2000-2018 retirement - sequence of returns

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
User avatar
nedsaid
Posts: 10676
Joined: Fri Nov 23, 2012 12:33 pm

Re: 2000-2018 retirement - sequence of returns

Post by nedsaid » Sun Aug 26, 2018 12:33 pm

randomizer wrote:
Sun Aug 26, 2018 11:04 am
HomerJ wrote:
Sun Aug 26, 2018 10:53 am
randomizer wrote:
Sun Aug 26, 2018 7:28 am
This doesn't tell me that I need (more) bonds. It just tells me what happened in one specific time interval. Nobody knows nothin', as they say. Sticking with 88:12 for now (still trying to accumulate).
Sure, this thread was about retirement and when one is withdrawing.

Nothing wrong with being heavy stocks or 100% stocks while accumulating.
Yeah, I know. You can strike my last sentence and still get my intended meaning though. You can construct almost any result you want by selecting the right dates. Many responders on this thread seem to be saying that it confirms for them the need to hold more bonds. But all it really does is say that bonds were great from 2000-2018. That's the specific conclusion. The general conclusion is that you can't predict the future.
My take on this is that bonds are less volatile and the returns more predictable than for stocks. Both the volatility and the returns from stocks less predictable than for bonds and unfavorable distributions of return for stocks can show up at precisely the wrong time for retirees. It seems better to have a large helping of bonds when heading into retirement, you never know when those awful bear markets will show up.
A fool and his money are good for business.

User avatar
Leif
Posts: 2499
Joined: Wed Sep 19, 2007 4:15 pm

Re: 2000-2018 retirement - sequence of returns

Post by Leif » Sun Aug 26, 2018 12:37 pm

B4Xt3r wrote:
Sun Aug 26, 2018 7:16 am
Thanks. I think this post needs to get a little more love than it has. This shows that 100% stock is the "safer" bet to me.
Remember this is not yearly returns. It is over an 18 year overlapping cycle for each data point. A lot can happen within that 18 years, particularly if you are making withdrawals. That was pointed out well at the beginning of this thread.
Last edited by Leif on Sun Aug 26, 2018 1:25 pm, edited 1 time in total.

garlandwhizzer
Posts: 2063
Joined: Fri Aug 06, 2010 3:42 pm

Re: 2000-2018 retirement - sequence of returns

Post by garlandwhizzer » Sun Aug 26, 2018 1:03 pm

Great post, Homer, thanks. I think one very important point to be made in addition to stock/bond mix is the enormous role that luck plays when it comes to picking the date to start retirement The 2000 - 2018 retirement started with the collapse of the tech bubble 2000 - 2003 and included the worst bear market since the Great Depression 2007 -2009. If both cases highly inflated asset bubbles (tech then real estate) burst and the first ten years was a "lost decade" for stocks where equity produced essentially no return at all while forced 4% withdrawals chewed away at declining portfolio assets. If on the other hand one had retired in 1990, 100% equity would have boomed for a full decade and the portfolio even with a 4% withdrawal rate would have been considerably larger in 2000 than when retirement started in 1990. The nice thing about having an equity/bond split is that during bull markets rebalancing tends to shift assets away from outperforming bubbles into less volatile assets. This may reduce long term returns but clearly it reduces volatility which is very important in retirement.

The question now is will the future going forward be more like a 1990 retirement or a 2000 retirement? One thing for certain is that we don't expect the robust bond market returns of 2000 - 2018 going forward. Less certain is what will happen with equities. Typically going forward after long bull markets like now, equity returns over the short/intermediate term are less robust. Typically going forward from a deep bear market equity returns going forward are quite robust. We're a long way from a dip bear market now. There is however a high level of uncertainty about future equity returns. It seems reasonable in retirement planning to have at least sufficient assets in bonds/cash/CDs to provide for living expenses for a significant period of years, more than long enough to ride out a long bad bear market and allow for stock recovery. The central thing to avoid in financial planning is selling equity into the depths of a bear market. 50/50, 60/40, or 70/30 will all work if you start with sufficient assets.

Garland Whizzer

curmudgeon
Posts: 1658
Joined: Thu Jun 20, 2013 11:00 pm

Re: 2000-2018 retirement - sequence of returns

Post by curmudgeon » Sun Aug 26, 2018 1:37 pm

siamond wrote:
Sun Aug 26, 2018 10:57 am
willthrill81 wrote:
Sun Aug 26, 2018 10:15 am
Yes, 2000 was definitely an outlier. Based on siamond's post, there were only six starting years since 1950 where 100% stock did not leave one with a larger real portfolio after 18 years of withdrawals than a 50/50 or 100% TBM portfolio.

But don't lose sight of the fact that the U.S. stock market, overall, has done very well since 1950, better than most other nations. The U.S. might repeat that performance, and it might not. What if the U.S. and the rest of the world's stock markets have relatively poor for 50 years? Would having a 50% allocation to bonds be adequate? I don't think that this scenario is likely at all, but I don't think that it's impossible either.
Yes, I agree. I would add that 18 years is kind of a weird time period to look at (I used it to be consistent with the OP, but it would be more proper to use 30 years or so). The OP's math is essentially a Safe Withdrawal Rate analysis, looking at the portfolio's balance trajectory and end value.
I really found this graph quite enlightening. The 18 year timeframe is interesting, because it doesn't follow the standard lengths used so often. It's useful to think of it from the perspective of someone who retired around age 60-65, and is now 78-83 and looking at their portfolio and future options as sort of a mid-course analysis. Do you annuitize? Stay the course? Maybe lifespan has caught up with you and it doesn't matter.

While the 100% bonds cases generally have the lowest values after 18 years, that's what you would generally expect, and many of the scenarios aren't necessarily bad from the very risk-averse perspective. My first instinct was to arbitrarily say that any scenario which had an ending value of less than $500K was "bad", while over $1M was "good". But if you just stuck your money away in bonds or equivalent, you might well expect to have spent down more than half of your initial (inflation-adjusted) value after 18 years. Being down to 30% or less (1950 and much of the 1960's) would be pretty uncomfortable, though.

It would be interesting to superimpose a graph of starting P/E ratios for the stocks, and starting TBM rates (or just 10-year T-note) for the bonds. The only significantly positive TBM scenarios came with pretty high starting interest rates. Current TBM rates much more resemble the 50's and early 60's. Maybe real rates would also be relevant. P/E relevance isn't quite so clear, as regulatory changes over time have shifted them to a degree.

User avatar
siamond
Posts: 4318
Joined: Mon May 28, 2012 5:50 am

Re: 2000-2018 retirement - sequence of returns

Post by siamond » Sun Aug 26, 2018 1:50 pm

curmudgeon wrote:
Sun Aug 26, 2018 1:37 pm
I really found this graph quite enlightening. The 18 year timeframe is interesting, because it doesn't follow the standard lengths used so often. It's useful to think of it from the perspective of someone who retired around age 60-65, and is now 78-83 and looking at their portfolio and future options as sort of a mid-course analysis.
Thanks. Your feedback is helpful to refine the decision to add this kind of graph in the next update of the Simba spreadsheet. I made it so that one can easily select the timeframe (30 years, 18 years, or whatever you want). I agree that a lot of research is focused on entire retirement periods, but heck, life goes on afterwards and one should re-assess the situation from time to time, and have proper tools to do so.
curmudgeon wrote:
Sun Aug 26, 2018 1:37 pm
It would be interesting to superimpose a graph of starting P/E ratios for the stocks, and starting TBM rates (or just 10-year T-note) for the bonds. The only significantly positive TBM scenarios came with pretty high starting interest rates. Current TBM rates much more resemble the 50's and early 60's. Maybe real rates would also be relevant. P/E relevance isn't quite so clear, as regulatory changes over time have shifted them to a degree.
I took a stab at studying the correlation between starting CAPE and Safe Withdrawal Rates in this blog post, extending past research from Michael Kitces. I didn't think at the time to extend the analysis to starting interest rates. Let me mull over this, this could be a nice winter project.

heyyou
Posts: 3196
Joined: Tue Feb 20, 2007 4:58 pm

Re: 2000-2018 retirement - sequence of returns

Post by heyyou » Sun Aug 26, 2018 2:01 pm

4% SWR with consistent annual inflation adjustments is so 1990s.
As mentioned far above, skip the inflation adjustments on down years, per Kitces and David Zolt's TPA.
Or McClung's:
Start with a 50/50 allocation, but spend only from bond funds until they are gone, if no opportunity for rebalancing at 20% growth of the retirement day value of the stock assets. 110% or 100% are okay but less optimal.
As with VPW, McClung also suggests spending a set % of the recent annual portfolio value. It helps to have lower necessary expenses, so the variable income is tolerable.

