Skip to Content
MIT News feature

The Economics of Retirement

The retirement landscape is changing. And James Poterba’s research shows that you probably aren’t saving enough.
February 18, 2015

When James M. Poterba started studying retirement finance, he knew he was venturing into fraught territory. Tens of millions of Americans compulsively check their 401(k) retirement accounts online and lie awake wondering how much they can save or, if they are already retired, how they can stretch fixed incomes. Even so, Poterba could not have known how readily people would share their anxieties.

“I am often asked, ‘Is it just too late? How am I doing, Jim?’” he says.

But Poterba is not in the business of giving retirement advice. The Mitsui Professor of Economics at MIT, he is a scholar who has studied how Americans save for retirement and how they fare financially as retirees. His research has revealed disquieting facts. For instance: over a recent two-decade span, among American households headed by someone aged 48 to 56, the proportion whose head has a defined-benefit pension plan has dropped from 50 percent to 30 percent. (In a defined-­benefit plan, employers guarantee retirees a fixed pension, typically based on length of employment.) The income certainty those pensions once provided to a substantial chunk of the population is now vanishing.

Perhaps more worrisome, Poterba and two colleagues found that roughly half of Americans die with less than $10,000 in assets beyond their annual retirement incomes. Some retirees are one health problem or market slump away from having almost no savings.

Meanwhile, American life expectancy has risen dramatically: from 51.5 and 58.3 years for men and women born in 1900, respectively, to an estimated 80.0 and 84.6 for men and women born in 2000. This otherwise welcome development means more people need to pay for more years of living after retirement.

“Only economists could turn the prospect of someone managing to live a very long time into something that is potentially problematic,” says Poterba, with a wry laugh.

Retirement is not problematic for everyone, of course. “There is tremendous heterogeneity in the elderly population—in their preparation for retirement, and also in their retirement spending needs,” Poterba observes. “There are disparities in health status, family support, longevity, and in how much people have saved.”

So there is no universal answer to the question “How much saving is enough?” Still, Poterba has produced calculations that estimate how much you must save in order to generate specific levels of a certain type of retirement income.

And while most of his work illuminates the national retirement picture as a whole, studying it can give people much-needed perspective on their own financial needs.

Getting to 65
One of the most important things economists have documented recently is how much the retirement landscape is changing. Americans are living longer, are more likely to be unemployed, and are working later in life when they can; many of those who had pension guarantees are losing them. To drive home the idea that increased longevity may affect your finances, Poterba notes that when a married couple reaches 65, there is roughly a 50-50 chance that at least one spouse will live to 90.

Then consider that to save, you must earn more than you spend. Since the 2008 economic crash, the earning part has grown harder for many Americans. Federal statistics show that participation in the labor force is at 63 percent, down from 66 percent in 2008—the lowest it has been since the 1970s. Inflation-adjusted wages have also stagnated at many income levels.

In another change, those who can keep working beyond age 65 are increasingly doing so, reversing a trend that once seemed inevitable. In 1970, 42 percent of men and 18 percent of women aged 65 to 69 were in the labor force. Twenty years later, in 1990, those figures had dipped to 26 and 17 percent, respectively. But another 20 years on, in 2010, the figures had rebounded to 37 percent and 27 percent; people do not want to give up jobs and careers, or cannot afford to.

There is no universal answer to the question “How much saving is enough?”

As Poterba has written, the reason for this trend is “most likely a combination” of financial need and other factors, such as improved health care. Still, the change is striking: in 2013, 52 percent of those older than 65 in the top quarter of the income distribution reported at least some earnings, compared with 12 percent who did so in 1988. A Gallup poll in 2014 found that Americans’ self-reported retirement age is 62, up from 57 in 1991, when Gallup started tracking the question.

Certainly there is logic to staying in the work force. “That’s very important, because there are two effects,” Poterba observes. “You add a couple of high-­savings years and defer by a couple of years the age at which you begin drawing down whatever you accumulated.”

