Sequence of returns risk while accumulating

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
User avatar
Topic Author
willthrill81
Posts: 17355
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Sequence of returns risk while accumulating

Post by willthrill81 » Mon Feb 19, 2018 5:08 pm

Sequence of returns risk (SRR), the risk that the particular sequence of returns that you experience as an investor will be disadvantageous to you achieving your investment goals, is something that most investors on this forum understand well. It is primarily due to SRR that the safe withdrawal rate for U.S. retirees has been identified as being around 4%, despite stocks' returning around 7% in real dollars over the period being investigated. If a big bear market hits early in one's retirement and doesn't rebound quickly, the retiree who doesn't adjust her spending accordingly will be 'damaging' the portfolio further and may reduce the portfolio's size to a point where there isn't enough left when the market eventually recovers. That's why the safe withdrawal rate research always indicates that this rate is lower than the long-term rate of return of volatile asset classes like stocks. Note that if there were no volatility to returns, then SRR would not exist.

It is often said that SRR is mainly a risk for retirees, and there is some truth to that. Once a person has retired, it may be difficult or impossible for them to reduce their portfolio withdrawals and/or to return to their prior work in order to make up for a shortfall in their income caused by them experiencing SRR. I won't get into the ways that SRR can be dealt with here as that has been addressed in many other threads.

What seems lost on many, though, is that SRR is also very real for those who are still in the accumulation phase (i.e. building their portfolio before retirement). Most of the investment calculators out there demonstrate how much you would have if you invested X dollars over Y time period at Z rate of return. The problem is that in the real world, returns are not constant, giving rise to SRR for accumulators.

To illustrate this point, below is a chart created in Portfolio Visualizer. It illustrates the result of a Monte Carlo simulation of a portfolio comprised of 60% U.S. total stock market and 40% intermediate term Treasuries with the initial portfolio being worth $1,000 and $1,000 invested monthly for twenty years and with the portfolio rebalanced annually. Historic returns from 1972-2017 were used to derive both the mean returns and standard deviation; the same was done for inflation. Monte Carlo simulators like this have a known tendency to overstate the extremes (i.e. 'tails'), so these results should be interpreted cautiously.

Image

It is readily visible that the difference in outcomes varies significantly, depending on the SR encountered. At the 10th percentile the investor has $291,762, while at the 90th percentile the investor has $648,275, more than twice as much. Further, if the portfolio had a greater allocation to stocks or if the time period used were longer than twenty years, the relative difference between these points would grow even larger.

At the 10th percentile, the dollar-weighted (i.e. what the investor actually experienced) real return over the entire period was just 2%, while the dollar-weighted real return at the 90th percentile was 9.4%. This illustrates how dollar-weighted returns, which are what accumulating investors actually experience, can greatly diverge from the time-weighted returns that are usually discussed.

Examine the above results after ten years of investing in the above scenario. The 10th percentile has a real portfolio balance of $122,750, while at the 90th percentile this balance is $209,724, a difference of less than $90k, though this means that the 90th percentile balance was 72% as large after just ten years as the 10th percentile was after twenty years. This difference alone is causing some of the observed difference after twenty years, but what really matters here are the returns experienced during the latter years of the accumulation period. At the 10th percentile, the portfolio balance increased by about 138% in the last decade, while the portfolio balanced increased by 209% in the last decade at the 90th percentile.

If the investor's goal was planning on 'average' returns (i.e. ending with the median ending balance of $436,650) and planned accordingly, they would have achieved that by year 16 at the 90th percentile but would only have 67% of that after 20 years at the 10th percentile.

What then are accumulating investors to do to address SRR? First, they should not plan on experiencing historically average returns over their accumulation period. On a time-weighted basis, the markets may deliver these historical returns, but they may do so in a disadvantageous way to you as an investor. By using lower returns for planning purposes, investors are more likely to achieve their investment goals within their allotted time. However, this means that, to the extent that the future resembles the past, investors are likely to be over-saving and may experience negative consequences during the accumulation period as a result, though this may mean that investors have a higher likelihood of achieving their goals early (e.g. early retirement).

Second, accumulating investors can reduce the negative consequences of SRR by lengthening their accumulation period. This might seem counter-intuitive given the above statement that the difference in very good and very poor outcomes expands as this period grows. While this is true, longer accumulation periods reduce the negative consequences of SRR because the longer that investors are 'in the market', the less dependent they are on the returns of any specific time period. An investor with a 40 year accumulation period is less dependent on time-specific returns than an investor with a 10 year accumulation period. Consequently, this means that those who are striving for early retirements will be more exposed to SRR than others because their accumulation period is shorter.

Third, higher allocations to fixed income, as opposed to equities, can be expected to reduce SRR due to their lower volatility. However, this comes at the expense of a lower expected return over the period as well and may lengthen the needed accumulation period.

Fourth, an obvious way to deal with SRR is to continue accumulating if one has not achieved one's investment goals during the specified period. Given that returns tend to be mean-reverting, such an investor might get a boost from better returns while still accumulating or in the early years of retirement, when SRR is greatest for that period as well.
“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

Grt2bOutdoors
Posts: 22216
Joined: Thu Apr 05, 2007 8:20 pm
Location: New York

Re: Sequence of returns risk while accumulating

Post by Grt2bOutdoors » Mon Feb 19, 2018 5:18 pm

+1 Great write-up.

Cliff Notes Version:

1) Save more.
2) Expect less.
3) There is no free lunch.
4) Keep working.
"One should invest based on their need, ability and willingness to take risk - Larry Swedroe" Asking Portfolio Questions

KlangFool
Posts: 15454
Joined: Sat Oct 11, 2008 12:35 pm

Re: Sequence of returns risk while accumulating

Post by KlangFool » Mon Feb 19, 2018 5:40 pm

OP,

1) In summary, saves more since working longer may not be a viable option.

2) And, for most people, it comes down to do not be "House Poor". Once a person buys the house that everyone expects you to buy, you are in debt to the house for a very long period of time. Savings go out of the window.

3) It is very simple but hard. If you save one year of annual expense every year, you can reach your number in 25 years even with 0% real return.

KlangFool

dbr
Posts: 32087
Joined: Sun Mar 04, 2007 9:50 am

Re: Sequence of returns risk while accumulating

Post by dbr » Mon Feb 19, 2018 6:31 pm

Thanks for posting that. I often wonder why people think the sequence of returns does not matter during accumulation. It is the same math with a different sign on the withdrawals. Note there is a planning difference that you get to see the outcome before you do something rash like retiring, while in the retired case you have to die to see the full detail of the outcome.

User avatar
Topic Author
willthrill81
Posts: 17355
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Sequence of returns risk while accumulating

Post by willthrill81 » Mon Feb 19, 2018 6:44 pm

dbr wrote:
Mon Feb 19, 2018 6:31 pm
Note there is a planning difference that you get to see the outcome before you do something rash like retiring, while in the retired case you have to die to see the full detail of the outcome.
Right. If investors see that SRR has hit them during accumulation, many can simply keep working and/or up their savings. If retirees experience the negative consequences of it, they likely have little to do but cut their spending and/or seek out alternative income sources.

