The Retirement Readiness Challenge: Measure what matters
Just as employees benefit from maximizing their deferrals to receive all potential matching dollars, plan sponsors benefit from maximizing their plan designs to receive the greatest rewards. Until recently, plan sponsors did not have a reliable way to answer big-picture questions about their retirement plans: Is it doing the job it's supposed to? How many employees are accumulating adequate savings for retirement? Is the plan design effective in achieving employer objectives while maximizing the potential for participant retirement success?
Milliman's Retirement Readiness Challenge (RRC) opens the door to getting answers to these questions. Data from the RRC is the backbone of a consultative process that enables plan sponsors to accurately gauge their plan performance. Armed with this data, they can make informed decisions, helping them create better outcomes for participants that are aligned with their organization's financial goals and retirement philosophy.
Measuring what matters
In recent years, Milliman employee benefits consultants have responded to an industry trend: Plan sponsors are depending more on their service providers to assure participants’ retirement readiness. The providers' response, by and large, has been to create new tools: iPhone apps to track fund assets, investor profile quizzes, or a vast array of communications. (See below, "Who is responsible for retirement plan results?")
We believed that plan sponsors didn’t really care so much about the tools themselves; they wanted to understand if the plan was succeeding. The team at Milliman decided to quit looking at the gadgets, and focus on finding the core answer that really matters: How is the employee population doing in terms of staying on track for retirement?
Our solution was to look at each individual participant. We created a formula that combines the key factors of wealth accumulation and projects them forward to retirement age. Then, for each individual, we posed the question, Is he or she going to make it?
How Milliman calculates retirement readiness
The drivers of accumulating assets for a successful retirement are specific to each client's program—overall demographics, contribution rates, company contributions, personal investment returns, and so on. These drivers are factored into a projection formula. We then add each participant's expected Social Security benefit to arrive at an estimate of the participant's annual retirement income. This figure is divided by the participant's final salary to arrive at a percentage that represents the participant's replacement income ratio.
The target replacement ratio is somewhat subjective and varies based on individual circumstances. The range of required replacement ratios varies depending on the study and assumptions. For example, some studies, which include recent health care costs, produce a higher required replacement ratio while others assume that reduced post-retirement consumption lessens the required ratio.
Whatever the ratio, the RRC is designed to serve as a proxy for the health of the total plan, not each individual participant. To accomplish this, Milliman simply calculates the percentage of employees that are on track to achieve an 80% replacement ratio target. This is the organization's RRC. The RRC report also segments the employee population by age, service, and compensation to provide a holistic view of the retirement program.
In practice, Milliman consultants start with a baseline report of each client's current state. Then we set up a number of alternative scenarios. For example, adding an auto-enrollment feature: How much would it cost and how would it impact plan health or RRC? Because we can use either the organization's actual data or census demographic information from other Milliman clients of similar type, size, and geographic region, we are able to present clients with a reliable estimate of the cost/benefit tradeoffs of virtually any scenario.
Alignment with core plan objectives
Using RRC helps the plan sponsor or the retirement committee prudently consider its retirement philosophy for its employee population. For example, the report might present a picture showing that only 35% of employees are on track to retire with enough money. Milliman consultants work with the plan sponsors to help them make informed decisions that align with retirement philosophies for each organization.
Seeing a baseline RRC number with only 35% of employees on track for retirement necessarily provokes a series of tough questions: Are we comfortable with 35%? Or do we want to improve our score? If so, do we want to be at 100%—or is 50% adequate?
In many cases, the client's real goal is not to reach a numerical score: It's a values-based goal. For example, a company may want to design a program that will enable employees hired at age 35 or younger, who contribute to the plan at a rate of 4% to 6% over their lifetimes, to be, in most cases, on track for retirement.
In response, Milliman can help determine the plan features needed to meet that goal. For example, the employer may have to match at least 3% of pay, or realign the match to encourage a certain deferral rate, in order to achieve retirement readiness for most long-term employees. If the plan sponsor indicates that the match is too expensive, we can show them alternatives. Perhaps the RRC goal could be reached by getting employees to contribute 8% of pay to the plan, or re-mapping participants to appropriate age-based investments could be considered. Plan provisions designed to enhance participation could be discussed. Whether or not the plan sponsor decides to enhance its program depends on the organization and its philosophy. The key point is that RRC provides hard numbers that help lead to an informed decision.
Plan-specific metrics
As a practical example, we'd like to summarize data from one of our Retirement Readiness Challenges, which happens to be a company with a relatively young employee population. We used the company's actual data to calculate its baseline RRC. The key observation was that only 9% of its employees were on track to hit the retirement readiness target. To improve this result, we modeled three different options.
The first was an adjustment to its plan through the addition of a 1% annual auto-increase feature. The auto increase would increment by 1% until participants were deferring 6% of pay. We used data from companies in the same industry with a similar employee profile to project the impact. In fact, this first option proved to be an effective strategy. RRC increased to 42%, and the added cost to the company was minimal, increasing the current matching obligation less than 4% (less than 0.03% of payroll). We should point out that the cost would be significantly higher in an organization with more highly compensated employees and a more typical matching formula. This plan sponsor didn’t realize the cost benefit of this simple plan provision option until the RRC quantified the results.
