NEW: Episode Transcript

Rick Unser:    Well, Josh, welcome back to the podcast. You've been on the podcast so many times I've lost count. But the good news is, I always have so much fun talking to you. So I'm really looking forward to chatting about this report that you guys put out about the longterm impacts of automatic enrollment.

Josh Dietch:    Thanks, I'm happy to be here. And similarly, I've had a lot of fun doing past podcasts, so it was a no brainer to come on again.

Rick Unser:    And I got to tell you, after having so many conversations with you, as I was reading through the report that you guys put out, I couldn't help but hear your voice in that. So, that made it even more fun to read. All right, so I guess what I thought might be interesting is there was a couple of statements that you guys made within the report that I think are just really interesting and I'd love to get you to just kind of elaborate on that. So the first one that kind of caught my eye is the evidence suggests that auto-enrollment alone does not create healthy longterm financial behaviors. In fact, the opposite is true. I mean, I'm a huge fan of automatic enrollment and I mean that seems like a pretty strong statement that also might go counterintuitive to what some people have thought. So can you maybe give us a little context on that one, or how did you guys come to that conclusion or statement in this process?

Josh Dietch:    Sure. Well I don't know that I'd go quite as far as saying that auto-enrollment results in unhealthy financial behaviors. But I think what the data illustrates is that there are consequences both intended and unintended when makes plan design decisions. And maybe it's helpful to just maybe kind of take a step back. How do we even start approaching this question? When I first started at T.Rowe, well over about it's coming up on two and a quarter years ago. I was at a policy or an academic forum actually in DC, and I was sitting there for a full day of listening to academics show their latest and greatest research on retirement.

Josh Dietch:    And afterwards this guy came up to me, he said, "I'm running my dissertation on auto features, automatic enrollment specifically, and I'm looking for some record keeping data." And I was so new in my tenure. I didn't know enough to just say no. So was really interested, "What are you looking at?" He said, "Well, I'm wondering if auto-enrollment actually has its intended effect over a working lifetime. What's the longterm effect? We know that it increases participation, because you take people that are not saving and we forced them to save, but does it actually result in lasting behaviors?" And I thought, wow. That's really interesting because it's sort of if you can answer that question, then you can do some pretty amazing modeling as to if a plan changes its design, what's the financial impact?

Josh Dietch:    Because you can predict the behavior. So what we did was we shared basically 10 years worth of record keeping data. So obviously all the participant indicative data is anonymized. So there's no way to tell who these people are and he ran several experiments on the data and you're right. There are some things that are counterintuitive to what you might expect. For example, who does auto-enrollment ultimately benefit? The other thing that was kind of counterintuitive is what are the lasting impacts in terms of behavior? So for example, and he didn't do this with our data exactly, he used data from the UK, the nest system where he could actually track individuals across employers.

Josh Dietch:    And so you had some plans where there was automatic enrollment and other plans where there was not and as you know, most of us are not likely to stay with the same employer over the course of our working lifetimes. Some plans offer it, some plans don't. And the researcher's name just for clarity purposes is Todd Zucman who now is with the National Bureau of Economic Research in MIT Sloan School of Management. What he found was that if I'm automatically enrolled and I go to my next employer that doesn't offer automatic enrollment, I'm less likely to participate than if I came from an employer that didn't offer auto-enrollment and the next employer didn't offer it either.

Josh Dietch:    So put it another way, I'm either conditioned to proactively opt in or I'm conditioned to assume that my employer is going to enroll me. And that has lasting consequence over one's working lifetime, because one can result in missed savings opportunities. And as we all know, one of the primary benefits of participating in a retirement plan early is compounding returns. And so that has a consequence. So the takeaway from that is that for employers whether or not they offer auto-enrollment is obviously a design choice, but it has impact not only on the people they employed today or in the future, but the people that they're recruiting into the firm and the previous experience that those employees might've had.

