The University of Wisconsin professor on the challenges of decumulation, the connection between defaults and abandoned retirement accounts, and the ‘hassle costs’ that can weigh on workers.
Our guest on the podcast today is Dr. Anita Mukherjee, associate professor in Risk and Insurance at the Wisconsin School of Business at the University of Wisconsin at Madison. Professor Mukherjee conducts research on public policy related to prisons and the opioid crisis, as well as household finance, retirement, and aging. Professor Mukherjee joined the Wisconsin School of Business after completing her Ph.D. in applied economics at the Wharton School at the University of Pennsylvania. She holds an M.S. in management science and engineering, a B.S. in mathematics, and a B.A. in economics, all from Stanford University. Prior to enrolling at Wharton, she spent two years working as a consultant to Oliver Wyman.
Abandoned Retirement Plans
“Lost and Found: Claiming Behavior in Abandoned Retirement Accounts,” by Anita Mukherjee and Corina Mommaerts, Retirement & Disability Research Center, 2020.
“Set It and Forget It? Financing Retirement in an Age of Defaults,” by Lucas Goodman, Anita Mukherjee, and Shanthi Ramnath, papers.ssrn.com, Oct. 19, 2022.
“Frictions in Saving and Claiming: An Analysis of Unclaimed Retirement Accounts,” by Anita Mukherjee and Corina Mommaerts, Retirement & Disability Research Center, December 2019.
Wealth Inequality and Retirement
“Inequality in the Golden Years: Wealth Gradients in Disability-Free and Work-Free Longevity in the United States,” by Hessam Bavafa, Anita Mukherjee, and Tyler Q. Welch, Journal of Health Economics, Sept. 26, 2023.
“The Effects of the Opioid Crisis on Employment: Evidence From Labor Market Flows,” by Anita Mukherjee, Daniel W. Sacks, and Hoyoung Yoo, papers.ssrn.com, July 27, 2022.
“The Effects of Naloxone Access Laws on Opioid Abuse, Mortality, and Crime,” by Jennifer L. Doleac and Anita Mukherjee, papers.ssrn.com, Dec. 5, 2022.
“Intergenerational Altruism and Retirement Transfers: Evidence From the Social Security Notch,” by Anita Mukherjee, The Journal of Human Resources, September 2022.
Medicaid and Financial Literacy
“Medicaid and Long-Term Care: The Effects of Penalizing Strategic Asset Transfers,” by Junhao Liu and Anita Mukherjee, papers.ssrn.com, Jan. 6, 2020.
“Building Financial and Health Literacy at Older Ages: The Role of Online Information,” by Anita Mukherjee, Hessam Bavafa, and Junhao Liu, The Journal of Consumer Affairs, January 2019.
Christine Benz: Hi, and welcome to The Long View. I’m Christine Benz, director of personal finance and retirement planning for Morningstar.
Jeff Ptak: And I’m Jeff Ptak, chief ratings officer for Morningstar Research Services.
Benz: Our guest on the podcast today is Dr. Anita Mukherjee, associate professor in risk and insurance at the Wisconsin School of Business at the University of Wisconsin at Madison. Professor Mukherjee conducts research on public policy related to prisons and the opioid crisis, as well as household finance, retirement, and aging. Professor Mukherjee joined the Wisconsin School of Business after completing her Ph.D. in applied economics at the Wharton School at the University of Pennsylvania. She holds an M.S. in management science and engineering, a B.S. in mathematics, and a B.A. in economics, all from Stanford University. Prior to enrolling at Wharton, she spent two years working as a consultant to Oliver Wyman.
Dr. Mukherjee, welcome to The Long View.
Dr. Anita Mukherjee: Thank you. Thank you for having me.
Benz: We’re excited to have you here. You have a diverse portfolio of research interests, public policy related to prisons and the opioid crisis, as well as household finance, retirement, and aging. Do you find that there are a lot of connections, a lot of interrelationships among those two broad sets of topics? And if so, what’s an example of that?
Mukherjee: I think for me, the draw to this set of topics has really been a focus on vulnerable populations. And you find those both among people growing older, with some uncertainty as to how the last years will look like, and also among people who are afflicted by drugs or by being involved in the criminal justice system. So, for me, it’s been vulnerable populations and also uniquely groups for whom policy is hugely impactful. I think both when we’re older, we’re very influenced by policies around healthcare, around retirement savings; and people who are in prison or involved in the healthcare system from different angles are also very involved and affected by policy very deeply. So, these have been reasons that I’ve focused on these populations.
