STUMP » Articles » Wisconsin Wednesday: Pew Research on Its Ability to Weather Bad Experience » 30 May 2018, 12:19

Continuing the look at the main Wisconsin pension plan — a risk-sharing plan that seems to have performed fairly well compared to other U.S. pensions.

Recently, a paper came out: Assessing the Risk of Fiscal Distress for Public Pensions: State Stress Test Analysis

Greg Mennis, Susan Banta, and David Draine

THE PEW CHARITABLE TRUSTS
901 E STREET NW
WASHINGTON, DC 20004

May 2018

This paper summarizes the results of a stress test simulation analysis on the largest government pension plans in 10 states under different economic scenarios and assumptions for policymaker behavior. The simulation model applies both deterministic and stochastic methods. The model also uses state-specific actuarial projections and revenue forecasts, as well as a common set of capital market assumptions. Results are calculated for a variety of actuarial and financial measures over 30 years, with particular attention to downside economic scenarios and different assumptions on how officials are likely to respond to resulting increases in pension costs based on past behavior. We find that poorly funded plans face the risk of unfunded liabilities and high costs, and in some cases, insolvency under scenarios where returns are lower than expected.

Conversely, we find that states with well-funded pension systems have achieved this result through a combination of fiscal discipline and adherence to policies specifically designed to manage the impact of market volatility in low-return scenarios. Finally, stochastic analysis of the state plans reveals that contribution policy — in addition to the timing of investment returns — is a significant factor influencing cost variability. We conclude that stress testing is not just an academic exercise, and recommend that it be a standard reporting practice for all public-sector retirement systems.

The various news items being published on this study have focused on the failing pensions (i.e., New Jersey) – I will get to that at the end of this post.

But they did look at the very-much-not-failing pensions of Wisconsin.

STRESS-TESTING WISCONSIN PENSIONS

Let’s look at what they found out about this particular state.

North Carolina and Wisconsin: Policies to Manage Cost and Volatility

Retirement systems in North Carolina and Wisconsin have consistently ranked among the best-funded
public pensions in the U.S., with 2016 reported funded ratios of 88 percent and 99
percent, respectively.⁵⁴ For this reason, North Carolina and Wisconsin are at a minimal risk of
fiscal distress. Both have maintained a commitment to fiscal discipline and have enacted policies
that strengthen the financial position of their public plans. However, the two states have
achieved fiscal health through different paths, which are reflected in different projected results
under stochastic analysis.

….
Wisconsin

Wisconsin has also demonstrated a commitment to making full actuarial contributions by
following a funding policy that calls for addressing unfunded liabilities more quickly than most
states.⁵⁶ In addition, the state’s unique risk-managed defined benefit plan mitigates most of
taxpayers’ exposure to market fluctuations. Specifically, the plan adjusts employee contributions
to account for unplanned costs, as well as sharing gains if investments outperform. The plan also
increases and decreases its post-retirement annuity benefits based on returns and funding
status.⁵⁷

Here are footnotes 56 & 57:

[56] Wisconsin calculates pension contributions using a funding method that amortizes unfunded liabilities over the future payroll of active employees, which effectively results in paying off pension debt faster than the 25- to 30-year amortization periods typically used among public pension plans.

[57] The Pew Charitable Trusts. (2017). Cost-Sharing Features of State Defined Benefit Pension Plans: Distributing risk can help preserve plans’ fiscal health. Available at http://www.pewtrusts.org/en/research-and-analysis/reports/2017/01/cost-sharingfeatures-of-state-defined-benefit-pension-plan

So here’s the deal — the Wisconsin plan has a large component of employee contributions to the pension.

EMPLOYEES SHARE A LARGE PART OF FUNDING BURDEN IN WISCONSIN

Here’s a histogram from all the public plans in the Public Plans Database, for Fiscal year 2016:

Wisconsin’s share for employees is just under 50% (I believe officially it’s 50/50, but there can be timing issues). The average for all the pensions in the database is about 26% (weighted by plan size).

Given that Wisconsin explicitly counts on employee contributions, they can’t make the usual dodge of thinking “30 years amortization is just fine because government is long-term” — if you’re dependent on the employee to pay down the shortfall, then you need to consider how many years those employees will be around to contribute. It would be interesting to see how well other plans with high employee contribution %s do.

LESS VOLATILITY FOR CONTRIBUTIONS IN WISCONSIN

Here is a key graph from the research paper:

You can see that Wisconsin has less variability for employer contributions. Because of the close tie between employer and employee contributions, it means less variability for employees as well.

WISCONSIN IS SPECIAL

The comment from the researchers:

The efficacy of risk-management policies in Wisconsin are further reflected in the detailed results
for all simulations included in Appendix II. The state’s results across our metrics are striking. For
example, the employer contribution rate does not exceed 25 percent of payroll in virtually any
year under any simulation; and the funding level drops below 80 percent in any forecasted year
in less than half of the simulations. Additionally, the largest year-over-year increase in
contribution rates across any of the 10,000 simulations we modeled under state policy was less
than 4 percent of payroll. However, Wisconsin’s policy does result in more variable outcomes for
workers — *employee contributions will rise or fall depending on plan funding, and investment
performance will determine the size of COLAs provided to retirees, or whether COLAs are
provided at all.*

oooOOOOoooo, do they make the fabled mistake of the 80 percent?

I’m not addressing that one today, but just something to think about it.

Adjusting both benefits and contributions in the face of adverse experience helps reduce problems.

Unfortunately, many/most U.S. public pensions have constraints on whether the public employees (foreget about the retirees) can have these adjustments made on either contributions or benefit levels.

We’ve seen COLA reductions for current participants (forget retirees) shot down in Illinois and New Jersey, but it seems to have stuck in Rhode Island.

We’ve seen greater employee contributions for current participants shot down in many states.

So Wisconsin’s example may not be all that helpful for current policy-makers… unless they realize what it would take to get from where they currently are, to where they’d like to go.

Still, it’s good to see that others are noticing the sustainable nature of pensions where all parties have some skin in the game.

When it’s expected that taxpayers are going to bear all the risk, and the public employees all the reward… there are problems.

OTHER COVERAGE OF PEW RESEARCH

I may pull out the not-so-good results to look at, but in the meantime, here are some items on the report:

• CIO: US Retirement Systems ‘More Vulnerable Than Ever’
• John Bury: NJ Stress Test Study
• CBS/AP: Study: Some public pensions funds could run dry in downturn

I will point out that these look at the failing pensions… and I want to say: some of these funds could run dry even without a market downturn.

I did a quick check of some of the “drop dead” dates against my own projections, and found that they were qualitatively the same. So yay.