Episode 284: Prof. Scott Cederburg: Challenging the Status Quo on Lifecycle Asset Allocation

27:34You're getting this thing that over longer horizons is getting riskier. It's becoming more correlated with stocks, and it's just not compensating you much on the average return side. Wow. Inflation is persistent and is not predictive of higher future bond returns, whereas with inequities, you have that mean reversion. Yeah, exactly. There's a paper that you may think of, I think it's from Cliff, from EQR, called International Diversification Works Eventually. They basically look at that like the short-term versus long-term correlation on domestic and international stocks. Yeah, exactly. It's my belief anyway that we should really be concentrating on return properties at the horizons that people actually care about.

28:14We know so much about monthly returns, but how many of us have a one-month holding period? So true. How important is social security? You're modeling social security. If you take that out, is 100 % equities still optimal? 100 % equity is still optimal regardless of what we would do with the Social Security piece. The reason for that, the distributions are all better throughout the distribution. So wealth of retirement, it's just giving you a better distribution of wealth of retirement compared with the other strategies. Retirement consumption, the differences across the strategies are all coming from the withdrawals from the strategies and not Social Security.

28:56the way we set up the simulation is we have eight parallel couples that all in each month have the exact same income as each other and the exact same contributions to Social Security and everything like that and the exact same longevity. The only thing that differs across the couples is the strategy that they pick. Social Security is mainly in our paper to give us maybe a little bit more realistic sense of utility-based calculations in retirement. Like it's not realistic to say that you go to zero consumption if you run out of money. There's still something there. It's just not going to be the lifestyle you were hoping for if you run out of savings.

29:35How would adding inflation protected bonds affect what you found? It's tough for us to like explicitly look at this in the historical data because these things just haven't existed for the entire history and they don't exist in all countries. even now. I think then you were saying that Canada has kind of phased out their program. These things, if you do have access to them, the way that we've tried to look at it is imagine that there is some hypothetical tips where you're just going to get, I'm probably going to say tips a lot. So treasury inflation protected securities in the US for tips. So imagine that you have these inflation-linked bonds that are just going going to pay you 1 % real rate.

30:22We try to do maybe 0 % or 2%. One of the risks with tips, if you're waiting until retirement to put part of your money in this, I think real rates right now are pretty high. So you would be getting a pretty good investment. But it wasn't that long ago, the real rates, even on long-term tips were negative. So there's a little bit of that kind of reinvestment risk that you have with the tips. As we looked at the trade-offs, If we put like 20 % or 40 % of money in tips after retirement, they're still not good for the wealth accumulation phase. But after retirement, putting 20 % to 40 % in tips, it was this trade-off where the ruin probability comes down.

31:04You're more likely to maintain consumption throughout your life if you're using a 4 % rule, but your bequest is going to go way down too because you're not generating nearly as much wealth. And then if you use one of these alternative withdrawal strategies where you're withdrawing more, if you make a lot of money in your portfolio, if you sort of link withdrawals to the savings, then it's a tradeoff of, do I want to avoid super low consumption at some point? Or do I want the possibility of getting more? And so there is just kind of a safety tradeoff that would be different across different investors, I think.

31:40And you're not modeling taxes, right? Which could be a problem for inflation-protective bonds? That's another practical issue with inflation-protective bonds is you pretty much, I mean, I think you want to have those things in a retirement account or perhaps the TIPS funds. I mean, those have a lot of volatility on kind of a year-by-year basis potentially, but there are kind of implementation issues with the TIPS. Same issue in Canada. Although, like you said, those are being phased out anyway, so we don't have to worry about them. How much more would this household have to save in target date funds to fund their retirement consumption to match that of someone investing in stocks through the full life cycle?

32:22Our investors in the base case are saving 10 % of their income as they go. And so we run this kind of utility comparison during retirement. The couples are getting utility from each month of consumption, and then they're getting some utility from bequest to their heirs. And so if we try and do this comparison, what would it take for me to save in a TDF to get the same kind of happiness in retirement as this all equity strategy? And you would have to save 14.1 % with the TDF throughout your entire life. So that's relative to the baseline of 10%. You have to save, in some sense, it's 4.1 % more, but that's 4.1 of your total income.

