Island Rapid Transit Operating Authority Philadelphia 2001 Proposal to Provide Actuarial and Consulting Services for Mi I I i m a n the City of Fort Collins General Employees' Retirement Plan d 11 1 -SYSTEM, OFFICE DATE RETAINED S I RTOA Sta ton Portland, OR 1983 Stratford, Connecticut Hartford 2001 Tacoma Employees' Retirement System Seattle 1975 Texas County and District Retirement System Seattle 1999 Tillamook County Portland, OR 1997 Town of Thurmont, Maryland Philadelphia 1993 TriMet Portland, OR 2005 Tulsa County Employees' Retirement System St. Louis 2003 Tuscaloosa - Hourly, City Of Atlanta 1994 West Hartford, Connecticut Hartford 1992 West Warwick, Rhode Island Hartford 2000 Wethersfield, Connecticut (Hartford 1992 Wichita Employees' Retirement System Omaha 2000 Windham Board Of Education, Connecticut Hartford 1997 Windsor, Connecticut Hartford 1993 Woodburn Portland, OR 1983 IProposal to Provide Actuarial and Consulting Services for E7 M i I I i m a n the City of Fort Collins General Employees' Retirement Plan Milliman's Public Sector Clients Uniformed Systems SYSTEM OFFICE DATE RETAINED Anchorage Police and Fire Retirement Plan, Alaska Portland, OR 1991 Antonia Fire Protection District Pension Plan and Trust St. Louis 2005 Arvada Fire Protection District Volunteer Pension Fund Denver 1994 Belgrade Fire Department Relief Association Denver 2008 Bloomington Fire Relief Association Minneapolis 1997 Boulder, Old Hire Police Plan Denver 2006 Boulder, Old Hire Fire Plan Denver 2006 Bristol Police Plan, Rhode Island Hartford 1997 City of Springfield Police Officers and Fire Fighters' Retirement Fund St. Louis 1999 Cumberland Police, Rhode Island Hartford 2002 Delaware Municipal Police and Fire Washington D.C. 1995 Delaware State Police Washington D.C. 1995 Delaware Volunteer Fire Washington D.C. 1995 Elsmere Police Pension Plan Philadelphia 1999 Estes Park Volunteer Firefighters Pension, Town of Denver 2003 Firemen's Pension Fund Seattle Guilford, Connecticut, Police New York 1999 Hastings, Nebraska, Fire and Police Omaha 1982 Iowa Peace Officers' Retirement, Accident and Disability System Omaha 1997 Madison Police, Connecticut Hartford 1995 Normal, Illinois, Fire and Police Chicago 1990 North Platte Police Officers Omaha Oakland Police and Fire System Portland, OR 1974 Omaha Firefighters' Benefit Association Omaha 1987 Omaha Police and Fire Omaha 1995 Overland Park, Kansas Fire Department Retirement Plan St. Louis 2005 Overland Park, Kansas Police Department Retirement Plan St. Louis 2005 Police and Firemen's Retirement Fund of the City of Berkeley, Missouri St. Louis 2001 Retirement Plan for Employees of Mehville Fire St. Louis 1995 Proposal to Provide Actuarial and Consulting Services for M i I I i ma n the City of Fort Collins General Employees' Retirement Plan i 1 d 1� SYSTEM OFFICE DATE RETAINED Protection District Shin lehouse Borough, Pennsylvania Philadelphia 1988 Southeast Weld Fire Protection District Firemen's Pension Fund Denver 1998 Three Forks Fire Relief Association Denver 2008 U. S. Coast Guard Health Washington D.C. 1994 U. S. Coast Guard Pension Washington D.C. 1994 U. S. Department Of Defense Washington D.C. 1994 West Haven Police Dept., Connecticut Hartford 1993 Wichita Police & Fire Retirement System Omaha 2000 Proposal to Provide Actuarial and Consulting Services for Mi I I i ma n the City of Fort Collins General Employees' Retirement Plan ri I 1 I r 1 Milliman's Public Sector Clients Audits SYSTEM OFFICE DATE Alaska State Public Employee and Teacher Systems Portland, OR 2002 Aurora, CO General Employee's Retirement System Denver, CO 2009 Austin Employees' Retirement System, City of Milwaukee 2002 Dearborn, City of MI Milwaukee 1993 Employees' Retirement Fund of the City of Dallas Seattle 1996 Denver Water Board Denver 2009 Franklin, City of Milwaukee 2002 Illinois Municipal Retirement Fund Milwaukee 2000 Missouri Local Government Employees Retirement System Milwaukee 1998 Missouri State Employees Retirement System Milwaukee 1997 New Mexico Retiree Health Care Authority Denver / Boise 2007 City of Phoenix Employee's Retirement System Denver / Omaha 2008 Teacher Retirement System of Texas Omaha 1998 State Pension & Funding Council Seattle 2002 -Washington Wisconsin Legislative Audit Bureau Milwaukee 2001 -Wyoming Retirement System Denver 2004, 2009 I Proposal to Provide Actuarial and Consulting Services for mp� i n 1 m a n the City of Fort Collins General Employees' Retirement Plan Milliman's Public Sector Clients Other Projects SYSTEM OFFICE DATERETAINED Allin town Fire District* Hartford 2004 Bend -La Pine School District Portland, OR Berkeley, City of San Francisco 1994 California Public Employees Retirement System CalPERS Portland, OR 1997 Central Unified School District San Francisco 2004 Charlestown, RI Hartford 2005 City of Albany Albany 2005 City of Amarillo Houston 2005 City of Buffal6 Albany 2005 City of Falls Church Washington D.C. 2005 City of Norwalk** Hartford 2004 County of Lassen San Francisco 1999 Denver Public Schools Denver Eugene Public Schools Portland, OR Fairfax County Water Authority Washington D.C. 1981 Florida Retirement System Washington D.C. 1986 Forest Grove, Oregon Portland, OR 1989 Fresno County Office of Education San Francisco 2002 Houghton County Medical Care Facility Chicago 1994 Indianapolis Public Transportation Corp. Indianapolis 2005 Iowa Judicial Retirement Fund, State of Omaha Ivy Tech State College Indianapolis 2005 Kings Canyon Unified School District San Francisco 2004 Maryland Municipal Special Work Washington D.C. Maryland State Legislature Washington D.C. Milwaukee, City of Milwaukee New Hampshire Judges Hartford 2002 New York State Local Employees' Retirement System Albany 1986 Newberg School District SERP Portland, OR North Broward Hospital District, Florida Atlanta 1983 NYC -Office Of Management & Budget Philadelphia Ohio Retirement Study Commission Philadelphia 1990 IProposal to Provide Actuarial and Consulting Services for lil"Milliman the City of Fort Collins General Employees' Retirement Plan 1 j F SYSTEM OFFICE DATE RETAINED Omaha Airport Authority, Nebraska Omaha 1985 Pennsylvania Public Employees Retirement Commission Philadelphia 1986 Port of Oakland San Francisco 1995 Public Agency Retirement Services Washington D.C. Public Agency Retirement System Washington D.C. Roseville, City of San Francisco 2002 Salem -Kaiser School District Portland, OR San Jose California,City Of Portland, OR San Mateo Community College District San Francisco 1996 San Ramon, City of San Francisco 2005 Santa Clara Valley Transit Authority San Francisco 1998 SEPTA Philadelphia 1996 St Helens Rural Fire District Portland, OR State Center Community College District, Fresno San Francisco 1997 State of Idaho Seattle State of Louisiana Atlanta Stephens County Hospital, Georgia Atlanta 1983 Stockton, City of San Francisco 2001 Texas Pension Review Board and Legislative Budget Board Houston 1979 The Board of Public Utilities Retirement Pension Plan Omaha 2005 The ConwayCorporation Omaha Town of Chester Hartford 2005 Town of Chester Hartford 2005 Town of Colonie Albany 2005 Town of Granby Hartford 2004 Town of Orange Hartford 2004 Town of Orange Police Hartford 2004 Travis County Atlanta Truckee Meadows Fire Protection District San Francisco 2001 Washoe County, Reno San Francisco 1996 West Haven First Fire District Hartford 2004 West Shore Fire District *** I Hartford 2004 Proposal to Provide Actuarial and Consulting Services for M i I I i m a n the City of Fort Collins General Employees' Retirement Plan APPENDIX D SAMPLE RESEARCH PUBLICATIONS This Appendix shows examples of publications produced by Milliman's Research Department and our Public Sector Specialty Group. O PERiScope contains information specifically geared to public sector issues. o Perspectives provides a communication of general interest on employee benefits and occasional Client Action Bulletins containing information on regulatory and legislative actions ' recommending appropriate actions are published periodically throughout the year. More publications and case studies can be found on our website at www.milliman.com. 1 Proposal to Provide Actuarial and Consulting Services for M i I I i ma n the City of Fort Collins General Employees' Retirement Plan 1 �i January 2009 K! Milliman D, D DPublic`Employee `Retirement Systems., , Public plan DB/DC choices Mark 011eman, FSA, MAAA This article is about choices: When given the choice, do public employ- ees choose a defined benefit (DB) plan or a defined contribution (DC) plan? Do employers give employees the chance to choose a second time? What happens when employees choose their own investments? Can employers choose to offer meaningful death and disability benefits to DC members? What are the implications of an employer choosing to change from a DB to a DC plan? This article looks at the recent experi- ence of statewide retirement systems to provide some answers. What do public employees choose? Many people claim that DC plans are more attractive to new employees than DB plans. Is this true? As a test, note that in the last 10 years, the seven statewide systems listed in Table 1 have begun giving new hires the choice between participating in a DB or a DC plan. Their experience indicates that public employees prefer DB plans. The percentage of new employees electing DC plans ranges from 3% in the Ohio Public Employee Retirement System to 26% in Florida. NEW HIRE ELECTIONS IN MOST RECENT COMPLETE YEAR Table 1 shows that many of the members going into a DB plan never submit an election and are placed in the DB plan by default. However, based on survey data, Florida found that "up to 45% of the defaulters may be using this option as their active election in the belief that by defaulting there could be no mistakes made in their plan choice." What is more, Table 2 shows that in Washington PERS—the only system where DB is not the default-63% of new members have actively chosen an all-DB plan (Plan 2) over the default of a combined DB and DC plan (Plan 3). Most of these DB/DC choice plans have had relatively stable election percentages in the short time they have existed. However, we do not know how the choices members make will change in the future. The stock market decline of 2000 to 2002 has certainly influenced many members. No doubt factors such as the future of the stock market and the experiences of people retiring with only DC plans will influence future member choices. The financial market experience of late 2008 may have some influence as well. SYSTEM DB BY DEFAULT DB ACTIVE DC ACTIVE COMBINED PLAN ENROLLMENTS ENROLLMENTS ACTIVE ENROLLMENTS COLORADO 39^/0 "` 43^/d 18% NOT OFFERED" FLORIDA 55% 19% 26% NOT OFFERED' MONTANA PERS 90% NOT SEPARATED" 10% NOT OFFERED` NORTH DAKOTA** 88% NOT SEPARATED` 12% NOT OFFERED` OH10"PERS 82% 13% 3% 2% OHIO TEACHERS 72% 14% 11% 4% SOUTH CAROLINA 80% NOT SEPARATED` 20% NOT OFFERED' " "NOT SEPARATED" MEANS ACTIVE DB ENROLLMENTS HAVE NOT BEEN SEPARATED FROM DEFAULT DB ENROLLMENTS. "NOT OFFERED" MEANS THERE IS NOT AN OPTION TO ENROLL IN A COMBINED DB/DC PLAN. "NORTH DAKOTA STATISTICS ARE FOR JANUARY 2001 THROUGH JUNE 2008. A member of Abelica Global milliman.com li I fl CUMULATIVE WASHINGTON PERS NEW HIRE ELECTIONS FROM MARCH 2O02 TO SEPTEMBER 2008 PLAN 3 COMBINED PLAN 3 COMBINED PLAN 2 ALL DB DB & DC DB & DC ACTIVE ACTIVE BY DEFAULT ENROLLMENTS ENROLLMENTS 19% 18% 63% Tables 1 and 2 summarize the experience of systems that allow their members to choose between a DB plan and a DC plan. Ohio and Washington state members also have the choice of a "combined" plan, where employer contributions fund a DB plan and employee contributions fund a DC plan. Washington state members do not have the option of an all -DC plan. What about do-overs? One plan design choice employers face