[Webinar] - Is Waiting Until 70 Always the Best Move? Discover the Risks Advisors Often Overlook

(00:06) Okay, welcome everyone to today's webinar on social security with Derek Tharp. We'll give everybody a couple minutes to get in the room. Hope everybody had a great Thanksgiving. We are uh got our first snow over the weekend here in Colorado and I think we're we're due for our first big snow tonight.
(00:33)  So, got to make sure I got the shovels out of the yard so they're not buried. We uh we've got our first snow day here, so my kids are home, too. Oh, so you had a real one today. Okay. Yeah. Nice. Anybody else? Any snow? Put it in the uh put it in the chat. Um little housekeeping before we get started. See, there's still people coming in. But, um, today's webinar is available for CFPCE.
(01:03)  So, make sure that you are on for 50 minutes or more. Um, and that you fill out the survey at the end. Um, please put your questions in the Q&A rather than in the chat. Feel free to use the chat. I see there's more snow days. Um, what is this snow day you speak of? Yes. I think we're we're due for like six inches tonight. That probably Well, it depends on the timing. We'll see.
(01:29)  Maybe it'll uh And we just had it dusting. Um so yeah, put put questions in the Q&A. I I think um Derek will be, you know, doing a full presentation, then we'll hit the the questions at the end. If there are questions that you um really want to see answered, go ahead and hit the uh the thumbs up, I think it is, or the whatever it is, the like um option, so that we can kind of see because we we typically are not going to get to all questions.
(01:56)  Um so, uh with that, so today we're talking about social security. This is kind of um I'd say one of the hottest topics I've seen in financial planning lately. Um and we've been talking a lot about it. We ran a um a master class over the summer. um which if you're interested in diving deeper into this topic, please check that out.
(02:17)  Um we can we can put the uh the link here in the uh in the chat. Um and it's one of the more lively places of of actual debate and disagreement, which I always like to see because it um it really means, you know, people care a lot. people, you know, this this matters and there are different um you know, opinions on how to properly um consider when to claim social security and some of the the math behind it and and really even just some of the conceptual stuff behind it.
(02:47)  So Derek has done I think you know recently more than anybody to dive really deeply into maybe the even some of the lesser considered parts of um of this topic and also do a good kind of survey of where people have been both in the financial planner world and in you know financial economists and all sorts of people. So I think this is going to be a really good uh really good webinar and so Derek take it away. Yeah.
(03:13)  Well, excited to be talking on this. Like Justin said, I think it's a it's an important topic. You know, really relevant for a lot of financial planners and our clients and thinking about how do we help people best navigate social security. And um you know, I'll say right off the bat, you know, we're really here thinking you're talking about rethinking this maybe delay to 70 rule.
(03:33)  I don't think um you know, for a lot of people it does make sense to delay. Um I'm certainly not here to say that delaying is a bad thing at all. can make a lot of sense. But I want to dive into maybe some of the the risks and the things that often aren't as maybe well appreciated as we um might consider.
(03:49)  So let's start off with a question just you know what is the right discount rate for a social security claiming analysis right? Might sound like a pretty simple question but I think you know actually as we dive into it there are some different thoughts on this. So probably the predominant view is that waiting until 70 is the best thing to do.
(04:07)  So there's a large segment both practitioners and researchers who proclaim that waiting is almost always the best route, right? We we see this very clearly. Um there might be some exceptions, maybe caveats like somebody has terminal illness or their spousal benefits, other things we might want to think about, but by and large there's a real strong tendency to say we really should be waiting until 70.
(04:30)  Um and I'd say particularly in the financial planning kind of industry realm, we see a lot of prominent researchers who've really argued for this approach, right? um they don't necessarily see a lot of conflicting opinions. Um and it's pretty strongly on the waiting until 70 side. Now there are some maybe areas of smaller disagreement.
(04:48)  So um you know even you know researchers when I'm talking about especially some of these prominent researchers they're not identical in their views. They have different you know nuances to what they're thinking. Here's just two examples right? So Michael Finina he's taken a pretty strong stance on you know what that right discount rate should be.
(05:06)  So he this is a quote from an article that says using a discount rate that incorporates risk premium is not appropriate for guaranteed future social security cash flows. Um look at somebody like Wade Fowl who I'd say his research has largely been pretty consistent with with Michaels. I don't think they have strong disagreements in this area but he has acknowledged that you know maybe discount rates ought to vary between people.
(05:29)  So he has a blog post acknowledging somebody with a 6% discount rate might come to different conclusions than someone with a 2% discount rate. Um, and so it really depends on kind of this personal discount rate rather than there's one discount rate that everybody should use. Now, maybe some larger disagreements. Um, probably the most prominent one I can find here, uh, looking at, uh, Michael Kites has some blog posts where he's looked at, you know, what sort of discount rate should you use when you're doing different types of retirement, uh, calculations. And there's a full article here. You could go read the whole article, but I actually just pulled out
(05:59) a little snippet here from the u the comment section where in the article Michael had made the case that you really you should use the expected rate of return on the portfolio that you're spending down. That would be the best place to start or what to use as a discount rate. Um Bill Meyer responded here said, you know, social security is like a tip.
(06:23)  So 8% discount rate, which was what would have been used in the article here, is too high, right? Okay. And Michael responds back, if the portfolio is earning 8% and that's being spent instead of social security, then the discount rate really should be 8%. That's the opportunity cost that somebody's paying in this case. Right? So this I would say is maybe the the strongest area of disagreement um that at least until recently um I had seen.
(06:47) Now these are really in some ways reflecting differing views, right? So there is um and there's actually a good social security article u from an article from the social security administration. I'll I'll reference to it later, but uh really there's kind of two different perspectives. One is more of an asset pricing or market valuation view.
(07:04)  Um and they would say here, you know, social security is a tip-like asset and should be discounted at tip-like rates. The other is more of an opportunity cost perspective, saying that well social security should really be discounted based on actual real world opportunity costs, right? What am I actually giving up in order to delay claiming my social security? And that's what we ought to use. Now, who's right? Uh I'm actually gonna we'll come back to this.
(07:26)  We'll we'll table this for a second because I actually want to take a step back and say, are we even right really asking the right questions? Because I think sometimes here, you know, there's a lot beneath this that's actually a much more kind of sophisticated question than it might get at first glance.
