Lifecycle Investing - Still the Right Strategy

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
User avatar
Topic Author
Ben Mathew
Posts: 1409
Joined: Tue Mar 13, 2018 11:41 am
Location: Seattle
Contact:

Lifecycle Investing - Still the Right Strategy

Post by Ben Mathew »

With the sharp market declines we experienced recently, young investors who invested very aggressively in the stock market (100% or more) may be having a hard time and may be questioning the wisdom of their aggressive stance. It's normal to be scared, but do try to keep it in perspective. The arguments for behind being aggressive when young is compelling.

Lifecycle investing tells us that you will minimize lifetime risk for a given return by spreading stock risk evenly across time. Since you have very little savings in the early years, that entails putting all of it (and maybe then some, via leveraging) into stocks. In return, you get to reduce your stock allocation when older. Basically, you end up with a steeper glidepath than what is typical. Yes, this means that if stocks fall 25% when you're young, you'll lose a large portion of your savings. The point to remember is that your current savings is not a lot of money. Most of your savings is in the future. Because of that, a 100% loss when you are 30 is smaller than a 50% loss when you are 60.

While this is true, the loss of a large portion of your savings when you are young can be very difficult to handle psychologically. So my advice for most people would be to stop at 100% stocks and not go push into leveraged territory, unless you are really understand the mechanics as well as your own psychology and can be really rational about it. I know that when I was younger, the $200K in savings we had cobbled together over a few years felt like a lot of money even though it wasn't very much from a lifecycle perspective. My perspective was skewed. I was invested 100% in stocks, and I was comfortable with that. Leverage would have been harder.

Whether you are 100% stocks or in leveraged territory, remind yourself of the principles behind lifecycle investing and why it's optimal from a risk management perspective to be aggressive when young. Paradoxical as it seems, you are taking on less lifetime risk. The plan is good. Stick to it. There will be many more ups and downs to come. Averaging over it all is the right way to go.

Good luck!
User avatar
nisiprius
Advisory Board
Posts: 42586
Joined: Thu Jul 26, 2007 9:33 am
Location: The terrestrial, globular, planetary hunk of matter, flattened at the poles, is my abode.--O. Henry

Re: Lifecycle Investing - Still the Right Strategy

Post by nisiprius »

As far as I know, nobody but Ayres and Nalebuff (and their followers) call this "lifecycle investing." They've appropriated a term that had a different meaning, just as they've appropriated Paul Samuelson's name for a strategy he explicitly denounced.
Annual income twenty pounds, annual expenditure nineteen nineteen and six, result happiness; Annual income twenty pounds, annual expenditure twenty pounds ought and six, result misery.
User avatar
Topic Author
Ben Mathew
Posts: 1409
Joined: Tue Mar 13, 2018 11:41 am
Location: Seattle
Contact:

Re: Lifecycle Investing - Still the Right Strategy

Post by Ben Mathew »

nisiprius wrote: Fri Mar 13, 2020 12:34 pm As far as I know, nobody but Ayres and Nalebuff (and their followers) call this "lifecycle investing." They've appropriated a term that had a different meaning.
Lifecycle models in economics generally refer to models that look at how to optimize consumption and savings over a person's lifetime. The simple model without risk simply says you use borrowing and savings to smooth consumption across time (transfer $ from high income middle years to low income years early and late). If you add in risk, you'll get some version of the model that Ayres and Nalebuff are working from. I think it's a perfectly appropriate use of the term "lifecycle" in economics.
nisiprius wrote: Fri Mar 13, 2020 12:34 pm just as they've appropriated Paul Samuelson's name for a strategy he explicitly denounced
Paul Samuelson's offhand comments were wrong. His earlier work that Ayres and Nalebuff reference is correct. I think Samuelson made the comment without really reading the book, and he did investors a huge disservice by his careless remarks. I wrote about Samuelson's potential misinterpretation of Ayres Nalebuff in this thread:
Ben Mathew wrote: Thu Mar 07, 2019 12:50 pm
bobcat2 wrote: Thu Mar 07, 2019 9:36 am
Ben Mathew wrote: Tue Mar 05, 2019 10:37 pmI can understand the reasoning behind what Samuelson wrote in 1969. I cannot understand the reasoning behind what he said in 2008.
I think the reasoning of what he said in 2008 is straightforward. Here is the kernel of what Samuelson said in 2008.

Many analysts argue that when you average over many investment periods, so favorable are the long-run returns of stocks that while you are still young, you should borrow substantially to hold large positions in stocks and you should do so because some kind of “stochastic dominance” is supposed to justify it.

I think I have written 27 articles rebutting this idea. It smacks of what I call the “Milton Friedman fallacy.” When that sage was a TIAA trustee before me, he believed that investing for a large number of future periods did, by some law of large numbers, mandate becoming more risk tolerant. The Milton Friedman fallacy is a simple one. Also called the Kelly criterion, it leads to the conclusion that, in contrast to utility theory, one should always maximize the geometric mean. It is the same as the 1738 Daniel Bernoulli conjecture that if you have a duel with your brother-in-law and you are faced with a stationary probability process—stationary through time—going to the geometric mean is the way to win. Being second in investing, unlike being second in dueling, is good, however, and very few attain it.


What is it about the above you have trouble understanding?

BobK
Paul Samuelson seems to be under the impression that Ayres and Nalebuff are falling victim to the fallacy of time diversification, a subject that Paul Samuelson has written about--apparently 27 times! But Ayres and Nalebuff are not making that error.

The fallacy of time diversification is the idea that if you hold stocks for a longer period of time, the ups and downs cancel each other and you end up with low risk in the long run. This is not correct. The longer you hold stocks, the more bets you are taking. So there's a higher chance of very good and very bad outcomes--i.e. more risk. Think of every year in the stock market as an independent bet with a positive expected return--say a coin toss where if it's heads, you get $10, and if it's tails, you lose $5. The more bets you make, the more your expected payout, but the more widely distributed your payouts will be. The bets don't cancel out. It is true that the average payout per bet is subject to the law of large numbers and becomes more and more tightly distributed around the expected payout of .5*($10)+.2*(-5)=$2.50. But the payout that you are going to be walking away in the end is not the average payout. It's the total payout. And since you are adding more and more bets over time, your actual payout, which is the total payout, is getting more and more widely distributed--i.e. more and more risky. In the same way, with a portfolio invested in the stock market, the longer you stay in the market, the less variance there will be in your average log return. But you are not eating your average log return. You are eating the actual dollars in your portfolio. And the actual dollars in your portfolio will be getting more and more widely distributed the longer you stay in the market. So you are increasing both your expected return and your risk by staying longer in the market.

But Ayres and Nalebuff do not make this fallacious argument for time diversification. Their version of time diversification is entirely correct. In terms of the coin toss analogy, what they are saying is not that you should just take more coin tosses. What they are saying is that you should spread your bets over as many coin tosses as you can, taking a smaller bet on each coin toss. It's a subtle but crucial difference. They are saying that instead of betting $1 on 10 coin tosses, bet $.50 on 20 coin tosses. This is real diversification. Both betting strategies will yield an expected payout of $2.50 * 10 = .5 * $2.50 * 20 = $25. But the $.50 bet spread across 20 bets will yield a payout that is more tightly distributed around the expected payout of $25 than the $1 bets concentrated on 10 coin tosses. The law of large numbers does apply in this case. It is possible to diversify across time by spreading your bets across time instead of concentrating it all on a few years. Time diversification works when done properly--smaller bets over a longer period of time instead of bigger bets over a shorter period of time.

This error of not recognizing the possibility of time diversification by reducing the bet and spreading it out over time deserves a name. I'd like to call it the "fallacy of the fallacy of time diversification."
User avatar
Steve Reading
Posts: 2668
Joined: Fri Nov 16, 2018 10:20 pm

Re: Lifecycle Investing - Still the Right Strategy

Post by Steve Reading »

Ben Mathew wrote: Fri Mar 13, 2020 12:33 pm With the sharp market declines we experienced recently, young investors who invested very aggressively in the stock market (100% or more) may be having a hard time and may be questioning the wisdom of their aggressive stance. It's normal to be scared, but do try to keep it in perspective. The arguments for behind being aggressive when young is compelling.

Lifecycle investing tells us that you will minimize lifetime risk for a given return by spreading stock risk evenly across time. Since you have very little savings in the early years, that entails putting all of it (and maybe then some, via leveraging) into stocks. In return, you get to reduce your stock allocation when older. Basically, you end up with a steeper glidepath than what is typical. Yes, this means that if stocks fall 25% when you're young, you'll lose a large portion of your savings. The point to remember is that your current savings is not a lot of money. Most of your savings is in the future. Because of that, a 100% loss when you are 30 is smaller than a 50% loss when you are 60.

While this is true, the loss of a large portion of your savings when you are young can be very difficult to handle psychologically. So my advice for most people would be to stop at 100% stocks and not go push into leveraged territory, unless you are really understand the mechanics as well as your own psychology and can be really rational about it. I know that when I was younger, the $200K in savings we had cobbled together over a few years felt like a lot of money even though it wasn't very much from a lifecycle perspective. My perspective was skewed. I was invested 100% in stocks, and I was comfortable with that. Leverage would have been harder.

