3 Pervasive Retirement Industry Myths

Ours is a complex and complicated business — constantly changing and evolving. And yet, there are key fallacies about today’s retirement system — and how it compares with what used to be — that will not go away.

Back in the good old days being “covered” by a pension plan meant you would actually get a full pension benefit.

We’re routinely told that “once upon a time” individuals used to work for a single employer their entire career, and that most of those workers were “covered by a workplace retirement plan, frequently a defined benefit pension.”

While defined benefit plans were certainly more common a generation ago, they were not as ubiquitous as is often assumed (see here).  

Moreover, while some workers did spend their working career at a single employer (and some still do, particularly in the public sector), the data show that for the very most part we have long been a nation of relatively short-tenured workers. How short? Well, the median job tenure in the United States — how long workers stay at one job — has hovered around five years for the past three decades. Indeed, according to the nonpartisan Employee Benefit Research Institute (EBRI), in recent years it has ticked up, to about 5.5 years, but that’s because women are staying in their jobs longer; job tenure for men has actually been dropping.

What that means is that even workers who were “covered” by a pension plan in the private sector weren’t working with that employer long enough to get much — or any — of that promised pension benefit.

Only half of American workers have access to a workplace retirement plan.

Speaking of coverage, this is one of those statements that, while technically accurate, is somewhat misleading. Applied to all workers, that is what the National Compensation Survey (conducted by the U.S. Department of Labor’s Bureau of Labor Statistics) indicates. But it includes all workers, including very young, very low-income, part-time and part-year workers.

If you focus on full-time, full-year wage and salary workers ages 21-64, an analysis by the Employee Benefit Research Institute (EBRI) noted that in 2013, two-thirds of those workers — workers who might reasonably be expected to be covered by a voluntary workplace retirement plan under current law — did, in fact, work for an employer that sponsored a plan.

The average 401(k) balance tells us…anything.

Let’s say I told you that the average 401(k) balance in a survey sampling was $130,000 — would that be good or not?

What if I then told you that our sampling consisted of an individual who is 25 years old and has a 401(k) balance of $5,000, and an individual who is 64 years old and has a 401(k) balance of $255,000? How might that change your response? Would that tell you anything meaningful about the retirement readiness of that group?

Of course not — but surveys and coverage of those surveys routinely purport to glean a sense of retirement readiness from those kind of numbers. Despite the reality that they are comprised of savings totals for workers with a wide range of age, tenure, and savings rates — totals that are simply added together, and then divided by the number of workers in the sample.

The math on 401(k) averages is easy. The conclusions often drawn from that math, iffy. At best.

So, the next time you’re at an industry event and hear someone (who should know better) repeat one (or more) of those statements, or interviewed by a reporter who puts one (or more) of those presumptions forth as “proof” of the current system’s shortfalls, keep in mind that the data tells a different story.

And one that isn’t told often enough.

- Nevin E. Adams, JD

"Staying" Power

When we moved into our last home―an old house, and one in which the prior family had lived for quite some time―we found a set of markings on a doorframe.  Markings that appeared to indicate the height―and growth patterns over time―of the children of the former owners.  We took note of this at the time, but those markings didn’t last long after we moved in.  After all, while our kids were still growing, and they were now going to live in the same house, there was little point in assessing their progress against that of the former residents.

We’ve noted before the shortcomings of metrics such as an “average” 401(k) balance (see “Above” Average, online here ), which generally aggregate the balances of participants in widely different circumstances of age and tenure―everything from those just entering the workforce who have relatively negligible 401(k) balances with those who may have been saving for decades.  While these averages can, over time, provide a sense of the general direction in which things are moving, they’re not generally a very accurate barometer when it comes to assessing actual retirement accumulations. That’s one reason why the annual updates of the EBRI/ICI 401(k) database also report average balances by age and tenure.

That said, people change jobs, employers change 401(k) providers, and so, even in a repository as comprehensive and long-standing as the 24-million-participant EBRI/ICI 401(k) database, those “averages,” of necessity, include the experience of different individuals over time.  In order to accurately assess the impact of participation in 401(k) plans over time, and to understand how 401(k) plan participants have fared over an extended period, it is important to analyze a group of consistent participants―a longitudinal sample.

