Kitces' "bond tent" to protect against sequence risk

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Post Reply
Topic Author
jmk
Posts: 579
Joined: Tue Nov 01, 2011 7:48 pm

Kitces' "bond tent" to protect against sequence risk

Post by jmk » Tue Jan 31, 2017 10:48 am

Good article by Michael Kitces on the "Portfolio Size Effect." Summary: As one nears retirement the majority of one's returns come from the portfolio itself rather than income contributions; this puts one in a vulnerable position from market drops prior to and in the first decade of retirement (sequence risk). So, contrary to most popular glide paths, one should allocate with a bond tent: increase one's bond allocation during ages 55-65 and then decrease bonds "back to normal" by age 79.

https://www.kitces.com/blog/managing-po ... -red-zone/


Image

jebmke
Posts: 10278
Joined: Thu Apr 05, 2007 2:44 pm

Re: Kitces' "bond tent" to protect against sequence risk

Post by jebmke » Tue Jan 31, 2017 10:59 am

I only skimmed the article. I don't fully understand the shape of the blue line. Why the sudden plunge at age 80? Some assumption about LTC perhaps?
When you discover that you are riding a dead horse, the best strategy is to dismount.

User avatar
CULater
Posts: 2652
Joined: Sun Nov 13, 2016 10:59 am
Location: Hic sunt dracones

Re: Kitces' "bond tent" to protect against sequence risk

Post by CULater » Tue Jan 31, 2017 11:04 am

Of course, a better way to do it is to follow a "safety first" policy. According to this, your goal would be to accumulate a relatively safe income "floor" constituted of TIPS, annuities, etc.(including your pension and SS payments) and if your portfolio has a surplus beyond that you can optionally invest that amount in riskier "growth" assets such as stocks. As you spend down your floor over time, then your equity allocation will naturally grow to a larger proportion of your total assets if you're not making withdrawals from it. This would look like a "tent" strategy.
On the internet, nobody knows you're a dog.

User avatar
Dutch
Posts: 1277
Joined: Thu Jun 27, 2013 2:12 pm

Re: Kitces' "bond tent" to protect against sequence risk

Post by Dutch » Tue Jan 31, 2017 11:37 am

Previous conversation about this topic
viewtopic.php?t=204677

Grt2bOutdoors
Posts: 22230
Joined: Thu Apr 05, 2007 8:20 pm
Location: New York

Re: Kitces' "bond tent" to protect against sequence risk

Post by Grt2bOutdoors » Tue Jan 31, 2017 11:48 am

CULater wrote:Of course, a better way to do it is to follow a "safety first" policy. According to this, your goal would be to accumulate a relatively safe income "floor" constituted of TIPS, annuities, etc.(including your pension and SS payments) and if your portfolio has a surplus beyond that you can optionally invest that amount in riskier "growth" assets such as stocks. As you spend down your floor over time, then your equity allocation will naturally grow to a larger proportion of your total assets if you're not making withdrawals from it. This would look like a "tent" strategy.
Correct. The words echoed by Taylor Larimore comes into play here "simplify". All these theories and articles are published for one purpose only - to create "news" and sell a person's services, there is nothing new here. It would be easier to say "if you can't afford to lose it then you should not be in the stock market" - but that will not bring new clients. The "tent" just obfuscates things and makes it look like the advisor or planner is earning their keep.
"One should invest based on their need, ability and willingness to take risk - Larry Swedroe" Asking Portfolio Questions

Theoretical
Posts: 1504
Joined: Tue Aug 19, 2014 10:09 pm

Re: Kitces' "bond tent" to protect against sequence risk

Post by Theoretical » Tue Jan 31, 2017 1:08 pm

https://www.kitces.com/blog/understandi ... d-decades/

Unless the bond tent is TIPS, and in light of the sequence of returns stuff in this article for extended drab returns, this seems somewhat contradictory. However, I do in general agree with a higher equity in retirement approach.

pkcrafter
Posts: 13894
Joined: Sun Mar 04, 2007 12:19 pm
Location: CA
Contact:

Re: Kitces' "bond tent" to protect against sequence risk

Post by pkcrafter » Tue Jan 31, 2017 1:17 pm

jebmke wrote:I only skimmed the article. I don't fully understand the shape of the blue line. Why the sudden plunge at age 80? Some assumption about LTC perhaps?
I don't know about the blue line, although it's supposed to represent portfolio value. So is Kitces suggesting that drawdown goes to 0 even with 60% equity? He is using 30 years of retirement, which is standard, but still I don't think it's a reasonable conclusion. Of course, if you didn't get off the throttle heading into retirement you could have a much more serious problem with bad sequence.