(07:45)  And and I'd say I'm even somebody who personally kind of that at first glance, you know, I just took the pretty standard perspective of yeah, delaying till 70 using these really low tips like discount rates that makes sense because of the risk return relationship. Um, and it really took some reflecting on like actually let's think pretty carefully about what we're doing here to bring me to at least thinking that there's a lot more nuance to this discussion than might meet the eye.
(08:09)  So let's take first a closer look at some of that research that really strongly supports claiming at 70. There are some common methodological consistencies between these especially in what you read in kind of the practitioner world you know journal financial planning those types of outlets that might be more commonly read by um adviserss rather than some of the more esoteric kind of economic journals that may not be quite on our our radar as much. So some of these common uh consistencies, one would be the use of an expected value framework. Um so here
(08:38) are just basically like a net present value. What is you know the probability somebody's going to receive some sort of future benefit? What is that future benefit? Um you know that's kind of what would fall under this broader expected value framework. Uh ultimately the use of 0% or very low discount rates.
(08:54)  Again these could be tips yields. So maybe it's a 1% reel or a 2% reel discount rate. Some studies even use negative tips yields. So some studies have actually even used negative discount rates saying that know a dollar that you receive today is worth less than a dollar you receive in the future.
(09:14)  Right? It's kind of a weird concept but if you're using tips yields and tips yields are currently negative that's where you'd end up. Uh also here addressing mortality risk through mortality adjusted benefits and and we'll get into a little bit more nuance of why I think there's a little bit more discussion that could be had around what should we be thinking about in terms of mortality risk.
(09:32)  Uh but these would be what I'd say are very baseline commonalities. So you see these across a number of different researchers, pretty consistent methods and pretty consistent results that you know claiming until or delaying until 70 U makes the most sense. But you know going back to this expected value consideration, it there is a whole other framework that we see very commonly in economics around expected utility theory.
(09:56)  And a common way to illustrate this would be say we give people a choice. they can choose between um $20 or a 25% chance to win $100. Now, under expected value theory, the same framework a lot of the more prominent research uses, A is irrational. You can't choose A because the expected value of B is greater. So, it's irrational to say, you know what, yeah, give me that guaranteed $20. I don't want to take the risk.
(10:20)  I just want something, you know, something in my pocket right now. But under expected utility theory, we don't know which one of these is rational, right? We do are able to account for things um like how much utility somebody would get and they might get more utility from those first dollars they receive.
(10:39)  And so actually the gamble in this case may not be worth it um or maybe it is right it depends on somebody's own way that they basically experience utility or satisfaction from the options presented. So under expected utility theory either one may be rational and I think this is really the better model for actually looking at human behavior um we do see again expected value theory dominates most of that well-known research but there are really issues with that right we know that expected value theory if we're applying that in other personal financial decision-m context right if
(11:09) we're looking at something like insurance well insurance would actually be irrational right because in this case if we're just calculating the what you're paying probability you have a loss. We're doing all the math. Well, premiums have to cover expected claims and also administrative expenses, profit margins, other things to get back that are in that.
(11:29)  Now, certainly we wouldn't say that insurance is irrational. In many cases, it makes sense to transfer some of that risk and avoid a catastrophic outcome by engaging in uh you know, paying a little bit of premium above and beyond the expected value of those costs. So again, expected utility theory is much better for actually answering questions like how do think of people think about u insurance purchases because it actually gets at that dynamic a little bit better. Um also when we look at things like loss aversion right there's all sorts of ways that you know
(12:01) we we might guard against um you know losses. Uh our you know if you think about even kind of a evolutionary psychology kind of perspective our ancestors right who were most attentive to risks. or somebody hears a rustling in the grass, right? If they were attentive to that risk, they were more likely to survive.
(12:19)  So, we do accept some false positives that we might, you know, might react to a risk that wasn't actually a risk. Maybe it was just the wind, but the chance of that wasn't a lion right in the grass might actually pay off. So, in many cases, we want to be a little bit conservative and we can express that through expected utility theory in a way we can't through expected value theory.
(12:44)  Uh Nim TB has also argued that expected value theory will fail if potential outcomes involve ruin right so if we have you know particularly catastrophic outcomes that are part of our uh you know distribution of outcomes we're looking at that could be a problem and I like this quote um here that you know you never cross a river that is on average 4T deep right and that's what we're getting at where if there is some sort of catastrophic sort of loss you know that is something we want to pay attention to but getting back to social security more in particular I think asking this question of you know what is ruin in a social
(13:14) security claiming context right because this is actually kind of a a trickier question than we might think now the most common one we'll hear about is that longevity protection the social security provides and obviously a very real benefit of social security so living a very long life right that would be one risk that's out there now is that a risk that really amounts to ruin in many cases I I I don't know I think it's a It's an open question.
(13:41)  Um, we do see especially when you look at the way a lot of people spend, Michael Kitsus has some good articles around if you actually use the 4% rule like how much do you have when you get to the end of a retirement period and for a lot of people especially when we use these heristics that might kind of plan around worst case scenarios.
(14:00)  Uh, we end up in situations where there really was no risk of running out of money in retirement, right? And if you're using a riskbased guardrails framework, if you're doing things like that, you really should be accounting for that risk in many ways that should be mitigating that um in a way that maybe your social security benefit whether it's a little bit higher or lower isn't actually having a huge impact and this ruin in terms of if you lived a long time.
(14:24)  But there are other risks here, right? So thinking about working longer than one had to and maybe missing out on prime retirement years, right? Okay. So, what if somebody, you know, does, you know, they they maybe don't enjoy their job, but they've stayed in it because they wanted to try and delay their claiming social security a little bit longer and they get a terminal illness, right? Those could be things that I think people might look back on with with a lot of regret.
(14:47)  And again, is it ruin? I I don't know if it's ruin, but certainly it is a very potentially negative outcome. Um, same thing with just getting a terminal illness. Maybe somebody was delaying claiming and maybe they left hundreds of thousands of dollars of potential benefits on the table. Now, the common retort to this is that, well, they're dying, so they don't actually need the money, which is fair in terms of actually preserving their own well-being in retirement, but there is still something here to be said for that's, you know, hundreds of thousands of dollars that could have went to a family member, could have went to a charity, could have went to different places that somebody cared about. Um,
(15:17) and ultimately what this is getting at, right, is this is a really hard question to answer because we're actually dealing with a two-sided risk problem, right? there is no inherently risk averse decision. You can't say that delaying claiming is necessarily more risk averse because it might address some risks but it's actually reducing some risk but increasing other risks.