33:08And it's like a 41 % increase in the amount that you're actually saving in order to achieve the same outcomes. Wow. Those are real numbers. Can you talk about the downsides of investing in an all equity strategy? The really big one is been sweeping this one under the rug up to this point, but it's drawdowns, intermediate drawdowns. If I'm kind of defining a drawdown, this is like a peak to trough change in your account balance, ignoring any withdrawals that you've done, just like what happened to your assets or your asset returns over a period. And so again, we're thinking in real terms, but as an example, in retirement, the largest drawdown that a couple will face in retirement on average across simulations for a TDF, there'll be some time on average where you're going to lose 38%.

34:01So you have a peak to trough dip of 38 % of your asset value. And for the internationally diversified stock portfolio, that's 50 % on average. So at some point, you're going to have a bigger drop, like stocks are more volatile over relatively short periods of time. And so you do have to kind of be prepared for it's going to be a little bit bumpy. But then all of those eventual outcomes that we were talking about, wealth of retirement, wealth of death, retirement consumption, probability of running out of money, all of those paths include those big drops in stocks. stocks but at least if you you have a big drop in stocks we see quite often in the data anyway that the stocks kind of come back whereas if bonds crash especially if it's just losing real value due to inflation it's never coming back and so the issue with bonds is if you do get hit that 30 loss over a two-year period and in bonds in the u.s i mean the only way that you can get some of that back if interest rates go back to zero, but you're never going to get back that inflation eroded portion of that wealth.

35:11So it's a much bigger average drawdown, but you're still better off on all evaluation metrics. It's very much like an eventual outcome versus sitting there and looking at your account balance every day. Those are going to be two very different experiences for you if you're one of these all equity investors. Man. Barring behavioral issues, you just alluded to, and we can come back to that later. Just objectively, how much of a risk do you think large intermediate losses actually are for people who are saving for and living through and funding their retirements? We force our investors in the simulation.

35:50They have no emotion. We tell them what strategy to do. They are very loyal to the strategy that we tell them to do. That becomes the key. It does come down to the behavioral stuff. I know US regulators are worried about drawdown risk. I think that's just something that we need to confront head on effectively and say like, this is something that exists. If we do a strategy that is likely to give us better outcomes, but has these intermediate drawdowns. okay, now we just need to figure out what are we going to do? Because the obvious issue is people panic and withdraw their money from risky assets when they crash and probably have the highest expected returns.

36:37At least with stocks, we think that after a crash, that's at least partially because expected returns are probably really high at that point. You have to stick it out. So keep going, Scott. What about in retirement when retirees often talk about the sequence of returns risk, which is really these intermediate losses. What does all this say about that? Our investors don't respond to sequence of return risk at all. Some of it definitely is going to be coming in in this ruin probability that we have. The 4 % rule is subject to the sequence of return risk. If you have an immediate crash in stocks in retirement, you're more likely to run out of money.

37:17that risk is baked into the results that we're getting. Different withdrawal strategies are going to handle that in different ways. One thing that we did at one point is just say, okay, so this is the scariest thing. You're a retiree, you're 100 % stocks, and then stocks crash immediately upon retirement. And so we actually looked and we're like, what if we just look at the worst outcomes in stocks for that first year of retirement? What's the best strategy to be in, it's still the internationally diversified stock strategy. Whoa, that's wild. And we can go even throughout the entire retirement period and say, okay, so the worst case scenario is that stocks do poorly while you're retired.

38:01What if we take the 20 % worst outcomes for stocks during your entire retirement period? So in that case, so the baseline ruin probabilities were 17 % for TDF, 8 % for the stocks, for this stock strategy. If we're in this 20 % set of worst outcomes, then it's 34 % for the TDF and 24 % for the stock strategy. It's worse to be in a period where stocks do really badly, but the TDF actually still does worse than the all equity strategy when stocks do poorly. And I think this goes back to if we look at the long-term relations across assets from our previous paper using these data that we talked about last year, over a 30-year period, stocks are going to not beat inflation with about a 12 % probability and bonds, that's 27 % probability.

38:52But if we look at a loss period for stocks, if stocks lose over a 30-year period, bonds lose 61 % of the time over that same 30-year period. The TDF has tried to get you away from the risk that stocks are going to go down by getting you into bonds. But if stocks go down, bonds probably go down. And so that's not a safe place to go. And stocks still like 4 % higher average return than bonds. We're not saying that this is all like safe and rosy outlook on everything. If you're in all stocks, we're just saying the data tell us that there's not good alternatives to that. With the potential exceptions of these annuities, like inflation-linked annuities or inflation-protected bonds are potentially going to be a helpful part of retirement.

