(audience applauding) - All right. Folks, welcome back. Most of you already know I am John Luskin and I'm gonna be moderating our withdrawal rate rumble today. We've got three very special guests. One was just interviewed about an hour ago, so he doesn't need much of an introduction, if he did anyway in the first place.

That's Mr. Bengen, creator of the so-called 4% rule. And then on the other side of the panel, we have Christine Benz, Director of Personal Finance and Retirement Planning at Morningstar. Her latest book is entitled How to Retire, 20 Lessons for a Happy, Successful and Wealthy Retirement. Christine was recently on the Bogleheads on Investing podcast with host Rick Ferry, talking about those 20 lessons.

So you folks can check that out on boglecenter.net. That's episode 73, the most recent episode. And then last but not least, our gentleman in the middle, Carsten Biggern-Jeske, retired in 2018 after a long career in academia, government and corporate America. On his blog, Early Retirement Now, he writes on all topics related to personal finance.

He's best known for his comprehensive series on safe withdrawal rates. Now folks, just like last time, I have some questions I prepared ahead of time, but I know you have your questions, so please write them down on piece of paper, hold them up, and then Mike Piper right over there, give us a wave, Mike.

He is gonna be collecting them. All right, let's get started with a quick review of all your research on retirement planning. Again, we recently just had Mr. Bingen talk about the 4% rule. Let's turn it over to Carsten and Christine. Tell us a little bit about what you found about sustainable spending in retirement.

- Yeah, so hi, I'm Carsten, and people often ask me, Carsten, you write so much about safe withdrawal rates, you must be really nervous about running out of money in retirement. And actually, it's the other way around. So people ask me, you must be really nervous about running out of money, and I say, no, it's actually the other way around, by doing your analysis right, and looking at your numbers, and looking at your plan, and kicking the tires from every possible angle that gives you more confidence to retire, and gives you more courage to actually take money out of your account, right?

So unlike maybe some other fire bloggers, I actually take money out of my account. So I get a little bit of revenue from my blog, but most of my retirement spending comes out of my portfolio. And so when I got ready to retire, I mean, obviously I knew about Bill's work, right?

It's legendary, and it's not Bill's fault, but it was too generic for me personally, right? I'm not a 30-year traditional retiree, retiring at age 65 with a flat spending profile. I had all sorts of other parameters, longer horizon, some positive cash flows, some negative cash flows in the future, right?

I had a bequest motive. I don't want to run out of money after 30 years, because that would still be considered a success in some retirement calculators. So I needed to personalize what I do for my personal retirement. And some people say, obviously 4% rule is a good starting point, and it is, but these personal parameters, they can shift it all the way down to maybe three and a quarter percent for somebody who is 30 years old and has a 60 or 65-year horizon, or you could go all the way up to 5% for somebody who retires a little bit later who has social security right around the corner.

And so I wanted to do that exercise right. So then my niche in the FIRE community is that, I mean, I like to do this personalized analysis both on the personal parameters, and then also I'm very careful and cautious because this is really near and dear to my heart, both as an economist and former finance, asset management professional.

It's asset valuations matter a lot, right? Especially for people in the FIRE community, right? Because we retire endogenously, and we tend to have the retirement date close to the market peak, right? Because we retire when we have enough, well, that's positively correlated with retiring right around the market peak.

And so all of these additional factors that go in, and so this is why I started my series, and it has been therapeutic to me that I feel more confident in retirement. And a lot of people that reach out to me, they say, yeah, I mean, you gave me the confidence to retire and they also knew about your work, but they said, I mean, you pushed me over the edge to also retire.

- Oh, don't blame me. (all laughing) - And again, we don't blame him, we almost blame other people misinterpreting your results. I mean, that's really what happens sometimes as a researcher. And because you had to keep it generic, right? You couldn't publish a safe withdrawal rate study for say your neighbor, right?

Who has all of these bells and whistles, you had to keep it generic and general. - Typically if my neighbor is John Gotti, yeah. (all laughing) - So anyway, so this is my little niche. I publish a lot on earlyretirementnow.com. I have a free early retirement toolkit. It's a spreadsheet.