is whether to give employees a chance to change their mind. This chance for a do -over has been referred to by some as the pension mulligan. Although Montana PERS, North Dakota, Vermont, and Washington state require new hires to make a one-time irrevocable decision, other systems do not. Colorado allows members to change their election one time in years two through five after hire. Ohio Teachers and South Carolina also allow members to change their election once in the first five years, but only from DC to DB. Florida allows members to change once at any time before retirement or termination of employment. Last, Ohio PERS allows members to change up to three times: once in their first five years of employment, once in their second five years, and once more at any time after 10 years of service through retirement. You might ask, "What do systems do when members change their mind?" Florida allows two choices when members switch from the DB to the DC plan. The members can either (1) freeze their current DB benefits based on service and salary to date and have future contributions accumulate in their DC accounts, or (2) convert their DB benefits into DC accounts based on the value of the normal retirement benefit. If a Florida member wants to switch from DC to DB, the member must pay the full cost based on either the present value or the actuarial accrued value, depending upon where the member has previous DB service prior to joining the DC plan. The DC account is used first. If there is more money than needed in the DC account, the member keeps the extra in the DC account. If there is not enough money in the DC account, then the member must pay the difference or stay in the DC plan. Ohio PERS, which allows up to three changes, takes a somewhat different approach. Changes are prospective only, but members transferring to the DB or combined plan have the option to purchase service in the new plan using their DC accounts. Frozen DB benefits are based on salary and service during DB membership only. The do -over could be particularly valuable when a member's situa- tion changes. As an example, the portability of a DC plan might be attractive to a teacher who does not expect to stay long in a position due to a military spouse who is frequently moved around the country. However, if the couple's plans change and they decide to settle down, the teacher might want to change to the DB plan. Can meaningful death and disability benefits be provided in a DC environment? Yes, meaningful death and disability benefits can be provided in a DC environment, but it will require supplemental contributions. Consider the choices three states havecmade to respond to the criticism that DC accounts do not provide adequate death and disability benefits. In Florida, where members choose between a DB and a DC plan, disabled members can choose to surrender their DC account balance and receive the same disability benefits as provided by the DB plan. This raises a question: Where does the money to finance this benefit come from? The answer is that the employer pays a separate charge ranging from 0.25% of pay for general members to 1.33% of pay for special risk members, and a side account is maintained to finance the difference between the cost of the disability benefits and the dollar amount of the DC accounts surrendered by the members. If DC members die in Florida, their death benefit is the DC account balance. Montana PERS has a similar provision where 0.30% of DC member pay is set aside to finance long-term disability benefits. Alaska has a different approach. Alaska public employees hired after July 1, 2006, all go into a DC plan. Here the occupational death and disability benefit is 40% of salary until normal retire- ment (50% of salary for the occupational death of police and fire members). The employer continues both the employer and employee contributions into a special occupational death and dis- ability trust account until the member reaches normal retirement, or until the date the member would have reached normal retire- ment in the case of occupational deaths. At normal retirement age, the 40% (or 50%) of salary benefit stops, and the member, or survivor, receives the DC account as well as the accumulated contributions from the occupational death and disability trust account with actual returns net of expenses. Employers make contributions into a separate fund to finance the extra benefit not provided by the DC account. 2 :: JANUARY 2009 I I r 11 SUMMARY The City of Fort Collins currently faces a wide variety of retirement plan issues, only a few of which are mentioned above. Valuation methods, funding methods, assumptions and benefit changes must be analyzed carefully in order for the City to make sure that benefits promised to participants are properly funded. Milliman has worked with public plans since 1947 and has assisted our clients with a wide variety of situations including the issues the City faces. The proposed project team of Pat Kahle, Joel Stewart and Francine Moyer is committed to maintaining Milliman's public retirement system practice through dedication to the GERC and dedication to Milliman's high standards of professional service. We are pleased to present our qualifications to the GERC in this proposal. ■ Proposal to Provide Actuarial and Consulting Services for 3 M i I I i ma n the City of Fort Collins General Employees' Retirement Plan 1 11 1 What happens when employees choose their own investments? Experience indicates the average employee directing his or her own investments earns lower investment returns than a statewide DB system. Here is the experience of two states. Nebraska's state and county employees hired between 1964 and 2003 had only a DC plan. During the same period, Nebraska main- tained separate DB plans for its school employees, state judges, and state patrol. Over the 20 years leading up to 2002, the average return in the DB plans was 11 % and the average return in the DC plans was between 6% and 7%. Why the significant difference? One reason '�i�E�I�COP same manner as the DB plan. However, unlike Washington's TAP, which is one of many investment choices, in Oregon's IAP there are no other investment choices, and so all DC money is invested to match the DB plan. Both Washington and Oregon provide members with a profes- sionally managed portfolio. Washington's approach leaves room for individual risk tolerance. For instance, members near retirement may not want to take as much risk. Oregon's approach ensures that all member funds are invested in a carefully managed portfolio. Either way, it is ironic that DC members may need to give up their ability to choose their own investments in order to earn returns is that nearly 50% of DC member contributions were invested in the competitive with DB plans. stable value fund. The stable value fund was the default for members not making a specific investment election. Although the stable value fund is very conservative and the investor's balance will not decrease, the investor also has a lower expected rate of return. Partially due to the lower returns, employees were receiving a replacement ratio of their pre -retirement income closer to 30% rather than the projected 50% to 60%. Nebraska has since decided that employees hired on or after Jan. 1, 2003, will go into a hybrid defined benefit plan. West Virginia had a similar experience. Teachers hired between 1991 and 2005 had only a DC plan. Teachers hired after July 1, 2005, go into a DB plan instead. One of the reasons for this change is that average DC returns lagged DB returns. As an example, dur- ing the seven years from 2001 to 2007, the DB plan outperformed the DC plan in both the best and worst markets. The DC return was higher in only one of the seven years Over the seven year period, the average DB return was 3.15% higher. Specific returns are shown in the appendix. Do DC members have to choose their own investments? Employees directing their own investments tend to earn lower invest- ment returns than statewide DB systems for a variety of reasons. DC members are part-time investors, whereas DB plans are managed by full-time highly trained professionals. DB plans have investment options that are generally not available to DC members, such as real estate, private equity, and hedge funds. DC members often lack dis- cipline and chase returns. Does this mean that DC members cannot earn the same investment returns as DB plans? No, DC members can earn exactly the same returns. Members of Washington state Plan 3 have the option to invest in the Total Allocation Portfolio (TAP), which mirrors the investments in the state DB plan and therefore earns the same returns. Washington has made the TAP the default investment option for Plan 3, and approximately 61 % of the members' DC assets are in the TAP option. The employee contributions of members in the Oregon Public Service Retirement Plan go into the Individual Account Program (IAP). Like Washington's TAP, Oregon's IAP money is invested in the Both the Washington and Oregon plans are hybrid plans where employer contributions fund a DB plan and employee contributions go into a DC plan. This is significant because the DB plan will provide some level of guaranteed income regardless of DC investment returns. Does changing to DC solve funding problems? In 1991, the West Virginia teachers' poorly funded DB plan was closed to new members. All new hires were put into a DC plan. This funding solution overlooked some important considerations: New members do not start with any unfunded obligation. Projected contributions for new members were worth more than the projected DB costs for those members. No unfunded obligations for existing members are reduced when new members go into a DC plan. As a result, the loss of new members made it more difficult to finance the unfunded obligations of the West Virginia Teachers' Retirement System (TRS). In 2003, West Virginia studied whether teacher retirement should be returned to a DB plan. Another factor in the decision was that 4,500 members who transferred from the DB to the DC plan in 1991 found it hard to retire after the bear market of 2000-2002. When also considering the lower average returns that were earned on the DC member accounts, the state decided that starting in 2005 all new hires would go into the DB plan to save money. After studying the issue, the state decided that funding a DB plan properly would be less expensive than a DC plan providing equivalent benefits. The state has shown discipline to achieve this proper funding, with extra contributions of $290.1 million in fiscal year 2006 and $313.8 million in fiscal year 2007 In addition, West Virginia completed a tobacco bond securitization in fiscal year 2007 and deposited $807.5 million of those proceeds into TRS as another special appropriation. Most recently, in June of 2008, the teachers in the DC plan were given the choice to switch to the DB plan. Seventy- eight percent chose to switch. JANUARY 2009 :: 3 I u I I� a West Virginia projected a $1.2 billion savings in the first 30 years due to moving new entrants from the DC to the DB plan. This relies on an assumed return of 7.5%. The Legislature asked what return would be needed to break even. The answer was 6.0%. In order for the DB plan to save money, a projected return of more than 6.0% was needed. The employer cannot avoid funding risk with a DB plan, but changing to a DC plan does nothing to take care of unfunded obligations. Some states require specific contributions to the DB plan as a percent of DC member pay in order to finance preexisting unfunded liabilities and to defray expenses. The systems include Colorado, Montana PERS, Ohio PERS, Ohio Teachers, and South Carolina. Details are in the appendix. What are the implications of these choices? The choices