Ptak: What was the initial research spark for you in these areas, and perhaps specifically in the area of household finance, retirement, and aging?
Mukherjee: For me, I think, household finance is something that I think every single one of us has to do on some level. And I think within that, what’s unique about retirement and aging to me is just the long-term planning that’s required there, and all the different frictions and biases that then enter along with that. So, it’s not just like planning for your next big thing, it’s planning for something that’s quite uncertain over maybe decades into the future. You don’t even know if you’ll live that long, you don’t know what your health state will be. Employers and policy attempts to help you plan with different sorts of benefits. But I think navigating the public and private resources and the complexity of the long-term horizon is really interesting for me. And I think just in conversations, one of my projects has been about these abandoned retirement accounts. It’s been also this lack of attention that we could have, at least in our younger years, around planning for retirement. You have different jobs, you’re not totally clear what savings you have with each job, and how those are all adding up and getting ported into later states of life. So, I think those challenges were really interesting for me to think about and study.
Benz: I want to follow up on the abandoned retirement accounts. You co-authored a paper on that topic. First off, could you talk about how you gauged abandonment? Because it seems like a lot of people, myself included, might really choose to tune out of their retirement accounts for long periods of time. So how did you know that something had really been left probably inadvertently?
Mukherjee: It’s a great question because this is something we had to define in some ways in the paper. This is co-authored with Shanthi Ramnath and Lucas Goodman, both economists in policy areas. And this paper, we defined abandonment by looking at people who failed to meet their required minimum withdrawals for at least three years in a row. Basically, with retirement savings, you’re allowed to accumulate them tax-free. But at some point, you have to withdraw them and pay taxes on them. And so that required a minimum distribution—usually hits at age 70. Now it’s at age 72. It will be age 75. But if at that point, we see people failing to withdraw their savings for one, two, or three years—in the paper we even look at up to 10 years—that’s when we start to categorize you as maybe you’ve abandoned these savings and you’re not planning to withdraw them at all.
Ptak: What happens to assets that have been abandoned in a 401(k) plan? What responsibilities does the plan sponsor have to reunite account owners with their accounts?
Mukherjee: This is a great question. It’s actually a muddy one. So, what happens to assets that have been abandoned? In principle, they should stay with the plan if they’re at least $5,000. And then at some point, if the plan determines that you have abandoned them, then they can send them to the state through a process called escheatment where they say, look, this is abandoned property. We’re not holding on to it anymore. It should be the state’s responsibility to then take it over and find the rightful owner. De facto, like the actual rules that the plan sponsors have to try and reconnect owners with their accounts is not very much. And there’s a lot of variation across plans and how much they do. At minimum, they have to send you notices to your last known mailing address. But many plans do much more, like they have staff that will Google you and try to locate you. But really, they just have to go by what you have given them, which is your address, maybe your phone number, and they give you repeated mailings there. But if you’ve moved out of state, or if you’ve moved even houses, those mail communication may not come to you. And it’s really nobody’s fault legally at that point.
Benz: How do the states get involved? In the paper, you talked about how a couple of different states, I think the examples were Wisconsin and Massachusetts, have taken varying approaches to helping reunite people with their abandoned retirement plan assets. Can you talk about that dimension?
Mukherjee: Some states when you have unclaimed property, like Wisconsin’s one of these where they will, if they see you as a taxpayer in the state, they will just automatically send you a check. So, using Social Security numbers, they will understand that OK, this person has unclaimed property, I know where this individual lives, because they just filed taxes. So, they’ll just send a check that automatically reunites that unclaimed property with the owner. In most states, however, like Massachusetts, and really all others, the property goes to the central unclaimed property fund. And then, they try to reunite people by putting in names in newspapers I think a couple times a year. There’s a large list of names that are published in state newspapers saying these individuals have unclaimed property. Every state has a searchable database by which you can go in and see if you have unclaimed property. So, there are ways that states try to help. But I think the problem with particularly these retirement accounts is that most of them don’t end up in unclaimed property to begin with. I think the stuff that is an unclaimed property is really interesting. And there’s such a large variety of assets there. But as far as retirement plans go, mostly, they stay with the plans that they were originally in. They don’t end up being sent over to unclaimed property departments.
Ptak: One of your team’s conclusions from this research was that defaulting participants in new plans through automatic enrollment seems to be exacerbating the problem of abandoned accounts. Can you expand on that?
