Author Topic: Target AA once fired  (Read 9695 times)

boarder42

  • Walrus Stache
  • *******
  • Posts: 7849
Target AA once fired
« on: December 02, 2016, 01:18:08 PM »
What's everyones target AA when fired and do you use a fixed or sliding scale.  I'm personally right now at a 90/10 fixed just based on cFIREsim but i'm open to new ideas if something creates better security.  including being coupled with a creative withdrawal strategy.

VoteCthulu

  • Bristles
  • ***
  • Posts: 409
Re: Target AA once fired
« Reply #1 on: December 02, 2016, 02:42:54 PM »
I'm not fired yet (2019 hopefully), but I currently have 90/10 and have no plans to change that in the next 20 years.

Classical_Liberal

  • Handlebar Stache
  • *****
  • Posts: 1151
  • Age: 43
Re: Target AA once fired
« Reply #2 on: December 02, 2016, 03:52:31 PM »
I have a love/hate relationship with portfolio questions on the forum. Certainly, I have my personal opinion on which portfolios are best for where one is at in accumulation/drawdown stages.  OP is not a beginner, I'm not sure you would benefit from anything I have to write, so a general comment & will follow with interest (no pun).

Portfolios are not one size fits all.  Do you own or rent, is SS in play, risk tolerance, how old are you, is it likely you may work some again, how much wiggle room in your budget, do you have kids/is a legacy important?  All of these really matter.  Generally, I think there is often a gross misunderstanding of how much a drawdown portfolio is benefited by reduced volatility.  CAGR is not king.

Edit: for my inability to grasp proper use of the english language.
« Last Edit: December 02, 2016, 03:54:41 PM by Classical_Liberal »

JustGettingStarted1980

  • Bristles
  • ***
  • Posts: 377
Re: Target AA once fired
« Reply #3 on: December 02, 2016, 04:40:29 PM »
Boarder,

I've given this question some amateur thought.

I'm a 90/10 kind of guy myself, but I think that when I FIRE in about 8 years, as I'll be living solely on my investments and not on any pension, SS, or real estate, I REALLY don't want to go back to work if there is some kind of economic crisis.

Thus, my withdrawal will be Variable Percentage Withdrawal, but also I think I'll have a 50/50 AA for the first year and then slowly increase the AA back to 80/20 by 5%/year over the next 6 years. The goal at the time of FIRE will be to protect my assets (because I already won the game), not make more assets (that I don't need, since I had already determined my FIRE number based on my wishes/desires for spend level).  So this is my way of protecting against the sequence of returns risk.

As for legacy, if I raise my kids right, their legacy will be to live there own damn lives the way they see fit, and if they get anything more than a great upbringing, a good moral education, and a college degree, they should feel lucky. :)

That's my plan this week, anyway.

JGS

ender

  • Walrus Stache
  • *******
  • Posts: 5028
Re: Target AA once fired
« Reply #4 on: December 02, 2016, 04:53:18 PM »
We're pretty close to 100% equities right now.

I figure that as we pay our mortgage that's sort of like bonds.

What we end up at will depend primarily on how much I work when we "FIRE." Given the massive advantages to part time working, we might put it all on black with 100% stocks, work for $20k/year, and roll with it.

boarder42

  • Walrus Stache
  • *******
  • Posts: 7849
Re: Target AA once fired
« Reply #5 on: December 02, 2016, 05:43:06 PM »
Briefly read some of these will read more thoroughly tomorrow.

Wanted to clarify the point isn't for me specifically but to create a discussion. There were some interesting concepts brought up in another thread and I'd like to create a big debate on this. We debate mortgage vs no mortgage til blue in the face but this is actually much much more important.

End goal is to optimize the AA to the nth degree   and maybe there are mixes we aren't even looking at. From a mid small cap variation BC the TSM is mostly large cap just due to the size.


Interest Compound

  • Pencil Stache
  • ****
  • Posts: 658
Re: Target AA once fired
« Reply #6 on: December 03, 2016, 10:41:54 AM »
100% World Cap-Weighted Stocks.

Currently: 52.6% USA / 47.4% International
Once Fired: Whatever the market decides :)

somers515

  • 5 O'Clock Shadow
  • *
  • Posts: 69
Re: Target AA once fired
« Reply #7 on: December 03, 2016, 01:25:38 PM »
Generally, I think there is often a gross misunderstanding of how much a drawdown portfolio is benefited by reduced volatility.  CAGR is not king.

+1 to this.  I've noticed that on this forum this point seems to get missed often.

Here are two links those reading this thread might find helpful.

http://jlcollinsnh.com/2014/06/10/stocks-part-xxiii-selecting-your-asset-allocation/

https://portfoliocharts.com/2015/11/17/how-safe-withdrawal-rates-work/

Interest Compound

  • Pencil Stache
  • ****
  • Posts: 658
Re: Target AA once fired
« Reply #8 on: December 03, 2016, 02:19:09 PM »
Generally, I think there is often a gross misunderstanding of how much a drawdown portfolio is benefited by reduced volatility.  CAGR is not king.

+1 to this.  I've noticed that on this forum this point seems to get missed often.

Here are two links those reading this thread might find helpful.

http://jlcollinsnh.com/2014/06/10/stocks-part-xxiii-selecting-your-asset-allocation/

https://portfoliocharts.com/2015/11/17/how-safe-withdrawal-rates-work/

All possible 50 year retirements using all data from 1871 with the 4% rule:



All possible 50 year retirements using all data from 1871 with the 5% rule:



And just for fun, all possible 50 year retirements using all data from 1871 with the 3% rule:


boarder42

  • Walrus Stache
  • *******
  • Posts: 7849
Re: Target AA once fired
« Reply #9 on: December 03, 2016, 02:52:22 PM »
Thanks compound. Cagr is king. But diminished after 90% equities

scottish

  • Handlebar Stache
  • *****
  • Posts: 1495
  • Location: Ottawa
Re: Target AA once fired
« Reply #10 on: December 03, 2016, 02:55:04 PM »
We have a five year bond ladder we can live off if stocks stink for 5 years.   I increase the bond holdings by $1000 per quarter every year.   Right now, we reinvest at $12K every quarter.

