Author Topic: Using the rising equity glidepath to reduce sequence of returns risk  (Read 6702 times)

dividendman

  • Handlebar Stache
  • *****
  • Posts: 1894
Is anyone doing this? An in depth description of what I'm talking about is here: https://www.kitces.com/blog/should-equity-exposure-decrease-in-retirement-or-is-a-rising-equity-glidepath-actually-better/

I'm considering going this route. Right now I'm 80/20 equities/bonds. However, I'm considering, once I pull the plug (within the next year... or OMY :( ) I would start with a 70/30 split, then shift my allocation, regardless of market activity, by 2% to stocks every year, until I got to a 90/10 allocation then stay there. This would probably mean spending the bond portion to live for the first 10 years unless the stock market had great returns or the bond markets tanked. Of course I would stay within the 4% rule. This would hopefully mitigate the sequence of returns risk for the first 10 years.

Is there any major problem with this that I'm overlooking?
« Last Edit: September 29, 2016, 04:21:19 PM by dividendman »

Jack

  • Magnum Stache
  • ******
  • Posts: 4725
  • Location: Atlanta, GA
Re: Using the rising equity glidepath to reduce sequence of returns risk
« Reply #1 on: September 29, 2016, 04:30:59 PM »
I've been meaning to start a thread on this idea, but never did (mainly because I haven't gotten around to simulating it yet). I'm thinking of doing a similar thing: maintain my current 100% equities until FIRE, then instantly drop down to 80/20 (or less), then slowly rise back to 100% equities after 10 years or so.

Clean Shaven

  • Pencil Stache
  • ****
  • Posts: 821
  • Location: Wild Wild West
Re: Using the rising equity glidepath to reduce sequence of returns risk
« Reply #2 on: September 29, 2016, 04:36:17 PM »
I'm interested in this concept too.  Read the Kitces article awhile back, and it makes sense.

I'm currently at around 80 (stock)/ 15 (bond)/ 5 (cash), but have been wondering whether to shift to a much more conservative initial allocation (say, 50 stock/ 45 bond/ 5 cash), and then do the "rising equity glide path) and adjust during the initial 10 years back up to an 80/20 (ish) split.

msilenus

  • Pencil Stache
  • ****
  • Posts: 524
Re: Using the rising equity glidepath to reduce sequence of returns risk
« Reply #3 on: September 29, 2016, 09:17:16 PM »
Some recent glidepath discussion:
http://forum.mrmoneymustache.com/investor-alley/changing-asset-allocation-as-fire-approaches/msg1243215/#msg1243215

dividendman, were you familiar with that thread already, or did you arrive at your 70/30 starting AA with a 10-year path from another direction?

dividendman

  • Handlebar Stache
  • *****
  • Posts: 1894
Re: Using the rising equity glidepath to reduce sequence of returns risk
« Reply #4 on: September 29, 2016, 11:07:03 PM »
Some recent glidepath discussion:
http://forum.mrmoneymustache.com/investor-alley/changing-asset-allocation-as-fire-approaches/msg1243215/#msg1243215

dividendman, were you familiar with that thread already, or did you arrive at your 70/30 starting AA with a 10-year path from another direction?

Thanks for the thread link! Very good info. No, I didn't read that thread first, so yes, I came from another direction. Basically I started with the 80/20 and always keeping that during RE. Then I figured I already have 5 years of living in my bonds (the 20%). Then I remembered that Mad Fientest blog entry where the 10 year correlation was pretty good at predicting the overall success rate, so, why not "glide" in a way that makes the first 10 years naturally take more of my bonds.

I just didn't see a (big) downside. Perhaps I should glide it all the way to 100/0 but I think I'd still want 10% in bonds just in case, so I ended up at 70/30 at the start of RE (which isn't too far off of my current 80/20 split). Basically I find it nice to skew my withdrawals to my bond portion until my portfolio has survived the dreaded first 10 years where things can really go south.

I have also considered other "rules". e.g. if the market goes up 15% over inflation (or some high number) then take from the stocks regardless of my "glide path" since my portfolio is more likely to survive and maybe do double the move on the glide path the next year (if required), because tilting more towards stocks after a huge gain doesn't seem logical either. So, I'm still throwing all of these things around.

I would like to have all of my rules established before I pull the fire plug so I don't convince myself I'm being logical later when I'm actually being emotional.

