August Asset/Liability Study Texas Municipal Retirement System

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

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

June 2, 2016 City #01160

June 2, 2016 City #00048

CITY OF LOS ANGELES DEPARTMENT OF WATER AND POWER INTRADEPARTMENTAL CORRESPONDENCE

May 30, 2014 City #00004

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

Texas Municipal Retirement System Actuarial Valuation Report as of December 31, 2017

Tacoma Employees Retirement System

Stochastic Analysis Of Long Term Multiple-Decrement Contracts

2011 Asset/Liability Study Preliminary Results

March 24, Board of Trustees Houston Municipal Employees Pension System 1201 Louisiana Suite 900 Houston, TX 77002

2014 Survey of GASB Accounting Measures for Public Pension Plans

City of San José Federated City Employees Retirement System

ACTUARIAL SECTION (UNAUDITED)

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

Pension Simulation Project Rockefeller Institute of Government

Arkansas Public Employees Retirement System Actuarial Valuation and Experience Gain/Loss Analysis June 30, 2017

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

Staying Ahead of the Investment Curve

F I R 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 ACTUARIAL VALUATION R E P O R T A S O F D E C E M B E R 3 1,

Teachers Retirement System of the State of Illinois

F I R E A N D P O L I C E P E N S I O N A S S O C I A T I O N

Review of October 1, 2017 Actuarial Valuation Results

Actuary s Certification Letter (Pension Trust Fund)

Actuary s Certification Letter (Pension Trust Fund)

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

City of Winter Springs Defined Benefit Plan Actuarial Valuation

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

Sheet Metal Workers' National Pension Fund

June 1, 2017 City #00004

City of Orlando Police Officers' Pension Fund

Charter Township of Independence. Other Post Employment Benefits

Maine Public Employees Retirement System State Employee and Teacher Retirement Program. Actuarial Valuation Report as of June 30, 2017

June 1, 2017 City #01180

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

Postemployment Health Insurance -- Sensitivity Tests Sensitivity Analysis RETIREE PREMIUM RATE DEVELOPMENT

The Water and Power Employees Retirement Plan of the City of Los Angeles ACTUARIAL EXPERIENCE STUDY

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

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

Texas Municipal Retirement System. August 22, Retiree Mortality Study. Joseph Newton Mark Randall. Copyright 2013 GRS All rights reserved.

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

CITY OF WINTER SPRINGS DEFINED BENEFIT PLAN ACTUARIAL VALUATION AS OF OCTOBER 1, 2008

There may also be changes specific to your plan such as contract amendments and funding changes.

Luke and Jen Smith. MONTE CARLO ANALYSIS November 24, 2014

There may also be changes specific to your plan such as contract amendments and funding changes.

Jackson County State of Michigan. Amended and Restated Comprehensive Financial Plan For Pension and Other Post-Employment Benefits

Fire and Police Pension Fund, San Antonio Actuarial Valuation and Review as of January 1, 2017

Report to Board of Administration

There may also be changes specific to your plan such as contract amendments and funding changes.

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

There may also be changes specific to your plan such as contract amendments and funding changes.

There may also be changes specific to your plan such as contract amendments and funding changes.

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

City of Albany Police and Fire Relief or Pension Fund

Target Date Glide Paths: BALANCING PLAN SPONSOR GOALS 1

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

There may also be changes specific to your plan such as contract amendments and funding changes.

Projected Results % $12,964,000 TBD % $14,311,000 TBD

Retirement. Optimal Asset Allocation in Retirement: A Downside Risk Perspective. JUne W. Van Harlow, Ph.D., CFA Director of Research ABSTRACT

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

Arkansas Judicial Retirement System Annual Actuarial Valuation and Experience Gain/(Loss) Analysis Year Ending June 30, 2018

Projected Results % $3,056,000 TBD % $3,453,000 TBD

City of Jacksonville General Employees Retirement Plan

Houston Police Officers Pension System ACTUARIAL VALUATION REPORT FOR THE YEAR BEGINNING JULY 1, 2017

Measuring Retirement Plan Effectiveness

Employer Contribution Rate % % (projected)

Fire and Police Pension Fund, San Antonio

The Water and Power Employees' Retirement Plan of the City of Los Angeles Actuarial Valuation and Review as of July 1, 2017

City of Hollywood General Employees Retirement System ACTUARIAL VALUATION REPORT AS OF OCTOBER 1, 2016

Cavanaugh Macdonald. The experience and dedication you deserve

Cavanaugh Macdonald. The experience and dedication you deserve

St. Paul Teachers Retirement Fund Association Actuarial Valuation as of July 1, 2018

PA Healthcare System Adopts a New Strategy to Tackle Financial Challenges

Hibernation versus termination

ORANGE COUNTY EMPLOYEES RETIREMENT SYSTEM. Review of Economic Actuarial Assumptions for the December 31, 2012 Actuarial Valuation

Massachusetts Water Resources Authority Employees Retirement System

Florida Retirement System

City of Jacksonville General Employees Retirement Plan Actuarial Valuation and Review as of October 1, 2016

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

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

March 25, Mr. Randall Blum Finance Director City of Eastpointe Eastpointe, Michigan Dear Mr. Blum:

December Mr. Randall Blum Finance Director City of Eastpointe Eastpointe, Michigan Dear Mr. Blum:

Projected Results % $415,000

The City of Omaha Police & Fire Retirement System

WYOMING JUDICIAL RETI R E M E N T S Y S T E M ACTUARIAL VALUATION R E P O R T FOR T H E Y E A R B E G I N N I N G J A N U A R Y 1,

Sheet Metal Workers' National Pension Fund

Orange County Employees Retirement System

Projected Results % $18,000

IMPERIAL COUNTY EMPLOYEES RETIREMENT SYSTEM. Review of Economic Actuarial Assumptions for the June 30, 2014 Actuarial Valuation

Projected Results % $1,830,000

ST. CLAIR COUNTY EMPLOYEES RETIREMENT SYSTEM

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

San Bernardino County Employees Retirement Association

The next regular meeting of the Retirement Board will be held at 8:30 a.m. on Thursday, March 15, 2018.

Projected Results % $68,000

Pennsylvania Municipal Retirement System

REVERSE ASSET ALLOCATION:

Report to Board of Administration

Cavanaugh Macdonald. The experience and dedication you deserve

Transcription:

August 2016 Asset/Liability Study Texas Municipal Retirement System

Table of Contents ACKNOWLEDGEMENTS... PAGE 2 INTRODUCTION... PAGE 3 CURRENT STATUS... PAGE 7 DETERMINISTIC ANALYSIS... PAGE 8 DETERMINISTIC SCENARIO ANALYSIS... PAGE 19 STOCHASTIC ANALYSIS... PAGE 21 APPENDIX 1: SUPPLEMENTAL STOCHASTIC EXHIBITS... PAGE 44 APPENDIX 2: SENSITIVITY ANALYSIS: VOLATILITY... PAGE 48 APPENDIX 3: SENSITIVITY ANALYSIS: CORRELATIONS... PAGE 51 APPENDIX 4: ASSUMPTIONS AND METHODS... PAGE 54 1

Acknowledgements PREPARED BY: JAMES VOYTKO, SENIOR CONSULTANT, RVK, INC. RYAN SULLIVAN, CONSULTANT, RVK, INC. MATTHIAS BAUER, CONSULTANT, RVK, INC. WITH THE COOPERATION OF: DAVID DOUGHERTY, LLC (RVK CONSULTING ACTUARY) GABRIEL ROEDER SMITH & COMPANY (SYSTEM ACTUARY) 2

