When You Assume...

The use of assumptions was the focus of a recent report from the Employee Benefit Research Institute (EBRI), which not only highlighted the assumptions used in two separate retirement studies, but illustrated a real problem with their results.

The EBRI report examined two earlier  studies — one by Pew Trusts, the other by the Center for Retirement Research at Boston College — that purported to show that the retirement prospects for Gen Xers were worse than those of the Baby Boomers. The EBRI report not only highlighted the questionable assumptions, but took the time to quantify the impact (see “EBRI Report Calls Out Pew, CRR on Retirement Conclusions”). Such public criticisms are a rarity, as evidenced by the media coverage accompanying the release of the EBRI report.

As recently as a few years ago, I would have assumed that findings published by credentialed individuals associated with a reputable institution of higher learning could be taken at face value, if only because I assumed that if their methodologies were flawed, their findings would be taken to task by other similarly credentialed individuals.
Sadly, that does not seem to be the case.

The Pew report, though it went to some pains to determine appropriate rates of return to project out the future balances of both Boomers and Gen Xers, chose to completely ignore any future contributions.  Not so big a deal for those on the cusp of retirement perhaps, but what about those Gen Xers (defined as those Individuals born between 1965 and 1974) who are (just) 40? That’s at least a quarter century of contributions — and earnings on those contributions — completely disregarded by the Pew assumptions. Think that might tend to produce a lower retirement readiness conclusion about that demographic?

As for the report by CRR, the EBRI report explains how it relies on “wealth-to-income patterns by age group from the 1983–2010 Federal Reserve Surveys of Consumer Finances (SCF).” That certainly sounds like a credible source, but it also happens to be based on self-reported information and a perspective of 401(k)-plan designs and savings trends that pre-date the impact of automatic enrollment plan features that followed the passage of the Pension Protection Act of 2006. Not that you’ll find that acknowledged in the report, even in the footnotes. And yet data from the CRR’s NRRI are routinely cited in industry reports.

Projections about the future inevitably require some assumptions, and the EBRI report does as well. But it not only lays out its assumptions in great detail, it provides a wide range of findings associated with those various assumptions so that readers can draw their own conclusions based on the scenario(s) they think most likely to occur.

There’s an old adage that points out the inherent dangers in relying overly much on assumptions. It cautions that one should “never assume, because when you assume, you make an ‘ass’ of ‘u’ and ‘me.’”

That goes double for those who blindly rely on research findings drawn from assumptions poorly constructed and/or undisclosed.

Nevin E. Adams, JD

• See an earlier analysis of the Pew Trust report here.

• You can find a more comprehensive explanation of some of the issues associated with the CRR’s NRRI here.

Model "Hones"

After a winter filled with predictions of winter weather that never quite measured up to the forecasts, the nation’s capital (and the Southeast) finally got a taste of what those in the Midwest and Northeast have been contending with for weeks.  Not that there haven’t been close calls here, but up until this week, the multitude of factors that have to come together to produce a significant snowfall here—hadn’t.  Much to the discomfiture of those who ply their trade making such predictions, it should be noted.

While Mother Nature can be notoriously fickle, technologies like Doppler radar are able to discern with greater precision real-time activity which can be fed into sophisticated computer models that draw on the experience of the past to extrapolate a series of potential outcomes—some deemed more likely than others.  Projected outcomes that, despite the recent experience here, do an increasingly accurate job of helping each of us plan our daily commute or that vacation a couple of weeks hence.

Arguably, predictions about retirement readiness are even more complicated than weather forecasting, though there are certain similarities.  Done properly, they require a foundation in knowing the actual resources available to individuals—and an ability, through the use of sophisticated computer models, to forecast the impact of current behaviors on future outcomes.  Those models must also consider the likelihood and timing of certain external factors, and their impact on resources, and to project if (or when) those resources will run short against lifespans and circumstances as unique and variable as the American population.

Of course, those computer models require updating on a regular basis to properly account for changes in assumptions (and realities).  A recent EBRI Issue Brief[1] noted that, due to the increase in financial market and housing values during 2013, the probability that Baby Boomers and Generation Xers would NOT run short of money in retirement increased between 0.5 and 1.6 percentage points (when aggregated by age cohort), based on the Employee Benefit Research Institute (EBRI) Retirement Readiness Ratings (RRRs).[2]

EBRI’s analysis, which analyzes the major factors that cause retirement outcomes to differ, and their impact, found that, among other things, eligibility for participation in an employer-sponsored 401(k)-type plan remains one of the most important factors for retirement income adequacy.  In fact, Gen Xers in the lowest-income quartile with 20 or more years of future eligibility in a defined contribution plan are half as likely to run short of money as those with no years of future eligibility.

