C.1. Capital Markets Research Group Asset-Liability Study Results. December 2016

Similar documents
2011 Asset/Liability Study Preliminary Results

Minnesota State Board of Investment. Legislative Commission on Pensions and Retirement

August Asset/Liability Study Texas Municipal Retirement System

State Universities Retirement System of Illinois (IL SURS)

2017 Capital Market Projections. Alaska Retirement Management Board. March 2, Paul Erlendson Senior Vice President

City of San José Federated City Employees Retirement System

Asset Allocation Study

Pension Glossary. 401(k) Plan A defined-contribution pension plan offered by many corporations.

Report on a Possible New Plan Design for the Shelby County Retirement System

Santa Barbara County Employees Retirement System. Actuarial Valuation as of June 30, Produced by Cheiron

Marin County Employees Retirement Association

Iowa Public Employees Retirement System Economic Assumptions Review

SELECTING A STRATEGIC ASSET ALLOCATION. San Diego County Employees Retirement Association. March 2014

Tulare County Employees Retirement Association

Ohio Police & Fire. Pension Fund. Investigation of Demographic and Economic Experience. Conduent Human Resource Services. Five-Year Period from

Topic Five: Case Study: Asset Allocation at the Texas Teacher Retirement System

ACTUARIAL VALUATION REPORT AS OF OCTOBER 1, City of Plantation General Employees Retirement System

Designing Outcome-Focused Defined Contribution Plans: Building Sustainable Income for Retirees

Subject: Actuarial Valuation Report for the Year Ending December 31, 2016

SOUTH BURLINGTON SCHOOL DISTRICT RETIREMENT INCOME PLAN. ACTUARIAL VALUATION as of October 1, 2015

Teachers Retirement System of the State of Illinois

Example Note Disclosure and RSI for Employer Participants of the Educational Retirement Board s Pension Plan (following the GASBS 68 Illustration #3)

TOWN OF LANTANA POLICE RELIEF AND PENSION FUND ACTUARIAL VALUATION REPORT AS OF OCTOBER 1, 2014

Florida Retirement System

ACTUARIAL VALUATION REPORT AS OF OCTOBER 1, City of Plantation Police Officers Retirement System

MOA Trust Fund Investment Flexibility June Michael J. O Leary CFA Executive Vice President Callan Associates Inc.

MISCELLANEOUS PLAN OF THE CITY OF MODESTO (CalPERS ID: ) Annual Valuation Report as of June 30, 2014

Employer Contribution Rate % % (projected)

Overview of Asset/Liability Process. City of Jacksonville Police & Fire Pension Fund

Teachers and State Employees Retirement System Principal Results of Actuarial Valuation as of December 31, 2016

TOWN OF LANTANA POLICE RELIEF AND PENSION FUND ACTUARIAL VALUATION REPORT AS OF OCTOBER 1, 2016

CITY OF HOLLYWOOD GENERAL EMPLOYEES RETIREMENT SYSTEM ACTUARIAL VALUATION REPORT AS OF OCTOBER 1, 2012

Pension Plan for Bargaining Unit Employees of TriMet

Conduent Human Resource Services Retirement Consulting. Public Employees Retirement System of New Jersey

Capital Market Assumptions

Investment Policy Statement

Actuarial Section ARLINGTON COUNTY EMPLOYEES RETIREMENT SYSTEM. Arlington County Employees Retirement System

Manager Comparison Report June 28, Report Created on: July 25, 2013

SAFETY PLAN OF THE CITY OF PASADENA (CalPERS ID: ) Annual Valuation Report as of June 30, 2014

MISCELLANEOUS PLAN OF THE CITY OF OAKLAND (CalPERS ID: ) Annual Valuation Report as of June 30, 2014

INVESTMENT POLICY STATEMENT CITY OF DOVER POLICE PENSION PLAN

ORLANDO UTILITIES COMMISSION PENSION PLAN ACTUARIAL VALUATION REPORT AS OF OCTOBER 1, 2016

ASSET ALLOCATION REPORT

Pension Simulation Project Rockefeller Institute of Government

PENSION SIMULATION PROJECT Investment Return Volatility and the Michigan State Employees Retirement System

Arizona PSPRS Pension Task Force Actuary 101

Teachers and State Employees Retirement System Principal Results of Actuarial Valuation as of December 31, 2017

UCRP and GEP Quarterly Investment Risk Report

Actuarial Valuation Report for the Employees Retirement System of the City of Baltimore

Wyoming Retirement System Actuarial Experience Study As of December 31, 2016

MISCELLANEOUS PLAN OF THE CITY OF OCEANSIDE (CalPERS ID: ) Annual Valuation Report as of June 30, 2015

BOYNTON BEACH POLICE PENSION FUND INVESTMENT PERFORMANCE PERIOD ENDING MARCH 31, 2011

City of Boynton Beach Municipal Police Officers Retirement Fund Actuarial Valuation Report as of October 1, 2018

STATE OF IOWA PEACE OFFICERS RETIREMENT, ACCIDENT AND DISABILITY SYSTEM. Five Year Experience Study For Period Ending June 30, 2016.