Rick Unser:    Now well said, and I saw a lot of what you said in the report too, and as I was reading it, I was like, wow, there's some really interesting conclusions here that don't get talked about a lot, but also some that, like you said, are a little counterintuitive. And I've been... One of the other ones I pulled out is you guys referenced an automatic enrollment paradox where though auto-enrollment is framed as a means to increase savings, the reality is that it's a better means to increase participation. Auto-enrollment suppresses savings compared to opt-in regimes.

Josh Dietch:    Yeah, that's maybe counterintuitive, but in fact, true. And when you start to think it through, it makes sense because what we know is if I opt you in that 3%, there's a better chance that you stay at 3% than if you have to choose at what rate you want to save. And so, there's been a lot of discussion over the last, I would say five plus years about how do we move those default rates beyond 3% because again, here's another example of perhaps unintended consequence. I'll ask you a rhetorical question, but do you know why 3% was so popular?

Rick Unser:    Yeah. So I think 3% was referenced in something that the government put out when PPA or something was introduced.

Josh Dietch:    So you've taken my rhetorical question and you've gotten most of it right. So it's actually, it was in the model language or the example given in Revenue Ruling 98-30 that said, "You can auto..." And this was pre-PPA actually. This was the 1998 where they said, "You could auto-enroll somebody at 3%," and because they put it in ink and fired up the Xerox machine, that's what people went to because there was almost endorsed by the IRS.

Rick Unser:    Yeah. And then that was another interesting point that you guys brought up in the paper too, was that I think a lot of people would point to the Pension Protection Act as the birth of automatic enrollment or maybe when automatic enrollment really got its beginnings. But that's not necessarily the case.

Josh Dietch:    Yeah. And it's interesting. I'll take my example a little bit further. So you had Revenue Ruling 98-30 and then you had the sort of ERISA preemption with the QDIA and then you had the QACA guidance on the safe harbor for ADP testing. So suddenly that 3%, that model language and example became 6% because of the QACA, the Qualified Automatic Contribution Arrangement language that said, "You could start it three but you had to get to six within I think four years, but could go no higher than 10%," and what's interesting it's when we look at our own record keeping data, we basically see at least within our client base, basically 61% of our clients that we're using our own enrollment pre-PPA or when it was enacted, 61% were at 3% and now that's fallen to 31% and almost a full third actually are at 6%. Now, whether or not they're taking advantage of the safe harbor, I won't have an opinion on that, but what it shows you is the power of putting something in a ruling or in a guidance letter that sort of says, "Well, if you were to do this, this would be okay."

Rick Unser:    Yeah. And I think one thing that as people think about automatic enrollment and as we're evolving this conversation, moving those default rates from three to four to six or whatever, I think a lot of people struggle with, "Well, okay I get it 3% not going to get people to retirement, but I'm kind of worried about what the impact on our plan will be if we increase the default rates." And I think you guys had some conclusions that you came to, in the report about that as well.

Josh Dietch:    Yeah, so what we looked at is, okay, if you're at 3% what happens if you increase that by 1%. You move it to 4%. And what we saw is that the relationship is basically one to one. So if I go from three to four, basically my participation rate will drop by 1%. If I go from three to six, it'll drop by almost 3%. But the real question you have to ask yourself is like, is there a greater good? Yes, by participation rate went down. But the number of employees that are actually saving at more meaningful rates is higher, and so there's sort of a trade off that a plan sponsor needs to consider.

Rick Unser:    And let me just make sure I'm following from some of what you just said there. So I'm an employer, I'm automatically enrolling my employees at 3% and I have a 93% participation rate?

Josh Dietch:    Yeah.

​

Rick Unser:    If I'm following this statistical aspect of this or if I'm going based on what you just said and, "Hey Josh, you're working with us as a provider, as a strategic partner, whatever. What happens if we go from 3% to 6% as our default rate for automatic enrollment?" Your answer is, "Well, because you're going from three to six that's a three point increase. We could reasonably expect that your participation might go from 93 to 90%."

Josh Dietch:    Exactly.

Rick Unser:    Okay.

Josh Dietch:    And let's just make the math that even I can do in my head. If you've got a hundred employees, is the benefit to those that are at the default greater than the two that you've lost?