It's not a program you have to pay for. I post this for free. You can download this and run your own simulations. And it will always stay free. People have asked me, "Can't we make this a for-pay product?" And I want to keep this free so people can utilize that without paying for that.

So anyway, so that's my blog and people seem to enjoy it. - So Karsten's blog is really amazing and comprehensive. Our team at Morningstar has been doing a state of retirement income study every year for the past three years. And we very much stand on the shoulders of people who have had a go at this research before.

So Bill Bengen's research is the starting point for everything that any of us have done in this space. But our team does Monte Carlo simulations, which were discussed in the previous session with Bill, where we attempt to kind of take a forward-looking view. We embed our Morningstar investment management team's capital markets assumptions, so their expectation of stock and bond returns, as well as their expected inflation over a 30-year horizon.

They tend today to be pretty conservative in terms of their U.S. equity forecasts, somewhat more sanguine about non-U.S. equity and fixed income return expectations have gotten a bit better with bond yields rising. So first year we did this research in 2021, the takeaway was 3.3% would be a good starting withdrawal percentage if you had a 30-year horizon and you wanted to take the same amount out, inflation adjusted every year throughout your retirement.

That was partly because bond yields, as you remember, were very, very low at that time. Equity valuations were very elevated. When we revisited in 2022, the number had lifted a little bit, in part because of better bond yields. And so it was 3.8% in 2022. In 2023, drum roll, it was 4%, was the starting safe withdrawal percentage that we came out with.

But the interesting component of that research, I think, is that we do look at a number of different flexible or dynamic strategies, including John Guyton's guardrails strategy that he was here to discuss yesterday. And what we do find, there are a lot of different variations on these dynamic strategies, but you do tend to get a lift in terms of your starting withdrawal if you're okay with the trade-off that you will potentially have to take less if the market goes down.

So we explore the pros and cons of the various dynamic strategies. One of them is that if you do have a strong bequest motive, dynamic strategy will tend to be less good for that because it's going to encourage you to spend more if your portfolio balance is up. So that's a trade-off.

And like Karsten said, it's very individual-specific. Some people might say that's not a concern. I'm kind of a last dollar person, they might say. And so in that case, you might be just fine with a dynamic strategy. So it is very, very personal, but we explore the pros and cons and the lift that you get with taking a dynamic strategy.

- Thank you, Christine. Mike, I'm ready for you. Let's talk about variable portfolio distributions in retirement. This one came from Greg Wilson from Vogleheads Facebook. He asks about how does one incorporate the go-go, slow-go, and no-go years into your distribution strategy? And before I let you guys answer that, I'll have one of you tackle go-go, slow-go, and no-go for those folks who aren't retirement planning nerds.

- Yeah, I couldn't hear the question, sorry. - It was about spending changes throughout the retirement life cycle and the kind of go-go, slow-go, no-go trajectory that people have-- - For spending? - Yeah, yeah. - Okay, that smile thing, yeah. Yeah. - I can hop on that. - Why don't you start on that?

- Because we actually did model that into our research. We had some research from EBRI, Employee Benefits Research Institute, that we used on retiree spending. So we modeled in real retiree spending, which does tend to show that pattern, that actually real spending tends to decline by one percentage point less than the inflation rate, if you sort of average it out.

And so we incorporated that into last year's research. So if you're thinking that your spending will trend down throughout your life cycle, you can take more earlier on. So the 4% sort of baseline finding became 5% when we incorporated that natural downturn in spending. But of course, retirees need to be okay with that trade-off.

One quick note on the retirement spending smile. This is something I discussed in my book with David Blanchett and so there's been so much discussion about this smile idea. So spending is going down and then going up. And so the upward trajectory obviously relates to healthcare expenses, specifically uninsured long-term care costs.

Catastrophic long-term care costs, David pointed out, actually accrue to a fairly small percentage of our retiree population. So it's a big number, but it is landing with a small percentage of retirees. The problem is you don't know which cohort you'll fall into, the person who doesn't need much long-term care or the person who needs a whole lot.

So anyway, just for what it's worth. - So I mean, I can answer the two. One is for me personally and how I would handle this academically if I wanted to model this in my sheet. So first of all, the smile doesn't start for me yet, right? Because it usually starts at age 65.