The rest is in equities.  10%-20%-30%-40%  for Asia, Europe, Canada, US.

Classical_Liberal

  • Handlebar Stache
  • *****
  • Posts: 1151
  • Age: 43
Re: Target AA once fired
« Reply #11 on: December 03, 2016, 03:47:45 PM »
@ Interest Compound, Point taken.

Counter point(s):

It's not just the end result that matters.  It's the ride along the way.  For some, lower highs and higher lows have a payoff.  For many others they are a requirement since they dont have it in them to handle 50% drawdowns and stay the course.

Cfiresim is awesome, but it's a hacksaw, not a scalpel.  There is more to life than VTI and BND.  Subsets of these and different asset classes each have their own history, correlations, and potentials.

If all one wants is fully FIRE and be successful, then by all means work until you have a 4%WR and go 90/10. Adjust for inflation and live life without ever looking at your portfolio balance, because risk is meaningless to you. Odds are highly stacked you'll be ridiculously wealthy when you die.  90/10 Fits great if you're a size 40 regular. 

Some folks have a more complex web of goals. Example: Plan to FIRE, travel the world for the first few years not working, then return home and work part time or have capital to start a business.  Is 90/10 appropriate for this person?  Would her interests be better served with a different portfolio initially to reduce sequence risk when she has zero income potential & higher expenses?  Why not adjust portfolio to meet needs of individual situations, using all available data, then adjust when situation changes?

For those of us who need some tailoring and dont mind a bit of surgery, there are many tactical moves that could significantly reduce time worked and satisfy a guarded nature to more appropriately meet specific goals. I'd rather work less, decrease volatility (sleep better), and increase success rates for my personal web of goals.  The sacrifice for this is more time spent managing the portfolio, a smaller chance of dying uber rich, and I'm down with that.

msilenus

  • Pencil Stache
  • ****
  • Posts: 517
Re: Target AA once fired
« Reply #12 on: December 03, 2016, 03:54:36 PM »
i'm open to new ideas if something creates better security.  including being coupled with a creative withdrawal strategy.

We had a conversation that digressed in this direction a couple months back.
http://forum.mrmoneymustache.com/investor-alley/changing-asset-allocation-as-fire-approaches/msg1241674/#msg1241674

somers515

  • 5 O'Clock Shadow
  • *
  • Posts: 69
Re: Target AA once fired
« Reply #13 on: December 03, 2016, 04:15:15 PM »
From the portfolio link I provided above:

"After publishing a few tools and articles based on safe withdrawal rates, one of the most common questions Iíve seen so far is some iteration if this:

Obviously higher returns support higher withdrawal rates.  Thatís why I invest in 100% stocks!  How can a lower-return portfolio possibly support higher withdrawal rates than a higher-return portfolio?

I admit the answer is fairly unintuitive, and explaining this without getting too deep into the weeds is a bit of a challenge."

People should read the whole article but if you won't the short answer is volatility.  If you take 4% every year then in years that your 100% stocks portfolio has stocks going down by a lot it will really hurt you when you take out 4% of the original portfolio (this is certainly true if stocks go down a lot during the first few years of your retirement).  I'm not saying that you shouldn't have a very healthy percentage of stocks.  I'm also not trying to argue with anyone here.  I just would hate for anyone reading this thread to look at those charts and not at least consider the links I provided.  The good news is with 25x your annual spend as your nest egg you will almost certainly be good whatever asset allocation you decide upon.  Thanks.

boarder42

  • Walrus Stache
  • *******
  • Posts: 7849
Re: Target AA once fired
« Reply #14 on: December 03, 2016, 05:01:59 PM »
almost certainly be good with whatever asset allocation is 100% false for a 4% SWR

100% gold fails alat
100% bonds fails alot
100% cash will fail

lots of AA thats fail alot more with less equities

your statement of the draw down really only applies early in FIRE unless markets fundamentally change and dont perform at all for a long time.  in which case all AAs will likely fail

somers515

  • 5 O'Clock Shadow
  • *
  • Posts: 69
Re: Target AA once fired
« Reply #15 on: December 03, 2016, 06:31:44 PM »
almost certainly be good with whatever asset allocation is 100% false for a 4% SWR

From context in this thread you know that I wasn't referring to 100% gold, bonds or cash.  I was trying to be nice with that sentence.  I should have added the word reasonable in front of asset allocation though.  That better?  Anyway my point was that people on this forum routinely advocate for 100% stocks during the drawdown phase.  I add to that conversation that you should consider volatility during the drawdown phase.  Read the links that I provided, they spell it all out thoroughly.  If you have a criticism with the articles linked then I'd be interested.

Classical_Liberal

  • Handlebar Stache
  • *****
  • Posts: 1151
  • Age: 43
Re: Target AA once fired
« Reply #16 on: December 03, 2016, 08:12:46 PM »
almost certainly be good with whatever asset allocation is 100% false for a 4% SWR

100% gold fails alat
100% bonds fails alot
100% cash will fail

lots of AA thats fail alot more with less equities

your statement of the draw down really only applies early in FIRE unless markets fundamentally change and dont perform at all for a long time.  in which case all AAs will likely fail

Obviously you are correct. 

Thought experiment, Person A starts drawdown 90-10 rebalancing and person B starts 60 VTI-10 BND-10 LTT-10 Gold-10 Cash rebalancing.  If everything goes within a standard deviation of norm, both are doing just fine 10 years later.  90-10 is significantly richer, but what purpose does that wealth serve? 

