Fact-Checking Cook and Ghilarducci On Retirement (Again)

Really, I don’t want to be doing this. I just want a fact-based discussion on retirement income security, so we can fix what needs fixing and stop worrying about things we shouldn’t be worrying about. Is that too much to ask?

Apparently, it is.

Back in August I published an entire study through the American Enterprise Institute that fact-checked 10 reasons Americans should get over their “denial” there’s a retirement crisis, from Christopher Cook and Teresa Ghilarducci. My conclusion – and I invite you to fact check me if you like – is that none of their 10 reasons really hold up.

I hoped that would be the end of that. But Cook and Ghilarducci have a new article out. Do they repeat some of the same old falsehoods? Sure.

But they also come up with new and even-falser falsehoods than before.

And yet, as much as I don’t like this, I’ve sworn to myself that an informed discussion of retirement savings is too important to let this stuff pass anymore. And so here we are.

Cook and Ghilarducci’s column, “Why Aren’t We Talking About America’s Retirement Crisis?” has been syndicated in newspapers across the country. Ghilarducci herself is one of the most prominent academics working in the retirement field today. And so it matters that this column is full of claims that aren’t simply mistaken by a decimal point or a matter of interpretation. It is filled with what I would call research negligence: claims that the authors either knew or should have know are misleading and false.

Let’s run through some of them.

“Nearly half of today’s middle-class adults will be poor or near-poor in retirement.”

This is simply untrue – no two ways about it.

For background, the Census Bureau defines “near poverty” as having an income between 100 and 125 percent of the federal poverty threshold. According to Census Bureau research, about 6.9 percent of age 65+ Americans have incomes below the poverty line. About 14.8 percent had incomes below 150 percent of the poverty line, which is above near-poverty. My guestimate is that around 12.9 percent of current seniors are either poor or near-poor. (This assumption doesn’t matter much, as we’ll see.) Moreover, all credible projections of future retirement incomes that I’m aware of find that elderly poverty will decline in coming decades. So 12.9 percent probably overstates the future number of poor or near-poor seniors.

Moreover, the vast majority of Americans who are poor in old age will also have bene poor prior to retirement. That same Census Bureau research shows that poverty declines significantly as households enter retirement. On net, more people move out of poverty in retirement than move into it. So Cook and Ghilarducci’s claim that half of middle-class Americans, who weren’t poor prior to retirement, will become poor or near-poor in old age, is just preposterous. I don’t see how you can possibly make the math work. Anyone purporting to be an expert on retirement incomes could tell you that it’s false.

Next up! Cook and Ghilarducci make a double-claim, that “Nearly half of older Americans have no retirement savings and must rely solely on Social Security in old age.”

But twice the claims means twice the errors.

The claim that “Nearly half of older Americans have no retirement savings” counts only retirement accounts. If you’re a federal, state or local government employee and you’ve got a (typically pretty generous) traditional pension, Cook and Ghilarducci say that’s not savings. Or, if you have assets outside of a retirement account that will provide income in retirement, such as a taxable investment account, real estate, a farm or small business, and so forth, that’s also excluded.

Why do Cook and Ghilarducci exclude those other sources of retirement income? Because if you include them, as the Federal Reserve’s Survey of Household Economics and Decision-making does, you’ll find, as the Fed did, that 88 percent of Americans aged 60 and over have retirement savings on top of Social Security. Look it up if you like.

Cook and Ghilarduci may counter that, sure, people might have those forms of retirement savings, but it’s not enough savings. If so, that’s what they should say. And I could counter that other Fed data show that retirement savings are at record high levels in every age, income, educational and racial/ethnic group. And at least then we’d be having an honest conversation.

It keeps going with the latter half of their sentence: “…and must rely solely on Social Security in old age.”

Seriously, I can’t even imagine that they’re making this claim. It has been, as lawyers like to say, “asked and answered.”

In 2007 – that’s 17 years ago – Social Security Administration researcher Lynn Fisher used IRS data matched to government household surveys to examine how heavily retiree households rely on Social Security. What did she find? Only 4.5 percent of elderly persons relied on Social Security for all of their income. In other words, literally one-tenth of the figure Cook and Ghilarducci claim. More recently, the Census Bureau analyzed the number of seniors who receive at least 90 percent of their income from Social Security. The Census Bureau’s answer: 12.2 percent, despite using a lower bar than Cook and Ghilarducci.

Next Cook and Ghilarducci drop this bomb on us: “About 79 percent of people aged 62 to 70 can’t afford their pre-retirement living standards.” I have no idea how they’re generating this figure. But I know that it’s incorrect.

For instance, economists Peter Brady and Steven Bass used IRS data to track American’s incomes from ages 55 through 72. To check Cook and Ghilarducci’s claims, I’ll compare income at 65 to income at 55. (The specific ages don’t matter very much.) For the typical household, income drops by only 12 percent from age 55 to 65. Retirement planners say that retirees need an income equal to 70 to 80 percent of their pre-retirement earnings. The typical 65-year old has a “replacement rate” of about 88 percent. For the poorest 25 percent of seniors, their incomes increase in retirement. I just don’t think the data support Cook and Ghilarducci’s claims.

Or, if you don’t like that kind of data, we can just ask people. The Fed’s Survey of Household Economies and Decision-making asks Americans to describe their financial security in one of four ways: “Finding it difficult to get by;” “Just getting by;” “Doing okay;” and “Living comfortably.” Among 52 to 61 year olds from 2019-2023, 76 percent said they were at least “Doing okay.” Among 62 to 70 year olds, 82 percent did so. And self-reported financial security continues to improve with age: at ages 75 and over, 86 percent said they were at least doing okay financially, the best in any age group. Moreover, if we look at objective questions regarding financial security – Can you pay a $400 expense with cash? Do you have a rainy day fund? Can you pay your bills? Are you carrying credit card or medical debt? – seniors are again financially healthier than younger Americans.

In short, Cook and Ghilarducci’s claim just doesn’t have any backing.

I could go on, but you get the point.

The issue isn’t necessarily that Cook and Ghilarducci are wrong in claiming Americans faces a retirement crisis and that I’m right in saying we don’t.

It’s in how that debate is being conducted. Cook and Ghilarducci have a habit of making claims that aren’t true and which by now they should know aren’t true. This simply isn’t how public policy debates should happen. Retirement is a hugely important issue both to households and to policymakers. They deserve to be given information they can rely upon.