So I still have a few years to go. I am actually more worried about the other way around that if you retire early, say 20 years before normal retirement age and you do only CPI, I forget about CPI minus 1%, you do only CPI and you're still in your prime years where you want to travel, you're engaged, you still want to buy the newest iPhones and computers and cars.

I would say it probably should be CPI plus X and plus X could be something like average per capita consumption growth, which is maybe one and a half percent. 'Cause if you don't do that, you don't adjust your spending the same way everybody else around you, your relatives, your neighbors, your friends would adjust their spending, right?

Because the CPI prices, well, the phone that you would have bought last year, well, the price for that goes down a lot this year, but I'm no longer buying the phone from a year ago or two years ago. I no longer, we no longer buy black and white television sets, right?

Or these boxy sets, right? We buy, so a lot of the CPI, the reason why the CPI is actually a little bit depressed has been in some of these technological goods are becoming cheaper, but they're becoming cheaper because of quality adjustments, right? So the actual out-the-store price is not going down as much as the CPI would suggest.

So I would personally be a little bit afraid of doing only CPI adjustment, at least at the beginning, but then I agree, obviously later in retirement, a lot of people slow down. And the way I would model this in my simulation sheet, there's actually one parameter where you can set a scaling parameter.

And the scaling means I can set this to one at the beginning and then I can increase by, say, if I want to do CPI plus 1%, I can let this grow by 1%. So it would model spending increases above inflation. And then later on, when that U shape, that smile happens, then I will take this down again and take it up again later when I'm maybe 85 or 90 years old.

And I mean, usually what happens is that if you're an early retiree and you're looking at, well, what is the potential for nursing home expenses later in retirement? It will have a relatively small impact on my initial withdrawal rate. But I can tell you that doing anything like the CPI plus X, it will take down my initial withdrawal rate for sure, because even though it's a very slow creep in your expenses, it definitely lowers your safe withdrawal rate.

And that's one of the things that we should worry about. Can we really afford to do only CPI plus zero? Do we have to do CPI plus X? And if you go way above CPI adjustments, it might mean that, especially for early retirees, you need to be more cautious and conservative there.

-Can everybody hear me with this mic, all right? -Yeah. -Okay, good. Probably what I'm going to say is going to echo what my colleagues have said earlier, but one thing I learned in my research over years is that until I've run the numbers, I'm better not off announcing any conclusions because this field constantly surprises me.

There are a lot of counterintuitive results that come out of the research. If your expenses -- and I haven't modeled an expense where you start high, drop, and come up, which is called a smile. And I guess the first question I ask is, "Your retirement kind of a Mona Lisa smile, or is it really a big laugh?" Yeah, we were going to drop sharply.

My instinctive reaction is to expect that the most important part of retirement are the first 10 years, my research indicated. What happens there to inflation, to your expenditures, to your withdrawals, to market returns, pretty much sets the pattern for the rest of retirement. So if you go into retirement and you're reducing expenses initially, I would expect it would have a favorable effect upon your initial withdrawal rate.

So I would expect that if you use the COLA method and you were going to come out with 5%, you might have significantly more if you were going to follow some kind of a smile, depending upon your sense of humor, you know, so. -Christine, in your research, you mentioned we can go from 4% to 5% if we assume a decrease in spending.

Are you also assuming an increase at the end, as well, an increase in spending? -So we did -- we looked at the ABRI research, and I believe it did have some elevation. But after talking to David, I think we did not incorporate, you know, the full effects that he has demonstrated with his SMILE research.

So because it's just not something that happens to everyone. Not everyone is hit with those big outlays later in life. -Got it. It's those outliers. So we have a related question, also from Boatlet's Facebook. This one is from Anthony, and he asks about how sustainable distributions in retirement change if you've got no Social Security income, and then that Social Security income kicks on.

How does that impact your distribution rate? -Yeah, I'm really -- I have, like, a dead spot here, if you could just repeat that for me. -How does Social Security income -- How does Social Security income -- -Oh, okay. How does that fit in? Yeah, I can start that. -I ignore it completely.

