Underfunded State Pensions The Size of the Problem, the Obstacles to Reforms, and Potential Paths Forward

Transcription

1 Underfunded State Pensions The Size of the, the Obstacles to Reforms, and Potential Paths Forward October 13, 2011 Thomas J. Healey & Carl Hess

2 Underfunded State Pensions Size of the Asset Values, Liabilities, & Unfunded Liabilities Valuation Method Controversy Causes of the Obstacles to Reform Potential Paths Forward Benefit Creep Loss in Market Value of Assets Political Hurdles Legal Hurdles Costing Principles of Retirement Plans Successful Reformers Benefit & Financing Design Changes 2

3 Underfunded State Pensions Size of the Asset Values, Liabilities, & Unfunded Liabilities Valuation Method Controversy 3

4 Trillions Determining Assets: Actuarial vs. Market Values $3.0 $2.5 $2.61 $2.46 $2.0 $2.01 $1.94 $1.5 $1.0 $0.5 $0.0 Assets of US Social Security Trust Fund Wilshire Consulting Valuation of State Pension Plans' Assets (Actuarial Value) Wilshire Consulting Valuation of State Pension Plans' Assets (Market Value) Novy-Marx and Rauh Valuation of State Pension Plans' Assets (Market Value) 4 Size of the

5 Market Value of Public Pensions Has Plummeted Market Value Funding Ratio of 126 Public Pension Plans 100% 90% 80% 70% 60% 50% 5 40% Size of the SOURCE: Wilshire Consulting Wilshire Report on State Retirement Systems: Funding Levels and Asset Allocation.

6 Liabilities: Determining Present Value $1,000 $900 Accrual of a $1,000 pension benefit payable at age 70 $800 $700 $600 $500 $400 4% 8% 6 Size of the $300 $200 $100 $ Age

7 Trillions Present Value of Liabilities Varies by Method of Calculation $10 $9 $9.16 $8 $7 $6 $5 $5.17 $4 $3 $3.11 $2.97 $2 $1 7 Size of the $0 PV of US Social Security Obligations Novy-Marx and Rauh PV of State Pension Plans' Liabilities Under Treasury Rate Wilshire Consulting PV of State Pension Plans' Liabilities Using Government Assumptions Novy-Marx and Rauh PV of State Pension Plans' Liabilities Using Government Assumptions

8 Trillions Unfunded Liabilities: $Billions or $Trillions $7 $6.55 $6 $5 $4 $3 $3.23 $2 $1 $1.10 $1.03 $ Size of the $0 PV of US Social Security Unfunded Obligations Novy-Marx and Rauh PV of State Pension Plans' Unfunded Liability Under Treasury Rate Wilshire Consulting PV of State Pension Plans' Unfunded Liability Using Market Value of Assets Novy-Marx and Rauh PV of State Pension Plans' Unfunded Liability Using Government Liability Assumptions Wilshire Consulting PV of State Pension Plans' Unfunded Liability Using Actuarial Value of Assets

9 Largest Unfunded Liabilities Based on Stated Liabilities California: $154.2 billion unfunded 68% funded Illinois: $85.4 billion unfunded 43% funded Ohio: $75.3 billion unfunded 61% funded New Jersey: $62.9 billion unfunded 49% funded Texas: $53.7 billion unfunded 70% funded Based on Treasury Yields California: $475.7 billion unfunded 41% funded Illinois: $219.1 billion unfunded 23% funded Ohio: $216.9 billion unfunded 35% funded Texas: $188.2 billion unfunded 40% funded New York: $166.4 billion unfunded 53% funded 9 SOURCE: Novy-Marx and Rauh. The Liabilities and Risks of State-Sponsored Pension Plans Size of the

10 Worst Funded Ratios Based on Stated Liabilities Based on Treasury Yields Oklahoma: 37% funded $20.3 billion unfunded Arkansas: 39% funded $12.7 billion unfunded Indiana: 43% funded $20.9 billion unfunded Illinois: 43% funded $85.4 billion unfunded Connecticut: 48% funded $22.4 billion unfunded Arkansas: 21% funded $30.2 billion unfunded Oklahoma: 22% funded $42.7 billion unfunded Rhode Island: 22% funded $21.1 billion unfunded Illinois: 23% funded $219.1 billion unfunded Indiana: 25% funded $46.9 billion unfunded 10 SOURCE: Novy-Marx and Rauh. The Liabilities and Risks of State-Sponsored Pension Plans Size of the