Using retirement day valuations to adjust the initial WD %, lower percentage if higher valuations, is almost common sense.

Each one of the above is the retiree living within his/hers means, not blindly overspending in spite of negative portfolio returns, as did the example's 2000 retiree.

McClung even tested his WDs by wrapping the 30 year data from 2014 back to 1926 to start over. His 2000 retiree had 15 years of our returns, then 15 years re-starting in 1926 (start of 2000 thru end of 1940).

User avatar
HomerJ
Posts: 11942
Joined: Fri Jun 06, 2008 12:50 pm

Re: 2000-2018 retirement - sequence of returns

Post by HomerJ » Sun Aug 26, 2018 4:29 pm

TravelforFun wrote:
Sun Aug 26, 2018 11:42 am
Original post ignores the fact that people will adjust their spending based on the growth of lack of growth in their portfolios. Also, people who start withdrawing from their portfolio before SS benefits kick in can reduce their portfolio withdrawal significantly after SS.

Finally, all the calculations I ran show you would never run out of money with a 3% SWR and my FI number is based on this rate.

TravelforFun
Everything you said correct.

3% worked, and starting with 4% but adjusting spending if necessary also worked.

If SS kicks in part way through your retirement, even 4% with 100% stocks starting in August 2000 would have done just fine.
The J stands for Jay

User avatar
HomerJ
Posts: 11942
Joined: Fri Jun 06, 2008 12:50 pm

Re: 2000-2018 retirement - sequence of returns

Post by HomerJ » Sun Aug 26, 2018 4:35 pm

garlandwhizzer wrote:
Sun Aug 26, 2018 1:03 pm
The question now is will the future going forward be more like a 1990 retirement or a 2000 retirement? One thing for certain is that we don't expect the robust bond market returns of 2000 - 2018 going forward. Less certain is what will happen with equities. Typically going forward after long bull markets like now, equity returns over the short/intermediate term are less robust. Typically going forward from a deep bear market equity returns going forward are quite robust. We're a long way from a dip bear market now. There is however a high level of uncertainty about future equity returns. It seems reasonable in retirement planning to have at least sufficient assets in bonds/cash/CDs to provide for living expenses for a significant period of years, more than long enough to ride out a long bad bear market and allow for stock recovery. The central thing to avoid in financial planning is selling equity into the depths of a bear market. 50/50, 60/40, or 70/30 will all work if you start with sufficient assets.
Great post. I especially agree that we shouldn't expect the same bond returns that we saw in 2000-2018. I would never suggest 100% bonds for anyone either.

I agree the trick is to have enough CDs/bonds/cash to get through the next bear market so you don't have to sell stocks that have dropped 50% to eat.

I plan to have a 3-year CD ladder, then the rest of my money 60/40 stocks/bonds I think.

Each year I'll cash out a CD, live on that money, and buy a new 3-year CD with the money generated from the 60/40 portfolio, rebalancing back to 60/40 (so selling the asset that did the best, and not touching the asset that did the worst that past year).
The J stands for Jay

User avatar
willthrill81
Posts: 6500
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: 2000-2018 retirement - sequence of returns

Post by willthrill81 » Sun Aug 26, 2018 4:39 pm

heyyou wrote:
Sun Aug 26, 2018 2:01 pm
4% SWR with consistent annual inflation adjustments is so 1990s.
I've never even heard a rumor of a person who actually implemented the '4% rule' to the letter. Virtually everyone reduces their withdrawals when their portfolio is suffering. And most increase their withdrawals during the good times (except for most Bogleheads, of course).

The '4% rule' and fixed withdrawal methods in general are merely useful starting or reference points for discussion and/or general insight to put people into the 'ballpark' of what's 'reasonable'. When someone asks "Do I have enough to retire" and they have 30x their annual spending, we can say that, based on extensive historical analysis, the answer is "very probably."
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings

pkcrafter
Posts: 13133
Joined: Sun Mar 04, 2007 12:19 pm
Location: CA
Contact:

Re: 2000-2018 retirement - sequence of returns

Post by pkcrafter » Sun Aug 26, 2018 7:36 pm

Nice work by Homer and all others who have contributed in the discussion. Caution, downer alert. :happy

Thoughts on risk. Risk being defined as having the money for something important when you need it.

High stock allocations may be OK during accumulation, but is 100% stock ever justified? I don't think so. At what time prior to retirement should we reduce risk, 5 years, 10?

Some might argue that if the situation is right and there is a low withdrawal rate, then risk reduction isn't needed. The charts we have been looking at go back to ~1945. Most investors may now think in terms of a severe crash as a 45% loss, but what happens if we go back to 1925? What if we call the 1929 event a once in a hundred year event?

Another concern is we are in the longest running bull market in history, and I would also add that this run is not quite 10 years in the running. In an investing life time, that's not very long. Is this making people nervous? No, it's making more new investors want to get in on it. We are seeing a lot of new members who desire allocations of 90-100% stock. On the other side there are posters asking whether we should do something about the current situation like get out of the market. And even more surprising, we are seeing a lot of posters who seem deeply worried about bonds, not stocks. Of course we should not get out of the market, but I am wondering is we should at least be aware of black swan events and Bernstein's shallow risk/deep risk.

Nassim Taleb's black swan events

http://www.visualcapitalist.com/black-s ... e-forever/

Here is Jason Zweig on Bernstein's Deep Risk. Sure, events like these are very remote, but investors need to be aware.

http://jasonzweig.com/shallow-risk-and- ... the-woods/

Here is an interesting investing recovery calculator

https://www.getsmarteraboutmoney.ca/cal ... -recovery/

My conclusion is stocks really are risky and the potential consequences need to be understood, and from that informed decisions can be made.
Last edited by pkcrafter on Mon Aug 27, 2018 8:00 am, edited 1 time in total.
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.

Random Walker
Posts: 3327
Joined: Fri Feb 23, 2007 8:21 pm

Re: 2000-2018 retirement - sequence of returns

Post by Random Walker » Sun Aug 26, 2018 8:54 pm

pkcrafter wrote:
Sun Aug 26, 2018 7:36 pm

High stock allocations may be OK during accumulation, but is 100% stock ever justified? I don't think so. At what time prior to retirement should we reduce risk, 5 years,
I think there is no fixed number of years. Instead I think one should get increasingly clear on realistic goals and distinguish needs from wants as he gets closer to retirement. Then, with an understanding of past returns, current valuations, future expected returns, start taking risk off the table as he sees fit. This seems more rational to me than a blind 1-2% less equity per year target date approach. With equity SD around 16-20%, seems to me that at times the investor should take more definitive action.
Currently there has been a record bull, portfolios have swelled likely more than investors expected, valuations are generous, and future expected returns modest. Not only is the Mean future expected return modest, the whole potential distribution of returns has shifted left. Might be a good time for someone within 5-10 years of retirement to take some substantial risk off the table. Strongly recommend William Bernstein’s short e book on lifecycle investing. Anyone within sight of retirement needs to read the very end. He discusses Murpjphy’s Law of Retirement. Also check out a thread here by the same name.

