One of the most intriguing retirement studies issued in 2012 was by economists James Poterba of MIT, Steven Venti of Dartmouth, and David Wise of Harvard.* Their study looked at wealth holdings among Americans in their late 60s. And one of the findings that I found most striking is the married/single divide in retirement savings.

Here’s a table showing the distribution of financial savings of households age 65–69 in 2008.  One line is for married couples; the second is for single-person households.   (The researchers use the term single-person household intentionally; it includes individuals who are single throughout their lives as well as divorced and widowed individuals.)

The total figure here includes all retirement accounts (e.g., IRAs, 401(k)s) and all taxable savings and investment accounts (e.g., bank, brokerage, mutual fund). It doesn’t include Social Security, pensions, or housing wealth—so it’s not a complete snapshot of adequate retirement resources.

Financial savings of households with those aged 6569 in 2008

Source: Poterba, Venti, and Wise (2012) Note: Includes all tax-deferred retirement accounts and all taxable savings and investment accounts.

Source: Poterba, Venti, and Wise (2012) Note: Includes all tax-deferred retirement accounts and all taxable savings and investment accounts.

Take a look at the median or 50th percentile is the midpoint—half of households are above the median and half below. The median married household has $111,600 in total savings. In contrast, the median single-person household has $12,500 in savings. That’s a huge difference, almost a factor of 10.

At the extremes, the top 30% of married households have savings of $332,400 or more; the top 30% of single-person households, $90,000 or more. Meanwhile, the bottom 30% of married households have less than $24,000 saved; the bottom 30% of single-person households, less than $800.

My guess is that if the entire distribution of American savings looked like the married couple numbers, we’d be having a very different conversation about Americans and their retirement saving ability. So what’s going on? What explains this sharp divide?

One possible explanation is divorce. When a couple separates, assets are divided, and savings can fall due to legal and other costs. If this were the main reason, we’d expect single-person figures to be just under half of those for married couples. But the gap is much wider.

Another explanation is being single throughout life. When you live alone, you don’t benefit from the economies of scale of sharing costs with another person in the household, and so you may save less over your lifetime for a given level of income.  If you lose your job, you don’t have the self-insurance that comes from having another household member with income and health benefits.

Another likely culprit is early death of a spouse. You may be familiar with this situation in your own family. One spouse, often the working male, becomes sick in his 50s or early 60s, loses work, and then dies prematurely. The healthier spouse, often the female, may have a lower income or may not be working. She spends savings on living expenses and her husband’s medical costs. The loss of savings accelerates if they lose health insurance. Long-term care such as a nursing home can also accelerate the loss of assets. Medicaid, which can be used to pay for nursing care, doesn’t kick in until the household depletes most of its savings.

One lesson to draw from this data is that retirement security is more than accumulating savings. You need also to protect against large, unexpected claims. That means having disability, life, and health insurance. The new health care act may help when you lose workplace coverage—but of course you’ll still need to buy a policy. Another lesson to draw if you’re single is that you have to do more, in the form of higher savings, and perhaps insurance coverage, than your married counterparts.

In many studies of retirement preparedness, getting divorced, becoming a widow or widower, or being single are risk factors associated with being financially unprepared for retirement. This important study reminds us why—and also suggests how, as individuals, we might counteract some of these risks.

* Poterba, James M., Steven F. Venti, and David A. Wise, “The Composition and Draw-down of Wealth in Retirement.” National Bureau of Economic Research Working Paper 17536,, Table 2.

Notes: All investing is subject to risk, including possible loss of principal.