Sorry about that. The folk -- Obviously, if you're doing a complete financial planning analysis for an individual for retirement, critical portion. But my whole job is -- I think is, from the beginning, you've been trying to maximize what you get out of your portfolio, irrespective of what other income you have, whether it be Social Security, annuity, or Pinterest, so forth.

I like to say, think of your portfolio as a cow that you bought at retirement, and you want to get milk from that cow for your whole retirement. You want to know as much as you can about the maintenance and care and the feed and the housing of that cow.

So you get the most milk, and you want to also know how much milk you can safely get out of that cow without killing it, you know, at least not so soon. So that's kind of a weird analogy, I know, but I think if you think of it like that, that's what I do, and I don't ignore about all the other animals in the barnyard, like the goats and the sheep and everything else.

I'm just focusing on the cow. -Yeah, so, I mean, we have to distinguish sometimes traditional versus early retirees, right? Traditional retirees often retire and then claim Social Security at the same time. So you basically -- Many people just do a separate analysis. They say, "Well, I have my Social Security income, and then I take my portfolio, and then I apply the 4% rule, and then I run with that." Obviously, depending on the size of your Social Security, you could also say, "Well, maybe Social Security already covers all of your mandatory expenses, so you can take some more chances with your portfolio." So in that sense, Social Security impacts your safe-withdrawal-rate analysis.

For us early retirees, obviously, we want to factor in Social Security. I mean, some people say, "Well, I ignore it," right? And then they ignore it, and then they say, "Well, first of all, maybe the government is going to default on it." I don't think the government will default on it.

They might do a little bit of a haircut. There's some kind of benefit formula changes. It's possible, but maybe you give it a small haircut. But even if you give it a small haircut -- So imagine that you know that 20 years into your retirement, you receive Social Security benefits that are worth 2% annually withdrawal rate compared to today's portfolio.

Well, obviously, you can't raise your safe-withdrawal-rate today by 2%. Has to be a little bit less than 2%, but I think it should not be zero, right? It has to be more than 2%. And, of course, by how much you can raise your today initial withdrawal, and then you would take it down as much as you get Social Security later.

So that is a numerical and quantitative exercise. You have to put this into my toolkit and then see what it is. It could be that it's maybe half of it, and you can already use half of it as long as later on in retirement, you reduce your withdrawals. So you think of that as a multistage retirement.

But instead of thinking of that as some kind of a bucket and a mental accounting thing where you do everything separately, I think it should be done as one unified, holistic exercise and that that's the clean and proper and mathematically optimal way. And, yeah, I mean, you should factor in Social Security.

Don't discount it. -I would also just say that for our 2024 research, that is one of the real-world scenarios that we're modeling out where someone delays Social Security, and so that necessitates higher withdrawals earlier on. And so we're examining the impact on what starting withdrawals would look like in that scenario and how that affects later withdrawals.

And, John, in response to your previous question about how we modeled in long-term care, I'm recalling that the EBRI research that we used did not actually show that upward trend on the right-hand side of the smile, and so it was a little bit in conflict with David Blanchett's research that showed the smile.

I remember we kind of looked at it and said, "Wait, it's not even really going up," even factoring in the fact that some retirees have high long-term care costs. -So use 5% only if you're lucky. Okay. All right, we've got a couple questions comparing Carsten's and Bill's sustainable distribution rates.

Carsten claims 4% is too high. Bill claims you can go higher than 4%. Discuss. Okay. -Again, it depends on your personal parameters, right? And it also depends on what our asset valuations right now. So we're doing this conference in September 2024, so we'll be reminded the CAPE is at 36.

Whether it's trailing earnings or forward-looking earnings, they are maybe not all-time highs but very close to all-time highs, so I would probably be a little bit more cautious. But as I said earlier, right, if you look at not the generic safe withdrawal rate exercise but you look at actual volunteers, actual people that do their math, say somebody who retires in their late 40s, early 50s, and Social Security isn't too far away, right, you should take that into account.

Very quickly, you're way above 4%, even at the market peaks, right? And absolutely, you can squeeze out a little bit more if you factor that in. And by the way, I'm already prepared for what Bill is going to say. I mean, maybe you can squeeze out a little bit more with more diversification.