A recent EBRI Issue Brief (jointly published with the Investment Company Institute), “401(k) Participants in the Wake of the Financial Crisis: Changes in Account Balances, 2007–2011,” did just that. Drawing on the actual activity of 8.6 million participant accounts, it found that at year-end 2011, the average 401(k) account balance of the consistent group of participants during 2007‒2011 was $94,482, or 60 percent larger than the average account balance of $58,991 among participants in the entire EBRI/ICI 401(k) database. The median (mid-point) 401(k) account balance among the consistent participants was $42,082 at year-end 2011, about two-and-a-half times the median account balance of $16,649 of participants in the entire EBRI/ICI 401(k) database.

Now, in any given year the change in a participant’s account balance is the sum of several factors: new contributions; total investment return on account balances; and withdrawals, borrowing, and loan repayments.  The influence of each of those factors on individual account balances varied according to individual circumstances; participants who were younger or had fewer years of tenure experienced the largest percentage increases in average account balance between year-end 2007 and year-end 2011, and, for those younger, smaller account balance participants, contributions were responsible for a significant percentage of the growth in their account balances (consistent participants in their 20s saw their accounts nearly triple between 2007 and 2011).  Older participants experienced smaller percentage gains, and most of that was from investment returns.  Moreover, some of those balances were doubtless affected by the initiation of retirement-related withdrawals.

Still, and as the EBRI Issue Brief outlines, the trends in the consistent group’s account balances highlight the accumulation effect―the staying power―of ongoing 401(k) participation.

- Nevin E. Adams, JD

“401(k) Participants in the Wake of the Financial Crisis: Changes in Account Balances, 2007–2011” is available online here.

Covered "Call"

Sooner or later, at just about every retirement plan conference, you’ll hear someone—and generally more than just one someone—cite the statistic that “only about half of working Americans are covered by a workplace retirement plan.”

It’s a data point that is widely and openly presented as fact—not only by those inclined to dismiss the current system as inadequate (or worse), but even by some of its most ardent champions, who see it as a call to action for expanded access to these programs. It’s drawn from the U.S. Census Bureau’s March 2012 Current Population Survey (CPS).(1) But does it tell the full story?

A recent EBRI Issue Brief notes that in 2011, 78.5 million workers worked for an employer/union that did not sponsor a retirement plan. Looking specifically at those who did not work for an employer that sponsored a plan, the report notes that:

• 8.9 million were self-employed (and were thus barred from having a plan by their own inaction).
• 6.2 million were under the age of 21 (below ERISA’s mandated coverage level).
• 3.9 million were age 65 or older (beyond “normal” retirement age).
• Just over 31 million were not full-time, full-year workers.
• 16.8 million had annual earnings of less than $10,000.

Taking those factors(2) into account, it’s not hard to see why a large percentage of the 78.5 million people who worked for an employer that did not sponsor a plan—roughly 66.8 million in 2011, based on the CPS estimates above—might not be covered by a workplace plan for reasons that have little to do with the efficiency or efficacy of the current retirement plan structure.

When you filter out the overlap between those categories—situations where workers fall into several of those categories simultaneously (for example, workers who are under age 21, have less than $10,000 in annual earnings, and who are not a full-time, full-year worker)—there are about 42.4 million workers whose lack of coverage might be attributed to being in one or more of those categories. And yes, that’s more than half of the “uncovered” workers in the CPS analysis.

Indeed, while claiming that “fewer than half of working Americans have access to a workplace retirement plan” might be technically accurate, doing so exaggerates the size of the coverage “gap”—and obscures factors that might actually help explain it.

Nevin E. Adams, JD

(1) A similar result can be gleaned from the National Compensation Survey from the Bureau of Labor Statistics.

(2) There are other factors linked to rates of participation. For example, the EBRI Issue Brief also notes a correlation between firm size and participation. See Figure 30 in “Employment-Based Retirement Plan Participation; Geographic Differences and Trends, 2010.”

The "Big" Picture

A recent EBRI Issue Brief examined trends in employment-based retirement plan participation, noting that in 2011, the percentage of all workers participating in an employment-based retirement plan was essentially unchanged from the year before. More specifically, the percentage of all workers (including part-time and self-employed) participating in an employment-based retirement plan¹ stood at 39.7 percent in 2011, compared with 39.8 percent in 2010.² At the same time, the percentage of full-time, full-year wage and salary workers ages 21–64 (those most likely to be offered a retirement plan at work) saw a slight decline, slipping from 54.5 percent in 2010 to 53.7 percent in 2011.