Paul
When times are good, investors tend to forget about risk and focus on opportunity. When times are bad, investors tend to forget about opportunity and focus on risk.

User avatar
ixohoxi
Posts: 131
Joined: Fri Nov 22, 2013 2:58 pm

Re: Kitces' "bond tent" to protect against sequence risk

Post by ixohoxi » Tue Jan 31, 2017 3:51 pm

jebmke wrote:I only skimmed the article. I don't fully understand the shape of the blue line. Why the sudden plunge at age 80? Some assumption about LTC perhaps?
I think he postulated a high withdrawal rate to illustrate a more dramatic difference in risk. He didn't spell out the assumptions for the graph.
Henceforth, content shall be my aim, and anticipation my joy. -Alfred Billings Street

User avatar
grap0013
Posts: 1885
Joined: Thu Mar 18, 2010 1:24 pm

Re: Kitces' "bond tent" to protect against sequence risk

Post by grap0013 » Tue Jan 31, 2017 6:21 pm

I see arguments for more stocks and arguments for less stocks in retirement. I'm just gonna save a moderate amount and probably retire with something like 30:20:50 stocks:alternatives:bonds and withdraw 4% nominal. Whatever that gives me I'm going to spend.
There are no guarantees, only probabilities.

heyyou
Posts: 3628
Joined: Tue Feb 20, 2007 4:58 pm

Re: Kitces' "bond tent" to protect against sequence risk

Post by heyyou » Tue Jan 31, 2017 10:08 pm

Spending a fixed real amount of the retirement day assets, without ever considering returns for the next 2-3 decades, has risk. Spending as though your retirement is equal to the worst case, has risk.

Retirees have lived within their means for 30 years prior to retirement. That needs to continue in retirement, by spending from recent portfolio value.

Northern Flicker
Posts: 5390
Joined: Fri Apr 10, 2015 12:29 am

Re: Kitces' "bond tent" to protect against sequence risk

Post by Northern Flicker » Wed Feb 01, 2017 2:06 am

It is based on this study which suggested a rising equity glide path in retirement to counter sequence of return risk.
Risk provides no guarantee of return.

randomguy
Posts: 8672
Joined: Wed Sep 17, 2014 9:00 am

Re: Kitces' "bond tent" to protect against sequence risk

Post by randomguy » Wed Feb 01, 2017 11:05 am

CULater wrote:Of course, a better way to do it is to follow a "safety first" policy. According to this, your goal would be to accumulate a relatively safe income "floor" constituted of TIPS, annuities, etc.(including your pension and SS payments) and if your portfolio has a surplus beyond that you can optionally invest that amount in riskier "growth" assets such as stocks. As you spend down your floor over time, then your equity allocation will naturally grow to a larger proportion of your total assets if you're not making withdrawals from it. This would look like a "tent" strategy.
You would need to show some math to convince me that you would have done better in the 1966-1981 time frame with your strategy. My impression is that it doesn't help one bit other than maybe sleeping at night. The cases where strategies like this do better, pretty much all of the gain is from reduction of spending, not better returns. I am also a bit suspect that this bond tent is going to suffer the same fate. The tent will reduce volatility (1973-1974, 2000-2002) during those years but if you are not getting good real returns, I am not sure how much difference you will see between this and a normal 60/40 type allocation.


You really need some plots of various 60/40 portfolios (over various time periods) and compare them to this strategy (which as the author points out has a ton of missing details) but my general impression from all of these studies is that you are looking at pretty small differences. You are not going from a 4% to 6% SWR by messing around with AA. To do that you would need to find some instrument that would give you a steady 2-3% real and that doesn't exist.