39:42But other than that, it's just sort of, you kind of have to take some risks and hope that you have a good outcome. Stocks and bonds will go down together in the long run because people will hear that and think, well, no, bonds go up when stocks crash, but that's like that month or maybe that year. But And in the long run, if domestic stocks do badly, bonds will also tend to do badly in the data. There are interesting kind of lags. I guess I don't even know if I mentioned how we're modeling our long-term returns, but we're using something called the block bootstrap approach, where we're taking this historical data and we're going to be randomly drawing periods from the historical data and coming up with a return sequence that lasts your entire life.

40:23but we're taking on average 10 years of consecutive data from the same country. So we'll get all four asset classes from the same country and taking 10 years on average, we're able to maintain some of these longer term asset return dependencies. I think a nice thing about the bootstrap is we don't even have to take a stance on what it is that we're trying to find in there. You're just asking the data, how does stuff behave over long periods? And there are these interesting lags, I think, between stocks and bonds, even stocks and inflation, where going back to the 70s, like Fama and Schwert are showing there's a negative relation between nominal stock returns and inflation at short horizons.

41:11But then if you look over longer horizons, it sort of has to be the case that if we have a ton of inflation over the next 50 years, stock prices are going to be really high. There's a major investment firm that reports in one of their white paper sort of things, a 70-year correlation between nominal stock returns and inflation reports a correlation of 0.1. In the state of the world where a loaf of bread costs$100 million in 50 years, I think the bread company is going to be worth more than$600 million. At some point, these two things have to be really tied together over the long run. But I think it's just very important.

41:48If you're trying to look at long-term returns, look at long-term returns. We mentioned investor behavior recently. There's a bunch of research that we looked at recently showing that investors underperform the asset classes they invest in and increasingly so with increasingly volatile assets. And for equities, depending on the paper you look at, it's like one and a half or 2 % per year. How do you think investors should factor that into their asset allocation thinking? I mean, it comes entirely from, are you pulling money out of the market at certain times? So this is, I mean, relative to like a buy and hold strategy, it all comes down to, are you committed to the strategy or not?

42:28I acknowledge it's a large problem to think about is that people are going to have to actually just sit there while their retirement account is crashing and like not respond to it. But that's what has to be done in order to realize these gains. From their perspective, it feels like they're watching the world end because that's what it always feels like. But in your sample, there are cases where the world hasn't ended, obviously, but where countries have economically ended, at least as represented by their stock market, and things have still worked out okay. It's still coming out in the data that these are the best outcomes.

43:08In some of these periods too, where just stocks have gotten hammered in some of these countries, but nominal bonds have not exactly been the greatest thing in the world either. You've said that a few times, that it's like, you're not saying stocks are safe. You're just saying, given the opportunity set of stocks and bonds, bonds look terrible in the long run. We have to behave like the data in your sample. That's the message. Turn off the human brain and behave like data. it's kind of interesting i don't know how to think about the entire range of policies that are feasible and possible the potential for just like reporting when i go on my fidelity account i swear it depends what the graphs look like in terms of if they look like this because things have been going down fidelity will try and find some range of time where it didn't go down the whole time and show me that graph.

44:05And then if things have been going well, it shows me a graph that's going up. I don't disagree with this sort of an approach. It's finding some way of kind of giving somebody a longer term view of their account performance versus just what's my dollar amount right now versus the highest number that I remember. That's kind of the problem that we face with this type of thing. Some sort of adaptive feedback loop. There are systems in some other countries too, where you're just invested in a pension fund and that pension fund invests your assets and you're just invested in that pension fund. And so that where the investors have a little bit less personal control over the asset allocation.

44:49I mean, it feels like you're losing control, but maybe that's not such a bad thing if you're kind of locked into a strategy. How important are return dependencies like mean reversion in stocks in your results? Super, super important. Mean reversion is really important. Both domestic stocks and international stocks, we see this mean reversion effect, so they get safer over longer horizons. But I think even potentially more important are these kind of unmodeled, and I don't even know exactly why they happen, but things like stocks and bonds becoming more correlated over time. We didn't have to explicitly model what was going to happen with inflation and currencies.

