Dear Public Retirement System Trustees, Chief Officers, and Executive Directors,
A few years ago I was speaking to a small group of educators in New England about their retirement benefits. We were talking generally about their expectations for retirement and whether or not they thought they’d have enough money to feel secure. We were nearly 10 minutes into the conversation when I realized that some of them were talking about their pension plan and others were only thinking about their supplemental 457 plans. One educator thought the pension system wouldn’t be around when he retired, similar to fears many have expressed about Social Security. Another said she didn’t think her pension would be very much because of weak investment returns. However, a third person, who had spent nearly two decades teaching in public schools said she assumed her pension would provide a decent benefit.
While each person expressed different sentiments about how much their benefits would be worth, they had one thing in common — none of them had a clear understanding of how their retirement benefits work.
I spend a lot of my time talking to public employees, and have found this lack of understanding about how retirement benefits work is very common. We know that many retirement systems offer some kind of service to explain how benefits work, but up take on this financial education is not always very strong. These teachers, police officers, and other workers that I meet are like most Americans — they don’t pay a lot of attention to the particulars of their retirement plan. They, somewhat reasonably, have a lot of other things to focus on, including their demanding jobs.
Our team at Equable Institute wanted to create a tool that would help public sector workers understand not just what benefits they have, but whether or not those benefits are serving them well. The result is our recently published Retirement Security Report, an interactive digital project that covers all major types of plans intended to provide primary retirement income, including: pensions, defined contribution plans, guaranteed return (sometimes called “cash balance”) plans, and hybrid plans. We use publicly available data about retirement plan provisions to measure the benefits offered by each plan. We then score those benefits and policies against retirement income security metrics that matter the most for individuals at various stages in their career.
The intention of the Retirement Security Report is not to compare retirement plans against each other. We recognize that different states have different priorities and goals with their retirement benefits, including varying economic conditions and enrollment levels in Social Security.
However, when we looked across the landscape of retirement benefits currently being offered to new members (as opposed to legacy plans or tiers of benefits that are closed to new members), we noticed some troubling findings.
Most importantly, the results of the Retirement Security Report indicate that the majority of individuals who enroll in a public retirement system are not being served well by their retirement benefits. Specifically, for individuals who will work up to 10 years covered by the same retirement plan, we found that only 13 out of 335 plans managed by statewide systems are putting their members on a path to retirement income adequacy. Additionally, for individuals who will work between 10 and 20 years as members of the same retirement plan, there are just 51 out of 335 plans that are serving their members well. Combined, this means that the majority of public employees that will serve 20 years or less as part of the same retirement plan are not receiving benefits that will place them on a path to a secure retirement income.
We acknowledge that this generally is not the fault of retirement systems that do not have control over benefit levels, but it is a policy challenge the leaders of retirement systems should take seriously.
The good news is that 3-out-of-4 plans are serving individuals that work a full career in the same retirement system well. This holds whether the retirement plans are pensions, defined contribution plans, hybrid plans, or some other benefit design. Generally speaking, individuals who are going to work for 25 to 40 years covered by the same retirement plan are going to wind up with quality benefits that support adequate retirement income.
But public retirement systems are not only supposed to serve individuals who will serve their full career covered by the same plan. At least, that is not what we believe at Equable. Moreover, it is typically not what most retirement systems or state legislatures say their goals are. If the objective of a retirement system is to provide all of its members with at least a path to adequate retirement income security, whether or not the members work their full career participating in that plan, then the majority of retirement systems in America are falling short.
The remainder of this letter spells out in some detail how we’ve defined our terms as part of this project and addresses what I anticipate will be some common questions. The intention is not to attack public retirement systems, but rather to provide a simple, data-driven assessment of the quality of retirement benefits that are currently being offered as a framework for both understanding the status quo and identifying areas for improvement. There are clearly quality pension plans, defined contribution plans, hybrid plans, and guaranteed return plans — but all of those plan designs can also be built in a way that doesn’t serve all participants.