Whether you are 100% stocks or in leveraged territory, remind yourself of the principles behind lifecycle investing and why it's optimal from a risk management perspective to be aggressive when young. Paradoxical as it seems, you are taking on less lifetime risk. The plan is good. Stick to it. There will be many more ups and downs to come. Averaging over it all is the right way to go.

Good luck!
Lifecycle Investing provides temporal diversification. Just like asset diversification, that means some times it’s detrimental. If we knew 2020 would have a drop, or Ex-US would underperform, we’d obviously go to cash in 2020 or stick to only domestic stock respectively. But then again, 2020 could’ve been an amazing year or Ex-US might crush domestic.

Since we don’t know, the rational choice is to be exposed to both Ex-US and domestic; to both 2020 and 2060 stock returns. The optimal choice IMO is to diversify.

I am very content with the strategy and need no convincing. Over the long term, I hope the benefits of diversification, both asset and temporal, eventually lead to higher returns and lower risk over my investing lifetime.

Either way, thank you for the kind words of encouragement. ^_^
"... so high a present discounted value of wealth, it is only prudent for him to put more into common stocks compared to his present tangible wealth, borrowing if necessary" - Paul Samuelson
rascott
Posts: 2407
Joined: Wed Apr 15, 2015 10:53 am

Re: Lifecycle Investing - Still the Right Strategy

Post by rascott »

I find the idea very logical, and wish I had known/ thought of it when I was a younger. I'm at a point now where 100% equity (with a small cap tilt) is plenty enough for me.
TheoLeo
Posts: 281
Joined: Mon Apr 02, 2018 11:39 am

Re: Lifecycle Investing - Still the Right Strategy

Post by TheoLeo »

So lifecycle investing basically is "age in bonds" on steroids? A glide path using high or very high stock allocation when starting out.
User avatar
Steve Reading
Posts: 2668
Joined: Fri Nov 16, 2018 10:20 pm

Re: Lifecycle Investing - Still the Right Strategy

Post by Steve Reading »

TheoLeo wrote: Fri Mar 13, 2020 3:41 pm So lifecycle investing basically is "age in bonds" on steroids? A glide path using high or very high stock allocation when starting out.
People always conveniently forget the second part; a much lower stock allocation when retired.

So age in bonds starting with higher allocations and ending with lower allocations so the number of real dollars in the market is much more even throughout decades, which “spreads” your risks by diversifying (getting exposure) to as many different decades as possible
"... so high a present discounted value of wealth, it is only prudent for him to put more into common stocks compared to his present tangible wealth, borrowing if necessary" - Paul Samuelson
TheoLeo
Posts: 281
Joined: Mon Apr 02, 2018 11:39 am

Re: Lifecycle Investing - Still the Right Strategy

Post by TheoLeo »

305pelusa wrote: Fri Mar 13, 2020 4:08 pm
TheoLeo wrote: Fri Mar 13, 2020 3:41 pm So lifecycle investing basically is "age in bonds" on steroids? A glide path using high or very high stock allocation when starting out.
People always conveniently forget the second part; a much lower stock allocation when retired.

So age in bonds starting with higher allocations and ending with lower allocations so the number of real dollars in the market is much more even throughout decades, which “spreads” your risks by diversifying (getting exposure) to as many different decades as possible
Would it be correct to say, that the steeper the glide path, the better the risk/reward propabilities are for a specified time frame in the future and the worse the risk /reward propabilities become for time periods before and after that specific time frame? A fixed allocation then has better risk/reward propabilities for the time before and after this specific time frame the steep glide path was designed for. If that is correct, then how exactly do you calculate the steepness to fit the time frame you need the cash flow? And how sensitive is this steep glide path if I need my cash flow earlier or later than calculated compared to a fixed allocation. Maybe a mathematically talented member can somehow quantify that?
Last edited by TheoLeo on Fri Mar 13, 2020 4:48 pm, edited 1 time in total.
User avatar
Steve Reading
Posts: 2668
Joined: Fri Nov 16, 2018 10:20 pm

Re: Lifecycle Investing - Still the Right Strategy

Post by Steve Reading »

TheoLeo wrote: Fri Mar 13, 2020 4:38 pm
305pelusa wrote: Fri Mar 13, 2020 4:08 pm
TheoLeo wrote: Fri Mar 13, 2020 3:41 pm So lifecycle investing basically is "age in bonds" on steroids? A glide path using high or very high stock allocation when starting out.
People always conveniently forget the second part; a much lower stock allocation when retired.

So age in bonds starting with higher allocations and ending with lower allocations so the number of real dollars in the market is much more even throughout decades, which “spreads” your risks by diversifying (getting exposure) to as many different decades as possible
Would it be correct to say, that the steeper the glide path, the better the risk/reward propabilities are for a specified time frame in the future and the worse the risk /reward propabilities become for time periods before and after that specific time frame? A fixed allocation then has better risk/reward propabilities for the time before and after this specific time frame the steep glide path was designed for. If that is correct, then how exactly do you calculate the steepness to fit the time frame you need the cash flow? And how sensitive is this steep glide path if I need my cash flow earlier or later than calculated compared to a fixed allocation. Maybe some mathematically talented members can somehow quantify that?
You want to have a somewhat constant real dollar exposure to equities over time. If you’re young and accumulating, you don’t have much in savings. So you need to invest more of your proportional financial wealth to put the same amount of real dollars in the market today than you will when you’re retired.

This steepness is calculated by adding your financial wealth to your human capital (the present value of all your future savings contributions) and then investing a proportion of THAT much bigger figure in stocks, today. That would require leverage when young.

I recommend people set up these glide paths conservatively. Meaning, assume a certain amount of your paycheck will be consumed even if you currently don’t see how. And only discount whatever is left out of that. That means that my allocation will only be optimal if some event forces me to spend more than I anticipated. So far that hasn’t been the case so I technically haven’t diversified temporally perfectly. That’s good enough for me.
"... so high a present discounted value of wealth, it is only prudent for him to put more into common stocks compared to his present tangible wealth, borrowing if necessary" - Paul Samuelson
Lee_WSP
Posts: 3776
Joined: Fri Apr 19, 2019 5:15 pm
Location: Arizona

Re: Lifecycle Investing - Still the Right Strategy

Post by Lee_WSP »

AlohaBill wrote: Fri Mar 13, 2020 4:45 pm I disagree. These anti bogleheads after running through a long tread want to start it up again. They just want to change boglehead philosophy. They got nothing better to do than troll here.
It's not really up to you whether or not the forum should or should not just be an echo chamber and exclude opposing view points. It's not up to me either. But as it stands, the rules are posted and opposing viewpoints are allowed.

Lifecycle investing is also not against the Bogle Philosophy per se. What tenet does it violate in your opinion?
User avatar
Uncorrelated
Posts: 1232
Joined: Sun Oct 13, 2019 3:16 pm

Re: Lifecycle Investing - Still the Right Strategy

Post by Uncorrelated »

TheoLeo wrote: Fri Mar 13, 2020 4:38 pm
305pelusa wrote: Fri Mar 13, 2020 4:08 pm
TheoLeo wrote: Fri Mar 13, 2020 3:41 pm So lifecycle investing basically is "age in bonds" on steroids? A glide path using high or very high stock allocation when starting out.
People always conveniently forget the second part; a much lower stock allocation when retired.

So age in bonds starting with higher allocations and ending with lower allocations so the number of real dollars in the market is much more even throughout decades, which “spreads” your risks by diversifying (getting exposure) to as many different decades as possible
Would it be correct to say, that the steeper the glide path, the better the risk/reward propabilities are for a specified time frame in the future and the worse the risk /reward propabilities become for time periods before and after that specific time frame? A fixed allocation then has better risk/reward propabilities for the time before and after this specific time frame the steep glide path was designed for. If that is correct, then how exactly do you calculate the steepness to fit the time frame you need the cash flow? And how sensitive is this steep glide path if I need my cash flow earlier or later than calculated compared to a fixed allocation. Maybe some mathematically talented members can somehow quantify that?
There is a trade off between the steepness of the glidepath and the end balance. If there are no additional deposits (i.e. all wealth is given up front), then the optimal allocation is a constant allocation. Optimal in this context means that the expected return is X, and that all possible glidepaths with the same expected return X have more risk (measured in variance) than the constant allocation.

If there are deposits or withdrawals from the portfolio, the optimal allocation is no longer a constant, but it's close. If you have $50k liquid net worth and $100k guaranteed future savings in the form of income, then your optimal allocation is a constant proportion of the liquid net worth + human capital. Again, optimal means that if the glidepath has an expected return of X, there is no other glidepath that has the same expected return but lower risk.

It's really a special case of merton's portfolio problem. In practice the solution is significantly more complex because it's a mult-asset problem that has no closed form soluton (to my knowledge). Even the Ayres and Nalebuff solution is a pretty crude approximation that only works in specific cases (single asset, constant leverage cost). However, I have an approximate solution that supports arbitrary combinations of assets: viewtopic.php?f=10&t=274390&start=800#p5036988. It turns out this is a very complex problem.
User avatar
Steve Reading
Posts: 2668
Joined: Fri Nov 16, 2018 10:20 pm

Re: Lifecycle Investing - Still the Right Strategy

Post by Steve Reading »

AlohaBill wrote: Fri Mar 13, 2020 4:45 pm I disagree. These anti bogleheads after running through a long tread want to start it up again. They just want to change boglehead philosophy. They got nothing better to do than troll here.
No trolling; this strategy is quite simply a free lunch. I’ve opened threads here and various posters have helped me extensively, Ben (OP) being one of the main ones.