Same scenario, but this time the trifecta of an early reversion to mean PE/10, slow growth, and above average inflation hit.  Ouch, right.  10 years in both are probably in trouble, but this time person B is richer.  Now what purpose does that wealth serve? 
« Last Edit: December 03, 2016, 08:24:08 PM by Classical_Liberal »

Classical_Liberal

  • Handlebar Stache
  • *****
  • Posts: 1151
  • Age: 43
Re: Target AA once fired
« Reply #17 on: December 03, 2016, 08:36:44 PM »
We had a conversation that digressed in this direction a couple months back.
http://forum.mrmoneymustache.com/investor-alley/changing-asset-allocation-as-fire-approaches/msg1241674/#msg1241674

But, we are having such a fun argument here!  I don't wanna ruin it by seeing that my thoughts are wholly unoriginal.

Interest Compound

  • Pencil Stache
  • ****
  • Posts: 658
Re: Target AA once fired
« Reply #18 on: December 03, 2016, 09:22:08 PM »
@ Interest Compound, Point taken.

Counter point(s):

It's not just the end result that matters.  It's the ride along the way.  For some, lower highs and higher lows have a payoff.  For many others they are a requirement since they dont have it in them to handle 50% drawdowns and stay the course.

Sure, that analysis only looks at the numbers, and the numbers assume people aren't idiots. But if we're going to take emotions into account, we can't just look at the downside. Those same people are also at risk of jumping around due to their portfolio "underperforming" during good times. Look at all the people jumping out of Betterment recently, because their portfolio is "underperforming" the market over the last 8 months or so. Be careful not to apply that thinking only to the rare bad times, it applies in good times as well.

Why do you believe you'd be more suseptible to one type of emotional mistake, and not the other? When you deviate from the market, people always find a way to act like idiots and let emotions ruin their portfolio.

Cfiresim is awesome, but it's a hacksaw, not a scalpel.  There is more to life than VTI and BND.  Subsets of these and different asset classes each have their own history, correlations, and potentials.

Now you're going into "if you can predict the future winners, put money in them now, and avoid the losers, you'll have more money!" territory. This has proven to be a loser's game.

If all one wants is fully FIRE and be successful, then by all means work until you have a 4%WR and go 90/10. Adjust for inflation and live life without ever looking at your portfolio balance, because risk is meaningless to you. Odds are highly stacked you'll be ridiculously wealthy when you die.  90/10 Fits great if you're a size 40 regular. 

Some folks have a more complex web of goals. Example: Plan to FIRE, travel the world for the first few years not working, then return home and work part time or have capital to start a business.  Is 90/10 appropriate for this person?  Would her interests be better served with a different portfolio initially to reduce sequence risk when she has zero income potential & higher expenses?  Why not adjust portfolio to meet needs of individual situations, using all available data, then adjust when situation changes?

Money is fungible. Money doesn't care about emotions. It doesn't matter why it's being spent. Here's what it looks like if you're spending an extra 10% ($4000 in this example) a year for the first 5 years traveling the world:



Here's what it looks like if you spend an extra 25% ($10,000 in this example) a year for the first 5 years traveling the world:



Here's what it looks like if you spend an extra 25% ($10,000 in this example) a year for the first 5 years traveling the world, then brought in $1000 a month from your part-time job, for 10 years:



Here's what it looks like if you spend an extra 25% ($10,000 in this example) a year for the first 5 years traveling the world, then spent $100,000 (inflation-adjusted) for start-up costs for your business, then brought in $1000 a month from your business for 10 years:



Here's what it looks like if you spend an extra 25% ($10,000 in this example) a year for the first 5 years traveling the world, then spent $100,000 (inflation-adjusted) for start-up costs for your business, then brought in $1000 a month from your business for 10 years, then spent $300,000 (inflation-adjusted) trying to fix the business, only to watch it fail anyway:



Here's what it looks like if you spend an extra 25% ($10,000 in this example) a year for the first 5 years traveling the world, then spent $300,000 (inflation-adjusted) for start-up costs for your business, then brought in $1000 a month from your business for 10 years:



Here's what it looks like if you spend an extra 25% ($10,000 in this example) a year for the first 5 years traveling the world, then spent $100,000 (inflation-adjusted) for start-up costs for your business, then lost $1000 a month from your business for 10 years:



...etc

For those of us who need some tailoring and dont mind a bit of surgery, there are many tactical moves that could significantly reduce time worked and satisfy a guarded nature to more appropriately meet specific goals. I'd rather work less, decrease volatility (sleep better), and increase success rates for my personal web of goals.  The sacrifice for this is more time spent managing the portfolio, a smaller chance of dying uber rich, and I'm down with that.

This is a dangerous game you're playing. On what are you basing your assertion that your "tactical moves" would result in a net-positive and not a net-negative?
« Last Edit: December 03, 2016, 09:27:04 PM by Interest Compound »

arebelspy

  • Administrator
  • Senior Mustachian
  • *****
  • Posts: 28239
  • Age: -999
  • Location: Seattle, WA
Re: Target AA once fired
« Reply #19 on: December 04, 2016, 05:08:27 AM »
Holy crap, this is some fun mathing, Interest Compound.  I think this is my second favorite IC contributions, after the GB thread.

Following!
We are two former teachers who accumulated a bunch of real estate, retired at 29, spent some time traveling the world full time and are now settled with three kids.
If you want to know more about us, or how we did that, or see lots of pictures, this Business Insider profile tells our story pretty well.
We (rarely) blog at AdventuringAlong.com. Check out our Now page to see what we're up to currently.

Tyler

  • Handlebar Stache
  • *****
  • Posts: 1160
Re: Target AA once fired
« Reply #20 on: December 04, 2016, 10:09:08 AM »
Cfiresim is awesome, but it's a hacksaw, not a scalpel.  There is more to life than VTI and BND.  Subsets of these and different asset classes each have their own history, correlations, and potentials.