Rick Unser:    Yeah. I think that's an interesting conversation for a lot of employers and I don't know, based on some of my experience too, it might not even be that 1%, if you go down that route. So yeah, I think that's a really interesting way to look at it, is kind of going back to your comment there about what is the greater good here and philosophically where do you want to land on this spectrum?

Josh Dietch:    Yeah. And you raise a point that it's fundamental to this whole discussion. It's what's the philosophy or the goal that the employer or the plan sponsor is working towards? Because the plan design really needs to reflect what that philosophy or goal is. That's the means by which they achieve it. We sort of have this point of view, if you will, what's your preference? What's the strategy to get there? And then how do you express that in terms of plan design.

Rick Unser:    And I think that, that has such a bigger, broader reaching benefit to it. And I don't know that a lot of employers kind of go through that philosophical aspect. And some of it I think is time-related. Some of it is just priorities in running a business, but you're 100% right. I think the more people that can take a minute and reflect and take a breath and say, "Okay, well, we have this retirement plan that we offer to our employees.

Rick Unser:    What are we trying to accomplish? How does it factor into recruiting retention? How does it factor into retirement outcomes? Is our plan something that we want people to stick around in after they leave us and use our plan as a retirement destination?" That came from a podcast interview I did yesterday, which was awesome. But I think some of those big picture questions are really, really important and can have a real impact and also can bring some clarity as well.

Josh Dietch:    Yeah, and those are conversations whether it's us and engaging with our clients or advisors or consultants. These are the conversations that should be heard because otherwise a plan in its design is a means to an end. If you don't know what the end is, then you're probably going to miss the target.

Rick Unser:    Yeah. One other thing that caught my eye in the report, and I think this dovetails to back to something you started to talk about, but we didn't dive in on it. Is, do initial nudges go far enough or is there more to be done? The answer-

Josh Dietch:    Oh yeah. That's a question I didn't answer.

Rick Unser:    Yeah. And I think here, this is where you started going, which is the answer lies in who ultimately benefits from automatic enrollment and why? What does that mean? I feel like you guys did some pretty unique work there, and I'd love to hear a little more about why you did it and what you found out.

Josh Dietch:    Yeah. So our research partner, Dr. Chuckman, what he did in his experiments was he looked at employers that implemented auto-enrollment at 3% and what do did was he tracked employees 12 months prior to the implementation and 36 months after. That way you could sort of track these people because remember he had 10 years worth of data, and what he wanted to understand was, what was the difference in behavior between those plans that offered auto-enrollment and the behavior before they had it with opt-in? And what he found, and this is really interesting, he creates this analysis where he looked at the ratio of savings to wages.

Josh Dietch:    And so when you have a range you can sort of carve it up however you want. But what he found was... Let's say you go below the median. So basically if it's a ratio of savings to wages, you can pretty much assume that it's minimal savings and low wages. And what he found is that if someone had to opt in, and this is at the 25th percentile, that over 36 months of cumulative savings would be 1.2%, as an average for that population. So what that is essentially telling you is that they're not saving. But if you were to look at it for those that experienced auto-enrollment, after 36 months it was 7.4%, was their cumulative savings.

Josh Dietch:    Now if we just look at the median, so for those that were auto enrolled, the cumulus savings at 3% was not surprisingly 9%, and those that had to opt in was actually 7.6%. Clearly not as much as those that were auto-enrolled, but what you begin to get a sense of is the answer to that. Who does it benefit and why? And the why here is the fact that they have some sort of economic elasticity. They have the wages to support savings of their own free volition. They make the choice to save, and they pretty much catch up over time. Sure that slope of savings is steeper, but the fact of the matter is that they do catch up. And if you go above the 75th percentile, they basically saved the same.

Josh Dietch:    And it's all above the default because they can save more than the default. And that gives you another insight that the default itself, if it's not set high enough, and there's a lot of research that illustrates this, it can actually serve as sort of a governor. If my employer's saying, 3% is it, then 3% must be a good number. And we know it's not. And so it goes back to the question, one of the unintended consequences of auto-enrollment, yet it's great at getting people to save that might not otherwise save. But for those that do save, it can actually tamp down the amount that they're saving because of an endorsement effect.