Dave

AlphaLess
Posts: 1061
Joined: Fri Sep 29, 2017 11:38 pm
Location: Kentucky

Re: 2000-2018 retirement - sequence of returns

Post by AlphaLess » Sun Aug 26, 2018 9:01 pm

garlandwhizzer wrote:
Sun Aug 26, 2018 1:03 pm
The question now is will the future going forward be more like a 1990 retirement or a 2000 retirement?
It could also be worse, as stocks are priced high, and bonds are priced high.
"You can get more with a kind word and a gun than with just a kind word." George Washington

AlphaLess
Posts: 1061
Joined: Fri Sep 29, 2017 11:38 pm
Location: Kentucky

Re: 2000-2018 retirement - sequence of returns

Post by AlphaLess » Sun Aug 26, 2018 9:06 pm

HomerJ wrote:
Sun Aug 26, 2018 4:35 pm
I agree the trick is to have enough CDs/bonds/cash to get through the next bear market so you don't have to sell stocks that have dropped 50% to eat.

I plan to have a 3-year CD ladder, then the rest of my money 60/40 stocks/bonds I think.
What is the proportion of assets held in the 3-year CD ladder?
Put another way, if you have a $1MM, 60% stock + 40% bond portfolio, then how much do you have in the CD ladder?
$50K
$100K?
$300K?
$500K?
"You can get more with a kind word and a gun than with just a kind word." George Washington

User avatar
HomerJ
Posts: 11942
Joined: Fri Jun 06, 2008 12:50 pm

Re: 2000-2018 retirement - sequence of returns

Post by HomerJ » Sun Aug 26, 2018 9:31 pm

AlphaLess wrote:
Sun Aug 26, 2018 9:06 pm
HomerJ wrote:
Sun Aug 26, 2018 4:35 pm
I agree the trick is to have enough CDs/bonds/cash to get through the next bear market so you don't have to sell stocks that have dropped 50% to eat.

I plan to have a 3-year CD ladder, then the rest of my money 60/40 stocks/bonds I think.
What is the proportion of assets held in the 3-year CD ladder?
Put another way, if you have a $1MM, 60% stock + 40% bond portfolio, then how much do you have in the CD ladder?
$50K
$100K?
$300K?
$500K?
12% in the 3-year CD ladder, I guess... 4% withdrawals. Cash out the coming-due CD, use that money for expenses that year, buy a new 3-year CD using stock/bond money, rebalancing.

So $120k of $1 million I guess, with $528k in stocks and $352k in bonds. (60/40 of $880k).

Maybe I should run the numbers how that would have worked out retiring in August 2000.
The J stands for Jay

AlohaJoe
Posts: 3930
Joined: Mon Nov 26, 2007 2:00 pm
Location: Saigon, Vietnam

Re: 2000-2018 retirement - sequence of returns

Post by AlohaJoe » Sun Aug 26, 2018 9:50 pm

siamond wrote:
Sun Aug 26, 2018 1:50 pm
I didn't think at the time to extend the analysis to starting interest rates. Let me mull over this, this could be a nice winter project.
It isn't quite the same topic but David Blanchett has a nice paper from 2015 called "Initial Conditions and Optimal Retirement Glide Paths". The paper is trying to add more insights to the argument about increasing vs decreasing equity glidepaths in retirement. He starts out by noting that models that don't take initial conditions into account can lead to misleading results.

He finds that
Increasing glide paths appear to perform best in higher-return environments, while decreasing glide paths perform better in lower-return environments.
So it isn't quite the same topic but it is another entry in the area of regime dependence or "initial conditions may dramatically affect whether a strategy looks good or not", which I think is generally an under-explored area of retirement research.

Along similar lines see the recent article "Warning: Stock and Bond Correlation Assumptions are Regime Dependent" which makes the case that stocks & bonds may appear uncorrelated when you take 100 year averages but that doesn't hold when you look at their correlations under various regimes. In particular, when bonds have negative real returns they become positively correlated, with a correlation of 41%.

marcopolo
Posts: 1280
Joined: Sat Dec 03, 2016 10:22 am

Re: 2000-2018 retirement - sequence of returns

Post by marcopolo » Sun Aug 26, 2018 10:07 pm

HomerJ wrote:
Sun Aug 26, 2018 4:35 pm

I plan to have a 3-year CD ladder, then the rest of my money 60/40 stocks/bonds I think.

Each year I'll cash out a CD, live on that money, and buy a new 3-year CD with the money generated from the 60/40 portfolio, rebalancing back to 60/40 (so selling the asset that did the best, and not touching the asset that did the worst that past year).
I am sure I am missing something, but it is not clear to me what you are accomplishing here.

If you always replenish the 3-year CD ladder from the best performing portion of your 60/40 AA, what does the CD ladder accomplish?
Aren't you just effectively withdrawing from the 60/40 portfolio.

To me it just seems like you have something more like a 50/50 portfolio, where part of the 50% fixed income happens to be in CDs, and you are never touching that?
Once in a while you get shown the light, in the strangest of places if you look at it right.

User avatar
HomerJ
Posts: 11942
Joined: Fri Jun 06, 2008 12:50 pm

Re: 2000-2018 retirement - sequence of returns

Post by HomerJ » Sun Aug 26, 2018 10:14 pm

marcopolo wrote:
Sun Aug 26, 2018 10:07 pm
HomerJ wrote:
Sun Aug 26, 2018 4:35 pm

I plan to have a 3-year CD ladder, then the rest of my money 60/40 stocks/bonds I think.

Each year I'll cash out a CD, live on that money, and buy a new 3-year CD with the money generated from the 60/40 portfolio, rebalancing back to 60/40 (so selling the asset that did the best, and not touching the asset that did the worst that past year).
I am sure I am missing something, but it is not clear to me what you are accomplishing here.

If you always replenish the 3-year CD ladder from the best performing portion of your 60/40 AA, what does the CD ladder accomplish?
Aren't you just effectively withdrawing from the 60/40 portfolio.

To me it just seems like you have something more like a 50/50 portfolio, where part of the 50% fixed income happens to be in CDs, and you are never touching that?
Good point.

Maybe if both stocks and bonds go down, I don't replenish the CD that year? :)

But yeah, think of it as a 50/50 portfolio with a mix of bonds and CDs on the fixed income side.
The J stands for Jay

User avatar
Leif
Posts: 2499
Joined: Wed Sep 19, 2007 4:15 pm

Re: 2000-2018 retirement - sequence of returns

Post by Leif » Sun Aug 26, 2018 11:05 pm

Deleted.
Last edited by Leif on Mon Aug 27, 2018 7:56 am, edited 1 time in total.

AlphaLess
Posts: 1061
Joined: Fri Sep 29, 2017 11:38 pm
Location: Kentucky

Re: 2000-2018 retirement - sequence of returns

Post by AlphaLess » Sun Aug 26, 2018 11:16 pm

HomerJ wrote:
Sun Aug 26, 2018 9:31 pm
12% in the 3-year CD ladder, I guess... 4% withdrawals. Cash out the coming-due CD, use that money for expenses that year, buy a new 3-year CD using stock/bond money, rebalancing.

So $120k of $1 million I guess, with $528k in stocks and $352k in bonds. (60/40 of $880k).

Maybe I should run the numbers how that would have worked out retiring in August 2000.
Looks like you have a 53% stock, and 47% bond / CD portfolio.

Questions:
- where are you going to get CD rates going back to 2000?
- shouldn't CD returns be lower than a diversified bond fund?
"You can get more with a kind word and a gun than with just a kind word." George Washington

User avatar
willthrill81
Posts: 6500
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: 2000-2018 retirement - sequence of returns

Post by willthrill81 » Mon Aug 27, 2018 12:27 am

AlphaLess wrote:
Sun Aug 26, 2018 11:16 pm
HomerJ wrote:
Sun Aug 26, 2018 9:31 pm
12% in the 3-year CD ladder, I guess... 4% withdrawals. Cash out the coming-due CD, use that money for expenses that year, buy a new 3-year CD using stock/bond money, rebalancing.