CITY OF CLEARWATER EMPLOYEES PENSION PLAN ACTUARIAL VALUATION REPORT AS OF JANUARY 1, 2016

Employees' Retirement Fund of the City of Fort Worth Revised Actuarial Valuation and Review as of January 1, 2014

P O L I C E M E N S A N N U I T Y A N D B E N E F I T F U N D O F C H I C A G O A C T U A R I A L V A L U A T I O N R E P O R T F O R T H E Y E A R E

NEW CASTLE COUNTY, DELAWARE EMPLOYEES PENSION PROGRAM

Quarterly Investment Review

July 31, The Board of Trustees City of Pontiac General Employees Retirement System Pontiac, Michigan

UNIVERSITY OF CALIFORNIA RETIREMENT PLAN ASSET AND RISK ALLOCATION POLICY

Preliminary Results of 2014 Actuarial Experience Study

Teachers Retirement Association of Minnesota. Review of Economic Assumptions

Missouri Department of Transportation and Highway Patrol Employees Retirement System (MPERS) Actuarial Valuation Report June 30, 2018

San Joaquin County Employees Retirement Association

Local Governmental Employees Retirement System Principal Results of Actuarial Valuation as of December 31, 2017

City of Fort Pierce Retirement and Benefit System Sixtieth Annual Actuarial Valuation Report for the Year Ending September 30, 2018

TriMet Defined Benefit Retirement Plan for Management and Staff Employees

State Teachers Retirement System of Ohio

Missouri Department of Transportation and Highway Patrol Employees Retirement System (MPERS) Actuarial Valuation Report June 30, 2017

San Joaquin County Employees Retirement Association

The Submission of. William M. Mercer Limited. The Royal Commission on Workers Compensation in British Columbia. Part B: Asset/Liability Study

Callan Associates Inc. Investment Measurement Service Quarterly Review City of Milwaukee Employees Retirement System March 31, 2011

MISCELLANEOUS PLAN OF THE METROPOLITAN WATER DISTRICT OF SOUTHERN CALIFORNIA (CalPERS ID: ) Annual Valuation Report as of June 30, 2013

DRAFT CAMPTON HILLS POLICE PENSION FUND. Actuarial Valuation as of May 1, 2017 LAUTERBACH & AMEN, LLP. GASB 67/68 Reporting

Setting Callan s Capital Market Projections

O A K L A N D C O U N T Y E M P L O Y E E S ' R E T I R E M E N T S Y S T E M

Employees Retirement System of the City of Baltimore

TEACHERS RETIREMENT BOARD. REGULAR MEETING Item Number: 7 CONSENT: ATTACHMENT(S): 1. DATE OF MEETING: November 8, 2018 / 60 mins

SAFETY POLICE PLAN OF THE CITY OF ANAHEIM (CalPERS ID: ) Annual Valuation Report as of June 30, 2015

S TAT E U NIVERSITIES R E T I REMENT SYSTEM OF I L L INOIS

CP#32-08 Investment Policy

October 7, The Board of Trustees City of Pontiac General Employees Retirement System Pontiac, Michigan

Volatility-Managed Strategies

MISCELLANEOUS PLAN OF THE COUNTY OF RIVERSIDE (CalPERS ID: ) Annual Valuation Report as of June 30, 2013

North Carolina Local Governmental Employees Retirement System Report on the Actuarial Valuation Prepared as of December 31, 2013

State Universities Retirement System of Illinois. Actuarial Valuation Report as of June 30, 2018

Introduction 1-2. Summary of Results and Comments 3-15

Investment Cost Effectiveness Analysis Norwegian Government Pension Fund Global

Florida Retirement System

North Carolina Local Governmental Employees Retirement System. Report on the Actuarial Valuation Prepared as of December 31, 2014

Target Funds. SEMIANNual REPORT

Amended as of January 1, 2018

VIRGINIA POLYTECHNIC INSTITUTE AND STATE UNIVERSITY OPTIONAL RETIREMENT AND CASH MATCH PLANS INVESTMENT POLICY STATEMENT

CITY OF GENEVA, ILLINOIS FIREFIGHTERS PENSION FUND ANNUAL FINANCIAL REPORT. For the Year Ended April 30, 2016

Teachers Retirement Association of Minnesota A Pension Trust Fund of the State of Minnesota. Actuarial

CITY OF TALLAHASSEE PENSION PLANS ACTUARIAL VALUATION REPORT AS OF OCTOBER 1, 2016

City of Albany Police and Fire Relief or Pension Fund

C I T Y O F S O U T H F I E L D E M P L O Y E E S R E T I R E M E N T S Y S T E M G A S B S T A T E M E N T N O S. 6 7 A N D 6 8 A C C O U N T I N G

Projected Results % $3,882,000 TBD % $4,538,000 TBD

Transcription:

December 2016 2016 Asset-Liability Study Results Capital Markets Research Group

Scope of the Project Asset/Liability Study Phase 1 Review MCERA s current investment program. Strategic allocation to broad asset classes. Important to distinguish between strategy (i.e. the target asset class/benchmark) and implementation (i.e. the way the manager constructs the portfolio). Set asset class, portfolio expectations. Return, risk, correlation, and other considerations. Evaluate potential new asset classes/strategies. Phase 2 Build integrated asset-liability model: Reflect 6/30/2015 valuation results; confirm model assumptions, review with actuary. Roll valuation results forward to 6/30/16 to begin projections. Deterministic projections assume valuation assumptions are achieved. Simulation apply Callan s capital market projections, insert capital market uncertainty, evaluate alternative investment strategies. Phase 3 Develop preliminary asset-liability results. Confirm decision variables; ascertain risk tolerance and effective investment time horizon. Callan internal peer review of the study s results. Ongoing review and interaction with staff. Develop the final asset-liability study. Present finalized asset-liability results to MCERA Board of Trustees. MCERA Board selects an appropriate asset allocation. 1

Timeline August/September 2016 Construct liability model in ProVal, starting with 2015 valuation results. October 2016 Callan presentation: Overview of study process, review of current program, set capital market expectations, evaluate potential new strategies (Phase 1). Complete liability model, integrate asset mixes and develop projections and simulations (Phase 2). December 2016 Callan presentation: Deliver refined asset-liability study results. Complete study, adoption by Board. 2

Process Overview

Why Conduct an Asset and Liability Study? The cornerstone of a prudent process for pension plan, endowment, and foundation trustees (and any individual investor) is a careful and thorough examination of their long-term strategic plan. Explicitly acknowledge change and uncertainty in the capital markets. Establish reasonable rate-of-return and risk expectations. Incorporate material changes in strategic plan policies and demographics Funding policy, benefit formula, eligibility, early retirement, COLA, decrement tables Reflect changes in regulations Public pension: GASB 67 and 68 Project and evaluate impact on assets, liabilities and funded status. Confirm an investment policy to meet return and risk objectives in relation to funding, accounting and policy goals. If no material changes have occurred, an asset allocation review should still be conducted every 3 5 years. 4