Rick Unser:    Yeah. And that kind of reminds me of another quote that I flagged in here, which is, "Auto-enrollment is a beginning, not an ending to create healthy longterm financial behavior among employees."

Josh Dietch:    Yeah. And obviously as an add on to that, when we have conversations around auto-enrollment, a fast follower to that conversation is auto escalation because that's the means by which you get people to save at more meaningful rates above the default, but giving people the optionality to opt out. It's sort of interesting. What I love about this is it's sort of the intersection of academia and practice. And Richard Thaler got the Nobel for this, but his collaborator was Cass Sunstein and in their book Nudge, they call this whole concept of auto-features. They framed it as libertarian paternalism. You're always free to say no.

Rick Unser:    Yeah, no, I had not heard that before, but it's a really good way to describe it.

Josh Dietch:    In retrospect, I wish I said, I said it. [inaudible 00:20:28].

Rick Unser:    Well, I know I've kind of flagged a couple things that I thought were eye-catching or interesting or maybe a little counterintuitive. What else did you guys review or what are some of the other conclusions that you came to that might be helpful employers to know about as relates to automatic enrollment, it's broader impacts, et cetera?

Josh Dietch:    Yeah. So a couple of things that I personally thought was interesting. We talked about sort of auto-enrollment being a learned behavior and higher defaults don't necessarily discourage savings. But one of the concepts that I think the insight that we just talked about where who benefits and why, it goes back to what's your philosophy? What are you trying to achieve with your plan and does that align to some sort of social preference? And what I mean by social preference is what's your approach towards the plan? Are you paternalistic?

Josh Dietch:    You want everybody to be treated equal? Do you want people to make their own decisions? Because those social preferences all have consequence as it relates to plan design. And let me give you an example, or I'll give you three examples. So let's say you believe that someone's choice to participate in a plan is in fact their choice. And you're not going to offer auto-enrollment, but you do offer a match. What you're essentially saying is, we'd prefer as a plan to spend our match dollars on older, higher paid employees because we know that older employees tend to participate more than younger employees. They're probably on more solid financial footing in life and retirement is not so far away.

Josh Dietch:    So they tend to save more because they're probably trying to catch up in some manner. So it's sort of, well, if I believe that it's someone's choice to participate, then I'm also saying that, that's how I want to spend my match budget. On the other end of the spectrum, if you say, "You know what, I want to treat everybody equally." Then that's a more of a paternalistic point of view. And you might set your match at the automatic enrollment default. And what you're essentially saying is, I want to treat my younger and lesser paid employees similarly to everybody. I'm going to take a little bit off the top and spread it wider. And that's more of a progressive approach.

Josh Dietch:    And in the middle, and this is our partner in this, Saha, he gave it a term and he's an economist, so this is a term only an economist can love, inequality adverse. And if I were to translate that to something that folks like you and me can understand, I'd call it a stretch match. I'll set the match threshold above the default so that there's some shared commitment to gaining those dollars. That to me was, it's a really interesting way of framing it because you sort of go from preference to how am I going to do this, to how does that translate to plan design?

Rick Unser:    I mean that was amazing and I just, I'm stealing all of that going forward. But that I especially liked the more easy definition of shared commitment. That's a really good way to phrase that. And I think, I mean, just using round numbers here, and I hate to stereotype, it's, "Hey, we match 50 on six or 50% of the first 6% somebody contributes and we're going to automatically enroll somebody at 3%." That's the shared commitment or a basic example of it. Probably a more optimal example would be a higher match target and a higher automatic enrollment start rate, but just as an example there.

Josh Dietch:    Yeah, I mean it's just like, it's a really simple and elegant way of sort of thinking through what's the sort of the yin and yang of I do this, this is what happens. And you can sort of start to see how that you might sort of apply that to not just sort of on paper scenarios, but in conversations.

Rick Unser:    And that's something too that I think as people are looking to communicate concepts with leadership teams, with management teams, with the C-suite, people might not be as hip to talk to about automatic enrollment or automatic increases or this level of target matching blah blah blah. But when you talk in some of those more broad terms, shared commitment, I think that's something that a management team can dig in and say, "Yeah, you know what, we like that, that makes sense and we get that," versus well, "We're going to automatically roll at this but we're going to match it that," I mean that maybe starts to get a little technical and you start losing some of the philosophical concepts behind it.