(15:40)  Right? So no matter which direction we move here, we're kind of just prioritizing or deprioritizing certain risks that we might be thinking about. So why don't we use expected utility theory? Well, there actually are some studies that that have used expected utility theory in a pretty robust way around claiming social security.
(16:00)  Um, again, those tend to be in more very economics heavy type journals. Maybe not the ones advisers read as much, but there there is some of that research out there. Um, I think a few reasons why it may not be as popular, especially in kind of the adviser um research is there are some real limitations of trying to do this as well.
(16:19)  First, you know, all utility is going to be subjective, right? So we know that people have different preferences. Nobody has the exact same identical preferences as somebody else. So all of this is going to be kind of abstracting into this really kind of complex aggregation of of a person that may or may not actually reflect reality.
(16:40)  Uh there is no single like true utility curve, right? People are actually quite complex and nuanced. Um and a lot of times the different risks and things that we face are very path dependent, right? Somebody might have a lot of regret regret with a certain outcome um if they made some earlier decisions that they may not have with that same outcome if they made other choices.
(16:58)  So it really gets to be quite complex in terms of figuring out okay well how happy or unhappy am I with an outcome. There's a lot of factors that are going to play into that. Uh and I think maybe a strength of expected value theory is that it just treats a dollar as a dollar.
(17:17)  Right? In a way, it says there's no need to try and untangle this complicated web of what money really means to people. We're just going to look at this very objectively. A dollar's a dollar. We don't have to worry about that. But I do think there's also maybe a bit of kind of the streetlight effect going on here where this is this our tendency to look for answers where they're easiest to find rather than where the truth is most likely to be.
(17:34)  Right? And so we have this common example here. um you know somebody's out looking for their keys in the street and you know they're looking under a street light and the police officer says did you lose them here and they say no the light is better here right so this is just that tendency where we might be looking where it's easiest to look instead of really getting to the heart of the matter which I think has happened in some respect to a lot of the considerations that we look at in social security so I want to go through kind of a list here of what I would consider to be some often ignored risks of delayed claiming Um, now I do
(18:06) want to emphasize too, obviously there are benefits of delayed claiming that aren't you know on this list. Things like um survivor benefits, spousal benefits. Um, there are some tax advantages to delaying claiming around Roth conversion. It's also just a little bit more efficient in the sense that only 85% of your income as a maximum is subject to taxes.
(18:25)  So there are definitely advantages, but here I'm really just trying to emphasize some of these risks that are also there that generally may maybe not fully accounted for or maybe not accounted for in ways that I think gives us a true sense of of what's going on. So the first one here being mortality risk.
(18:43)  Now mortality risk, you know, this is just that risk that somebody dies before reaching the point where they've come out ahead from delaying. Now, most of us understand this, but I think just looking at a little case study here can be helpful to see what that could actually amount to in terms of dollars. So, let's say we have James here. He's 61 years old, single.
(19:02)  He chooses to claim his $2,710 monthly benefit early next year at 62. Um, he expects inflation to average 3%. He expects to earn 8% on his portfolio. This would amount to a 4.85% real return um on his portfolio if his pension covers his spending needs. um you know and he could be you know somebody who lives really modestly relative to other income they have. So that's kind of the assumption we're making here.
(19:26)  He's in a position where he might be claiming his benefit but actually investing every check in a taxable brokerage account, right? He doesn't actually have a need for it, but he still claimed it or decided to claim his early. Now if he passes away at age 70, he would actually have based on these assumptions about $398,000 in this taxp brokerage account.
(19:48)  Now, this is set up to be kind of the uh the extreme example here, right? In terms of basically giving him maximum benefit, seeing what that could grow at, assuming a pretty, you know, I wouldn't say aggressive, right? We could make this even more extreme and assume he had an all stock portfolio or something like that.
(20:06)  But we're assuming maybe a historical return consistent with like a 60/40 portfolio, but this is almost $400,000. That could be money that's left behind if he were to pass wet, right? If he would have instead said, "I'm going to delay to 70." And he died right upon reaching 70, that is a real risk, right? That's $400,000 um of mortality risk that really doesn't get accounted for in the way that mortality adjustments happen in most most studies.
(20:29)  So, you know, I would say, well, should we really just ignore this? Um I I don't know that we should, right? I think it's fair for some people to say, you know what, I don't like that trade-off, right? I might value the the chance of being able to have some more funds that I could give to somebody that I could use for myself, whatever it might be.
(20:48)  Um, and I think that's a fair perspective for somebody to have for themselves. Um, and you know, I also think there's this question around kind of regret. What would somebody regret more? And again, no one uniform answer to this, but I think for some people, let's say they get a terminal diagnosis at 69 and they delayed claiming social security, right? I think there will be some like I made the wrong decision.
(21:12)  Now obviously that person's facing much bigger issues than just their social security benefit in terms of you probably wrestling with their mortality and things like that. But it's a situation where you know I do think there realistically might be some regret that I kicking yourself I should have claimed earlier. Now let's say somebody lives a very long life.
(21:28)  They lived in 95 but they claimed at 62, right? Is that person going to be kicking themselves because they didn't delay till 70? Maybe that it could happen. Um, but I I just think that's hard to see in a lot of cases and particularly cases where people were navigating retirement in a way that they were mitigating that risk.
(21:46) They ever ran out of money, right? That obviously the scenario where somebody would care very much at 95 would be if they spent down everything else, they have nothing, they're living solely off of their social security benefit. Then that would be a situation obviously where where you would care about that.
(21:59) But I think it's fair to just kind of pose that question and think about what would people regret more because you know there there is that regret component or this psychological comfort that we have when it comes to decisions we make about our money. But the key point here you know just like we don't ignore longevity risk right that's a real benefit of delaying claiming social security is you have this inflation protected guaranteed income source that is now larger if you delay your benefit.
(22:24) We shouldn't ignore mortality risk. And now a fair objection might be well Derek doesn't you know existing research account for mortality risk and yes this is true right so oftent times future benefits are going to be mortality adjusted but the way that's um you know really applied in practice is looking at okay there's a very low probability somebody's going to pass away in the earlier years of retirement we're going to reduce those benefits a little bit more we'll reduce the benefits much more when they're later out and there's nothing wrong with that from a quantitative perspective that's
(22:54) perfectly reasonable thing to do. But I think it does kind of understate that reality of, you know, going back to the the example of case study where somebody might pass away at 70 and there's that $400,000 that was left on the table because nobody actually receives mortality adjusted benefits in the real world.