So $120k of $1 million I guess, with $528k in stocks and $352k in bonds. (60/40 of $880k).

Maybe I should run the numbers how that would have worked out retiring in August 2000.
Looks like you have a 53% stock, and 47% bond / CD portfolio.

Questions:
- where are you going to get CD rates going back to 2000?
https://www.bankrate.com/banking/cds/hi ... 1984-2016/

AlphaLess wrote:
Sun Aug 26, 2018 11:16 pm
- shouldn't CD returns be lower than a diversified bond fund?
Not necessarily. From the perspective of the retail investor, CDs can be a bit of a free lunch over the bond market. Larry Swedroe has said that retail investors may be better served with CDs than bonds.

For instance, the yield (and expected return for the next 10 years) of the TBM is currently about 3.0%. You can get five year CDs that pay as much as 3.5% right now.
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings

CurlyDave
Posts: 755
Joined: Thu Jul 28, 2016 11:37 am

Re: 2000-2018 retirement - sequence of returns

Post by CurlyDave » Mon Aug 27, 2018 1:22 am

Luckygirl wrote:
Sat Aug 25, 2018 1:31 pm
Would these results be similar if you were taking RMD’s starting in 2000? Every year the amount you have to take goes up.
There is no law that you must spend the RMD. You can withdraw and then re-invest after paying taxes on the withdrawal.

JustinR
Posts: 781
Joined: Tue Apr 27, 2010 11:43 pm

Re: 2000-2018 retirement - sequence of returns

Post by JustinR » Mon Aug 27, 2018 4:50 am

B4Xt3r wrote:
Sun Aug 26, 2018 7:16 am
siamond wrote:
Sat Aug 25, 2018 4:08 pm
Mostly for the sake of testing a new feature planned for the next update of the Simba backtesting spreadsheet, let me add my 2 cents...

I took Homer's basic parameters (18 years time period, compare 100% stocks with 100% bonds, then 50/50). $1M to start with, $40k fixed (inflation-adjusted) withdrawal at the beginning of the year. Only difference is that the chart below shows the portfolio end value in inflation-adjusted dollars instead of nominal dollars.

Then let's vary the start date (Jan 1st 1950, Jan 1st 1951,..., Jan 1st 2000). Here is the outcome (click to see a larger display). Starting year on the X axis, portfolio end value on the Y axis.

For sure, 100% stocks certainly had a couple of unpleasant outcomes, including the 2000 starting year. But 100% bonds had a lot more unpleasant outcomes (e.g. in a time period where interest rates went up). "Bonds are for safety"... ah, good one! Then yeah, a more balanced 50/50 portfolio would have been less traumatic, although in the late 60s, whatever you did, if you stuck to fixed withdrawals, you were going down the drain...

PS. if you're pondering what happened to the red point for starting years like 1950/51 or 1982/83, well, it's (literally) off the chart. I cut the chart at $5M for readability.

Thanks. I think this post needs to get a little more love than it has. This shows that 100% stock is the "safer" bet to me.
Except literally no one said you should go 100% bonds.

It doesn't matter if 100% stocks does great a lot of the times. That's expected.

Retirement isn't about doing great, it's about avoiding catastrophic results at all costs. So the answer, like everyone in this thread has been saying, is you need something in-between.

I think you took the wrong lesson from that post.

ignition
Posts: 202
Joined: Sun Dec 11, 2016 11:28 am

Re: 2000-2018 retirement - sequence of returns

Post by ignition » Mon Aug 27, 2018 5:16 am

pkcrafter wrote:
Sun Aug 26, 2018 7:36 pm
Some might argue that if the situation is right and there is a low withdrawal rate, then risk reduction isn't needed. The charts we have been looking at go back to ~1945. Most investors may now think in terms of a severe crash as a 45% loss, but what happens if we go back to 1925? What if we call the 1929 event a once in a hundred year event?
1929 wasn't that bad surprisingly enough. A fixed 4% inflation adjusted withdrawal rate would have depleted the portfolio after 25 years.

A 3% withdrawal rate would have worked fine for 30 years with an ending balance of 1.5M inflation adjusted.

pkcrafter
Posts: 13133
Joined: Sun Mar 04, 2007 12:19 pm
Location: CA
Contact:

Re: 2000-2018 retirement - sequence of returns

Post by pkcrafter » Mon Aug 27, 2018 8:48 am

Random Walker wrote:
Sun Aug 26, 2018 8:54 pm
pkcrafter wrote:
Sun Aug 26, 2018 7:36 pm

High stock allocations may be OK during accumulation, but is 100% stock ever justified? I don't think so. At what time prior to retirement should we reduce risk, 5 years,
I think there is no fixed number of years. Instead I think one should get increasingly clear on realistic goals and distinguish needs from wants as he gets closer to retirement. Then, with an understanding of past returns, current valuations, future expected returns, start taking risk off the table as he sees fit.

I think evaluating an investor's position, goals, needs, wants, (need and ability), will provide reasonable guidelines, but I also think that past returns, valuations, expected returns are too unreliable to use for short term decisions like deciding when to, and how much to fine tune risk reduction on the threshold of retirement.

This seems more rational to me than a blind 1-2% less equity per year target date approach. With equity SD around 16-20%, seems to me that at times the investor should take more definitive action.

Currently there has been a record bull, portfolios have swelled likely more than investors expected, valuations are generous, and future expected returns modest. Not only is the Mean future expected return modest, the whole potential distribution of returns has shifted left. Might be a good time for someone within 5-10 years of retirement to take some substantial risk off the table. Strongly recommend William Bernstein’s short e book on lifecycle investing. Anyone within sight of retirement needs to read the very end. He discusses Murpjphy’s Law of Retirement. Also check out a thread here by the same name.

Yes, this makes sense. The "lost decade," which HomerJ opened with, was a consequence of crazy valuations which ended with the NASDAQ losing 78% and the S&P lost 44%. Wild run ups can result is lengthy recoveries because of investors bad experiences and subsequent extremely defensive behavior.

Dave
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
HomerJ
Posts: 11942
Joined: Fri Jun 06, 2008 12:50 pm

Re: 2000-2018 retirement - sequence of returns

Post by HomerJ » Mon Aug 27, 2018 9:06 am

ignition wrote:
Mon Aug 27, 2018 5:16 am
pkcrafter wrote:
Sun Aug 26, 2018 7:36 pm
Some might argue that if the situation is right and there is a low withdrawal rate, then risk reduction isn't needed. The charts we have been looking at go back to ~1945. Most investors may now think in terms of a severe crash as a 45% loss, but what happens if we go back to 1925? What if we call the 1929 event a once in a hundred year event?
1929 wasn't that bad surprisingly enough. A fixed 4% inflation adjusted withdrawal rate would have depleted the portfolio after 25 years.

A 3% withdrawal rate would have worked fine for 30 years with an ending balance of 1.5M inflation adjusted.
And again, 4% worked for 50/50 portfolio during the Great Depression. Bonds did quite well to temper the stock market drop for an investor back then.

100% stocks at 4% withdrawal has failed a couple of times (but still lasted 20-25 years, so "failed" could be fixed with a few cut-backs in spending). 98% of time, you end up super rich. Far richer than the guys with bonds. Those are pretty good odds. But you also have to have nerves of steel.

And just because you had nerves of steel the last time we experienced a crash WHILE YOU STILL WORKING, that doesn't mean you'll have the same nerves of steel when it happens as you're pulling money out. Very different experiences.

But it looks like 4% (even at 100% stocks) with some flexibility in spending is very safe. 4% with some bonds may not require cut-backs in spending during a bad market. Or as ignition point out, 3% has always worked.

