Benefit Bonds: The New Paradigm and Market Outlook for 2009 Girard Miller Senior Strategist PFM Group April 30, 2009 Bond Buyer Conference 2009 Public Financial Management and the PFM Group
Copyright and intellectual property rights The information presented here is copyrighted, proprietary data. Commercial redistribution, reproduction or re-use of this material without express written permission is expressly prohibited.
Topics Drivers of Benefit Bond Decisions Business Cycle and Benefit Bonds Depression-era Research On Relative Returns POB Trusts Bonding Examples Strategy and the Need for Legislation
POB Arbitrage: Traditional Model 4,500,000,000 Arbitrage Example 4,000,000,000 3,500,000,000 3,000,000,000 Arbitrage versus Cost of Debt and Actuarial Rate; better off by issuing debt, provides opportunity to pay off debt with excess returns Projected Wealth ($) 2,500,000,000 2,000,000,000 Arbitrage versus Cost of Debt, but earn less than Actuarial Rate; better off by issuing debt 1,500,000,000 1,000,000,000 500,000,000 No arbitrage earned; worse off by issuing debt 0 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 Time Period (years) 4
Traditional POB Model Borrow at a taxable muni rate below the pension fund s actuarial return assumption E.g, sell at 5.5% and target 7.5-8% earnings Business cycle was irrelevant if assumed POB spread was deemed sufficient Borrow as much as possible Fully fund the UAAL Invest bond proceeds in pension fund Business-as-usual 5
What Went Wrong: Timing Is Everything 1800 1600 Those who sold POBs and OPEB-OBs at market peaks were doomed at the outset 1400 1200 1000 800 600 1996 1998 2000 2002 2004 2006 2008 S&P 500 City of New Orleans POBs Oakland County OPEB Bonds New Jersey POBs Wisconsin & Contra Costa County POBs Source: Yahoo Finance S&P 500 Chart 6
Timing Is Everything Especially After Interest Costs! 1800 1600 Very few issuers have now cleared their hurdle rates 1400 1200 1000 800 600 1996 1998 2000 2002 2004 2006 2008 S&P 500 City of New Orleans POBs Oakland County OPEB Bonds New Jersey POBs Wisconsin & Contra Costa County POBs Source: Yahoo Finance S&P 500 Chart 7
GFOA Policy Statement Considerable caution advised Concerns: Market risk Debt capacity Non-traditional use of debt (not infrastructure) Reinvestment scams After market meltdown in 2008-09, risk-averse finance officers are expected to remain cautious in the use of benefits bonds, regardless of potential market opportunity. 8
The New Benefits Bonds Paradigm Business cycle analysis is essential Why sell bonds to buy bonds? Don t over-size the issue Protect investors and taxpayers Covenants on benefits increases Assign surplus earnings to debt repayment and market volatility reserves, not benefits increases Mitigate market risks The New Benefits Bonds Paradigm is premised on risk management. 9
Drivers of Benefit Bond Decisions Traditional Deal Sizing Amount to fully fund the pension Financial Analysis Current Taxable Bond Rates Stochastic Reinvestment Probability Analysis Investment Management Traditional asset allocation (60/40 stocks/bonds) Risk Mitigation None And, of course, some deals were done for budgetary reasons. 10
Drivers of Benefit Bond Decisions The New Paradigm View Deal Sizing Based on stage in business cycle Optimal recessionary sizing: 80 to 85% for POBs 65% for OPEB-OBs, preferably split in two issues Financial Analysis Primary driver of deal equity market risk/reward characteristics based on stage in business cycle Dynamic reinvestment scenario analysis 11
Drivers of Benefit Bond Decisions The New Paradigm View (cont d) Investment Management Equity-only initially, no arbitrage in selling bonds to buy bonds Time-based migration to normal asset allocation Overfunding Risk Mitigation Restrictive covenants to preclude benefits give-aways Excess returns available to pay down debt Protective POB trusts 12
The Business Cycle and the Benefit Bonds Window Benefits Bond Window 13
And Because This Isn t a Garden- variety Recession.. Historical research on equity returns vs bond yields during and after the Great Depression Mean-reversion persisted: 30 year performance spreads were the largest in all Ibbotson history, several years >10% annually Risks remained in 5- and 10-year holdings Worst case in Depression era was 20-year return from pre-crash 1929 to 49 So, the ONLY case of 30-year underperformance was period just ended in 08-09 14
And Now. PFM s proprietary Depression-era research And Implications for Benefits Bonds in the 21 st Century Here is the data: Source: Ibbotson Associates 15
What Monte Carlo Analysis Says Data source: Ibbotson Associates, Inc. 16
It Even Worked Over 20 Years Except From 29-49 and 82-02 Data source: Ibbotson Associates, Inc. 17
When Horizon Is 10 Years, Cracks In The Armor Emerge Data source: Ibbotson Associates, Inc. 18
And Major Pain In 5-Year Periods Data source: Ibbotson Associates, Inc. 19
Now, The Composite of All That Data -- In Absolute Terms Data source: Ibbotson Associates, Inc. 20
Next, The Composite In Relative Terms Data source: Ibbotson Associates, Inc. In shorter periods, stocks underperformed 1/3 of the time 21
Fixed-rate Bonds vs Equities Over 5 And 30 Year Periods Significant short-term risks Data source: Ibbotson Associates, Inc. 22
Variable-rate Bonds vs Equities Over 5 And 30 Year Periods Data source: Ibbotson Associates, Inc. 23
Variable vs Fixed: Excess Returns Over 30 Years Average Excess Return for Fixed Rate Borrower (1926 1978) = 6.2% Average Excess Return for Variable Rate Borrower (1926 1978) = 6.4% Data source: Ibbotson Associates, Inc. 24