Meanwhile, the diversification of retirement income has become even more elusive. While it was never universal, the ideal “three-legged stool” portfolio—Social Security income, an employer-provided private pension plan, and savings, including 401(k) plans—now exists mostly among the well off. Just 12 percent of people over 65 in the lower half of the national income distribution have employer-­provided private pension income. Whereas the 401(k) was once touted as a supplement to the defined-benefit pension, it is now often the replacement for it. And instead of an annuity income stream from a pension, more people today get an account balance at retirement and must decide how to consume that balance as they age. In all, roughly one-quarter of the population has a portfolio that balances on all three legs of the retirement stool.

A large number of Americans do own a home—about 64 percent of residences were owner-occupied as of October, according to the U.S. Census Bureau. That can be a good source of savings, or at least provide a psychologically easy way to save. “The mortgage payment is creating the savings for you in the house,” Poterba says. “It’s not a decision that requires you to say, ‘I’m going to take [some] percent of my income and put it in a savings account each month.’”

If you do rely on putting money in an account, however, you might be surprised by how much consistent saving it will take to fund a comfortable retirement. In the Richard T. Ely lecture at the annual meeting of the American Economic Association in January 2014, Poterba presented new calculations that help clarify how much people should be stashing away.

Assume you are continuously employed with a salary rising 1 percent annually when adjusted for inflation. In order to purchase an annuity—a regular annual payout—replacing half of your salary at age 65, you must save 14.8 percent of your income if you begin saving at 25, given a 2 percent return on essentially risk-free savings such as U.S. Treasury bonds. You must save 20.7 percent of your income if you don’t start saving until age 35, and 32.7 percent of your income if you don’t start saving until 45.

“The results often suggest savings rates that are higher than most people would expect,” Poterba says—and not just for annuities. As he noted in the Ely lecture, “The calculations suggest that savings rates like the median 401(k) contribution rate are unlikely to enable prospective retirees to replace half of their final earnings.” Translation: Yes, most folks should be setting more aside.

Citizen of economics
Poterba, 56, became a leading economist of taxation and public finance after getting his PhD at Oxford and joining the MIT faculty in 1983, at age 25. Last November he received the Daniel M. Holland Medal from the National Tax Association, a lifetime achievement award. He still studies many taxation issues.

It was in the 1990s that he began studying retirement savings, sometimes in tandem with economists Steven Venti of Dartmouth and David Wise of Harvard. (This was a natural extension of his work, since many retirement savings vehicles—including IRAs and 401(k)s—are constructed around tax incentives. “They are creatures of the tax code,” Poterba says.) The first paper they published together, in 1994, examined whether IRAs and 401(k)s were “crowding out” other types of saving. “We did a lot of research, and we never found substantial crowd-out,” says Wise. He calls their ongoing partnership “a really easy collaboration,” and he says that when they are finishing a paper, he trusts Poterba to make sure it is sound.

“Jim is the closer,” he says. “I always like to have him look last.”

Poterba, however, says it’s he who is getting an “education” in retirement issues from other scholars, punctuated by some original research of his own. Such collegiality is typical: he is a highly active citizen in the republic of economics. A former head of MIT’s Department of Economics, he has organized public events at MIT and is a de facto historian of the celebrated department. His wife, Nancy Rose, is also an MIT economics professor; she is now on leave serving as chief economist of the Department of Justice’s antitrust division.

Roughly half of ­Americans die with less than $10,000 in assets beyond their annual retirement incomes.

Beyond MIT, Poterba is an elected fellow of the American Academy of Arts and Sciences, and since 2008 he has been president of the National Bureau of Economic Research (NBER), the Cambridge-based network that supports scholarship, disseminates research papers, and hosts annual conferences. He is also one of eight people on the NBER’s Business Cycle Dating Committee, which makes quasi-official judgments about when the U.S. economy is considered to be in recession.

“He’s been a force in the world,” says Robert Hall, PhD ’67, the Stanford macro­economist who chairs the business cycle committee.

Poterba started doing NBER-backed research as an undergraduate at Harvard. (He also worked for Lawrence ­Summers, then an assistant professor at MIT, with whom he has since coauthored over a dozen papers.) “The NBER had always been really important for my growth as a scholar,” he says. “It’s been a lot of fun.”