The real danger is a crash two days after you retire. Having SOME bonds/cash that you can spend for a couple of years while you wait for the market to recover seems prudent.

Or go back to work. May be hard in a recession. But at least you're still young at that point, and you were just working a few months ago, so it's not like your skills are outdated.
The J stands for Jay

actx
Posts: 17
Joined: Fri Mar 23, 2018 2:30 pm

Re: 2000-2018 retirement - sequence of returns

Post by actx » Mon Aug 27, 2018 9:24 am

Posts like this make me consider putting the majority of my retirement in an offering I have in my 401k and 401k excess from my employer. . .there is an option they provide for retirement that is a fund that returns 3.5% per year. It is "guaranteed" as far as that goes but during 2008/2009 . .3.5%. During 2000-2002. . .3.5%. For the last 35 years it has been avail it returns 3.5%.

So, what are your thoughts on removing a huge amount of risk and moving to this fund? I have just over $2M in those accounts which would throw off $70k/year without touching principal. I have another $1.5M in a very aggressive DFA portfolio.

User avatar
willthrill81
Posts: 6500
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: 2000-2018 retirement - sequence of returns

Post by willthrill81 » Mon Aug 27, 2018 9:30 am

actx wrote:
Mon Aug 27, 2018 9:24 am
Posts like this make me consider putting the majority of my retirement in an offering I have in my 401k and 401k excess from my employer. . .there is an option they provide for retirement that is a fund that returns 3.5% per year. It is "guaranteed" as far as that goes but during 2008/2009 . .3.5%. During 2000-2002. . .3.5%. For the last 35 years it has been avail it returns 3.5%.

So, what are your thoughts on removing a huge amount of risk and moving to this fund? I have just over $2M in those accounts which would throw off $70k/year without touching principal. I have another $1.5M in a very aggressive DFA portfolio.
It would certainly be an improvement over TBM, which is currently yielding 2.58%.
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings

pkcrafter
Posts: 13133
Joined: Sun Mar 04, 2007 12:19 pm
Location: CA
Contact:

Re: 2000-2018 retirement - sequence of returns

Post by pkcrafter » Mon Aug 27, 2018 9:40 am

ignition wrote:
Mon Aug 27, 2018 5:16 am
pkcrafter wrote:
Sun Aug 26, 2018 7:36 pm
Some might argue that if the situation is right and there is a low withdrawal rate, then risk reduction isn't needed. The charts we have been looking at go back to ~1945. Most investors may now think in terms of a severe crash as a 45% loss, but what happens if we go back to 1925? What if we call the 1929 event a once in a hundred year event?
1929 wasn't that bad surprisingly enough. A fixed 4% inflation adjusted withdrawal rate would have depleted the portfolio after 25 years.

A 3% withdrawal rate would have worked fine for 30 years with an ending balance of 1.5M inflation adjusted.
Are you using 4% wr from the original portfolio value or the value after the fall of 88%? I guess Bengen says the 4% wr was successful.

Article on swr

http://abovethecanopy.us/does-the-4-per ... till-work/

https://www.morningstar.com/videos/7336 ... awal-.html

https://retirementresearcher.com/asset- ... wal-rates/





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.

ignition
Posts: 202
Joined: Sun Dec 11, 2016 11:28 am

Re: 2000-2018 retirement - sequence of returns

Post by ignition » Mon Aug 27, 2018 9:49 am

HomerJ wrote:
Mon Aug 27, 2018 9:06 am
And again, 4% worked for 50/50 portfolio during the Great Depression. Bonds did quite well to temper the stock market drop for an investor back then.

100% stocks at 4% withdrawal has failed a couple of times (but still lasted 20-25 years, so "failed" could be fixed with a few cut-backs in spending). 98% of time, you end up super rich. Far richer than the guys with bonds. Those are pretty good odds. But you also have to have nerves of steel.

And just because you had nerves of steel the last time we experienced a crash WHILE YOU STILL WORKING, that doesn't mean you'll have the same nerves of steel when it happens as you're pulling money out. Very different experiences.

But it looks like 4% (even at 100% stocks) with some flexibility in spending is very safe. 4% with some bonds may not require cut-backs in spending during a bad market. Or as ignition point out, 3% has always worked.

The real danger is a crash two days after you retire. Having SOME bonds/cash that you can spend for a couple of years while you wait for the market to recover seems prudent.

Or go back to work. May be hard in a recession. But at least you're still young at that point, and you were just working a few months ago, so it's not like your skills are outdated.
Yes, I agree you would have to have nerves of steel in a 1929 scenario. I'm planning to hold 10-20% bonds in retirement and spend only 3% of my portfolio when there is a stock market crash in the first years of retirement.

Also remember 50/50 isn't totally safe. It would have failed for a few periods starting in the 60's. In fact 100% stocks had a slightly higher success rate than 50/50 according to cFiresim (95.73% vs 94.02%). Flexibility is important no matter which allocation you choose.

ignition
Posts: 202
Joined: Sun Dec 11, 2016 11:28 am

Re: 2000-2018 retirement - sequence of returns

Post by ignition » Mon Aug 27, 2018 9:52 am

pkcrafter wrote:
Mon Aug 27, 2018 9:40 am
ignition wrote:
Mon Aug 27, 2018 5:16 am
pkcrafter wrote:
Sun Aug 26, 2018 7:36 pm
Some might argue that if the situation is right and there is a low withdrawal rate, then risk reduction isn't needed. The charts we have been looking at go back to ~1945. Most investors may now think in terms of a severe crash as a 45% loss, but what happens if we go back to 1925? What if we call the 1929 event a once in a hundred year event?
1929 wasn't that bad surprisingly enough. A fixed 4% inflation adjusted withdrawal rate would have depleted the portfolio after 25 years.

A 3% withdrawal rate would have worked fine for 30 years with an ending balance of 1.5M inflation adjusted.
Are you using 4% wr from the original portfolio value or the value after the fall of 88%? I guess Bengen says the 4% wr was successful.

Article on swr

http://abovethecanopy.us/does-the-4-per ... till-work/

https://www.morningstar.com/videos/7336 ... awal-.html

https://retirementresearcher.com/asset- ... wal-rates/





Paul
I used 4% from the original portfolio value and used cFiresim to calculate.

Random Walker
Posts: 3327
Joined: Fri Feb 23, 2007 8:21 pm

Re: 2000-2018 retirement - sequence of returns

Post by Random Walker » Mon Aug 27, 2018 10:14 am

HomerJ wrote:
Mon Aug 27, 2018 9:06 am
.

And just because you had nerves of steel the last time we experienced a crash WHILE YOU STILL WORKING, that doesn't mean you'll have the same nerves of steel when it happens as you're pulling money out. Very different experiences.
WOW! I think that is one important and powerful statement! Im not yet retired. I’ve been proud of my steel nerves through 2000-2002 and 2007-2008, but I’ve been in accumulation phase. And I think we got a bit lucky with the rapid “V Shaped” recovery from 2007-2008. Steel nerves in withdrawal phase is a whole different league. That’s why most of us should cool off the portfolios.

Dave

garlandwhizzer
Posts: 2063
Joined: Fri Aug 06, 2010 3:42 pm

Re: 2000-2018 retirement - sequence of returns

Post by garlandwhizzer » Mon Aug 27, 2018 12:41 pm

actx wrote:

there is an option they provide for retirement that is a fund that returns 3.5% per year. It is "guaranteed" as far as that goes but during 2008/2009 . .3.5%. During 2000-2002. . .3.5%. For the last 35 years it has been avail it returns 3.5%.
The problem with this is that it does not take into account future inflation. All products which guarantee fixed nominal returns in the future provide merely the illusion of safety, not real safety. Inflation is now 2.9%, so this product currently offers a real return of 0.6% in inflation adjusted dollars which is less at present than either of Vanguard's TIPS funds offers. The difference is that TIPS offer future inflation protection which this does not. Of course no one knows what inflation will be in the future and most analysts believe it will be low by historical standards, somewhere around two percent, due to strong secular forces of globalization/technology. The majority of analysts are however frequently wrong about any future prediction and it seems unwise to me to put a lot of your eggs in your long term basket without protection from inflation. If you're planning for the long term you want real inflation adjusted returns, not nominal returns. The main attraction of investment products that offer fixed nominal returns in the future is more psychological than economic. It's okay IMO to put a small/modest portfolio allocation into such products as long as you have other assets that grow over time and keep up with inflation.

