You’re Probably Saving Enough for Retirement

The headlines are meant to alarm you. “Millions of older workers are nearing retirement with nothing saved,” CBS News reports. “Only one in 10 low-income workers between the ages of 51 and 64 had any retirement savings in 2019, says the New York Post, citing “a troubling report recently published by the US Government Accountability Office.” Meanwhile, nearly all high-income Americans are saving, an inequality the GAO attributes to the federal tax preference for retirement-plan contributions.

All of this is confused, at best. The report analyzes households of workers 51 to 64 using the Federal Reserve’s Survey of Consumer Finances. The SCF data do show that only around 1 in 10 households in the bottom fifth of incomes in 2019 reported having a retirement account, such as a 401(k) or IRA. The report prompted Sens. Sheldon Whitehouse and Bernie Sanders to lament “the huge portion of Americans who lack any retirement savings at all.”

But the GAO misinterprets the Fed data in three important ways that, once corrected, paint a much more encouraging picture.

First, most of those the report calls low-income “older workers” aren’t actually working. While 87% of middle- and 92% of high-income households were working, only 33% of these low-income households in 2019 had a single person earn a single dollar of wages. This is because retirement savings replace wages lost after retirement. Nonwage sources of income—including welfare payments and investment income—don’t cease in retirement. Other government agencies, such as the Social Security Administration and the Congressional Budget Office, avoid this problem by analyzing Americans by earnings instead of income.

Second, even low-income households with working members shouldn’t necessarily be saving apart from Social Security. Of low-income households in this age range in 2019, 92% had total earnings below $20,000, the vast majority of which would be replaced by Social Security’s progressive benefits. If these households did save, many would lose means-tested benefits such as Medicaid or food stamps. Most of the nonsaving households are acting rationally given the incentives they face.

Third, the report unnecessarily limits its investigation of “retirement savings” to balances in retirement accounts. This ignores the more than $17 trillion in benefits Americans have accrued under traditional pensions, which remain almost universal in the public sector.

Here is a more realistic figure: Among all households of workers 51 to 64 earning more than $20,000 in 2019, 80% either have a retirement account or are entitled to traditional pension benefits. Even among households with total earnings between $20,000 and $40,000, 49% have a formal retirement plan. Simply adopting reasonable parameters multiplies the report’s headline finding fivefold.

Yes, retirement savings are highly unequal. But it isn’t because, as the report asserts, “for each dollar contributed to a retirement account, a taxpayer subject to the highest marginal tax rate deducts 37 cents compared to 10 cents for a taxpayer subject to the lowest tax rate.” The federal tax preference for retirement savings isn’t a deduction, like mortgage interest. Taxes are merely deferred until money is withdrawn. Economist Peter Brady of the Investment Company Institute finds that because tax rates drop more in retirement for low earners than for high earners, the federal tax subsidy per dollar saved is typically higher for low earners than high earners.

The lion’s share of the retirement-savings tax preference flows to high-income households simply because high earners save far more for retirement. That’s because Social Security replaces a far smaller share of their pre-retirement earnings. While a low-income household might receive Social Security benefits replacing 70% to 80% of its pre-retirement wages, a high earner making $325,000 would receive a Social Security replacement rate of only around 15%. Even lacking the federal subsidy for retirement savings (which some economists argue does little to increase retirement savings), high earners would save far more than low earners.

It’s easy to make the data dance to a particular tune if that is the analyst’s desire. A more nuanced picture of the U.S. retirement system shows that while it isn’t perfect, most Americans who need to be saving for retirement are doing so.

Mr. Biggs is a senior fellow at the American Enterprise Institute.