(23:12)  You either get your benefit or you don't. And so I don't know that a, you know, terminally ill 72-year-old is likely to take a whole lot of comfort in knowing that, well, you know, they delayed their uh delaying still did increase their mortality adjusted benefits when in reality that isn't what they ultimately got.
(23:30)  So might have been a good bet, might have been a smart thing to do. Things just didn't play out for them. But I think there's still real um, you know, risk. it's often not accounted for uh because you know maybe something like a s uh stress testing or looking at different simulations might actually better address that particular risk.
(23:50)  Now another risk that's often kind of commonly overlooked I think is sequence of returns risk. So here this comes from um Wade Vow's retirement planning guide book uh very almost famous you know chart when it talks about sequence of returns risk and what we see and you see here sequence of returns risk really peaks at time zero here on the the chart which would be years after retirement.
(24:13)  So right at retirement we see that sequence of returns risk is the highest and then it gradually declines really that first 5 to 10 years after retirement we see those highest levels of sequence of returns risk. Now, one thing that maybe not be maybe may not be quite as well appreciated is the way that this actually changes when we're talking about um sequence of returns risk specifically or or delaying claiming social security specifically.
(24:37)  And that's in large part because what Justin Fpatrick and myself we've referred to as the retirement distribution hatchet, the way that we get this frontloading of portfolio withdrawals often when somebody is delaying claiming social security, right? So if they're using a riskbased guard rails approach, they're trying to smooth out their spending in general.
(24:56)  That's going to be a situation where delaying claiming is actually going to put pretty high strain on the portfolio in those earlier years. That's also going to do that in a way that again exacerbates that sequence of returns risk. So we go back to this chart, right? And this is actually understating the sequence of returns risk because usually here it's just using something like a kind of constant inflation adjusted spending assumption or something along those lines to get out what we see here. Whereas this is very different.
(25:19)  Now we're talking about not only do we have sequence of returns risk that we had in the baseline, but we've really exacerbated that because we're um inflating how hard we're pulling from the portfolio in those early years. And especially that you know first seven years of retirement or so for many people that hatchet that blade of the hatchet they're might be taking distribution rates at four times five times or more than that what they'd be taking later in retirement.
(25:43)  You know, we can even think of the kind of logical extreme here where that blade period of the hatchet is just somebody basically delaying claiming as long as they can, right? They're spending down their portfolio 100%. Until they get to the point where they have to turn on their social security benefit, right? So, that might actually be a situation where now we're talking about somebody spending down 100% of their portfolio over an 8-year or less timeline.
(26:06)  And that's going to have a ton of sequence of returns risk to it in a way that we we don't see as much um when we're thinking about just a 30-year typical retirement period. So, let's look at one example, you know, of what we could see here. So, we have Mike who's 62, single, um, and considering claiming social security.
(26:25) Let's say Mike has a $500,000 IRA allocated 6040. He's eligible for a $2,200 monthly benefit beginning at age 62. um he is comfortable using a risk-based guards approach like we're fans of at income lab and that approach will help keep his spending on track throughout retirement.
(26:44)  So a question we might ask is well what happens if we stress test him going through a period like retiring just before the global financial crisis right what would we actually see in that situation now interestingly first let's look at spending so here you can see um the gold line is showing the claiming at 62 his income starts out a little bit lower um so he does take a little bit of a reduction in how much he can initially spend uh because of that u but it's not huge you can see these are pretty close in terms of total spending levels. Uh, interestingly, as it gets later in life, that actually ends up flipping um
(27:17) because the um what happens with the portfolio, which we'll take a look at in a second. But the biggest result here, I would say, is that these spending levels aren't that different, right? This is actually pretty similar spending whether um he's claiming at 62 or 70.
(27:36)  It doesn't actually have a huge difference in how much he could spend from something like a riskbased guardrails perspective. But now let's look at what happens with the portfolio. So in this case you can see when he again the gold line is showing the claiming at 62. So here we're showing you know in both cases saw a pretty significant draw down of the portfolio there. But since the social security benefit helped reduce what had to be pulled out of the portfolio in these years and allowed more of the portfolio to recover and come back. You can see that it's a pretty big difference here. So go all the way out
(28:06) to 2025 here and now he would have $600,000 in his portfolio if he claimed um at 62 whereas he only has $200,000 in his portfolio from claiming at 70. Of course he has a larger social security benefit right there. There's trade-offs to this but um I think for a lot of people this is a meaningful consideration um in particular because there's a lot of optionality built into having a portfolio where you can take a lump sum.
(28:34)  There's different things you can do with a portfolio that you can't do with just a social security benefit. And so some people might prefer to preserve that. Other people may not care and that's perfectly fine as well. U but we see the gap here is is quite sizable. So um again just summarizing had about that $400,000 difference um in what was in the portfolio based on the the claiming wage.
(29:00)  Now, an important thing to note here is that we actually as planners do have some pretty good tools to help mitigate this risk, right? So, one of the nice things about social security is you maintain that option to turn on your benefit at any time, right? So, retirees might take the approach of, well, actually, I'm I'm going to start out delaying, but if we see markets tumble, we see markets decline, then somebody might want to change their mind, right? So, we they could turn on their benefit to reduce the strain on the portfolio if they hit tough markets.
(29:24) Um, of course, we can also retroactively claim up to six months of benefits, right? So, we can even sort of reverse some of our delays. So, this is a nice place where I think, you know, we don't have to set a strategy and, you know, write it in stone. It can be something that really we can be dynamic and we can adjust.
(29:41)  And in some cases, if we go through a big downturn, this might be a tool we really want to consider that, okay, maybe somebody really did think they were going to delay claiming, but once we see what happens in the market, maybe that would be a situation where turning that benefit on might actually make sense.
(29:58)  Now, I want to shift gears a little bit to the policy risk, um, which is a risk that retirees care a lot about. So a 2025 AARP report found that only 36% of Americans felt that they were confident in Social Security's future. So a lot of pessimism around what might be there to actually pay out terms of benefits.
(30:21)  Um a lot of that fear does come from the depletion of the Social Security trust fund. Now an important nuance of this is within the same study they were able to um basically identify that a lot of the people they were surveying didn't quite understand actually what would happen right if if the social security trust fund would deplete and they didn't realize that's actually actually something that's planned for and intended.