11 Five Cities with the Worst Pension Underfunding Per Household Net Pension Assets ($B) Stated Liability ($B) (% funded) Liability using Treasury Yield ($B) and (% funded) Unfunded Liability/ Household (Treasury yield) Chicago $2.2 $46.3 (47%) $66.6 (32%) $41,966 New York City $92.6 $155.8 (59%) $214.8 (43%) $38,886 San Francisco $11.9 $16.3 (73%) $22.6 (53%) $34,940 Boston $3.6 $7.4 (49%) $11.0 (33%) $30,901 Detroit $4.6 $8.1 (57%) $11.0 (42%) $18,643 SOURCE: Novy-Marx and Rauh. The Crisis in Local Government Pensions in the United States Calculations performed in June, Size of the

12 Five Cities with the Worst Pension Underfunding Per Household Net Pension Assets ($B) Stated Liability ($B) (% Funded) Liability using Treasury Yield ($B) and (% Funded) Unfunded Liability/ Household (Treasury Yield) Chicago $2.2 $46.3 (47%) $66.6 (32%) $41, Size of the New York City $92.6 $155.8 (59%) $214.8 (43%) $38,886 A resident in Chicago has to worry not only about the city s unfunded pension liability of $41,966 per household, but also the State of Illinois unfunded pension liability of $46,152 per household. This leaves each Chicago household responsible for $88,118 in unfunded public pension liabilities. SOURCE: Novy-Marx and Rauh and US Census San Francisco $11.9 $16.3 (73%) $22.6 (53%) $34,940 Boston $3.6 $7.4 (49%) $11 (33%) $30,901 Detroit $4.6 $8.1 (57%) $11 (42%) $18,643

13 Different Accounting Practices Yield Dramatically Different Results Historical GASB Approach Future liabilities are discounted at a rate of anticipated return on pension fund assets, currently 8%. Plans report asset values as an actuarial average over several years, mitigating effect of market upheavals (a technique known as smoothing ) Advocates argue that a discount rate based on a projected return is appropriate for governments always able to meet their financial obligations. Fair Value Method Future liabilities are valued at whatever price an insurance company operating in a competitive market would charge to assume responsibility for those obligations. Assets are valued at whatever amount investors are willing to pay for them (this lack of smoothing creates wide variation in asset values). If taxable municipal or Treasury bond yields are used to discount liabilities under this method, the rate becomes roughly half the 8% rate used under the GASB approach (dramatically growing the size of liabilities). Advocates argue that because pension payments are such a sure thing, they should be discounted with a riskless or near riskless rate (e.g., the Treasury rate or taxable municipal bond rate). 13 Size of the

14 Opinions on Discount Rates The annualized return for the past eighty years on a plain vanilla 60/40 domestic stock/bond portfolio was 8%. This period includes the financial crisis of the 1930s as well as the credit debacle in Gary Findlay MOSERS There is no professional disagreement [among economists]. The only appropriate way to calculate the present value of a very low-risk liability is to use a very low-risk discount rate. Donald L. Kohn Vice Chairman, Federal Reserve 14 Size of the

15 Underfunded State Pensions Size of the Asset Values, Liabilities, & Unfunded Liabilities Valuation Method Controversy Causes of the Benefit Creep Loss in Market Value of Assets 15

16 Public Employees Have Higher Total Compensation than Private Employees $75,000 Average Compensation in Real 2010 Dollars $70,000 $65,000 $60,000 $55,000 $50, US Private Industry Employees 2010 State and Local Government Employees 16 Causes of the SOURCE: Salaries are calculated by taking the Bureau of Economic Analysis (BEA) figures for compensation of employees by industry (Table 6.2D) and dividing it by BEA s full-time equivalent employment by industry figures (Table 6.5D). The resulting salaries are then adjusted to 2010 dollars through the use of purchasing power adjustment. See

17 Comparing US Inflation to 3.5% COLA 6.0% 5.0% 4.0% 3.0% 2.0% 1.0% 0.0% % US inflation SOURCE: US Inflation Calculator. Historical Inflation Rates: , Causes of the Pension COLA

18 Benefit Creep: Hidden Benefit Increases Pension padding (final year overtime) Special benefits (NYC Christmas bonuses) Special disability pensions Double dipping 18 Causes of the

19 03/31/ /30/ /31/ /30/ /31/ /30/ /31/ /30/ /31/ /30/ /31/ /30/ /31/ /30/ /31/ /30/ /31/ /30/ /31/ /30/ Value of S&P 500 Index Loss in Market Value of Assets $1, $1, $1, $1, $ $ $ Value of S&P 500 Index 100% 95% 90% 85% 80% 75% 70% 65% 60% 55% 50% Market Value Funding Ratio of 126 State Pension Plans SOURCES: Standard & Poor's (S&P 500): Wilshire Consulting Wilshire Report on State Retirement Systems: Funding Levels and Asset Allocation. 19 Causes of the

