Related Links

• Savings After Retirement
• Why Do Couples and Singles Save During Retirement?

Transcript

Tim Sablik: My guest today is John Bailey Jones, vice president of microeconomic analysis in the Research Department at the Richmond Fed. John, welcome back to the show.

John Bailey Jones: It's great to be back.

Sablik: We've got a special Valentine's Day topic for our listeners today: how do singles save in retirement compared to married couples? Okay, maybe it's not a very romantic question. But give us a break, we're researchers.

John, this topic is actually something that you've been researching for several years now. I was curious, what drew you to this question?

Jones: A big part of my research agenda is looking at the economics of health and aging, in particular at the behavior of older households. My co-authors and I have written a number of papers on what is called the "retirement savings puzzle." Why do older households draw down their wealth so slowly? The paper we're going to talk about today falls into this category.

Why do I care about this issue in general? Well, retirees are a large and growing part of the population. Their behavior is also of a personal interest to me as I think about what my parents did as they got older and I start to contemplate what I might do as I grow older.

Sablik: One thing that stood out to me immediately from your paper is the fact that previous research into saving behavior during retirement has tended to focus on singles. Why is that, and what did that research find?

Jones: Well, the literature initially focused on singles because it was easier. It's easier to think through a problem of a single person rather than a couple. It's easier to program an economic model on a computer — there are fewer moving parts. Typically, you start out with the easier problem and then work up to the more complicated ones.

What we found regarding the saving of singles is that in the United States, at least, medical expenses can explain a substantial part of the saving of older singles. This is because medical expenses tend to be backloaded toward the end of people's lives. To give just a simple example, you might see an 80-year-old being frugal in order to be able to pay for long-term care at age 88.

This research has further found that many people also saved to leave bequests, especially people with more wealth.

The third motive, which I will not talk about much today but has been uncovered by other researchers, is that many older people hold on to their family homes because they really prefer to live there. It's a place where they spent most of their lives. It's a place that they are familiar with.

A key point is that these motives may apply simultaneously. For example, my wife and I may set aside assets to pay for long-term care when we get older. But if I pass away without needing that care, I'm happy to leave those assets to my heirs. Either way, the money goes to a good purpose.

Sablik: That's a good segue to your paper where you look at differences between singles and couples. What is different about saving behavior of retired couples compared to singles?

Jones: First of all, they're just differences in scale — two people rather than one. So, couples have more income and more expenses.

But I'd say the fundamental difference is that most couples eventually become singles. Couples have to figure out how much wealth they want to leave to the surviving spouse if one member dies. Do they prefer to consume a lot while they are together? Or, do they want to make sure the survivor is well taken care of? Throughout this, they also have the option of transferring wealth to other heirs.

What we see in practice is that couples do leave a lot of wealth to surviving spouses. But when the first spouse dies, about one third of households also leave wealth to non-spousal heirs, usually children. So, it's not surprising that our simulation model finds that couples seem more heavily motivated by bequests than do singles. That's kind of a big difference.

Part of this is due to the tendency of married people to be richer. Our model is set up so that bequests are a luxury good. That means it's a good that rich people tend to spend a higher share on — you can think of mansions and yachts. But part of the difference is that couples have more opportunities to leave bequests. They can leave wealth to surviving spouses, who can then choose to hold on to that wealth and bequeath it when they die, or the first spouse to die can leave bequests to non-spousal heirs at the time of his or her death.

Sablik: Just to clarify, in your data couples are specifically married couples? Do you look at non-married couples as well?

Jones: Couples are married couples. We throw cohabiting people out of the dataset because our framework doesn't know what to do with them.

For the time being, it's a fairly safe thing to do. In the next 20 to 30 years, we're going to have more cohabitation. It matters in a lot of ways. For example, Social Security benefits traditionally have been spousal benefits. There's not a common law marriage provision for that.

Sablik: This brings me to my next question. How does all this research on savings and retirement connect to the policy work that the Fed does, or to fiscal policy in general?

Jones: In order to meet its dual mandate, the Fed needs to understand the forces that drive inflation and unemployment. The saving decisions of older households, who now hold about 40 percent of wealth, are important drivers. So, if you want to have a good grasp of the economy as a whole, this is one of the sectors you have to keep your eye on.

On the fiscal side, the federal government spends a lot of money on programs for older households — Social Security, health insurance programs like Medicare and Medicaid. These amounts are large and they're generally thought to be unsustainable, so reform is needed. We need to assess how reforming these programs would affect older households and, in turn, how the response of older households to these reforms would affect the economy.

Suppose the government makes Medicare insurance less generous. Will households reduce their non-medical consumption to cover higher medical expenses — say, maybe go out to eat a little less — or will they spend funds that otherwise would have bequeathed? These are really important questions because they're probably some pretty major reforms on the horizon.

Sablik: Are there any other questions related to saving and retirement that you're hoping to explore in future research?

Jones: I'm currently working on some projects, trying to produce better estimates of medical expense risk. We've talked about how medical spending is a big concern and a driver of saving for a lot of people.

A big challenge in measuring medical expense risk is trying to capture the persistence of medical spending over time. If you get hit with a big medical bill this year, are you more likely to get hit with a big bill next year? You're concerned about year-to-year expenses, but you're also concerned about expenses over your remaining lifetime and how they compare to what you might have in savings.

Another concern is predicting how much of a person's medical expenses will be paid out of pocket by the person themselves and how much will be picked up by insurers. From a scientific perspective, this is really challenging. There's a lot of variation on the shares and they have proven really hard to predict.

Sablik: That sounds interesting. We'll definitely keep an eye out for that research. John, thanks so much for coming on to talk with me today.

Jones: It's always a pleasure.