Introduction RVK, Inc. (RVK) has prepared this report for the Texas Municipal Retirement System (TMRS) to: o Present projected valuation results with respect to the funded status of the Plan. o Present projected benefit payments of the Plan. o Investigate asset mixes to determine those which best serve to protect and increase funding levels, while providing adequate liquidity for benefit payments. The valuation projections are shown using both a deterministic and stochastic process. The deterministic process provides an open group analysis of projected valuation results based on a fixed set of future assumptions (see summary in the Assumptions and Methods section of this report). The stochastic process provides an open group analysis of projected valuation results under many capital market environments based on expected asset returns and inflation, and their expected volatility. Using a Monte Carlo simulation technique, both assets and liabilities are assumed to vary stochastically, linked together by changes in inflation. Expected values, variances of the returns and inflation, and correlations are used to generate 2,000 trials to produce a distribution of potential outcomes. A stochastic analysis can answer questions about the best/worst case outcomes along with the probability of such outcomes. 3

Introduction (continued) What is an Asset/Liability Study? Investment programs and the strategy they seek to implement (Investment Policy) do not exist in a vacuum. They seek to satisfy one or more investment objectives and operate within a plan framework that includes the investment objectives (Benefit Policy) and plan funding (Contribution Policy). The purpose of an Asset/Liability Study is to examine how well alternative investment strategies (i.e., differing asset allocations) address the objectives served by the Plan the Plan s liabilities in the context of the Plan s funding streams the Plan s Contribution Policy. It is the only standard analysis that fully links all three aspects of the Plan s key financial drivers. In doing so, it creates an important guidepost for the actual asset allocation for the Plan; the asset allocation chosen by the Plan s fiduciaries will likely reflect the nature of the liabilities but also numerous other factors including risk preferences, liquidity, implementation constraints, etc. For the TMRS Asset/Liability Study, we assume the objectives are: 1. Fund all participants benefits over time. 2. Assure sufficient liquidity to pay benefits at all times. 3. Foster a stable contribution stream consistent with objectives 1 and 2. 4. Achieve adequate returns without accepting unnecessary or imprudent levels of risk. An Asset/Liability Study is NOT... An actuarial study of the TMRS liabilities that is the purview of the Plan s actuary. A prescription for Plan benefits that is the purview of the elected representatives. Contribution Policy Investment Policy Asset Liability Analysis Benefit Policy An assessment of the affordability of contribution levels that is the purview of the elected officials and their constituents. The sole determinant of the final asset allocation adopted for the Plan there are a number of factors, including insights from an Asset/Liability Study, which will bear on the optimal asset allocation. 4

Introduction (continued) Asset/Liability Studies in Practice... Begin with a forecast of the financial liabilities (i.e., benefit obligations). Include a baseline estimation of the financial contributions to the Plan over time. Compare alternative investment strategies (i.e., total fund asset allocations to the Plan s financial needs). Draw conclusions regarding how well various investment strategies satisfy the Plan s financial needs. This Asset/Liability Study... Uses data from the December 31, 2015 TMRS Actuarial Valuation provided by the Gabriel Roeder Smith & Company (GRS) to project pension liabilities. Uses the actuarial cost method and the actuarial assumptions described in the December 31, 2015 TMRS Actuarial Valuation prepared by GRS. Compares these specific investment strategies (A) the Current Allocation (as of April 30, 2016), (B) the Target Allocation, (C) a conservative illustrative portfolio (100% Fixed Income), (D) a diversified portfolio with reduced risk and a diversified portfolio with increased risk relative to the Target Allocation (Lower Risk and Higher Risk), and (E) an aggressive illustrative portfolio (100% Equity). Assumes the Plan s current benefit policy throughout the entire projection period changes to the benefit policy are the purview of the elected representatives. Note: Does not assume any actuarial adjustments that may take place in future years. 5

Introduction (continued) Key Takeaways Assets are available to cover 84% of the System s liabilities as currently estimated by the System s actuary. This equates to a shortfall of approximately $4.7 billion. The funding ratio on a market value basis is expected to gradually increase to approximately 98% by 2035 from the current value of 84%. The present funding level is unquestionably a strong financial position relative to most other public pension plans. This Study suggests that continued diversification in the investment strategy of the System s assets is desirable. This Study does not suggest changes to the current investment strategy in place. This Study does not suggest changes to the long-term strategic target allocation. The incremental cost of additional volatility does not justify the potential increase in median outcomes. Reducing volatility increases contributions and does not improve the median outcome. 6

Current Status A summary of the Plan follows: Valuation Date December 31, 2015 Assumed Rate of Return 6.75% Market Value of Assets (MVA) $23.7 billion Actuarial Value of Assets (AVA) $24.3 billion Actuarial Accrued Liability (AAL) $28.4 billion Billions $35 $30 $25 $20 $15 $10 $5 $0 $23.7 Market Value of Assets Current Status As of December 31, 2015 $28.4 Actuarial Accrued Liability $4.7 Deficit Market Value (MVA/AAL) 84% Demographics Actuarial Value (AVA/AAL) 86% Active 106,894 120,000 100,000 80,000 60,000 106,894 56,481 50,707 Retirees and Beneficiaries 56,481 Inactive Vested 50,707 40,000 20,000 0 Active Retirees and Beneficiaries Inactive Vested 7

Deterministic Analysis This section provides an analysis of the Plan s assets, liabilities, funded status, and benefit payments based on a fixed set of future assumptions. Each analysis that follows in this deterministic section rests on the critical assumptions below and must be read and interpreted with them in mind particularly assumptions #2, #3 and #4. The deterministic assumptions are as follows: 1. Current Plan provisions (see Summary of Benefit Provisions in Section 7 of the December 31, 2015 actuarial valuation report prepared by GRS.) 2. The participant data used in the December 31, 2015 actuarial valuation prepared by GRS. Active participant data was grouped to speed processing. 3. Actuarially assumed rate of return on Plan assets for all projection years: 6.75%. 4. For 2016, employer contributions were assumed equal to 12.63% of system-wide pay. For all other years of the projection, employer contributions are assumed to equal: (1) normal cost less expected employee contributions, plus (2) an amortization of the unfunded actuarial accrued liability (UAAL). The unfunded liability as of December 31, 2015, was amortized over 21-years to approximate the Plan s equivalent single amortization period of 20.6 years. New experience gains and losses were amortized over 23 years to approximate the Plan s methodology that calculates costs by city, and includes offsetting of experience losses. 5. Asset Valuation Method Assumes a 10-year phase-in of actual versus expected returns. 6. Assumes demographic experience projected in accordance with the assumptions used in the December 31, 2015 actuarial valuation prepared by GRS. 7. Open group analysis: level active populations of Fire, Police, and Other subgroups. New active participants entering the Plan are assumed to have similar characteristics to recently hired participants. 8

Deterministic Analysis (continued) Demographics Following are the projected number of active and inactive participants at the beginning of each Plan year from 2014 through 2035 (2015 is actual). These projections are based on an open group analysis. Using the actuary s assumptions for death, termination, retirement, and disability, current participants are assumed to leave the Plan in the future. The number of total inactive participants (Retirees and Beneficiaries and Vested Inactive) increases by approximately 45% during the 20-year projection period shown. Projected Demographics 300,000 Active Retirees and Beneficiaries Inactive Vested 250,000 200,000 150,000 100,000 50,000 0 At Plan Year Ending Total Population Annual Percent Change N/A 2% 2% 2% 2% 1% 1% 1% 1% 1% 1% 1% 1% 1% 1% 1% 0% 1% 1% 0% 0% 9

Deterministic Analysis (continued) Benefit Payments The Plan s projected annual benefit payments are shown in the chart below. The projected benefit payments are expected to increase by about 232% over the next 20 years. As a percentage of the market value of Plan assets, benefit payments are expected to remain roughly constant through the end of the projection period (see page 13). Projected Benefit Payments Billions $4.0 $3.5 $3.0 $2.5 $2.0 $1.5 $1.1 $1.3 $1.5 $1.6 $1.7 $1.7 $1.9 $2.0 $2.1 $2.2 $2.3 $2.5 $2.6 $2.8 $2.9 $3.0 $3.2 $3.4 $3.5 $3.6 $3.7 $1.0 $0.5 $0.0 For the Plan Year Ending 10