Where Does Asset Allocation Fit in Strategic Planning? Evaluating the interaction of the three key policies that govern a defined benefit plan with the goal of establishing the best investment policy Investment Policy How will the assets supporting the benefits be invested? What risk/return objectives? How to manage cash flows? Investment Policy Funding Policy Funding Policy How will the benefits be funded? What assumed investment return? How are deficits amortized? What actuarial methodologies are applied to dampen contribution volatility? Benefits Policy What type/kind of benefits? What level of benefit? When and to whom are they payable? Benefits Policy 5

Defining MCERA s Risk Tolerance Factors Critical to Decision-Making Size of the Plan Current funded status Expected funding requirements Plan status (open to new participants; existing members still accrue benefits) Time horizon Liquidity needs: Benefit payment less contributions Funding policy can impact liquidity needs Liability growth rates Willingness to take risk: Sensitivity to size of contribution or contribution volatility Financial ability to take risk 6

Liability Model

Asset/Liability Study Process Liability Model and Projected Cash Flows Pension Plan Equation: Benefits + Expenses = Investment Return + Contributions Callan builds the liability model Uses data from plan actuary (Cheiron) Liability Assumptions Funding Policy Employee contributions Employer contributions Benefit Policy Benefit formulas Cost of living increases Demographics Ratio of Active vs Retirees Average age Population growth Salary increases Mortality table longevity risk management Discount rate 8

Liability Model and Key Actuarial Assumptions Variable As of 6/30/2015 Total Actuarial Liability Value $2,469.1mm Key Actuarial Assumptions Investment Return 7.25% Price Inflation 2.75% Description Market Value of Assets $2,066.2mm Unfunded Actuarial Liability $402.8mm Market Funded Status 83.7% Employer Contribution for FYE 2016 32.63% Employer Contribution for FYE 2017 32.22% Salary Scale COLA 3.0%, plus longevity & promotion 2%-4% caps, vary by plan and tier Asset-liability projections are based on the 6/30/2015 actuarial report for the MCERA Plan. 2015-16 investment experience is reflected in projections. Total plan return July 2015 June 2016 = 2.26%; CPI = 1.0% Employer contributions shown above are blended rates incorporating multiple plan groups and tiers, and reflect the employers share of normal cost plus substantial contributions to pay down the unfunded actuarial liability. Employee contributions are in addition to the rates shown above, and vary by plan group. 9

Plan Membership Active & Inactive Member Count, Annual Payroll 10,000 9,000 8,000 7,000 6,000 5,000 4,000 3,000 2,000 1,000 $450 $400 $350 $300 $250 $200 $150 $100 $50 0 $0 2016 2017 2018 2019 2020 2021 2022 Plan Membership Projected Payroll (Millions) 2023 2024 2025 2026 2027 2028 2029 2030 2031 2032 2033 2034 2035 2036 Active Members Inactive Members Annual Payroll (Right Scale) Active members are held constant. Future new hires replace exits due to retirement, death, disability, and withdrawal. Active membership is constant (implies 0% workforce growth). Payroll adjusted to be consistent with Callan s capital market assumptions. 10

Actuarial Liability $4,500 $4,000 $3,500 $3,000 $2,500 $2,000 $1,500 $1,000 $500 50% 40% 30% 20% 10% $0 0% 2016 2017 2018 2019 2020 2021 Actuarial Liability (Millions) Active Liability Percentage 2022 2023 2024 2025 2026 2027 2028 2029 2030 2031 2032 2033 2034 2035 2036 Active Liability Inactive Liability Active Liability / Total Liability (Right Scale) Total plan liability grows by 3% annually over the next 10 years. Inactive liabilities grow faster than active liabilities (3.6% vs. 1.6%). Active liability falls from 33% of total liability to 28% by 2026, and to 26% by 2036. 11

Actuarial Liability - Simulation $3,800 Actuarial Liability $3,600 $3,400 $3,200 $3,000 $2,800 $2,600 $2,400 5-Year 10-Year Percentile 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 Growth Growth 97.5th $2,552 $2,720 $2,838 $2,952 $3,069 $3,182 $3,303 $3,413 $3,529 $3,643 $3,759 4.5% 4.0% 75th 2,552 2,664 2,761 2,858 2,956 3,055 3,151 3,242 3,340 3,436 3,529 3.7% 3.3% 50th 2,552 2,630 2,711 2,795 2,879 2,963 3,044 3,128 3,212 3,295 3,380 3.0% 2.9% 25th 2,552 2,586 2,651 2,721 2,796 2,865 2,937 3,007 3,086 3,155 3,223 2.3% 2.4% 2.5th 2,552 2,496 2,531 2,568 2,625 2,670 2,727 2,776 2,821 2,874 2,939 0.9% 1.4% Range 0 224 307 384 444 513 576 636 708 769 820 3.6% 2.5% The actuarial liability increases 2.5% per year over the 10 year forecast horizon. The Plan s liabilities are sensitive to changes in inflation and the resulting impact on salaries. Based on Callan s 10-year capital market expectations, the expected liability return is 6.8%. The liabilities are growing at a rate slower than the full interest cost of 7.25% since Callan s inflation expectation of 2.25% is lower than the actuary s assumed inflation of 2.75% 12