Mukherjee: I should pause for a second, and maybe talk about the motivation more generally for this paper, and which links to this. I think a big motivation is just that individuals today have more and more jobs as they enter retirement. They’ve experienced more and more jobs, and each of these jobs may have had retirement plans. And so, the question is, really, do people pay attention and consolidate and manage their various retirement accounts? It’s much more fragmented now than it might have been decades ago, where individuals just had one job and one pension, or one main account to manage. Now, I think the average individual entering retirement has had 12 jobs with retirement accounts. And so, the question is, do you really know about all of them? Do you manage them?
And so when we’re looking at the automatic-enrollment aspect, one of the huge, huge pushes that has been very successful in raising retirement savings has been automatic enrollment, where employers from day one of you joining a job will default you into some kind of savings. So it’s cut out of your paycheck and into some kind of a 401(k) or 403(b) plan where you’re saving for retirement. And what we find in the paper is at least for the lower balances, where people are leaving behind savings from an old job between $1 and $5,000, we find that a lot of those types of plans seem to have abandoned rates at, I think it was close to 30%. And we think a lot of the problem might stem from people never even knowing that they were in a retirement account. So clearly, automatically enrolling people is beneficial in the sense that people now have savings, at least passively, that they’re accruing into retirement. But a downside, at least as people job hop is that you’re getting these payroll deductions for benefits that you might not even know about.
And when we think about people not claiming these later in life, it’s a little bit different than not claiming something like unemployment insurance. It’s your own savings that were cut from your paycheck that you’re not accessing. And so, we think defaults are clearly important. They play a big role in getting people engaged with retirement saving. But we have to be a little bit careful about it, especially for people with smaller balances from maybe short-term jobs where maybe the individual didn’t have the attention or the financial knowledge to really understand their own benefits. And these savings might end up lost forever.
Benz: You examined the factors that tend to be associated and not associated with abandoning a retirement account. Account size, which you’ve already alluded to, sounds like it’s pretty important. So, if I have a significant amount of money in a plan, I’m more likely to pay attention to it and know that it’s there. But what other factors are in the mix that you looked at in the paper?
Mukherjee: There’s a couple. I think one is financial sophistication. This project was quite exciting because one of the co-authors, Lucas Goodman, he’s at the Office of Tax Analysis. So, we had access to really all U.S. tax records for this project. We could measure financial sophistication by looking at, did you file a tax return? Did you report investment income? Things like that. And there we see that at least some of these measures of financial sophistication are negatively related to abandonment. Some other factors were, I think, the plan’s ZIP code. So essentially, there’s a lot of plan variation, like I was saying earlier. Some plans, if they’re well-resourced, will look for you. And thus, that account will be less likely to be abandoned. But other plans, if they’re just doing the bare minimum of sending a notice to their last known address, if you had your account with that plan, you may end up much more likely to have an abandoned account. So, we think financial sophistication is a big one, and the plan characteristics or plan location appears to be a big one as well.
Ptak: Is there any way that defaults could better serve participants to help address this problem we’ve been talking about? For example, put in place defaults for the time when the participant separates from service, not just when they’re hired?
Mukherjee: I think this is a great question and something that people are thinking a lot about. There are some defaults when people separate. So that’s partly where we got this variation from, in the sense of if you leave behind less than $1,000 in retirement savings at a job, you are automatically cashed out with the penalty. If you leave behind $1,000 and $5,000, you’re automatically put into a balance-preserving plan. And if you leave behind more than $5,000, you get to stay with the plan. So, there are already some defaults. I think other ideas that could work maybe is increasing the threshold for cash-outs without penalty. If you’re leaving a job with a small pocket of savings, it’s not likely to be very meaningful as it rolls into retirement, particularly as you think about all the different management fees. Maybe allowing people to just cash out some of those savings with the lower penalty could be helpful. Maybe forcing people to engage in some way, or increasing the amount of time you have to roll over retirement savings—things like that could be certainly helpful.
But there is this general tension that defaults are intended to be passive. Like they’re intended to not get your attention. That’s partly why they work. But then as people at some point need to awaken to shifting from saving to using up that money, then you suddenly have to remember that you have all these savings or be reminded that you have all these savings. And so, I think some kind of a default that maybe when somebody separates from service that at least gets your attention in some way as to your benefits could certainly be very helpful.