Deterministic Analysis (continued) Contributions The Plan s projected contributions, expressed as total dollar contributions, are shown in the chart below. The results assume the contribution policy remains unchanged, and that the Plan s assets return precisely the actuarially assumed rate each year without exception for all projection years. Projected Contributions $2.5 Employer Contribution Employee Contribution Billions $2.0 $1.5 $1.0 $1.1 $1.2 $1.2 $1.3 $1.3 $1.3 $1.4 $1.4 $1.4 $1.5 $1.5 $1.5 $1.6 $1.6 $1.7 $1.7 $1.7 $1.8 $1.8 $1.9 $1.9 $0.5 $0.0 For the Plan Year Ending Annual Percent Change N/A 6% 3% 2% 2% 3% 3% 3% 2% 3% 3% 2% 2% 2% 3% 3% 3% 2% 2% 3% 3% 11

Deterministic Analysis (continued) Contributions The Plan s projected contributions, expressed as a weighted average percentage of salary, are shown in the chart below. The results assume the contribution policy remains unchanged, and that the Plan s assets return precisely the actuarially assumed rate each year without exception for all projection years. Projected Contributions (as a weighted average % of Salary) 25% Employer Contribution Employee Contribution 20% 19% 20% 20% 20% 20% 20% 20% 20% 20% 20% 20% 20% 20% 20% 20% 20% 20% 20% 20% 20% 20% 15% 10% 5% 0% For the Plan Year Ending 12

Deterministic Analysis (continued) Payout (benefit payments/market value of assets) The Plan s projected payout ratios are shown in the chart below. The payout ratio is expected to remain constant through the end of the projection period. The results assume the current contribution policy remains unchanged and that the Plan s assets return precisely the actuarially assumed rate each year without exception for all projection years. Projected Payout (Projected Benefit Payments/Projected Market Value of Assets) 10% 8% 6% 4% 5% 5% 5% 6% 6% 6% 6% 6% 6% 6% 6% 6% 6% 6% 6% 6% 6% 6% 6% 6% 6% 2% 0% For the Plan Year Ending 13

Deterministic Analysis (continued) Benefit Payments/Contributions The Plan s projected benefit payments divided by projected contributions are shown in the chart below. The results assume the contribution policy remains unchanged, and that the Plan s assets return precisely the actuarially assumed rate each year without exception for all projection years. Projected Benefit Payments/Projected Contributions 200% 180% 160% 140% 120% 100% 80% 60% 40% 20% 0% 98% 188% 189% 190% 190% 124% 128% 132% 137% 145% 150% 152% 155% 159% 167% 172% 181% 173% 175% 112% 118% For the Plan Year Ending 14

Deterministic Analysis (continued) Actuarial Accrued Liabilities and Market Value of Assets The Plan s projected actuarial accrued liabilities and market value of assets are shown in the chart below. The results assume the contribution policy remains unchanged, and that the Plan s assets return precisely the actuarially assumed rate each year without exception for all projection years. The relative disparity between the market value of assets and Plan liabilities is expected to decrease by 70% through the end of the projection period. The funded ratio (based on market value of assets) is expected to gradually increase to 98% by the end of the projection period. This is shown more clearly on the following pages. Projected Market Value of Assets and Projected Actuarial Liabilities $70 $60 $50 Market Assets Actuarial Accrued Liability Billions $40 $30 $20 $10 $0 For the Plan Year Ending 15

Deterministic Analysis (continued) Deficit (market value of assets actuarial accrued liabilities) The Plan s projected deficit of assets is shown in the chart below. The results assume the contribution policy remains unchanged, and that the Plan s assets return precisely the actuarially assumed rate each year without exception for all projection years. The disparity between the market value of assets and Plan liabilities is expected to decrease by the end of the projection period by 70%. 16

Deterministic Analysis (continued) Actuarial (actuarial value of assets/actuarial accrued liability) The Plan s projected actuarial funded ratio is shown in the chart below. The Plan is expected to end the projection period at approximately 98% funded. The results assume the contribution policy remains unchanged, and that the Plan s assets return precisely the actuarially assumed rate each year without exception for all projection years. Projected Actuarial 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% 86% 87% 87% 88% 88% 88% 89% 89% 90% 90% 91% 91% 92% 92% 93% 94% 95% 95% 96% 97% 98% For the Plan Year Ending 17

Deterministic Analysis (continued) Market (market value of assets/actuarial accrued liability) The Plan s projected market funded ratio is shown in the chart below. The Plan is expected to end the projection period at approximately 98% funded. The results assume the contribution policy remains unchanged, and that the Plan s assets return precisely the actuarially assumed rate each year without exception for all projection years. Projected Market 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% 84% 85% 85% 86% 87% 87% 88% 88% 89% 90% 90% 91% 92% 92% 93% 94% 94% 95% 96% 97% 98% For the Plan Year Ending 18

Deterministic Scenario Analysis Full Funding Implied Returns The figure below shows the projected investment return for the total fund needed to bring the Plan to 100% funding (on a market value basis) in 10 and 20 years, respectively. The results assume all other actuarial assumptions are precisely met over the time periods shown and that these returns are earned for every year, without variance. Actuarially assumed rate of return 6.75% Projected Annual Rate of Return Needed to Reach Full Funding 9% 8% 7% 6% 5% 4% 3% 2% 1% 0% 7.9% 6.9% 10 Years 20 Years 19

Deterministic Scenario Analysis (continued) Sensitivity Analysis Decreased Return Under the deterministic analysis presented in the preceding pages, the Plan is projected to have a market funded ratio of 98% in 20 years. The table below summarizes the projected funded ratio and other key statistics in 2035 assuming the Plan experiences an annualized investment return of 100 basis points lower (5.75%) than the current actuarially assumed rate of return (6.75%). The values assume all other actuarial assumptions are exactly met. The original values are also presented in the table for comparison. Actuarially Assumed Rate of Return Value in 2035 Reduced Return (100 bps) Impact of Reduced Return Projected Payout 6% 7% 1% Projected Employer Contributions (billions) $1.3 $1.8 $0.5 Projected Benefit Payments/Projected Total Contributions 190% 150% -40% Projected Actuarial Accrued Liabilities (billions) $61.5 $61.5 $0.0 Projected Market Value of Assets (billions) $60.1 $50.8 ($9.3) Projected Deficit (billions) $1.4 $10.6 $9.2 Projected Market 98% 83% -15% 20 Year Cumulative Total Projected Cumulative Employer Contributions (billions) $21.4 $24.8 $3.4 Values in impact column may be impacted by rounding. 20

Stochastic Analysis In the previous section of this report, we assumed the Plan operated going forward with certain knowledge of the future investment returns earned by the Plan s assets. This section introduces the element of uncertainty in those future investment returns. This part of the analysis examines Plan assets and liabilities under many capital market environments based on expected future asset returns and inflation, and their expected volatility. Using a Monte Carlo simulation technique, both assets and liabilities are assumed to vary stochastically, linked together by changes in inflation. Using the current expected values and variances of the returns and inflation, along with their correlations, 2,000 trials are generated to produce a distribution of results. A stochastic analysis can answer questions about the best/worst case outcomes along with the probability of such outcomes. This is contrasted with the deterministic analysis that provides an expected value if all current Plan assumptions are exactly met. 21