(30:43)  So there is some maybe ignorance here that's you know fueling some of this and that's certainly an area where better educating people can be very useful. Um, but regardless of, you know, how accurate some of the beliefs are, the reality is their policy risk is a real risk that's on clients minds. Um, and I think advisers are generally less worried about the policy risk.
(31:02)  I think advisers are generally better just more knowledgeable of some of the adjustments that could be made to the Social Security Administration uh or to Social Security benefits to kind of shore that up. Um, things like, you know, raising how much uh income gets taxed for Social Security purposes.
(31:20)  um making adjustments to future retirees benefits, things like that that may not impact retirees today. But I think it is something that you know there are maybe some kind of lurking risks that even if we may not realistically see the big drop um right when the trust fund depletes, although that is still a risk. I mean Congress could decide to do nothing which kind of on brand for Congress in some ways to uh who who knows what would happen.
(31:45)  But um I think you know there are some other less overt risks that still could be quite large and especially for types of clients that advisers work with. And so some of those could be tax changes that could reduce the net benefits. So you can imagine a future world where yeah the the benefits paid out exactly the same but maybe there's a change in how social security benefits are taxed that could have some sort of negative implication.
(32:09)  Again, I think that's probably even more realistic for wealthier, more affluent individuals who have higher income levels. Um, there's also proposals like we saw under the Obama administration to move to like a chained CPI. Now, economists love chain CPI. It actually makes a lot of economic sense when you think about substitution effect and kind of the way that CPI might overstate inflation.
(32:29)  uh because the common example, right, we have the substitution effect being that um you know, if the price of chicken doubles, but the price of beef stays the same, people don't buy the exact same basket of chicken and beef, right? They're going to substitute some. They're going to buy relatively less chicken, relatively more beef.
(32:46)  And so, in that way, if we hold the basket of goods that makes up the CPI fixed, we are maybe overstating, you know, what um inflation is actually there. So, it's actually a very economically sound proposal the Obama administration put forward. It was not popular and it was quickly um kind of removed from consideration, but it's the type of thing that I think we could see come back because there is again actually pretty sound basis for it.
(33:11)  Um and that could be something that you know may not have that immediate effect but cumulatively over the course of a um full retirement period could be meaningful. So ultimately again I think we should consider this um you know how people weigh these risks is going to vary based on their own preferences and their own views but um definitely is something that shouldn't necessarily be totally wiped away and disregarded.
(33:33) Now getting to opportunity cost, you know, here we're really getting back to kind of the core of some of that disagreement we started out with, you know, Michael Kits versus Bill Meyer perspective and some of that there. And I want to go back to this view that, you know, there's really two different ways to look at this.
(33:52)  There's an asset pricing or market valuation view that's saying that social security is a tip-like asset, so it should be discounted at tip- like rates. And then there's this opportunity cost view which says that social security should be discounted based on whatever actual real world opportunity cost somebody face um is facing.
(34:11)  And so a 2025 CFA institute study found that real returns on a 60/40 portfolio were about 4.9% from 1901 to 2022. And this is based on US returns. Um, so if we're just trying to look, you know, just historically kind of what real returns have we seen from a portfolio, that could be something closer to 4.9%. Now, some people will want to use different forward-looking returns.
(34:35)  because there's a lot of ways you can change the assumptions here, but you know again just looking at it purely from a historical basis um in some ways uh you know we can differ on exactly how we would define the expected return in the 60/40 portfolio but that's going to be our sort of best guess of what somebody would actually experience. Now both sides have valid points right on one hand tips returns from a risk matching perspective you know that they do make a lot of sense uh that is more kind of the true uh nature but we also have to consider where are most retirees actually taking distributions right and the kind of the best way theoretically
(35:08) to defer claiming social security would be um especially if you're going to take this tips perspective would be to use something like a tips ladder um where you set up a ladder you fully um fund the deferral of Social Security benefits from that ladder. In that case, I think TIPS returns are actually a very good um opportunity cost to use.
(35:27)  It would make sense to use that because that's the asset that's being displaced. Uh but for many retirees, it's actually impossible to fully fund social security delay just from their bonds or their tips within a 60/40 portfolio. Right? So, if somebody doesn't have enough in assets to do that, it simply isn't going to be um a reality for them.
(35:50)  So going back to these perspectives though we have to think you know what question are we actually asking here and with that asset pricing market valuation framework we're saying what's the market valuation on my social security benefits basically if some hypothetical Wall Street trader you know could buy your benefits from you what would they be willing to pay and the opportunity cost question is what do I need to give up to get a higher social security benefit and in financial planning I think we actually generally care a lot more about this latter question right it really is more that opportunity cost perspective And that's that's relevant in a way that I think um might be underappreciated for
(36:20) people in the camp of saying tips yields are the only thing we can ever use. Also want to draw attention to again this is I've mentioned it previously but this is a um a paper that comes from the social security administration bulletin.
(36:37)  Uh, in this quote in particular, it says, "The claiming decision poses 97 monthly choices between ages 62 and 70, each with different guaranteed fixed future cash flow stream um that can be appropriately valued by the participant for whatever reason and according to how much more he or she values sooner payments rather than later ones, right? And this one is actually this article is getting directly at this issue of this mislication of an asset pricing framework to an opportunity cost question and really saying that you know actually it is okay for people to use discount rates that aren't tips. Um that's very justified and it comes down
(37:09) to individual preferences and how much an individual values these earlier payments versus later ones as well as the other risks that we've really talked about here. So you know common objection right and this is where a lot of the conversation recently I said has really been fixed is kind of well the risk is different and yes that is true right we're not doing an apples to oranges comparison or we are doing an apples to oranges comparison if we're comparing um you know expect or risky portfolio returns to delaying claiming social security right they have a different risk profile but even in this case right
(37:44) even if we do want to account for that different risk an important technical note is that we wouldn't jump straight to tips, right? In economics, we talk about what you call certainty equivalent returns, which is basically asking the question, what guaranteed rate of return would somebody need uh to make them indifferent between a risky and a risk-free investment? And this is really important because these numbers vary by individuals, right? So, David might accept 2% while Michael might require 4.8%. Right? There is no right answer.
(38:13) We are all different in how much we value certainty um uh or not. And so this is an important nuance that again even if we want to risk adjust we don't necessarily jump straight to one. It's still a personal consideration based on how much somebody personally values the certainty that comes with it.