Variable-rate Arbitrage Beat Fixed Strategy In Depression And Thereafter ( 26 78 Study) Average Excess Return for Variable Rate Borrower (1926 1978) = 6.4% Average Excess Return for Fixed Rate Borrower (1926 1978) = 6.2% Data source: Ibbotson Associates, Inc. These data support the consideration of multi-modal strategies 25
Stock Market: Relative Performance In The Depression Query: Did it matter when leverage was initiated? When was too late? Data source: Ibbotson Associates, Inc. 26
30 Year Holdings Worked Any Time In The Depression & Even Better In Early 40s Data source: Ibbotson Associates, Inc. 27
Best Results Were Long-term Investments Until 1938 Data source: Ibbotson Associates, Inc. 28
In Fact, The Depression Produced The Highest Long-term Arbitrage Windows In History! Data source: Ibbotson Associates, Inc. 29
With Superior 30-year Returns Early In 30s And Better 20-year Returns In The 40s. Data source: Ibbotson Associates, Inc. 30
And It Didn t Much Matter When One Started, Until Stocks Had Tripled From 1932 Bottom Data source: Ibbotson Associates, Inc. The higher returns in later years support a 2-issue strategy If the Great Recession analogy holds water 31
Informative Results in Early 30s: Shorter Horizon Underperformed When Prices Were Lowest Data source: Ibbotson Associates, Inc. Likewise the lower returns in earlier years support a 2-issue strategy 32 if the Great Recession analogy holds water
Implications of Depression - Era Research Record spreads may be feasible but double-dip risks must be addressed Strong case for multiple issuance rather than betting the ranch multiple windows may arise Especially with OPEB-OBs as start-ups Metrics for a reset, not a cycle Measurement from bottom, not the top GDP expansion phase would begin before reaching prior capacity peak Risks of False Positive grow each month after economy stabilizes especially in 5- and 10-yr periods Query: How do responsible financial advisors avoid market timing liability? 33
Why POB Trusts? The trillion-dollar public-pension fund conundrum Public pension funds have awarded employees and retirees a straddle option at zero cost: Heads employees win Tails taxpayers lose The put option has now cost taxpayers $1 trillion of unfunded liabilities When markets underperform actuarial assumptions, employers/taxpayers absorb amortized UAAL 100% The call option gives employees permanent, irreversible, constitutionally protected benefits increases, when markets outperform the strike 34
The Legacy of This Straddle Option: Skyrocketing Employer Costs Chart source: NASRA public funds survey 2007 Employer costs have increased 40% and employee costs remained constant Heads employees win, tails taxpayers lose 35
Remedy: The POB Trust If taxpayers underwrite the put option, then they should own the call option POB trust preserves excess returns for taxpayers To redeem bonds and reduce future costs Query: At market peaks, when plans are fully funded, which is the rational policy? Awarding permanent, irreversible, constitutionally protected benefits increases that increase liabilities, or Permanently reducing liabilities and costs to taxpayers -- who underwrite all risks of public defined benefits plans 36
Deployment of Proceeds: POB Trust Fund Model Place proceeds in a separate trust Generally outside pension fund Invested primarily in equities Over life of bond issue, migrate toward normal mix Limit allowed uses of Trust assets/earnings: Contribution to pension fund Pay off trust-related bond principal Avoids employee/retiree raids on overfunding Requires actuarial cooperation Preferable to treat POB trust as a valuation asset Otherwise may require a cash mgmt account for annual payments to pension fund, which reduces long-term returns 37
Deal Structure: Options Fixed rate callable versus non-callable Fixed rate level debt versus level principal COPs: when traditional debt is restricted Variable rate Medium Term Notes COPs: when traditional debt is restricted Multi-Modal Availability of call feature is a key structural decision as well as historical research on variable costs vs equity returns 38
Deal Structure: Multi-modal Example 120 100 Uniform to Fill Constraint $ Millions 80 60 40 20 This white space is the future - value savings - 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 Series 2009A Series 2009B Series 2009C Interest and Ongoing Costs Amortization Fill Constraint 39
Legislative Challenges and Solutions Many states lack OPEB authority generally Bonds, trusts and investments Some states have pension and OPEB investment limitations (e.g., maximum % in equities) May be compelled to sell bonds to buy bonds by asset allocation requirements in state law Even worse, upper Midwest laws promote issuing bonds to raise taxes and invited dubious arbitrage strategies Needed: broader legislation Solution: Model Legislation See Girard Miller s Governing.com column: http://www.governing.com/articles/0902gmillere.htm 40
Summary 2009 Market Cycle will drive POB and OPEB Bonds Traditional POB model now in question PFM s New Paradigm approach more pragmatic, thoughtful and prudent Advance preparation is critical because of uncertain timing of window of opportunity with a limited duration When will taxable municipal market reopen? How quickly will economy bottom out/recover? 41
Benefits Bonds Wisdom Those who fail to learn from history. are doomed to repeat it - George Santayana and Winston Churchill 42
Girard Miller s Contact Info and Governing Columns Link Phone 310.795.1354 Email: girardinmalibu@charter.net or millerg@pfm.com Governing.com: Column index link: http://www.governing.com/articles/gmiller.htm 43