Stochastic Analysis (continued) Long-Term Return and Risk Assumptions In order to perform a stochastic analysis and create asset allocation alternatives, it is necessary to estimate, for each asset class, its probable return and risk. The expected returns are our best estimates of the average annual percentage increases in values of each asset class over a prospective long period of time, and assumed to be normally distributed. The risk of an asset class is measured by its standard deviation, or volatility. If asset returns are normally distributed, two-thirds (67%) of all returns are expected to lie within one standard deviation on either side of the mean. For example, we expect Global Equity (50/50 Equity) to return, annually on average, 7.95% with a standard deviation of 18.40%, meaning that two-thirds of the time we expect its return to lie between -10.45% (= 7.95 18.40) and 26.35% (= 7.95 + 18.40). Moreover, we expect 95% of all return outcomes to lie within two standard deviations of the mean return, implying only a one-in-twenty chance that the return on Global Equity will either fall below -28.85% or rise above 44.75%. The risk and return assumptions used in this study are outlined in the below table and chart: Asset Class Arithmetic Return Assumption Standard Deviation Assumption 50/50 Equity 7.95 18.40 Intermediate Duration Fixed Income 3.50 6.00 Non-Core Fixed Income 6.65 12.50 Custom Real Return 6.65 8.70 Custom Real Estate 7.25 14.00 Diversified Hedge Funds 6.50 9.50 Private Equity 10.25 25.50 Airthmetic Return (Annualized, %) 20.00 18.00 16.00 14.00 12.00 10.00 8.00 6.00 4.00 2.00 0.00 0.00 3.00 6.00 9.00 12.00 15.00 18.00 21.00 24.00 27.00 30.00 33.00 Risk (Annualized Standard Deviation, %) 50/50 Equity Int FI Non-Core FI Custom RR Custom RE DHF Priv. Eq. 22

Stochastic Analysis (continued) Correlation Between Asset Classes Creating a diversified portfolio of asset classes enables the investor to achieve a high rate of return while minimizing volatility of the portfolio. As defined on the previous page, volatility is risk or standard deviation. By minimizing the volatility of a portfolio, we produce asset returns that vary less from year to year. Diversification exists because the returns of different asset classes do not always move in the same direction, at the same time, or with the same magnitude. Correlation values are between 1.00 and 1.00. If returns of two asset classes rise or fall at the same time and in the same magnitude, they have a correlation value of 1.00. Conversely, two asset classes that simultaneously move in opposite directions, and in the same magnitude, have a correlation value of 1.00. A correlation of zero indicates no relationship between returns. The assumed correlations are largely based on historical index data, with some qualitative analysis applied. For instance, where appropriate, we have weighted current history more heavily. The correlation matrix used in this study is shown below: 50/50 Equity Int. Duration Fixed Income Non-Core Fixed Income Custom Real Return Custom Real Estate Diversified Hedge Funds Private Equity 50/50 Equity 1.00-0.01 0.83 0.52 0.29 0.70 0.79 Int. Duration Fixed Income -0.01 1.00 0.34 0.16-0.10 0.14-0.28 Non-Core Fixed Income 0.83 0.34 1.00 0.79 0.15 0.65 0.64 Custom Real Return 0.52 0.16 0.79 1.00 0.30 0.55 0.61 Custom Real Estate 0.29-0.10 0.15 0.30 1.00 0.27 0.57 Diversified Hedge Funds 0.70 0.14 0.65 0.55 0.27 1.00 0.70 Private Equity 0.79-0.28 0.64 0.61 0.57 0.70 1.00 The fact that the correlations shown in the table are nearly all positive does not imply that these asset classes do not diversify one another. Their correlations are significantly less than 1.00, meaning we expect a measurable number of instances when the underperformance of one or more of the asset classes will be offset by the outperformance of others. This point is demonstrated on the following pages, which illustrate that diversification into less correlated asset classes can decrease the expected overall volatility of a portfolio. 23

Stochastic Analysis (continued) Efficient Portfolios Each frontier portfolio (optimal allocation) is created using target rates of return both above and below the projected rate of return for the current allocation. This range illustrates the trade-off between return and risk; additional return can only be achieved by undertaking additional risk. The table below shows the possible optimal allocations given the selected asset classes and their constraints listed under Min and Max. The table shows the Current (as of April 30, 2016) and Target Allocations and highlights four additional portfolios (100% Fixed Income, Lower Risk, Higher Risk, and 100% Equity) for consideration throughout this study. Min Max 1 2 3 4 5 6 7 8 9 10 The combination of Real Return, Real Estate, Diversified Hedge Funds, and Private Equity is limited to 50%. Current Allocation 100% Fixed Income Lower Risk Target Allocation 50/50 Equity 0 100 0 0 11 14 6 21 23 28 41 50 44 0 25 35 45 80 Int. Duration Fixed Income 0 100 65 50 39 26 8 0 0 0 0 0 29 100 20 10 5 0 Non-Core Fixed Income 0 100 0 0 0 10 36 29 27 22 9 0 7 0 20 20 15 0 Custom Real Return 0 100 14 22 28 27 16 17 6 0 0 0 6 0 10 10 5 0 Custom Real Estate 0 100 11 14 17 22 29 33 30 24 13 0 7 0 10 10 10 0 Diversified Hedge Funds 0 100 10 14 5 2 5 0 0 0 0 0 8 0 10 10 10 0 Private Equity 0 100 0 0 0 0 0 0 14 26 37 50 0 0 5 5 10 20 Total 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 Capital Appreciation 0 0 11 24 42 50 64 76 87 100 51 0 50 60 70 100 Capital Preservation 65 50 39 26 8 0 0 0 0 0 29 100 20 10 5 0 Alpha 10 14 5 2 5 0 0 0 0 0 8 0 10 10 10 0 Inflation 25 36 45 48 45 50 36 24 13 0 12 0 20 20 15 0 Expected Return 4.65 5.14 5.64 6.13 6.63 7.12 7.62 8.11 8.61 9.10 6.35 3.50 6.57 7.02 7.48 8.41 Risk (Standard Deviation) 4.95 5.15 6.03 7.34 8.79 10.43 12.69 15.20 17.92 20.80 10.19 6.00 9.85 11.51 13.53 19.01 Return (Compound) 4.53 5.01 5.47 5.88 6.27 6.62 6.88 7.06 7.16 7.17 5.87 3.33 6.12 6.41 6.64 6.78 Return/Risk 0.94 1.00 0.94 0.84 0.75 0.68 0.60 0.53 0.48 0.44 0.62 0.58 0.67 0.61 0.55 0.44 RVK Expected Eq Beta (LCUS Eq = 1) 0.11 0.13 0.21 0.28 0.32 0.42 0.56 0.70 0.87 1.03 0.52 0.06 0.46 0.55 0.67 0.99 RVK Liquidity Metric (T-Bills = 100) 66 58 59 55 44 47 44 43 47 49 75 85 60 61 61 75 Higher Risk 100% Equity 24

Stochastic Analysis (continued) Efficient Frontier The risk of each alternative allocation is plotted against the horizontal axis, while the return is measured on the vertical axis. The line connecting the points represents all the optimal portfolios subject to the given constraints and is known as the efficient frontier. The upward slope of the efficient frontier indicates the direct relationship between return and risk. Efficient Frontier 10.00 9.50 10 9.00 9 8.50 8.00 7 8 100% Equity Return (Annualized, %) 7.50 7.00 6.50 6.00 5.50 2 3 4 5 6 Higher Risk Target Allocation Lower Risk Current Allocation 5.00 1 4.50 4.00 3.50 100% Fixed Income 3.00 4.00 5.00 6.00 7.00 8.00 9.00 10.00 11.00 12.00 13.00 14.00 15.00 16.00 17.00 18.00 19.00 20.00 21.00 22.00 Risk (Annualized Standard Deviation, %) 25

Stochastic Analysis (continued) Asset Mixes Outlined below are the Current (as of April 30, 2016) and Target Allocations and four other mixes to be examined in this stochastic analysis. The expected return, expected risk (as measured by standard deviation), and RVK Liquidity Metric, for each is also shown. Asset Class Current Allocation 100% Fixed Income Lower Risk Target Allocation Higher Risk 100% Equity 50/50 Equity 44% 0% 25% 35% 45% 80% Intermediate Duration Fixed Income 29% 100% 20% 10% 5% 0% Non-Core Fixed Income 7% 0% 20% 20% 15% 0% Custom Real Return 6% 0% 10% 10% 5% 0% Custom Real Estate 7% 0% 10% 10% 10% 0% Diversified Hedge Funds 8% 0% 10% 10% 10% 0% Private Equity 0% 0% 5% 5% 10% 20% Total Equity 44% 0% 30% 40% 55% 100% Expected Return 6.34% 3.50% 6.57% 7.02% 7.48% 8.41% Expected Risk 10.18% 6.00% 9.85% 11.51% 13.53% 19.01% RVK Liquidity Metric 75 85 60 61 61 75 26