45:30But in the data, it shows up that over longer periods of time, exchange rate fluctuations tend to offset some of your local inflation and make international stocks safer over time. We think it's really important. We're thinking about long-term investors. We want to know about long-term returns and you kind of have to just go grab the data and make sure that whatever we're saying about long-term returns is consistent with what's going on in the data. Can you talk about how your results change when you switch to using monthly returns? We ran one analysis. Our base case, we're using this full developed country sample and we're using this block bootstrap where we're trying to preserve the time series dependencies in the data.

46:14So we're taking 10 years on average of data. There's an alternative for, I guess, a couple of alternatives. One, we could use the US data. And then two, we could use what we would call like an IID bootstrap. That's just going to assume that months are independent of each other. And so in that bootstrap, you just go and we have 30 ,000 months of data and you just go pick one of the 30 ,000 and you go pick a different one of the 30 ,000 and you string these all together. And what that does is it breaks up any of those time series dependencies in returns. If we look across these, you can then sort of think of four different approaches.

46:51We have our base, which is develop country sample, block bootstrap. But then you could think about doing, well, US sample, block bootstrap. Or we can do our develop country sample with the IID bootstrap or the US sample with the IID bootstrap. So we can go across those two dimensions, get kind of fill in those four boxes. And what's interesting is if we do either switch from developed country sample to U.S. sample or switch from block bootstrap to IID, the gaps kind of close between, say, the TDF and the all equity strategy. They close a little bit. All of the ruin probabilities come down relative to our base case.

47:35but the stock strategy actually still dominates. If you're changing either of those two assumptions, the stock strategy still dominates. It's only when you go to the US IID approach. So you're saying, whatever, we have 130 years worth of US data and you say, okay, that's sufficient. And that's what I want to learn from. And then you're taking the stance that returns don't have any time series properties. So in that scenario, stocks still tend to be better at wealth accumulation, but there are lower ruin probabilities on some of these stock bond age-based strategies. There are investors who would prefer, in that case, get into bonds.

48:19But that does rely on a couple of pretty strong assumptions. I mean, we can argue about US versus developed countries sample. And we always have this argument about whether the US is special, whether other developed countries are exactly the same or not. And there's a debate to be had about that. I think the really tough debate is returns are not IID. We know that volatility is time varying and we can all like see these patterns, even just like looking at the market for a few days, you know, that volatility varies over time. And there's a lot of evidence that there's variation of expected returns and mean reversion.

48:56And we see continuation and bond returns and all this sort of stuff. So we think maintaining those aspects of the data is really important. If you're going to do that, then it actually doesn't matter anymore whether you're using the developed country sample or the US sample, you're getting the same answer, where the all equity strategy is safer than the TBS. Wild. That part's crazy. I think a lot of people use either monthly returns or just simple Monte Carlo that is assuming IID and that you would get the result that you're describing, which is a result everybody gets. A lot of the academic literature, we assume that returns are normally distributed, IID or something like this.

49:36So it does kind of extend around. There's another approach that seems to be pretty commonly used in industry that's like vector autoregressions, where there's an attempt there to model out long-term outcomes. But the structure of these things tends to be that you're going to use pretty short-term data to estimate stuff. and then you try and estimate the dynamics. I think a very tough thing for a vector autoregression to pick up though is there's so much reliance on these short-term correlations to figure out how things are related over long horizons that I just don't see how they're going to get stock bond correlation to go from 0.15 at a short horizon to 0.5 at a long horizon.

50:16It's just such a dependence on these really short-term correlations and I think that that's probably where a lot of things are getting missed. Given all this, and I mean, obviously, this is a new paper, so people didn't have the information. Maybe that's part of the answer. But why do you think retirees are doing stuff that's more similar to the traditional advice? Everybody's doing what they're told. Your listeners are thinking a lot about their asset allocation and doing their own analyses and all these sorts of things. If we look across investors like retirement savers in the US, some numbers by Vanguard recently, I mentioned these QDIAs where employers can put this thing as the default and then they're free of liability.

51:00So 98 % of employers have a target date fund as their default. And then if you look at the investors, 83 % of investors hold at least some of their money in a target date fund, and 59 % have all of their money in one target date fund. So it's three-fifths of Americans are just doing the thing that they're being defaulted into, or they are taking that as financial advice. And most people don't really think about it that much. So if that's the status quo, then that's the status quo, and that's going to tend to be what people do. Yeah, it makes sense. I want to ask about home country bias. This paper is mind-blowing for all of the reasons that we just talked about.