Benz: I’m curious, and it’s really beyond the scope of this paper, but did the work in this area get you thinking about how our system could better serve people who are in the decumulation phase? You mentioned earlier that most of us have many jobs throughout our lifetimes. It’s just a lot more complicated to figure out retirement decumulation. Do you feel like there needs to be more work done at that life stage versus at the front end where it seems like we’ve solved for a lot of the problems in terms of getting people to save better and sooner? It’s more in the decumulation phase where people could really use some help.
Mukherjee: Yeah, I think that’s definitely part of the motivation for this project. My Ph.D. advisor, Olivia Mitchell, always told me decumulation phase is where we don’t know as much. And so, I’ve always had an eye toward questions in that space. And I think definitely part of decumulation is just knowing your assets. To begin with, that’s partly where this project came into play. And I think there’s a huge interest in there. Like, even in terms of how people engage with their benefits. You do take a lump sum out; do you take an annuity? Do you take some kind of a regular monthly payment until the money draws out? I think there’s a lot of questions around how people can best decumulate. And I know there’s a strong financial advising space there. And I think there’s lots more work to do on how to get people to do it optimally. My own work has been interested in step one of that, which is how do you at least know exactly what you’ve had as you enter retirement, which seems kind of obvious. But I think with all the different hassle costs, and employers you might have worked for may no longer exist, so the plan that you had may have changed hands multiple times. I’m interested in all those hassle costs and frictions and just managing and keeping track of your accounts to begin with, which informed the decumulation phase.
Ptak: I wanted to shift and talk about wealth inequality in retirement. It’s been another focus in your research. That people with more wealth tend to live longer, I think has been well established. You worked on some recent research that digs into that a bit further, however. Specifically, you and your co-researchers found that not only do wealthier people live longer, but they also have more healthy disability-free years than lower income people do. Can you discuss what’s going on there?
Mukherjee: This paper actually just got published in the Journal of Health Economics, just this week. So, we’re excited about this work. In this paper, we were really interested in the relationship of wealth with the quality of longevity after age 65. I think certainly our motivation, as you said, the fact that people with wealth live longer is known, but how are those longer lives broken down into healthy and unhealthy years, or work and not working years was not known. And so for our project, we were really interested, for example, do the most wealthy live very long because they’re able to extend six states of life, like by having access to good healthcare, or is it that they are just generally healthier and experience also a lot of good years, post age 65? This is one of my first descriptive type of papers. We’re really just interested in laying down some facts about how people live after age 65 across the wealth quartile and across years in the U.S.
And here we find that the wealthiest quartile of people do live, not just longer lives after age 65, but it’s also healthier lives, and years that enable them to work longer. We’re really seeing, when we think about inequality in the U.S., we’re seeing a lot of these inputs to inequality where the wealthiest can keep working for longer, they can keep having healthier lives, keep drawing down their savings over a longer period of time. We’re seeing some evidence of deepening inequality in the U.S. through this study.
Benz: I’m curious, and this is something you looked at in the paper, can you talk about wealth and education and whether they’re interchangeable in this discussion of longevity? If someone has high educational attainment, but maybe they’re not in the top quartile for wealth, are they likely to exhibit some of these same patterns in terms of disability-free, healthy life expectancy as the people who have high incomes and maybe are top quartile from a standpoint of income?
Mukherjee: So this is something, we looked at in the paper, as you noted, we looked at education, mostly because a challenge with looking at the effect of wealth on health is that they are interrelated. Let’s say you have a disease at age 40, you may draw down your wealth to take care of that disease and to manage that disease. Maybe that disease also shortens your life span. So, what we might incorrectly see in the data is that people who arrive with less wealth to age 65, when we see that they live less long afterward, maybe it’s because they had a health shock earlier in life. And so, this deep connection between health and wealth is something that makes it hard to make statements about the effect of wealth on health in older ages. What we did in the paper is that we looked at education and the reason why education is often, you could say preferred, but at least interesting, is because you can’t spend down your education. So, if you get a big disease at age 40, you can’t change your education.
You can’t spend it to take care of your health. And so, for that reason, we think education is another proxy for socioeconomic status, just like wealth is, that is usually fixed earlier in life. And when we look at the trends of healthy life expectancy with education instead of wealth, we find the same types of patterns. So people who are highest education, that’s who a lot of these gains in life expectancy, healthy life expectancy and extended years of work, are accruing to. I wouldn’t say that they’re fully interchangeable because they’re not exactly the same for all the reasons you might imagine. You could be wealthy without education, you could have high education and not high wealth and so forth, but they tend to both select on, they’re both imperfect proxies for some measure of socioeconomic status, which is maybe what really is related to these outcomes we’re interested in. So, for that reason, education and wealth do behave similarly when we look at their effect on disability-free life expectancy.