Stochastic Analysis (continued) Projected Actuarial (actuarial value of assets/actuarial accrued liability); 20 Years The graph below shows the distribution of possible actuarial funded ratios twenty years from now, assuming the six different asset mixes highlighted on the prior pages. The results assume the current contribution policy remains unchanged for all projection years. 400% Projected Actuarial December 31, 2035 350% 300% 95th 250% 200% 150% 75th Median 100% 25th 50% 0% Current Allocation 100% Fixed Income Lower Risk Target Allocation Higher Risk 100% Equity 5th Current Allocation 100% Fixed Income Lower Risk Target Allocation Higher Risk 100% Equity Liability 5th $26.1 56% $28.9 53% $24.6 58% $25.5 56% $26.8 54% $31.2 46% 25th $17.6 71% $23.7 61% $15.9 73% $15.5 73% $15.5 74% $16.6 72% Median $9.7 84% $20.0 67% $8.2 86% $5.9 90% $3.5 94% ($2.4) 104% 75th ($0.9) 101% $16.3 73% ($2.7) 104% ($9.0) 114% ($16.0) 125% ($35.8) 155% 95th ($28.1) 144% $9.0 86% ($31.6) 148% ($49.5) 174% ($71.5) 214% ($153.3) 342% 27

Stochastic Analysis (continued) Projected Market (market value of assets/actuarial accrued liability); 20 Years The graph below shows the distribution of possible market funded ratios twenty years from now, assuming the six different asset mixes highlighted on the prior pages. The results assume the current contribution policy remains unchanged for all projection years. 450% Projected Market December 31, 2035 400% 350% 300% 250% 95th 75th 200% Median 150% 100% 25th 50% 0% Current Allocation 100% Fixed Income Lower Risk Target Allocation Higher Risk 100% Equity 5th Current Allocation 100% Fixed Income Lower Risk Target Allocation Higher Risk 100% Equity Liability 5th $30.4 48% $34.1 45% $29.1 50% $29.8 48% $31.5 46% $35.8 39% 25th $20.9 65% $28.3 53% $19.2 68% $18.2 70% $17.8 70% $18.9 69% 50th $11.2 82% $24.6 59% $9.4 85% $6.2 89% $3.4 95% ($3.7) 106% 75th ($2.4) 104% $20.7 67% ($4.5) 107% ($11.9) 118% ($19.7) 131% ($42.3) 166% 95th ($34.3) 151% $12.6 80% ($38.2) 156% ($58.2) 190% ($87.1) 234% ($182.3) 389% 28

Stochastic Analysis (continued) Projected Market and Maximum 1 Year Investment Loss (market value of assets/actuarial accrued liability) The tables below show the probability that the Plan will be at various funding levels for each of the six different asset mixes highlighted on the prior pages. The tables also illustrate the maximum 1 year investment loss each portfolio is expected to experience during the given time period. The results assume the current contribution policy remains unchanged for all projection years. 5 Years Probability of Full Probability of < 84% Probability of < 60% Maximum 1 Year Funding in 2020 (Current) Funding in 2020 Funding in 2020 Investment Loss Current Allocation 17% 57% 7% -25% 100% Fixed Income 1% 86% 5% -18% Lower Risk 18% 55% 5% -24% Target Allocation 23% 52% 8% -28% Higher Risk 28% 51% 10% -33% 100% Equity 34% 49% 17% -46% 10 Years Probability of Full Probability of < 84% Probability of < 60% Maximum 1 Year Funding in 2025 (Current) Funding in 2025 Funding in 2025 Investment Loss Current Allocation 23% 57% 14% -25% 100% Fixed Income 1% 93% 29% -18% Lower Risk 25% 54% 13% -25% Target Allocation 31% 50% 14% -28% Higher Risk 35% 49% 16% -33% 100% Equity 40% 46% 22% -46% 20 Years Probability of Full Probability of < 84% Probability of < 60% Maximum 1 Year Funding in 2035 (Current) Funding in 2035 Funding in 2035 Investment Loss Current Allocation 28% 53% 18% -28% 100% Fixed Income 0% 97% 53% -21% Lower Risk 32% 49% 15% -25% Target Allocation 40% 44% 15% -29% Higher Risk 46% 40% 16% -33% 100% Equity 54% 35% 18% -46% 29

Stochastic Analysis (continued) Projected Payout (expected benefit payments/market value of assets); Current Allocation The graph below displays the range of possible payout ratios over the next twenty years, assuming the Plan s assets are allocated according to the Current Allocation. The results assume the current contribution policy remains unchanged for all projection years. The median annual benefit payment as percentage of the market value of assets is expected to range between 5% and 7%. The worst-case scenario could reach 13% or higher. 20% 15% 10% 5% Projected Payout Current Allocation 5th 25th Median 75th 95th 0% For the Plan Year Ending Median 5% 5% 6% 6% 6% 6% 6% 6% 6% 7% 7% 7% 7% 7% 7% 7% 7% 7% 7% 7% 7% 30

Stochastic Analysis (continued) Projected Payout (expected benefit payments/market value of assets); 100% Fixed Income The graph below displays the range of possible payout ratios over the next twenty years, assuming the Plan s assets are allocated according to the 100% Fixed Income portfolio. The results assume the current contribution policy remains unchanged for all projection years. The median annual benefit payment as percentage of the market value of assets is expected to range between 5% and 10%. The worst-case scenario could reach 13% or higher. 20% 15% 10% 5% Projected Payout 100% Fixed Income 5th 25th Median 75th 95th 0% For the Plan Year Ending Median 5% 6% 6% 6% 6% 7% 7% 7% 8% 8% 8% 8% 9% 9% 9% 9% 10% 10% 10% 10% 10% 31

Stochastic Analysis (continued) Projected Payout (expected benefit payments/market value of assets); Lower Risk The graph below displays the range of possible payout ratios over the next twenty years, assuming the Plan s assets are allocated according to the Lower Risk portfolio. The results assume the current contribution policy remains unchanged for all projection years. The median annual benefit payment as percentage of the market value of assets is expected to range between 5% and 7%. The worst-case scenario could reach 12% or higher. 20% 15% 10% 5% Projected Payout Lower Risk 5th 25th Median 75th 95th 0% For the Plan Year Ending Median 5% 5% 6% 6% 6% 6% 6% 6% 6% 6% 6% 7% 7% 7% 7% 7% 7% 7% 7% 7% 7% 32

Stochastic Analysis (continued) Projected Payout (expected benefit payments/market value of assets); Target Allocation The graph below displays the range of possible payout ratios over the next twenty years, assuming the Plan s assets are allocated according to the Target Allocation. The results assume the current contribution policy remains unchanged for all projection years. The median annual benefit payment as percentage of the market value of assets is expected to range between 5% and 7%. The worst-case scenario could reach 13% or higher. 20% 15% 10% 5% Projected Payout Target Allocation 5th 25th Median 75th 95th 0% For the Plan Year Ending Median 5% 5% 6% 6% 6% 6% 6% 6% 6% 6% 6% 6% 6% 7% 7% 7% 7% 7% 7% 7% 7% 33

Stochastic Analysis (continued) Projected Payout (expected benefit payments/market value of assets); Higher Risk The graph below displays the range of possible payout ratios over the next twenty years, assuming the Plan s assets are allocated according to the Higher Risk portfolio. The results assume the current contribution policy remains unchanged for all projection years. The median annual benefit payment as percentage of the market value of assets is expected to range between 5% and 6%. The worst-case scenario could reach 14% or higher. 20% 15% 10% 5% Projected Payout Higher Risk 5th 25th Median 75th 95th 0% For the Plan Year Ending Median 5% 5% 6% 6% 6% 6% 6% 6% 6% 6% 6% 6% 6% 6% 6% 6% 6% 6% 6% 6% 6% 34