From his well-informed vantage point, Poterba says he sees economists more than ever tracking not just how people accumulate wealth but what happens to those assets during retirement. “The reason that’s shifted is that the Baby Boomers have been moving along their life course,” he says. “When that large group of the population was in early middle age, there was tremendous interest in the saving element … Today, as the Baby Boomers are approaching retirement, more interest has shifted to understanding what role these assets will play after individuals stop working.”

In the retirement years
Poterba, Venti, and Wise have been part of that shift. Their paper on savings at the time of death, published in 2012, found that 46 percent of Americans die with less than $10,000. It also showed how much finances can vary depending on life circumstances: about 52 percent of single people in the study who were single throughout retirement had less than $10,000 in assets, compared with just 36 percent of people who had started retirement in two-person households but later became single, and 26 percent of people in two-person households.

“These results raise fundamental questions about the health of the U.S. retirement savings system,” wrote Harvard economist David Laibson in a published commentary on the paper.

Such empirical findings also add nuance to a basic concept of retirement finance, the life-cycle diagram pioneered by former MIT professor (and Nobel Prize winner) Franco Modigliani. In this picture, assets accumulate until retirement, then decline smoothly until death. In a 2011 paper, Poterba, Venti, and Wise showed that barring death of a spouse, net worth in the median senior-citizen household in the 1990s and early 2000s did not decline immediately upon retirement: it remained even or actually rose between ages 65 and 80, as retirees cut spending to match their retirement incomes.

“The revisionist view of what’s happening is that many households are holding wealth as a precaution against uncertainties and shocks that may hit them later on,” says Poterba.

Those “shocks” are often medical, as an ever-growing body of retirement studies shows. But health issues do not always loom largest in retirement finance, Poterba warns. “Had we not lived through the Great Recession recently, we might be tempted to say that’s a bigger consideration,” he says. “But anyone who saw their equity investment in an IRA or 401(k) decline by 35 percent in 2008 and 2009 would know there are also financial-­market shocks that can be pretty important.”

The volatility in the housing market of the last decade may also lead to yet another change in the retirement landscape. “The Baby Boomers’ parents benefited from a generally rising housing market,” Poterba notes, adding: “I think for many people home equity serves as a buffer … but they don’t feel comfortable financing day-to-day consumption out of it. Depending on the overall financial preparation of households in the next two to three decades, we may see people who need to dip into their housing equity much more substantially.”

Economists would like to study many more of the details related to these late-in-life scenarios. When do retirees sell their houses, and why? To what extent do informal networks (family, friends, local organizations) provide off-the-books support for older people? For his part, Poterba has launched new projects that study the relationship between wealth and health for seniors and ask whether more education leads to better retirement outcomes, independent of other factors.

The more we know about the many paths people take through retirement, Poterba says, the harder it may be for policymakers to craft “one-size-fits-all” financial solutions. But by increasingly putting these issues under the lens, economists can help us see that retirement planning does not begin at 40 or end at 65. It usually requires foresight and financial flexibility for decades, right through the final years of most people’s lives. Still, more data will probably never stop people from approaching retirement economists with those two eternal questions: “How am I doing? Is it too late?”

Keep Reading

Most Popular

Large language models can do jaw-dropping things. But nobody knows exactly why.

And that's a problem. Figuring it out is one of the biggest scientific puzzles of our time and a crucial step towards controlling more powerful future models.

OpenAI teases an amazing new generative video model called Sora

The firm is sharing Sora with a small group of safety testers but the rest of us will have to wait to learn more.

Google’s Gemini is now in everything. Here’s how you can try it out.

Gmail, Docs, and more will now come with Gemini baked in. But Europeans will have to wait before they can download the app.

This baby with a head camera helped teach an AI how kids learn language

A neural network trained on the experiences of a single young child managed to learn one of the core components of language: how to match words to the objects they represent.

Stay connected

Illustration by Rose Wong

Get the latest updates from
MIT Technology Review

Discover special offers, top stories, upcoming events, and more.

Thank you for submitting your email!

Explore more newsletters

It looks like something went wrong.

We’re having trouble saving your preferences. Try refreshing this page and updating them one more time. If you continue to get this message, reach out to us at customer-service@technologyreview.com with a list of newsletters you’d like to receive.