Ptak: Let’s discuss the lowest-income quartile, if we can, where there’s a connection to your other work about opioids. That cohort has been hit disproportionately by so-called deaths of despair over the past few decades, including drug-related deaths and suicide. Can you discuss what you’re seeing in that part of the population?
Mukherjee: This is a massive area of work in economics and our recent Nobel Prize to also focus on a lot of this work. In my particular paper that we’re talking about, we condition on living to age 65. We look at people after age 65. So, a lot of this is resolved by that age. In a way, I think the life expectancy for people who have drug addiction, opioid abuse, and so forth is quite low, unfortunately. And many of those individuals wouldn’t be in our sample. But certainly, more broadly speaking, I think these have driven, I think, some of the first declines in longevity in the U.S. in a long time. They’re hugely important, large enough to affect kind of the aggregate statistics on life expectancy, even when cut by subgroups. And something that I think we’re still grappling with, the opioid crisis continues. And so, I don’t think this has been solved in any way. And something that is important, and I think really complicates a lot of the analysis that I and others are trying to do in an important way, we try to address it by looking at different cuts of the population and so forth.
But really, these drug-related deaths are quite … I think what’s unique about them, from my own prior work and others’ prior work, is that they cut across all social groups, they cut across race, they cut across everything. So, it’s not as easy to study. And I think there’s a lot of interest in trying to better understand policies around the opioid epidemic and how we can contain it, and consequently reduce drug-related deaths and suicide. Some of my own prior work has looked at opioid policies, like prescribing limits, and particularly like triplicate state protection, where some states already had rules that made it much harder to prescribe opioids than in other states. And those seem to be very powerful and impactful. But yeah, I think there’s no solution here; this is something that’s ongoing.
Benz: Thinking broadly about longevity, is it too simplistic to think that the lower income, say, quartile of our population has experienced some declines in longevity, whereas the higher income, say, quartile, they’ve just sailed along and even had longevity gains through this period, even though longevity broadly for our population seems to have stalled out or maybe even gone downward a little bit?
Mukherjee: I think that what you summarized is really what this research finds. That these gains that we’re having in longevity are accruing only to the top-wealth quartile. So, it’s in a way not simplistic. That is the basic finding. And I think, related to the previous question, I think the lowest-income quartile has experienced either stagnation or maybe some decline due to all these other factors going on. But I think that fact that a lot of these gains are only really accruing to the richest people in society, highlights some need to put attention on that. Why is that? And how can we experience gains more equally as society?
Ptak: Your research noted that people in the lowest-income quartile tended to overestimate their life expectancy. What are the implications of that tendency?
Mukherjee: We looked at some of these questions around what we call sort of subjective mortality and subjective expectations about living and healthy living. And we do find some excess optimism among the lowest-income quartile. I think the main implications there are really around saving and maybe health investments. So if you think that you’re going to live a lot longer than you will, maybe you oversave, maybe you don’t invest in health. I think that’s the bigger one. Maybe you don’t invest in health as much as you should. If you wanted to change that trajectory, maybe there’s some really great returns to health investments today that people are not making because they think they’ll be OK. Those are the most salient ones that come to mind. But there could certainly be other implications in terms of how people plan even, caregiving. You might buy long-term care insurance, maybe you don’t need to, and people don’t overall. But I think generally it highlights some disconnect between how these gains in longevity are being spread across the population. People think that they’re being more evenly spread than they are.
Benz: I wanted to discuss the Social Security dimension of this. Social Security I think is, a lot of us would agree, that the best-performing part of our system for retirees. It’s progressive and it’s designed to replace a higher share of lower-income workers’ pay than is the case for higher-income workers. But the fact is that wealthier people live longer, and they often get more years of income from Social Security and that undermines that progressive nature. So, are there any adjustments to Social Security that you think might adjust that, acknowledging that you’re not an expert in Social Security? But do you have any thoughts on that?