51:40This part, I'm very interested to get some answers here. The optimal portfolio is 50 % domestic, 50 % international. You show in the appendix that going down to 35 % domestic is marginally better, but that 35 % domestic, that's still a massive home country bias for investors outside the US. What is driving that? Why is there such a large optimal allocation to domestic stocks? And sorry to keep going on the question, but we talked earlier about how much better international stocks look in the data. I just don't get it. So a few aspects, starting out with maybe the potential of at least having some domestic relative to international.

52:19There's still a long correlation, not huge. There is certainly positive correlation and relatively large, but it's not 0.9 correlations or anything once we're out at 30-year horizons. So there's some just plain old diversification benefit. I think part of it, it's a little tough to pin all the million couples down on why they preferred what, but I think a good part of it is probably currency stuff. If you think about over a long horizon, like a 30-year period, if your currency appreciates over that whole period, then that tends to hurt your international investments. because in the US, I would be selling dollars to buy the foreign currencies to buy some stuff, and then eventually I would have to buy dollars back again.

53:06And if the dollar has strengthened, I'm able to buy fewer dollars. So if we look at those cases where the currency has appreciated, it also tends to be the case that currency is appreciating in a country because the economy is doing fairly well and the stock market's doing fairly well, typically in those same sort of periods. So we do see that domestic stock returns are better than international stock returns if your currency appreciates over a long period. And then if the currency depreciates, the opposite on both of these. I get to buy back cheap dollars. So my international investments do really well.

53:40And then the domestic market has probably suffered a little bit if the currency is weakening. So I think that's probably part of what's going on. One thing that we looked at a little bit in our previous paper, because all the stuff that we're talking about now is completely unhedged on the currency side for the international stuff. It's a little bit tough to perfectly look at this in the historical data because we have to make assumptions. We don't have currency derivative data for this entire history. So we would have to make some assumptions about currency hedging and stuff. It might be possible that if you were able to partially hedge your international portfolio or something.

54:22I don't think you want to fully hedge the currency risk in your portfolio because it has this nice property of offsetting local inflation. But that currency exchange rate volatility, if you're really loaded up on international stuff, might become a little bit of an issue. So it may be the case that partially hedging that pushes down the domestic piece. Interesting. Okay. So partially hedging some of your foreign currency exposure might push down the optimal allocation to domestic stocks. It's speculation, but I think if that's where some of the risk is coming from, then it seems to make sense.

54:56I don't think you want to get 100 % of that volatility out of your portfolio, but I don't think we ever looked at partial hedging. We looked at either currency hedged or unhedged. John Campbell has a paper, I think, looking at that. We asked him about it when he was on about optimal hedging, and I think it was partial hedges were better than unhedged or fully hedged. So that is interesting speculation. Man. On the currency effect, are domestic stocks hedging local consumption because of the currency exposure? Is that a way to say that? It's going to tend to be the case, I guess, that the domestic stock market is going to be doing better in these times with relatively better local economic growth.

55:45It would probably also be relatively lower local inflation if the currency is appreciating. That's kind of the times the domestic stocks would be doing better is relatively good times. But perhaps another way that might affect this, we haven't yet built in any correlation between our labor income and stock market outcomes. So it's also possible we're planning on doing that and basically building in some correlation structure between human capital and stock market stuff just to see whether that pushes anybody off of stocks. If that's all at the local level, that tends to be at the domestic stock level that could push people a little bit away from domestic stocks there too.

56:33When you take into account any correlation between domestic stocks and labor income, that might push down the home country bias. I think also, if we're looking at the economic magnitudes of some of these things, one of the things that we've looked at is basically the whole range of, what if I go 5 % domestic, 10 % domestic, all the way up to 100 % domestic? If you take this 50-50 domestic international investor, and you force them to do 100 % domestic, they're angry. This couple saving 10 % now feels like they have to save 16%. They do not want to be 100 % domestic stocks. But this 50-50 couple is indifferent between being 50-50 and 20-80.