We've heard a lot about small-cap value investing. In historical simulations, absolutely, you can show that small-cap value buyers would have raised your safe withdrawal rate. I personally, I'm a little bit of a skeptic, so I'm in the Rick Ferry camp and Jim Dally because I would be afraid. I wouldn't tie my retirement to the small-cap value buyers picking up again right around the time I'm retiring.

And what if it doesn't? What if it deflates again? I would probably be, so if you talk about an alpha source, right, somebody could come up with some alpha source that increases my retirement spending. I want that alpha source to be relatively reliable over every single market cycle. So I don't have the luxury of saying, oh, yeah, I mean, every once in a while, we have some 30-year drought periods, but then it really takes off.

Well, after 30 years, you might be dead or bankrupt. But anyways, I'll let Bill weigh in on that one. So how do you want to raise that? I want to be sure I answer the right question. So you're asking, what specifically are you asking? So I'm the more conservative guy, and you say with the right planning, you can increase your safe withdrawal rate.

- No, that's question based. - 5% plus. - Yeah, there are, I mentioned in previous sessions, there are factors. I've identified four factors, which I call free lunches in retirement investing. Some of them are not so surprising. The top list is diversification. Every capable investor does that. You don't put all your money in Nvidia stock and ride it till the end of your life.

You know, you have other investments. Although that would not have been too bad for the last couple of years, I got to admit. Another one is rebalancing your account. That can add a significant amount to your, and you need to do that to get to the 5% level. Another consideration is to slightly tilt your portfolio towards a higher returning assets.

Don't do this in an excessive way. For example, I start with a portfolio of five asset classes, each have 11%. And I decide I'm gonna put maybe 14% in the micro caps and the small cap stocks, which have quite a bit higher returns over the long term than some of the other asset classes.

That'll give you a significant bump to your, like a quarter of a percent. And all these factors, when you put them in, add up and historically they've been proven really not to generate any additional risk over above what you would normally take. So I say if, you know, food's on the table, let's eat it.

You know, enjoy the meal. - So relatedly, we've got S&P 500, we've got bonds. What are the other asset classes that we are assuming is in the portfolios when you guys are calculating your distribution rates? - Now I have to admit, I didn't hear that right. Could you explain it to me, Bill?

(audience laughing) - I can repeat. So what other asset classes are you using in your forecast of safe withdrawal rates? - Right, so in my toolkit, I have U.S. stocks, international stocks. So on U.S. fixed income, I have short-term, something like three-month T-bill, 10-year bonds, 30-year bonds. And I have gold and commodities.

And I have the two pharma-French factors, the value and size factor. And you can play around with that. And for me, personally, I think that you can pretty much span everything by just doing the 10-year treasury versus U.S. stocks and play around with the allocation. And what I found is that portfolio 75% S&P 500 and 25% intermediate bonds, so something like the IEF, I believe.

So seven to 10 years ETF. That would hedge against kind of everything that the market throws at you, whether it's a demand shock recession like the Great Depression in 1929 to the 1930s, or this long extended event between the 1960s and 1982 with the sluggish market and then very inflationary episode with poor bond returns and also modestly poor equity returns.

So everything that the market throws at you, you can capture with that and get a halfway decent safe withdrawal rate. And now you can obviously micromanage this and find the absolute best allocation that covers all the different angles. Sounds like overfitting to me. So if I keep it relatively simple, I mean, at least I know going forward, we have an equity premium that's going to pay generously, I hope.

We have a bond term premium that pays you a little bit more than cash. There's a little bit of diversification between stocks and bonds, so it's not going to totally wipe out your portfolio in the next sequence of return event. So I'm relatively comfortable doing the 75/25. I'm really hesitant to find anything that is super optimized where I throw in all sorts of other factors like small cap and value and small cap value and energy and gold.

Actually gold at the margin, at least would have performed quite well. Of course, always with the caveat that in this country, we weren't even allowed to hold gold, but I mean, you can throw the return series at the simulations, obviously, with a little bit of the asterisks that actually gold would have.

I mean, a small allocation, something like 5% to 10% would have done a pretty good job at hedging quite intriguingly, both deflationary and inflationary recessions. But again, I don't want to stress that too much and I'm not doing it myself. I kind of did the math. If I look at my portfolio, if I wanted to put 10% into gold, I mean, I don't think I would be comfortable moving that into gold.