discussed in this article have many implications. Public employees have overwhelmingly chosen DB plans over DC plans. This implies that DB plans are more attractive than DC plans to public employees. This is not surprising, as public employees tend to have long service. Some systems have chosen to allow their members a second choice. This do -over could help an employee reverse a bad decision. Some systems have chosen to provide meaningful death and disability benefits in a DC environment; however, supplemental contributions are required. Employees tend to earn less when they choose their own investments. However, this can be countered in a DC plan by using an alternative like Washington state's TAP or Oregon's IAP, where the DC assets are invested in the same manner as the DB assets. Choosing to change from a DB to a DC plan does not solve funding problems. In the final analysis, it's a question of accumulation and distribution. The accumulation of contributions and investment earnings determines available retirement income. A plan that maximizes investment earnings maximizes the benefits provided by contributions. Public employees are choosing plans that provide lifetime distributions. There is not yet much experience on how many DC members have been able to make their assets last a lifetime. The distribution phase and the loss of longevity risk pooling in retirement is probably the hardest obstacle for DC plans to overcome. The consequences of outliving one's assets are severe. DC plans rarely measure whether assets accumulated will provide adequate retirement income. How many employees can be sufficiently educated and empowered to navigate the risks of pre -retirement accumulation and postretirement distribution? There often seems to be a choice between the employer bearing all the risk of funding a defined benefit and the member bearing all the risk of accumulating sufficient assets to last a lifetime. However, there are some choices that share risk between employers and employees, such as the combined DB/DC plans in Washington, Oregon, and Ohio, and DB plans where contribution increases are shared by employees. More choices are needed where risk is shared, or better yet reduced, and adequate retirement benefits are provided for a reasonable cost. Further details are provided in the appendix available on Milliman's Web site. Mark 011eman is a principal and consulting actuary in Milliman's Seattle office. This publication is intended to provide information and analysis of a general nature. Application to specific circumstances should rely on separate professional guidance. Inquiries may be directed to: Brent Banister, Editor; 1120 South 101st Street, Suite 400, Omaha, NE 68124-1088; (402) 393-9400; periscope@milliman.com. Offices in principal cities worldwide JANUARY 2009 ©2009 Milliman, Inc. All Rights Reserved 1 January 2009 r 1 I a•:i � � ,d�, ,'p��� iwl�i �iV�rPN�iP' i ON] lug III a �l dl��pkV r s � liii hll�d II ili,ai u Appendix - Public plan DB/DC choices This appendix provides further details. STATE SYSTEMS REFERRED TO IN THIS ARTICLE SYSTEM CURRENT PLAN 0 Milliman EFFECTIVE DATE ALASKA PERS & TRS DC JULY 1, 2006 1COLORAOO`PERA _ DB/DC-tHbICE' JANUARY 1, 2006 FLORIDA RS DB/DC CHOICE JULY 1, 2002 MONTANA'PERS [QB/DC:CHOICE "JULY ,1, 2002, NEBRASKA PERS HYBRID DB JANUARY 1, 2003 NORTH DAKOTA'PERS pB%DCC 'OICE,(LIMITED GROUP) JANUARY 1,2000 OHIO PERS DB/DC/COMBINED CHOICE JANUARY 1, 2003 OHIO STIRS : , ZB%DC%COMBINEDCHOICE ­JULY2; 2001„ OREGON PERS COMBINED AUGUST 29, 2003 LSOUTH,CAROLINA'RS .. DB/DC CHOICE JULY 1,2001 VERMONT SRS DB/DC CHOICE (LIMITED GROUP) JANUARY 1, 1999 WASHINGTON'STATE DB%COMBINED CHOICE MARCH 1 2002 WEST VIRGINIA TRS DB JULY 1, 2005 Systems with supplemental contributions The following systems have contributions paid as a percentage of DC member salaries that are not credited to DC member accounts. Supplemental contributions required to fund DB liabilities show that introducing a DC plan does not reduce the unfunded liabilities of the existing DB plan. Colorado PERA Amortization Equalization Disbursement (AED)—The AED has been 0.5% of pay in 2006, 1 % in 2007, and 1.4% in 2008. It is scheduled to increase 0.4% each year to a maximum of 3% in 2012. • Supplemental Amortization Equalization Disbursement (SAED)—The SAED is 0.5% in 2008, and is scheduled to increase 0.5% each year to a maximum of 3% in 2013. • In Colorado, the AED and SAED are both contributions to the DB plan to account for adverse selection. Both are applied to both DB and DC payroll, The AED is paid by employers. The SAED, although technically an employer contribution, is considered to be an employee contribution because it comes out of the employee compensation package. Both grade down when trust funds reach 100% funding. Florida RS • To fund supplemental disability benefits for DC members, a contribution ranging from 0.25% of DC member pay for general members to 1.33% of DC member pay for special -risk members is paid by employers into a separate side account. • Employers contribute 0.05% of pay to fund communication and adminis- tration. • In Florida, there is no payment as a percent of DC member salaries to fund DB unfunded liabilities. A member of Abelica Global milliman.com IP E R G c®mPutific Emolofreo Retirement Srs[ems _ Montana PERS The following contributions are made by Montana PERS employers as a percent of DC member pay: • A Plan Choice Rate (PCR) contribution equal to 2.505% of pay is made to the DB plan to prevent DB costs from increasing due to financing unfunded liabilities over a smaller payroll and increases in the normal cost rate due to antiselection. The PCR was 2.37% from inception at July 1, 2002, until July 1, 2007, when it increased to 2.505% of pay. • A payment of 0.30% is made to finance long-term disability benefits. • A payment of 0.04% is made to the education fund. Ohio PERS A contribution of 0.77% of pay for members in the all DC plan is made to the DB plan by the employer in 2008 as a "mitigation rate" The board reviews the mitigation rate annually, and it can vary between 0% and 6%. The highest level to date is 0.77%. Ohio STRS • 3.5% of pay from employer contributions for all DC members is used to pay for the unfunded liabilities of the DB plan. South Carolina RS • The South Carolina Retirement System currently collects 4.24% of the employer contribution and may retain an amount as determined by the director to defray any reasonable expenses incurred in performing services regarding the plan. This amount has changed as follows: 3.05% for FYE 6/30/2007, 4.06% for FYE 6/30/2008, and 4.24% for FYE 6/30/2009. Further system details The following section provides a brief summary of information relevant to this article for each system. Alaska Starting July 1, 2006, Alaska's public employee and teachers defined benefit plans are closed. New hires will go into the defined contribution plan. The default percent of pay contribution rates are 5% employer and 8% employee in PERS and 7% employer and 8% employee in TRS. Additional employee contributions may be elected once in the first 24 months of hire subject to the IRS maximums in Section 415. Alaska teachers do not participate in Social Security and many Alaska public employers, like the state of Alaska, have opted out of Social Security participation. Colorado Public Employee Retirement Administration (PERA) Starting Jan. 1, 2006, Colorado allowed new employees to choose between the PERA DB plan, the PERA DC plan, and three other state -offered DC plans. Members have a 60-day election window and can then change their minds once between the PERA DB and PERA DC plans either way in years two through five after retirement. If a member changes to the DC plan, the DB benefit is frozen based on service and salary to the date of the change and the member participates in the DC plan going forward. If the member changes to the DB plan, the member has the option to purchase his or her original time in the DB plan after one year based on actuarial value. The DB and DC plans require the same employer and employee percentage of pay contributions. The base contribution rates are 10.15% employer and 8% employee for state and school employees, and 12.85% employer and 10% employee for state troopers. For DB members, 1.02% of pay from the base employer contribution is used to fund retiree healthcare instead of pension benefits. For DC members, the 1.02% of pay goes into the members' DC accounts as part of the employer contribution and it is up to the members to pay for healthcare when they retire. The AED and SAED supplemental contributions described earlier are in addition to these base contribution rates. Table 5 is a historical record of the choices of new hires in Colorado PERA. Florida Retirement System (FRS) Starting July 1, 2002, Florida allowed new employees to choose between a DB plan and a DC plan. Members have a six-month election window and can change their minds once at any time before retirement or termination. Details of how the switch is treated are given in the main body of the article. There are no employee contributions to either the DB or the DC plan. Employer contributions to members' DC accounts range from 9% of pay for general members to 20% of pay for special risk. Employer contributions to fund additional disability benefits for DC members range from 0.25% of pay for general members to 1.33% of pay for special -risk members. Employers contribute 0.05% of pay to fund communication and administration. DC accounts vest 100% at one year of service. DB benefits vest 100% at six years of service. Accounts and benefits are 0% vested before these dates. 2 :: JANUARY 2009 u i� Table 6 is a historical record of the choices of new hires in Florida. Florida has an active education campaign. DC elections have increased each year and the overall DC election percentage of 26% in the year ending June 30, 2008, is the highest of any system in this study. Montana Public Employees' Retirement System (PERS) Starting Jan. 1, 2002, Montana PERS allowed new employees to choose between a DB plan and a DC plan. Members have 12 months after hire to make a one-time irrevo- cable decision between the DB plan and the DC plan. The DB and DC plans require the same employer and employee percentage of pay contributions. Employers contribute 7.035% of pay. Employees contribute 6.90% of pay. Employer DC contribu- tions can be broken down as 4.19% to the DC account, 2.505% plan choice rate (DB funding), 0.30% for long-term disability benefits, and 0.04% for the education fund. The entire employee contribution is credited to the DC account. Table 7 is a historical record of the choices of new hires in Montana PERS. Members not making a choice are placed in the DB plan by default; however, statistics are not available on what portion of new hires entering the DB plan did so by default. North Dakota Public Employees Retirement System (NDPERS) Starting Jan. 1, 2000, North Dakota allowed nonclassified state employees to choose between a DB plan and a DC plan. As only nonclassified state employees are eligible, there were only 291 members in the DC plan as of July 1, 2008. Members have six months after hire to make a one-time irrevo- cable decision between the DB plan and the DC plan. The DB and DC plans require the same employer and employee percentage of pay contributions. Employers contribute 4.12% of pay and employees contribute 4% of pay for a total contribution of 8.12% of pay. Table 8 shows that about 12% have actively elected the DC plan and 88% have either actively elected the DB plan or have not made a choice and have been placed in the DB plan as the default. Breakouts by year and the portion of DB elections that were active versus default are not available. Ohio Public Employees Retirement System (OPERS) Starting