Mukherjee: It’s an interesting question. There are ways you could change. I think the angle that I have been more interested in is how do we get the gains in longevity to be more spread across the population so that Social Security is in fact progressive as it’s intended to be? But I think there could be ways in which maybe at the time that you decide on your claiming you are given some accurate life expectancy information and allowed to make decisions around distributions. I think the concern has been generally that people claim too early and take too much lump sum-type distributions rather than annuities. I wouldn’t say this is an issue maybe as much in practice as it is by design. So, in principle, if everybody was taking an annuity at the right time at the normal retirement age, the full retirement age, then you might see that the wealthier people end up getting more years of the annuity. I think in practice, many people claim early, many people take lump sums. So, there’s some undoing of that concern.
Ptak: One point that some of our previous guests have made is that retirees are generally pretty happy across the wealth spectrum. And that undermines the notion that the U.S. is in the midst of or on the brink of a retirement crisis. How would you respond to assertions like those that, retirees are a happy lot in general?
Mukherjee: That’s a good thing, right? Overall, I think it’s good. I’m a faculty member in our risk and insurance department. And so being, I think, around insurance economics constantly, it’s the risks that everybody’s happy until something happens. So, I think there are these risks of, you become disabled, or you thought your kids would do caregiving for you, but they’re not going to anymore. There are risks of you can no longer be independent; you suddenly don’t have enough money, like nursing home care costs are much larger than you anticipated. I think all those types of risks are really where at least my research hopes to play a role. It’s one thing to be happy, and it’s good to be happy. But I do think we’re in a crisis in the sense that I think the savings that people have may not be sufficient to afford the types of nursing home care, types of care for the duration that people might want them. And so, I think at that point, for example, some of my prior research looks at what happens. So, it may be then that the burden shifts to the next generation, maybe. In particular, it’s often like an adult daughter who steps out of the workforce to do caregiving.
So that could be a solution. But then you’re perpetuating some inequalities to the next generation as a way to deal with some gaps in retirement planning from the previous generation. I think people may be happy, but I think how people are equipped to handle the risks. Some are uncertain, but more and more you can predict on some level, whether you might need nursing home care, and whether you can afford it. I think those are where the happiness maybe is less stable, and more critical to really think about planning for those tail risks.
Benz: I would love to dial out and look beyond the U.S. and ask, to what extent are some of the trends that we’ve been discussing the relationship between wealth and longevity and healthy lifetimes? To what extent are those a U.S. phenomenon or a U.S. problem? And to what extent do other countries, I suppose, especially developed countries, have a similar pattern in play?
Mukherjee: From our paper, when we did the review, it does seem to be a general developed-country fact. So not everybody has looked at specifically the breakdown of longevity post 65, as relates to wealth as we have. But generally, I think this is something true in developed countries. And so, this relationship between wealth and longevity, but also wealth and healthy longevity, and wealth and maybe the ability to work seems to be strong and growing over time, and present in modern economies. And so, when we think about that, I think the real way to maybe begin to address it is to think about the healthcare system. Who goes to the doctor, who gets treatment? I think one of the studies, and here, I’m stepping out of my own research for a bit. But I think even things like statins, which are known to be life extending, are they prescribed across the wealth spectrum? And I don’t know the answers to those things. But my guess is that management of chronic disease is probably a big factor in these gradients in the U.S. and beyond. And so, I think these are prevalent enough and stable enough in different settings that we have to now think harder about why this is.
Ptak: You’ve mentioned long-term care a couple times during our conversation, so wanted to turn to that for a few moments. Some of your research has looked at Medicaid and long-term care, and specifically whether people offload assets to become Medicare eligible. What did you find?
Mukherjee: This was a paper that we looked at the specific rule that Medicaid has, where you can become eligible for Medicaid only once you have, net of your house and several exclusions, have $2,000 in assets. And so, we find that people who exhibit some sense that they might need nursing home care in the future. So, this was a survey question, a dataset, they asked: With what probability do you think you’ll need nursing home care in the future? And we find that people who anticipate needing it do give more money to their kids in advance. And so we find that to be some evidence that people are offloading their assets by giving them to children to become Medicaid eligible for when they do need long-term care, because otherwise you have to give it to the nursing home.
You can spend it down through either nursing home care, or you can spend it down by just giving it away and then becoming eligible. There are some policies to try and limit that behavior. Because it’s not the spirit of the asset test. And so, what we find is that in the policy where they basically changed the period in which they will look to see whether you’ve done things like that. Whether you’ve just given away money to kids, whether we see the Deficit Reduction Act change that look-back period from three years to five years. And in response, we see that people actually do offload assets earlier than five years. So, people are paying attention to these look-back periods and thinking about depleting assets in a way that can make you eligible for nursing home care. I think a benefit of giving to your children, although you lose some agency over it, is that if it turns out maybe you don’t need nursing home care, you might be able to get some of that money back either directly or in the form of care. And so, we find some kind of interplay between Medicaid rules and people’s management of assets.