I think that delaying SS until age 70 is one of the more powerful means available to most to reduce SRR in both accumulation (decreases the needed portfolio size) and decumulation (helps to better floor income than starting SS earlier).
“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

JBTX
Posts: 5955
Joined: Wed Jul 26, 2017 12:46 pm

Re: Sequence of returns risk while accumulating

Post by JBTX » Mon Feb 19, 2018 8:36 pm

Will

Your OP shows modeling that the bottom 10% is 2% real return over 20 years. I am guessing that 3% would get you to around bottom 20%, give or take.

Given that, is it unreasonable to expect 3% real, on average, when you are starting at very high CAPE10 and below average long term interest rates?

Also, what conceptually is sequence of returns risk during accumation. Overall, buying high for much of the period would give you less shares. Also, I would tend to think high valuations during the early part of accumulation would be most problematic since these early contributions tend to give the biggest boost to long term portfolio performance.

Random Walker
Posts: 4399
Joined: Fri Feb 23, 2007 8:21 pm

Re: Sequence of returns risk while accumulating

Post by Random Walker » Mon Feb 19, 2018 9:10 pm

Here is an option. The variability in potential outcomes is due to the standard deviation of possible returns in any given year. If one can keep the expected return of the portfolio constant, but decrease the expected and realized volatility, then the dispersion of the final outcomes will be smaller. Moreover, the portfolio’s compounded return will be closer to the simple weighted average return of portfolio components because volatility drag is lessened. How can expected return be held constant and expected volatility lessened? By diversifying across sources of risk / sources of return that have correlations less than 1. This is possible with international diversification on the equity side, diversification across factors, and diversifying into alternatives such as variance risk premium, alternative lending, reinsurance, time series momentum, style premia. I’ve brought this up in the past in two prior threads.

Expanding On The Rewards Of Multiple Asset Class Investing
viewtopic.php?f=10&t=231257&p=3778438#p3778438

Do You Think About Volatility Drag On Your Portfolio?
viewtopic.php?f=10&t=199092&newpost=3046609

Dave

jmk
Posts: 577
Joined: Tue Nov 01, 2011 7:48 pm

Re: Sequence of returns risk while accumulating

Post by jmk » Mon Feb 19, 2018 9:55 pm

dbr wrote:
Mon Feb 19, 2018 6:31 pm
Thanks for posting that. I often wonder why people think the sequence of returns does not matter during accumulation. It is the same math with a different sign on the withdrawals. Note there is a planning difference that you get to see the outcome before you do something rash like retiring, while in the retired case you have to die to see the full detail of the outcome.
One reason it matters more during retirement is that SRR matters more at that point. This is intuitive: If the market crashes 50% when you first start it won't matter much to your long term totals; but if the market crashes 50% right before you retire it affects things enormously. Also, in retirement you don't have the ability to work more to make up for losses. Finally, many people tend to retire at the end of a bull market when valuations are high.

But those who argue SRR mattes more at retirement are not saying it doesn't matter in accumulation. Just that it matters most in the years just prior to retirement and in the years just after.

User avatar
Topic Author
willthrill81
Posts: 17355
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Sequence of returns risk while accumulating

Post by willthrill81 » Mon Feb 19, 2018 10:10 pm

JBTX wrote:
Mon Feb 19, 2018 8:36 pm
Will

Your OP shows modeling that the bottom 10% is 2% real return over 20 years. I am guessing that 3% would get you to around bottom 20%, give or take.

Given that, is it unreasonable to expect 3% real, on average, when you are starting at very high CAPE10 and below average long term interest rates?
I'd say that it's not unreasonable to use 3% real as a basis for one's planning, though I think that the likelihood of this actually taking place over the next 20 years is too pessimistic. Kitces found that 30 year returns following years where CAPE10 was in the top 20% experienced real returns only 1% below the historic 7% average, while the inverse was true when CAPE10 was in the bottom 20% at the starting point. Over the long-term, current valuations haven't had much impact on returns, certainly not enough to be useful for planning purposes.
JBTX wrote:
Mon Feb 19, 2018 8:36 pm
Also, what conceptually is sequence of returns risk during accumation. Overall, buying high for much of the period would give you less shares. Also, I would tend to think high valuations during the early part of accumulation would be most problematic since these early contributions tend to give the biggest boost to long term portfolio performance.
It would be a very real SRR problem for an accumulator to experience great returns for most of their accumulation period but poor returns for the last decade of accumulation (e.g. 2000-2009). Remember that the final decade of accumulation and the first decade of retirement are the times in an investor's tenure where SRR can really hit them hard.
“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
Topic Author
willthrill81
Posts: 17355
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Sequence of returns risk while accumulating

Post by willthrill81 » Mon Feb 19, 2018 10:16 pm

Random Walker wrote:
Mon Feb 19, 2018 9:10 pm
Here is an option. The variability in potential outcomes is due to the standard deviation of possible returns in any given year. If one can keep the expected return of the portfolio constant, but decrease the expected and realized volatility, then the dispersion of the final outcomes will be smaller. Moreover, the portfolio’s compounded return will be closer to the simple weighted average return of portfolio components because volatility drag is lessened.
This is why glidepaths are put forward by many; portfolios with more bonds are less susceptible to SRR than portfolios that are very stocky heavy. The problem with this approach is that it reduces the expected return on the portfolio just when those higher returns, if they do occur, could have the greatest positive impact on your portfolio.
Random Walker wrote:
Mon Feb 19, 2018 9:10 pm
How can expected return be held constant and expected volatility lessened? By diversifying across sources of risk / sources of return that have correlations less than 1. This is possible with international diversification on the equity side, diversification across factors, and diversifying into alternatives such as variance risk premium, alternative lending, reinsurance, time series momentum, style premia. I’ve brought this up in the past in two prior threads.

Expanding On The Rewards Of Multiple Asset Class Investing
viewtopic.php?f=10&t=231257&p=3778438#p3778438

Do You Think About Volatility Drag On Your Portfolio?
viewtopic.php?f=10&t=199092&newpost=3046609

Dave
Yes, diversifying can be a great means of reducing SRR. Personally, I currently dabble in peer-to-peer lending and am a trend follower. Trend following, in particular, as been shown by academics to have greatly reduced the negative impact of SRR without compromising upside potential.
“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

randomguy
Posts: 8660
Joined: Wed Sep 17, 2014 9:00 am

Re: Sequence of returns risk while accumulating

Post by randomguy » Mon Feb 19, 2018 10:31 pm

jmk wrote:
Mon Feb 19, 2018 9:55 pm
dbr wrote:
Mon Feb 19, 2018 6:31 pm
Thanks for posting that. I often wonder why people think the sequence of returns does not matter during accumulation. It is the same math with a different sign on the withdrawals. Note there is a planning difference that you get to see the outcome before you do something rash like retiring, while in the retired case you have to die to see the full detail of the outcome.
One reason it matters more during retirement is that SRR matters more at that point. This is intuitive: If the market crashes 50% when you first start it won't matter much to your long term totals; but if the market crashes 50% right before you retire it affects things enormously. Also, in retirement you don't have the ability to work more to make up for losses. Finally, many people tend to retire at the end of a bull market when valuations are high.