Jan. 1, 2003, OPERS allowed new employees to choose between an all-DB plan (the Traditional Pension Plan), an all -DC plan (the Member -Directed Plan), and the Combined Plan. PERG OPrE In the Combined Plan, employer contributions fund DB benefits and all member contributions are credited to DC accounts. Members have three chances to change their minds about their choice —once in the first five years after hire, once five to 10 years after hire, and once at any time after 10 years from hire and before retirement. Changes are prospective only, but members transfer- ring to the all-DB or combined plan have the option to purchase service in the new plan using their DC accounts. Service purchases are based on service in the plan the member is opting out of; must use the DC account first; and if the DC account is less than the total cost, then the member may still purchase all service with an additional lump sum, rollover, or payroll deduction. Frozen DB benefits are based on salary and service during DB membership only. The employer contribution is 14% of pay and the employee contribution is 10% of pay for all three plans and for all groups. Members in the all -DC and combined plans have all employee contributions credited to their DC accounts. However, a portion of the employer contribution is used to fund retiree health benefits (4.5% of pay in 2008). Also, the mitigation rate, which is currently 0.77% of pay, comes out of the 14% employer contribution and is not credited to DC accounts. Table 9 is a historical record of the choices of new hires in OPERS. State Teachers Retirement System of Ohio (STRS) Starting July 1, 2001, STRS allowed new employees to choose between an all-DB plan, an all -DC plan, and a combined plan. In the combined plan, employer contributions fund DB benefits and all member contributions are credited to DC accounts. Members have a six-month election window. After the member is put in the all-DB plan either by default or by active election, he or she cannot elect out. All changes after the first six months are effective at the end of the fiscal year following the fourth anniversary of the hire date. Members must positively elect to stay in the com- bined or all -DC plan at the end of the fifth fiscal year of participation or they will default into the all-DB plan. If members change into the all-DB plan, they forfeit their DC accounts and are treated as if they had been in the all-DB plan since hire. There are no changes after the end of the fifth fiscal year of participation after hire. The employer contribution is 14% of pay and the employee con- tribution is 10% of pay for all three plans. Members in the all -DC and combined plans have all employee contributions credited to their DC accounts. However, a portion of the employer contribu- tion to the all -DC plan is used to fund unfunded liabilities for the all-DB plan (3.5% of pay in 2008). JANUARY 2009 :: 3 H I I PEGIca 1 'Ftiblic Em oe RetirementS�to Table 10 is a historical record of the choices of new hires in STRS of Ohio. Oregon Public Service Retirement Plan (OPSRP) Oregon has chosen that starting Aug. 29, 2003, all new hires go into a combined pension plan with two components: the defined benefit pension program and the defined contribution Individual Account Program (IAP). The pension program provides a defined benefit equal to 1.5% of final average earnings (1.8% for police officers and firefighters) for every year of service and is funded entirely by employer contributions. The IAP is funded entirely by the employee contributions, which are 6% of pay. All IAP assets are invested in the same portfolio as the DB assets; there is no difference. Employees have no choice in how IAP assets are invested. As a result, the members' DC accounts earn the same return, positive or negative, as the DB assets. Earnings are credited annually to member accounts. Administrative fees are deducted from the fund's earnings as part of the annual crediting process. Members receive an annual statement after interest is credited each year. South Carolina Retirement Systems South Carolina allows new employees to choose between a DB plan and a DC plan. This arrangement was made effective over the period from July 1, 2001, to July 1, 2003, varying by group. DC members choose between four authorized investment provid- ers. Members must choose investment options from their chosen investment provider. Members may change investment providers during the annual open -enrollment period subject to the invest- ment provider's contractual limitations. Members have a 30-day election window after hire to choose between the DB plan and the DC plan. During their first five years, members can change from the DC plan to the DB plan. Members cannot change from the DB plan to the DC plan. If a member changes to the DB plan during this five-year period, the member has the option to purchase his or her original time in the DB plan. The cost is 16% of the member's highest career salary for each year of service. The member has the option, but is not required, to use his or her DC account for these service purchases. The DB and DC plans require the same employer and employee percentage of pay contributions. Employers contribute 9.24% of pay. Employees contribute 6.50% of pay. Five percent of employer DC contributions are deposited to the DC account; the South Carolina Retirement System currently collects 4.24% of the employer contribution and may retain an amount as determined by the director to defray any reasonable expenses incurred in performing services regarding the plan. The entire employee contribution is credited to the DC account. I _ 4 :: JANUARY 2009 Table 11 is a historical record of the choices of new hires in South Carolina. Like most other systems, the DB plan is the default election. It is interesting to note that the percent of new hires elect- ing DC varies widely by group. The percent of higher education employees choosing DC has varied from 32% to 37%, whereas the DC choice for other groups has only varied from 11 % to 16%. Vermont Starting Jan. 1, 1999, all new exempt state employees were given a choice between a DB plan and a DC plan. In addition, beginning in July of 2000, the governing body of employers in the Vermont Municipal Employees' Retirement System (VMERS) can elect to offer employees a choice between a DB plan and a DC plan. To date, about 77 of the over 400 VMERS employers have chosen to offer this choice to their employees. Employees make a one-time irrevocable choice at hire. In the state DC plan, employers contribute 7% of pay and employ- ees contribute 2.85% of pay. In the VMERS DC plan, employers contribute 5% of pay and employees contribute 5% of pay. Statistics on the percentage of members electing the DC plan or DB plan are not available. Washington State Department of Retirement Systems Starting March 1, 2002, Washington state allowed new hires in the Public Employees' Retirement System (PERS) to choose between an all-DB plan (Plan 2), and a combined plan (Plan 3). In the combined plan, employer contributions fund DB benefits equal to 1 % of final average earnings for each year of service and all member contributions are credited to DC accounts. Starting July 1, 2007, new hires in the Teachers' Retirement System (TRS) and the School Employees' Retirement System (SERS) were given the same choice between Plan 2 and Plan 3. Members have 90 days after hire to make a one-time irrevocable decision between the all DB plan and the combined plan. At the same time the plan election is made in the first 90 days, members in the combined plan (Plan 3) also choose between six employee contribution -rate options. Once the employee contribution -rate option is chosen, it cannot be changed as long as the member remains with the same employer. If members sepa- rate from the employer, they may change their contribution rate with the next employer. All employee contributions are credited to the DC account. The six employee contribution options in the combined plan are as follows: Option A: 5% of pay contribution at all ages Option B: 5% to age 35, 6% at ages 35 to 44, 7.5% at ages 45 and up Option C: 6% to age 35, 7.5% at ages 35 to 44, 8.5% at ages 45 and up F� I Option D: 7% of pay contribution at all ages Option E: 10% of pay contribution at all ages Option F: 15% of pay contribution at all ages Employees who do not make an election in the first 90 days after hire are placed in the combined plan (Plan 3) with employee con- tribution option A. Approximately 58% of combined plan members are in option A, with the remainder spread fairly evenly between the other five contribution options. One of the DC investment options is the Total Allocation Portfolio (TAP), which mirrors the investments in the state DB plan and therefore earns the same returns. Washington has made the TAP the default investment option for Plan 3 and approximately 61 % of the members' DC assets are in the TAP option. Starting in October of 2008, target date funds managed by an outside pro- vider are also available. The target date funds allocate investments without the member's involvement and automatically change the asset mix as the member moves closer to retirement. Table 12 shows that approximately 63% of the PERS members hired between March 1, 2002, and Sept. 26, 2008, have actively chosen the all DB plan over the combined plan, which is the default. Breakouts of choices by year are not available. West Virginia Teachers Retirement System The following chronology of the West Virginia TRS fills in some holes not described in the article. • 1941—West Virginia TRS was established as a DC plan. • 1960s and 1970s—DB benefits were added to counter the inad- equate DC benefits, but the benefits were never properly funded. • 1991—The DC plan (TDC) was established for new hires in response to funding problems and 4,500 former DB partici- pants also switched from the DB to DC. • 2003—Many of the 4,500 who switched felt misled and said they could not afford to retire. Other DC members were also not satisfied. • 2005—The state decided that a given level of benefits could be funded for a lower cost through a DB plan. Average DC returns had been lower than DB returns in both up and down markets. Changing to a DC plan did not solve the state's funding problems. All members hired after July 1, 2005, go into the DB plan instead of the DC plan. West Virginia projected a $1.2 billion savings in the first 30 years due to moving new entrants from the DC to the DB plan. • 2006 and 2007—Special appropriations of $290.1 million in FY2006 and $313.8 million in FY2007 were deposited. In addition, West Virginia completed a tobacco bond securitization in FY2007 and deposited $807.5 million of those proceeds into TRS as another special appropriation. All these amounts were in addition to the regular contribution determined under the ARC, which was converted to a level dollar amortization (from level percentage of payroll). Clearly, West Virginia is demonstrating a new DB contribution discipline. • 2008—DC members are given the option to switch to the DB plan. Of those DC members, 78.6% (14,925 members) chose to switch to the DB plan. Surprisingly, the switch, which was expected to cost the state up to $78 million before the elections were made, is now expected to save the state about $22 million. Fewer older TDC members than expected transferred. More young TDC members than expected transferred. Fifty percent of those over 70 transferred. Sixty-nine percent of those age 65 to 69 transferred. Eighty-one percent of those 45 to 64 transferred. Seventy-six percent of members under age 40 transferred. Table 4 shows the investment returns for the seven years ended June 30, 2001, through June 30, 2007. The seven-year average DB return was 3.15% higher than the average DC return. DB investments did better in both the best and worst investment years. The average DC return was only higher in 2003 when DC investments averaged 4.84% and DB investments earned 4.75%. JANUARY 2009 :: 5 WEST VIRGINIA TEACHERS' DC RETURNS COMPARED TO DB YEAR ENDING JUNE 30 2001 DC PLAN -2.60% DB PLAN -0.25% �2002 2003 4.84% 4.75% 2004 8:83% 15A8% 2005 6.33% 10.56% 2006 6173% �9.55oh 2007 11.85% 1 Z43% 7 YR'AVERAGE 4.59% 7.74% I COLORADO PERA NEW HIRE CHOICES' (EFFECTIVE DATE: JANUARY 1, 2006) DB BY DEFAULT DB ACTIVE ENROLLMENTS DC ACTIVE ENROLLMENTS "2006 37% '48% 14% 2007 39% 43% 18% 1/08r"5%08 35% 43% - 11% 'BASED ON 11,200 NEW HIRES. FLORIDA RETIREMENT SYSTEM NEW HIRE CHOICES` (EFFECTIVE DATE. JULY 1, 2002) DB BY DEFAULT DB ACTIVE ENROLLMENTS DC ACTIVE ENROLLMENTS 9/,02 -J6/03 86% 6% 8% 7/03 - 6/04 73% 11% 16% 7/04 -'6L05 .61W 118% 21,% - 7/05 - 6/06 59% 19% 22% 7/06 -;6/07 18% 18% 24% 7/07 - 6/08 55% 19% 26% ' AT JUNE 30, 2008 THERE ARE 609,888 DB MEMBERS AND 95,392 DC MEMBERS. 