Garland Whizzer

actx
Posts: 17
Joined: Fri Mar 23, 2018 2:30 pm

Re: 2000-2018 retirement - sequence of returns

Post by actx » Mon Aug 27, 2018 1:26 pm

LUCK. . .this is the thing that most of us leave out. .

This guy at this site: http://www.retirementoptimizer.com

has written a very interesting retirement calculator and more importantly a 500 page book "Unveiling the Retirement Myth" which is a combination of experience, analysis (lots of analysis and models) and very eye opening. The book is $10 I think and is well worth it. Net/net the MOST important determination of a successful retirement is luck. . .(assuming you adequate $, etc).

Note, I have no affiliation with the author. . .just enjoyed his book.

User avatar
willthrill81
Posts: 6500
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: 2000-2018 retirement - sequence of returns

Post by willthrill81 » Mon Aug 27, 2018 1:30 pm

garlandwhizzer wrote:
Mon Aug 27, 2018 12:41 pm
actx wrote:

there is an option they provide for retirement that is a fund that returns 3.5% per year. It is "guaranteed" as far as that goes but during 2008/2009 . .3.5%. During 2000-2002. . .3.5%. For the last 35 years it has been avail it returns 3.5%.
The problem with this is that it does not take into account future inflation. All products which guarantee fixed nominal returns in the future provide merely the illusion of safety, not real safety. Inflation is now 2.9%, so this product currently offers a real return of 0.6% in inflation adjusted dollars which is less at present than either of Vanguard's TIPS funds offers. The difference is that TIPS offer future inflation protection which this does not. Of course no one knows what inflation will be in the future and most analysts believe it will be low by historical standards, somewhere around two percent, due to strong secular forces of globalization/technology. The majority of analysts are however frequently wrong about any future prediction and it seems unwise to me to put a lot of your eggs in your long term basket without protection from inflation. If you're planning for the long term you want real inflation adjusted returns, not nominal returns. The main attraction of investment products that offer fixed nominal returns in the future is more psychological than economic. It's okay IMO to put a small/modest portfolio allocation into such products as long as you have other assets that grow over time and keep up with inflation.

Garland Whizzer
So are you against TBM?
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings

User avatar
HomerJ
Posts: 11942
Joined: Fri Jun 06, 2008 12:50 pm

Re: 2000-2018 retirement - sequence of returns

Post by HomerJ » Mon Aug 27, 2018 1:31 pm

actx wrote:
Mon Aug 27, 2018 1:26 pm
LUCK. . .this is the thing that most of us leave out. .

This guy at this site: http://www.retirementoptimizer.com

has written a very interesting retirement calculator and more importantly a 500 page book "Unveiling the Retirement Myth" which is a combination of experience, analysis (lots of analysis and models) and very eye opening. The book is $10 I think and is well worth it. Net/net the MOST important determination of a successful retirement is luck. . .(assuming you adequate $, etc).

Note, I have no affiliation with the author. . .just enjoyed his book.
How much money we have throughout and at the end of our retirement is hugely dependent on luck. No one can accurately predict the next 30 years.

How SUCCESSFUL our retirement is (financially) can be far less dependent on luck with good planning.

We can retire right into a bear market (unlucky), but with proper planning, we can still be successful.
The J stands for Jay

marcopolo
Posts: 1280
Joined: Sat Dec 03, 2016 10:22 am

Re: 2000-2018 retirement - sequence of returns

Post by marcopolo » Mon Aug 27, 2018 1:51 pm

Random Walker wrote:
Mon Aug 27, 2018 10:14 am
HomerJ wrote:
Mon Aug 27, 2018 9:06 am
.

And just because you had nerves of steel the last time we experienced a crash WHILE YOU STILL WORKING, that doesn't mean you'll have the same nerves of steel when it happens as you're pulling money out. Very different experiences.
WOW! I think that is one important and powerful statement! Im not yet retired. I’ve been proud of my steel nerves through 2000-2002 and 2007-2008, but I’ve been in accumulation phase. And I think we got a bit lucky with the rapid “V Shaped” recovery from 2007-2008. Steel nerves in withdrawal phase is a whole different league. That’s why most of us should cool off the portfolios.

Dave
Agree. I think this is a very important observation. I similarly pat myself on the back about staying the course, and even rebalancing into equities in the 2000-2002 and 2007-2009 recessions. But, I was in accumulation mode and making very good income. I retired recently, and hope i will have similar fortitude when (not if) the severe down turn comes). But, i do recognize that the emotional aspects will be very different when i am withdrawing vs. accumulating. Recall the long thread started by Sheepdog that illustrates how difficult it can be to stay invested in the face of a dwindling portfolio.

To help with this, i have dialed my equity weighting down from 70% to about 50% in the last couple of years. Hopefully, that is sufficient to provide a cushion when the times comes. At some point, I may consider a rising equity glide path to get back to 70%, so i guess I am kind of doing a "bond tent" around the retirement date.
Once in a while you get shown the light, in the strangest of places if you look at it right.

pkcrafter
Posts: 13133
Joined: Sun Mar 04, 2007 12:19 pm
Location: CA
Contact:

Re: 2000-2018 retirement - sequence of returns

Post by pkcrafter » Mon Aug 27, 2018 2:15 pm

ignition wrote:
Mon Aug 27, 2018 9:52 am
pkcrafter wrote:
Mon Aug 27, 2018 9:40 am
ignition wrote:
Mon Aug 27, 2018 5:16 am
pkcrafter wrote:
Sun Aug 26, 2018 7:36 pm
Some might argue that if the situation is right and there is a low withdrawal rate, then risk reduction isn't needed. The charts we have been looking at go back to ~1945. Most investors may now think in terms of a severe crash as a 45% loss, but what happens if we go back to 1925? What if we call the 1929 event a once in a hundred year event?
1929 wasn't that bad surprisingly enough. A fixed 4% inflation adjusted withdrawal rate would have depleted the portfolio after 25 years.

A 3% withdrawal rate would have worked fine for 30 years with an ending balance of 1.5M inflation adjusted.
Are you using 4% wr from the original portfolio value or the value after the fall of 88%? I guess Bengen says the 4% wr was successful.

Article on swr

http://abovethecanopy.us/does-the-4-per ... till-work/

https://www.morningstar.com/videos/7336 ... awal-.html

https://retirementresearcher.com/asset- ... wal-rates/





Paul
I used 4% from the original portfolio value and used cFiresim to calculate.
:thumbsup
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.

pkcrafter
Posts: 13133
Joined: Sun Mar 04, 2007 12:19 pm
Location: CA
Contact:

Re: 2000-2018 retirement - sequence of returns

Post by pkcrafter » Mon Aug 27, 2018 2:20 pm

Random Walker wrote:
Mon Aug 27, 2018 10:14 am
HomerJ wrote:
Mon Aug 27, 2018 9:06 am
.

And just because you had nerves of steel the last time we experienced a crash WHILE YOU STILL WORKING, that doesn't mean you'll have the same nerves of steel when it happens as you're pulling money out. Very different experiences.
WOW! I think that is one important and powerful statement! Im not yet retired. I’ve been proud of my steel nerves through 2000-2002 and 2007-2008, but I’ve been in accumulation phase. And I think we got a bit lucky with the rapid “V Shaped” recovery from 2007-2008. Steel nerves in withdrawal phase is a whole different league. That’s why most of us should cool off the portfolios.