Contributions Assuming 6.9% Return % of Expected Payroll 35.0% 30.0% 25.0% 20.0% 15.0% 10.0% 5.0% 0.0% 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 2031 2032 2033 2034 2035 2036 Charts assume plan earns a 6.87% return, consistent with Callan s capital market projections (Current Target mix). Normal cost as % of payroll rises gradually from 12% to 12.4% by 2026, remains constant for the following ten years. Normal cost in dollars increases 2.9% per year over next ten years, as payroll increases. $100.0 $90.0 $80.0 $70.0 $60.0 $50.0 $40.0 $30.0 $20.0 $10.0 Employer Normal Cost Rate Amortization Cost Closed-period amortization of Unfunded Liability at 6/30/2013 ends in 2030, resulting in sharp drop in amortization cost. Remaining cost is for UAL from 2009 (22 years remaining). Contribution projection is very similar to that calculated using 7.25% return and 2.75% inflation assumed in the actuarial valuation. $0.0 2016 2017 2018 2019 2020 2021 2022 2023 2024 $ Millions 2025 2026 2027 2028 2029 2030 2031 2032 2033 2034 2035 2036 Normal Cost ($) Amortization Cost 13

Funded Status AAL and MVA (Millions) $4,500 $4,000 $3,500 $3,000 $2,500 $2,000 $1,500 $1,000 $500 120% 100% 80% 60% 40% 20% Funded Status Top chart assumes plan earns a 6.9% return, consistent with Callan s capital market assumptions. Funded status returns to 100% over 20 year projection period. $0 0% 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 2031 2032 2033 2034 2035 2036 Actuarial Accrued Liability (AAL) Market Value of Assets (MVA) Funded Status (Right Scale) 120.0% Funded Status 100.0% 80.0% 60.0% 40.0% 20.0% Actuarial Callan Bottom chart compares funded status using Callan s capital market assumptions versus actuarial assumptions; projections are virtually identical. 0.0% 14

Liquidity Needs Chart assumes plan earns a 6.9% return. Benefits & Contributions (Millions) $300 $250 $200 $150 $100 $50 $0 2016 2018 2020 2022 2024 2026 2028 2030 2032 2034 2036 Benefit Payments Contributions Net Outflow / Assets (Right Scale) 10.0% 9.0% 8.0% 7.0% 6.0% 5.0% 4.0% 3.0% 2.0% 1.0% 0.0% Net Outflow / Assets Net Outflow Flow = Funding Contributions Benefit Payments Expenses Liquidity needs help define the appropriate time horizon for investments and shape the ability to commit to volatile and/or illiquid asset classes. Net cash flow as a % of plan assets is a useful indicator of liquidity needs. A ratio below 5% is viewed as manageable and should not impact asset allocation. A ratio between 5% and 10% bears careful watch and may necessitate strategies to manage cash flow needs, and could impact asset allocation, especially exposure to illiquid assets. Net cash flow is negative and rises from 2.8% of plan assets to 3.6% in 10 years (2026). Liquidity needs in this range are manageable under the current investment policy for the next 10 years. We project net outflow to rise to 5.2% of assets in 20 years. 15

Asset Modeling

2016 Capital Market Expectations Return and Risk Summary of Callan s Long-Term Capital Market Projections (2016 2025) Asset Class Equities Index PROJECTED RETURN 1-Year 10-Year Arithmetic Geometric* Real PROJECTED RISK Standard Deviation Sharpe Ratio Projected Yield 10-Year Geometric* 2015-2024 Standard Deviation Geometric* Delta Broad Domestic Equity Russell 3000 8.85% 7.35% 5.10% 18.70% 0.353 2.40% 7.60% 19.00% -0.25% Large Cap S&P 500 8.60% 7.25% 5.00% 17.95% 0.354 2.50% 7.50% 18.30% -0.25% Small/Mid Cap Russell 2500 9.85% 7.55% 5.30% 22.75% 0.334 1.90% 7.85% 22.95% -0.30% Global ex-u.s. Equity MSCI ACWI ex USA 9.55% 7.55% 5.30% 21.30% 0.343 2.70% 7.80% 21.45% -0.25% International Equity MSCI World ex USA 9.00% 7.25% 5.00% 20.05% 0.337 3.00% 7.50% 20.20% -0.25% Emerging Markets Equity MSCI Emerging Markets 11.15% 7.60% 5.35% 27.85% 0.320 1.70% 7.90% 27.95% -0.30% Fixed Income Short Duration Barclays G/C 1-3 2.60% 2.60% 0.35% 2.25% 0.156 2.80% 2.40% 2.25% 0.20% Domestic Fixed Barclays Aggregate 3.05% 3.00% 0.75% 3.75% 0.213 4.60% 3.00% 3.75% 0.00% Long Duration Barclays Long G/C 4.30% 3.70% 1.65% 11.40% 0.197 5.30% 3.20% 11.40% 0.50% TIPS Barclays TIPS 3.10% 3.00% 0.75% 5.30% 0.160 4.20% 3.00% 5.30% 0.00% High Yield Barclays High Yield 5.40% 5.00% 2.75% 10.50% 0.300 8.00% 5.00% 11.10% 0.00% Non-U.S. Fixed Barclays Global Aggregate ex US 1.80% 1.40% -0.85% 9.20% -0.049 4.00% 2.30% 9.40% -0.90% Emerging Market Debt EMBI Global Diversified 5.00% 4.60% 2.35% 9.90% 0.278 6.50% 4.70% 10.00% -0.10% Other Real Estate Callan Real Estate 7.20% 6.00% 3.75% 16.45% 0.301 5.00% 6.15% 16.50% -0.15% Private Equity TR Post Venture Cap 13.15% 8.15% 5.90% 32.80% 0.322 0.00% 8.50% 33.05% -0.35% Hedge Funds Callan Hedge FOF Database 5.55% 5.25% 3.00% 9.30% 0.355 2.25% 5.25% 9.30% 0.00% Commodities Bloomberg Commodity 4.40% 2.75% 0.50% 18.50% 0.116 2.25% 2.75% 18.50% 0.00% Cash Equivalents 90-Day T-Bill 2.25% 2.25% 0.00% 0.90% 0.000 2.25% 2.25% 0.90% 0.00% Inflation CPI-U 2.25% 1.50% 2.25% 1.50% 0.00% * Geometric returns are derived from arithmetic returns and the associated risk (standard deviation). Source: Callan Associates 17