6 :: JANUARY 2009 I I H S c o p 1E I Public Eimlovee Nhimme nf MONTANA PERS NEW HIRE CHOICES (EFFECTIVE DATE: JULY 1, 2002) DC ACTIVE ENROLLMENTS 7/64 —6/,05 1,9% 1 7/05-6/06 10% 7/,06-'007 1,09/0 7/07-6/08 10% NORTH DAKOTA PERS NEW HIRE ELECTIONS FROM JANUARY 2001 THROUGH JUNE 2008* (EFFECTIVE DATE: JANUARY 1, 2000) DB BY DEFAULT DC ACTIVE ENROLLMENTS �88% 12% * THERE ARE 291 MEMBERS IN THE DC PLAN AS OFJULY 1, 2008. OHIO PERS NEW HIRE CHOICES* (EFFECTIVE DATE: JANUARY 1, 2003) DB BY DEFAULT DB ACTIVE DC ACTIVE COMBINED PLAN ENROLLMENTS ENROLLMENTS ACTIVE ENROLLMENTS 1004 j84%. 110/0 3 :2% 2005 84% 10% 3% 3% 2006 83% 12% 3% 2 ow 2007 82% 13% 3% 2% 1108 -6/08 79% 15%, 4% '2% - BASED ON 228,234 NEW HIRES. I 11 I I JANUARY 2009 :: 7 OHIO TEACHERS (STRS) NEW HIRE CHOICES' (EFFECTIVE DATE: JULY 1, 2001) DB BY DEFAULT DB ACTIVE ENROLLMENTS DC ACTIVE ENROLLMENTS COMBINED PLAN ACTIVE ENROLLMENTS 7/01­_6/.04' 69%. 7/04 — 6/05 70% 16% 11% 4% 4%05 - 06 72% 13% 11% 4% 7/06 — 6/07 72% 13% 11% 4% 4/07'4 6106" 71% 14% 1'1% -BASED ON 123,781 NEW HIRES. I SOUTH CAROLINA RETIREMENT SYSTEMS PERCENT OF NEW HIRES ELECTING DEFINED CONTRIBUTION' (EFFECTIVE DATES: JULY 1, 2001 TO JULY 1, 2003) HIGHER ED. K - 12 SCHOOLS STATE AGENCIES OVERALL `7/04 =76/M '32%. 14%. 7/05 — 6/06 34% 14% 12% 18% 7/06 — 6/07 37% '15% ' 13% 19% 7/07 — 6/08 35% 16% 13% 20% ' BASED ON 128,459 NEW HIRES. I 11 CUMULATIVE WASHINGTON PERS NEW HIRE ELECTIONS FROM MARCH 2O02 TO SEPTEMBER 2008 PLAN 3 COMBINED PLAN 3 COMBINED DB & DC PLAN 2 ALL DB DB & DC BY DEFAULT ACTIVE ENROLLMENTS ACTIVE ENROLLMENTS 18o/. 1180/0 '63% i� 8 :: JANUARY 2009 1 IJ 11 SECTION 2 PROPOSAL REQUIREMENTS 1. Three references of work similar to this project, preferably municipalities or public agencies within Colorado. System Reference Metro Wastewater Reclamation District Mr. Alan Cohen 6450 York Street Finance Officer Denver, Colorado 80229 (303) 286-3000 City of Boulder Ms. Cappie Fine 1777 Broadway Boulder, CO 80302-5221 (303) 441-3014 Northern Colorado Water Conservancy District Ms. Minerva Lee 220 Water Avenue Head, Human Resources Berthoud, CO 80513 970 622-2212 2. Resumes of key personnel involved. Include information about credentials held, professional organization memberships, involvement with Employee Retirement Plans. Indicate who would provide backup if the primary actuary could not complete the study. Your primary contact will be a team of three professionals in Milliman's Denver office who will also perform the majority of the work. The principal and managing actuary will be Pat Kahle. Joel Stewart will provide consulting expertise, technical support, and will be available I on a daily basis in the event Pat is not available. Francine Moyer will provide the technical support for the work. I The project team will include an internal senior consultant, Pat Beckham, a principal in Milliman's Omaha office. Pat will be an additional resource for the GERP as needed. If something were to happen to Pat Kahle, Pat Beckham would become the lead consultant for City. In the event Pat Beckham was not available, there are a number of public sector consultants in the Portland and Seattle offices that could step in to assist. Pat Kahle Joel Stewart Francine Moyer Pat Beckham ■ Proposal to Provide Actuarial and Consulting Services for 4 M i I I i ma n the City of Fort Collins General Employees' Retirement Plan I I I 1 I A PER O S CO P 'Public Employee Reiiiement syslems References Damsel, Bill, Assistant Director, Defined Contribution, Ohio Public Employees Retirement System Davis, Tammy, Assistant Director, Payroll and Financial Reporting, South Carolina Retirement System Green, Garry, Research and Education Section Administrator, Florida Retirement System Hardesty, Michelle, Assistant Director Retirement Services Division, Washington Department of Retirement Systems Lea, Kathy, Retirement Manager, Alaska Division of Retirement and Benefits Mackee, Dennis, Director of Communications for the State Board of Administration of Florida Mandel, Harry, Board Actuary, West Virginia Consolidated Public Retirement Board Maurek, David, Chief Operating Officer, Colorado Public Employees' Retirement System Nebraska Public Employees Retirement Systems, Retirement Roundup, State & County Employees, Vol. 17, No. 5, Summer 2002 Orr, Dale, Actuarial Analysis Section Manager, Oregon Public Service Retirement Plan Reinhardt, Bryan, Research Analyst, Benefits Planner, North Dakota Public Employees Retirement System Russell, Gary, Director, Member Services, State Teachers Retirement System of Ohio Samson, Kathy, Defined Contribution Plans and Education Services Bureau Chief, Montana Public Employee Retirement Administration Webster, Cynthia, Director, Vermont State Retirement System This publication is intended to provide information and analysis of a general nature. Application to specific circumstances should rely on separate professional guidance. Inquiries may be directed to: Brent Banister, Editor; 1120 South 101st Street, Suite 400, Omaha, NE 68124-1088; (402) 393-9400; periscope@milliman.com Offices in principal cities worldwide January 2009 C 2009 Milliman, Inc. All Rights Reserved I I 1 11 August 2009 Milliman D D D D Publlc Employee Retirement :Systems Using Risk Profiles to Manage Public Pension Funding Goals By Joshua Davis, FSA, EA, MAAA, Scott Porter, FSA, EA, MAAA, and Karen Steffen, FSA, EA, MAAA Most plans have historically used the concept of the plan's funded ratio, the simple relationship between the plan's Actuarial Value of Assets (AVA) and its Actuarial Accrued Liability (AAL) to measure progress toward funding pension obligations. As long as the funded ratio increased over time, many plans were satisfied with the arrangement. One consequence of this approach was that saving too much was discouraged. If a plan were close to 100% funded or overfunded, governments often felt the plan had "too much" money and that other uses for these public funds were more urgent. The concept of being more than 100% funded often surfaced within the public plan sector during the investment boom of the 1990s when accumulated assets grew much faster than their obligations. Before that time, as long as some progress was being made on increasing the funded ratio, a plan was considered to be doing well. Many decision makers in the late 1990s had perhaps forgotten (or never heard of) the market decline in the 1970s that necessitated a spate of increased contributions. The funded ratio measurement currently fails to recognize two key components to the detriment of all stakeholders. It does not recognize to what extent a plan is subject to market volatility, a shortcoming brought into focus by the recent economic turmoil, nor does it adequately reflect a plan's maturation process. New funding policies and actuarial models must evolve to help manage the investment risk on the liability side of public pension plans' balance sheets, because neither the Traditional Actuarial Practice (TAP) disclosures nor any new Market Value of Liabilities (MVL) disclosures give policy makers enough information to make sound funding decisions as a stand-alone measurement. Public pension plan policy makers need to shift their focus from simply managing the plan's funded ratio to managing the plan's risk profile. Seeking a multidimensional funding measurement A multidimensional measurement is needed by the pension community to determine future investment decisions, contribution rates, and benefit plan changes. Assessing a plan's risk profile is an important first step and should take into account the expected maturing of the plan. Establishing and implementing an effective formal funding policy is critical for public plans going forward. An effective funding policy needs to be able to answer the following: • What contribution rate can the public employer maintain to sustain the plan? • How will the government maintain a well -funded plan? • How will the plan maintain stable employee and employer contribution rates? • How should the funding policy react when the plan is more than 100% funded? • What are the risks if things do not go as planned (i.e., asset values fall and the employer is unable to maintain the contributions required to sustain the plan)? Well thought-out funding policies, based on a plan's risk profile, need to be devised in advance. They will help plan trustees better appreciate and handle the actual risks in their plans. Key factors contributing to a plan's risk profile Public pension plans can be assigned a spot on a risk continuum. At the riskiest end are underfunded, mature plans invested in volatile securities with little room in the budget to adjust for potentially higher contributions. At the least risky end are overfunded plans with a healthy active member population, whose assets are invested in a dedicated bond portfolio matching their obligations, and which have the ability to adjust their budget for changes in contribution requirements. A member of Abelica Global milliman.com I I Ll I I �ER�Sco�E Both MVL and TAP measures are silent about some major risks embedded in pension plans, including contribution volatility, future investment risk, and the maturity of the plan. Contribution volatility — How volatile will future contribution rates be? Can the employer adapt to volatile contribution requirements? At what point can the employer no longer afford the benefits promised? Answers to these questions are based on the underlying characteristics of the plan's demographics. Moreover, answering them should help the employer adopt policies that meet these goals. An employer with a growing active member population is a healthy employer with the means to adapt to volatile contributions. As an employer's growth stabilizes, its budget policies have more trouble accommodating volatile contributions. Its risk profile has changed. An employer who understands its budget risks should adopt policies to manage its pension plan accordingly. How do trustees adapt to changes in their plan's risk profile, before they occur, through their funding policies? This is the $64,000 question and depends on all the other factors below. Future investment risk — What investments are backing the required future cash flows and what kind of investments are they? The nature of investment risk and return, and the historical difference in performance and volatility among such asset classes as equities, bonds, and commodities, play a major role in a plan's risk profile. (These differences are well documented and so will not be discussed in detail here.) However, it is worth noting that historically, public plans' investing in equities led to a significant increase in investment revenue that helped keep the level of contributions down. However, recent events prove that the risks assumed with these equity -based