Now you're going into "if you can predict the future winners, put money in them now, and avoid the losers, you'll have more money!" territory. This has proven to be a loser's game.

Nah -- Classical_Liberal is just rightfully pointing out that there's more to retirement backtesting than studying only two indices and projecting the results to everything else.  For example, I know you invest heavily in international stocks which cFIREsim completely ignores.  Maybe your own portfolio would have followed the same trend in the charts, but maybe not.  Not all stocks and bonds are created equal. 

In any case, I admire all of the thought and research you put into your investment decisions and retirement contingency planning.  Keep doing what you're doing. 
« Last Edit: December 04, 2016, 10:59:01 AM by Tyler »

Interest Compound

  • Pencil Stache
  • ****
  • Posts: 658
Re: Target AA once fired
« Reply #21 on: December 04, 2016, 12:51:32 PM »
Cfiresim is awesome, but it's a hacksaw, not a scalpel.  There is more to life than VTI and BND.  Subsets of these and different asset classes each have their own history, correlations, and potentials.

Now you're going into "if you can predict the future winners, put money in them now, and avoid the losers, you'll have more money!" territory. This has proven to be a loser's game.

Nah -- Classical_Liberal is just rightfully pointing out that there's more to retirement backtesting than studying only two indices and projecting the results to everything else.  For example, I know you invest heavily in international stocks which cFIREsim completely ignores.  Maybe your own portfolio would have followed the same trend in the charts, but maybe not.  Not all stocks and bonds are created equal. 

In any case, I admire all of the thought and research you put into your investment decisions and retirement contingency planning.  Keep doing what you're doing.

The charts from CFiresim were posted in response to:

Generally, I think there is often a gross misunderstanding of how much a drawdown portfolio is benefited by reduced volatility.  CAGR is not king.

and

Why not adjust portfolio to meet needs of individual situations, using all available data, then adjust when situation changes?

We have data of both Stocks (volatile) and Bonds (non-volatile) going back to 1871. Looking at this data, I set out to determine if reducing portfolio volatility increased the portfolio's success rate, or otherwise benefited the drawdown of the portfolio.

It did not.

Then I set out to determine if tailoring the withdrawals to meet the needs of various individual situations would result in reduced volatility increasing the portfolio's success rate, or otherwise benefiting the drawdown of the portfolio.

It did not.

It may be obvious to you or me, but the lesson is easy to miss; Reducing volatility does not necessarily result in improved drawdown of the portfolio, even when some sequence of returns risk shows up. If you're trying to pick "Subsets of these and different asset classes" that both reduce volatility and benefit portfolio drawdown, that most definitely involves predicting future winners.

And yes, predicting future winners has proven to be a loser's game.

Classical_Liberal

  • Handlebar Stache
  • *****
  • Posts: 1151
  • Age: 43
Re: Target AA once fired
« Reply #22 on: December 04, 2016, 01:05:31 PM »
Sure, that analysis only looks at the numbers, and the numbers assume people aren't idiots. But if we're going to take emotions into account, we can't just look at the downside. Those same people are also at risk of jumping around due to their portfolio "underperforming" during good times. Look at all the people jumping out of Betterment recently, because their portfolio is "underperforming" the market over the last 8 months or so. Be careful not to apply that thinking only to the rare bad times, it applies in good times as well.

Why do you believe you'd be more suseptible to one type of emotional mistake, and not the other? When you deviate from the market, people always find a way to act like idiots and let emotions ruin their portfolio.
Money is fungible. Money doesn't care about emotions. It doesn't matter why it's being spent. 

These are excellent points!  One does have to ask themselves what they would do if their portfolio is underperforming (personally, this is something I've run through my head many times).  Money is nonemotional, but people are not, hence the disconnect.  No matter how many examples of Cfiresim math is provided, most people are not tolerant of 40+ percent dips to their life savings.  This can ultimately lead to an intellectual trap in the same way someone who panic sells gets stuck in an emotional trap. This is why investors & market timing fails so miserably. It happens over & over again, yet those few who CAN tolerate the risk ignore these real world facts. 

Each individual needs to have a portfolio the meets their unique needs, including emotional ones.  Granted, 40 or 50 years ago tailoring a personal portfolio to individual situation would have been expensive as hell (transaction costs, tax consequences, etc).  Times have changed!  Today low cost funds are available for more specific asset classes, 401K, roth, IRA's are available to rebalance portfolios without tax consequences and with low to no transaction costs.

Now you're going into "if you can predict the future winners, put money in them now, and avoid the losers, you'll have more money!" territory. This has proven to be a loser's game.

I'll defer to Tyler's answer above, I'll never be able to articulate my position better than such a heavy weight. I'll also reiterate his point that given your particular international equity weighting (which I'll assume you chose to meet your personal situation,  risk tolerance, and  diversification needs.  Since its not 90/10) those results wouldnt mean a darn thing. 

Here's what it looks like if you spend an extra 25% ($10,000 in this example) a year for the first 5 years traveling the world:...

I agree with Arebelspy, awesome math work!  Your responses are  always impressive, I should have known better than to provoke :) I've read through many of your contributions in the past and am grateful for them!

I could go to Tyler's site and run a bunch of scenarios as well.  If I did take that time, I'm sure I would be criticized that the data is limited to 1972+ and it would be a legit criticism. However, if data limitations are to be considered, shouldn't we also consider the limitations of cfiresim?  The fact it lumps all stock and bonds into two single categories is a huge limitation!  Neither tool is useless, but both have limitations. One thing that neither tool considers is the emotional side of investing the majority of individual investors doubtlessly endure. Nor can the tools identify when life goals will change, for early retirees this will likely happen often.  When it happens, each person should examine their current financial situation and investment allocation to make sure it is meeting their new needs; mathematically, emotionally, realistically for the near and long term.

This is a dangerous game you're playing. On what are you basing your assertion that your "tactical moves" would result in a net-positive and not a net-negative?