Benz: Another of your research interests is financial literacy and also health literacy and how to improve it. And you had a really interesting paper where you looked at someone’s savviness with respect to being able to do online searching and how that affects their literacy level, say financial literacy. Can you discuss that?
Mukherjee: This was a paper that I was really excited about in part because it gets at some of my longer-term goals of trying to think about the hassle costs in healthcare and in financial management. There’s just so many logins now for every single account, do people really figure that out? Do they create them, remember them, log in, roll things over, keep track of their savings, things like that. And so I think in this paper, our main interest was A, to really document to what degree do people, at least older people—again, this is using the health and retirement study, which focuses on older individuals in the U.S.—to what degree do people exhibit internet literacy? And then within that, how does being literate online affect your financial literacy and your health literacy? And so we were just interested … It seems like there is quite a strong relationship—people who are better at the internet, at navigating the internet, logging into things, being able to search for information, they do seem to have higher financial literacy and higher health literacy. And we think that’s important. In the sense when we think about these inequities and the digital divide and all that stuff, both geographically and among older people and younger people, being able to stay technologically engaged seems to help inform changing information about finances and health. So, it does seem to matter quite a lot.
Ptak: Is it possible that this problem is exclusive to the current population of older adults, some of whom have only been using the internet for part of their lives and will improve with each successive generation?
Mukherjee: I think certainly, the comfort with technology increases with each generation, but I think there will always be probably a gradient between younger and older and adoption and comfort with technology. Like just yesterday, I was at Whole Foods, and they asked me if I wanted to enable my thumbprint for pay, and I said no. But maybe a younger person would say yes, and if they say yes everywhere, maybe that makes their financial management much easier. You can just use biometrics to log in to everything, and maybe older people will be slower to adopt that type of thing. So I think that there will always be some kind of gradient. But certainly you’re right, like our generation growing older will be very comfortable with email and all that stuff, but maybe not with the most newest forms—like the multiple-factor authentication is very common. I don’t know the adoption of that across the age gradient. Or, just keeping your finances secure but also maintaining access to them—I think will improve, but there’s likely to stay some divide, both among young and old, and also among old in terms of the comfort with that technology.
Benz: I wanted to ask about the role of career in all of this. It seems like if you have a job in the information economy, you’re just naturally going to spend more time doing searches on this or that, and you’d be better at it when you need information on other things, related to your finances, for example. Can you discuss that? And also, I know there’s been research done on the fact that people with certain types of jobs, desk jobs, for example, oftentimes just have more discretionary time to do things. They’re related, personal, details, things they need to follow up on or call a doctor or whatever it is. The job just gives you more leeway to do that stuff. Can you talk about how career fits into all this?
Mukherjee: I don’t have my own research on this, but what you’re saying very much resonates. I think in line with all these hassle costs, in terms of how hard it is to navigate and keep track of everything. Certainly, people who have both the literacy to do so in terms of, not just reading, writing literacy, but literacy of really the online portals and being able to maybe have a password manager, things like that. I think certainly people with jobs that enable being able to do some of that more easily, either because you’re doing it anyways for work or because you have some slack time during the day—you’re already at your computer, I think certainly helps. And I think that’s really the markers of that in our data when we look at them is sometimes in financial literacy. And that seems to mitigate more of it. So if somebody’s very financially literate but does not have a desk job, that seems to be OK, but often, as you mentioned, those might be correlated. But certainly, no, I think staying employed in general—prior research has shown that that itself has beneficial effects on mortality and other stuff. So it’s maybe useful to stay employed for cognitive reasons and also to maintain your skills at managing your own personal finances.
Ptak: How do you decide which topics are worthy of further research?
Mukherjee: That’s a good question. I’m not sure. I make my own decisions. Ideally, I think the research question that I’m studying is interesting regardless of the finding. So, I try not to pick questions where I’ll be disappointed if the data shows something. It should be interesting either way, so sometimes that leads me to a descriptive paper like the health-wealth gradients that we discussed earlier. Sometimes it leads me … The abandoned accounts paper, for example, there was part of it was descriptive. We just knew nothing about the extent of abandoned accounts, but then beyond that, we were able to do a little bit more.
