Americans are notoriously bad at saving money for the future, especially compared with their elders, who had not only grown thrifty during the Great Depression but also often got a pension from paternalistic employers. Most people dying these days, on the other hand, are lucky if they had a 401(k) plan with any kind of employer match. So it’s no great surprise that about 40 percent of Americans die more or less broke.

What is surprising is that in a nation of non-savers about 60 percent of people (mainly the nation’s “haves”) manage to sock away enough to bequeath money to their heirs. Exactly how they manage this, or why, hasn’t been clear. With one-third of US wealth held by households headed by people aged 65 or older — and the average age of the population expected to keep rising — their financial decisions will matter more and more for the economy as a whole.

Digging into the question of why people save in retirement, economists writing in the Annual Review of Economics say that it’s not clear how many bequests — some of which are enormous — are intentional. Some retirees may want to pass on something to the next generation, but many seem to be anticipating bad times: huge medical bills or expensive nursing homes.

Knowable spoke with review coauthor John Bailey Jones, a research adviser at the Federal Reserve Bank of Richmond, who studies how people spend money over their lifetimes. (The views he expresses are his own, not those of his employer.) This conversation has been edited for length and clarity.

How can a bequest be unintentional?

An unintentional bequest occurs when you are saving money for some contingency, whether it's medical expenses or full-time nursing care, or perhaps you're really concerned that you might live for a really long time, and then events unfold in such a way that you don’t have to make those expenditures. And, of course, an intentional bequest is one whose foremost purpose is to transfer funds to your heirs.

Why are these hard for researchers to distinguish?

If you’re just looking at the bequests themselves, it’s not clear whether they’re intentional or unintentional. All we know from looking at the bequests data is that people with low amounts of income and wealth don’t save at all, and therefore usually leave no bequest — whereas wealthier, higher-income people do save, and do end up leaving bequests. But for people who do save, all the possible scenarios end with notable bequests.

The first story is of saving against the risk of medical expenses. Because those expenses are large, people will have to save a fairly significant amount — and if those expenses don’t occur, they’ll end up leaving a lot. However, these expenses are so large that there’s a class of people that just don’t even bother saving for it. If you’re a person who’s got $150,000 in the bank at age 70, you may just say, “I’m not going to starve myself, because if I go in a nursing home, two years later I’ll run out of money and will be on Medicaid anyway.” So in that story — the story of unintentional bequests — people with a fair amount of income and wealth will save a fair amount. Others don’t save at all.

The other story is you care about your children’s inheritance — you will leave something if you can, but you’re not going to starve yourself. So it depends on how much money you have.

Say you’re a person who’s got $100,000 in the bank, but your kid has gone to college and she’s got an MBA. You’re not going to try to bequeath a lot of money to this kid, because frankly, the kid’s doing pretty well for herself, and it would be better served for you.

But say you’re really rich and you’ve got that same kid: She’s got an MBA, but she’s also got college loans and she’s trying to buy a house in California. At that point, if you’re rich, you’re like: “I can take care of myself, but beyond that, I’m going to take care of my kid after I die.” So in the second story, you’re leaving bequests for your kids, but typically even in this scenario where you are well-off, this is not your very first priority. You want to take care of yourself, and to make sure your spouse is taken care of. You only worry about these bequests once you attain a certain level of wealth.

So you’ve got these two stories, and we write down economic models, and the two models generate very similar-looking bequests.

Two nearly identical graphs show the amount of money left in bequests among those who said they wanted to leave money to their heirs and those who did not. Savings during retirement may often be motivated by factors other than bequests, but many leave money behind in any case.

Studies of the amounts of money left to heirs show little difference between the people who planned to leave bequests and those who unintentionally left money behind — according to both actual amounts (blue line) and economic modeling (red line).

In terms of the intentional bequests, if people really want to give money to their kids, why not give more while they’re still alive?

Most of us who do this work believe that both sorts of bequest motives are present at a time: It may be that mostly you’re interested in leaving money to your kids, but you are also concerned about being hit with a big medical bill, so that’s why you wait. Another reason is that for a lot of people, their principal asset is their house, and houses are pretty illiquid. And older people seem to have a very strong desire to stay in their house.

There’s also something called the strategic bequest motive — which is, basically, you don’t hand over the bequests until you die, just to make sure that your kids take care of you.

Is there a way to tell whether a bequest was strategically motivated?

