America needs a fact-based pension debate. Part 2.

In Part 1 of my response to Senator Bernie Sanders and economist Teresa Ghilarducci, I explained why Sanders’ favorite and most compelling fact – that just about half of Americans approaching retirement haven’t any savings – is totally incorrect, as Sanders is must have known for years.

But it is not a few single factoid. Sanders and Ghilarducci argue that the U.S. pension system is just not working and make several other claims which can be equally false, albeit for a distinct reason. The U.S. retirement system—each private savings and Social Security—is each enormously necessary and enormously expensive. This is just not personal for me. I like Teresa lots. And who knows, possibly Bernie and I’d get along well. He’s grumpy, I’m grumpy.

But we absolutely need to make use of accurate data to grasp whether the U.S. pension system is working well. But just one side of the retirement debate does this.

Consider these numbers from Teresa’s article, two of which quote Sanders while the third comes from Teresa:

• “Approximately 52 percent of Americans ages 65 and older live on less than $30,000 a year, and one in four live on less than $15,000 a year.”
• “Nearly 17 million older Americans ages 65 and older are financially insecure and living below 200% poverty – nearly a third; more than 10 percent are in poverty.”
• “The poverty rate among older people has stubbornly remained at around 10 percent for the past 30 years, increasing from 8.9 to 10.2 percent from 2020 to 2022.”

All of those claims cite data from the US Census Bureau’s Current Population Survey (CPS). Given these figures, only a idiot – me, for instance – would deny the looming pension crisis.

And yet each of those claims is fake, a incontrovertible fact that the Census Bureau itself acknowledges. Accurate data tells a totally different story.

Let’s start originally. In 2012 Wall Street Journal op edSylvester Schieber and I identified that the recent population survey found that Americans ages 65 and older receive $228 billion in income from IRAs, employer-sponsored retirement plans, and pensions. However, IRS data showed that Americans received $568 billion from the identical sources. Now Americans don’t tell the IRS what they receive more income than they really have. This signifies that the CPS, the source of Sanders and Ghilarducci’s facts, only captures about 40% of the advantages paid by private retirement plans.

The US Census Bureau took these concerns seriously. In 2017, two census economists, Adam Bee and Joshua Mitchell, published a groundbreaking study that linked CPS survey responses to IRS records. This linkage allowed the 2 researchers to match CPS-reported income for a given household with IRS data.

What Bee and Mitchell found was startling and possibly probably the most significant finding in retirement research within the last decade: the median retiree’s true income in 2012 was 30 percent higher than measured by the CPS. The actual poverty rate for older people was not the 9.1% reported within the CPS, but only 6.9%. Since 2017, the Bee Mitchell study has been cited nearly 150 times by other scientists.

And the Census Bureau didn’t stop there: In 2023, it released the primary edition of its National Experimental Wellbeing Statistics (NEWS

EWS
), an ongoing project that uses “all available survey, administrative and trade data” to “provide the best possible estimates of our nation and economy.” The data release from Initials NEWS showed: “For households aged 65 and over [in 2018]The median household income is 27.3 percent higher and poverty is 3.3 percentage points lower than in [CPS] Poll estimates.” That’s a giant deal.

Social Security Administration economists confirmed This approach produces similar results. For example, while the CPS reported that only 47% of households over age 65 received advantages from a personal pension plan, IRS data showed that 69% actually received advantages from a personal pension plan.

And private retirement advantages are at the foundation of the CPS’ income measurement problem. The essential reason: The Census Bureau defines “income” as payments received “on a regular basis.” This signifies that needs-based withdrawals from a retirement account will not be counted as “income.” And as seniors increasingly receive advantages from retirement accounts, the CPS’s underestimation of retirees’ true income has increased over time.

Bernie Sanders and Teresa Ghilarducci apparently live in a world where this research has not yet taken place.

What does more precise data tell us about Sanders and Ghilarducci’s claims?

The publicly released Census NEWS data doesn’t allow us to accurately confirm Sanders’ claim that 52% of those over 65 have incomes of lower than $30,000 and one in 4 have incomes of lower than $15,000. But 52% means a figure near the median, and we all know that the true median income for seniors is 27% higher than within the CPS data cited by Sanders. So Sanders’ claim is nearly definitely false.

What about Sanders’ incontrovertible fact that nearly a 3rd of older Americans have incomes below 200% poverty? Well, Bee and Mitchell’s study found that 33.4% of seniors had incomes below 200% of the poverty level, in response to CPS. But IRS data showed only 23.6% did so. So incorrect again.

What about Teresa’s claim that senior poverty “has remained stubbornly constant at about 10 percent for the past 30 years?” That’s perhaps the largest problem. According to Bee and Mitchell of the Census Bureau, in 1990, 9.7% of seniors had incomes below the poverty line. By 2018, the poverty rate amongst older people had fallen to six.4%. Teresa says that over the identical period, the danger of falling into poverty has fallen by greater than a 3rd. That’s a totally different narrative.

Now I do not expect everyone to pay attention to these data issues. For example, the Social Security Administration claimed for years, based on CPS figures, that as much as 36% of seniors received just about all of their income from Social Security. When Bee and Mitchell checked these statistics using IRS data, they found that only 12% received 90% or more of their income from Social Security. That’s an enormous difference, but you possibly can’t expect the person on the road to know the research.

But Sanders chairs a Senate committee and has a staff capable of manufacturing a 21-page report with 94 footnotes. Ghilarducci is a tenured professor specializing in retirement issues.

They should know higher than to cite data that we all know of course is inaccurate. We owe people a greater retirement debate than they’re getting.