(38:31)  I won't go into this in too much detail but did include it here um in case somebody's watching the recording wants to pause it. I think I put together this table that really kind of shows okay on the the lower end of the side here we have our at uh market value framework right where tips yields are the one right answer.
(38:48)  Uh but on the upper side here uh we kind of have this bound for what you know investment risk could be anywhere from really tips yields up to let's say historical 60/40 portfolio returns. Um but even beyond that we might start here and then layer in additional risks above and beyond that that could push the risk higher or lower in those levels.
(39:06)  So again won't go too go too deep there but that's another consideration. And ultimately my take I do think when we're talking about opportunity cost the expected return on whatever displaced asset um is being spent down is a very reasonable first approximation right there's a few reasons technical reasons why I say this um you know is it you we're glossing over some of the risk adjustment but I think once you also weigh some of the other risk factors that are commonly not factored into this uh these types of analyses kind of moves the needle back in that direction and for a lot of people it's a very
(39:40) practical way place to start. It also has a nice benefit of really giving us kind of our best guess of what the future might be, right? Especially if we're doing some sort of modeling um we're using software long-term projections, it is going to be more of our best guess of what would actually happen.
(39:57)  Um moving on to regret risk. Um some things I might go a little bit quick too just because I want to be mindful of of time here. Um this one in particular, uh so Justin Fitzpatrick's written about this Peter Sandman's framework.
(40:15)  um you know risk how it's kind of hazard and outrage but here hazard in this framework is some objective quantifiable danger maybe like the social security benefits could be cut by 21%. And outrage is actually this psychological factor that can make certain risks feel unbearable. Right? So I think for some people they really feel this sense of like social security specifically I paid in all these years and I got nothing in return or I got less than I as if they were to pass away something like that where they might get nothing in return and that can feel almost unbearable to some people. Other people may not care at all. Right? This is again a situation where we can't just
(40:46) treat people all the same. It depends on what their own values are. Um, but just because it's mathematically equivalent loss between two people does not mean that they're going to have the same sort of impact or how that feels to them. Um, health span risk.
(41:06)  Uh, this is a one that I really have been giving a lot more thought to uh lately especially after um when it was actually a few years ago now, but when I read dive to zero from Bill Perkins, that one's really stuck with me. And I think just the way that when we talk about health span, right, and the way that um, you know, we we may be able to spend a dollar, you know, at 65 or 95, but we're going to have a different health context at 65.
(41:30)  There's more things we're going to be able to do physically, mentally, we may be more capable of doing things with that, right? And if we're using some of these tools or some of the research that uses a 0% discount rate, that assumes $10,000 at 62 is identical to an extra $10,000 at 95. But I think, you know, people are not only more capable of using that extra 10,000 at 62, but when we start to think of what Bill Perkins refers to as the memory dividends, kind of the lifetime benefit you receive from those experiences, that's also greater at a younger age as well. Um, so I think really thinking about our health and how we can use resources, you know, that's
(42:02) another factor generally overlooked in a lot of research. Um, spending optionality or flexibility. So one thing that a portfolio provides that social security benefits don't is the ability to use sort of a lump sum or an expense right so it's you can think of these as real options right kind of like options we have in financial markets but real options and life being that flexibility to act whenever an opportunity arise um and preserving that portfolio value really does help preserve that opportunity to say yes if some sort of
(42:33) once in a-lifetime opportunity comes up right so let's say I don't know your your daughter's favorite uh musician is going to be playing in Paris, right? You can't ask social security to send you five months of benefits in advance to go see that. Whereas you could take a lump sum distribution from a portfolio. Now, of course, there's borrowing. There's other things we could do.
(42:52)  Um but you know, a lot of people, especially when we talk about bigger expenses, may not want to be borrowing in retirement. And again, that's just an optionality that you have with a portfolio that you don't have with social security benefits. Um and particularly when we think about the extreme case.
(43:09)  So, if we think about somebody who's fully spending down their portfolio um just to get to the point where they've maximized their social security benefit um that that actually can happen for households that have lower port initial portfolio balances. If they really want to maximize their benefit, that's what they're doing. They're spending their portfolio down to zero. Then they're turning their social security benefit on.
(43:28)  And that's going to take away a lot of optionality. It is going to provide a nice, you know, a larger kind of guaranteed lifetime income for them. But when we run the numbers and do some of the stress testing, I don't know if it's the best way for for everybody. Um, obviously overall wealth should play a role here, too.
(43:46)  So, if we're talking about oftentimes this is most contentious for, you know, um, Americans in maybe $2 million portfolios and below where there is some of these trade-offs. If we're talking about deca millionaires, right, they can maintain flexibility regardless of whether they claim social security or not, right? It's not a consequential decision for them. So, that could be a very different sort of situation.
(44:04) Um and then underspending risk is one that actually I think again deserves a lot more attention than it gets. And we see that people think about social security income different than portfolio distributions. So social security income is actually one of the easiest retirement income sources to spend. Uh Blanchett and had a really good study on this where they looked at the the HRS.
(44:24) They found that um uh retirees spent about 80% of their guaranteed income but only 50% of what could sustainably be spent down from a portfolio. So people much more freely spent guaranteed income compared to uh portfolio income and I think again putting myself in my own advisor context working with clients especially the clients I run into tend to be people who are good savers. They're diligent.
(44:54)  I'm spending a lot more time oftentimes talking about underspending risk and the fact that, you know, I'm encouraging people to take more income because they could spend more, they could do more things if they wanted to. Of course, people don't have to do that, but that's where I spend a lot of my conversations more so than meeting with people who, you know, are at any real genuine risk of depleting their portfolios in retirement.
(45:13)  So, I think it's worth thinking about, okay, well, if claiming earlier could have a meaningful impact on how much somebody's willing to spend or even when they would retire, like some people like to couple that decision and say, you know, I'm not going to or um as soon as I retire, I am going to claim my social security benefit.
(45:32)  So, they might work longer just because they psychologically have linked those two together. Um, again, people like Blanch and Fina, that same article that I I mentioned, they've argued that purchasing annuity in retirement can increase spending confidence. I think, you know, that that's true. I think there are some frictions there when it comes to actually getting people to purchase an annuity.
(45:51)  That lump sum, taking a lump sum and converting it to a dollar value is a hard pill for a lot of people to swallow. I do think that same logic ought to apply though to social security, right? that claiming social security earlier can give clients confidence to retire earlier, to spend more.