Dave
Yes, I've described it as suddenly finding yourself on a tightrope with no net below! When you have had income for over 30 years of working steady and suddenly it's not there any longer, you can feel a little vulnerable.


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.

actx
Posts: 17
Joined: Fri Mar 23, 2018 2:30 pm

Re: 2000-2018 retirement - sequence of returns

Post by actx » Mon Aug 27, 2018 2:41 pm

willthrill81 wrote:
Mon Aug 27, 2018 1:30 pm
garlandwhizzer wrote:
Mon Aug 27, 2018 12:41 pm
actx wrote:

there is an option they provide for retirement that is a fund that returns 3.5% per year. It is "guaranteed" as far as that goes but during 2008/2009 . .3.5%. During 2000-2002. . .3.5%. For the last 35 years it has been avail it returns 3.5%.
The problem with this is that it does not take into account future inflation. All products which guarantee fixed nominal returns in the future provide merely the illusion of safety, not real safety. Inflation is now 2.9%, so this product currently offers a real return of 0.6% in inflation adjusted dollars which is less at present than either of Vanguard's TIPS funds offers. The difference is that TIPS offer future inflation protection which this does not. Of course no one knows what inflation will be in the future and most analysts believe it will be low by historical standards, somewhere around two percent, due to strong secular forces of globalization/technology. The majority of analysts are however frequently wrong about any future prediction and it seems unwise to me to put a lot of your eggs in your long term basket without protection from inflation. If you're planning for the long term you want real inflation adjusted returns, not nominal returns. The main attraction of investment products that offer fixed nominal returns in the future is more psychological than economic. It's okay IMO to put a small/modest portfolio allocation into such products as long as you have other assets that grow over time and keep up with inflation.

Garland Whizzer
So are you against TBM?
My question also. . .why have any in TBM or similar when I can lock 3.5% for at least my bond/stable portfolio option.

User avatar
willthrill81
Posts: 6500
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: 2000-2018 retirement - sequence of returns

Post by willthrill81 » Mon Aug 27, 2018 2:44 pm

actx wrote:
Mon Aug 27, 2018 2:41 pm
willthrill81 wrote:
Mon Aug 27, 2018 1:30 pm
garlandwhizzer wrote:
Mon Aug 27, 2018 12:41 pm
actx wrote:

there is an option they provide for retirement that is a fund that returns 3.5% per year. It is "guaranteed" as far as that goes but during 2008/2009 . .3.5%. During 2000-2002. . .3.5%. For the last 35 years it has been avail it returns 3.5%.
The problem with this is that it does not take into account future inflation. All products which guarantee fixed nominal returns in the future provide merely the illusion of safety, not real safety. Inflation is now 2.9%, so this product currently offers a real return of 0.6% in inflation adjusted dollars which is less at present than either of Vanguard's TIPS funds offers. The difference is that TIPS offer future inflation protection which this does not. Of course no one knows what inflation will be in the future and most analysts believe it will be low by historical standards, somewhere around two percent, due to strong secular forces of globalization/technology. The majority of analysts are however frequently wrong about any future prediction and it seems unwise to me to put a lot of your eggs in your long term basket without protection from inflation. If you're planning for the long term you want real inflation adjusted returns, not nominal returns. The main attraction of investment products that offer fixed nominal returns in the future is more psychological than economic. It's okay IMO to put a small/modest portfolio allocation into such products as long as you have other assets that grow over time and keep up with inflation.

Garland Whizzer
So are you against TBM?
My question also. . .why have any in TBM or similar when I can lock 3.5% for at least my bond/stable portfolio option.
If you are guaranteed 3.5% and have full liquidity, that's certainly a better option than TBM these days, which is projected to return its yield, about 2.6%, over the next decade. Even if TBM had a slightly higher yield than 3.5%, the stable value fund might be a better option because of its lack of volatility.
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings

ColoRetiredGirl
Posts: 131
Joined: Mon Nov 13, 2017 11:40 pm
Location: Colorado

Re: 2000-2018 retirement - sequence of returns

Post by ColoRetiredGirl » Mon Aug 27, 2018 2:45 pm

Deleted
Last edited by ColoRetiredGirl on Mon Aug 27, 2018 3:04 pm, edited 1 time in total.

pkcrafter
Posts: 13133
Joined: Sun Mar 04, 2007 12:19 pm
Location: CA
Contact:

Re: 2000-2018 retirement - sequence of returns

Post by pkcrafter » Mon Aug 27, 2018 2:54 pm

My question also. . .why have any in TBM or similar when I can lock 3.5% for at least my bond/stable portfolio option.
Because it violates the fundamental rule of diversification.

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
willthrill81
Posts: 6500
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: 2000-2018 retirement - sequence of returns

Post by willthrill81 » Mon Aug 27, 2018 2:55 pm

ColoRetiredGirl wrote:
Mon Aug 27, 2018 2:45 pm
B4Xt3r wrote:
Sun Aug 26, 2018 9:40 am
HomerJ wrote:
Fri Aug 24, 2018 12:28 pm

50/50 seems like a good balance for retirement.
After seeing Siamond's post above, do you still stand by this statement?
Lurking here but there is a big difference between ‘retirement/retired’ and ‘ accumulating’ for retirement. I can easily see taking on more risk e.g. 88/12 with a long horizon but once retired you need, or at least, I would play if safe with the 50/50 AA.
Once many folks leave their primary career, reentering it down the line due to poor portfolio performance can be difficult or impossible. This leads many to be 'extra' conservative. OTOH, poor portfolio performance would be very unlikely to require you to need all of the income you made prior to retirement. For instance, if you need $50k for expenses, taking a $25k job would enable you to cut your portfolio withdrawals in half, which could be a huge help.

No answers here.
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings

User avatar
willthrill81
Posts: 6500
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: 2000-2018 retirement - sequence of returns

Post by willthrill81 » Mon Aug 27, 2018 2:58 pm

pkcrafter wrote:
Mon Aug 27, 2018 2:54 pm
My question also. . .why have any in TBM or similar when I can lock 3.5% for at least my bond/stable portfolio option.
Because it violates the fundamental rule of diversification.

Paul
Maybe, but we don't diversify just for the sake of diversification. If a single entity that I had great confidence in would guarantee me the returns I need to achieve my goals (i.e. TIPS), why bother with diversification? Further, we don't worry about diversification if someone wants to put all of their bond holdings into U.S. Treasuries.

I'm not saying that anyone should put all of their fixed income holdings into a single company, but there's a lot to be said for a guaranteed return.
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings

User avatar
Will do good
Posts: 701
Joined: Fri Feb 24, 2012 8:23 pm

Re: 2000-2018 retirement - sequence of returns

Post by Will do good » Mon Aug 27, 2018 3:58 pm

Thank you HomerJ for these charts and discussion. I feel good with my 60/40 portfolio.

Lindyhopper
Posts: 10
Joined: Wed Aug 08, 2018 3:34 pm

Re: 2000-2018 retirement - sequence of returns

Post by Lindyhopper » Mon Aug 27, 2018 6:56 pm

TY OP for this.

AlphaLess
Posts: 1061
Joined: Fri Sep 29, 2017 11:38 pm
Location: Kentucky

Re: 2000-2018 retirement - sequence of returns

Post by AlphaLess » Mon Aug 27, 2018 8:28 pm

willthrill81 wrote:
Mon Aug 27, 2018 12:27 am
AlphaLess wrote:
Sun Aug 26, 2018 11:16 pm
HomerJ wrote:
Sun Aug 26, 2018 9:31 pm
12% in the 3-year CD ladder, I guess... 4% withdrawals. Cash out the coming-due CD, use that money for expenses that year, buy a new 3-year CD using stock/bond money, rebalancing.