Stochastic Analysis (continued) Projected Payout (expected benefit payments/market value of assets); 100% Equity The graph below displays the range of possible payout ratios over the next twenty years, assuming the Plan s assets are allocated according to the 100% Equity portfolio. The results assume the current contribution policy remains unchanged for all projection years. The median annual benefit payment as percentage of the market value of assets is expected to range between 5% and 6%. The worst-case scenario could reach 16% or higher. 20% 15% 10% 5% Projected Payout 100% Equity 5th 25th Median 75th 95th 0% For the Plan Year Ending Median 5% 5% 6% 6% 6% 6% 6% 6% 6% 6% 6% 6% 6% 6% 6% 6% 6% 6% 6% 6% 6% 35

Stochastic Analysis (continued) Cumulative Contributions to Date; Current Allocation The graph and table below show the range of projected cumulative contributions over the next twenty years, assuming the Plan s assets are allocated according to the Current Allocation (highlighted on the prior pages). The results assume the current contribution policy remains unchanged for all projection years. Billions $50 $45 $40 $35 $30 $25 $20 $15 $10 $5 $0 Projected Cumulative Employer Contributions to Date Current Allocation For the Plan Year Ending 5th 25th Median 75th 95th 5th $0.7 $1.6 $2.5 $3.5 $4.6 $5.8 $7.1 $8.6 $10.1 $11.8 $13.6 $15.6 $17.5 $19.7 $22.1 $24.7 $27.4 $30.3 $33.3 $36.6 $40.1 25th $0.7 $1.6 $2.4 $3.3 $4.3 $5.3 $6.4 $7.6 $8.9 $10.3 $11.8 $13.4 $15.1 $16.8 $18.7 $20.7 $22.8 $24.9 $27.2 $29.7 $32.2 Median $0.7 $1.5 $2.4 $3.2 $4.1 $5.0 $6.0 $7.1 $8.2 $9.3 $10.5 $11.7 $13.0 $14.5 $15.9 $17.5 $19.1 $20.7 $22.5 $24.4 $26.2 75th $0.7 $1.5 $2.3 $3.1 $4.0 $4.8 $5.6 $6.5 $7.4 $8.3 $9.2 $10.2 $11.1 $12.1 $13.1 $14.0 $15.2 $16.1 $17.2 $18.3 $19.5 95th $0.7 $1.5 $2.3 $3.0 $3.8 $4.4 $5.1 $5.7 $6.3 $6.9 $7.6 $8.2 $8.8 $9.5 $10.1 $10.8 $11.4 $12.1 $12.9 $13.7 $14.4 36

Stochastic Analysis (continued) Cumulative Contributions to Date; 100% Fixed Income The graph and table below show the range of projected cumulative contributions over the next twenty years, assuming the Plan s assets are allocated according to the 100% Fixed Income portfolio (highlighted on the prior pages). The results assume the current contribution policy remains unchanged for all projection years. Billions $50 $45 $40 $35 $30 $25 $20 $15 $10 $5 $0 Projected Cumulative Employer Contributions to Date 100% Fixed Income For the Plan Year Ending 5th 25th Median 75th 95th 5th $0.7 $1.6 $2.5 $3.4 $4.5 $5.7 $7.0 $8.4 $10.0 $11.6 $13.4 $15.4 $17.6 $19.9 $22.4 $25.1 $28.0 $31.1 $34.2 $37.7 $41.2 25th $0.7 $1.6 $2.4 $3.3 $4.3 $5.4 $6.5 $7.8 $9.1 $10.6 $12.2 $13.9 $15.8 $17.8 $20.0 $22.3 $24.8 $27.5 $30.2 $33.3 $36.5 Median $0.7 $1.5 $2.4 $3.2 $4.1 $5.1 $6.2 $7.3 $8.6 $9.9 $11.3 $12.9 $14.6 $16.4 $18.4 $20.5 $22.7 $25.1 $27.7 $30.4 $33.2 75th $0.7 $1.5 $2.3 $3.2 $4.0 $4.9 $5.9 $6.9 $8.0 $9.2 $10.5 $11.9 $13.4 $15.0 $16.7 $18.5 $20.4 $22.5 $24.7 $27.2 $29.8 95th $0.7 $1.5 $2.3 $3.1 $3.8 $4.6 $5.5 $6.4 $7.3 $8.3 $9.4 $10.5 $11.7 $13.0 $14.3 $15.8 $17.2 $18.9 $20.6 $22.5 $24.5 37

Stochastic Analysis (continued) Cumulative Contributions to Date; Lower Risk The graph and table below show the range of projected cumulative contributions over the next twenty years, assuming the Plan s assets are allocated according to the Lower Risk portfolio (highlighted on the prior pages). The results assume the current contribution policy remains unchanged for all projection years. Billions $50 $45 $40 $35 $30 $25 $20 $15 $10 $5 $0 Projected Cumulative Employer Contributions to Date Lower Risk For the Plan Year Ending 5th 25th Median 75th 95th 5th $0.7 $1.6 $2.5 $3.5 $4.6 $5.7 $7.0 $8.4 $9.9 $11.6 $13.4 $15.2 $17.2 $19.3 $21.5 $23.9 $26.6 $29.4 $32.5 $35.3 $38.5 25th $0.7 $1.6 $2.4 $3.3 $4.3 $5.3 $6.4 $7.6 $8.8 $10.2 $11.6 $13.1 $14.7 $16.4 $18.2 $20.0 $22.0 $24.1 $26.3 $28.6 $31.1 Median $0.7 $1.5 $2.4 $3.2 $4.1 $5.0 $6.0 $7.0 $8.1 $9.2 $10.4 $11.6 $12.9 $14.2 $15.6 $17.1 $18.6 $20.2 $21.9 $23.6 $25.3 75th $0.7 $1.5 $2.3 $3.1 $4.0 $4.8 $5.6 $6.5 $7.4 $8.3 $9.2 $10.1 $11.0 $11.9 $12.8 $13.8 $14.7 $15.7 $16.6 $17.6 $18.6 95th $0.7 $1.5 $2.3 $3.0 $3.8 $4.5 $5.1 $5.8 $6.4 $7.0 $7.6 $8.2 $8.9 $9.5 $10.2 $10.8 $11.5 $12.2 $12.9 $13.7 $14.4 38

Stochastic Analysis (continued) Cumulative Contributions to Date; Target Allocation The graph and table below show the range of projected cumulative contributions over the next twenty years, assuming the Plan s assets are allocated according to the Target Allocation (highlighted on the prior pages). The results assume the current contribution policy remains unchanged for all projection years. Billions $50 $45 $40 $35 $30 $25 $20 $15 $10 $5 $0 Projected Cumulative Employer Contributions to Date Target Allocation For the Plan Year Ending 5th 25th Median 75th 95th 5th $0.7 $1.6 $2.5 $3.5 $4.6 $5.8 $7.2 $8.7 $10.2 $11.9 $13.7 $15.7 $17.7 $19.9 $22.2 $24.7 $27.5 $30.5 $33.4 $36.3 $39.6 25th $0.7 $1.6 $2.4 $3.3 $4.3 $5.3 $6.4 $7.6 $8.9 $10.2 $11.7 $13.2 $14.9 $16.6 $18.3 $20.2 $22.1 $24.1 $26.3 $28.6 $30.9 Median $0.7 $1.5 $2.4 $3.2 $4.1 $5.0 $6.0 $7.0 $8.0 $9.1 $10.3 $11.4 $12.6 $13.9 $15.3 $16.6 $18.1 $19.5 $21.1 $22.6 $24.1 75th $0.7 $1.5 $2.3 $3.1 $3.9 $4.7 $5.6 $6.4 $7.2 $8.0 $8.8 $9.6 $10.4 $11.2 $12.0 $12.8 $13.6 $14.6 $15.4 $16.3 $17.2 95th $0.7 $1.5 $2.3 $3.0 $3.7 $4.4 $5.0 $5.6 $6.1 $6.7 $7.4 $7.9 $8.6 $9.2 $9.8 $10.4 $11.1 $11.7 $12.4 $13.2 $13.9 39