Agreed, life is a dangerous situation, as are unpredictable and emotional humans. Every decision we make has the potential for "net positives" and "net negatives", that's life.  If a person is saving for a specific goal, say to purchase a house in 3 years, would you recommend 90-10 for their savings because it always does better in the long run?  of course not!  That money should be in CD's.  What happens when the same emotional, unpredictable human decides two years later that they want to rent instead, should that money remain in CD's? 

I realize I'm not going to change IC's opinion and write this simply as a counterpoint to the purely intellectual long term view.  Not everyone's retirement consists of save 25X spending then draw down 4% inflation adjusted every year while golfing, living in a paid off home in the burb's, and raising a family with a pareto optimized life.  FIRE to Mustachians is a very wide range of goals and possibilities that will likely change over time. Freedom of time, location and pursuit personal goals means AA should adjust to meet each individuals goals, while taking into account individual personalities.

Tyler

  • Handlebar Stache
  • *****
  • Posts: 1160
Re: Target AA once fired
« Reply #23 on: December 04, 2016, 02:10:24 PM »
If you're trying to pick "Subsets of these and different asset classes" that both reduce volatility and benefit portfolio drawdown, that most definitely involves predicting future winners.

I disagree.  Where you see truly diversified portfolios as "predicting future winners", I see well-balanced options prepared for every economic outcome (including ones where stocks struggle for decades).  Research into these more diverse portfolios (looking well beyond intermediate treasuries as the only other option) shows that the improved consistency of an economically agnostic portfolio can absolutely benefit withdrawal rates. FWIW, according to my calculations your 50/50 US/International split adds about 0.5% to the SWR over the US market alone likely for this very reason.  :)

I'm happy to see how confident you are in your all-stock portfolio.  I'm personally more comfortable with a more diverse retirement portfolio less dependent on the stock market for success.  There's more than one good way to structure a retirement portfolio, and ultimately the best choice is the one that you'll stick with for the long haul. 
« Last Edit: December 04, 2016, 02:18:34 PM by Tyler »

Classical_Liberal

  • Handlebar Stache
  • *****
  • Posts: 1151
  • Age: 43
Re: Target AA once fired
« Reply #24 on: December 04, 2016, 02:32:53 PM »
@ Tyler and off topic.  The changes and additions to your site are awesome.  I find the Retirement spending calc is particularly helpful in my case.  Thanks again for all the work you do over there!

Interest Compound

  • Pencil Stache
  • ****
  • Posts: 658
Re: Target AA once fired
« Reply #25 on: December 04, 2016, 04:05:11 PM »
If you're trying to pick "Subsets of these and different asset classes" that both reduce volatility and benefit portfolio drawdown, that most definitely involves predicting future winners.

I disagree.  Where you see truly diversified portfolios as "predicting future winners", I see well-balanced options prepared for every economic outcome (including ones where stocks struggle for decades).  Research into these more diverse portfolios (looking well beyond intermediate treasuries as the only other option) shows that the improved consistency of an economically agnostic portfolio can absolutely benefit withdrawal rates. FWIW, according to my calculations your 50/50 US/International split adds about 0.5% to the SWR over the US market alone likely for this very reason.  :)

I'm happy to see how confident you are in your all-stock portfolio.  I'm personally more comfortable with a more diverse retirement portfolio less dependent on the stock market for success.  There's more than one good way to structure a retirement portfolio, and ultimately the best choice is the one that you'll stick with for the long haul.

The question is not if it can. The question is if it's reasonable to expect to succeed at it. If someone today intends to structure their retirement portfolio to both reduce volatility and benefit portfolio drawdown, using "Subsets of these and different asset classes", here are some of the possible choices they'd have:

Prosperity:
  • Total US Stock Market
  • Large Cap Value
  • Large Cap Growth
  • Large Cap Blend
  • Mid Cap Value
  • Mid Cap Growth
  • Mid Cap Blend
  • Small Cap Value
  • Small Cap Growth
  • Small Cap Blend
  • Micro Cap
  • Total International Stock Market
  • Emerging Market
  • Pacific Market
  • Europe Market
  • International Value
  • International Small
  • S&P 500
  • Dividend Appreciation
  • High Dividend
  • Value Index
  • Growth Index
  • FTSE Social Index
  • Extended Market
Inflation:
  • Short Term Tips
  • Long Term Tips
  • Gold?
  • REITS?
  • Silver?
  • Coal?
  • Platinum?
  • Diamonds?
  • Oil?
  • Natural Gas?
  • Coffee?
  • Sugar?
  • Copper?
  • Wheat?
  • Cotton?
(There are arguments for each one of these.)

Deflation:
  • Long Term Bond Index
  • Long Term Treasuries
  • Long Term Government Bonds
  • Long Term Corporate Bonds
  • Long Term Municipal Bonds
Recession:
  • Money Market
  • Short Term Bond index
  • Short Term Treasuries
  • Short Term Tips
  • Ultra Short Term Bonds
  • CDs
  • Savings accounts
Their success or failure, most certainly hinges on picking future winners from this group of 50+ choices. Your Golden Butterfly used hindsight to choose winners, successfully reducing volatility and benefiting portfolio drawdown. Before you can even begin to discuss such portfolios, some questions need to be answered:
  • How many combinations of these choices would've failed at this task? In other words, how many wrong answers to this question are there?
  • How many combinations of these choices would've succeeded at this task? In other words, how many correct answers to this question are there?
  • How much do these wrong and correct answers shift each year?
  • Is the potential benefit worth the risk?
Of anyone here, I think you are uniquely suited to actually answer these questions :)

I put myself in the shoes of a 1972 investor, and using all information that would've been available to me at the time, tried to make the same accomplishment; reduce volatility while benefiting portfolio drawdown. I used all the same principles, even hindsight, to make my selection. I came up with:

Pure Story 1:

Prosperity = US Stock Market
Inflation = TIPS
Deflation = Long Term Treasuries
Recession = Short Term Treasuries

Pure Story 2:

Prosperity = US Stock Market
Prosperity x2 = International Stock Market
Inflation = TIPS
Deflation = Long Term Treasuries
Recession = Short Term Treasuries

and Pure Story 3 (The Growth Butterfly)

Prosperity = Large Cap Growth
Prosperity x2 = Large Cap Growth
Inflation = TIPS
Deflation = Long Term Treasuries
Recession = Short Term Treasuries

All 3 spectacularly failed:










Without knowing the future, this proved to be an incredibly difficult task. I was usually successful at reducing volatility lower than that of a comparable 3-fund portfolio, but was never able to benefit portfolio drawdown. All my portfolios ran out of money.