So that's why I want to do something very bare, very bare bones, nothing over-fitted. And if it's not over-fitted, I'm more confident that it actually works going forward and out of sample. - Our asset allocation, we do vary. So we look at all different equity bond allocations when we do the research.

At an intra-asset class level, we hold those steady. We don't vary them. Our main thing that we're looking at is the core asset classes and how varying those exposures change up the safe withdrawal rate. The kind of stunning takeaway for our base case last year was so we came up with this 4% safe withdrawal percentage, but it corresponded with a portfolio with just 20% to 40% equity.

And I was like, go back, let's do that again because that doesn't make sense. But really what it is is with what we're asking the simulator to provide for us is we're saying we want this sort of fixed paycheck, unwavering in retirement. And so given where yields were last year when we ran the simulations, it basically says back to us, okay, if that's what you want, if you never want to waver how much you take from the portfolio, you just want the same real amount, go for the fixed income today.

It was basically saying let's lock that down because of the much higher standard deviation with equities. So that was kind of an interesting takeaway. So the sort of secondary takeaway is that if you are willing to vary your withdrawals, you can take a higher starting withdrawal, as I said, and the highest safe withdrawal with a dynamic strategy tends to correspond with a higher equity allocation.

So just some interesting findings in our most recent research. - Yeah, absolutely, which brings me to a related question. You mentioned that optimal mix, at least for this year's study being between 20 and 40% bonds because you make an assumption of high yields from that bond portfolio. Carson, Bill, tell us about the optimal stock to bond mix if you want to get the highest distribution rate from your portfolio.

- Yeah, so as I said, I think 75/25 seems to be a pretty good mix. If you go too far on the bond side, it helps you during the Great Depression, but then it totally kills you during the 1970s and '80s, and then vice versa. If you're too aggressive on stocks, it's going to really kill you during the Great Depression.

And then, so you almost want to look at what's the, you look at the minimum between the two episodes, and then you maximize the, and again, this sounds a little bit like maybe overfitting, sausage making, you almost lose your appetite when you see sometimes how sausage is made and how statistics are made, and then also how economic forecasts are made and how safe withdrawal rate simulations are made.

Sometimes you don't want to know the details, but I mean, this is how I came to the 25/75, right? You look at what does the best, more robust that gives you a hedge against both a deflationary recession and an inflationary recession. Now, as I said before, you can play a little bit more with maybe we don't want to have all bonds that have duration, right?

Maybe we have a little bit more fixed income, but it's in cash, so something like three-month T-bills. You could play around with that too, but if I'm just tied to the stock versus bond, I found that the 75/25 seems to be pretty good, and then there's always this, I'm glad Clifford Essner is not here, so he would always point out, well, why don't you just take the tangency portfolio and lever that up?

Yeah, I mean, if you're very financially savvy, you can do that, but the average retail investor wouldn't be able to do that. So if you are bound to a portfolio without leverage and only stocks and bonds, I would propose 75/25. - Yeah, do you want me to? Yeah, my research indicates that between about 46% equities and about 70% equities, it doesn't make too much difference what you choose.

You're going to have approximately the same withdrawal rate. You get outside that range, like 75%, I'd like to see your numbers, I really haven't, I'd probably learn a lot what you're saying, but 75% is a slight penalty, but it's not a lot. So I can't disagree strongly with what you're saying.

I do want to make one point. I think that retirees, my research is based on a buy and hold strategy, or it assumes that. That's because it's very difficult to analyze anything that isn't buy and hold, but I'm of the firm opinion that retirees, particularly early in their retirement, the first 10 years, should not be buy and hold investors necessarily.

I think they should be more risk managers. I think their primary concern would be preserving their nest egg from a huge bear market, like we had in 2008, where people saw they lost 30, 35% of their portfolio. It came back eventually because of very aggressive actions by the Fed, but I just want to express the opinion, and I know there's a lot of people who disagree with me.