But those who argue SRR mattes more at retirement are not saying it doesn't matter in accumulation. Just that it matters most in the years just prior to retirement and in the years just after.
Sequence matters but so does total return. Historical 50 year returns (call it a lifetime of market investing) vary from 7% to 13% nominal. That is huge given what a 1% change will do to account balances over that time frame. Now when you through in the variable portfolio value (i.e. the sequence part), you get differences. But I bet you end up with a lot more money with 13% with a bad sequence than you do with 7% with a good sequence.:) This is well known.

SSR handling in accumulation is easier to deal with as you can manipulate more levers from saving more, being more aggressive, to working longer.

randomguy
Posts: 8660
Joined: Wed Sep 17, 2014 9:00 am

Re: Sequence of returns risk while accumulating

Post by randomguy » Mon Feb 19, 2018 10:34 pm

KlangFool wrote:
Mon Feb 19, 2018 5:40 pm

2) And, for most people, it comes down to do not be "House Poor". Once a person buys the house that everyone expects you to buy, you are in debt to the house for a very long period of time. Savings go out of the window.

I am bit curious. How did all your coworkers and friend lose money in the washington, DC housing market? In general the market has had decent returns over the past 20 years. If I had to guess the problem wasn't being house poor. It was spending that house equity.

User avatar
Watty
Posts: 18722
Joined: Wed Oct 10, 2007 3:55 pm

Re: Sequence of returns risk while accumulating

Post by Watty » Mon Feb 19, 2018 10:36 pm

willthrill81 wrote:
Mon Feb 19, 2018 5:08 pm
Second, accumulating investors can reduce the negative consequences of SRR by lengthening their accumulation period. This might seem counter-intuitive given the above statement that the difference in very good and very poor outcomes expands as this period grows. While this is true, longer accumulation periods reduce the negative consequences of SRR because the longer that investors are 'in the market', the less dependent they are on the returns of any specific time period. An investor with a 40 year accumulation period is less dependent on time-specific returns than an investor with a 10 year accumulation period. Consequently, this means that those who are striving for early retirements will be more exposed to SRR than others because their accumulation period is shorter.
There are a couple of problems with this. As others have mentioned a person may not be able to work longer even if they want to. When I was going through my 50's a saw a lot more people than I would have expected run into career or health setbacks. Planning earlier retirement has a huge advantage in that you may be able to work longer if you need to.

I did not have a real specific plan but I was generally targeting being able to retire by the time I turned 60. By the time I was in my early 50's my numbers were looking pretty good so I cut my retirement savings back to be just enough to get my employers 401k match and I started spending more on things like travel. I retired a few years ago by the time I turned 59 but I could have worked longer if I needed to.

User avatar
Topic Author
willthrill81
Posts: 17355
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Sequence of returns risk while accumulating

Post by willthrill81 » Mon Feb 19, 2018 10:52 pm

Watty wrote:
Mon Feb 19, 2018 10:36 pm
willthrill81 wrote:
Mon Feb 19, 2018 5:08 pm
Second, accumulating investors can reduce the negative consequences of SRR by lengthening their accumulation period. This might seem counter-intuitive given the above statement that the difference in very good and very poor outcomes expands as this period grows. While this is true, longer accumulation periods reduce the negative consequences of SRR because the longer that investors are 'in the market', the less dependent they are on the returns of any specific time period. An investor with a 40 year accumulation period is less dependent on time-specific returns than an investor with a 10 year accumulation period. Consequently, this means that those who are striving for early retirements will be more exposed to SRR than others because their accumulation period is shorter.
There are a couple of problems with this. As others have mentioned a person may not be able to work longer even if they want to. When I was going through my 50's a saw a lot more people than I would have expected run into career or health setbacks. Planning earlier retirement has a huge advantage in that you may be able to work longer if you need to.

I did not have a real specific plan but I was generally targeting being able to retire by the time I turned 60. By the time I was in my early 50's my numbers were looking pretty good so I cut my retirement savings back to be just enough to get my employers 401k match and I started spending more on things like travel. I retired a few years ago by the time I turned 59 but I could have worked longer if I needed to.
It's very true that an accumulator may not be able to continue working, just as a retiree might not be able to cut their spending. There is no one-size-fits-all approach to addressing SRR. But yes, the statement of mind you highlighted is correct. Simply being exposed to SRR doesn't mean that it's guaranteed to kick you in the teeth, it just means that the dispersion of results is bigger. It could just as easily go the other way and reward you in a big way. Those who were in the final decade of accumulation during the 1990s got an extremely favorable sequence of returns.

That being said, I think that a lot of 50-somethings who get laid off and can't secure another position of approximately equal pay shouldn't let that alone 'force' them into early retirement. Even if they need to take a large pay cut and can't save anything more for retirement, they can at least pay for some of their expenses with some kind of employment, that will take unnecessary strain off of their portfolio.
Last edited by willthrill81 on Mon Feb 19, 2018 10:55 pm, edited 1 time in total.
“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

JBTX
Posts: 5955
Joined: Wed Jul 26, 2017 12:46 pm

Re: Sequence of returns risk while accumulating

Post by JBTX » Mon Feb 19, 2018 10:53 pm

willthrill81 wrote:
Mon Feb 19, 2018 10:10 pm
JBTX wrote:
Mon Feb 19, 2018 8:36 pm
Will

Your OP shows modeling that the bottom 10% is 2% real return over 20 years. I am guessing that 3% would get you to around bottom 20%, give or take.

Given that, is it unreasonable to expect 3% real, on average, when you are starting at very high CAPE10 and below average long term interest rates?
I'd say that it's not unreasonable to use 3% real as a basis for one's planning, though I think that the likelihood of this actually taking place over the next 20 years is too pessimistic. Kitces found that 30 year returns following years where CAPE10 was in the top 20% experienced real returns only 1% below the historic 7% average, while the inverse was true when CAPE10 was in the bottom 20% at the starting point. Over the long-term, current valuations haven't had much impact on returns, certainly not enough to be useful for planning purposes.
JBTX wrote:
Mon Feb 19, 2018 8:36 pm
Also, what conceptually is sequence of returns risk during accumation. Overall, buying high for much of the period would give you less shares. Also, I would tend to think high valuations during the early part of accumulation would be most problematic since these early contributions tend to give the biggest boost to long term portfolio performance.
It would be a very real SRR problem for an accumulator to experience great returns for most of their accumulation period but poor returns for the last decade of accumulation (e.g. 2000-2009). Remember that the final decade of accumulation and the first decade of retirement are the times in an investor's tenure where SRR can really hit them hard.
I haven’t thought about it a lot, but it makes sense.

If you invested a lump sum, the sequence of returns makes no difference. However where it makes a difference is in periodic investments going forward. You would rather have low returns early to purchase more shares and then have the values shoot up at the end. That way you end up with more shares.