20 The Effects of Benefit Creep $90,000 $80,000 $70,000 $60,000 $50,000 $40,000 $30, Causes of the Benefits annually increased at rate of 3.5% COLA Benefits annually increased at historical rate of inflation SOURCE: US Inflation Calculator. Historical Inflation Rates: , 2 September 2, 2011,

21 Underfunded State Pensions Size of the Causes of the Obstacles to Reform Asset Values, Liabilities, & Unfunded Liabilities Valuation Method Controversy Benefit Creep Loss in Market Value of Assets Political Hurdles Legal Hurdles 21

22 Political Pressure to Increase Pension Benefits Public pension plan sponsors face intense political pressure to maintain the status quo or to increase plan benefits. The pressure is greatest when government coffers are full and pension plans appear to be fully funded. 22 Obstacles to Reform

23 Public Employee Unions Spend Big on Elections American Federation of State County & Municipal Employees (AFSCME) reportedly donated $87.5 million to candidates in 2010 elections (Wall Street Journal, October 10, 2010). Biggest spender outside of political parties. In Wisconsin, changes in collective bargaining laws (including increased state employee pension contributions) triggered massive protests and recall elections during which an estimated $40 million was spent (Reuters, August ). There were only 9 recall elections, but spending doubled the total of Wisconsin s 116 state legislative races in 2010 (Appleton, Wisconsin Post Crescent, August 7, 2011). 23 Obstacles to Reform

24 Legal Obstacles to Public Pension Reform Public Pensions as Contracts Constitutional protection of past and future benefit accruals: Alaska, Arizona, Illinois, and New York Constitutional protection of past benefit accruals: Michigan, Hawaii, and Louisiana Non-Constitutional contract protection of past and future benefit accruals: California, Kansas, Massachusetts, Nebraska, Oregon, Vermont, Washington, and West Virginia Non-Constitutional contract protection of past benefit accruals: Arkansas, North Carolina, and Oklahoma Public Pensions as a Gratuity or Property Interest Gratuity: Indiana and Texas Property Interest: Connecticut and New Mexico 24 Obstacles to Reform SOURCE: Monahan, Amy B. "Public Pension Plan Reform: The Legal Framework

25 Recent Court Decisions: A Harbinger of Things to Come? Recent court decisions in Colorado and Minnesota upheld public pension reforms enacted by state legislatures and governors. Neither state has the most restrictive legal approach to pension obligations, but both have traditionally viewed plan benefits as either contracts or legally enforceable promises. These Court decisions convey a sense of possibility to reform-minded states, even those that give contractual-type protection to accrued public pension benefits. 25 Obstacles to Reform

26 The Need for Reform Is Pressing Pritchard, Alabama 2009: First city in the US to default on its public pension obligations Pension beneficiaries now receive only $357/month in benefits Central Falls, RI The city declared Chapter 9 bankruptcy in August, 2011 Unfunded liability of $80M Court-appointed receiver Illinois If IL used Treasury rate to discount: 77% underfunded $284.8B in total pension liabilities (which constitutes 717% of Illinois annual tax revenue) 26 Obstacles to Reform

27 Underfunded State Pensions Size of the Asset Values, Liabilities & Unfunded Liabilities Valuation Method Controversy Causes of the Obstacles to Reform Potential Paths Forward Benefit Creep Loss in Market Value of Assets Political Hurdles Legal Hurdles Costing Principles of Retirement Plans Successful Reformers Benefit & Financing Design Changes 27

28 Calculating a Pension Benefit Typical Retirement Formula Retirement benefits: Based on final average compensation (FAC) and years of credited service; the maximum benefit is 75% of FAC FAC is calculated by taking the last 12 months of salary Participants receive 2% of FAC for each year of service COLA: Benefits increase by 2.5% annually, effective on January 1 following the first full year of retirement Benefit Calculation: FAC of $5,929; retired at age 60; 30 years of service. (30 years of service) (2%) = 60% 60% ($5,929 FAC) = $3,557 $3,557 (12 months) = $42,688 annual pension benefit (with annual 2.5% COLA) 28 Potential Paths Forward

29 Costing Principles of Retirement Plans 1 Costs should be apportioned over beneficiaries working lifetimes in a reasonable relation to the value of the deferred compensation Risk must be explicitly priced across time periods. All benefits will be payable when due. There must be a catch-up mechanism for costs not met. There must be a catch-up mechanism for experience different than assumed. 29 Potential Paths Forward

30 Costing Principles Illustrated $1,000 $900 Accrual of a $1,000 pension benefit payable at age 70, funded from age 50 to 65 $800 $700 $600 $500 Accum 4% Accum 8% Accum 4% $400 $300 $200 $ $0 Potential Paths Forward Age