As long as this community is so helpful in implementing superior strategies, I’d like to stick around and listen to them.

If it’s not your cup of tea, can I suggest you leave the thread? This will let those who actually care learn the approach better. If you think the thread violates standards in this forum, please flag it to the Mods and I will gladly leave the forum if that’s what they decide. If they’re fine with it, I’d love to keep discussing it and maybe helping others. Ideally without posters telling me how this doesn’t belong in this website :)
"... so high a present discounted value of wealth, it is only prudent for him to put more into common stocks compared to his present tangible wealth, borrowing if necessary" - Paul Samuelson
YearTrader
Posts: 105
Joined: Thu Jan 02, 2020 6:46 am

Re: Lifecycle Investing - Still the Right Strategy

Post by YearTrader »

I found it hard to estimate my future cash flow / human capital and risk tolerance / utility function. Also AFAIK I'd need to estimate the long term expected risk and return of asset classes in my investing universe.

How sensitive is this strategy to these input parameters? I think it's still a valid strategy as long as the overall variance of long term investment return can be reduced.
gone_forever
Posts: 4
Joined: Mon Mar 09, 2020 2:02 pm

Re: Lifecycle Investing - Still the Right Strategy

Post by gone_forever »

Makes total mathematical sense -- I've not been investing long and when I think about whether or not I need to take any action to stop loss; I just remind myself that I have 20-30 years of contributions left. How would I feel when I only had 10? I've been learning what I can about this as fast as I can; excellent timing with the market being down.

What are the options for doing this in a retirement account (IRA?) How sub-optimal is just using a LETF like UPRO? I know 2:1 is the ideal, but it seems like 4 phases (3:1 leverage to 2:1 leverage being an intermediate stage) makes just as much sense as 3.
Caduceus
Posts: 2810
Joined: Mon Sep 17, 2012 1:47 am

Re: Lifecycle Investing - Still the Right Strategy

Post by Caduceus »

Ben Mathew wrote: Fri Mar 13, 2020 12:33 pm
Lifecycle investing tells us that you will minimize lifetime risk for a given return by spreading stock risk evenly across time. Since you have very little savings in the early years, that entails putting all of it (and maybe then some, via leveraging) into stocks. In return, you get to reduce your stock allocation when older. Basically, you end up with a steeper glidepath than what is typical. Yes, this means that if stocks fall 25% when you're young, you'll lose a large portion of your savings. The point to remember is that your current savings is not a lot of money. Most of your savings is in the future. Because of that, a 100% loss when you are 30 is smaller than a 50% loss when you are 60.
This sounds like the sort of thing cooked up by academics who are better at publishing theories than actually getting rich. But I'm quite curious about the mechanics.

Is there a valuation component to this strategy? I don't see why time diversification would be such a great thing if you're leveraging up as a young investor when the stock market is richly priced and expensive, and then taking proportionately less leverage than a traditional investor during some later period when the markets may be much more cheaply priced. I'm pretty sure if you were to test this strategy by starting out young investors during periods where markets were trading at multiples of 30 CAPE versus young investors who leveraged up during periods where markets were trading at multiples of 15-20 CAPE that the results would be very different. So, my question is: is there a valuation component to this strategy. Or is it just: you're 25 years old, go lever up. I hope it's not the latter, because that just sounds stupid.

IIRC, there was an infamous thread where someone initiated this strategy for a period of time where markets were at all time highs. It didn't end well. The impact to lifetime total wealth was significant.

What's the proposed form of levering? Futures?
User avatar
Topic Author
Ben Mathew
Posts: 1409
Joined: Tue Mar 13, 2018 11:41 am
Location: Seattle
Contact:

Re: Lifecycle Investing - Still the Right Strategy

Post by Ben Mathew »

305pelusa wrote: Fri Mar 13, 2020 2:23 pm Lifecycle Investing provides temporal diversification. Just like asset diversification, that means some times it’s detrimental. If we knew 2020 would have a drop, or Ex-US would underperform, we’d obviously go to cash in 2020 or stick to only domestic stock respectively. But then again, 2020 could’ve been an amazing year or Ex-US might crush domestic.

Since we don’t know, the rational choice is to be exposed to both Ex-US and domestic; to both 2020 and 2060 stock returns. The optimal choice IMO is to diversify.

I am very content with the strategy and need no convincing. Over the long term, I hope the benefits of diversification, both asset and temporal, eventually lead to higher returns and lower risk over my investing lifetime.

Either way, thank you for the kind words of encouragement. ^_^
:beer

I sometimes see comments that ask where are the "100% stocks" and "leveraged" posts when the market is down. I felt it's useful to let people know that the lifecycle strategy is an approach designed to work well through both the ups and the downs.
User avatar
Topic Author
Ben Mathew
Posts: 1409
Joined: Tue Mar 13, 2018 11:41 am
Location: Seattle
Contact:

Re: Lifecycle Investing - Still the Right Strategy

Post by Ben Mathew »

rascott wrote: Fri Mar 13, 2020 2:42 pm I find the idea very logical, and wish I had known/ thought of it when I was a younger. I'm at a point now where 100% equity (with a small cap tilt) is plenty enough for me.
That just about describes where I am at too. Except I tilt to value as well.
abracadabra11
Posts: 197
Joined: Sat May 01, 2010 2:09 pm

Re: Lifecycle Investing - Still the Right Strategy

Post by abracadabra11 »

AlohaBill wrote: Fri Mar 13, 2020 4:12 pm Meh! Doesn’t belong in this forum.
There are plenty here that would disagree. Lifecycle investing, though likely not appropriate for many investors (because of the various prerequisites laid out in the book), is worth discussing and considering. The ceiling for most glidepaths has been 100% stock, but there's no reason to think that higher levels of allocation don't provide sufficient premium for the risk involved. If anything, the idea behind lifecycle investing pushes us to ask - what's the ideal ceiling for stock allocation.

I'll add that I've read the book and I don't find it particularly well written (and I'm not a fan of the few 'advertisements' in the book).
User avatar
vineviz
Posts: 8571
Joined: Tue May 15, 2018 1:55 pm

Re: Lifecycle Investing - Still the Right Strategy

Post by vineviz »

YearTrader wrote: Fri Mar 13, 2020 5:13 pm I found it hard to estimate my future cash flow / human capital and risk tolerance / utility function. Also AFAIK I'd need to estimate the long term expected risk and return of asset classes in my investing universe.

How sensitive is this strategy to these input parameters? I think it's still a valid strategy as long as the overall variance of long term investment return can be reduced.
I think you'd be surprised at how easy it is get a reasonably good approximation of these things, especially using the downloadable spreadsheets on the Ayres and Nalebuff website: http://www.lifecycleinvesting.net/resources.html

The PV of your future retirement savings, for example, requires just one piece of data (your current income) and two assumptions (your future savings rate and a discount rate).
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch
User avatar
bluquark
Posts: 1186
Joined: Mon Oct 22, 2018 2:30 pm

Re: Lifecycle Investing - Still the Right Strategy

Post by bluquark »

I tend to agree with Bernstein that although it sounds good in theory, the problem is that it ignores non-market risks.

The first risk is the possibility of an ill-timed episode of unemployment or disability requiring much earlier drawdowns than planned. Especially a problem for those with dependents.

The second one is the possibility of overestimating one's risk tolerance and polarizing to a conservative strategy after a crash. Young investors are likely to have an especially hard time knowing their true risk tolerance. And because it's a niche strategy that goes against conventional wisdom and indeed will sound actively reckless to your friends, there will be more opportunities to question it without much community support.

And I'd caution against feeling that because you are still confident in the midst of this crash, your risk tolerance is now known. During the Great Depression and Japan's lost decades the market just kept slowly declining year after year, causing almost everyone to eventually come to believe that the country had entered a permanent zero-stock-return era. If you sell to 80% bonds at the market bottom, you have ended up with the worst possible strategy.

There are a Spock-like few who will be able to meet this psychological test and also have no dependents and a really stable job (e.g. federal government). But this is such a tiny population that this strategy doesn't seem like something to advocate except as an interesting theoretical exercise.
70/30 portfolio | Equity: global market weight | Bonds: 20% long-term munis - 10% LEMB
User avatar
vineviz
Posts: 8571
Joined: Tue May 15, 2018 1:55 pm

Re: Lifecycle Investing - Still the Right Strategy

Post by vineviz »

Caduceus wrote: Fri Mar 13, 2020 5:16 pm This sounds like the sort of thing cooked up by academics who are better at publishing theories than actually getting rich. But I'm quite curious about the mechanics.

Is there a valuation component to this strategy? I don't see why time diversification would be such a great thing if you're leveraging up as a young investor when the stock market is richly priced and expensive, and then taking proportionately less leverage than a traditional investor during some later period when the markets may be much more cheaply priced. I'm pretty sure if you were to test this strategy by starting out young investors during periods where markets were trading at multiples of 30 CAPE versus young investors who leveraged up during periods where markets were trading at multiples of 15-20 CAPE that the results would be very different. So, my question is: is there a valuation component to this strategy. Or is it just: you're 25 years old, go lever up. I hope it's not the latter, because that just sounds stupid.
If you're curious I definitely suggest seeking out the book. It's old enough that used copies are relatively cheap.