I hope that the Retirement Security Report can be a jumping off point for assessing the quality of benefits being offered and whether all members of state administrated retirement plans are being served well.
How Do We Define Adequacy and the Quality of Public Retirement Plans?
Individuals and families have “retirement security” when they have access to enough income during their retirement years that they can live comfortably and meet their basic needs. Many people aspire to more than just the basics, and have a wide range of preferences about how they want to live in retirement. Expenses can vary too, depending on housing, health care costs, and dependents. But, at a simple level, retirement security is about having adequate income during post-work life.
The art and science of individual financial planning requires digging in to these complexities and having more than just a simple rule of thumb that every person should follow. However, when assessing retirement plans at a general level for all participants in the plan, straightforward and generalized targets are necessary. Therefore, we define adequate retirement income as reaching a 70% replacement of pre-retirement salary by age 67 (the normal Social Security retirement age for most individuals in the future).
The Retirement Security Report models out the benefits provided by a given retirement plan and measures the plan’s projected benefit accumulation against our benchmark for adequacy.
For example, if a pension plan has a normal retirement age of 62 and a multiplier of 2%, then an individual hired at age 25 will have reached a 74% replacement rate five years before our adequacy benchmark requires them to. If the plan has members also enrolled in Social Security we add those benefits to the accumulation too.
Naturally, lots of details matter. For example, the definition of pre-retirement salary is a meaningful part of the benchmark (for every retirement plan we use their current definition of final average salary or highest average salary; for those without a published definition we define pre-retirement income as the final year before retirement). We have also taken a lot of steps and gathered several rounds of peer feedback to ensure we’ve addressed many of these important elements to ensure that the final result is rooted in a rigorous methodology.
We recognize that no single public retirement plan is generally expected to be the only source of retirement income savings, particularly for individuals enrolled in Social Security. This is part of the reason why our adequacy benchmark is focused on a relatively reasonable 70% replacement rate (which we adjust for Social Security enrollment) and isn’t accounting for health care costs. However, most public sector retirement plans do frame themselves as the primary source of retirement income. Therefore, we think it is reasonable that participants in a public plan be able to reach this minimum standard for adequacy with their government-sponsored retirement plan plus Social Security (if eligible).
What About Other Definitions of Adequacy?
There is plenty of debate in the financial literature about what kind of retirement income replacement rates constitute reasonable targets (and even whether replacement rates are the best way to think about targets for retirement income security). Depending on whether a financial planner is counting costs for dependents, a fully paid off home, or health care costs, projected replacement rate targets tend to range between 60% and 80% — and those individuals who anticipate particularly high health care costs in retirement may be encouraged to target a 100% replacement rate.
While the baseline analysis of the Retirement Security Report uses a 70% replacement rate target, we do not believe this is truly a universal goal. It is simply a reasonable median goal. Every family has their own particular financial circumstance that could mean a modified goal. To this end we have included in our charts that measure the value of state administrated retirement plans lines that show how those benefits stack up against 60% and 80% replacement rate thresholds too.
What Plans are Analyzed by the Retirement Security Report?
This first edition of the Retirement Security Report is focused on plans and tiers of plans that are open to new members and are offered by statewide retirement systems. We will add select municipal retirement systems in the coming months, as well as some legacy plans (e.g. those with active members but that are not open to new participants).
We are also focused only on plans that are intended to be a primary source of retirement income; there are numerous supplemental plans that are not included in this analysis — though we note which systems offer such benefits.
We include all defined contribution plans, guaranteed return plans, and hybrid plans that are primary retirement plans. For pension plans, we started with the statewide plans that have at least $1 billion in actuarially accrued liabilities, but we will gradually add all statewide retirement systems in the coming months.
Are Pension Plans Really Supposed to Provide Retirement Security to Every Participant?