Yes, there are lots of ways to structure a retirement portfolio, I'm simply highlighting the pitfalls of making your retirement dependent on picking future winners. Choose wrong, and you're in trouble.

Tyler

  • Handlebar Stache
  • *****
  • Posts: 1160
Re: Target AA once fired
« Reply #26 on: December 04, 2016, 05:38:18 PM »
The question is not if it can. The question is if it's reasonable to expect to succeed at it.

Peace.  I'm not going to argue.

I will simply state that based on the best data available, each of the portfolios on my site (including the Classic 60-40 and Three-Fund portfolio) have supported higher SWRs than the US stock market alone. Your Global Stock 50/50 has done the same!  So nobody has to make up portfolios themselves and hope to get lucky -- simply follow the advice of smart people like Bernstein, Swensen, and Browne and you'll already be on the right track. I have also provided plenty of data to explain how it works.  It's all based on sound economic and mathematical principles, and while some portfolios may do better than others in the future I think it's reasonable to expect the benefit of intelligent diversification in retirement portfolios will continue. 

You are also free to dismiss all of those portfolio authors and invest however you feel most comfortable.  Just like you dismissed cFIREsim and the Trinity study and put your money in international stocks with a VPW.  ;)  Seriously -- I have no problem with your plan.  Different people require different portfolios, and clearly you've found a good one for you. 
« Last Edit: December 04, 2016, 06:03:44 PM by Tyler »

boarder42

  • Walrus Stache
  • *******
  • Posts: 7849
Re: Target AA once fired
« Reply #27 on: December 04, 2016, 05:39:25 PM »
IC how did you come to a 50/50 us to int AA. how does it compare to a 90/10 us stock to bond

boarder42

  • Walrus Stache
  • *******
  • Posts: 7849
Re: Target AA once fired
« Reply #28 on: December 04, 2016, 05:44:29 PM »
Tyler why does your data only go back to 72. Seems to limit the sample size considerably

Tyler

  • Handlebar Stache
  • *****
  • Posts: 1160
Re: Target AA once fired
« Reply #29 on: December 04, 2016, 05:47:47 PM »
Tyler why does your data only go back to 72. Seems to limit the sample size considerably

https://portfoliocharts.com/withdrawal-rates-faq/

Interest Compound

  • Pencil Stache
  • ****
  • Posts: 658
Re: Target AA once fired
« Reply #30 on: December 04, 2016, 06:04:47 PM »
IC how did you come to a 50/50 us to int AA. how does it compare to a 90/10 us stock to bond

I've purposely not discussed my personal portfolio (well, outside of the first post where I answered your question), because it's not relevant to any of my recent posts. I wish Tyler and Classical_Liberal weren't aware of my personal asset allocation, as I believe it'd result in a more constructive conversation where they wouldn't think I'm just trying to defend my 100% stock portfolio and justify my personal choices.

That said, I think that line of conversation has come to a close, so sure, we can discuss my non-relevant personal asset allocation :) I think I can boil it down to two schools of thought:
  • Passive investing has a deep humility at its core--the aim is not to separate winners from losers, but rather to hold the entire market. (Why I only hold total market funds, and hold international at market-weight)
  • Volatility is only temporary, but you can permanently cripple your portfolio trying to avoid it. (Why I'm 100% stocks)
How does it compare to 90/10? 90/10 will likely have less volatility, as bonds are inherently less volatile. Not because backtesting tells you it's less volatile, but because of what bonds inherently represent. If your 90/10 doesn't include international, then you're forgoing about half of the world. I'll spare you my rant on International stocks :) I'll just say I can't think of a good reason to justify that.

arebelspy

  • Administrator
  • Senior Mustachian
  • *****
  • Posts: 28239
  • Age: -999
  • Location: Seattle, WA
Re: Target AA once fired
« Reply #31 on: December 04, 2016, 06:17:28 PM »
The changes and additions to your site are awesome.

They have all been, up until the recent data obfuscation.  :(


(Yes, I read the reasoning behind it. No, I did not find it compelling.  Unless there is more Tyler is not saying.)
We are two former teachers who accumulated a bunch of real estate, retired at 29, spent some time traveling the world full time and are now settled with three kids.
If you want to know more about us, or how we did that, or see lots of pictures, this Business Insider profile tells our story pretty well.
We (rarely) blog at AdventuringAlong.com. Check out our Now page to see what we're up to currently.

brooklynguy

  • Handlebar Stache
  • *****
  • Posts: 2207
  • Age: 39
Re: Target AA once fired
« Reply #32 on: December 05, 2016, 10:28:08 AM »
The changes and additions to your site are awesome.

They have all been, up until the recent data obfuscation.  :(


(Yes, I read the reasoning behind it. No, I did not find it compelling.  Unless there is more Tyler is not saying.)

Ditto.  I didn't even really understand the reasoning (available here), let alone not find it compelling, and would be grateful if Tyler would be willing to elaborate, either here in this thread or somewhere else if that's more appropriate.