31 Successful Reforms Utah New Jersey Colorado Atlanta, GA Detroit, MI Federal Government Reduce, Suspend, or Eliminate COLA Creation of Hybrid DC/DB Creation of Just DC Plan Closing DB Plan to New Employees Push Back Normal Retirement Push Back Early Retirement (or Penalize) Increase Employee Contributions Increase Employer Contributions Creation of Target Funding Ratio Decrease Benefit Calculation (smaller final average compensation or multiplier) (Optional) 31 Potential Paths Forward

32 The Utah Example The 2008 financial crisis blew a 300% hole in the pension system With a $6.5 billion unfunded pension liability, Utah requested that state actuaries project out pension costs for 40 years The state requested actuarial projections considering market returns of 6%, 7%, 7.5%, and 8.5% Modeled scenarios included: Standard option (increase contribution rates) Do-nothing option (freeze contribution rates at existing levels) Delay option (freeze contribution rates for 3 or 5 years, then increase contribution rates) Regardless of the scenario, the data showed the state could not simply grow out of its problem SOURCE: Utah State Senator Dan Liljenquist and Utah: A Case Study for Pension Reform, The Pelican Post Potential Paths Forward

33 The Utah Solution Closed existing pension plan to new participants New employees choose between a defined contribution plan and a hybrid defined benefit/defined contribution plan State contribution to the hybrid plan is capped at 10% of payroll; participants must fund the balance of the Annual Required Contribution No defined benefit increases until system is 100% funded SOURCE: Utah State Senator Dan Liljenquist and Utah: A Case Study for Pension Reform, The Pelican Post Potential Paths Forward

34 Federal Government as an Example Federal Employees Retirement System (FERS) was enacted in 1986 FERS supplemented Social Security with a basic annuity plan (Defined Benefit) and a thrift-savings plan (Defined Contribution) for all federal employees hired after 1983 The older Civil Service Retirement System (CSRS) remained in effect as a pension for federal employees hired before 1984 who chose not to switch to FERS 34 Potential Paths Forward

35 Potential Benefit Design Changes 1) Increase final salary averaging period to five years Lowers the cost of funding by 1.3% of payroll Overall cost reduction of 5% 2) Change to the practice of averaging pay over the course of a career Lowers the cost by 7.4% of payroll Overall cost reduction of 33% 3) Raise the retirement age to 65 Lowers the cost by 5.7% of payroll Overall cost reduction of 25% 4) Eliminate COLA Lowers the cost by 5.3% of payroll Overall cost reduction of 24% 5) 5% employee contribution Each 1% of employee salary contribution lowers taxpayer cost by 1% of payroll A 5% employee contribution lowers taxpayer cost by 5% Enacting Changes 2 5 in combination lowers the cost by 19.1% of payroll, an overall cost reduction of 85% for taxpayers. 35 Potential Paths Forward

36 Cost Reduction Contribution of Potential Benefit Design Changes 90.00% 80.00% 70.00% 60.00% 50.00% 40.00% 30.00% 20.00% 10.00% 0.00% 9.78% 32.89% 22.22% 20.00% Eliminate the COLA Change to the practice of averaging pay over the course of a career Employee contributes 5% of salary Raise the retirement age to 65 Total Cost Reduction of 85% 36 Potential Paths Forward

37 Potential Financing Changes Eliminate the intergenerational risk transfer. ARC calculations could be based on: (1) A risk-free discount rate for liabilities associated with current retirees; and (2) A discount rate reflecting the asset allocation invested to match the liabilities associated with current workers (i.e., future retirees). Require amortization of deficits over reasonable periods. On the basis of an individual participant, amortization should not last much longer than an employee s remaining work life. Make payment of the Actuarially Required Contribution a legal requirement. Fix and monitor the size of a pension plan s funding ratio. Funding ratio should be monitored through required stress-testing in actuarial valuations. 37 Potential Paths Forward

38 38 Financing Changes, Illustrated $1,000 $900 $800 $700 $600 $500 $400 $300 $200 $100 $0 Potential Paths Forward Accrual of $1,000 pension benefit payable at age 70, funded from age 50 to 65 Accum 4% Accum 8% Accum 4% Set funding targets to appropriate levels Fund shortfalls over reasonable periods Age

39 Key Takeaways 1. The problem is huge trillions, not billions 2. Significant obstacles to reform Political hurdles Legal hurdles Individual solutions required in 50 states 3. Potential ways to reduce deficits through benefit changes: Raise retirement age Change accrual rates and pensionable pay Reduce/eliminate cost-of-living adjustments 4. Potential to address unfunded liabilities through financing changes: Better costing Increased employer contributions Increased employee contributions 39 Conclusion

40 Questions? Contact Kevin Nicholson, 40