I don't want to suggest that you use market timing, which is trying to get the exact bottom to sell, to buy an exact top to sell. I don't think anyone in God's creation can do that reliably, but I think there are third-party services that offer advice in risk management, where as risk increases in the stock market, you reduce your allocation, and as more favorable conditions arise, you increase your allocation.

I just think, to me, that makes sense to preserve capital and not expose yourself to something we haven't seen before. Who knows what the next bear market will look like. - So I think even, Vogel probably would caution at market timing, but I think I agree with you that, for example, something like a reverse glide path in retirement, right, where you haven't, so people call it differently, reverse glide path, or a bond tent, right, which the bond tent is the two parts, the two legs of the glide path into retirement, and then during retirement, you start shifting up again.

So that means you want to have the maximum bond exposure at the time of your retirement, and then, almost as a hedge against sequence of returns, then you shift back into equities. And do you know why? Bill Bernstein's chart, did you see that one? But the negative correlation in stock returns, not year to year, not month to month, not day to day, but over larger time frames, over 10 years, 15 years, 30 years, if you have bad luck in your first 10 to 15 years, stock returns are then going to be much higher over the next 15 years, at least on average, and well, guess what, that's exactly where your glide path ends, and then you have higher equity allocation, right, when equities turn around again.

So that's where that intuition comes from. And then there's actually multiple ways of justifying that. One is this market history, but even with completely, basically, Monte Carlo returns, where you don't have any kind of mean reversion or negative serial correlation, is what it's called in statistics, even with completely random Monte Carlo draws from a normal distribution, you still have this impact.

And it has to do with, when you go into retirement, you're phasing out your, when you go into your last few years of allocation, you want to go lower risk, and then once you're in retirement, so think of your retirement savings as almost like an implicit bond allocation, and you phase this out.

To overcome that, you do the glide path into retirement, and then when you're in retirement, think of your withdrawals as an implicit negative bond allocation, right, because you're taking money out. It's almost like a bond, like a fixed income flow. I mean, it's literally fixed withdrawals, so I think of that as negative bond allocation.

And as you roll this out, and say you, and then as a bequest motive, you have, well, whatever you have left over, you leave to your kids, and you want to maximize that. And as you phase out your withdrawal, so that's the opposite effect, and now you go back into equities again, because you have less of this negative bond allocation.

So this glide path, bond tend and reverse glide path, it's very intuitive, even though target date funds do the opposite, and this hasn't really reached the target date industry, and there might be some legal reasons why they can't do it, but it's actually mathematically, and then from a financial history point of view, there are rationales that you want to have this reverse glide path in retirement.

- And Carsten, is that something you've looked at in your research, bond tend, and has that increased sustainable distribution rates? - But then again, don't get too delusional, right? I mean, it's not like, it's not like if your, if your bare bones safe withdrawal rate is 3.5%, it's not like with that reverse glide path, you can suddenly raise that to 4% or 4 1/2%.

And we are talking about something like 20 basis points, which is huge, right? Because 20 basis points, that could be something like six, 7% of your retirement budget. I mean, at least in historical simulations, you have a little bit of a higher initial withdrawal rate if you do this bond tend business.

Yes, absolutely. - Super. Bill, we've got a follow-up question for you. What is the fourth free lunch, only three were mentioned? - Yeah, that was the fourth free lunch, the glide path. - Yeah, exactly, yeah. - All right. Okay, at least two speakers mentioned that there is no safe fixed withdrawal rates and that you will have to adjust withdrawals as markets adjust.

Do you agree or disagree? - Well, I don't, I don't know that you have to, but our research would point to the value of doing so, that if you are willing to make, say annual, at least an assessment, you might not always have to do a course correction, but if you're willing to check in annually, I think that's a good practice.

And potentially you will have to take haircuts now and again, but you may be able to take raises as well, depending on what sort of dynamic spending system you use. - Yeah, absolutely. So for example, this is probably another way in which your research is misunderstood, that they think that a fixed withdrawal rate or a fixed withdrawal amount CPI adjust, means that this is a set it and forget it kind of deal.

And of course you want to check your financial plan occasionally. And I always view it this way. I mean, what if every year you re-retire, right? I mean, you look at your portfolio and you do another safe withdrawal rate analysis and you look at, of course, what would be really dangerous and probably way too conservative is what if your portfolio is down?