Tyler

  • Handlebar Stache
  • *****
  • Posts: 1160
Re: Target AA once fired
« Reply #33 on: December 05, 2016, 12:00:41 PM »
Short story -- I've been doing a lot of work behind the scenes researching data sources and getting feedback from people who do this kind of thing for a living.  While I also dislike limiting information, the goal is to avoid stepping on any legal toes so that I can continue doing what I enjoy.  That said, the changes are not final and I'm researching ways to responsibly minimize the impact.

If you have any more questions, please contact me via PM.  It's complicated, and I'd rather not get into too much detail.  Thanks for caring enough to ask!  :)

Interest Compound

  • Pencil Stache
  • ****
  • Posts: 658
Re: Target AA once fired
« Reply #34 on: December 05, 2016, 10:14:06 PM »
So nobody has to make up portfolios themselves and hope to get lucky -- simply follow the advice of smart people like Bernstein, Swensen, and Browne and you'll already be on the right track.

Ok, Browne is the only one listed whose portfolio has enough live data (data obtained only after the portfolio was published), to approach an early-retiree's possible investment horizon, so let's see if following his advice would've successfully put us on a track of reducing volatility while benefiting portfolio drawdown.

Here's a chart of someone retiring with the 4% rule, starting the year Browne's "Permanent Portfolio" was created, compared with US Stocks:

(Current year withdrawal: $106,371.40)




How about with a 6.5% withdrawal rate (current year withdrawal $172,853.52)?




For fun, Browne's "Permanent Portfolio" before it was created in 1981:

4% rule (1981 withdrawal: $86,985.65):



6.5% rule (1981 withdrawal: $141,351.67):



Again it seems if you're trying to pick "Subsets of these and different asset classes" that both reduce volatility and benefit portfolio drawdown, it definitely involves predicting future winners. Without knowing the future, this proves to be an incredibly difficult task. Even with Survivorship Bias working in his favor, Browne couldn't get it right. What chance do I have?

Interest Compound

  • Pencil Stache
  • ****
  • Posts: 658
Re: Target AA once fired
« Reply #35 on: December 05, 2016, 10:21:44 PM »
I will simply state that based on the best data available, each of the portfolios on my site (including the Classic 60-40 and Three-Fund portfolio) have supported higher SWRs than the US stock market alone.

All possible 50 year retirements using all data from 1871 with the 4% rule:




Please see my previous posts in this thread for more examples.
« Last Edit: December 05, 2016, 10:24:50 PM by Interest Compound »

Tyler

  • Handlebar Stache
  • *****
  • Posts: 1160
Re: Target AA once fired
« Reply #36 on: December 06, 2016, 01:01:53 AM »
I appreciate your feedback, Interest Compound.

For anyone interested, the PC Retirement Spending calculator runs the same analysis that Interest Compound did for the PP but does it for every start year simultaneously (not just two) and even allows you to pick different portfolios and withdrawal methods.  That will give you a nice picture of the full range of options and outcomes and you can decide for yourself what makes sense.

It's absolutely true that for 50-year retirements 100% US stocks worked best with a constant dollar withdrawal method when there are no other options than intermediate treasuries.  While I can't offer direct 50-year comparisons, it's also true that other portfolios supported higher perpetual withdrawal rates that can last just as long.   Browse the various options and you'll see what I mean.  Of course not every portfolio may make sense for you and you may even completely disagree with a few, but perhaps one will resonate.  Actually -- take the time to plug Interest Compound's portfolio into the calculators to see how that has done as well.  Maybe that's the best option for you, too. 

I hear the objection to identifying good portfolios ahead of time.  I respectfully disagree.  It's not about being "best" but being consistently good enough, and some portfolios are absolutely more consistent than others.  I personally choose to invest in more than just stocks and am a very happy early retiree.  If anyone would like more information feel free to PM me, as at this point I prefer to share than to argue the same points back and forth.  No matter what direction you choose and what reason you choose it, I wish everyone a happy and long retirement!  :)

DavidAnnArbor

  • Handlebar Stache
  • *****
  • Posts: 2176
  • Age: 54
  • Location: Ann Arbor, Michigan
Re: Target AA once fired
« Reply #37 on: December 06, 2016, 07:21:15 AM »
thank you for the Socratic method of dialogue

boarder42

  • Walrus Stache
  • *******
  • Posts: 7849
Re: Target AA once fired
« Reply #38 on: December 06, 2016, 07:26:17 AM »
the time line of tyler's data is what concerns me (yes i read your write up) but picking an AA based solely on 40 years or so of data and saying hey it worked for the last 40 why not the next is where i get turned around.  however you tool is quite useful to play with and see that steady withdrawal is possible with different AA's. just wish the data went back farther

Classical_Liberal

  • Handlebar Stache
  • *****
  • Posts: 1151
  • Age: 43
Re: Target AA once fired
« Reply #39 on: December 06, 2016, 08:20:42 AM »
the time line of tyler's data is what concerns me (yes i read your write up) but picking an AA based solely on 40 years or so of data and saying hey it worked for the last 40 why not the next is where i get turned around.  however you tool is quite useful to play with and see that steady withdrawal is possible with different AA's. just wish the data went back farther

I agree with Boarder42 that the lack of longer term data is a weak point.  Also any portfolio with a heavy gold concentration (like Brown's) has the "one off" event of a changing structure of world reserve currency (USD) and the ability for Americans to again legally hold gold. This should be carefully measured.

What concerns also concerns me is the enormously long bull market of US equities and almost simultaneous bull of US Treasuries through most of the 80's and 90's.  That data impacts all post WWII (ie modern) 50 year simulations to a large degree. Is that type of long term multidecade growth (albeit with a few temporary bears) repeatable going forward?  I'm not anymore convinced of that as I am of a repeat of gold performance in the early 70's. Just a thought.... One that lead me to slightly alter my portfolio plans after more detailed research.

Interest Compound

  • Pencil Stache
  • ****
  • Posts: 658
Re: Target AA once fired
« Reply #40 on: December 06, 2016, 09:20:48 AM »
I hear the objection to identifying good portfolios ahead of time.  I respectfully disagree.  It's not about being "best" but being consistently good enough, and some portfolios are absolutely more consistent than others.