While those movements were very small, the increase in the number of workers participating in 2011 halted a three-year decline. Moreover, it’s not as though this gauge has shown a steady trend, even in recent history. When you take into account all workers, the percentage participating in an employment-based retirement plan reached 44.4 percent in 2000, but declined to 39.7 percent in 2006, before increasing to 41.5 percent in 2007—the highest level since 2004, before then slipping back to 39.6 percent in 2009.³

When you look inside the overall numbers, we find that some categories examined had increases in the probability of workers participating and others showed decreases. Not only does the status as a full-time or part-time worker have a major impact on participation rates, the report notes, so does the demographic characteristics of the individual worker, the type and size of their employer, and even their physical location (workers in the South and West were less likely to participate in a plan than those in other regions of the country, for example).

Looking at the overall percentage of females participating in a plan, you might notice that it was lower than that of males—but when you control for aspects such as work status or earnings, the female participation level actually surpasses that of males. If you look only at the participation rate of Hispanics as a group, they appear to lag other groups significantly in terms of retirement plan participation. However, it turns out that only the nonnative Hispanics actually have participation levels substantially below those of all other workers. In fact, nonnative-born Hispanics had substantially lower participation levels than native-born Hispanics, even when controlling for age and earnings.

In responding to big issues, it’s sometimes tempting to look at data only in the aggregate, and in the press of time, to draw broad conclusions from trend lines over remarkably recent history, trying to get a sense of the “big picture.” But as the data above suggest, a “better” picture is often drawn from an analysis of the component parts that make up that big picture, balanced with an appreciation for longer-term trends.

Nevin E. Adams, JD

¹ The number of workers participating in an employment-based retirement plan increased from 60.7 million in 2010 to 61.0 million in 2011, returning to the 2009 level, the second lowest since 1997 and well below the 67.1 million workers who participated in a plan in 2000, the peak year for the number of workers participating in a plan from 1987–2011.

² “All workers” is the broadest work force population group, including those not covered by a retirement plan. A more restrictive definition of the work force, which more closely resembles the types of workers who generally must be covered by a retirement plan in accordance with the Employee Retirement Income Security Act of 1974 (ERISA), is the work force of full-time, full-year wage and salary workers ages 21–64. Under this definition, 60.8 percent of these workers worked for an employer sponsoring a plan, and 53.7 percent of them participated in a retirement plan.

³ An important public-policy topic associated with an analysis of employment-based retirement plan participation is the number of workers who are not participants, as well as the number of those who work for employers/unions that do not sponsor a plan. For example, when taking into account only workers who work full-time, full-year, make $10,000 or more in annual earnings, and work for an employer with 100 or more employees, 15.1 million (or 25.2 percent of the defined population) would be included among those working for an employer that did not sponsor a plan. Another way to look at this last number is that 74.8 percent of workers with those characteristics worked for an employer that did sponsor a retirement plan in 2011. This is explained in more detail on page 30 of the November 2012 EBRI Issue Brief, “Employment-Based Retirement Plan Participation: Geographic Differences and Trends, 2011,” online here. See also “’Under’ Covered,” online here.

'Under' Covered?

One of the more pervasive statistics bandied around about the voluntary retirement system is that only about half of working Americans are covered by a workplace retirement plan.

It’s a data point that is widely and openly presented as fact—not only by those inclined to dismiss the current system as inadequate, but even by some of its most ardent champions, who see that result as a call to action for expanded access to these programs.

There’s only one problem: It doesn’t tell the whole story.

A 2011 EBRI report found that in 2010, 77.6 million workers worked for an employer/union that did not sponsor a retirement plan and 91.0 million workers did not participate in a plan. However, focusing in on employees who did not work for an employer that sponsored a plan, 9.0 million were self‐employed.

Of the remaining 68.5 million:

• 6.2 million were under the age of 21, and
• 3.7 million were age 65 or older.
• 32.0 million (approximately) were not full‐time, full‐year workers, and
• 17.2 million had annual earnings of less than $10,000.

Now, admittedly, many of these workers would fall into several of these categories simultaneously (they might, for instance be under age 21, make less than $10,000 in annual earnings, and not be a full‐time, full‐year worker).

But if you adjust these numbers so that only workers who work full-time, full‐year, make $10,000 or more in annual earnings, and work for an employer with 50 or more employees, only 17.4 million workers (or 26.7 percent) would be included among those working for an employer that did not sponsor a plan.

Of course, another way to look at this last number is that 73.3 percent of these workers with those characteristics worked for an employer that DID sponsor a retirement plan in 2010.

And that’s a lot more than 50 percent.

- Nevin E. Adams, JD