portfolios need to be better managed or at least better understood. In hindsight, one of the biggest problems public plans faced during the market upswing in the late 1990s was what to do with the "excess" money, if they reached or exceeded the 100% funded ratio. For decades, the underlying assumption was that once a plan reached the 100% funding goal, that full funding level would persist and maybe even grow. The concept that any funds greater than 100% were "excess" or available for other needs arose in many places around the country. Few trustees understood the risks associated with the investments that generated these excess assets, and that perhaps these funds would be best saved for a rainy day. The key questions employers and all stakeholders should be asking are these: Are the investments consistent with employer's ability to adjust to volatile contribution requirements? Are potential volatile contribution requirements in keeping with the evolution of the plan's demographics? In reviewing valuation reports from the past 20+ years for several public retirement systems, we have found that as the plan's demographics have matured, plans have increased their exposure to riskier asset classes, and at the same time budget constraints have gotten tighter. One might believe that this is the opposite reaction employers and trustees should have given the circumstances. Now the question is: Can employers and trustees afford to reduce the risk in their pension plan portfolios? Can they afford not to? Maturity of the plan — How mature is a plan? How is that maturity level changing and how might those changes affect the contribution and investment risks over time? For many public trustees, one concept used in setting its funding policies is intergenerational equity, (i.e., the cost for one generation of taxpayers is consistent with the cost for prior and future generations of taxpayers). Current actuarial methods do not take into account expected changes in the maturity of the plan and how those changes might adjust the contribution and investment risks over time. How does one measure plan maturity? And how should adjustments in the risks be reflected in the funding policies of today to be consistent with the concept of intergenerational equity? The following measurements can be used to assess a plan's maturity level: 1. Active -to -retiree ratio The first is a simplified measure of the number of active members versus the number of retirees. Today an increasing percentage of employees in the public sector are collecting retirement benefits. The average active -to -retiree ratio has decreased by more than one- third, from 3.1 to 1.9 active members per retiree from 1985 to 2005. In addition, the life expectancy for the average 60-year-old male and female member has increased by 11 % and 4%, respectively. 2. Percentage of retiree liability to total plan liability A second measure weights these head counts by their liabilities (i.e., the percentage of the plan's accrued liabilities attributable to retirees compared to the plan's overall liabilities). Over the last 20 years, the percentage of actuarial accrued liabilities attributable to retirees has increased from approximately 33% to almost 50% for the systems reviewed. 3. Accrued liability to annual benefit payment ratio The third measure is the ratio of a plan's accrued liabilities to its current year's benefit payments. This method provides a simplified measure of how many years of benefit payments are covered, excluding future investment income and contributions. Over the last 20 years, the ratio has declined from slightly more than 30 to slightly more than 20. Exhibit 1, on page 3, charts these three measures of plan maturity from 1985 to 2005 for 10 statewide systems. Alternative funding policies Two alternative funding measurement options could promote fuller recognition of a plan's investment risks while taking into account the maturation process of retirement systems. These alternative funding policies are designed to align a plan's underlying portfolio risks with its liabilities and its maturity. They would encourage plan trustees to focus annually on the plan's risk profile when devising asset allocation and funding strategies, to reduce contribution volatility and to strengthen the plan's long-term viability. 2 :: AUGUST 2009 I I EI I 1 Option 1: Fund in excess of 100% One alternative, based on a plan's risk profile, is to fund the plan in excess of 100%, with the surplus used to offset possible future losses. This strategy provides a cushion against market volatility and helps reduce contribution volatility. The risk profile of the investments and the plan's inherent demographic risks determine how much of a cushion is needed. This option has an added advantage of using the same terminology and concepts as the current 100% funded ratio goal. It uses the same measurement criteria and terms but recalibrates them to a new ultimate funding goal. As common asset portfolios are, on average, expected to deviate about 10% a year, a corresponding buffer for the active member liability should be established as protection against a market downturn. Because retired members have a growing liability without corresponding contributions, two standard deviations or 20% is advisable for the portion of the liability attributable to retirees. If a plan has 50% of its obligations (AAL) earmarked for retired members, then an ultimate funding goal of 1 15% is appropriate, 1 10% for the active member obligations and 120% for the retired member obligations. For a maturing plan, the percentage of the AAL earmarked for retired members is increasing over time. Therefore, if a system decides to maintain its current asset allocation strategy in the future, the ultimate funding goal will increase over time, resulting in additional contributions. On the other hand, if the system decides to adjust its asset allocation strategy to be less risky over time to coincide with the changing maturity of the plan, additional contributions will be required as less investment income would be expected. In either case, this additional dimension of the evolving plan's maturity should be taken into account today in determining asset allocation and funding strategies. Just as the Entry Age Normal Cost method requires higher payments early in members' careers to spread the costs over their working lifetimes, setting a funding goal higher than 100% EXHIBIT 1 - MEASURES OF PLAN MATURITY MEN 01NOp will allow employers and employees to pay more when the market is good and tax revenues are up than when the market and tax revenues are both down. Exhibit 2, on page 4, illustrates the point that excess returns in one year could be accumulated and used to offset deficient returns in another year, resulting in more stable contributions over time. The chart displays hypothetical actuarial rates of return of a portfolio composed of 60% equity and 40% bonds based on the historical investment results reported by Ibbotson for the last 78 years. The actuarial returns are based on portfolio gains and losses smoothed over a five-year period. The market value return for 2008 was computed to be -19% for the total fund. It is interesting to note that with five-year smoothing, the only time a negative return occurred was during World War II and the Great Depression. Information for modeling stock and bond performance was taken from the Ibbotson SBBI 2008 Classic Yearbook published by Morningstar, Inc. Politically, it would be important to understand that funds in excess of the 100% measurement could not be used for benefit improvements or decreased employer contributions. As a minimum, the Normal Cost contributions should continue to be paid when the 100% funded ratio level has been reached until a sufficient surplus has been attained. Option 2: Modified liability -driven investment (LDI) strategy To pursue a modified liability -driven investment (LDI) strategy is another possibility. This option focuses on a plan's maturity level in deciding how to adjust the asset allocation mix between equities and bonds to reduce the effect of market volatility. A plan would invest in fixed income securities with staggered maturities to cover annual benefit payments for the foreseeable future, (i.e., for anywhere from five to 15 years). The assets backing these cash flows should not be "chasing" equity risk premium. A plan with a lower risk profile would cover a shorter time horizon with its bond holdings. SYSTEM ACTIVE TO RETIREE RATIO % OF AAL FOR RETIREES 1985 2005 1985 2005 RATIO OF AAL TO ANNUAL BENEFIT PAYMENTS 1985 2005 A 2.9 1.3 36% 60% 30.4 18.1 B 5.1 1 I I 25` " ':45 :: 35.6 '23.9 C 3.1 2.4 35 41 28.8 25.6 D 11.7, 1A 39 49 ;25:2 18:0 E 2.4 1.6 34 44 26.8 20.2 F 12.6 14 39 '45 27.0 43:6 G 2.3 1.5 36 57 30.6 20.1 H W 2.2 34 51 25:9 •20.8 1 4.4 2.6 34 47 25.8 19.8 1 2:6 2:0 22 39 45.2 i24.9 AUGUST 2009 :: 3 11 I PE LJ U O S O P LE 1 'Public Employ e.Retirement Systems As a plan matures (i.e., as a larger percentage of the plan's liabilities are owed to members nearing retirement age or in retirement), a higher percentage of the plan's assets would be invested in a dedicated bond portfolio maturing within this foreseeable future time horizon. For the rest of the assets, a balanced, diversified portfolio would be maintained. The maturity of the plan and the risk profile of the remaining investments determine how many years out to set this time horizon. In a bull market, where the balanced, diversified portfolio grew in value, additional years of benefit payments could be backed by bonds through the sale of profitable assets in the diversified portfolio, thereby "de -risking" the plan. Based on the desired risk profile, each year the board will have to decide how many years of benefit payments to secure with dedicated bonds. This mechanism would encourage the trust to engage in an annual exercise in risk management for the plan. This would turn the discussion away from increasing benefits to managing the risk inside the plan to secure the benefits already promised. Boards should no longer use an annual funded ratio as the sole decision -making factor. Funded ratios will be based on a select and ultimate interest rate structure. The select period and associated interest rate will be based on the number of years that benefit payments are backed by bonds and the average yield of those bonds if they were held to term. The ultimate interest rate will be the traditional long-term median return of the diversified portfolio. If the number of years of benefit payments backed by bonds increases, the overall discount rate will be reduced, leading to a lower funded ratio but a reduced risk profile. This is known as "buying down the risk" On the other 25% 20% 15% 10% 5% 0% -5% hand, if the number of years backed by bonds decreases, the funded ratio may increase, but the risk profile of the plan will also rise. Over the years, as the plan matures, more of the liabilities will be backed by bonds regardless of how the other assets are invested. Conclusion The new focus for public plan sponsors should be on maintaining a target risk profile and funding it accordingly. The trustees' main job is to manage the risk profile of the plan. How are they going to structure the plan to meet its cash flow requirements and portfolio growth 30 years in the future? Plans need to decide what level of benefits the system wants to provide for future generations first. A current risk profile analysis should be performed and a targeted risk profile should be decided upon. Then a contribution rate is set that has a high likelihood of covering the required cash flows in the future. Finally, the plan's risk profile will guide the choice of investments. In the future, actuaries must make sure that the funding policy and investment