Josh Dietch:    Yeah. I mean because you can always model it, but unless you really understand what you're doing, the modeling itself has less relevance and it allows you to sort of have a more strategic discussion rather than you'll eventually get to the brass tacks of what's it going to cost.

Rick Unser:    Well that was certainly a good one. What else is in there that you think is important for people to pay attention to or maybe draw out of the conversation?

Josh Dietch:    So one of the things that Dr. Chuckman did with our data, and this was sort of going back to his basic thesis was, what's the lifetime effect of auto-enrollment? And that, again, it gets back to the who does it benefit and why? And so he developed this really robust modeling. And the interesting thing about modeling and the way that academics or economists look at this is, they're not just trying to create a black box, but they're trying to use data to see if the model can predict behavior. Because in our case, we gave him a data set of 4 million participant in records representing 600 plans over a 10 year period.

Josh Dietch:    So there was a lot of Plato so to speak, to sit down with. And so, he constructed this model that sort of took into account the characteristics of our retirement market. So that's people saving, that's social security, that's taxes. It takes into account personal preferences like I'd rather spend today versus save for tomorrow. Their willingness to take risks. Is there a switching cost to not saving to saving and vice versa. What does it cost me to live on a daily basis? Things like housing or other things related to that.

Josh Dietch:    Or even just what's the probability that I'm going to live to 80, and then lastly decisions around wealth. And so he constructed this model and we actually had him model it looking at if one more to invest in a... I'll call it a target date investment. So there's an asset allocation that shifts over time. And so, using the underlying participant data, he modeled this out to full retirement age. And what we see is that below the median, the lifetime effect of auto-enrollment compared to opt-in, I should say that's an important caveat, is not that significant. So, that first tells us one thing that the defaults in general are not high enough.

Rick Unser:    Well, and let me dig in on that a little bit because I think if I'm following you, and please correct me if I'm wrong, what you're saying is if someone has over their lifetime income that is in the lower 25th percentile, so you're... I don't know, 25, $30,000 equivalent somewhere in that range of income over their working life. Am I going down the right path on that?

Josh Dietch:    Yeah. Absolutely.

Rick Unser:    Then what to me, I'm hearing you say is, "Okay, if that individual or that group of individuals is automatically enrolled, their end result of replacement income in retirement doesn't really move that much because, A, they're getting a large benefit from social security as a percentage of their income-"

Josh Dietch:    No. It's actually the opposite. It's profoundly improved. Yeah. These people probably wouldn't be saving in the first place. And in fact, so like in the model that we provide in the research, we modeled it out at a 6% default and for the lowest paid people that could have an impact anywhere from 12 to 40% more wealth at retirement. I mean that's a huge difference. On the other hand at 6%, the lifetime effect for those above the median is not that high. So it shows what it really illustrates is that at some point there probably needs to be a consideration that just a blanket default doesn't really serve all constituencies equally.

Rick Unser:    I guess expanding on that, if I'm an employer and I have 5,000 employees and I have a whole range of people in terms of income and everything else potentially is the way I'm now thinking about automatic enrollment, that depending upon what income band you fall in, you might get enrolled at a different percent?

Josh Dietch:    I don't know that I'd be willing to go that far at this point, but I do think that how you educate and communicate with participants might vary. Because you sort of think about it this way. There's the effect of wealth over time, but there's also the impact or the relevant replacement that social security represents. Truthfully for the lowest paid employees, social security represents a higher percentage replacement of income in retirement than it does for higher paid employees. And so you may want to message differently based on that reality.

Rick Unser:    Yeah. And that was kind of where I was going with my statement earlier was I thought what you were saying was, yes automatic enrollment can have an impact on the kind of lower paid, but in the end social security is going to represent such a high percentage that it's marginal. Sorry, that's where I thought you were going. But again, you corrected me obviously.