57:20So like 20 % domestic, 80 % international. And even all the way down to 5 % domestic, 95 % international. The savings rate is still only 10.6%. Once you get on that side of the midpoint, it's a lot flatter over the outcomes. So the difference between 35 % and 10 % in domestic stocks is not going to be super, super huge from an economic perspective. And so then it probably comes down to what makes you feel most comfortable. Like in the paper, we've been talking about the 50-50 because one, it's kind of a simple rule of thumb. And we are like specializing to the US in the sense that we have US longevity, US social security, and then the US is roughly 50 % of the global market.

58:12And so something around market weights seems to make sense. for the euro zone i mean i think because so much of this is kind of currency driven i think if you're in the euro zone you probably want to think of the euro zone as your domestic market at this point in history and so that 35 in the euro zone probably doesn't seem too outlandish so that's again probably roughly in line with market weights the smaller markets like Canada, so you own currency, but smaller market, maybe these results, maybe bump it up a little bit, bump something up, do a little bit more home bias or something like that.

58:52Or maybe it's a little bit of a justification if you are doing home bias. But if you really feel like doing market weights, I also wouldn't argue too much with that. As long as you're not doing 100 % domestic, then you're probably doing kind of okay. Would adding in additional cost by owning international stocks push the optimal home bias further? I think certainly that would push it in that direction. The US at this point, everything's pretty good. My friends around town, I tell them put half each in two ETFs. So VTI and VXUS, get them domestic stocks and international stocks. And I think the weighted average fee on those is about five basis points per year.

59:37We don't really look at fees in our setup, but certainly going to have an impact. I still think if I'm kind of eyeballing the distributions, I think that the international diversification is still going to be better on the left tail than even if we're taking some fees out, it's still going to be better on the left tail than a lot of the other strategies. but just depending on the size of the fee gap, it's going to lose average performance relative to domestic. Can you talk about how the findings in this paper relate to the findings in your paper on safe withdrawal rates? Because in that paper, you show the 100 % stocks, although it was domestic, and I just gave the answer that away, is dominated by 60-40.

1:00:22But can you just talk about that? So that paper, we had all domestic strategies. So it was domestic stock bond approaches, And 60-40 is better than domestic stocks, largely because of that left tail type of risk that you get in the domestic stock distribution. So that was what we were looking at in that paper. So basically, safe withdrawal rate is better at 60-40 than it is at 100 % domestic. But once you add in international stocks, an equity portfolio dominates on safe withdrawal rate. Yeah, the geographical diversification is better than the stock bond diversification. Do you have, by chance, the safe withdrawal rate at the 5 % ruin probability for the 50-50 portfolio?

1:01:03So we ran 3 % and 5 % rule cases. I can interpolate between 3 % and 4%, and it's about roughly 3.4 % is the withdrawal rate. It's much, much higher than for the target date fund or where any of these other things are still going to be much lower withdrawal rates. The TDF is still whatever it is, 2.3 % or something like that. So if you are willing to do this and you are willing to not touch your portfolio, then we would argue that a little bit higher withdrawal rate, kind of this 3.4 % is potentially feasible. And that's based on the average mortality? It's for the couples in our simulation. So it's uncertain longevity.

1:01:52But with the 50-50 equity strategy, roughly 3.4 % withdrawal rate. It's a big jump. Willing and able, of course. I'm curious, Scott, what aspects of all this work surprised you? The relative safety of 100 % equity even during the retirement period, I think, is initially counterintuitive. We've been working with this for so long. We've seen a lot of the tail risk for bonds, especially in real terms and stuff like that. We had a sense of what we were going to be getting in this, just that investors are pretty much completely not interested in holding bonds at any point was a bit of a surprise. I think the other thing that was surprising to us as we were doing it, if I'm telling the story of us writing the paper, a thing that was more difficult than I thought was reconciling with the conventional wisdom.

1:02:47Where do I find results that tell me that this TDF is better than the all equity strategy? From our base case, we're doing like these one-off change this assumption, change this assumption, change this assumption. And we're not finding any single assumption that gets us back to a TDF being an optimal strategy. It took this combination of US data and an IID bootstrap. So ignoring the time series properties of return to get us there. I'm probably revealing how slow I am at research, but I think that was like three or four weeks to be like, maybe I should run this combination of US and IID. But then that was finally the thing where you're like, oh, that is the thing that makes the TDF look safe in retirement is just this very special modeling approach.