One component of the approach is to rely on something the authors call the "Samuelson Share" to derive your current optimal equity exposure. Two key inputs to the Samuelson Share are the current levels of CAPE and the VIX index, which has the effect of reducing equity exposure when stocks are expensive and/or market volatility is very high.

You could quibble with the precise approach they use, but directionally it's pretty sound.
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch
User avatar
LadyGeek
Site Admin
Posts: 68701
Joined: Sat Dec 20, 2008 5:34 pm
Location: Philadelphia
Contact:

Re: Lifecycle Investing - Still the Right Strategy

Post by LadyGeek »

I removed an off-topic post and several replies. As a reminder, see: General Etiquette
We expect this forum to be a place where people can feel comfortable asking questions and where debates and discussions are conducted in civil tones.

At all times we must conduct ourselves in a respectful manner to other posters. Attacks on individuals, insults, name calling, trolling, baiting or other attempts to sow dissension are not acceptable.
Wiki To some, the glass is half full. To others, the glass is half empty. To an engineer, it's twice the size it needs to be.
Caduceus
Posts: 2810
Joined: Mon Sep 17, 2012 1:47 am

Re: Lifecycle Investing - Still the Right Strategy

Post by Caduceus »

vineviz wrote: Fri Mar 13, 2020 7:40 pm
One component of the approach is to rely on something the authors call the "Samuelson Share" to derive your current optimal equity exposure. Two key inputs to the Samuelson Share are the current levels of CAPE and the VIX index, which has the effect of reducing equity exposure when stocks are expensive and/or market volatility is very high.

You could quibble with the precise approach they use, but directionally it's pretty sound.
Why would you reduce your equity exposure when market volatility is high? Historically, VIX would be high at times of huge downside volatility, like in 2008-2009 - those were good times to be leveraging up your equity exposure, not down. Historically the folks who've bought when VIX has been high have done better than those who did not.
User avatar
LadyGeek
Site Admin
Posts: 68701
Joined: Sat Dec 20, 2008 5:34 pm
Location: Philadelphia
Contact:

Re: Lifecycle Investing - Still the Right Strategy

Post by LadyGeek »

I am very confused. Can someone explain the difference between "lifecycle investing" as discussed here and Life-cycle finance as discussed in the wiki?

If they have nothing to do with each other, that's fine. I'm just having some difficulties to understand what is being spread over the investor's timeline.
Wiki To some, the glass is half full. To others, the glass is half empty. To an engineer, it's twice the size it needs to be.
User avatar
vineviz
Posts: 8571
Joined: Tue May 15, 2018 1:55 pm

Re: Lifecycle Investing - Still the Right Strategy

Post by vineviz »

Caduceus wrote: Fri Mar 13, 2020 7:53 pm
vineviz wrote: Fri Mar 13, 2020 7:40 pm
One component of the approach is to rely on something the authors call the "Samuelson Share" to derive your current optimal equity exposure. Two key inputs to the Samuelson Share are the current levels of CAPE and the VIX index, which has the effect of reducing equity exposure when stocks are expensive and/or market volatility is very high.

You could quibble with the precise approach they use, but directionally it's pretty sound.
Why would you reduce your equity exposure when market volatility is high? Historically, VIX would be high at times of huge downside volatility, like in 2008-2009 - those were good times to be leveraging up your equity exposure, not down. Historically the folks who've bought when VIX has been high have done better than those who did not.
The primary goal of this strategy is risk management, not return maximization or market timing.
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch
Caduceus
Posts: 2810
Joined: Mon Sep 17, 2012 1:47 am

Re: Lifecycle Investing - Still the Right Strategy

Post by Caduceus »

vineviz wrote: Fri Mar 13, 2020 7:57 pm
Caduceus wrote: Fri Mar 13, 2020 7:53 pm
vineviz wrote: Fri Mar 13, 2020 7:40 pm
One component of the approach is to rely on something the authors call the "Samuelson Share" to derive your current optimal equity exposure. Two key inputs to the Samuelson Share are the current levels of CAPE and the VIX index, which has the effect of reducing equity exposure when stocks are expensive and/or market volatility is very high.

You could quibble with the precise approach they use, but directionally it's pretty sound.
Why would you reduce your equity exposure when market volatility is high? Historically, VIX would be high at times of huge downside volatility, like in 2008-2009 - those were good times to be leveraging up your equity exposure, not down. Historically the folks who've bought when VIX has been high have done better than those who did not.
The primary goal of this strategy is risk management, not return maximization or market timing.

OK. To be honest, I was genuinely curious at first, and now I'm not. Using volatility as a proxy for risk when this is a strategy that purports to be a long-term strategy for temporal diversification makes zero sense. The strategy has a very strange idea of what risk entails - its version of "managing risk" entails buying shares using levered funds and buying only when volatility is low (when fundamental valuations are likely higher), both things that increase rather than reduce risk (real risk, which is the permanent impairment of capital, and not short-term volatility).

Another serious question: Did the authors of this paper/strategy employ this strategy themselves - how wealthy are they? IIRC, the only poster I know on this forum who employed some version of this on an infamous thread flamed out spectacularly, costing an immense amount of total lifetime wealth.
User avatar
Steve Reading
Posts: 2668
Joined: Fri Nov 16, 2018 10:20 pm

Re: Lifecycle Investing - Still the Right Strategy

Post by Steve Reading »

Caduceus wrote: Fri Mar 13, 2020 5:16 pm is there a valuation component to this strategy. Or is it just: you're 25 years old, go lever up. I hope it's not the latter, because that just sounds stupid.
The latter describes me perfectly. I'm 25 and I levered up regardless of market valuations. I saw no reason to delay temporal diversification just to try to avoid a market peak. I don't believe valuations can truly predict markets any ways.
It sounds counterintuitive but it's well-supported by the math, theory and historical results.
Caduceus wrote: Fri Mar 13, 2020 5:16 pm What's the proposed form of levering? Futures?
You could use options, futures, margin or even Leveraged ETFs. I strongly prefer the first three for reasons I've detailed many times in my thread:
viewtopic.php?f=10&t=274390
Caduceus wrote: Fri Mar 13, 2020 7:53 pm
vineviz wrote: Fri Mar 13, 2020 7:40 pm
One component of the approach is to rely on something the authors call the "Samuelson Share" to derive your current optimal equity exposure. Two key inputs to the Samuelson Share are the current levels of CAPE and the VIX index, which has the effect of reducing equity exposure when stocks are expensive and/or market volatility is very high.

You could quibble with the precise approach they use, but directionally it's pretty sound.
Why would you reduce your equity exposure when market volatility is high? Historically, VIX would be high at times of huge downside volatility, like in 2008-2009 - those were good times to be leveraging up your equity exposure, not down. Historically the folks who've bought when VIX has been high have done better than those who did not.
Theoretically, at a higher volatility (and the same level of expected return), then stocks are less attractive. Which makes sense; if two portfolios had the same expected return but one was less volatile, the less volatile is preferable.

However, a lot of Lifecycle Investing theory relies on normal distributions, random walks and constant volatility. I don't think the same equations apply if you're trying to constantly change your allocation due to changes in volatility; that was never in the initial assumptions of Merton.

So I personally use long-term volatility (about 19% Stand. Dev for stocks). During times of much higher volatility, I simply remain the course. That's my way of dealing with the fact that the real world does not behave like the Merton assumptions.
"... so high a present discounted value of wealth, it is only prudent for him to put more into common stocks compared to his present tangible wealth, borrowing if necessary" - Paul Samuelson
User avatar
Steve Reading
Posts: 2668
Joined: Fri Nov 16, 2018 10:20 pm

Re: Lifecycle Investing - Still the Right Strategy

Post by Steve Reading »

Caduceus wrote: Fri Mar 13, 2020 8:07 pm The strategy has a very strange idea of what risk entails - its version of "managing risk" entails buying shares using levered funds and buying only when volatility is low (when fundamental valuations are likely higher), both things that increase rather than reduce risk (real risk, which is the permanent impairment of capital, and not short-term volatility).
I personally agree. Over the short term (days and weeks), stocks don't behave like normal distributions. Standard deviation is not an appropriate measure of risk (it misses the kurtosis, case in point this week) and hence the Merton allocation breaks down.
My preferred method is to just use long-term volatility (where stocks are somewhat normally distributed due to the Central Limit Theorem), apply Lifecycle Investing allocation to set my long-term strategic allocation with that long-term measure of risk, and just "ignore" the short term fluctuations. Others will do something else I guess 0_o

I don't use LETFs for similar related reasons.
Caduceus wrote: Fri Mar 13, 2020 8:07 pm Another serious question: Did the authors of this paper/strategy employ this strategy themselves - how wealthy are they? IIRC, the only poster I know on this forum who employed some version of this on an infamous thread flamed out spectacularly, costing an immense amount of total lifetime wealth.
My understanding is that they're currently employing it but they're older so they're not using 2:1 leverage or anything.

Market Timer blew up with this strategy for various contraindications, including the fact that he wasn't saving yet (he was still a student), he started with quite a bit of leverage (which forced him to sell as the market dropped) and actually made some pretty undiversified bets (on various financials).