2016 Capital Market Expectations Correlation Coefficient Matrix Key to Constructing Efficient Portfolios Broad US Equity 1.000 Large Cap 0.997 1.000 Small/Mid Cap 0.965 0.940 1.000 Global ex-us Equity 0.882 0.879 0.853 1.000 Non-US Equity 0.852 0.850 0.820 0.986 1.000 Em Mkts Equity 0.861 0.855 0.840 0.933 0.860 1.000 Defensive -0.240-0.230-0.260-0.254-0.230-0.280 1.000 US Fixed -0.108-0.100-0.130-0.123-0.105-0.150 0.870 1.000 Long Duration 0.136 0.138 0.121 0.106 0.119 0.069 0.729 0.925 1.000 TIPS -0.050-0.045-0.065-0.053-0.045-0.065 0.530 0.580 0.527 1.000 High Yield 0.640 0.640 0.610 0.629 0.610 0.610-0.170 0.020 0.220 0.060 1.000 Non-US Fixed 0.014 0.050-0.100 0.013 0.060-0.090 0.480 0.510 0.542 0.340 0.120 1.000 Em Mkt Debt 0.579 0.580 0.550 0.550 0.530 0.540-0.120 0.030 0.159 0.150 0.600 0.010 1.000 Real Estate 0.735 0.730 0.715 0.669 0.650 0.645-0.140-0.020 0.190 0.005 0.560-0.050 0.450 1.000 Private Equity 0.948 0.945 0.915 0.934 0.905 0.905-0.240-0.190 0.062-0.100 0.640-0.060 0.560 0.710 1.000 Hedge Funds 0.797 0.795 0.765 0.760 0.735 0.740-0.120 0.080 0.303 0.055 0.570-0.080 0.540 0.600 0.770 1.000 Commodities 0.167 0.165 0.165 0.177 0.170 0.175-0.220-0.120-0.042 0.100 0.100 0.050 0.190 0.200 0.180 0.210 1.000 Cash Equivalents -0.043-0.030-0.080-0.040-0.010-0.100 0.300 0.100-0.049 0.070-0.110-0.090-0.070-0.060 0.000-0.070 0.070 1.000 Inflation -0.011-0.020 0.020 0.010 0.000 0.030-0.200-0.280-0.284 0.180 0.070-0.150 0.000 0.100 0.000 0.200 0.400 0.000 1.000 Broad US Eq Large Cap Sm/Mid Cap Global ex-us Non-US Equity Em Mkt Eq Defens US Fixed Long Duration TIPS High Yield Non-US Fixed Em Mkt Debt Real Estate Private Equity Hedge Funds Comm Cash Equiv Inflation Relationships between asset classes is as important as standard deviation. To determine portfolio mixes, Callan employs mean-variance optimization. Return, standard deviation and correlation determine the composition of efficient asset mixes. Source: Callan Associates 18

MCERA Asset Classes - Return and Risk Asset Class 10-Year Compound Return Projected Standard Deviation Broad Domestic Equity 7.35% 18.70% Global ex-us Equity 7.55% 21.30% Domestic Fixed Income 3.00% 3.75% Private Equity 8.15% 32.80% Real Assets 6.10% 13.75% Cash Equivalents 2.25% 0.90% Total Real Assets portfolio: 8% private real estate, 7% public real assets Public real assets = 25% TIPS, 25% Commodities, 25% REITs, 25% Natural Resource Equity. 19

MCERA Asset Classes - Correlation Broad Dom Equity Broad Domestic Equity 1.00 GlobalxUS Equity Global Ex-US Equity 0.88 1.00 Domestic Fixed Domestic Fixed -0.11-0.12 1.00 Real Assets Real Assets 0.80 0.76-0.05 1.00 Private Equity Private Equity 0.95 0.93-0.19 0.78 1.00 Cash Equivalents Cash Equivalents -0.04-0.04 0.10-0.02 0.00 1.00 Inflation -0.01 0.01-0.28 0.23 0.00 0.00 1.00 Inflation Total Real Assets portfolio: 8% private real estate, 7% public real assets Public real assets = 25% TIPS, 25% Commodities, 25% REITs, 25% Natural Resource Equity 20

MCERA - 2016 Efficient Mixes Maximum Real Asset Allocation of 15% Alternative Asset Mixes - 15% Real Assets Maximum Final Min Max Asset Class 6/30/2016 Target Alloc Alloc Mix 1 Mix 2 Mix 3 Mix 4 Mix 5 Broad US Equity 31.4% 32% 0% 100% 18% 22% 27% 32% 36% Broad International Equity 19.8% 22% 0% 100% 14% 17% 20% 23% 27% Broad US Fixed Income 22.8% 23% 0% 100% 50% 40% 31% 22% 12% Real Assets 16.9% 15% 0% 15% 13% 15% 15% 15% 15% Private Equity 9.1% 8% 0% 100% 5% 6% 8% 9% 10% Totals 100% 100% 100% 100% 100% 100% 100% Expected Return 6.88% 6.87% 5.72% 6.17% 6.58% 6.95% 7.28% Real Return 4.63% 4.62% 3.47% 3.92% 4.33% 4.70% 5.03% Risk (Standard Deviation) 14.57% 14.55% 9.23% 11.10% 13.02% 14.95% 16.91% % equity 60% 62% 37% 45% 54% 64% 73% % fixed income 23% 23% 50% 40% 31% 22% 12% % real assets 17% 15% 13% 15% 15% 15% 15% Mixes are constrained to hold a maximum of 15% real assets. No new asset classes included. The current target mix is efficient and lies on the efficient frontier. Real assets expands the real estate allocation category to include other real assets, all publicly traded: TIPS, commodities, natural resource equity and REITs. Real estate remains the core, with added diversification. 21