However, this whole sequence of returns discussion implies a level of mean reversion. If you have many of years of good returns it is likely that they will be followed with bad returns. If returns were completely random, then even if you had 20 years of good returns then whether or not the next 10 is good or bad is the same probability as if you had 20 bad years preceding.

I think from other posts you indicate that markets are ultimately mean reverting - if so I agree (forgive me if I have mistaked your position ). However if you think returns are mean reverting, I would think that CAPE10 would be mean reverting too. The reason CAPE10 isn’t a great predictor or future returns is because of 1992-2018 has mostly been well above average, yet has had better than average returns. By the same token, you have also had many years of above average returns strung together.

I guess my point is being concerned with sequence of returns seems to imply long term mean reversion, which I would think would apply to CAPE10 too. I’m sorry if I’m not being clear here, I’m having trouble succinctly articulating this.

BanquetBeer
Posts: 446
Joined: Thu Jul 13, 2017 5:57 pm

Re: Sequence of returns risk while accumulating

Post by BanquetBeer » Mon Feb 19, 2018 10:53 pm

Don’t like the monte Carlo aspect. Distorts the top and bottom ends (what was used for comparison). I am curious on the impact of timing in retirement savings. If your savings timeline is 25 years how do historic SRR impact your final balance if it occurs year 0, 5, 10, etc. for what will likely be a 10 year period.

I’d imagine that if you extend the non monte Carlo simulation beyond retirement for 10 years the results will look different.

I would think the earlier you have low SRR the better (more shares followed by high growth) but in reality savings amounts go up with age (started work saving $10k vs 15 years later saving 30k, etc)

Lots of variables to consider but the Monte Carlo simple tells you the obvious - if returns are low you will have less money than if returns are high.

KlangFool
Posts: 15454
Joined: Sat Oct 11, 2008 12:35 pm

Re: Sequence of returns risk while accumulating

Post by KlangFool » Mon Feb 19, 2018 10:58 pm

randomguy wrote:
Mon Feb 19, 2018 10:34 pm
KlangFool wrote:
Mon Feb 19, 2018 5:40 pm

2) And, for most people, it comes down to do not be "House Poor". Once a person buys the house that everyone expects you to buy, you are in debt to the house for a very long period of time. Savings go out of the window.

I am bit curious. How did all your coworkers and friend lose money in the washington, DC housing market? In general the market has had decent returns over the past 20 years. If I had to guess the problem wasn't being house poor. It was spending that house equity.
randomguy,

<< In general the market has had decent returns over the past 20 years. >>

In my area, Northern VA, the house's price had recovered substantially since 2008/2009. But, it is still below the 2004/2005 level.

<<If I had to guess the problem wasn't being house poor. It was spending that house equity.>>

What house equity? If the household income is 150K and the house is 500K to 600K, most of the money is going towards house payment and interest. The household could not save any money in 401K and so on. He/she will pay a lot of taxes. Then, you throw in long-term unemployment or under-employment in the 40s and 50s, it does not take much to fall over the cliff.

In my affluent (annual median household income of 150K) neighborhood, I am a small minority that can afford to pay for my kids' college education. My peer's children have to take on the student loan. The major difference between me and my peer is in the house. My house is cheaper and I only buy a house when my net worth is 2.5 times the house price. They bought a more expensive house and much earlier than me.

KlangFool

User avatar
Topic Author
willthrill81
Posts: 17355
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Sequence of returns risk while accumulating

Post by willthrill81 » Mon Feb 19, 2018 11:01 pm

JBTX wrote:
Mon Feb 19, 2018 10:53 pm
However, this whole sequence of returns discussion implies a level of mean reversion. If you have many of years of good returns it is likely that they will be followed with bad returns. If returns were completely random, then even if you had 20 years of good returns then whether or not the next 10 is good or bad is the same probability as if you had 20 bad years preceding.

I think from other posts you indicate that markets are ultimately mean reverting - if so I agree (forgive me if I have mistaked your position ). However if you think returns are mean reverting, I would think that CAPE10 would be mean reverting too. The reason CAPE10 isn’t a great predictor or future returns is because of 1992-2018 has mostly been well above average, yet has had better than average returns. By the same token, you have also had many years of above average returns strung together.

I guess my point is being concerned with sequence of returns seems to imply long term mean reversion, which I would think would apply to CAPE10 too. I’m sorry if I’m not being clear here, I’m having trouble succinctly articulating this.
Yes, market returns definitely have a mean reversion tendency. But it can take a long time for returns to mean revert and even longer for valuations to mean revert. As you note, CAPE10 has been above its historic average for U.S. equities since 1992 with only a brief interruption in 2008-2009. Others have noted that valuations can only give a crude idea of future expected returns because they do not mean revert in a reliable fashion at all.

And it is possible to be on the wrong side of SRR during both accumulation and decumulation. I don't recall which specific periods where this was particularly bad in the U.S., but one could have good returns early in accumulation when it doesn't really benefit you, poor returns for the latter stages of accumulation and early stages of retirement, and then good returns again in the latter stages of retirement when your portfolio is already largely depleted.
“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
Topic Author
willthrill81
Posts: 17355
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Sequence of returns risk while accumulating

Post by willthrill81 » Mon Feb 19, 2018 11:02 pm

BanquetBeer wrote:
Mon Feb 19, 2018 10:53 pm
Don’t like the monte Carlo aspect. Distorts the top and bottom ends (what was used for comparison). I am curious on the impact of timing in retirement savings. If your savings timeline is 25 years how do historic SRR impact your final balance if it occurs year 0, 5, 10, etc. for what will likely be a 10 year period.

I’d imagine that if you extend the non monte Carlo simulation beyond retirement for 10 years the results will look different.

I would think the earlier you have low SRR the better (more shares followed by high growth) but in reality savings amounts go up with age (started work saving $10k vs 15 years later saving 30k, etc)

Lots of variables to consider but the Monte Carlo simple tells you the obvious - if returns are low you will have less money than if returns are high.
Yes, there are problems with Monte Carlo analysis that I pointed out in the OP. I simply used them for illustrative purposes. However, I only discussed the 'middle 80%' of returns predicted by the analysis in order to avoid at least some of the 'too fat' tails that Monte Carlo analysis is known for producing.
“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

randomguy
Posts: 8660
Joined: Wed Sep 17, 2014 9:00 am

Re: Sequence of returns risk while accumulating

Post by randomguy » Mon Feb 19, 2018 11:42 pm

KlangFool wrote:
Mon Feb 19, 2018 10:58 pm
randomguy wrote:
Mon Feb 19, 2018 10:34 pm
KlangFool wrote:
Mon Feb 19, 2018 5:40 pm

2) And, for most people, it comes down to do not be "House Poor". Once a person buys the house that everyone expects you to buy, you are in debt to the house for a very long period of time. Savings go out of the window.

I am bit curious. How did all your coworkers and friend lose money in the washington, DC housing market? In general the market has had decent returns over the past 20 years. If I had to guess the problem wasn't being house poor. It was spending that house equity.
randomguy,

<< In general the market has had decent returns over the past 20 years. >>

In my area, Northern VA, the house's price had recovered substantially since 2008/2009. But, it is still below the 2004/2005 level.