Are you gonna re-retire with this lower portfolio, again, at the 4% rule? Probably not, because now suddenly your withdrawals are just going to be as volatile as your portfolio. So you can probably do some kind of an adjustment. And there's even an economic and financial rational, right? Because if the stock market is down, now probably the CAPE ratio is down too, right?

So your portfolio may be down, but maybe now you can increase your safe withdrawal rate a little bit. Not by as much to completely undo the effect from your portfolio drop, but at least partially. And yeah, so I think it is absolutely, it's the financially, and I think it's also the mathematically sound thing to do to reassess your retirement plan.

And there's a whole, I studied economics and you talk a lot about dynamic optimization, but there's a whole, it's called the Bellman principle of optimality that every optimal plan has to be optimal at every possible stage in the future. So, and yeah, absolutely, you should revisit what your retirement plan looks like every, maybe not every month, but every year, check if you're still on plan.

And for example, I did this, right? I mean, I went through one correction in 2018, right out of the gates, a bear market in 2020 and another bear market in 2022. And so every time I checked, well, yeah, I mean, portfolio is slightly down, but my spending is that still in line with where it should be.

And then of course you also look at market valuations. Yeah, portfolio is down, but equity valuations are much better. And I always make this case, if the stock market is already down by 30%, I don't have to target again, a 4% rate, because that's calibrated to the worst case scenario of the 1929 bear market.

It's not like we are going to tag on another 80% drop from that 30% drop already. So take into account that, well, maybe don't compare today's safe withdrawal, if your portfolio is down, don't compare today's withdrawal rate to the peak before the 1929 bear market, but well, 1929 minus also 30 or 40%, where was the portfolio then?

And then check what would have been the safe withdrawal rate in that month, far away from that peak. So, and I mean, you definitely keep your sanity if you do that. Now, if you recalibrate every time with a 4% drop, that might become stressful, right? So you want to do it, you want to reassess, but reassess it in a sane way and don't get too conservative that way.

- Yeah, can I go next? - Okay. - Three years ago, I developed what I call a two factor model, where to develop the safe withdrawal rate for an individual, I look at current inflation rate, whatever the regime, what we think it'll be the next five years, and look at the valuations of the stock market using that Schiller-Kate cyclically adjusted PE ratio.

And I found out that those two factors used together, I have about an 85% correlation with what safe withdrawal rates were in the past. Well, that means about 15% of the time, it's going to give you the wrong answer. Half of that time, about seven and a half percent, the withdrawal rate that emerges from that process is going to be too low.

An example would be, let's say in the mid to late 80s, we would have forecasted a much lower withdrawal rate than was actually possible. And what happened, of course, is that from the mid to late 80s, which had fairly normal stock market valuations, valuations more than doubled into the 2000 dot-com bubble.

So investors had in their early retirement an enormous tailwind, which ballooned their portfolios beyond anyone could expect. And of course, since it happened early in retirement, it was maximum benefit. There are other cases where it'll give you the wrong answer, it'll give you the answer that's too high. So it's essential that an investment, what I call just not a withdrawal rate, it's an entire plan, and it's a retirement withdrawal plan, has to be monitored just like any plan, whether it be investment plan or a state plan, insurance plan, and you have to be prepared to recognize when that plan has deviated in a dangerous way that it might actually cause a downfall of the plan and recognize what steps you have to take.

And very briefly, if you run into an unexpectedly deep bear market early in retirement, probably your best strategy is do nothing because the market will probably recover from there. That's the one good thing about bear markets, they don't last forever. You know, they're gone and they recover. However, if you encounter a period of extended, unexpected high inflation, get ready to run for the hills, folks.

You're gonna have to do something to you actively to manage your withdrawal rate because that is a very bad danger because you're gonna elevate your withdrawal rates and it's gonna continue. You're stuck with those high dollar withdrawals for the rest of retirement. There is no respite. So, and there are other variations of it, but that's my overall view, take on things.

- Wonderful, thank you. Well, folks, that is gonna be it for our panel. Let's give them a round of applause. (audience applauding) - Thank you. - Thank you. - Good work. - Thank you. - Enjoyed it.