And beyond that, I’m an empirical microeconomist, so I try hard to look for questions that can be tested with appropriate data. Ideally, there’s some sorts of variation in the data that allows us to determine some kind of causal relationship. So, for example, in the abandoned accounts paper, there are these thresholds at $1,000 and $5,000 for when you’re leaving behind money with your old job at which you get treated differently. You get a cash-out, or you stay with the plan, or you get into this other plan that’s created by default. So, those types of thresholds really allowed us to learn a lot about, for example, the role of defaults and eventual abandonment. But I think keeping an eye out for these policy cutoffs and appropriate data. And I think most of all the question that I find interesting because a lot of my projects, they end up as being a little line on a resume, but they sometimes have three years of phone calls and research behind them, three or more years. And so, I think it should be something that also I enjoy reading about and learning about, not just at work, but just generally thinking about.
Benz: You referenced household finance earlier and maybe you can talk about what falls under that umbrella? Also, do you think that some realms of household finance have perhaps been eclipsed by the whole retirement and investing discussion? Do we maybe underplay some of the smaller decisions that households make with respect to their finances?
Mukherjee: My own research agenda for the next several years, I think I’m really interested in these. I’ve referenced them a couple times in this conversation, but I call them “hassle costs,”—this information or management cost, we could think about it in terms of all the different benefits we have. I think, there’s just so much logging in and logging out, not just in finance, but you see it a lot in healthcare too. Like for every doctor’s appointment, there’s five different logins from different providers. And so, I think trying to be reasonable about what people can actually do in that space and manage it over an entire lifetime, and how to reduce those hassle costs is really quite interesting for me. And also thinking about how much hassle costs prevent people from engaging with their own finances.
Like you couldn’t figure it out or you delay or procrastinate or something because it’s unpleasant. And I think the other area that seems to have been under-researched that I’m also interested to study is this idea of changing your habit from saving to spending. I think many people, especially among maybe the more wealthy—you get so trained to save that you’re scared to ever spend. So, thinking about how people make decisions about that juncture that, OK, now I’m officially going to go from saving to spending—when that is for an individual, for a couple, for a family, what unit are people thinking about as they make that decision—I think is not heavily studied. I think we’ve gotten good at getting people to save or at least better. But in terms of thinking about the decumulation, as you said, one of the first questions there is, just when does decumulation begin? And I think evidence shows it’s not exactly just at retirement. You may have adult children you’re thinking about, you may have a spouse who’s at a different career juncture than you. So, thinking about how people actually decide when to make that switch has not been heavily studied.
Benz: Well, Dr. Mukherjee, we are interested in seeing your forthcoming research on that topic. Thank you so much for taking time out of your schedule to be with us today.
Mukherjee: Thanks again for having me. This was really fun for me to reflect on the different research questions.
Ptak: Thanks again.
Benz: Thanks for joining us on The Long View. If you could, please take a minute to subscribe to and rate the podcast on Apple, Spotify, or wherever you get your podcasts.
You can follow us on Twitter @Christine_Benz.
Ptak: And @Syouth1, which is S-Y-O-U-T-H and the number 1.
Benz: George Castady is our engineer for the podcast and Kari Greczek produces the show notes each week.
Finally, we’d love to get your feedback. If you have a comment or a guest idea, please email us at TheLongView@Morningstar.com. Until next time, thanks for joining us.
(Disclaimer: This recording is for informational purposes only and should not be considered investment advice. Opinions expressed are as of the date of recording. Such opinions are subject to change. The views and opinions of guests on this program are not necessarily those of Morningstar, Inc. and its affiliates. While this guest may license or offer products and services of Morningstar and its affiliates, unless otherwise stated, he/she is not affiliated with Morningstar and its affiliates. Morningstar does not guarantee the accuracy, or the completeness of the data presented herein. Jeff Ptak is an employee of Morningstar Research Services LLC. Morningstar Research Services is a subsidiary of Morningstar, Inc. and is registered with the U.S. Securities and Exchange Commission. Morningstar Research Services shall not be responsible for any trading decisions, damages or other losses resulting from or related to the information, data analysis, or opinions, or their use. Past performance is not a guarantee of future results. All investments are subject to investment risk, including possible loss of principal. Individuals should seriously consider if an investment is suitable for them by referencing their own financial position, investment objectives and risk profile before making any investment decision.)