1:03:36Man. So the lifecycle investing advice that everybody gets is basically based on the survivorship bias and easy data bias sample that everyone looks at. it seems to be the case that that's heavily influential, like the US data are just heavily influential on this. And then the modeling, how we're modeling long-term returns, I think is super important for this problem. You tested withdrawal rates from three to 5%, and at 5%, stocks level lower probability of ruin than the other alternatives. Yeah. Wow. This is off the top of my head. I want to say the TDF was up to like 35 % ruin probability or something.

1:04:18And the diversified stock strategy was at 16, 17 % something. Is that in the paper? We have it in the appendix, or at least a picture in the appendix. I have to dig that up. There are a lot of appendices in the paper. Yeah. I didn't read all of them yet. Crazy. All that points to the fact that it continues to be a much lower probability of ruin at higher withdrawal rates. Again, just has me thinking about whether sequence of return risk is really something that people should be as concerned about as they are. Those withdrawal rates are the things that are most subject to the sequence of return risk.

1:04:55Those rules are what open you up to that. And our investors are doing nothing to mitigate that risk. They're just literal ones and zeros in a machine. So therefore, behavior is what really matters? Moderating behavior is going to be central in all of this stuff. If anybody's ever actually going to do this type of an approach, I mean, I'm doing this approach, but I probably am a little better than most at remembering the long-term aspects of these things. This research, as with your other research, but this one even more so really touches a lot of the advice that we give to clients. So from where you sit and in your opinion, how do you think these findings should affect how financial advisors talk to their clients?

1:05:41I think our goal with this paper is just sort of opening up the conversation even a little bit more. I mean, there's always talk about what life cycle investing should look like, but let's just even have more of the conversation. I think on the very like technical geeky side, I think there should be more conversation on how do we model long-term returns and how do we make sure that our models that we're putting into these things, like in industry and academia, how do we make sure that the models that we're putting in are generating long-term returns that match the properties of actual long-term returns?

1:06:16That's a really important part just from a technical perspective, because that's where a lot of the advice ends up coming from. And then at sort of a client level, I think there's a couple of psychological aspects, perhaps, I think are going to be really important in figuring all this out. One is it still every time I present anything on this, people are like, but I just want to have some safe bonds over here. It's like, well, that would be great, except they don't really exist with the exception of potentially tips in the US. but bonds seem so safe to everybody. And that's just ingrained in everybody's mind.

1:06:53And I think that's going to be an issue that needs to be overcome if we're pushing people off of bonds a little bit is just sort of changing the perception around that. The second thing, just being like, if we implement this stuff, those drawdowns are going to happen. And then everybody has to get talked off the ledge and very much long-term focus, outcome-based focus sort of stuff. I think these are all challenges. There's whatever it is,$600 billion per year in new retirement savings every year in the US,$600 billion. And we're finding large differences in outcomes, like economically large differences.

1:07:34This is just a huge problem. And I think we should dedicate some resources to making sure that we're getting as close as possible to kind of doing the right thing. The volatility piece is so interesting. I was talking to someone earlier about liquid private equity investments. And one of the things they said that I found just fascinating related to this discussion is that the best thing about a product like that is going to be that it's going to be less volatile than public equities because the underlying assets are valued less frequently. So he's like, it's not less risky, but it's going to look less risky.

1:08:05And behaviorally, that's going to be good for investors. Like that's such a weird roundabout way of getting to a better place for investors. And it probably costs a bunch more. Some of the reporting stuff, if you can report smoother things or more long-term things, that's advantageous. It would be interesting with liquid private equity stuff to see the extent to which the investors are guessing at what the current value of the private assets that are only being valued every once in a while, but the price of that fund probably doesn't have to be exactly equal to the stated net asset value. That would be a really interesting view on how investors are working with very limited information to come up with current prices for things.

1:08:46That would be interesting. It's just so funny how the psychology plays into decisions where it's like, even if we know something is objectively optimal, people still won't do it just because it moves up and down a lot from day to day. Well, Scott, I think that's all the questions we have. This has been an incredible conversation. This paper is awesome. It's changed the way that I'm thinking about stuff. I think it'll change the way a lot of people think about their portfolios. So it's a big contribution. Very well done. Awesome. Yeah. Thank you guys so much. It was great talking to you. Yeah. Great to see you, Scott.

1:09:17Thanks again.