The whole conversation is regarding the goal to "reduce volatility and benefit portfolio drawdown". Not being "best" or "consistently good enough", and not debating our personal investment choices. What you and I personally choose to invest in, is irrelevant to this conversation.

I appreciate your feedback, Interest Compound.

For anyone interested, the PC Retirement Spending calculator runs the same analysis that Interest Compound did for the PP but does it for every start year simultaneously (not just two) and even allows you to pick different portfolios and withdrawal methods.  That will give you a nice picture of the full range of options and outcomes and you can decide for yourself what makes sense.

No. My whole point is that it's incredibly difficult to predict the future. My whole point is that using hindsight to create a portfolio almost never works if you're trying to repeat the past. Pointing people to a calculator that mixes hindsight (where the Permanent Portfolio does amazingly), with actual performance (where it flops), does not accurately portray the experiences of anyone who has actually followed Browne's advice, taken their hard-earned-cash, and invested in this portfolio.

Pointing anyone interested in "reducing volatility and benefiting portfolio drawdown" to such information, inaccurately implies it has a good track record in accomplishing this task for people who have followed his advice...leading the reader to believe it's easy to predict the future, which is just wrong.

You want me to drop it? Add an option to all your calculators (and make it default), that only shows that portfolio's performance after the year it was created.

boarder42

  • Walrus Stache
  • *******
  • Posts: 7849
Re: Target AA once fired
« Reply #41 on: December 06, 2016, 09:56:19 AM »
is there any good data on mid cap and small cap stocks going all the way back over time. 

Interest Compound

  • Pencil Stache
  • ****
  • Posts: 658
Re: Target AA once fired
« Reply #42 on: December 06, 2016, 10:21:10 AM »
is there any good data on mid cap and small cap stocks going all the way back over time.

Firecalc has small cap going back to 1927, unfortunately no mid cap:



http://firecalc.com then click "Your Portfolio" up top.

boarder42

  • Walrus Stache
  • *******
  • Posts: 7849
Re: Target AA once fired
« Reply #43 on: December 06, 2016, 11:13:49 AM »
Has anyone looked at larger bond percentage and holding the more volatile yet much higher upside small caps. Using the binds to smooth it

Tyler

  • Handlebar Stache
  • *****
  • Posts: 1160
Re: Target AA once fired
« Reply #44 on: December 06, 2016, 01:37:26 PM »
Has anyone looked at larger bond percentage and holding the more volatile yet much higher upside small caps. Using the binds to smooth it

I recommend reading stuff by Larry Swedroe, as that's the exact type of portfolio strategy he writes about a lot.  "Reducing The Risk Of Black Swans" is a famous book on the subject.  Regarding retirement, you can look at his Swedroe Min Fat Tails portfolio using my numbers or model something kinda similar (but not exactly the same) with the custom Firecalc options.
« Last Edit: December 06, 2016, 01:39:06 PM by Tyler »

boarder42

  • Walrus Stache
  • *******
  • Posts: 7849
Re: Target AA once fired
« Reply #45 on: December 06, 2016, 01:58:44 PM »
man thats fun... thats what i wanted to see.  and what i wanted this conversation to spark.

standard
2MM
80k annual spend
40 years

so 100% S&P 500 fails 3 times
80smv/20LTT fails 1 time
60smv/20LTT fails 0 times never falls below 2MM and has huge upside.

i guess the question is why arent higher risk investments like this with low risk options proposed more

boarder42

  • Walrus Stache
  • *******
  • Posts: 7849
Re: Target AA once fired
« Reply #46 on: December 06, 2016, 02:02:03 PM »
i mean we get higher success less failure and more upside with data back to 1927.  which is a large sample size.

Interest Compound

  • Pencil Stache
  • ****
  • Posts: 658
Re: Target AA once fired
« Reply #47 on: December 06, 2016, 02:13:50 PM »
man thats fun... thats what i wanted to see.  and what i wanted this conversation to spark.

standard
2MM
80k annual spend
40 years

so 100% S&P 500 fails 3 times
80smv/20LTT fails 1 time
60smv/20LTT fails 0 times never falls below 2MM and has huge upside.

i guess the question is why arent higher risk investments like this with low risk options proposed more

This is precisely what we've been discussing. I'm repeating myself at this point, but since you're asking the same question...

Because these strategies require predicting future winners. Without knowing the future, this proves to be an incredibly difficult task. Choose wrong, and you're in trouble.

boarder42

  • Walrus Stache
  • *******
  • Posts: 7849
Re: Target AA once fired
« Reply #48 on: December 06, 2016, 02:20:51 PM »
man thats fun... thats what i wanted to see.  and what i wanted this conversation to spark.

standard
2MM
80k annual spend
40 years

so 100% S&P 500 fails 3 times
80smv/20LTT fails 1 time
60smv/20LTT fails 0 times never falls below 2MM and has huge upside.

i guess the question is why arent higher risk investments like this with low risk options proposed more

This is precisely what we've been discussing. I'm repeating myself at this point, but since you're asking the same question...

Because these strategies require predicting future winners. Without knowing the future, this proves to be an incredibly difficult task. Choose wrong, and you're in trouble.

i understand you cant know the future but all we are doing here is basing our investments off what has happened in the past.  even your investments are still based on what happened in the past if the stock markets for the world returned net 0% over the last 100 years you wouldnt have your strategy either. correct?

boarder42

  • Walrus Stache
  • *******
  • Posts: 7849
Re: Target AA once fired
« Reply #49 on: December 06, 2016, 02:24:50 PM »
so to that effect why not hold all 4 asset classes equally with some bonds.  small, small value, large, large value ... running this with 40% bonds has no failures in 40 years either. no failures and at 4.5% SWR