Are Americans’ Retirement Plans Doomed, Or Is This Survey A Bunch Of Baloney?

This is where Twitter works reasonably well: Virginia Postrel linked to a blog post in the Washington Post, titled “Almost 20 percent of people near retirement age have no retirement savings,” I followed the link, which itself linked to the study done by the Federal Reserve, “Report on the Economic Well-Being
of U.S. Households in 2013,” so now I have something interesting to write about. Postrel also asks whether the doom-and-gloom results are really any different than in the past, which is also a topic I ought to know something about, and combines my two favorite sorts of topics: looking at numbers and at history.

But, as I looked at this report, some of the results also just did not make a lot of sense.

So here we go:

This report consists not of analysis of data but of a survey commissioned by the Federal Reserve Board. This is an entirely new survey, so there isn’t any directly comparable historical data to reference. The data was collected by sending out e-mails to a randomly selected group of individuals from “KnowledgePanel” — which claims to be a top-notch way of randomly selecting individuals. (Digging a bit, it seems that this is a market research company that pays prospective survey-takers on a per-survey basis, and offers other prizes and bonus points; an article here claims that they’re a “scam” based on the fact that his experience was that over time they sent increasingly long surveys for the same per-survey payment, and that one of the prizes is a spin on the “prize wheel,” which the blogger says is a fixed never-win game.) In this case, participants were paid $10 for the completed survey, and the survey had about a 60% return rate, with 4,134 completed surveys. The responses were also adjusted to correct for non-representativeness of respondents.

And maybe this method of conducting the survey explains some of the results.

Let’s start with the headline figure: 20% of Americans between the ages of 55 and 64 have no retirement savings. The odd thing is that this is a figure from the executive summary but it doesn’t appear in the actual report, which splits the data into 45 – 59 and 60+. But that figure makes no sense in any case. The question was, “What type(s) of retirement savings or pension do you (or your spouse/partner) have?” and the answers/answer choices were:

18–29 30–44 45–59 60+ Overall
No retirement savings or pension 50.5 27.8 23.0 15.4 30.9
Social Security Old-Age benefits 17.5 31.5 46.4 67.6 36.3
401(k), 403(b), thrift or other defined contribution pension plan through an employer
30.3 52.8 47.9 37.1 43.7
Defined benefit pension through an employer (i.e., pension based on a formula, your earnings, and years of service)
7.0 16.0 27.0 25.9 18.2
Individual Retirement Account (IRA) 11.2 23.5 29.2 31.9 23.0
Savings outside a retirement account (e.g., a brokerage account, savings account)
15.4 19.3 28.6 33.3 22.7
Real estate or land 4.4 8.9 16.2 20.5 11.3
Other 1.7 3.4 4.1 4.1 3.2
Total number of respondents 3,163

(Sorry for the lousy formatting.)

So the first thing that stands out is that nearly everyone ought to have Social Security benefits — only a very small number of people aren’t eligible either on their own earnings record or a spouse’s earnings record. (I thought at first that they might be correcting to include illegal immigrants, but that seems unlikely.) Does the fact that 2/3s of the oldest group said they have Social Security benefits but only 17.5% of the youngest said so, mean that the youngest group doesn’t believe Social Security will be around? I’d say that the existence of Social Security as a type of retirement benefit didn’t occur to them, but it was an answer choice, with the instruction to select as many as were applicable. In any case, the invalidity of the Social Security answers makes the “no savings/pension” responses invalid.

What we’re left with are very odd responses for the other items as well. It’s true that in my work, I deal with medium-sized to large employers (or maybe large to very large, depending on the label), but the 25.9% for older employers and 7.0% for the youngest both seem very small, even when recognizing people with part-time and erratic work histories. At their heyday, DB pensions were the norm; as recently as 2 decades ago, 80% of the medium to large employers who are the clients at my firm provided them, and, though that figure has shrunk to maybe 25%, older workers were generally grandfathered (that is, continuing to accrue benefits), or, if not, by law their existing accruals were protected. Even now, public sector workers (except, if I’m not mistaken, certain federal government workers) continue to receive pensions, and they’re a significant portion of the overall workforce. Besides which, the question asks whether the respondent or his/her spouse/partner has such a benefit which ought to increase the affirmative responses significantly compared to just asking about the respondent.

Also note that not everyone responded. Was this just that people were allowed to send in half-completed surveys, and this came late enough in the survey that people got tired of the questions? Or were the no-answers really additional “no savings” responses?

(That being said, I do, of course, believe there are serious issues with Americans’ preparedness for retirement; last fall, when everyone was talking about public pensions as the “other pension crisis,” I called this issue the “other other pension crisis.“)

So that’s the retirement section.

There’s one other piece to the study that looks odd, and I’ll just address it briefly here: the question of other savings.

One question asked, “Have you set aside emergency or rainy day funds that would cover your expenses for 3 months?” and about 40% of respondents said that, yes, they had — about 1/3 of all age groups except the 60+ group, who answered “yes” at a rate of 56%.

Another question asked, “Suppose that you have an emergency expense that costs $400. Based on your current financial situation how would you pay for this expense?” Only 38.6% of respondents said that they had enough cash to pay for this expense immediately, which would mean that, taking these two answers together, 40% of the population has a 3 month rainy day fund, and the remainder have squat — which doesn’t seem right. Does the answer lie in another choice to that survey question: “Put it on my credit card and pay it off in full at the next statement,” which 27.7% selected? — or, since multiple answers were allowed, was this selected in part by a subset of the 38.6%? Was this short-term loan option strategic and practical, a reflection of practices of charging everything for convenience anyway? Other response options were taking a loan, borrowing from friend/family, getting a payday loan, selling something, or just not paying for the expense at all.

So the bottom line is this: I’m disappointed in this report. I haven’t really learned anything and, despite assurances that this is top-of-the-line market research, I’m doubtful.

Are Americans’ retirement plans doomed, or is this survey a bunch of baloney?