I'm doing it a little differently. I currently have an income and have been contributing to the strategy for close to a year. This is really key. I also started with much less leverage such that when markets did drop (like now) I can maintain the same exposure and participate on the upside. LETFs would not let you do that (they effectively lose exposure after market drops) so another reason I stay away. And I'm betting only on broad market index funds.
"... so high a present discounted value of wealth, it is only prudent for him to put more into common stocks compared to his present tangible wealth, borrowing if necessary" - Paul Samuelson
User avatar
Steve Reading
Posts: 2668
Joined: Fri Nov 16, 2018 10:20 pm

Re: Lifecycle Investing - Still the Right Strategy

Post by Steve Reading »

LadyGeek wrote: Fri Mar 13, 2020 7:55 pm I am very confused. Can someone explain the difference between "lifecycle investing" as discussed here and Life-cycle finance as discussed in the wiki?

If they have nothing to do with each other, that's fine. I'm just having some difficulties to understand what is being spread over the investor's timeline.
Lifecycle Investing here:

What's less risky, investing 1M real dollars in 2020 and 1M real dollars in 2050, or just 2M in 2050? I argue the former is much less risky. Instead of depending purely on the returns of 2050, you're depending half and half on two different years that have no effect on each other. This is diversification from a time perspective.

As accumulators, we begin with few assets (little in the market) and end up with lots in the market. That's like the latter scenario above, undiversified across time. Even if you went 100% in stocks when young, the number of dollars in the market pales in comparison to the number of dollars in the market in your 50/50 retirement. So the strategy uses leverage early on to try to achieve a more even exposure to the market.

I read the wiki and that looks like something different entirely. It seems like a strategy where instead of just spending a lot when you make a lot (and little when you make little) you borrow and/or save as needed to smooth out consumption.
"... so high a present discounted value of wealth, it is only prudent for him to put more into common stocks compared to his present tangible wealth, borrowing if necessary" - Paul Samuelson
Caduceus
Posts: 2810
Joined: Mon Sep 17, 2012 1:47 am

Re: Lifecycle Investing - Still the Right Strategy

Post by Caduceus »

305pelusa wrote: Fri Mar 13, 2020 8:21 pm
Caduceus wrote: Fri Mar 13, 2020 8:07 pm The strategy has a very strange idea of what risk entails - its version of "managing risk" entails buying shares using levered funds and buying only when volatility is low (when fundamental valuations are likely higher), both things that increase rather than reduce risk (real risk, which is the permanent impairment of capital, and not short-term volatility).
I personally agree. Over the short term (days and weeks), stocks don't behave like normal distributions. Standard deviation is not an appropriate measure of risk (it misses the kurtosis, case in point this week) and hence the Merton allocation breaks down.
My preferred method is to just use long-term volatility (where stocks are somewhat normally distributed due to the Central Limit Theorem), apply Lifecycle Investing allocation to set my long-term strategic allocation with that long-term measure of risk, and just "ignore" the short term fluctuations. Others will do something else I guess 0_o
A couple of points, in no particular order.

1) Standard deviation is not an appropriate measure of risk not because it misses kurtosis, but because it has absolutely nothing to do with the inherent valuation of stocks. A stock that fluctuates a great deal is neither more or less risky than a stock that does not fluctuate. You must think about what stocks represent. Stocks represent fractional interests in real businesses. They have real assets, real income, real liabilities. Their true value is connected to the fundamental business performance, not whether the market constantly changes its mind on what the business is worth, which often have very little to do with assessments of the business value.

Let me give you a real world example to explain why risk-as-volatility is an inherently ridiculous idea. During the tech bubble, there were companies with stock prices that hadn't moved much in a decade - highly stable, boring, mildly profitable companies that did nothing but pay out dividends. When people got crazy with all things technological, some CEOs had the bright idea to change the company name to something obviously technological. So, MIS International became Cosmoz.com. Guess what? As expected, its volatility suddenly exploded. Same company. Same business. Same profits. Same everything. Only thing that changed? The name. And, consequently, the volatility. So you expect me to believe that the company is riskier than it was before because its volatility has increased? In financial markets, stocks can experience volatility for a variety of reasons that have absolutely nothing to do with their inherent business value. Why should the frequency with which the market changes its mind on the value of a given stock or index have any bearing on its future value? And it doesn't. Finance academics were once enamored with high-beta stocks. When that didn't work, they went to low-beta stocks. Guess what - there's a reason you don't hear about beta very much any more as a sales pitch.

2) I would be very, very careful about using mathematical tools to model financial reality. The stock market is not a physical system and it does not - and will never - obey the rules of the model you set up to describe it. Historical and implied volatilities will change because the historical conditions that have underpinned those datasets have changed, so your estimates of volatility are useless. You can use math to describe the past - those relationships are static and won't change. But it is a fool's errand to assume that it can and will be applicable to the future.

I encourage you to read two great books in finance. The first is Roger Lowenstein's When Genius Failed: The Rise and Fall of Long Term Capital Management. You will love it. It explores how and why a fund started by Nobel Prize winning mathematicians/academics and some of the brightest minds on Wall Street ended in one of the worst debacles in financial history. The crash of their fund brought down not only themselves but the financial markets. The second book is Andrew Ross Sorkin's Too Big To Fail. It tells the story of how the very same mathematicians who created elegant and complicated Value At Risk models did not understand the models themselves, and could not explain why they weren't describing what was actually happening with the banks' portfolios. If you're the type of person who's good at math, it's going to be very seductive to think you're bringing the right tools to the job of investing. You're not.

We've had long enough time periods to know that stocks selected with the capital asset pricing model in mind haven't done well.

Ultimately, the proof is in the pudding. Look at the people who've actually made the most money, consistently, over long periods. There are the index investors, like the Bogleheads, and then the value investors, like Warren Buffett and Seth Klarman and Lou Simpson, Walter Schloss, etc. That's a fact. Other things have come and gone - temporary fads. No one's talking about quant funds like AQR these days on the board.
User avatar
LadyGeek
Site Admin
Posts: 68701
Joined: Sat Dec 20, 2008 5:34 pm
Location: Philadelphia
Contact:

Re: Lifecycle Investing - Still the Right Strategy

Post by LadyGeek »

305pelusa wrote: Fri Mar 13, 2020 8:30 pm
LadyGeek wrote: Fri Mar 13, 2020 7:55 pm I am very confused. Can someone explain the difference between "lifecycle investing" as discussed here and Life-cycle finance as discussed in the wiki?

If they have nothing to do with each other, that's fine. I'm just having some difficulties to understand what is being spread over the investor's timeline.
Lifecycle Investing here:

What's less risky, investing 1M real dollars in 2020 and 1M real dollars in 2050, or just 2M in 2050? I argue the former is much less risky. Instead of depending purely on the returns of 2050, you're depending half and half on two different years that have no effect on each other. This is diversification from a time perspective.

As accumulators, we begin with few assets (little in the market) and end up with lots in the market. That's like the latter scenario above, undiversified across time. Even if you went 100% in stocks when young, the number of dollars in the market pales in comparison to the number of dollars in the market in your 50/50 retirement. So the strategy uses leverage early on to try to achieve a more even exposure to the market.

I read the wiki and that looks like something different entirely. It seems like a strategy where instead of just spending a lot when you make a lot (and little when you make little) you borrow and/or save as needed to smooth out consumption.
That's a good explanation, thank you.
Wiki To some, the glass is half full. To others, the glass is half empty. To an engineer, it's twice the size it needs to be.
User avatar
vineviz
Posts: 8571
Joined: Tue May 15, 2018 1:55 pm

Re: Lifecycle Investing - Still the Right Strategy

Post by vineviz »

Caduceus wrote: Fri Mar 13, 2020 8:07 pm OK. To be honest, I was genuinely curious at first, and now I'm not. Using volatility as a proxy for risk when this is a strategy that purports to be a long-term strategy for temporal diversification makes zero sense. The strategy has a very strange idea of what risk entails - its version of "managing risk" entails buying shares using levered funds and buying only when volatility is low (when fundamental valuations are likely higher), both things that increase rather than reduce risk (real risk, which is the permanent impairment of capital, and not short-term volatility).
Reducing equity exposure when volatility is high is actually an entirely reasonable strategy, and not just for behavioral reasons. Despite common misconceptions, the benefits of diversification (temporal or otherwise) don't flow from spreading capital across different assets but rather from spreading risk across different sources: the former is merely a means to achieve the latter. Because of volatility's well-documented tendency to cluster, a strategy that maintains a relatively constant amount of volatility tends to be more efficient than one that maintains a relatively constant allocation of capital.

This is why target volatility strategies generally prove superior to fixed allocations with similar long-term average variance.
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch
Silence Dogood
Posts: 1447
Joined: Tue Feb 01, 2011 9:22 pm

Re: Lifecycle Investing - Still the Right Strategy

Post by Silence Dogood »

Ben Mathew wrote: Fri Mar 13, 2020 5:54 pm I sometimes see comments that ask where are the "100% stocks" and "leveraged" posts when the market is down. I felt it's useful to let people know that the lifecycle strategy is an approach designed to work well through both the ups and the downs.
Specifically regarding leverage, wasn't that the strategy attempted by market timer?

Or are you advocating for something different?