One thing to look at is the division of assets. Which heir is the principal caregiver? Do they get a larger bequest? And the data show they do — but if you’re the principal caregiver for a parent, you’re not earning much above minimum wage, in terms of what you might eventually inherit. So there’s something of a strategic bequest motive, but at the end of the day, love seems to matter.

So are you saying that the parent is just offsetting whatever income the child had to forego, as opposed to using the bequest as a strategic motivator?

Right, maybe they’re compensating the kid for opportunity cost, or maybe they’re just doing what they can. It’s often understood that people’s long-term-care needs do impose pretty major burdens on their children.

It seems that if someone is in a position to leave a large bequest, they’re also probably in a position to not burden the children as much, since they can buy the best professional long-term care.

“So there’s something of a strategic bequest motive, but at the end of the day, love seems to matter.”

John Bailey Jones

Exactly. A number of studies suggest that in a lot of cases, the best way to deal with the problem of long-term care would be to compensate the familial caregivers: If you need long-term care and you can’t afford to hire professional care, it might be better if the government could compensate one of your kids for their lost income, rather than cutting you a check and having you go to a nursing home or hire some other service.

Let’s go back to intentional versus unintentional bequests. What are some ways economists have tried to tease those apart?

One thing is to just ask people. Financial-product companies sometimes interview richer people and ask, “What is the most important thing that you’re saving for?” And typically, the top thing is medical expenses. Andrew Caplin, a behavioral economist at New York University, has done a bunch of surveys, presenting people with these hypothetical types of insurance. So they say, we’ll offer you something that looks like a really well-designed long-term-care policy, or we’ll offer you something that looks like life insurance. If you really wanted to leave a bequest, maybe you just buy life insurance, right? So the evidence from that typically suggests that it’s actually the medical expenses that are more important.

Bar charts show a breakdown of the values of inheritances in the United States between 1995 and 2016. More than half of bequests were valued at less than $50,000. Only 2 percent were for a million or more, with the rest falling in between. A second chart shows that this 2 percent of large bequests contained 40 percent of the total dollars transferred.

An analysis of inheritances received in the United States from 1995 to 2016 shows that the majority were for less than $50,000 (top). Very few bequests were for more than $1 million, but these large inheritances accounted for a full 40 percent of the total inherited dollars in the study (bottom).

Another argument researchers have teased out is, well, if people really care about medical expenses at the end of life and especially long-term care, why don’t they buy insurance against it? And they typically don’t, and that would suggest that maybe people are saving for bequests instead. Of course, there are a number of issues with how well these insurance policies work.

A really interesting approach is to see what you do with your money when there’s no uncertainty about long-term medical expenses. What do you do when a doctor tells you you’ve got six months to live? You know that if you’re thrifty, six months later, you will have a lot of money that you can leave to your kids. If you don’t care about giving money to your kids, and you only care about yourself and with having enough money to cover your medical expenses, then basically, you can spend that money.

This researcher in Norway found that when widowed people were given a cancer diagnosis, before they died they made a lot of transfers to relatives or other people they cared about, and then mostly left bequests. There wasn’t a large amount of personal consumption. Maybe that’s because if you have cancer, you don’t feel like living large. I don’t know. But it was a fascinating study. Unintentional bequests happen because life is uncertain, so if you can find these strange little situations where the uncertainty is resolved, then maybe you can figure out people’s intentions.

Talking about people’s intentions raises the question of how much thought people are actually putting into their decisions. Is there any reason to believe people are making a rational calculation about how much they will need to save?

It’s a little tricky, because there’s no accounting for taste. Some research shows that wealthier older people really want to make sure that they are well taken care of if they need to go to a nursing home. And who’s to say that’s not rational to save for? If you’ve been in a low-grade nursing home, you know it’s a pretty horrible place.

What’s the role of Medicare in allaying concerns about medical expenses?

While Medicare covers a lot of acute medical expenses, it doesn’t cover long-term care. So if you come down with Alzheimer’s and you have to spend five years in a nursing home, Medicare is going to pay for virtually none of that. Government-provided long-term care comes through Medicaid, and that’s means-tested, so you have to run down most of your wealth to get it. And there are also concerns about quality. So rich people view Medicaid as an extremely unsatisfactory alternative.

What’s your personal approach to bequest plans?

I’m paying for my kids’ college flat-out. So rather than bequests, I’m just making sure they come out with a B.A. and no debt. As a parent, it’s more important to make sure that your kids get an education than it is to make sure that they get some money when you die and they’re 50 years old.

Editor's note: This story was updated on February 12, 2019, to clarify that Jones' views are his own, and not those of the Federal Reserve Bank of Richmond.