policy jive with the plan's risk profile, because the risk profile —not the AVA/AAL funded ratio alone —will give plan decision makers a fuller and more adequate picture of the state of the plan, its future needs, and its long-range viability. This article is based on a paper presented at the SOA's Public Pension Finance Symposium held May 18-19, 2009. To access the complete paper (and other papers presented at the symposium), please follow this link: http://www.soa.org/meetings-and-events/ handoutsl2009-chicago-public-pension-pa pers. aspx. 1930 1935 1940 1945 1950 1955 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 O Actuarial Return - 8.00% ' 60140 equity/bond split "' 8% interest rate assumption This publication is intended to provide information and analysis of a general nature. Application to specific circumstances should rely on separate professional guidance. Inquiries may be directed to: Brent Banister, Editor; 1120 South 101st Street, Suite 400, Omaha, NE 68124-1088; (402) 393-9400; periscope@milliman.com. Offices in principal cities worldwide AUGUST 2009 ©2009 Milliman, Inc. All Rights Reserved I 1 WINTER 2009/2010 The Transparency of Target Date Model Portfolios The DB and NDCP Funding Conundrum Considering a Healthcare Dependent Eligibility Audit? K7 Milliman PERSPECTIVES D, Current`Issues in Employee Benefits The Transparency ®f Target Date Model Portfolios Charles Hodge, Investment Services Consultant Target date funds burst on the scene relatively recently and have been embraced by both defined contribution (DC) plan sponsors and their employees. Target date funds respond to plan sponsors' concerns that many participants need investment guidance and simple investment choices. They appeal to plan participants because they require only a few decisions based on easy -to -understand criteria, such as when the participant intends to retire. (Thus, a 2020 target date fund is designed for someone who is scheduled to retire in or near the year 2020.) 2008 was a significant year for target date funds. The negative market performance clearly impacted the return, and exposed the risk, of target date funds. Additional transparency and disclosure is very likely for these funds, whether legislated by the government or demanded by the market. As of September 2009, there were more than 388 target date funds offered by 45 mutual fund companies. Most of them, however, were being offered by single -fund families, with all the underlying asset classes being managed by that one fund company. A new and perhaps better kind of target date fund has been developed, called a Target Date Model portfolio, which potentially offers advantages over other target date mutual funds. The idea behind a Target Date Model portfolio is the possibility of constructing target date funds using a DC plan's existing investment options. Not only are these Target Date Model portfolios tailored to the individual DC plan, but they can also be adapted to the individual plan's participants. What are Target Date Models? A Target Date Model is an allocation strategy, using the core options in a sponsor's 401(k) plan, whose asset mix becomes more conservative as the participant ages. Target date funds in general Several ideas are incorporated into target date funds to help participants balance both their market risk and longevity risk. Target date funds are diversified portfolios, designed to reduce the volatil- ity of participants' investments. And they use varying methodologies (or "glide paths") to adjust risk over time as plan participants age and approach retirement. The funds' asset allocations are designed to have a higher exposure to risk, and thus earn a greater expected return, when participants have a longer time horizon before they retire. They seek to protect assets (with more restrained growth opportunities) at or near retirement. The specific glide path used varies from fund company to fund company. The range of target date asset classes and styles varies as well. For example, the U.S. equity allocation among all the target date funds listed in Morningstar (for September 2009) varies from 75% to 0%. Among just the funds targeted for 2030 and longer, the domestic equity allocation ranges from 74% to 0%, international exposure ranges from 40% to less than 1 %. Among all target date funds, the energy sector exposure ranges from 45% to 1% of equity and Imilliman.com J 1 the financial sector exposure from 84% to 10/b. Among the 2030 and longer funds, energy represents anywhere from 45% to 3% of domestic equity and financials represent from 34% to less than 1 %. Style and capitalization vary among the different families as well. For funds targeting retirement in 2030 and beyond, 75% of the funds had an average equity style categorized as large -cap blend, 13% had a large -cap growth style, and another 3% had a mid -cap blend. The average quality of the bond holdings also varied from AAA to B. A range of possible asset allocations exists across fund families. Target date funds also differ according to the age at which the asset mix becomes static. Many target date strategies assume all of a participant's income needs will be covered by an asset mix that is fixed on the day of retirement. Others recognize the need to address increasing life expectancy by using a slightly more aggressive glide path that has participants continue to invest a percentage of their assets in equities into retirement. Some only arrive at their most conservative mix of assets when the retiree reaches age 75. What is a glide path?: The glide path is a predermined rate at which a portfolio's equity exposure decreases over time. The portfolio is designed to have greater risk (and volatility) in ;the participant's younger years, and . less risk near and in retirement. Target Date Model portfolios may offer some fiduciary protection In 2006, target date funds got an important endorsement from the federal government. The Pension Protection Act (PPA) of 2006 created guidelines giving fiduciary protection to DC plan sponsors that adopt an automatic enrollment arrangement if they use a qualified default investment alternative (QDIA). In 2007, the Department of Labor issued regulations stipulating that "targeted retirement date" funds do indeed qualify as QDIAs. However, that is not to say that the fiduciary protection is absolute. Plan sponsors still bear fiduciary responsibility for the choice of whichever target date fund they use as their plan option, if they include one. A Target Date Model portfolio offers plan sponsors greater certainty about what exactly they, as fiduciaries, are offering their plan participants. Other advantages of these model portfolios Specifically, Target Date Model portfolios offer a wide range of advantages over single -fund -family target date funds: • they are constructed using individual plan sponsors' existing retirement plan investment options; • these underlying assets are selected according to their likelihood of representing the best in their class, regardless of which invest- ment management firm runs them; individual underlying assets can be replaced if and when it seems prudent to do so; • the fees associated with the underlying assets are known; • lower -cost index funds and investment trusts can be used; and • the glide path is known. In general, with a Target Date Model, an individual participant's port- folio is adjusted annually. For example, the portfolio's equity alloca- tion might decrease from 49% to 47% of the portfolio with the fixed - income allocation increasing by this same 2%. In contrast with off -the -shelf target date funds, Target Date Model portfolios are more flexible, more transparent, and their costs are more easily controlled. Some disadvantages of single -fund -family target date funds While one -stop shopping seems appealingly simple, target date funds offered by a single -fund family are by definition undiversified by fund family. All of the underlying assets in the target date fund are typically run by managers employed by the same company, with the same philosophy and same research resources. Moreover, target date mutual funds offered by a single -fund family have these additional disadvantages: no opportunity for best -in -class selection beyond the few (but not all) asset classes in which they excel; no plan sponsor cost control regarding the choice of underlying assets; and no glide path transparency. Target Date Model portfolios are multi -manager One benefit of a Target Date Model portfolio is the ability to control costs using a multi -manager design. For example, plan sponsors can choose to use both active and passive managers, funds or collective trusts with lower fees, and investments with and without recordkeeping fee offsets. The advantages of this are many. Building a portfolio across fund families, using best -in -class investments, plan sponsors can trim expenses associated with the plan's investment line-up, improve the overall quality of the fund, lower unwanted risk, and even pick a glide path that fits their participants' risk tolerance. Ease of evaluating performance and changing managers Target date funds are all very different in terms of the assets used in their portfolios, asset mixes, time horizons, and their glide path methodologies, which makes them hard to compare with one another. Comparisons to a peer group may indicate a difference in performance or risk, manager success or failure, or just systematic differences in the way the funds are built. In contrast, the evaluation of Target Date Model portfolios is relative- ly simple. Because they are constructed using the known investment 2 :: WINTER 2009/2010 J o [EH[E[FQf3 PERSPECTIVES Gurrem Issue n Employee Beia6ts Features for Participant Portfolios Target Date Funds Target Date Models InvestMap Simplifies employee communications regarding fund objectives Yes Yes Yes Gan reduce total number of funds No Offers best -in -class funds within a professionally built portfolio No Yes Yes Allows ability to, control+ouerallexpense` No Yes Can change underlying managers No Yes Yes :DIversifiedacrossfundfamily '> - , t _ '` No `' Yes Yes Simplifies employee investment allocation decision Yes Yes Yes Can £bed mplemmented ,w'thouf clumrsy risk ques#ionnai es Yes , , Yes Yes j Allocation based on maximizing wealth and reducing the probability Maybe Yes Yes of plan participants outliving their savings 14 Adfusts asset allocation.oy�r time , a , ' ? Yes'` Typical age at which asset allocation becomes static 60-65 75 75 1 1 Full disclosure of glide path No Yes Yes Participant can easily adjust risk profile or position on glide path No No Yes options in a plan sponsor's particular plan, their evaluation mirrors Responding to the economic crisis the evaluation of the underlying assets. Rather than try to evaluate Target date funds did not escape the economic turmoil of the past the target date model's performance as a whole, plan sponsors can year, and some people are calling for greater disclosure about fees, concentrate on the performance of its various component parts. If better explanations to plan participants about possible risks, and and when underperforming investments of the plan are replaced, more transparency about the glide path being followed. Target Date that component of the target date models is also replaced. Model portfolios provide solutions to each of these concerns. Tailoring participant portfolios While target date models are typically implemented in 5- or 10-year increments, options exist for tailoring portfolios to each partici- pant's individual age. An example of this capability is offered by InvestMapTM, which is Milliman's platform -specific answer to model portfolios and participant customization. It generates a different port- folio and glide path for each