(46:09)  And if that is true for somebody psychological, it's at least a tool that we should have in our toolbox to think about um and be be open to. Um and this is just another quote here from this particular study. Um it says, "Retirees who are behaviorally resistant to spending down savings may better achieve their lifestyle goals by increasing their share of wealth allocated to annuitized income.
(46:27)  This could take the forms of delaying claiming social security, retirement benefits, choosing a job, employer or pension, or purchasing an income annuity. Um, but the reason I say I've kind of emphasized some of these things here is that actually delaying claiming social security, right? If you're going to do that and you're doing that, you know, funding through your portfolio instead of continued work, you have to spend down your portfolio, right? So, if you're already behaviorally resistant to spending down your savings, well, that's what you have to do to claim social
(46:51) security benefits. um that's what you have to do to purchase an income annuity. You have to spend down your portfolio and convert that into a stream. What's kind of unique about social security benefits is we don't have a lump sum option, right? You can't take a lumpsum social security benefit.
(47:08)  And so it's actually kind of avoids some of those behavioral dynamics that I think sometimes people might be hesitant um to doing that. Right? If you gave people lumpsum options, there probably would be a lot of people that would take that lump sum in social security and then underspend that. whereas you don't have that option.
(47:24)  And so if it's going to be something somebody more freely spends, that is something that's worth consideration. So, okay, how do we tie this all together? Um, and I'm going to keep this pretty high level, you know, because I think there's a few different things. So, I think when we think about this, we should start with maybe this kind of baseline opportunity cost, right? What assets being displaced? I think there's a pretty good basis for that that we can start thinking there, especially if we're trying to choose a discount rate or some sort of analysis. Um, that could be different, right? So, if somebody's
(47:53) going to spend down a TIPS ladder, then certainly use TIPS yields. Um, if somebody's going to spend down a 60/40 portfolio, maybe that's what we want to use instead. But then from there, we can add or subtract to this baseline based on how other factors may or may not relate to a client's situation.
(48:12)  Um these case studies are in my kitus article um uh on the on this decision. I won't go over them in depth. More I'll just very high level say case study one is designed to be a situation who where somebody has a um a lot of factors that would raise that risk factors that would raise maybe their their discount rate.
(48:33)  They'd be wanting to account for these more. So they do fear benefit cuts. Um they're worried about dying early. they um are going to spend from a 6040 portfolio. They have um um you know these different factors that would all kind of align in that direction. So again, I won't go into all the details here, but um slides are here if you want to see them, but we might, and these numbers are not meant to be rigid, right? This is just meant to be like a conceptual illustration where we could go through and we could say, okay, so we started with the expected rate of
(49:02) return on the portfolio, they'd spend down. Um let's maybe call that 4% just to be conservative. Then maybe we add to it for mortality risks. That's relevant to that. We add to sequence of returns risk. We add regret risk, health span risk. Um you know, longevity risk. you could take away from that because there would be longevity protection provided to them.
(49:20)  We might kind of go through an exercise like that and net out a more kind of client specific discount rate. In this case, uh that came out to be about 8 and a half%. Um now this here um see if I can zoom in a little bit and see the uh screenshot from income lab a little bit better. But you know this is a place where you can go into the opportunity cost. You could put in the 8 and a half%.
(49:44)  Um you could also put in a customized reduction. So here we use a 10% benefit reduction, not necessarily the full projected assuming Congress does nothing, but maybe in 2033 we see a 10% cut and we can get a customized kind of heat map for when it makes the sense for them to claim and with their particular discount rate um you know with their unique factors in their plan does look like both of them claiming early might be something really worth considering. Um another thing in income lab I like to really look at is this um look at the lower number and the higher
(50:13) number. I find especially, you know, this is meant to be more extreme in terms of u definitely is a case where maybe claiming early makes sense. We'll look at another one that's the reverse of that. Um but I find for actually more a lot of middle of the road plans that I'm running for clients.
(50:30)  When I look at the numbers here on each end of the spectrum, they're not that different. Right? It might be something where it's the best or the optimal claiming decision once we use a more reasonable discount rate is 20 30 $40,000 higher lower than the other. Right? And so that's a I think worth considering too because then it's like okay well if it doesn't make a big difference but maybe somebody has a strong preference for one or the other then maybe that should also be factored in.
(50:54)  Have this other example here Spock and Tara they meant to be like hyperrational um you know they do everything by the textbook you know they they follow all the kind of um funding from a tips ladder all those different things that you would ask of somebody plus they have a $10 million portfolio and they want to spend relatively low relative to that.
(51:12)  um they have very long longevity concerns. Spock's father Sar lived to 202 but Spock's okay just planning to 100 right when we look at a plan like this this is a case where a very low discount rate is going to make a lot of sense. Um so here right where maybe most of these risks mortality risk maybe still a little bit there but sequence of returns risk is not a factor regret risk is not a factor health span risk was not a factor right all these things that were not factors where maybe we would even want to use a negative discount rate in this particular case right where everything nets out and maybe a negative
(51:41) is a very reasonable outcome um and in this case the kind of client specific discount rate negative 3.5% um actually can't even do that in income lab I think the the lowest you put in is zero Um maybe that's something that could be could be adjusted. Um but this is just meant to be an illustration of you know in this case delaying till 70 makes absolute sense uh for Spock and Tara and it really depends on these client specific factors and what matters to them.
(52:09)  Now one other quick approach I won't go into this in any detail but um some research I'm actually working on is we could use this expected utility framework um like a very detailed academic model um and so this is describing all the different types of risks and things that in a certain model um I've been working on you know we use epscen recursive utilities stoastic investment returns medical expenditure chocks longevity risk policy risk the quest motives all these different things that are generally overlooked in a lot of the analyses
(52:37) And here this is actually showing the optimal claiming ages. So table three is showing that for households with up to 200,000 age 62 was the optimal um on a kind of a baseline model that's showing model one then it flipped to age 70.
(52:56)  But model 4 is kind of the most behaviorally realistic model that we put in where people value things like front-loaded consumption um and they have this source uh dependent utility of spending. So they get more satisfaction spending social security income than they get spending their portfolio income. And that actually flips the results to claiming to 62 until you get to about a million dollar portfolio where then it bumps out to an age 65.