Not surprisingly, the report also noted that the risks of a long life and high long term-care costs drive huge variations in retirement income adequacy—and suggested that annuities and long-term care insurance could mitigate much of the variability in retirement income adequacy at or near retirement age.

Of course, while a broad-based forecast is sufficient for the local weather, and a broad-based sense of the nation’s retirement readiness is powerful fodder for purposes of setting public policy goals, individual forecasts—in order to be accurate—require more customized inputs.[3]  

Nevin E. Adams, JD

---

[1] “What Causes EBRI Retirement Readiness Ratings™ to Vary: Results from the 2014 Retirement Security Projection Model®” is available online here.

[2] EBRI’s proprietary Retirement Security Projection Model^® (RSPM), unlike many other models, takes into account a combination of deterministic expenses from the Consumer Expenditure Survey (as a function of age and income) as well as health insurance and out-of-pocket, health-related expenses, plus stochastic expenses from nursing home and home-health care (at least until the point such expenses are covered by Medicaid).  A chronology of the EBRI Retirement Security Projection Model^® is available online here.  

[3] Fortunately, individuals who want a sense of their retirement readiness can check out EBRI’s BallparkE$timate,^® along with the other materials available at www.choosetosave.org

Safety "Net"

I’m one of those travelers who absolutely dreads cutting it to the last minute. Not that I haven’t been forced to do so, from time to time, but I’m generally the one chomping at the bit to get to the airport, or to hit the highway an hour before anyone else. In my defense, on more than one occasion that “cushion” has been the difference between catching a flight or not. Planning that only considers a “best” or “normal” scenario too often overlooks the unexpected—and sometimes that margin of error is all you have.

For over a decade EBRI has modeled the nation’s potential retirement savings shortfall, and the EBRI Retirement Readiness Ratings™ provide an assessment of how many Americans are at risk of running short of money for needed expenses in retirement. In contemplating expenses, that model considers the regular expenses of living in retirement, as well as uninsured medical expenses, and the potential costs of nursing home care.

However, we have also documented and quantified the role of Social Security, defined benefit and private retirement accounts on retirement income adequacy for Baby Boomers and Gen Xers with an eye toward replacing their preretirement wages and income. While this more traditional focus on income replacement may misstate an individual’s actual post-retirement financial situation, many financial planners work with this goal as a starting point, and it can provide valuable insights particularly when—as is the case with EBRI’s projections—it is able to leverage actual 401(k) data from the unique EBRI/ICI 401(k) database, the largest such repository in the world.

Indeed, based on a recent EBRI analysis, between 83 and 86 percent of workers with more than 30 years of eligibility in a voluntary enrollment 401(k) plan are simulated to have sufficient 401(k) accumulations that, combined with current levels of Social Security retirement benefits, will be able to replace at least 60 percent of their age-64 wages and salary on an inflation-adjusted basis.

When the threshold for a financially successful retirement is increased to 70 percent replacement of age-64 income, 73–76 percent of these workers will still meet that threshold, relying only on 401(k) and Social Security combined. At an 80 percent replacement rate, 67 percent of the lowest-income quartile will still meet the threshold; however the percentage of those in the highest-income quartile deemed to be “successful” relying on just these two retirement components slips to 59 percent, reflecting the progressive nature of Social Security.

As positive a result as that seems for many, when the same analysis is conducted for automatic enrollment 401(k) plans (with an annual 1 percent automatic escalation provision and empirically derived opt-outs), the probability of success increases substantially: 88–94 percent at a 60 percent threshold; 81–90 percent at a 70 percent replacement threshold; and 73–85 percent at an 80 percent threshold.

That’s not quite the doomsday crisis scenario portrayed by many of the headlines in vogue today, though EBRI’s projections still show that a large number of Americans—even among those eligible for a 401(k) plan for 30 years—won’t be able to replace that pre-65 salary even at the various levels modeled, based on current savings patterns.

It does, however, illustrate the impact that changes in those current savings behaviors can have—and it underscores the significant role of Social Security as a vital safety net for the nation’s retirement security.

Nevin E. Adams, JD

“The Role of Social Security, Defined Benefits, and Private Retirement Accounts in the Face of the Retirement Crisis” is available online here.