Josh Dietch:    Yeah, I mean, and the other thing to consider just going back to what the data tells us is that if not for auto-enrollment, many of these folks who thought most probably wouldn't be saving. Or if they are saving, they're coming into the system much later in life where they've lost the benefit of compounding returns. So if you sort of think about it, if I sort of put the academic hat back on, one would say, this is a patient policy because it works really well over a long period of time because of that fact, that you... They've early and often and you reap the reward 40 years down the road.

Rick Unser:    That makes sense. Let me pick away at a couple maybe just statistical questions that we seem to get a lot. And I'd be curious because I've heard different things from different people on the podcast. What do you see right now as the percentage of plans that are using automatic enrollment and what direction is that trending? And then maybe the same question for auto increases?

Josh Dietch:    I have a good answer for the auto-enrollment, the auto increase it's a little different. I don't have exact figures, but what we see within our client base, roughly 60% of large plans, very large plans use auto-enrollment, but very few small employers do. And in some ways, I can think of a lot of sort of rational reasons for that. For a small employer, the cost of match is not... Well the cost of match for any employer is not insignificant. But again, you sort of get to who do you want to benefit from that match and why?

Josh Dietch:    And I think, large employers kind of look at this from a total compensation perspective and the match is just one element of their compensation philosophy, and it's easier to sort of build into a budget and they have people that actually can model that out or they have a consultant or advisor that can help them model that out. Whereas I think for a small employer there's a little bit more risk of the unknown involved. In terms of auto-escalation, I don't have the figures in front of me, but we definitely see a trend of greater adoption. The problem is that there's a much more firmly held belief around sort of a philosophical objection to making auto-escalation the default. Yeah. They're more likely to offer it, but as opt-in.

Rick Unser:    Yeah, no, that makes sense. And I guess, when you think about automatic enrollment, one thing I've heard from a couple of guests is we've kind of hit a plateau on the number of plans that are going to offer automatic enrollment. Did that come out in the data or have you seen that or do you believe that anecdotally in some of the things that you're seeing or do you think there's still more room for adoption and growth of automatic enrollment plans in the future?

Josh Dietch:    I mean there's clearly opportunity for growth because it would be growth among small plans. Would that move the needle in terms of participants that are impacted? Less so, because the market is just so top heavy in terms of the largest plans representing the vast majority of assets and participants, but the fewest number of plans. I agree with that. The fact of the matter is though is I think the challenge with smaller plans remains and from a provider's standpoint, there's certainly commercial opportunity in that. But I think more broadly, just as in terms of social policy, getting more people to save more, I'm not sure what the downside to that is, if we're in the business of helping people achieve financial security in retirement.

Rick Unser:    And we talked about this a little bit earlier, but one of the other questions that we get, and I think some of it is more honestly academic in nature than it is reality in nature, but is there a point of diminishing returns where you get either a default rate for automatic enrollment or a rate for auto-escalation where you get to that point where the opt-outs are much increased or much higher or kind of go away from that one to one relationship that we were talking about?

Josh Dietch:    Yeah. The amount of juice you get isn't worth the squeeze.

Rick Unser:    Yeah, exactly. And I'm just curious, did you guys look at that at all or is there anything that you have anecdotally where, "Hey, 4% is fine, 6% is fine, but 15%..." I'm obviously exaggerating, "... is probably too much."

Josh Dietch:    Yeah, it's interesting. So no, we didn't look at it, but I do have a point of view. What's interesting is that increasing the limit on defaults has been an item in some of the legislation that's been proposed over the past few years. And so I'm absolutely certain there is a theoretical limit. I don't know that from implementation standpoint where anywhere near where that might be. I mean it's interesting. Our financial planners kind of use as a rule of thumb that between employer and employee contributions you want to be able to save roughly 15%. So if plans 10 years ago were defaulting at 3% and now more and more they're moving to 6% clearly we've got some room to run before we run into that problem.

Rick Unser:    And another thing that just kind of jumped out at me in your report and maybe some of this ties together is where you talked about where do we stand in this broader conversation about retirement savings? And I think in a study you guys did in 2018 you said you guys surveyed a thousand participants on your record keeping platform and asked them how much should you be saving and how much are you saving? And I think you guys had some pretty interesting conclusions as a result of that survey in line of questioning.