(53:15)  Right? And we can even tie that to kind of discount rates that would be equivalent to that. And you can see again anywhere 5.9% to 3.1%. But I I won't go too far into details. I could do a whole presentation on that. Um but ultimately a conclusion there is no one discount rate. Um, for many Americans, that rate is not 0%.
(53:33)  And I think I'll just jump straight to the the questions here because I think we've got a few of those. So, uh, thank you for your time. All right, Derek, that was awesome. Um, so many so many parts of that that I I took some notes on for sure. Um, let's get into sequence of returns and the effect on the portfolio balance first.
(53:56) So, um, a couple people were asking about this, but delaying social security strains the portfolio early, which, you know, you showed, um, we did have some questions about how to show that in income lab. So, maybe I'll, um, show that quickly as you answer, but, um, but this person, uh, Brian, correctly says, but it provides greater relief later via higher benefits later on.
(54:20)  Shouldn't we, uh, account for both? And you know, I mean, fairly sure your example did, but um maybe just addressing that and if you let me share, I can just show and people because asking where to put um Yep. Yeah. I mean, I think definitely we do want to account for both, right? And I I would put that more in actually the you know, it's generally more the longevity risk, right? You're usually not getting hit by the sequence of returns risk could be a factor.
(54:45)  I mean, exactly how we're defining all the these different things could could vary. But, you know, when we're talking about somebody who lives to 100, 105, you know, a very long life expectancy, I'd put that more under the longevity risk category. Um, whereas that sequence of returns risk is more like that very catastrophic front-end um decline, but definitely it is something we'd want to factor in both. Yeah. I mean, just to see how to do it.
(55:08) So, I'm I'm using here the retirement stress test. So I'll show you in a second how to do it in the social security optimizer as well. I'm doing it here because I I created one where uh claiming a 62 versus claiming a 70 different you know I can even show the withdrawal dollar amounts right.
(55:26)  So claiming at 70 I have to withdraw more and then I withdraw less. So that's exactly what Brian was talking about correctly. Um interesting um the withdrawal percentages are actually not um hugely different there. Um but what's interesting is you know early on here blue is 62 gold is 70 and so the difference between the plans you know this is starting in 1999 2008 we got a $342 difference this is today's dollar so not taking inflation into account but down here you know we're down to a $243 difference uh $345,000 difference rather so if you live long enough they will
(56:03) eventually kind of come back toward each other which is exactly what you were talking about, Brian. So then the question is just like it you get into more the mortality risk, a life health span risk, right? Is the is having the larger difference earlier when you're more likely to be alive and healthy? Um I think that's the the u the issue there. And by the way, you can see the same kind of thing if you go to the social security optimizer.
(56:29) Um there is a uh a stress test option and you can just click on that and then you can choose two two claiming dates to compare and you'll see exactly the kind of thing I was just showing you. So for example here it's 70 and 67 and maybe I want to compare it to you know 62 and 62 and then I can I can see that.
(56:58)  So, um, let's go to some other questions. Um, the next question was, well, what's, you know, you showed the $400,000 loss, or maybe not loss, but sort of, you know, you're giving up $400,000. What's the loss if someone lives to 95? I think with that stress test, you can basically do that. And in the example we have, I mean, there may not even be you're not giving up anything potentially in terms of the, you know, if you're measuring it as um wealth you can pass on.
(57:27) Yep. Yeah. It's a I would take a look at that in the stress test. I think those longer term projections that's and play around with like a bad outcome. You can play around with retiring somebody during the global financial crisis more like a good outcome and you can kind of look at some different possibilities that are there. Yeah.
(57:45)  a couple good comments that I really just add to what you're saying of like, hey, it's more complicated, right? Somebody said, well, what if somebody has high interest credit card debt? Wouldn't, you know, if you took social security and used it to pay off that debt? Yeah. Okay. I mean, that definitely goes into it, right? That's higher.
(58:04)  Uh, and somebody's talking about using the money to buy life insurance, but you know, you could get really deep into some of these, which it sounds like you are. Maybe we should have a webinar where you go over your new research. Um, let's see here. There's button you can scroll through as well if there's any we could hit. Uh, um, I see the one isn't delay claiming purchasing a deferred annuity which might reduce underspending risk.
(58:29)  I mean, that's where you have to talk about when how are you funding that, right? Because if you're in the short term, you're going to have to spend down your portfolio in order to fund that unless you're still working, right? So that is the factor like if somebody's still working and I did see another question about still working as well um you know and that's certainly the earnings test acts as a discourage for people who are uh before full retirement age um you know if you want to continue working fully um and earn at a level that would wipe out your benefit. You wouldn't want
(58:54) to claim at 62 and then go do that. But at full retirement age then you do have a little bit more flexibility there. Yeah, that's right. Yeah. Somebody notes work your heart out after 67. um if you really want to. Um all right.
(59:17)  And actually Ryan asks, you know, when we're running different strategies and financial planning software, you know, like we were just showing an income, aren't we generally using the opportunity cost view, I think what he means there is the kind of the exchange, you know, the replacement thing. Yep. Yeah. I think that's actually a dynamic that's often pretty underappreciated that a lot of tools, I mean, even the income lab tool, you go in there, it's going to default you to a 0% kind of discount rate.
(59:36)  Um, you do that analysis, but then you step over to doing um, you know, if you just set up different scenarios and you're looking at different scenarios, there's kind of a difference there because you are implicitly using the portfolio's expected return as the discount rate when you're comparing long-term scenario comparisons versus when you're just doing a standalone calculation. and a lot of tools will default you to kind of a 0% type of comparison there.
(59:59)  Um, I think that's pretty underappreciated, but yeah, that's a that's a good point that was made there. Absolutely right. Yeah. In fact, I think the one difference that you pointed out and maybe we can close with this that I found really eye opening was okay, you know, if you look at kind of the economics research a lot of times those researchers are looking at like, hey, how would I value this thing if it were an asset? like you said if some hedge fund could do you know viatical settlements on social security pretty sure that doesn't exist but um you know what would they value it as
(1:00:29) which is just a very very different question than how does it affect you in your life and I think you rightly said actually the second one is the one clients want to look at the first one is kind of a not all that interesting like academic exercise um real life so I think that was a good good point um so with that make sure you uh fill out the survey at the end with your uh number and let us know um you know any any comments and and things. Sorry we didn't get to everybody's uh questions, but I'll end by thanking Derek very much
(1:01:01) for the great presentation and hope everyone has a great week. Thanks everyone. Thanks.