MCERA - 2016 Efficient Frontier Nominal Return Maximum Real Asset Allocation of 15% Plan s assumed investment return = 7.25% MCERA s asset allocation target is an optimal allocation, since it lies on the efficient frontier depicting risk and return. Current target is a well-diversified portfolio that includes fixed income, public equity, private equity and real assets, including private real estate. 22

MCERA - 2016 Efficient Frontier Real Return Maximum Real Asset Allocation of 15% Plan s assumed real investment return = 4.25% MCERA s long term nominal return assumption of 7.25%, inflation assumption of 2.75% and real wage growth of 0.25% suggest a long term real return target of 4.25%. Callan s 10-year return expectation for the target asset allocation is 6.87%, and combined with our inflation assumption of 2.25%, yields a real return expectation of 4.62%, higher than that assumed in the actuarial valuation. 23

Projected Rates of Return One Year Maximum Real Asset Allocation of 15% 50% Range of Projected Rates of Return Projection Period: 1 Year Optimization Set: 2016 Real Assets const 15 Annual Rates of Return (%) 40% 30% 20% 10% 0% (10%) (20%) 49 49 44 47 49 50 50 7.25% (30%) 6/30/16 Final Target Mix 1 Mix 2 Mix 3 Mix 4 Mix 5 5th Percentile 25th Percentile Median 75th Percentile 95th Percentile 35.0% 18.3% 6.9% (2.6%) (16.8%) 35.1% 18.1% 6.8% (2.5%) (16.9%) 22.6% 12.6% 5.2% (0.6%) (9.2%) 27.0% 14.6% 5.8% (1.3%) (12.0%) 31.6% 16.8% 6.4% (2.0%) (14.8%) 36.0% 18.7% 6.9% (2.9%) (17.6%) 40.8% 20.6% 7.6% (3.8%) (20.0%) Prob > 7.25% 49.3% 49.5% 44.4% 46.9% 48.8% 49.6% 50.4% 24

Projected Rates of Return Five Years Maximum Real Asset Allocation of 15% 50% Range of Projected Rates of Return Projection Period: 5 Years Optimization Set: 2016 Real Assets const 15 Annual Rates of Return (%) 40% 30% 20% 10% 0% (10%) (20%) 47 47 37 44 46 48 50 7.25% (30%) 6/30/16 Final Target Mix 1 Mix 2 Mix 3 Mix 4 Mix 5 5th Percentile 25th Percentile Median 75th Percentile 95th Percentile 18.7% 11.3% 6.6% 2.0% (3.3%) 18.6% 11.4% 6.6% 2.0% (3.2%) 12.7% 8.7% 5.4% 2.7% (0.9%) 14.7% 9.7% 5.9% 2.5% (1.7%) 17.0% 10.7% 6.3% 2.2% (2.6%) 19.1% 11.6% 6.8% 2.0% (3.4%) 21.2% 12.5% 7.2% 1.7% (4.4%) Prob > 7.25% 47.3% 47.3% 37.3% 43.7% 46.4% 47.7% 49.7% 25

Projected Rates of Return Ten Years Maximum Real Asset Allocation of 15% 50% Range of Projected Rates of Return Projection Period: 10 Years Optimization Set: 2016 Real Assets const 15 Annual Rates of Return (%) 40% 30% 20% 10% 0% (10%) (20%) 48 48 31 40 44 49 52 7.25% (30%) 6/30/16 Final Target Mix 1 Mix 2 Mix 3 Mix 4 Mix 5 5th Percentile 25th Percentile Median 75th Percentile 95th Percentile 14.8% 10.0% 7.1% 3.8% (1.0%) 14.8% 10.0% 7.0% 3.8% (1.0%) 10.6% 7.7% 5.7% 3.8% 0.6% 12.2% 8.5% 6.2% 3.9% 0.1% 13.7% 9.4% 6.7% 3.9% (0.5%) 15.1% 10.2% 7.1% 3.8% (1.1%) 16.5% 11.0% 7.4% 3.7% (1.8%) Prob > 7.25% 47.6% 47.8% 31.0% 40.2% 43.6% 48.5% 51.9% 26

2016 Capital Market Expectations Nominal vs Real Return Expectations Reduced Across All Asset Classes The expected return for the MCERA Policy Target Mix is 6.87%, below the 7.25% return assumed in the actuarial valuation. However, the Plan still had a reasonable chance of achieving this result over 10 years (almost 50% probability). In addition, the real return embedded in the valuation (7.25% - 2.75% inflation 0.25% wage growth = 4.25%) is actually lower than Callan s expected real return (6.87% - 2.25% inflation = 4.62%). While return expectations are lower for the next five- to ten-year horizon, MCERA will need to retain a strong orientation toward risk assets (equity) in pursuit of return to achieve its funding goals. Whether the plan should pursue more or less exposure to risk assets than the current policy target mix should not be unduly influenced by subdued expectations for the shorter-term 5-10 year horizon. We do not believe investors are likely to be compensated for greater risk taking in the shorter term. 27

One-Year Drawdown Analysis Current Asset Classes 0.0% 0.0% One-Year Worse-Case Drawdown -10.0% -20.0% -30.0% -40.0% -50.0% -25.9% -30.9% -35.9% -38.7% -39.9% -44.0% One-Year Impact to Funded Status -10.0% -20.0% -30.0% -40.0% -50.0% -30.6% -35.3% -39.9% -42.6% -43.7% -47.5% -60.0% Mix 1 Mix 2 Mix 3 Target Mix Mix 4 Mix 5-60.0% Mix 1 Mix 2 Mix 3 Target Mix Mix 4 Mix 5 The graph on the left depicts the worse-case (97.5 th percentile) drawdown based on simulated annual returns over the next 10 years. The graph on the right depicts the impact to funded status given a one-year worse-case drawdown and an expected liability return of 6.8%. Benefits and contributions are not reflected. E.g. Mix 1 Impact to Funded Status = (1-.259) / 1.068 1 = -30.6% 28