<<If I had to guess the problem wasn't being house poor. It was spending that house equity.>>

What house equity? If the household income is 150K and the house is 500K to 600K, most of the money is going towards house payment and interest. The household could not save any money in 401K and so on. He/she will pay a lot of taxes. Then, you throw in long-term unemployment or under-employment in the 40s and 50s, it does not take much to fall over the cliff.

In my affluent (annual median household income of 150K) neighborhood, I am a small minority that can afford to pay for my kids' college education. My peer's children have to take on the student loan. The major difference between me and my peer is in the house. My house is cheaper and I only buy a house when my net worth is 2.5 times the house price. They bought a more expensive house and much earlier than me.

KlangFool
400% over the past 25 years.
https://wtop.com/business-finance/2017/ ... ince-1991/

But lets assume you don't care about 25 years terms.

District Measured notes from 2002 to 2016, which included the recession price dip, single-family home values in D.C. still rose 147 percent, compared to 55 percent for the entire metro area and 36 percent nationwide.

I am sure some one who bought at the absolute peak didn't do as well but I know plenty of people who have bought from 1995-2010 and they have all done fine with their home investment. And in my area prices are definitely 30%+ above those of a peak. Of course there are not 500k homes for sale in my hood:)

I am very suspect that difference between success and failure was the house. A 500k mortgage, taxes and so on is going to be on the order 4k payment. That isn't going to be enough to be the difference between success and failure over 10 years (and if it is over 10 years, they making money on their house). Seems far more likley they had additional spending issues.

CurlyDave
Posts: 1550
Joined: Thu Jul 28, 2016 11:37 am

Re: Sequence of returns risk while accumulating

Post by CurlyDave » Mon Feb 19, 2018 11:50 pm

1. I think that in accumulation SR Risk is the wrong term to use. It should be SR Effect. As I approach retirement I know exactly what my portfolio is, and have a very good idea of what it will be at retirement. I must adjust my lifestyle choices so that my portfolio accommodates them.

2. It would be really interesting to see a plot of final portfolio vs. starting year. I expect it would be a fairly smooth curve, but maybe not.

n00b590
Posts: 33
Joined: Mon Dec 04, 2017 3:35 pm

Re: Sequence of returns risk while accumulating

Post by n00b590 » Tue Feb 20, 2018 12:02 am

Unless I'm missing something, it seems your Monte Carlo simulations show the effects of variability in the actual time-weighted returns as well, not just in the sequence of returns. They are both risks to keep in mind, but I don't see how you can draw any conclusions on sequence risk specifically when your simulations conflate the two.

That being said, Lifecycle Investing introduces the idea of time diversification, using leverage when young in an attempt to minimize the sequence of returns risk while in the accumulation stage.

User avatar
siamond
Posts: 5317
Joined: Mon May 28, 2012 5:50 am

Re: Sequence of returns risk while accumulating

Post by siamond » Tue Feb 20, 2018 12:14 am

willthrill81 wrote:
Mon Feb 19, 2018 5:08 pm
What seems lost on many, though, is that SRR is also very real for those who are still in the accumulation phase (i.e. building their portfolio before retirement). Most of the investment calculators out there demonstrate how much you would have if you invested X dollars over Y time period at Z rate of return. The problem is that in the real world, returns are not constant, giving rise to SRR for accumulators.
Here is where I think the confusion comes from. Personally I view the definition of 'sequence of returns' as:
- Year1 returns R1
- Year2 returns R2
- Year3 returns R3
- Year4 returns R4

Now, let's shuffle the sequence of returns.
- Year1 returns R4
- Year2 returns R1
- Year3 returns R2
- Year4 returns R3

Do you expect a different outcome for $X invested at the beginning of the period? Certainly not. The initial amount X will become:
X*(1+R1)*(1+R2)*(1+R3)*(1+R4)

And since multiplication is commutative, this is exactly the same as:
X*(1+R4)*(1+R1)*(1+R2)*(1+R3) --- or any other reshuffling of the sequence of returns

What I think the OP means is that annual cash additions (during accumulation) or annual withdrawals (during retirement) DO create an outcome that depends on the sequence of returns. Which is very true.

Unfortunately, the formulation "if you invested X dollars over Y time period at Z rate of return" is actually somewhat ambiguous! One interpretation (initial investment) does not depend on the sequence of returns, the other (annual investments) does. Words, words, words...

User avatar
Oicuryy
Posts: 1383
Joined: Thu Feb 22, 2007 10:29 pm

Re: Sequence of returns risk while accumulating

Post by Oicuryy » Tue Feb 20, 2018 12:47 am

siamond's formulas for calculating SSR, the sustainable spending rate of constant annual withdrawals, can also be used in calculating the ending value of constant annual contributions. Just multiply the constant contribution amount by Gn/SSR.

Ron
Money is fungible | Abbreviations and Acronyms

User avatar
Topic Author
willthrill81
Posts: 17355
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Sequence of returns risk while accumulating

Post by willthrill81 » Tue Feb 20, 2018 1:03 am

siamond wrote:
Tue Feb 20, 2018 12:14 am
What I think the OP means is that annual cash additions (during accumulation) or annual withdrawals (during retirement) DO create an outcome that depends on the sequence of returns. Which is very true.
Yes, that was my point. For lump sum investing with no planned withdrawals, there is no SRR. But exceptionally few of us are in that position. :wink:

As an aside, this is directly related to the topic of time-weighted versus dollar-weighted returns. All of the returns listed for mutual funds, for instance, are time-weighted (i.e. they assume a lump sum invested over a certain time period), rather than dollar-weighted (i.e. what an accumulating investor would experience). To their credit, this is the only way they can realistically tackle the issue of providing returns because dollar-weighted returns obviously depend greatly on how many dollars are invested when.

Several years ago, Kitces demonstrated how the distinction between time-weighted and dollar-weighted returns was not considered in the now famous DALBAR study that erroneously came to the conclusion that most investors only get a fraction of the market's returns.
Suppose the stock market doubled in year one and then stayed flat for nine years. Over the 10-year period, the market return is 7.2% a year (“rule of 72”). If an investor invested $1,000 every year in an index fund that exactly matched the market, the investor would have $11,000 at the end of 10 years. Only the first $1,000 doubled. The other $9,000 had a zero return. As a result, the investor’s dollar-weighted return is only 1.7% a year for 10 years.

The big difference between the market’s 7.2% per year return and the investor’s 1.7% per year dollar-weighted return isn’t caused by any performance chasing or bad market timing. The investor is just faithfully investing in an index fund for the long term. When the market did well in year one, the investor simply didn’t have much money invested to catch the good return.

Now suppose the stock market stayed flat for nine years and then doubled in year 10. Over the 10-year period, the market return is still 7.2% a year. If an investor invested $1,000 every year in an index fund that exactly matched the market, this investor would have $20,000 at the end of 10 years, resulting in a dollar-weighted return of 12.3% a year for 10 years. It’s higher than the market return because in the year when the market return was high, the investor had $10,000 invested versus only $1,000 invested in the previous example.