Stochastic Analysis (continued) Cumulative Contributions to Date; Higher Risk The graph and table below show the range of projected cumulative contributions over the next twenty years, assuming the Plan s assets are allocated according to the Higher Risk portfolio (highlighted on the prior pages). The results assume the current contribution policy remains unchanged for all projection years. Billions $50 $45 $40 $35 $30 $25 $20 $15 $10 $5 $0 Projected Cumulative Employer Contributions to Date Higher Risk For the Plan Year Ending 5th 25th Median 75th 95th 5th $0.7 $1.6 $2.5 $3.6 $4.8 $6.0 $7.5 $9.0 $10.7 $12.4 $14.3 $16.3 $18.4 $20.7 $23.3 $26.0 $28.8 $31.7 $34.8 $37.9 $41.4 25th $0.7 $1.6 $2.4 $3.3 $4.3 $5.4 $6.5 $7.7 $9.0 $10.4 $11.9 $13.4 $15.1 $16.9 $18.6 $20.6 $22.5 $24.5 $26.6 $28.9 $31.2 Median $0.7 $1.5 $2.4 $3.2 $4.1 $5.0 $6.0 $7.0 $8.0 $9.1 $10.2 $11.3 $12.5 $13.7 $15.0 $16.3 $17.6 $19.0 $20.3 $21.7 $23.0 75th $0.7 $1.5 $2.3 $3.1 $3.9 $4.7 $5.5 $6.3 $7.0 $7.7 $8.5 $9.2 $10.0 $10.7 $11.4 $12.2 $13.0 $13.8 $14.6 $15.5 $16.4 95th $0.7 $1.5 $2.3 $3.0 $3.6 $4.3 $4.8 $5.4 $6.0 $6.6 $7.2 $7.8 $8.3 $9.0 $9.6 $10.2 $10.8 $11.5 $12.2 $12.9 $13.6 40

Stochastic Analysis (continued) Cumulative Contributions to Date; 100% Equity The graph and table below show the range of projected cumulative contributions over the next twenty years, assuming the Plan s assets are allocated according to the 100% Equity portfolio (highlighted on the prior pages). The results assume the current contribution policy remains unchanged for all projection years. Billions $50 $45 $40 $35 $30 $25 $20 $15 $10 $5 $0 Projected Cumulative Employer Contributions to Date 100% Equity For the Plan Year Ending 5th 25th Median 75th 95th 5th $0.7 $1.6 $2.6 $3.8 $5.2 $6.6 $8.2 $10.0 $11.8 $13.9 $16.0 $18.2 $20.6 $23.3 $26.0 $29.0 $32.1 $35.4 $38.8 $42.4 $45.8 25th $0.7 $1.6 $2.4 $3.4 $4.4 $5.6 $6.8 $8.1 $9.4 $10.9 $12.5 $14.2 $15.9 $17.8 $19.8 $21.8 $23.8 $25.9 $28.1 $30.4 $32.9 Median $0.7 $1.5 $2.4 $3.2 $4.1 $5.0 $6.0 $7.0 $8.0 $9.0 $10.1 $11.2 $12.3 $13.5 $14.7 $15.8 $16.9 $18.1 $19.3 $20.6 $21.7 75th $0.7 $1.5 $2.3 $3.1 $3.8 $4.6 $5.2 $5.9 $6.6 $7.2 $7.9 $8.6 $9.3 $10.0 $10.7 $11.5 $12.3 $13.1 $13.9 $14.8 $15.6 95th $0.7 $1.5 $2.2 $2.9 $3.5 $4.0 $4.6 $5.2 $5.7 $6.3 $6.9 $7.5 $8.1 $8.7 $9.3 $9.9 $10.5 $11.1 $11.8 $12.5 $13.1 41

Stochastic Analysis (continued) Employer Contributions (as a weighted average percentage of salary) The tables below show the range of required employer contributions (as a weighted average percentage of salary) assuming the six different asset mixes highlighted on the prior pages. The results assume the current contribution policy remains unchanged for all projection years. 5 Years Required Employer Contribution for Plan Year Ending 2020 5th 25th 50th 75th 95th Current Allocation 20% 16% 14% 13% 11% 100% Fixed Income 19% 16% 15% 14% 13% Lower Risk 20% 15% 14% 13% 11% Target Allocation 21% 16% 14% 13% 9% Higher Risk 22% 16% 14% 12% 9% 100% Equity 25% 17% 14% 11% 8% 10 Years Required Employer Contribution for Plan Year Ending 2025 5th 25th 50th 75th 95th Current Allocation 27% 20% 16% 12% 8% 100% Fixed Income 27% 22% 19% 17% 14% Lower Risk 26% 20% 16% 12% 8% Target Allocation 27% 20% 16% 10% 8% Higher Risk 29% 20% 15% 9% 8% 100% Equity 33% 22% 15% 8% 7% 20 Years Required Employer Contribution for Plan Year Ending 2035 5th 25th 50th 75th 95th Current Allocation 39% 29% 20% 8% 7% 100% Fixed Income 42% 35% 30% 26% 19% Lower Risk 38% 28% 19% 8% 7% Target Allocation 39% 27% 17% 7% 7% Higher Risk 40% 28% 14% 7% 7% 100% Equity 45% 28% 8% 7% 7% 42

Stochastic Analysis (continued) Drawing Inferences The tables below compare the projected market funded ratios five, ten, and twenty years from now, under the median (50 th percentile), worst-case (5 th percentile), and best-case (95 th percentile) scenarios, assuming the six different asset mixes highlighted on the prior pages. The table also displays for comparative purposes the median, peak, and trough projected payout ratios and cumulative employer contributions for the six asset mixes being examined. Market in Year 5 Cumulative Employer Payout s 5 Years Contributions in Year 5 (Billions) Year 5 Years 1 to 5 50th 5th 95th 50th 5th 95th Median Peak Trough Current Allocation 80% 58% 115% $5.0 $5.8 $4.4 6% 8% 4% 100% Fixed Income 74% 60% 90% $5.1 $5.7 $4.6 7% 8% 5% Lower Risk 82% 59% 116% $5.0 $5.7 $4.5 6% 8% 4% Target Allocation 83% 57% 124% $5.0 $5.8 $4.4 6% 9% 4% Higher Risk 84% 55% 134% $5.0 $6.0 $4.3 6% 9% 4% 100% Equity 85% 47% 163% $5.0 $6.6 $4.0 6% 10% 3% Market in Year 10 Cumulative Employer Payout s 10 Years Contributions in Year 10 (Billions) Year 10 Years 1 to 10 50th 5th 95th 50th 5th 95th Median Peak Trough Current Allocation 79% 52% 130% $10.5 $13.6 $7.6 7% 10% 4% 100% Fixed Income 65% 51% 86% $11.3 $13.4 $9.4 8% 10% 5% Lower Risk 82% 54% 130% $10.4 $13.4 $7.6 6% 10% 4% Target Allocation 84% 51% 146% $10.3 $13.7 $7.4 6% 10% 4% Higher Risk 86% 49% 165% $10.2 $14.3 $7.2 6% 11% 3% 100% Equity 88% 42% 222% $10.1 $16.0 $6.9 6% 13% 2% Market in Year 20 Cumulative Employer Payout s 20 Years Contributions in Year 20 (Billions) Year 20 Years 1 to 20 50th 5th 95th 50th 5th 95th Median Peak Trough Current Allocation 82% 48% 151% $26.2 $40.1 $14.4 7% 13% 4% 100% Fixed Income 59% 45% 80% $33.2 $41.2 $24.5 10% 13% 5% Lower Risk 85% 50% 156% $25.3 $38.5 $14.4 7% 12% 4% Target Allocation 89% 48% 190% $24.1 $39.6 $13.9 7% 13% 3% Higher Risk 95% 46% 234% $23.0 $41.4 $13.6 6% 14% 3% 100% Equity 106% 39% 389% $21.7 $45.8 $13.1 6% 16% 2% 43