Two-Year Drawdown Analysis Current Asset Classes 0.0% 0.0% Two-Year Worse-Case Drawdown -10.0% -20.0% -30.0% -40.0% -50.0% -60.0% -31.8% -37.7% -43.6% -46.9% -48.3% Mix 1 Mix 2 Mix 3 Target Mix -53.1% Mix 4 Mix 5 Two-Year Impact to Funded Status -10.0% -20.0% -30.0% -40.0% -50.0% -60.0% -70.0% -40.2% -45.4% Mix 1 Mix 2 Mix 3 Target Mix -50.6% -53.4% -54.7% -58.8% Mix 4 Mix 5 The graph on the left depicts the worse-case (97.5 th percentile) drawdown based on two years of simulated returns. Drawdown is the cumulative return from peak to trough. Thus, the above chart reflects two years of consecutive negative returns. The graph on the right depicts the impact to funded status given a two-year worse-case drawdown and an expected liability return of 6.8%. Benefits and contributions are not reflected. E.g. Mix 1 Impact to Funded Status = (1-.318) / 1.068^2 1 = -40.2% 29

Three-Year Drawdown Analysis Current Asset Classes 0.0% 0.0% Three-Year Worse-Case Drawdown -10.0% -20.0% -30.0% -40.0% -50.0% -60.0% -70.0% -43.1% -47.2% -51.3% -53.3% -55.0% Mix 1 Mix 2 Mix 3 Target Mix -58.7% Mix 4 Mix 5 Three-Year Impact to Funded Status -10.0% -20.0% -30.0% -40.0% -50.0% -60.0% -70.0% -53.3% -56.6% -60.0% -61.6% -63.1% Mix 1 Mix 2 Mix 3 Target Mix -66.1% Mix 4 Mix 5 The graph on the left depicts the worse-case (97.5 th percentile) drawdown based on three years of simulated returns. Drawdown is the cumulative return from peak to trough. Thus, the above chart reflects three years of consecutive negative returns. The graph on the right depicts the impact to funded status given a three-year worse-case drawdown and an expected liability return of 6.8%. Benefits and contributions are not reflected. E.g. Mix 1 Impact to Funded Status = (1-.431) / 1.068^3 1 = -53.3% 30

Time Horizon for Capital Market Expectations and Asset-Liability Analysis Open, active pension plans have very long term liabilities, and necessarily should maintain a long term perspective for investment strategy. Callan s asset-liability analysis typically focuses on a planning cycle of 5-10 years, incorporating current market conditions and the path from these short term conditions to long term expectations. Over much of Callan s history, the difference between our shorter-term expectations and our long term numbers was modest; for most planning purposes our short term and long term expectations were the same. Current conditions, particularly in the fixed income markets, suggest substantial difference in capital market expectations depending on time horizon, and the path from the current conditions to the long term expectations. Theme of the current Callan 10-year projections: The path to a rational set of long-term capital market outcomes is likely through an ugly shorter term period of rising interest rates, capital losses in fixed income, and volatile equity markets. 31

Integration of Assets and Liabilities

Simulate Financial Condition Liability Modeling Build Liability Model Asset Projections Define Capital Market Projections Deterministic Projections Create Asset Mix Alternatives Simulate Financial Condition Define Risk Tolerance Select Appropriate Target Mix 33

After the Modeling How to Make a Decision? Potential decision variables include: The range of actuarial liability Present value of future contributions Range of the market (or actuarial) value of Plan assets Funded Ratio Liquidity and cash flow needs Present value of future unfunded liability Ultimate Net Cost Ultimate net cost combines contributions paid in over the planning horizon plus the value of the unfunded liability at the end of the projection period. A discussion of goals and objectives for MCERA s financial future will inform all three major policies: benefits, funding and investments. 34

Market Value of Assets for Current Target Mix $7,000 Market Value of Assets ($mm) $6,000 $5,000 $4,000 $3,000 $2,000 $1,000 $0 50 th Percentile 97.5 th Percentile Percentile 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2.5th $2,079 $2,749 $3,163 $3,459 $3,894 $4,269 $4,668 $4,973 $5,353 $5,946 $6,314 25th 2,079 2,370 2,571 2,766 2,925 3,074 3,246 3,405 3,595 3,765 3,922 50th 2,079 2,188 2,285 2,372 2,485 2,581 2,665 2,777 2,895 3,007 3,148 75th 2,079 1,991 1,998 2,020 2,064 2,130 2,182 2,253 2,346 2,424 2,471 97.5th 2,079 1,581 1,486 1,458 1,412 1,412 1,432 1,491 1,502 1,529 1,574 Range 0 1,167 1,677 2,001 2,482 2,857 3,236 3,482 3,850 4,417 4,739 The expected outcome is the 50 th percentile, a 50/50 chance of occurrence. The worse case scenario is the 97.5 th percentile; a 1 in 40 chance of occurrence. For example, there is a 1 in 40 chance (2.5% probability) that the 6/30/2026 market value of assets will be $1,574 mm or less. 35

6/30/2026 Market Funded Status by Policy Mix 290% 1/1/2026 Funded Status (MVA / AL) 240% 190% 140% 90% 40% Percentile Target Mix 1 Mix 2 Mix 3 Mix 4 Mix 5 2.5th 192% 127% 145% 172% 199% 233% 25th 116% 95% 102% 110% 119% 128% 50th 93% 84% 87% 90% 94% 98% 75th 73% 72% 72% 72% 73% 74% 97.5th 48% 54% 52% 49% 47% 45% Expected Return 6.9% 5.7% 6.2% 6.6% 7.0% 7.3% Standard Deviation 14.6% 9.2% 11.1% 13.0% 15.0% 16.9% More aggressive mixes are expected to have a higher funded status at the end of 10 years but will have a lower funded status in a worse-case scenario (97.5 th percentile). All mixes except Mix 2 are expected to improve the plan s funded status at 6/30/15 of 83.7%. 36