Depending whether the market had higher returns in the beginning or in the end, investors are seen either as dumb or smart even when they made no effort to time the market.
https://www.kitces.com/blog/does-the-da ... uest-post/
“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

SGM
Posts: 3074
Joined: Wed Mar 23, 2011 4:46 am

Re: Sequence of returns risk while accumulating

Post by SGM » Tue Feb 20, 2018 3:17 am

The assumption that I would invest the same amount every year for x years was never something I considered. Each year with few exceptions I invested more than the prior year. Raises would come, mortgages eventually get paid off. Periodically I boosted my income by additional training. Self employment became an option. Limits on tax deferred accounts kept going up.

Market downturns made me look for more ways to squeeze a little extra into the market.

Maybe it was all luck, but I think it had more to do with hard work and ability to save and staying the course. I embraced many BH principles before i had heard of BHs.

carolinaman
Posts: 4038
Joined: Wed Dec 28, 2011 9:56 am
Location: North Carolina

Re: Sequence of returns risk while accumulating

Post by carolinaman » Tue Feb 20, 2018 7:20 am

willthrill81 wrote:
Mon Feb 19, 2018 10:52 pm
Watty wrote:
Mon Feb 19, 2018 10:36 pm
willthrill81 wrote:
Mon Feb 19, 2018 5:08 pm
There are a couple of problems with this. As others have mentioned a person may not be able to work longer even if they want to. When I was going through my 50's a saw a lot more people than I would have expected run into career or health setbacks. Planning earlier retirement has a huge advantage in that you may be able to work longer if you need to.

I did not have a real specific plan but I was generally targeting being able to retire by the time I turned 60. By the time I was in my early 50's my numbers were looking pretty good so I cut my retirement savings back to be just enough to get my employers 401k match and I started spending more on things like travel. I retired a few years ago by the time I turned 59 but I could have worked longer if I needed to.
It's very true that an accumulator may not be able to continue working, just as a retiree might not be able to cut their spending. There is no one-size-fits-all approach to addressing SRR. But yes, the statement of mind you highlighted is correct. Simply being exposed to SRR doesn't mean that it's guaranteed to kick you in the teeth, it just means that the dispersion of results is bigger. It could just as easily go the other way and reward you in a big way. Those who were in the final decade of accumulation during the 1990s got an extremely favorable sequence of returns.

That being said, I think that a lot of 50-somethings who get laid off and can't secure another position of approximately equal pay shouldn't let that alone 'force' them into early retirement. Even if they need to take a large pay cut and can't save anything more for retirement, they can at least pay for some of their expenses with some kind of employment, that will take unnecessary strain off of their portfolio.
Many older workers are unable to continue their careers due to job loss, illness or need to care for loved ones. I agree that those who suffer job loss should find some employment, even at much lower salaries. But there are many more older workers who do not have that option due to health issues or need to care for loved ones.

"Working longer: Good aspiration but not a plan. Research shows that about half of workers leave the work force earlier than they had expected due to health issues, job loss or the need to care for a loved one. If possible, have a contingency plan." – Mark Miller, Morningstar

KlangFool
Posts: 15454
Joined: Sat Oct 11, 2008 12:35 pm

Re: Sequence of returns risk while accumulating

Post by KlangFool » Tue Feb 20, 2018 8:16 am

randomguy wrote:
Mon Feb 19, 2018 11:42 pm

I am very suspect that difference between success and failure was the house. A 500k mortgage, taxes and so on is going to be on the order 4k payment. That isn't going to be enough to be the difference between success and failure over 10 years (and if it is over 10 years, they making money on their house). Seems far more likley they had additional spending issues.
randomguy,

Come on. Which part do you not understand? It never ends with the house. It only starts with the house,

1) In this area, the 500K to 600K house is a 3,000 square feet SFH.

2) The median income is 150K. So, you have the dual professional (Engineers) making 80K each inter-mix with single income director (200K) and so on in the neighborhood.

3) You are surrounded by brand new SUVs. Could you drive an old used car in this neighborhood?

4) Pre-school childcare is about $1,200 per month per kid.

5) Your neighbors went on European vacation this summer.

6) Your neighbors bought a new car for their teenager.

KlangFool

RRAAYY3
Posts: 926
Joined: Thu Jan 17, 2013 12:32 pm

Re: Sequence of returns risk while accumulating

Post by RRAAYY3 » Tue Feb 20, 2018 8:34 am

So ... save money? K got it

FrankLUSMC
Posts: 191
Joined: Tue Nov 28, 2017 1:43 pm

Re: Sequence of returns risk while accumulating

Post by FrankLUSMC » Tue Feb 20, 2018 8:46 am

KlangFool wrote:
Mon Feb 19, 2018 5:40 pm
OP,

1) In summary, saves more since working longer may not be a viable option.

2) And, for most people, it comes down to do not be "House Poor". Once a person buys the house that everyone expects you to buy, you are in debt to the house for a very long period of time. Savings go out of the window.

3) It is very simple but hard. If you save one year of annual expense every year, you can reach your number in 25 years even with 0% real return.

KlangFool
I get the feeling that you got hurt in the housing bubble? Or have you rented your whole life?
Have you calculated your savings (estimated?) by renting vs. owning/buying/mortgaging?

livesoft
Posts: 70085
Joined: Thu Mar 01, 2007 8:00 pm

Re: Sequence of returns risk while accumulating

Post by livesoft » Tue Feb 20, 2018 9:03 am

The Early Retirement Now articles on safe withdrawal rates discuss sequence of return risk in a big way. I believe it turns out that the same effect that SRR has for early retirees is actually quite beneficial for accumulators. Market drops helps accumulators in big ways for the same reasons it hurts early retirees.

Here is the first of 2 parts on Sequence of Return Risk from ERN:
https://earlyretirementnow.com/2017/05/ ... turn-risk/
We have benefited greatly from Sequence of Return Risk!
Yes, you heard that right. The ERN family has benefited from SRR over the last decade or so. You can probably already see where we are going with this, so let’s do the following more thorough calculation.
If you haven't read ALL the parts, then I would encourage you to do so.
Last edited by livesoft on Tue Feb 20, 2018 9:16 am, edited 1 time in total.
Wiki This signature message sponsored by sscritic: Learn to fish.

User avatar
iceport
Posts: 4405
Joined: Sat Apr 07, 2007 4:29 pm

Re: Sequence of returns risk while accumulating

Post by iceport » Tue Feb 20, 2018 9:16 am

willthrill81 wrote:
Mon Feb 19, 2018 10:10 pm
JBTX wrote:
Mon Feb 19, 2018 8:36 pm
Also, what conceptually is sequence of returns risk during accumation. Overall, buying high for much of the period would give you less shares. Also, I would tend to think high valuations during the early part of accumulation would be most problematic since these early contributions tend to give the biggest boost to long term portfolio performance.
It would be a very real SRR problem for an accumulator to experience great returns for most of their accumulation period but poor returns for the last decade of accumulation (e.g. 2000-2009). Remember that the final decade of accumulation and the first decade of retirement are the times in an investor's tenure where SRR can really hit them hard.
That's what I was thinking also. When I read JBTX's question, I instantly thought of an old Gummy tutorial that ended with a bit showing how DCA does not guarantee better returns:

"Now, if you could just pick the years when the stock or index went DOWN then UP, (rather than UP then DOWN) you'd come out ahead with DCA..."