Appendix 1: Supplemental Stochastic Exhibits Projected Actuarial (actuarial value of assets/actuarial accrued liability); 5 Years The graph below shows the distribution of possible actuarial funded ratios five years from now, assuming the six different asset mixes highlighted on the prior pages. The results assume the current contribution policy remains unchanged for all projection years. 150% Projected Actuarial December 31, 2020 125% 100% 75% 95th 75th Median 50% 25% 0% Current Allocation 100% Fixed Income Lower Risk Target Allocation Higher Risk 100% Equity 25th 5th Current Allocation 100% Fixed Income Lower Risk Target Allocation Higher Risk 100% Equity Liability 5th $11.0 69% $10.2 71% $10.6 70% $11.4 68% $12.5 64% $15.6 56% 25th $7.0 81% $7.7 79% $6.7 81% $6.9 81% $7.1 80% $8.1 77% Median $5.0 86% $6.2 83% $4.8 86% $4.7 87% $4.6 87% $4.5 88% 75th $3.1 91% $4.8 86% $3.1 91% $2.7 93% $2.1 94% $0.3 99% 95th $0.1 100% $3.2 91% $0.3 99% ($1.4) 104% ($4.0) 111% ($12.6) 133% 44

Appendix 1: Supplemental Stochastic Exhibits (continued) Projected Market (market value of assets/actuarial accrued liability); 5 Years The graph below shows the distribution of possible market funded ratios five years from now, assuming the six different asset mixes highlighted on the prior pages. The results assume the current contribution policy remains unchanged for all projection years. 175% Projected Market December 31, 2020 150% 125% 95th 100% 75th 75% Median 50% 25th 25% 0% Current Allocation 100% Fixed Income Lower Risk Target Allocation Higher Risk 100% Equity 5th Current Allocation 100% Fixed Income Lower Risk Target Allocation Higher Risk 100% Equity Liability 5th $14.8 58% $14.0 60% $14.2 59% $15.0 57% $16.1 55% $18.7 47% 25th $10.5 70% $11.5 68% $10.2 72% $10.3 71% $10.6 70% $11.8 66% 50th $7.0 80% $9.4 74% $6.6 82% $6.2 83% $5.8 84% $5.4 85% 75th $2.2 94% $7.2 80% $2.0 94% $0.6 98% ($0.9) 103% ($4.5) 113% 95th ($5.6) 115% $3.9 90% ($6.0) 116% ($9.0) 124% ($12.9) 134% ($23.7) 163% 45

Appendix 1: Supplemental Stochastic Exhibits (continued) Projected Actuarial (actuarial value of assets/actuarial accrued liability); 10 Years The graph below shows the distribution of possible actuarial funded ratios ten years from now, assuming the six different asset mixes highlighted on the prior pages. The results assume the current contribution policy remains unchanged for all projection years. 200% Projected Actuarial December 31, 2025 175% 150% 95th 125% 100% 75% 75th Median 50% 25th 25% 0% Current Allocation 100% Fixed Income Lower Risk Target Allocation Higher Risk 100% Equity 5th Current Allocation 100% Fixed Income Lower Risk Target Allocation Higher Risk 100% Equity Liability 5th $17.1 61% $17.4 60% $16.2 63% $17.2 60% $18.4 57% $22.1 49% 25th $11.6 74% $13.9 69% $10.9 75% $11.0 75% $11.3 74% $12.8 71% Median $7.3 83% $11.3 75% $6.9 85% $6.3 86% $5.9 87% $5.3 88% 75th $2.6 94% $8.7 80% $2.1 95% $0.3 99% ($1.8) 104% ($7.9) 118% 95th ($6.8) 116% $4.5 90% ($7.0) 116% ($12.1) 126% ($18.5) 141% ($38.3) 187% 46

Appendix 1: Supplemental Stochastic Exhibits (continued) Projected Market (market value of assets/actuarial accrued liability); 10 Years The graph below shows the distribution of possible market funded ratios ten years from now, assuming the six different asset mixes highlighted on the prior pages. The results assume the current contribution policy remains unchanged for all projection years. 250% 225% 200% 175% 150% 125% 100% 75% 50% 25% 0% Current Allocation 100% Fixed Income Projected Market December 31, 2025 Lower Risk Target Allocation Higher Risk 100% Equity 95th 75th Median 25th 5th Current Allocation 100% Fixed Income Lower Risk Target Allocation Higher Risk 100% Equity Liability 5th $20.6 52% $21.4 51% $19.9 54% $20.7 51% $21.7 49% $25.2 42% 25th $14.9 66% $18.0 59% $14.3 67% $14.4 67% $14.7 66% $15.9 63% 50th $9.1 79% $15.2 65% $8.1 82% $7.1 84% $6.3 86% $5.1 88% 75th $0.9 98% $11.9 73% $0.2 100% ($2.9) 107% ($6.0) 113% ($14.1) 131% 95th ($13.8) 130% $6.5 86% ($14.7) 130% ($21.4) 146% ($28.8) 165% ($55.8) 222% 47

Appendix 2: Sensitivity Analysis: Effect of Higher Volatility This section provides a sensitivity analysis of the original stochastic projections by assuming the risk (as measured by standard deviation) of each asset class is doubled. These modified assumptions are outlined in the table below, compared to the original values: Asset Class Arithmetic Return Assumption Standard Deviation Assumption Standard Deviation Assumption Doubled 50/50 Equity 7.95 18.40 36.80 Intermediate Duration Fixed Income 3.50 6.00 12.00 Non-Core Fixed Income 6.65 12.50 25.00 Custom Real Return 6.65 8.70 17.40 Custom Real Estate 7.25 14.00 28.00 Diversified Hedge Funds 6.50 9.50 19.00 Private Equity 10.25 25.50 51.00 RVK supports the recommendations based on the original assumptions shown in the Stochastic Analysis section of this report. However, this stress-testing illustrates that potential increased capital market volatility does not change the asset allocation recommendations, based on the current status of the Plan. Instead it simply widens the range of potential results, exacerbating the potential best and worst-case scenarios. 48

Appendix 2: Sensitivity Analysis: Effect of Higher Volatility (continued) Projected Market and Maximum 1 Year Investment Loss (market value of assets/actuarial accrued liability) The tables below show the probability that the Plan will be at various funding levels for each of the six different asset mixes highlighted on the prior pages. The tables also illustrate the maximum 1 year investment loss each portfolio is expected to experience during the given time period. The results assume the current contribution policy remains unchanged for all projection years. 5 Years Probability of Full Probability of < 84% Probability of < 60% Maximum 1 Year Funding in 2020 (Current) Funding in 2020 Funding in 2020 Investment Loss Current Allocation 31% 55% 23% -47% 100% Fixed Income 11% 71% 20% -34% Lower Risk 31% 54% 22% -45% Target Allocation 35% 52% 24% -51% Higher Risk 37% 51% 27% -58% 100% Equity 41% 50% 33% -72% 10 Years Probability of Full Probability of < 84% Probability of < 60% Maximum 1 Year Funding in 2025 (Current) Funding in 2025 Funding in 2025 Investment Loss Current Allocation 34% 53% 30% -47% 100% Fixed Income 11% 77% 39% -34% Lower Risk 36% 51% 28% -46% Target Allocation 40% 50% 30% -51% Higher Risk 42% 49% 31% -58% 100% Equity 45% 48% 34% -72% 20 Years Probability of Full Probability of < 84% Probability of < 60% Maximum 1 Year Funding in 2035 (Current) Funding in 2035 Funding in 2035 Investment Loss Current Allocation 43% 47% 27% -50% 100% Fixed Income 9% 81% 47% -40% Lower Risk 44% 46% 25% -47% Target Allocation 49% 42% 25% -52% Higher Risk 52% 40% 25% -58% 100% Equity 57% 36% 26% -72% 49