Recently I read a post advocating for a fixed asset allocation for life. I find it interesting how different these two strategies are.
Caduceus
Posts: 2810
Joined: Mon Sep 17, 2012 1:47 am

Re: Lifecycle Investing - Still the Right Strategy

Post by Caduceus »

vineviz wrote: Fri Mar 13, 2020 9:27 pm
Caduceus wrote: Fri Mar 13, 2020 8:07 pm OK. To be honest, I was genuinely curious at first, and now I'm not. Using volatility as a proxy for risk when this is a strategy that purports to be a long-term strategy for temporal diversification makes zero sense. The strategy has a very strange idea of what risk entails - its version of "managing risk" entails buying shares using levered funds and buying only when volatility is low (when fundamental valuations are likely higher), both things that increase rather than reduce risk (real risk, which is the permanent impairment of capital, and not short-term volatility).
Reducing equity exposure when volatility is high is actually an entirely reasonable strategy, and not just for behavioral reasons. Despite common misconceptions, the benefits of diversification (temporal or otherwise) don't flow from spreading capital across different assets but rather from spreading risk across different sources: the former is merely a means to achieve the latter. Because of volatility's well-documented tendency to cluster, a strategy that maintains a relatively constant amount of volatility tends to be more efficient than one that maintains a relatively constant allocation of capital.

This is why target volatility strategies generally prove superior to fixed allocations with similar long-term average variance.
I think it would be helpful for forum members for you post a real-life example. If not your own portfolio, then a reasonable approximation of what that portfolio would look like, at this current moment, on March 13. So that we can understand the mechanics of how it's levered, how much it's levered, how you're "spreading risk" - and most importantly, so that it's fully transparent and we can assess how this portfolio performs going forward.

Looking forward to the actual portfolio composition as of March 13. Thanks
ChrisBenn
Posts: 454
Joined: Mon Aug 05, 2019 7:56 pm

Re: Lifecycle Investing - Still the Right Strategy

Post by ChrisBenn »

Silence Dogood wrote: Fri Mar 13, 2020 9:39 pm
Ben Mathew wrote: Fri Mar 13, 2020 5:54 pm I sometimes see comments that ask where are the "100% stocks" and "leveraged" posts when the market is down. I felt it's useful to let people know that the lifecycle strategy is an approach designed to work well through both the ups and the downs.
(...)

Recently I read a post advocating for a fixed asset allocation for life. I find it interesting how different these two strategies are.
Actually in some ways not that different; the argument here would be for a fixed asset exposure over ones lifetime.
User avatar
Steve Reading
Posts: 2668
Joined: Fri Nov 16, 2018 10:20 pm

Re: Lifecycle Investing - Still the Right Strategy

Post by Steve Reading »

Silence Dogood wrote: Fri Mar 13, 2020 9:39 pm
Ben Mathew wrote: Fri Mar 13, 2020 5:54 pm I sometimes see comments that ask where are the "100% stocks" and "leveraged" posts when the market is down. I felt it's useful to let people know that the lifecycle strategy is an approach designed to work well through both the ups and the downs.
Specifically regarding leverage, wasn't that the strategy attempted by market timer?

Or are you advocating for something different?

Recently I read a post advocating for a fixed asset allocation for life. I find it interesting how different these two strategies are.
This strategy is identical to Siamond's during retirement. Just a fixed allocation (say 40/60). But during accumulation, our jobs act like bonds. So to have a fixed asset allocation (again, 40/60), you count that job as the bond. And it's a VERY large bond with respect to our savings for most of us young accumulators. So we invest the savings with leverage to try to balance out the large job-like bond and maintain the desired allocation (40/60).
"... so high a present discounted value of wealth, it is only prudent for him to put more into common stocks compared to his present tangible wealth, borrowing if necessary" - Paul Samuelson
Silence Dogood
Posts: 1447
Joined: Tue Feb 01, 2011 9:22 pm

Re: Lifecycle Investing - Still the Right Strategy

Post by Silence Dogood »

305pelusa wrote: Fri Mar 13, 2020 10:17 pm
Silence Dogood wrote: Fri Mar 13, 2020 9:39 pm
Ben Mathew wrote: Fri Mar 13, 2020 5:54 pm I sometimes see comments that ask where are the "100% stocks" and "leveraged" posts when the market is down. I felt it's useful to let people know that the lifecycle strategy is an approach designed to work well through both the ups and the downs.
Specifically regarding leverage, wasn't that the strategy attempted by market timer?

Or are you advocating for something different?

Recently I read a post advocating for a fixed asset allocation for life. I find it interesting how different these two strategies are.
This strategy is identical to Siamond's during retirement. Just a fixed allocation (say 40/60). But during accumulation, our jobs act like bonds. So to have a fixed asset allocation (again, 40/60), you count that job as the bond. And it's a VERY large bond with respect to our savings for most of us young accumulators. So we invest the savings with leverage to try to balance out the large job-like bond and maintain the desired allocation (40/60).
So, if a young investor contributes to a target retirement fund, while also paying down debt (e.g., a mortgage) throughout their lifetime, are they, in effect, doing what you are advocating for?
User avatar
Steve Reading
Posts: 2668
Joined: Fri Nov 16, 2018 10:20 pm

Re: Lifecycle Investing - Still the Right Strategy

Post by Steve Reading »

Silence Dogood wrote: Fri Mar 13, 2020 10:42 pm
305pelusa wrote: Fri Mar 13, 2020 10:17 pm
Silence Dogood wrote: Fri Mar 13, 2020 9:39 pm
Ben Mathew wrote: Fri Mar 13, 2020 5:54 pm I sometimes see comments that ask where are the "100% stocks" and "leveraged" posts when the market is down. I felt it's useful to let people know that the lifecycle strategy is an approach designed to work well through both the ups and the downs.
Specifically regarding leverage, wasn't that the strategy attempted by market timer?

Or are you advocating for something different?

Recently I read a post advocating for a fixed asset allocation for life. I find it interesting how different these two strategies are.
This strategy is identical to Siamond's during retirement. Just a fixed allocation (say 40/60). But during accumulation, our jobs act like bonds. So to have a fixed asset allocation (again, 40/60), you count that job as the bond. And it's a VERY large bond with respect to our savings for most of us young accumulators. So we invest the savings with leverage to try to balance out the large job-like bond and maintain the desired allocation (40/60).
So, if a young investor contributes to a target retirement fund, while also paying down debt (e.g., a mortgage) throughout their lifetime, are they, in effect, doing what you are advocating for?
A home is a financial asset that also acts like a "bond" in that it "pays you rent" (via you not paying any rent) reliably and consistently. So the additional bond-like home asset directly offsets the negative-bond-like mortgage. So buying a house with a mortgage vs renting actually makes no change on a temporal diversification perspective (although they are both better than paying off the house immediately, at least from a temporal perspective).

So it can really boil down to the target retirement fund. This has better temporal diversification than just contributing to a fixed allocated fund. But one can do even better by "breaking the ceiling" and going more aggressive at the start (say 200%) and more conservative at the end (say 20%).
"... so high a present discounted value of wealth, it is only prudent for him to put more into common stocks compared to his present tangible wealth, borrowing if necessary" - Paul Samuelson
User avatar
market timer
Posts: 6384
Joined: Tue Aug 21, 2007 1:42 am

Re: Lifecycle Investing - Still the Right Strategy

Post by market timer »

For people just starting their careers, the lifecycle investing idea is reasonable, but not terribly meaningful. The book opens with the example of a recent Yale Law grad, probably earning close to $200K/year, and deciding how to invest his first $5K. He decides to buy a SPY LEAP call offering 2:1 leverage in his Roth IRA. This is back in early 2010, so it was an investment that paid off well. Personally, I think it is debatable whether the law grad should use the money to pay down student loans, keep an emergency fund, or lever up, but this particular decision is not likely to have a material impact on his wealth in 40 years.

Let's fast forward 15 years. This is where things get really interesting and where the book does not spend much time. That Yale Law grad is now a partner at a law firm and 40 years old. He has a portfolio worth $2M, and his income has recently ramped up to $700K/year (pre-tax). He has a family and his contribution to household expenses totals $150K/year. Now the decision to leverage is no longer an academic exercise--there are real consequences to his decision.

The lifecycle investing approach says he should maintain 2x leverage and buy $4M in equities, as lifetime wealth is >$10M. However, this lawyer is a bit concerned about job security and is not sure his motivation is enough to last another 15-20 years. Moreover, he only needs about $2.5M in a balanced 60/40 portfolio to meet his needs, factoring in that living expenses will drop after the kids leave the house.

How should this 40-year-old invest?
User avatar
Uncorrelated
Posts: 1232
Joined: Sun Oct 13, 2019 3:16 pm

Re: Lifecycle Investing - Still the Right Strategy

Post by Uncorrelated »

YearTrader wrote: Fri Mar 13, 2020 5:13 pm I found it hard to estimate my future cash flow / human capital and risk tolerance / utility function. Also AFAIK I'd need to estimate the long term expected risk and return of asset classes in my investing universe.

How sensitive is this strategy to these input parameters? I think it's still a valid strategy as long as the overall variance of long term investment return can be reduced.
It's true that lifecycle investing requires you to estimate all those parameters, but a constant asset allocation also requires an estimation of risk tolerance and long term expected risk and return. Maybe your estimated risk tolerance and return are widely inaccurate, but in that case there is no reason to assume that a constant asset allocation is any better, since it (implicitly) depends on the same estimates.