Josh Dietch:    Yeah, there's a delta. There's no sugarcoating it. People intuitively know they should be saving more. But life intervenes. But what's interesting, and this is really interesting, what it also told us was... And I don't have the exact figures in front of me, but our participants were more likely to identify that 15% number than a similar study we did that was nationally representative. And what was fascinating to me was even though they might not be at 15%, I think on average they're at least within the sample of that survey, they were closer to 11%.

Josh Dietch:    They came up with the higher number than the nationally representative sample, which told us that the messaging that we put in front of them, they're hearing it. And so if they're absorbing the messaging that at least gives you an inclination that well now we just have to figure out how do we get them to do the next best thing for them? What's the next healthy behavior they can adopt so that they can save more and be closer to achieving some of their financial savings goals? To me it was sort of good news, bad news.

Rick Unser:    Yeah. And that was definitely one of the takeaways that I had, which is for a long, long time, 10% was what people had ingrained in their minds, et cetera. And as an industry we've been trying to move the needle closer to 15%. So I was happy to hear that, like you said, people are identifying with 15% and they're shifting their sites upward a little bit, acknowledging they might not be there, but at least shifting their sights upwards a little bit to recognize that, that's probably a more realistic target for savings rate if they want to replace a high enough percentage of their income when they want to make that transition into retirement.

Josh Dietch:    Yeah, I mean, one of the things just philosophically that we've strongly believe in, is that so much of the marketing that goes on within our industry is about scaring people into doing something. And let's face it, the numbers that one needs to achieve to retire or pay for healthcare in retirement or pay for college, they're big numbers. And if we just continue to scare people, they're just going to shut down. It's like how can you break down these really big complex problems and make it manageable? So one of the things that we did last year was we looked at... There's a lot of sort of, you need to have X number of dollars by such and such an age in order to be on track.

Josh Dietch:    And we just fundamentally believed that that was wrong. Sure, you needed significant savings at different levels over time, because ultimately that's what's going to enable your consumption in retirement, but it's a range. Because to use too many syllables, it's heterogeneous. Not everybody spends the same, in the fact of the matter is if you save less today, you can always save more tomorrow. We don't think that, that's the best way to do it, but we shouldn't shame people if that's what their best option is. And it's a way to sort of communicate with people that these are not insurmountable challenges, but there are things that you can plug away sort of slowly and steadily.

Josh Dietch:    I'll give you another great example of some work colleague of mine, Shibicto Bernagy, did on the cost of healthcare in retirement, and I won't stray too far, I promise. But the standard story is that you're going to need somewhere between 350 and $400,000 for a couple of pay for healthcare in retirement. And another rhetorical question, but what do you think the two biggest determinants of that number are?

Rick Unser:    How long you live.

Josh Dietch:    Yeah. That's one.

Rick Unser:    And how healthy you are not you are?

Josh Dietch:    Actually that is a good component. I was going to say inflation. So, that's the problem I get with asking a financial professional these questions. So what if I said to you, how much do you think you're going to pay your cable provider over retirement?

Rick Unser:    Yeah.

Josh Dietch:    $86,000 you know what? Healthcare seems like a pretty good deal. Moreover, just like cable, the bill for healthcare isn't due on day one. It's due monthly in the form of premium. And if you look at the breakdown between premium and out of pocket, three quarters of it to 80% of it is premium. And to your point, there is the volatility of shocks, if you will. And we've done some really good research on that as well, but it's like you can't just take these big numbers and go, "How do I deal with that?" You got to break it down so that people can sort of manage what's in front of them without being just so blown away that they shut down.

Rick Unser:    As a client and good friend of mine says, "How do you eat an elephant? One bite at a time."

Josh Dietch:    One bite at a time. Absolutely. So what you're saying is, we're both in the business of teaching people how to eat elephants.

Rick Unser:    Absolutely. I don't know that, that's very humane and friendly in reality, but as an analogy, I think it makes a lot of sense.