I do think I'm over thinking it though, cause if the market took a big prolonged shit I think I'd take more action than just my portfolio allocation because I'm a scaredy cat.

msilenus

  • Pencil Stache
  • ****
  • Posts: 524
Re: Using the rising equity glidepath to reduce sequence of returns risk
« Reply #5 on: September 29, 2016, 11:37:34 PM »
Weird.  Yeah, I got to 70/30:10 by just plugging various things into cFireSim until it wasn't getting any better.  I was expecting a higher bond starting point with shorter glide durations to be good, so my intuition was pretty far off relative to backtesting.  If 10 years is optimal (I have no idea if it is) then the way you came at it provides a potential explanation for why.    (The 70/30 coincidence seems like happenstance.)

IIRC, my results (which I don't totally trust --not very rigorous) suggested that 100/0 was a better endpoint than 90/10, but not by much.  I think there are theoretical reasons to argue for both some-bonds and for no-bonds.  I wouldn't try to talk you into 100/0 if you're leaning toward 90/10.

Jack

  • Magnum Stache
  • ******
  • Posts: 4725
  • Location: Atlanta, GA
Re: Using the rising equity glidepath to reduce sequence of returns risk
« Reply #6 on: September 30, 2016, 08:11:23 AM »
IIRC, my results (which I don't totally trust --not very rigorous) suggested that 100/0 was a better endpoint than 90/10, but not by much.  I think there are theoretical reasons to argue for both some-bonds and for no-bonds.  I wouldn't try to talk you into 100/0 if you're leaning toward 90/10.

Neither would I. I'm 100/0 on the theory that long-term returns are higher for stocks and I don't really care about volatility, but I occasionally wonder if a small bond allocation (5-10%) might increase my overall return by facilitating rebalancing.

Livewell

  • Stubble
  • **
  • Posts: 179
  • Location: SF Bay Area
Re: Using the rising equity glidepath to reduce sequence of returns risk
« Reply #7 on: September 30, 2016, 09:44:57 AM »
This is a very interesting method, thanks for sharing.  I've been thinking about this type approach but trying to determine the best starting and ending percentages for said buckets...  this helps.  I do think this model would be good mentally for me during those first few years of FIRE, knowing I wouldn't be touching stocks at all and having a plan in place to mitigate the early sequence of returns risk.  Will have to do some modeling around what works best for me, curious to see what others think!

From the article:  "Of course, in a scenario where a retiree is following a rising equity glidepath, and it turns out market returns are good – and the client suddenly winds up with far more wealth than ever anticipated – there’s always an option to change paths at that point, whether by bringing spending up or by taking equity risk off the table. Though technically they’ll be so far ahead that a rising equity glidepath isn’t a “risk” (i.e., even with a bear market, they’re far enough ahead with few enough years remaining that there’s no more failure risk), certainly some clients will decide that after a great bull market that changes their planning outlook and ability to achieve their goals, it may be prudent to get more conservative at that point. But for clients who go through a poor market in the early years, the rising equity glidepath becomes key to being able to sustain through the second half of retirement when the good returns finally arrive (in case the double-digit earnings yields on equities aren’t enticing enough at that point!). In other words, the rising equity glidepath becomes “heads you win, tails you don’t lose” (and in the latter scenarios, the client can change to whatever they want later instead)."

dude

  • Handlebar Stache
  • *****
  • Posts: 2369
Re: Using the rising equity glidepath to reduce sequence of returns risk
« Reply #8 on: September 30, 2016, 11:38:04 AM »
This is worth reading on the subject:

https://www.onefpa.org/journal/Pages/MAR15-Retirement-Risk,-Rising-Equity-Glide-Paths,-and-Valuation-Based-Asset.aspx

Takeaway to me seems to be a 60/40 allocation (stocks/T-bills) is a superior, and FAR simpler, way to go.

Northwestie

  • Handlebar Stache
  • *****
  • Posts: 1224
Re: Using the rising equity glidepath to reduce sequence of returns risk
« Reply #9 on: September 30, 2016, 01:19:00 PM »
I find this section of the site has some of the most practical - nut & bolts information.  Thanks all.