The real question is: is it possible to make a better estimate of future cash flow than $0? If the answer is yes, then lifecycle investing results in higher utility. If the answer is no, then both lifecycle investing and a constant asset allocation end up with the same asset allocation.

A constant asset allocation isn't model free, it implicitly assumes exactly $0 cash flow, a fixed risk tolerance (to be exact, a CRRA) and certain return assumptions.
Last edited by Uncorrelated on Sat Mar 14, 2020 6:58 am, edited 1 time in total.
User avatar
Steve Reading
Posts: 2668
Joined: Fri Nov 16, 2018 10:20 pm

Re: Lifecycle Investing - Still the Right Strategy

Post by Steve Reading »

market timer wrote: Sat Mar 14, 2020 12:38 am
The lifecycle investing approach says he should maintain 2x leverage and buy $4M in equities, as lifetime wealth is >$10M. However, this lawyer is a bit concerned about job security and is not sure his motivation is enough to last another 15-20 years. Moreover, he only needs about $2.5M in a balanced 60/40 portfolio to meet his needs, factoring in that living expenses will drop after the kids leave the house.

How should this 40-year-old invest?
The conundrum you show (I think) is that the lawyer might want to keep working until he gets to 10M. But who knows, maybe he burns out and doesn’t? I’ve thought about this and the way I’ve handled it is that I temporally diversify whatever I know I will save. So in this scenario, lawyer knows he’ll need to hit 2.5M at 60/40, so he might reasonably set his current 2M and ~75/25.

If he ends up working extra because he likes working in general, then these extra savings won’t be as temporally diversified. So what? He doesn’t depend as much on them either.

I like the idea to temporally diversify only with conservative estimates of future cash flows, estimates you feel pretty confident you’ll make. So you’ll only be perfectly diversified if you do end up having an emergency, decide to retire early, etc. That’s good enough for me.
"... so high a present discounted value of wealth, it is only prudent for him to put more into common stocks compared to his present tangible wealth, borrowing if necessary" - Paul Samuelson
User avatar
Topic Author
Ben Mathew
Posts: 1409
Joined: Tue Mar 13, 2018 11:41 am
Location: Seattle
Contact:

Re: Lifecycle Investing - Still the Right Strategy

Post by Ben Mathew »

Caduceus wrote: Fri Mar 13, 2020 5:16 pm Is there a valuation component to this strategy? I don't see why time diversification would be such a great thing if you're leveraging up as a young investor when the stock market is richly priced and expensive, and then taking proportionately less leverage than a traditional investor during some later period when the markets may be much more cheaply priced.
No valuation component, but you can add it if you wish. So you'd have with a steep glide age based path based on age, which you could modify up or down based on current valuations. The lifecycle approach would be agnostic about whether that's a good idea or not, just like it would be agnostic about whether certain portions of the stock market provide excess returns.
User avatar
Topic Author
Ben Mathew
Posts: 1409
Joined: Tue Mar 13, 2018 11:41 am
Location: Seattle
Contact:

Re: Lifecycle Investing - Still the Right Strategy

Post by Ben Mathew »

- duplicate post -
User avatar
Topic Author
Ben Mathew
Posts: 1409
Joined: Tue Mar 13, 2018 11:41 am
Location: Seattle
Contact:

Re: Lifecycle Investing - Still the Right Strategy

Post by Ben Mathew »

bluquark wrote: Fri Mar 13, 2020 7:24 pm I tend to agree with Bernstein that although it sounds good in theory, the problem is that it ignores non-market risks.

The first risk is the possibility of an ill-timed episode of unemployment or disability requiring much earlier drawdowns than planned. Especially a problem for those with dependents.
Estimating future savings is a challenge. But a fixed allocation is optimal only if there are no more savings. That would be a bad assumption for most people, who end up having a career where they are able to save over much of their working age years. Yes, things could go worse than expected. But it seems more realistic to plan for continued contributions, and then change course as new information comes in.

Being conservative in your estimates about future savings (as 305pelusa is doing) is fine. But at least model your future savings and base your asset allocation on somewhat realistic expecations. Zero future savings is not realistic for most people.
Caduceus
Posts: 2810
Joined: Mon Sep 17, 2012 1:47 am

Re: Lifecycle Investing - Still the Right Strategy

Post by Caduceus »

Ben Mathew wrote: Sat Mar 14, 2020 9:27 am
Caduceus wrote: Fri Mar 13, 2020 5:16 pm Is there a valuation component to this strategy? I don't see why time diversification would be such a great thing if you're leveraging up as a young investor when the stock market is richly priced and expensive, and then taking proportionately less leverage than a traditional investor during some later period when the markets may be much more cheaply priced.
No valuation component, but you can add it if you wish. So you'd have with a steep glide age based path based on age, which you could modify up or down based on current valuations. The lifecycle approach would be agnostic about whether that's a good idea or not, just like it would be agnostic about whether certain portions of the stock market provide excess returns.
Anytime you can borrow at a rate that is lower than the future expected returns of an asset, you can expect to do well a majority of the time and utterly fail a small proportion of the time. This is no different than firms like Exxon borrowing at 3% to retire shares that are trading at a 10% cost of equity. The basic financial principles are not difficult to understand, and there would be no need to "jazz" it up by using phrases like "lifecycle" investing.

But the approach of borrowing at a lower cost than future expected returns is based on the assumption that you are right about future expected returns. I do not buy for a second that a strategy that is agnostic about market returns over an investor's lifetime will do well - the valuation component is what makes the leverage valuable. It would be utterly stupid to borrow at 5% to invest in long-term bonds yielding 2%, so why would you borrow at 5% to invest in equities so richly priced that the equity coupon is lower than your borrowing cost?

I do not think it is a coincidence at all that the two posters who've decided to embrace this strategy both happened to do so within, what, like one year of historic bear markets, when stocks have had a bull run and volatility was nice and low? They initiated this strategy precisely during a time when the equity coupon didn't justify the risk taken.

I've noticed that the most important thing has not yet been answered: Can we see an example of what such a portfolio would look like as of March 13, 2020? Give us a real-life portfolio: the exact positions and percentages, how the leverage is structured so we can assess the real cost, its performance since inception, etc. Bring the portfolio into the light - if you are right, it will outperform in the decades to come, so this should be very simple.
User avatar
Steve Reading
Posts: 2668
Joined: Fri Nov 16, 2018 10:20 pm

Re: Lifecycle Investing - Still the Right Strategy

Post by Steve Reading »

Caduceus wrote: Sat Mar 14, 2020 10:11 am
Anytime you can borrow at a rate that is lower than the future expected returns of an asset, you can expect to do well a majority of the time and utterly fail a small proportion of the time. This is no different than firms like Exxon borrowing at 3% to retire shares that are trading at a 10% cost of equity. The basic financial principles are not difficult to understand, and there would be no need to "jazz" it up by using phrases like "lifecycle" investing.
When Enron borrowed at 3% to invest at 10%, they had higher returns and higher risk. This strategy produces higher returns at lower risks. This is because of the temporal diversification effect that Enron isn't participating in. That's why we "jazz" it up with a unique phrase, because it is unique.
Caduceus wrote: Sat Mar 14, 2020 10:11 am But the approach of borrowing at a lower cost than future expected returns is based on the assumption that you are right about future expected returns. I do not buy for a second that a strategy that is agnostic about market returns over an investor's lifetime will do well - the valuation component is what makes the leverage valuable.
If one borrows at the risk-free rate, like I am, then this strategy is completely agnostic to valuations or future expected returns. Given that I will take X amount of stock risk throughout my life, it is optimal to spread it out, period. If the risk free rate/borrow rate was higher than expected equity returns, not only should you not use Lifecycle Investing: you shouldn't invest in stocks at all, ever.

To the extent borrowing costs are higher than risk-free rate, then you'd still implement Lifecycle Investing but not leverage as much. If borrowing costs were higher than stock expected returns but the risk-free rate wasn't (ex: your only source of borrowing would be credit cards) then you might do Lifecycle Investing without leverage. Start a 100% stocks at ramp as appropriate.

Caduceus wrote: Sat Mar 14, 2020 10:11 am I do not think it is a coincidence at all that the two posters who've decided to embrace this strategy both happened to do so within, what, like one year of historic bear markets
Human beings are excellent at finding patterns where there are none.