Cumulative Contributions ($) Ten Years $1,800 10-Year Cumulative Contributions $1,600 $1,400 $1,200 $1,000 $800 $600 $400 Percentile Target Mix 1 Mix 2 Mix 3 Mix 4 Mix 5 97.5th $1,527 $1,384 $1,432 $1,497 $1,542 $1,590 75th 1,087 1,077 1,081 1,090 1,088 1,089 50th 834 937 900 859 826 791 25th 587 754 685 615 581 553 2.5th 455 510 481 465 452 444 97.5th-50th 693 447 531 638 716 799 The graph illustrates the reward-risk trade-off of the alternative mixes in contribution space. The median shows expected contributions and the reward for taking more risk. The 97.5 th percentile shows the worse-case contribution, the cost of taking on risk. In a worse-case scenario, contributions are higher for a more aggressive asset mix. Contributions in the best-case result are held at normal cost plus the amortization of unfunded liability. Outsized returns may reduce the UAL to zero, but the normal cost is assumed to be contributed in all instances. The result is that the best-case cumulative contributions cannot be driven below normal cost. 37

Ultimate Net Cost 6/30/2016 Ultimate Net Cost $4,000 $3,000 $2,000 $1,000 $0 -$1,000 -$2,000 -$3,000 -$4,000 -$5,000 Percentile Target Mix 1 Mix 2 Mix 3 Mix 4 Mix 5 97.5th $3,167 $2,875 $2,984 $3,096 $3,186 $3,272 75th 1,905 1,986 1,952 1,929 1,904 1,880 50th 1,051 1,488 1,346 1,173 1,016 854 25th 156 962 706 377 69-270 2.5th -2,481-291 -933-1,855-2,760-3,789 Ultimate Net Cost (UNC) = 10-Year Cumulative Contributions + 6/30/2026 Unfunded Actuarial Liability UNC captures what is expected to be paid over 10 years plus what is owed at the end of the 10 year period. The majority of UNC for MCERA is comprised of contributions. More aggressive mixes lower UNC in the expected case but result in greater UNC in a worse case scenario. 38

Ultimate Net Cost Expected (50th Percentile) Ultimate Net Cost $800 $900 $1,000 $1,100 $1,200 $1,300 $1,400 $1,500 Risk versus Reward: Ultimate Net Cost over 10 Years Mix 1 Increase in Expected Cost Mix 2 Mix 3 Increase in Worse Case Cost Current Target Mix 4 Mix 5 $1,600 $2,850 $2,900 $2,950 $3,000 $3,050 $3,100 $3,150 $3,200 $3,250 $3,300 Worse-Case (97.5th Percentile) Ultimate Net Cost Ultimate Net Cost (UNC) = 10 year cumulative contributions (2016-2025) + 1/1/2026 Unfunded Liability What you paid over 10 years + what you owe at the end of 10 years An approximate linear risk-reward trade-off exists between the alternative mixes. The current Target mix is optimal based on Callan s 10-year capital market expectations. 39

Decision Factors Factor Return Objective Description Current actuarial assumed investment return is 7.25%. Only Mix 5, with 12% in fixed income, is projected to attain 7.25% over 2016-2025. However, the real return expectation for the fund is 4.25%, which is lower than Callan s real return expectation for the current target. Time Horizon Indefinite (plan is open) Liquidity Needs Actuarial Methodology Contribution Risk Risk Tolerance Liability Growth Funded Status Liquidity needs are moderate and will remain below 5% over the next 10 years, so not a material concern to the plan s investment strategy. Normal Cost plus closed period amortization of any UAL. Assets are not smoothed; actuarial value = market value Trade-off between lower median contribution rates and higher worse case contribution rates. Contributions are responsive to asset performance, especially on the downside. Normal cost contribution limits how far good performance can reduce contribution. Risk tolerance is the ability and willingness to take risk. Consider worse-case results for projected funded status, ultimate net cost and annual returns. Liabilities are growing steadily (3% over next 10 years). Liability return is 6.8%, using Callan s 2.25% inflation assumption. Plan funding is projected to rise to 94% in 10 ten years and reach 100% by 2030, under current benefit and funding policy. 40

Recommendation Substantial changes have been made to MCERA funding and benefit policy since the last asset/liability study in 2011. Elimination of asset smoothing and closed period amortization of the UAL makes contribution rates responsive to asset performance. Size of plan assets relative to payroll also heightens sensitivity of plan sponsor to adverse market results and the impact on supplemental cost to cover an unfunded liability. Normal cost contribution policy limits how far good investment performance can reduce contributions. Liability growth is linked to the actuarial discount rate (7.25%) and implies a nominal return target for the Plan. Capital markets will be challenged to generate a long-term return of 7.25%; however, the real return target of 4.25% embedded in the valuation is actually below Callan s expectation for the current Target, countering the pressure to pursue a higher nominal return. The plan has taken significant steps to close the funding deficit through contribution policy. Pursuing a higher expected return to further assist with closing the Plan deficit will expose it to greater contribution volatility. Liability return is projected to average 6.8% over the next 10 years, in line the nominal return expectation for the current target. Taking less risk than the current target would reduce contribution volatility and worse case outcomes, at the cost of lower expected return. The current target is a well-diversified portfolio that includes exposure to stocks, bonds, private equity and real estate and real assets, and can be retained as a reasonable policy. MCERA will need to retain a strong orientation toward growth in pursuit of return to achieve its funding goals Our analysis suggests that the return and risk position of the target can be expected to meet the funding needs of the Plan as articulated in the valuation, given our expectations for capital market performance. 41