So $120k of $1 million I guess, with $528k in stocks and $352k in bonds. (60/40 of $880k).

Maybe I should run the numbers how that would have worked out retiring in August 2000.
Looks like you have a 53% stock, and 47% bond / CD portfolio.

Questions:
- where are you going to get CD rates going back to 2000?
https://www.bankrate.com/banking/cds/hi ... 1984-2016/

AlphaLess wrote:
Sun Aug 26, 2018 11:16 pm
- shouldn't CD returns be lower than a diversified bond fund?
Not necessarily. From the perspective of the retail investor, CDs can be a bit of a free lunch over the bond market. Larry Swedroe has said that retail investors may be better served with CDs than bonds.

For instance, the yield (and expected return for the next 10 years) of the TBM is currently about 3.0%. You can get five year CDs that pay as much as 3.5% right now.
I agree with everything you said

But OPs strategy has a 3-year CD ladderer, which will have a 1.5 year or less duration at any time.

Also, in hindsight, it might be easy to datamine for the 'sweet-spot' CD rates, but in real time, it is harder to do.

When back-testing a strategy, you can not pick the highest rate.

And how many times did your CD back-test hit a bank that went under. So even though you got your principal + earned interest back, you lost the ability to continue investing that that sweet rate.
"You can get more with a kind word and a gun than with just a kind word." George Washington

User avatar
willthrill81
Posts: 6500
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: 2000-2018 retirement - sequence of returns

Post by willthrill81 » Mon Aug 27, 2018 9:07 pm

AlphaLess wrote:
Mon Aug 27, 2018 8:28 pm
willthrill81 wrote:
Mon Aug 27, 2018 12:27 am
AlphaLess wrote:
Sun Aug 26, 2018 11:16 pm
HomerJ wrote:
Sun Aug 26, 2018 9:31 pm
12% in the 3-year CD ladder, I guess... 4% withdrawals. Cash out the coming-due CD, use that money for expenses that year, buy a new 3-year CD using stock/bond money, rebalancing.

So $120k of $1 million I guess, with $528k in stocks and $352k in bonds. (60/40 of $880k).

Maybe I should run the numbers how that would have worked out retiring in August 2000.
Looks like you have a 53% stock, and 47% bond / CD portfolio.

Questions:
- where are you going to get CD rates going back to 2000?
https://www.bankrate.com/banking/cds/hi ... 1984-2016/

AlphaLess wrote:
Sun Aug 26, 2018 11:16 pm
- shouldn't CD returns be lower than a diversified bond fund?
Not necessarily. From the perspective of the retail investor, CDs can be a bit of a free lunch over the bond market. Larry Swedroe has said that retail investors may be better served with CDs than bonds.

For instance, the yield (and expected return for the next 10 years) of the TBM is currently about 3.0%. You can get five year CDs that pay as much as 3.5% right now.
I agree with everything you said

But OPs strategy has a 3-year CD ladderer, which will have a 1.5 year or less duration at any time.

Also, in hindsight, it might be easy to datamine for the 'sweet-spot' CD rates, but in real time, it is harder to do.

When back-testing a strategy, you can not pick the highest rate.

And how many times did your CD back-test hit a bank that went under. So even though you got your principal + earned interest back, you lost the ability to continue investing that that sweet rate.
We cannot know the 'sweet spot' for anything without the benefit of hindsight.

There are several banks offering 3.0% for three year CDs right now. Further, it is always worthwhile to shop around for the best current CD rates when making a new deposit. The best offer next month is often different from the best this month.
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings

User avatar
willthrill81
Posts: 6500
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: 2000-2018 retirement - sequence of returns

Post by willthrill81 » Mon Aug 27, 2018 9:21 pm

Since we're looking at 18 year withdrawal periods, I thought it would be worthwhile to look at the oldest 18 year period available in Portfolio Visualizer, 1972-1990. This demonstrates that the 'bonds are for safety' idea has been far from always true.

Portfolio 1 is 100% U.S. TSM, portfolio 2 is 100% intermediate-term Treasuries (TBM doesn't go that far back, and ITT is a decent proxy for TBM), and portfolio 3 is 50/50 of the two. The portfolio begins with $10k and uses '4% rule' withdrawals (i.e. 4% of initial starting balance adjusted subsequently for inflation every year) and annual rebalancing.

Image

The ITT portfolio certainly wasn't as volatile as the TSM portfolio, which was down a whopping 53.5% in real dollars by the end of 1974. But by Oct., 1981, the ITT portfolio was down almost 61% in real dollars, after fewer than ten years of withdrawals. That being said, all three of the portfolios were in deep drawdowns by the end of 1981. By the end of 1990, the TSM and 50/50 portfolios were nearly in a dead heat ($544k and $549k, respectively, in 1972 dollars), while the ITT portfolio was down to just $374k in real dollars.

If we extended this 18 year period out to 30 years, we'd see that the ITT portfolio finished with just $100k of real dollars left (2.5 years of withdrawals), while the TSM portfolio had $1.04M (more than at the start!) and the 50/50 had $716k.

This illustrates that while bonds may not be as volatile as stocks, volatility alone is a very poor measure of risk, especially in the withdrawal phase. It also illustrates that even during a very turbulent first decade of withdrawals, a 50/50 portfolio that was rebalanced annually did very well, even though things looked pretty badly around 1982.
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings

garlandwhizzer
Posts: 2063
Joined: Fri Aug 06, 2010 3:42 pm

Re: 2000-2018 retirement - sequence of returns

Post by garlandwhizzer » Mon Aug 27, 2018 9:34 pm

willthrill81 wrote:

If you are guaranteed 3.5% and have full liquidity, that's certainly a better option than TBM these days, which is projected to return its yield, about 2.6%, over the next decade. Even if TBM had a slightly higher yield than 3.5%, the stable value fund might be a better option because of its lack of volatility.


There are several problems with stable value funds. Here's one from TheStreet.com:
Low Transparency

Stable value funds do not adhere to rigorous reporting rules like mutual funds or other investment products in 401(k) plans. They do not have to report their portfolio holdings nor all related costs, which sap total returns. And so a stable value fund with an advertised return of 3% may only yield 2%. One point may sound minuscule but it's 33% less yield than what investors expected to get.

Insurance companies have to reveal the management fees they take on stable value funds, but they don't have to disclose the profits they earn from the spread, which is the difference between the amount insurers earn on stable value funds' assets and the payment to plan participants. Insurance firms typically make about 200 basis points on the spread, Morningstar estimates. Their methods in disclosing fees is very inconsistent, making assessing them on an apples-to-apples basis for advisers and plan sponsors impossible. Stable value funds have to be analyzed on an individual basis rather than simply looking at returns and yields.
So in effect companies offering these products take their own management fees plus typically 200 basis points (2%) off the top before they give you the 3.5%. That means one of two things. On average they are finding fixed income products that are yielding over 5.5% (3.5 + 2). No such safe fixed income products exist in the marketplace today. In short there is hidden risk. They do not have to report what their holdings so you don't know what risk you're taking on. You really don't know where the 3.5% comes from whereas if you buy TBM you can go to the Vanguard website and find out exactly what its holdings are and how they are rated, i.e. what the risk level is.

In addition the 3.5% is fixed over time like an annuity. That may looks fine if rates fall in the future but what if rates rise in the future and you can buy 10 Treasuries with significantly higher yields? The average yield on 10 year Treasuries over the last 60 years is greater than 6%. Although currently it is widely accepted that low rates are here to stay for a long period of years that guarantees nothing whatsoever about future inflation or interest rates. When inflation and rising rates do hit big time there is typically one thing present. Everyone including the experts are surprised and caught off guard. Personally, I put little faith in anyone's predictions about the future of markets.

Fixed rate future yield promises are an attractive lure the financial industry uses to hook investors and line their own pockets. Annuities and stable value funds are examples of this IMO.

Garland Whizzer

Post Reply