Caduceus wrote: Sat Mar 14, 2020 10:11 am They initiated this strategy precisely during a time when the equity coupon didn't justify the risk taken.
I initiated this strategy back when the PE ratio was ~25. That's at least a 4% yield. If earnings grew just to keep pace with inflation, that's a 4% real yield, well above my borrowing costs of 1.77% nominal, about 3.5% with taxes included. I initiated it when there was a reasonable equity risk premium (6% - 3.5% = 2.5% given a 2% inflation). And earnings tend to outpace inflation so the premium was even larger. And I expected some TLHing along the way to lower the tax bill too. Of course, humans have excellent hindsight bias and now this crash was evident apparently. I don't rely on such fallacies; I feel confident when I initiated and I feel confident now as well.
Caduceus wrote: Sat Mar 14, 2020 10:11 am I've noticed that the most important thing has not yet been answered: Can we see an example of what such a portfolio would look like as of March 13, 2020? Give us a real-life portfolio: the exact positions and percentages, how the leverage is structured so we can assess the real cost, its performance since inception, etc. Bring the portfolio into the light - if you are right, it will outperform in the decades to come, so this should be very simple.
I've been very forthcoming as to this strategy in my thread; I've documented my leveraged positions since the beginning. It's a little weird to see you ask this since I know you know about that thread. You asked me how I was doing just 2 days ago 0_o:
viewtopic.php?p=5089052#p5089052

If you have more specific questions on my portfolio, feel free to ask there.
"... so high a present discounted value of wealth, it is only prudent for him to put more into common stocks compared to his present tangible wealth, borrowing if necessary" - Paul Samuelson
Caduceus
Posts: 2810
Joined: Mon Sep 17, 2012 1:47 am

Re: Lifecycle Investing - Still the Right Strategy

Post by Caduceus »

305pelusa wrote: Sat Mar 14, 2020 10:50 am
Caduceus wrote: Sat Mar 14, 2020 10:11 am I've noticed that the most important thing has not yet been answered: Can we see an example of what such a portfolio would look like as of March 13, 2020? Give us a real-life portfolio: the exact positions and percentages, how the leverage is structured so we can assess the real cost, its performance since inception, etc. Bring the portfolio into the light - if you are right, it will outperform in the decades to come, so this should be very simple.
I've been very forthcoming as to this strategy in my thread; I've documented my leveraged positions since the beginning. It's a little weird to see you ask this since I know you know about that thread. You asked me how I was doing just 2 days ago 0_o:
viewtopic.php?p=5089052#p5089052

If you have more specific questions on my portfolio, feel free to ask there.
I see the updates on the overall portfolio value. And I know that you're down -49% now. But I don't see the details of what the portfolio is actually invested in, the exact proportions, and the details about how the leverage is being executed and the associated borrowing costs. Why not put that right at the top in the first post, in the same section that you update changes to portfolio value? Full transparency that way.

I don't want to give the sense that I wish you any ill-will. I think it's a silly strategy but I hope you succeed. I do not believe you fully comprehend how risky this is, given the way you've rationalized away why earlier posters have failed at this.
User avatar
vineviz
Posts: 8571
Joined: Tue May 15, 2018 1:55 pm

Re: Lifecycle Investing - Still the Right Strategy

Post by vineviz »

Caduceus wrote: Fri Mar 13, 2020 9:49 pm I think it would be helpful for forum members for you post a real-life example. If not your own portfolio, then a reasonable approximation of what that portfolio would look like, at this current moment, on March 13. So that we can understand the mechanics of how it's levered, how much it's levered, how you're "spreading risk" - and most importantly, so that it's fully transparent and we can assess how this portfolio performs going forward.

Looking forward to the actual portfolio composition as of March 13. Thanks
The strategy we're discussing is highly adaptable and is designed to help YOU manage YOUR investment strategy to suit YOUR particular circumstances, as well as the realized returns that YOU receive.

Plus, it's entirely customizable: if you don't feel comfortable with 2:1 leverage then feel free to cap it at 1.5x instead. If you don't want the allocation to vary based on current market volatility then use a constant VIX as the input to your Samuelson Share.

For kicks, though I ran a handful of simulations using their tool to give you some idea of what kind of equity allocations this approach might present. I kept RRA, savings rate, etc. constant and plotted the glide path over five different series of market returns. For comparison, I also include in the graph the allocations from Vanguard Target Retirement funds at the same ages.

Image

The level of leverage typically decreases pretty quickly, dropping considerably in the first 5-10 years. In these examples, Ayres & Nalebuff's approach results in a MORE conservative asset allocation than Vanguard after age 40 which is an effect of the superior risk-smoothing.
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch
langlands
Posts: 752
Joined: Wed Apr 03, 2019 10:05 pm

Re: Lifecycle Investing - Still the Right Strategy

Post by langlands »

[deleted duplicate]
Last edited by langlands on Sat Mar 14, 2020 11:24 am, edited 2 times in total.
langlands
Posts: 752
Joined: Wed Apr 03, 2019 10:05 pm

Re: Lifecycle Investing - Still the Right Strategy

Post by langlands »

Caduceus wrote: Fri Mar 13, 2020 9:18 pm
305pelusa wrote: Fri Mar 13, 2020 8:21 pm
Caduceus wrote: Fri Mar 13, 2020 8:07 pm The strategy has a very strange idea of what risk entails - its version of "managing risk" entails buying shares using levered funds and buying only when volatility is low (when fundamental valuations are likely higher), both things that increase rather than reduce risk (real risk, which is the permanent impairment of capital, and not short-term volatility).
I personally agree. Over the short term (days and weeks), stocks don't behave like normal distributions. Standard deviation is not an appropriate measure of risk (it misses the kurtosis, case in point this week) and hence the Merton allocation breaks down.
My preferred method is to just use long-term volatility (where stocks are somewhat normally distributed due to the Central Limit Theorem), apply Lifecycle Investing allocation to set my long-term strategic allocation with that long-term measure of risk, and just "ignore" the short term fluctuations. Others will do something else I guess 0_o
A couple of points, in no particular order.

1) Standard deviation is not an appropriate measure of risk not because it misses kurtosis, but because it has absolutely nothing to do with the inherent valuation of stocks. A stock that fluctuates a great deal is neither more or less risky than a stock that does not fluctuate. You must think about what stocks represent. Stocks represent fractional interests in real businesses. They have real assets, real income, real liabilities. Their true value is connected to the fundamental business performance, not whether the market constantly changes its mind on what the business is worth, which often have very little to do with assessments of the business value.

Let me give you a real world example to explain why risk-as-volatility is an inherently ridiculous idea. During the tech bubble, there were companies with stock prices that hadn't moved much in a decade - highly stable, boring, mildly profitable companies that did nothing but pay out dividends. When people got crazy with all things technological, some CEOs had the bright idea to change the company name to something obviously technological. So, MIS International became Cosmoz.com. Guess what? As expected, its volatility suddenly exploded. Same company. Same business. Same profits. Same everything. Only thing that changed? The name. And, consequently, the volatility. So you expect me to believe that the company is riskier than it was before because its volatility has increased? In financial markets, stocks can experience volatility for a variety of reasons that have absolutely nothing to do with their inherent business value. Why should the frequency with which the market changes its mind on the value of a given stock or index have any bearing on its future value? And it doesn't. Finance academics were once enamored with high-beta stocks. When that didn't work, they went to low-beta stocks. Guess what - there's a reason you don't hear about beta very much any more as a sales pitch.

2) I would be very, very careful about using mathematical tools to model financial reality. The stock market is not a physical system and it does not - and will never - obey the rules of the model you set up to describe it. Historical and implied volatilities will change because the historical conditions that have underpinned those datasets have changed, so your estimates of volatility are useless. You can use math to describe the past - those relationships are static and won't change. But it is a fool's errand to assume that it can and will be applicable to the future.

I encourage you to read two great books in finance. The first is Roger Lowenstein's When Genius Failed: The Rise and Fall of Long Term Capital Management. You will love it. It explores how and why a fund started by Nobel Prize winning mathematicians/academics and some of the brightest minds on Wall Street ended in one of the worst debacles in financial history. The crash of their fund brought down not only themselves but the financial markets. The second book is Andrew Ross Sorkin's Too Big To Fail. It tells the story of how the very same mathematicians who created elegant and complicated Value At Risk models did not understand the models themselves, and could not explain why they weren't describing what was actually happening with the banks' portfolios. If you're the type of person who's good at math, it's going to be very seductive to think you're bringing the right tools to the job of investing. You're not.

We've had long enough time periods to know that stocks selected with the capital asset pricing model in mind haven't done well.

Ultimately, the proof is in the pudding. Look at the people who've actually made the most money, consistently, over long periods. There are the index investors, like the Bogleheads, and then the value investors, like Warren Buffett and Seth Klarman and Lou Simpson, Walter Schloss, etc. That's a fact. Other things have come and gone - temporary fads. No one's talking about quant funds like AQR these days on the board.
I think your bias against mathematics and applications of quantitative methods to finance is understandable, but unwarranted. Unfortunately, quant finance gets a lot of bad press, mostly because there are a lot of charlatans. People in the quant hedge fund industry know that there is a world of difference between a place like Rentech (https://en.wikipedia.org/wiki/Renaissance_Technologies) and a place like AQR. It's honestly a joke that they can both be considered "quant funds." It's not an accident that AQR is the poster child of quant funds either: successful funds do everything they can to avoid the spotlight. It's also not an accident that Rentech was founded by a first rate mathematician and AQR by a student of Eugene Fama. (Look at the backgrounds of D.E. Shaw and Two Sigma founders as well, and you will begin to notice a pattern.). Quite frankly, academic empirical finance gives quant finance a bad name. (I'm not saying that everything they do is bad, but the seriousness with which they promulgate both their solid and speculative ideas could easily make a skeptic throw the baby out with the bath water.).

May basic point is that ultimately, mathematics is just a language, an extremely precise language to communicate ideas. It can't be wrong, it can only be misapplied. It often is.
Last edited by langlands on Sat Mar 14, 2020 12:07 pm, edited 1 time in total.
Post Reply