A common refrain that I’ve heard in my experience working with public plans is that they aren’t really designed to work for everyone because they are not supposed to be for everyone. For example, when I’ve raised a concern that individuals who only work a few years covered by a public retirement system are not served well by back-loaded benefit accumulation designs or long vesting periods on employer contributions to individual accounts, I’ve been told that these are not people that the retirement system is designed for. “Retirement benefits are a reward for career service” or “retirement plans are for those who stay” are readily available retorts.
And maybe retirement systems are not supposed to be for everyone. But if that is the case, they should not be advertised as such and should be more clearly marketed as only for select individuals. Ultimately, policymakers should be making decisions about the exact objectives and intent of the retirement benefits that their budgets fund and laws authorize. But retirement system trustees and directors have a lot of capacity to let participants in their plans know whether the benefits being offered are really for everyone.
If, on the other hand, public retirement systems really are supposed to be serving all of their members, then it is important to look carefully at different demographic groupings of individuals to assess just how many of them are being placed on a path to retirement income security. The results of our report suggest that very few plans are working well for individuals that will spend 10 years or less covered by the same plan. Further, it also indicates that even many of those who work up to 20 years are not getting enough from their retirement plans to be on a path to adequacy.
Why Are Plans for “Short-Term Workers” Getting Such Low Scores?
The methodology we’ve used involves measuring plan performance against a replacement rate metric. The benchmark asks that individuals build an equal amount of replacement rate income throughout the course of their careers. So a 70% replacement rate by age 67 gives a 25-year old 42 years to build that income, inclusive of SSA where appropriate. If SSA is going to cover 30% of the replacement rate, then the worker is aiming to build about 1% of the replacement rate income each year.
Of course, early on in a workers career people make less money than later on. So the percentage of salary contribution necessary to build that 1% of replacement rate income is larger than later on. This is why our scoring methodology does not require a retirement plan to exactly hit the benchmark every year in the first 20 years, but only get close enough to it (e.g. get at least 75% of the way to the benchmark) such that in the later years of a worker’s career the magic of compounding interest and/or the back-loaded design of a pension plan is going to enable the worker to achieve the adequacy goal.
Our approach does aim to provide a more comprehensive assessment of how well a retirement plan is serving its members than just benchmarking value of benefits as they accumulated. Plan members also should care about the eligibility of benefits, flexibility of benefits, and inflation protection of benefits. There are trade-offs related to different plan design provisions, which is why we don’t compare plan design concepts against each other but consider each on its own terms. But even when we factor all of those elements in the simple reality is that most plans in the public sector today are just not well designed for those who will only spend a portion of their career in public service.
How Do We Analyze Retirement Benefits Provided in Lump Sums?
It was important to measure all plan designs on a level playing field. Full methodological details are available in the paper we’ve published summarizing the Retirement Security Report data, but generally when it was necessary to look at account balances (natural to defined contribution and guaranteed return designs) we measured the net present value of pension benefits; and when it was necessary to look at replacement rates in annualized income terms (natural to pension plans) we annuitized the value of lump sums using conservative values.
How are the Costs of Plans Factored Into the Scoring Methodology?
The costs of plan provisions is not the focus of this project. They matter a lot in the grand scheme, of course, and there might be trade-offs that reasonably favor lower benefits for lower costs in certain jurisdictions. But the scores from this report are not intended to reflect the underlying costs of the benefits themselves. The objective here is to measure whether the benefits being offered are meeting their promises not just in a literal legal sense, but in a figurative moral sense. Are public sector workers enrolled in deferred compensation plans being called “retirement plans” actually being put on a path to adequate retirement income security?
We encourage comment and input from anyone on this first edition of the Retirement Security Report. It has been designed as a living project that will be periodically updated and expanded. Any ideas on improvements we can make to help better achieve the purpose of the project will be more than welcome. Any ways that we can help with promoting and fostering discussions within specific plans about whether or not they are truly meeting the objectives articulated and set out by the plan we will be more than happy to assist with.
Executive Director, Equable Institute