Image

The effect would be even more dramatic if you include the possibility of rebalancing into equities during a severe crash.
"Discipline matters more than allocation.” ─William Bernstein

dbr
Posts: 32087
Joined: Sun Mar 04, 2007 9:50 am

Re: Sequence of returns risk while accumulating

Post by dbr » Tue Feb 20, 2018 9:21 am

siamond wrote:
Tue Feb 20, 2018 12:14 am
willthrill81 wrote:
Mon Feb 19, 2018 5:08 pm
What seems lost on many, though, is that SRR is also very real for those who are still in the accumulation phase (i.e. building their portfolio before retirement). Most of the investment calculators out there demonstrate how much you would have if you invested X dollars over Y time period at Z rate of return. The problem is that in the real world, returns are not constant, giving rise to SRR for accumulators.
Here is where I think the confusion comes from. Personally I view the definition of 'sequence of returns' as:
- Year1 returns R1
- Year2 returns R2
- Year3 returns R3
- Year4 returns R4

Now, let's shuffle the sequence of returns.
- Year1 returns R4
- Year2 returns R1
- Year3 returns R2
- Year4 returns R3

Do you expect a different outcome for $X invested at the beginning of the period? Certainly not. The initial amount X will become:
X*(1+R1)*(1+R2)*(1+R3)*(1+R4)

And since multiplication is commutative, this is exactly the same as:
X*(1+R4)*(1+R1)*(1+R2)*(1+R3) --- or any other reshuffling of the sequence of returns

What I think the OP means is that annual cash additions (during accumulation) or annual withdrawals (during retirement) DO create an outcome that depends on the sequence of returns. Which is very true.

Unfortunately, the formulation "if you invested X dollars over Y time period at Z rate of return" is actually somewhat ambiguous! One interpretation (initial investment) does not depend on the sequence of returns, the other (annual investments) does. Words, words, words...
Exactly so. Sequence of returns involves two things not to be confused with other things:

1. The investor is adding or withdrawing money at different points in time.
2. The compound growth of the returns experienced comes out the same. This is not to be confused with experiencing a different average return between one person's history and another. The outcome one gets is dependent in the first place on the average return that person gets and after that on the sequence of returns for the same average.

KlangFool
Posts: 15454
Joined: Sat Oct 11, 2008 12:35 pm

Re: Sequence of returns risk while accumulating

Post by KlangFool » Tue Feb 20, 2018 9:25 am

FrankLUSMC wrote:
Tue Feb 20, 2018 8:46 am
KlangFool wrote:
Mon Feb 19, 2018 5:40 pm
OP,

1) In summary, saves more since working longer may not be a viable option.

2) And, for most people, it comes down to do not be "House Poor". Once a person buys the house that everyone expects you to buy, you are in debt to the house for a very long period of time. Savings go out of the window.

3) It is very simple but hard. If you save one year of annual expense every year, you can reach your number in 25 years even with 0% real return.

KlangFool
I get the feeling that you got hurt in the housing bubble? Or have you rented your whole life?
Have you calculated your savings (estimated?) by renting vs. owning/buying/mortgaging?
FrankLUSMC,

<<I get the feeling that you got hurt in the housing bubble? >>

1) No. Many of my peers are financially destroyed by being "House Poor".

<<Or have you rented your whole life?>>

2) No.

<<Have you calculated your savings (estimated?) by renting vs. owning/buying/mortgaging?>>

3) Do you buy the same house that you are renting? Please answer this question first.

4) Most folks rent the house (A) one level below the house (B) that they are willing to buy. Then, they use the cost of renting (B) to justify their purchase of (B). So, for most folks, the calculation is a fraud. They were never to rent (B) under any circumstances. They only use renting (B) to justify their increase in housing expense.

So, it is between 400K townhouse (A) versus 600K SFH(B).

5) In my case, I bought the same house (A) that I am willing to rent. And, the PITI is 20% to 30% lower than renting. Hence, I did not increase my housing expense by renting.

KlangFool

ryman554
Posts: 1279
Joined: Sun Jan 12, 2014 9:44 pm

Re: Sequence of returns risk while accumulating

Post by ryman554 » Tue Feb 20, 2018 9:32 am

randomguy wrote:
Mon Feb 19, 2018 11:42 pm
I am very suspect that difference between success and failure was the house. A 500k mortgage, taxes and so on is going to be on the order 4k payment. That isn't going to be enough to be the difference between success and failure over 10 years (and if it is over 10 years, they making money on their house). Seems far more likley they had additional spending issues.
I track this well, since I am the poster boy for being on the cusp of house-poor, but have managed to luckily earn my way out of it. Twice. I don't plan on making it three times.

If you have a mortgage on the order of 500-600k, a relatively frugal household is spending ~$85-90k/year, with about 1/2 of that being the house PITI. That's $30-40k/year on living expenses, which is comfortable, but not extravagant by any stretch, for a family. This notably does not include income taxes or savings or things like pre-school or, as Klang noted, European vacations or new cars to keep up with the Jonses.

At 150k/year (or even 200k/year), state and federal and FICA taxes are a big chunk, say 35-40k. So you're left with ~$20-30k for the rest. Yes, the folks who can't save do have a spending problem. They could save instead of blowing the 20-30k of "fun money". Better still to save pre-tax and get more mileage out of it. But you could also be much less in house and be able to do both and not be so stretched. So, while the house is not the only problem, it's a big part of it.

User avatar
2pedals
Posts: 1054
Joined: Wed Dec 31, 2014 12:31 pm

Re: Sequence of returns risk while accumulating

Post by 2pedals » Tue Feb 20, 2018 9:35 am

Great Post OP

I have a question, is it easy to produce a plot that accumulates based on an increasing investment? For example you start with an investment of $1,000/mo. but in real life that is not practical. In real life most people when they start out they have less disposable income. Once they start saving income is low and disposable income is very low. In general, as their career matures they get more disposable income and can increase retirement savings. For example if one at 30 years old were to save 10% of 75k salary (or $650/mo.) and as they get older there salary at 50 year old increases to 175k and save 20% (or $2,912/mo.). Assume a straight line increase in savings from 650/mo. to 2,912/mo.. What would the sequence of risk be then? Much, much more in my estimation.
Last edited by 2pedals on Tue Feb 20, 2018 10:07 am, edited 2 times in total.

randomguy
Posts: 8660
Joined: Wed Sep 17, 2014 9:00 am

Re: Sequence of returns risk while accumulating

Post by randomguy » Tue Feb 20, 2018 10:04 am

KlangFool wrote:
Tue Feb 20, 2018 8:16 am
randomguy wrote:
Mon Feb 19, 2018 11:42 pm

I am very suspect that difference between success and failure was the house. A 500k mortgage, taxes and so on is going to be on the order 4k payment. That isn't going to be enough to be the difference between success and failure over 10 years (and if it is over 10 years, they making money on their house). Seems far more likley they had additional spending issues.
randomguy,

Come on. Which part do you not understand? It never ends with the house. It only starts with the house,

1) In this area, the 500K to 600K house is a 3,000 square feet SFH.