Livewell

  • Stubble
  • **
  • Posts: 179
  • Location: SF Bay Area
Re: Using the rising equity glidepath to reduce sequence of returns risk
« Reply #10 on: September 30, 2016, 03:18:08 PM »
Did some quick math for our situation, essentially:

We get to our 4% FIRE number (estimate 2020)
Right now we are about 95% stocks, we would move to 75% overall
5% would be cash
20% would be bonds
50% would be in taxable stock accounts
25% would be in non-taxable stock accounts (various IRA's, our 401K- to be rolled)

To pay for our spending the first 10 years, we would use a combination of cash (first), dividends from taxable stock account (assume 2%), interest from bonds (assume 2%), and finally bond equity.   Assuming appreciation of 2.3% of our taxable stocks and our IRA stocks (plus 2% dividends in these accounts), we would end up with the same amount of money we started with, plus have the bonus of mitigating the initial sequence of returns risk.   Of course this assumes we don't make any side money or reduce spending.   While there is no way the sequence of returns will be that linear, I have a lot of confidence it would work out over 10 years.   Likely it will be much higher and we'll have to spend more.   :^)

I like it.   This is a plan that will enable me to sleep at night in those initial years.   Now back to getting to the FIRE number.....   :^)

CanuckExpat

  • Magnum Stache
  • ******
  • Posts: 2994
  • Age: 41
  • Location: North Carolina
    • Freedom35
Re: Using the rising equity glidepath to reduce sequence of returns risk
« Reply #11 on: October 16, 2016, 08:44:46 PM »
I think we'll be doing this, partially out of simplicity. We have a bit of a cash buffer, I plan to spend out of that, and not rebalance. That should over time become a rising equity glidepath

steveo

  • Handlebar Stache
  • *****
  • Posts: 1928
Re: Using the rising equity glidepath to reduce sequence of returns risk
« Reply #12 on: October 17, 2016, 03:56:04 AM »
Did some quick math for our situation, essentially:

We get to our 4% FIRE number (estimate 2020)
Right now we are about 95% stocks, we would move to 75% overall
5% would be cash
20% would be bonds
50% would be in taxable stock accounts
25% would be in non-taxable stock accounts (various IRA's, our 401K- to be rolled)

To pay for our spending the first 10 years, we would use a combination of cash (first), dividends from taxable stock account (assume 2%), interest from bonds (assume 2%), and finally bond equity.   Assuming appreciation of 2.3% of our taxable stocks and our IRA stocks (plus 2% dividends in these accounts), we would end up with the same amount of money we started with, plus have the bonus of mitigating the initial sequence of returns risk.   Of course this assumes we don't make any side money or reduce spending.   While there is no way the sequence of returns will be that linear, I have a lot of confidence it would work out over 10 years.   Likely it will be much higher and we'll have to spend more.   :^)

I like it.   This is a plan that will enable me to sleep at night in those initial years.   Now back to getting to the FIRE number.....   :^)

This is basically what I intend to do. I have long service leave and I reckon that is going to be 2-3 years living expenses. Then we use cash and dividends/interest. Then bond equity. If the markets boom them we can pull out of our equity positions.

The only difference between us is that we are using 75% equities now until we reach FIRE. I also intend to FIRE in 2020.

dividendman

  • Handlebar Stache
  • *****
  • Posts: 1894
Re: Using the rising equity glidepath to reduce sequence of returns risk
« Reply #13 on: October 17, 2016, 10:31:21 AM »
This is worth reading on the subject:

https://www.onefpa.org/journal/Pages/MAR15-Retirement-Risk,-Rising-Equity-Glide-Paths,-and-Valuation-Based-Asset.aspx

Takeaway to me seems to be a 60/40 allocation (stocks/T-bills) is a superior, and FAR simpler, way to go.

Interesting article, thanks!

The historical data and their analysis seem to make a case for holding cash equivalents (T-bills) instead of bonds (to avoid interest rate risk).

Has this topic (cash equivalents vs. bonds for the non-equity portion of an asset allocation) been discussed before? (I couldn't find anything on this forum using Google)

The section I highlighted is contradictory. T-bills are not cash equivalents and are in fact just government bonds. Did you mean to say very short term T-bills (gov bonds) - they would behave a bit more like cash, but would still have some interest rate risk.

I'm holding bonds (BND) because historically they retain their purchasing power better than cash.

Livewell

  • Stubble
  • **
  • Posts: 179
  • Location: SF Bay Area
Re: Using the rising equity glidepath to reduce sequence of returns risk
« Reply #14 on: October 17, 2016, 10:45:56 AM »

This is basically what I intend to do. I have long service leave and I reckon that is going to be 2-3 years living expenses. Then we use cash and dividends/interest. Then bond equity. If the markets boom them we can pull out of our equity positions.

The only difference between us is that we are using 75% equities now until we reach FIRE. I also intend to FIRE in 2020.

Glad that is making sense to you too. 

One other thing I forgot to mention is my current savings plan, which is to continue to fund taxable and 401k accounts with equities then fund the bond portion.  My thinking is this would provide even more time before we need the money from taxable equities and provide a little adjustment time as the Fed raises rates before buying bonds.