The second option combined the 1% auto-increase with an auto-enrollment feature that, by default, placed all employees in the plan, contributing 2%. This option had a much higher projected cost, which increased the current matching obligation by 51% (or 0.31% of payroll). Implementing this second option would result in a plan level RRC of 62%.
Finally, the third option mapped participants' investment selections into age-appropriate asset allocations. This option comes at no explicit cost to the plan sponsor. It does, however, require significant communication to explain the rationale for the change. Our model showed that this third option would result in raising RRC to 43%.
If Option 1 (automatic deferral increases) and Option 3 (re-mapping to age-based investments) were combined, the resulting plan level RRC would exceed 55% at the minimal cost associated with Option 1. If the employer's objectives were to only look at cost-neutral solutions, this combination provides the highest increase in RRC at a very low cost.
The actual RRC report shows far more information. It breaks down the results of each option by quartile. This enables the plan sponsor to see the percentages of employees that are on track to accumulate 25%, 50%, and 75%, respectively, of the replacement income they will need to retire.
The population is also segmented by age, salary, contribution rate, and many other variables. The options that Milliman can model are virtually unlimited. For instance, we could model the results of targeted communications campaigns, based on our historical response rate. In addition to communications, we can also model a full range of plan design changes, as well as changes to investment options.
From the client's perspective, working with the RRC is an iterative process. Once the plan sponsor decides that the organization is committed to improving its score, we start identifying likely options. We collaborate with the client, putting a variety of features into the formula and modeling the costs and benefits associated with each one.
What is the right thing to do?
Ultimately, the right course of action for each program is a judgment call. Some organizations might be completely comfortable with an RRC of 35%. The plan sponsor could say, "We have a good plan. We don't have a lot of people taking advantage of it, but we can't afford to raise the match. Bottom line, it's totally up to the employees if they’re going to participate or not."
Other clients might have a sense that their brand requires more commitment to ensuring retirement security. In effect, the employer might say, "If you come to work for us and you spend your career with us, we want to make sure that we have an excellent retirement program in place and that you will utilize it appropriately. We want to ensure that you'll be able to retire comfortably, because we believe it’s the right thing to do."
Who is responsible for retirement plan results?
Since the inception of the 401(k), employee benefits providers and plan sponsors have tried a variety of approaches to engage participants. The most significant impact of the evolution from defined benefit (DB) to defined contribution (DC) plans was the transition in responsibility. In a DB plan it was clear that employers bore responsibility for results; after the transition to DC, the employer's role was merely providing tools for employees. Because employers were legally restricted from providing investment direction, throughout the 1970s and 1980s retirement program efforts were limited to education in investment basics—teaching employees about the value of pre-tax deductions and the rudiments of how to build a diversified portfolio.
In the next phase of the transition, which corresponded to the bubble years of the 1990s, employers began adding tools that employees could use to calculate how much they needed to save annually. Without expert help, participants could determine what final lump sum—the so-called "number"—would be required, and what annual rate of return they would need from their investments to reach that goal.
But as the markets corrected in the last decade, employers realized that these efforts were not producing the desired results. In response, many plan sponsors began turning to their 401(k) provider for more effective ways to manage their plans.
The bottom line is that plan sponsors are looking for methods that impact retirement readiness results. The nascent realization is that the tools themselves don't matter as much as the methodology for interacting with participants—using whatever works in getting them truly ready for retirement. This reflects a shift in thinking about who is going to be ultimately responsible for participants' successful retirement outcomes.
Milliman welcomes this market evolution as we are well positioned to help our clients optimize their plan with a clear understanding of the desired goals, the data, and the cost implications. The Milliman Retirement Readiness Challenge (RRC) is typical of the innovative way we approach our work. We look at our clients' data and the goals they want to reach, and we create analytical models that help us make sense of the information. We inventory the options available, then sit down with the client as a trusted advisor working to accomplish its goals most efficiently.
Our analysis showed that, in spite of the industry's best efforts in providing employees with tools and applications, communications, and education, a lot of participants still don't do what they're supposed to do. In other words, it has proven to be extremely difficult to educate people into retirement success.
We reasoned that if education was not working, plan sponsors needed to focus on elements of the program they can actually control that will produce successful retirement outcomes for participants. Fortunately, the Pension Protection Act of 2006 legalized provisions such as auto-enrollment, auto-increase, and age-appropriate qualified default investment alternatives that plan sponsors can utilize to influence outcomes for employees.
The Retirement Readiness Challenge can make a difference because it allows plan sponsors to see the impact of these provisions from a cost-benefit perspective. Unlike all the other tools on the market, RRC lets plan sponsors control—and take responsibility for—retirement plan results. That's a lot different than just handing tools to employees and hoping for the best.