individual according to the specific year of retirement. It offers a specific portfolio for every retirement year. In addition, a participant can choose to adjust the risk profile so that the "portfolio age" can be more conservative or more aggres- sive than the participant's actual age. (With most target date funds, an investor's decision is driven solely by the retirement date, not according to an individualized risk tolerance assessment.) Life :expecfa ncyy Most"target date;funds focus on the'date of . retirement. With a current life expecfancyof1 85 for women and 82 for men, employees retiring at age 65 need some 20 years of income in retirement. There is much to be said in favor of target date funds generally, but plan sponsors should consider the further advantages of Target Date Model portfolios or tools that can be used to customize portfolios for individual participants. Target Date Model portfolios (and InvestMap) are diversified by investment management company as well as asset class and style. They can help plan sponsors measure and control fees and enhance performance through best -in -class investments. And plan participants can adapt portfolios to their individual needs. Target Date Model portfolios are ultimately the more transparent investment. Exactly what they are composed of (the underlying assets) and how the asset allocation will change over time (the glide path methodology) are known, making them relatively easy to under- stand, implement, explain to plan participants, and evaluate. Charles Hodge is an investment services consultant in Milliman's Dallas office. This article was peer reviewed by Randy Mitchell, who is also an investment consultant with Milliman in Dallas. WINTER 2009/2010 :: 3 Ll F1 o [E[N]E[F ff3 PERSPECTIVES rCurre t I - - EmPloyae Benefita' Dominick Pizzano, Compliance Consultant The Pension Protection Act of 2006 (PPA) and Internal Revenue Code section 409A(b)(3) created a mandatory funding connection between a plan sponsors defined benefit (DB) plan and any nonqualified deferred compensation plan (NDCP) it might have. That is, if circumstances arise that could either actually or potentially have a negative effect on the funding of the DB plan, the law prohibits the sponsor from "funding" an NDCP for certain designated employees. Basically, a plan sponsor may not fund the one (the NDCP) without adequately funding the other. The statute's intent seems clear enough, but unfortunately as written it raises a host of questions, including: • Which employees are actually affected by the restriction? • What does "fund" mean in this context? • How does the law affect a sponsor's ability to pay NDCP benefits? • What are the ramifications for deferral -only plans? This article will attempt to disentangle these and other critical issues, while encouraging sponsors to develop an action plan before their DB plans' at -risk status places their NDCPs at risk of 409A noncompliance. (Throughout this article, "employer" means the N DCP sponsor and any other employers in the same control group.) When do restrictions apply? NDCP restrictions apply during the appropriately named "restricted period," which goes into effect: when the "employer" is a debtor in a federal or state bankruptcy proceeding; six months before or after the date that an underfunded DB plan of the employer is terminated; or during any period when an employer's DB plan is "at -risk," which generally means the plan has more than 500 participants and the assets of the plan represent less than a certain percentage of the value of the benefits under the plan (i.e., 65% for 2008, 70% for 2009, 75% for 2010, and 80% for 2011 and beyond). Are all NDCP participants affected by the funding restraints? The funding restraints only apply to an individual who is an "applica- ble covered employee" of the employer. This term includes not only presently "covered employees" of the employer but also captures any employees that were "covered employees" of the employer at the time of their termination of employment. The term "covered employee" has a two -pronged definition: The principal executive officer (or an individual acting in that capacity during the last completed fiscal year) or an employee whose total compensation is required to be reported to share- holders under the Securities Exchange Act of 1934 by reason of the employee being among the three highest compensated officers for the taxable year (not counting the principal executive officer), as described in tax code section 162(m)(3) and clarified by IRS Notice2007-49; or 2. An officer, director, or shareholder who owns 10% or more of a publicly traded company's equity (i.e., an individual subject to the requirements of section 16(a) of the Securities Exchange Act of 1934). The application of these definitions to NDCP sponsors that are publicly traded is clear cut; the extent, if any, of their applicability to private companies is the subject of debate. Some analysts have argued that private companies are completely exempt. Their rationale is that such companies generally would not be subject to the Securities Exchange Act of 1934, which would by definition exclude them from coverage under prong #2 of the above definition. In prong #1, they argue that the reference is to section 162(m)(3), which is part of section 162(m) and deals with the $1 million deduction limit for publicly traded companies. Accordingly, they ask, "How can this definition apply to private companies if it is from a code section governing publicly held entities?" Getting a legal opinion before relying on that interpretation is advisable. If prong #1 applies to private companies, it may only apply on a limited basis, given that they have no employees "whose total compensation is required to be reported to shareholders under the Securities Exchange Act of 1934 by reason of the employee being among the three highest compensated officers." Thus, in a private company, if anyone is affected, it would only be the principal executive officer. The IRS has not confirmed or denied a total exemption for private companies and no timetable for clarifying guidance has been announced, so the conservative approach for such companies (absent the above -referenced legal opinion) may be to treat the principal executive officer as an "applicable covered employee." What constitutes noncompliance? Once the NDCP sponsor has entered a "restricted period" and the "applicable covered employees" are identified, a final question remains that may be the most puzzling: "What funding produces noncompliance?" However, before examining possible solutions 4 :: WINTER 2009/2010 11 to this riddle, it is worth noting that NDCPs must consist solely of a "promise to pay" a future benefit and, thus, are technically "unfunded." Accordingly, amounts deferred by participants or future benefits provided by the sponsor cannot be formally segregated to guarantee the plan's future obligations. If the sponsor becomes insolvent, the deferred amounts are considered a part of the sponsor's assets and, therefore, subject to the claims of its creditors. This "unfunded" feature is a prerequisite for NDCPs to maintain their pre-tax and tax -deferred status. That being said, many NDCP sponsors do set aside dollars that provide liquidity for future nonqualified benefit obligations. However, this "informal funding" may only consist of placing assets in a separate corporate account or, if greater protection is desired, a rabbi trust, which prevents the plan sponsor from reassuming control over these assets. Whether held in trust or not, informally funded assets must remain general assets of the company and thus subject to the claims of creditors. The restrictions create two noncompliance traps that can be tripped by such informal funding: Trap #1: The Funding Prohibition During the "restricted period," no assets may be set aside or reserved (directly or indirectly) in a trust (or other arrangement as determined by the Treasury Secretary) or transferred to such a trust or other arrangement for purposes of paying deferred compensa- tion of an applicable covered employee. This applies equally to any NDCP plan of the employer. However, this trap does not apply to any assets that are "so set aside" before the restricted period. Accordingly, once the NDCP sponsor has entered the restricted period, Trap #1 prohibits the sponsor from any further prefunding of NDCP obligations —whether through contributions to an actual trust (e.g., rabbi trust) or using a separate corporate account established for this purpose. Because any amounts that were contributed before the beginning of the "restricted period" are not affected, a typical rabbi trust, funded before a restricted period begins, should not be affected by this trap. This is the case as long as no additional fund- ing is made during the restricted period, and the trust's provisions or operations do not set off Trap #2. Trap #2: The "Restricted" Prohibition The employer's NDCP may not provide that assets will become exclusively earmarked for the provision of NDCP benefits in connection with a DB plan restricted period (or other similar financial measure determined by the Treasury Secretary). Furthermore, plan sponsors may not take any action that results in assets being restricted in this manner. This trap is even more inclusive than Trap #1 in its potential to snare the unwary as it does not limit its noncompliance target only to those individuals captured by the "applicable covered employee" definition. It thus creates potential penalties for all NDCP partici- pants. In addition, Trap #2 does not contain an exemption for assets already transferred before the restricted period and therefore could also reach arrangements that were fully funded before the beginning of a restricted period. [ JH[E[Fff S PERSPECTIVES f]' Cu lsa"e Em ye -- While clarifying guidance is needed, one common interpretation of Trap #2 is that it is intended to prohibit such features as the "springing" funding of unfunded trusts (i.e., no assets or only a small amount of assets are deposited in the trust until a specified event or condition occurs). Another example would be a revocable rabbi trust that becomes irrevocable. In these cases, the change in status of the trust or other arrangement would be deemed to be "in connection with" a restricted period or some other financial event identified by the Treasury and IRS (e.g., the sponsor's financials declining to a certain level). NDCPs and trusts containing triggers that might not comply with these provisions should be reviewed promptly and amended as necessary, because the mere presence of such triggers (as well as their use) might be construed as running afoul of the statute. Practitioners and NDCP sponsors alike are struggling to ascertain the reach of these restrictions. While it is clear that sponsors cannot make deposits to a rabbi trust or other similar arrangement on behalf of applicable covered employees during a restricted period, the wording of the statute leaves much open to interpretation regard- ing what other sponsor actions are prohibited. Are the rules limited strictly to such "prefunding"? What about a situation in which the NDCP sponsor does not set aside any funds but merely pays ben- efits as they become due? Are these plans prohibited from paying benefits during a restricted period? Unfortunately, the exact mean- ing of the terms "set aside," "transfer," "other arrangement," and "in connection with" have not been defined at this point and may not be defined by IRS guidance for another year or two (at the very least). For example, does the "in connection with" in Trap #2 mean "as a result of or will the IRS apply it broadly so as to have it mean "during"? Under the former interpretation, it would seem that the payment of WINTER 2009/2010 :: 5