Josh Dietch:    Yeah, exactly. I'm not sure. I'm not sure if I've cited up for that one.

Rick Unser:    Well, buddy, this has been a ton of fun. I also know you guys are doing a lot of other great work. What comes to mind that either you've put out recently, or maybe that's sort of in the queue for people to look out for going forward, and is there a place to find it, whether it's old or maybe coming down the pipe?

Josh Dietch:    Yeah, so I just mentioned my colleague, Shibicto's work on retiree health care. He's got another paper coming out any day now on the impact of shocks. So you're 70 years old and you trip and fall and break a bone. What does the cost look like? What's the persistency of it? What's the probability of that event happening? Looking through different lenses, whether it be age, gender, the magnitude of the shock, it's a really interesting paper. And actually he's going to follow it up with a piece on what one can effect to spend in the last couple of years of life.

Josh Dietch:    And these are kind of... I don't want to say morbid, but they're not the most pleasant subjects, but they're all things to understand if you're going to plan for retirement. And it's a really interesting series. I'm actually starting to look at student loan debt. There's a lot of information out there that we know the magnitude of student loan is outstanding. It's one point $6 trillion. It's a big number. But I think what's not well understood is participants' or savers' behaviors. There's a lot of buzz in the industry about student loan debt repayment programs, facilitated through record-keepers or planned designs. But there's consequences to paying down debt instead of saving for retirement. And one way I sort of look at it is, if I get $20,000 worth of college loans, it's roughly a car payment, but if I had a hundred thousand dollars, that's a mortgage.

Josh Dietch:    And rationally my approach to dealing with each should be different, but they're not. There was some research done by Jeff Psalmsabarker at Boston college that showed the retirement savings wasn't really impacted by the magnitude of the student loan debt, which to me is an irrational response. And so what I'm trying to better understand is what people's attitudes towards that debt are, and how that translates to behavior. I hope to have something out in the first quarter of next year. We're just starting to do the research, sort of a standard process, review the academic literature, see what our record keeping system tells us and then we do some survey work.

Rick Unser:    And is to the extent that people want to find some of that, is that available on a non-firewalled or non-paywalled T.RowePrice website?

Josh Dietch:    Yeah. Absolutely. There's on the retirement plans. As a large company we have many websites and distribution channels but the vast majority of is available on the retirement plans website. I believe the URL is retirement for all, but if you just Google T.RowePrice retirement for all, it will take you to the site and it's under research findings. I also did a study about a year ago on looking at the relationship between a high functioning retirement plan and corporate profitability. You had sort of mentioned some of the standard reasons why someone might offer a plan and sort of attracting and retaining talent are two things that sort of come to mind.

Josh Dietch:    But if you sort of go back to that idea around a retirement plan is sort of a means to an end. And if you want to create engagement among employees, the reason for that engagement is you think it will help your business operate better. And we're able to show that there is in fact a correlation between the retirement planning and corporate profitability. It's not causal, it is a correlation. But again, it's another way of looking at the retirement plan as more central to a business. It's just one of many strategies that are things that a business can do to drive an outcome. So, that was another piece of interesting research. And in addition to those things, we'll keep coming up with ideas.

Josh Dietch:    As you mentioned earlier, we do an annual study that looks at savings. We've got some stuff we're working on right now that not only looks at people that are saving in retirement plans, but people that have decided to not participate in their employer's plan or another group that they don't have access to a plan. And what are some of the similarities and some dis-similarities between the two and how do you sort of provide the benefits to those that aren't currently benefiting from the system? Because we definitely do see 401k plan and DC plans in general as an incredible benefit that has long lasting impact, and sort of our goal not only commercially but just because it's the right thing to do, is to increase access, and the impact that those plans can have on people's lives.

Rick Unser:    Well said, buddy, as always, was a pleasure chatting and going through all this stuff. I really liked a lot of the conclusions you guys came to on auto-enrollment and we'll link to all that on the show page and everything else. But I really appreciate you taking the time and I can't wait to dive into some stuff down the road with you.

Josh Dietch:    Anytime. It was my pleasure and thank you for inviting me.