User avatar
Doc
Posts: 9411
Joined: Sat Feb 24, 2007 1:10 pm
Location: Two left turns from Larry

Re: Kitces' "bond tent" to protect against sequence risk

Post by Doc » Wed Feb 01, 2017 11:31 am

CULater wrote:Of course, a better way to do it is to follow a "safety first" policy. According to this, your goal would be to accumulate a relatively safe income "floor" constituted of TIPS, annuities, etc.(including your pension and SS payments) and if your portfolio has a surplus beyond that you can optionally invest that amount in riskier "growth" assets such as stocks. As you spend down your floor over time, then your equity allocation will naturally grow to a larger proportion of your total assets if you're not making withdrawals from it. This would look like a "tent" strategy.
That's exactly what I did. I established a ten year TIPS ladder to cover bare minimum expenses above pension & SS so I would have ten years for the market to recover. The intent was to roll it if there was no market crisis. It was partially depleted during the Lehman crisis because I did rebalance to maintain my original AA and now it is now almost rebuilt as a mixed nominal Treasury + TIPS ladder but with some double rungs as I am not willing to commit to buying tens in this market.
A scientist looks for THE answer to a problem, an engineer looks for AN answer and lawyers ONLY have opinions. Investing is not a science.

User avatar
galeno
Posts: 1640
Joined: Fri Dec 21, 2007 12:06 pm

Re: Kitces' "bond tent" to protect against sequence risk

Post by galeno » Wed Feb 01, 2017 6:04 pm

So. As a retired 60 y/o couple. If we are at 40/60. Assume an AWR = 4.0%

Assume a real net CAGR = 3.0% for stocks and 0% for FI. We spend FI for the next 10 years.

After 5 years our port will look like: $46/$40 = 54%/46%.

After 10 years our port will look like: $53/$20 = 73%/27%.

Would it not be wise to rebalance from equities during this 10 year period so to NOT be at 73%/27%?
Last edited by galeno on Tue Feb 07, 2017 11:43 am, edited 1 time in total.

Ari
Posts: 601
Joined: Sat May 23, 2015 6:59 am

Re: Kitces' "bond tent" to protect against sequence risk

Post by Ari » Thu Feb 02, 2017 4:43 am

galeno wrote:Would it not be wise to rebalance from equities during this 10 year period so to NOT be at 80%/20%?
Depends on a lot of factors. The "bonds first" approach, where you spend down bonds until you're 100% equity and then live on that, has historically performed better than the rebalancing strategy, but you never know about the future. I would personally feel comfortable with 80% equity in retirement (but you'd have to ask me again in a decade or two, as I'm far from retired today), but it's hardly something for everyone.
All in, all the time.

randomguy
Posts: 8672
Joined: Wed Sep 17, 2014 9:00 am

Re: Kitces' "bond tent" to protect against sequence risk

Post by randomguy » Sun Feb 05, 2017 4:01 pm

Ari wrote:
galeno wrote:Would it not be wise to rebalance from equities during this 10 year period so to NOT be at 80%/20%?
Depends on a lot of factors. The "bonds first" approach, where you spend down bonds until you're 100% equity and then live on that, has historically performed better than the rebalancing strategy, but you never know about the future. I would personally feel comfortable with 80% equity in retirement (but you'd have to ask me again in a decade or two, as I'm far from retired today), but it's hardly something for everyone.
They are not recommending spending bonds down to 0. The paper is suggesting going from say 60/40 to 30/70 and then back to 60/40. Does it work better? Who knows. They sure aren't providing enough info to suggest you get much different results than just holding say 50/50. Given the growth and shrinkage of the AA in bonds, I am guessing you rarely see big differences. Going conservative 10 year before retirement and 10 years after is going to cause a huge portfolio drag in most cases. How often you win (i.e. how often are stocks lower 10 years after you going conservative) in absolute dollars is hard to say.

Ari
Posts: 601
Joined: Sat May 23, 2015 6:59 am

Re: Kitces' "bond tent" to protect against sequence risk

Post by Ari » Mon Feb 06, 2017 2:02 am

randomguy wrote:They are not recommending spending bonds down to 0.
I know that, of course. My point was that if spending bonds down to zero is better than static rebalancing, then it stands to reason that spending bonds down to less than zero should perform somewhere in between.
All in, all the time.

Post Reply