2) The median income is 150K. So, you have the dual professional (Engineers) making 80K each inter-mix with single income director (200K) and so on in the neighborhood.

3) You are surrounded by brand new SUVs. Could you drive an old used car in this neighborhood?

4) Pre-school childcare is about $1,200 per month per kid.

5) Your neighbors went on European vacation this summer.

6) Your neighbors bought a new car for their teenager.

KlangFool
So what your saving is an expensive house was a smart decision for these people. Instead of spending money on things with no value, they were "saving" it by paying off the mortgage. If you drop these peoples housing payments from 4k to 2k, the added money isn't going to end up in a 401(k). It is getting spent. Again this isn't a house poor situation. This is a spending issue.

And of course you can drive a used car in that neighborhood. Nobody cares what you drive. I live in a 1.2 million neighborhood and nobody has ever said anything about driving a 7 year old 30k MSRP car. Heck most of them do the same thing. You make conversation about with the people who just bought the new 80k car but nobody really cares. It is just something to talk about.

FrankLUSMC
Posts: 191
Joined: Tue Nov 28, 2017 1:43 pm

Re: Sequence of returns risk while accumulating

Post by FrankLUSMC » Tue Feb 20, 2018 10:10 am

KlangFool wrote:
Tue Feb 20, 2018 9:25 am
FrankLUSMC wrote:
Tue Feb 20, 2018 8:46 am
KlangFool wrote:
Mon Feb 19, 2018 5:40 pm
OP,

1) In summary, saves more since working longer may not be a viable option.

2) And, for most people, it comes down to do not be "House Poor". Once a person buys the house that everyone expects you to buy, you are in debt to the house for a very long period of time. Savings go out of the window.

3) It is very simple but hard. If you save one year of annual expense every year, you can reach your number in 25 years even with 0% real return.

KlangFool
I get the feeling that you got hurt in the housing bubble? Or have you rented your whole life?
Have you calculated your savings (estimated?) by renting vs. owning/buying/mortgaging?
FrankLUSMC,

<<I get the feeling that you got hurt in the housing bubble? >>

1) No. Many of my peers are financially destroyed by being "House Poor".

<<Or have you rented your whole life?>>

2) No.

<<Have you calculated your savings (estimated?) by renting vs. owning/buying/mortgaging?>>

3) Do you buy the same house that you are renting? Please answer this question first.

4) Most folks rent the house (A) one level below the house (B) that they are willing to buy. Then, they use the cost of renting (B) to justify their purchase of (B). So, for most folks, the calculation is a fraud. They were never to rent (B) under any circumstances. They only use renting (B) to justify their increase in housing expense.

So, it is between 400K townhouse (A) versus 600K SFH(B).

5) In my case, I bought the same house (A) that I am willing to rent. And, the PITI is 20% to 30% lower than renting. Hence, I did not increase my housing expense by renting.

KlangFool
KlangFool,

Thanks for the insight. I think peer pressure of keeping up with friends/family/neighbors contributes quite a bit to the "House poor" condition.
Sequence of the market also. All risks all the time.
Wife and I realize the need to downsize within the next 10 years. But we too have kept our PITI lower than a rent expense for similar house.
Good luck to you.

User avatar
Sandtrap
Posts: 9677
Joined: Sat Nov 26, 2016 6:32 pm
Location: Hawaii No Ka Oi , N. Arizona

Re: Sequence of returns risk while accumulating

Post by Sandtrap » Tue Feb 20, 2018 10:17 am

Outstanding write-up.
Thanks for the hard work to post it.
Indeed, SRR is a valid concern for anyone no matter what stage of life.
But still, all the more devastating to those in the "retirement red zone" (60+) where the combination of ill health, limited or no employment options, and other maladies due to advancing age all make for unrecoverable personal and financial "black swans".
j :D
Wiki Bogleheads Wiki: Everything You Need to Know

Random Walker
Posts: 4399
Joined: Fri Feb 23, 2007 8:21 pm

Re: Sequence of returns risk while accumulating

Post by Random Walker » Tue Feb 20, 2018 10:40 am

Reference was made above I think to “Murphy’s Law Of Retirement”. Late accumulators benefit from a bull market, see their portfolios swell, and then decide to retire. They are retiring after a long bull, equity valuations high, future expected returns low, whole potential distribution of returns shifts left, potential good outcomes less good, potential bad outcomes more bad. These retiring beneficiaries of recent bull market are at increased risk of retiring into an adverse sequence of returns.
Dave

User avatar
Topic Author
willthrill81
Posts: 17355
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Sequence of returns risk while accumulating

Post by willthrill81 » Tue Feb 20, 2018 12:03 pm

2pedals wrote:
Tue Feb 20, 2018 9:35 am
Great Post OP
Thanks!
2pedals wrote:
Tue Feb 20, 2018 9:35 am
I have a question, is it easy to produce a plot that accumulates based on an increasing investment? For example you start with an investment of $1,000/mo. but in real life that is not practical. In real life most people when they start out they have less disposable income. Once they start saving income is low and disposable income is very low. In general, as their career matures they get more disposable income and can increase retirement savings. For example if one at 30 years old were to save 10% of 75k salary (or $650/mo.) and as they get older there salary at 50 year old increases to 175k and save 20% (or $2,912/mo.). Assume a straight line increase in savings from 650/mo. to 2,912/mo.. What would the sequence of risk be then? Much, much more in my estimation.
It's not easy to compute the effect of increasing contributions over time (very few calculators do it), but it can be done semi-manually in Excel. But yes, the more that contributions are 'back-end loaded', the greater the potential impact of SRR. It comes back to the same idea that you're really depending on the returns you experience during the last decade or so of accumulation.
“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
Topic Author
willthrill81
Posts: 17355
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Sequence of returns risk while accumulating

Post by willthrill81 » Tue Feb 20, 2018 12:16 pm

Sandtrap wrote:
Tue Feb 20, 2018 10:17 am
Outstanding write-up.
Thanks!
Sandtrap wrote:
Tue Feb 20, 2018 10:17 am
But still, all the more devastating to those in the "retirement red zone" (60+) where the combination of ill health, limited or no employment options, and other maladies due to advancing age all make for unrecoverable personal and financial "black swans".
j :D
It's for this reason that I believe that age 60 (or younger) might be a preferred 'planned' retirement age for the purposes of becoming financially independent than the typical 65, if feasible. This could help to compensate for the above risks, all of which are very real. Further, if one can become FI without SS, it can then provide another level of safety beyond one's portfolio.
“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
Topic Author
willthrill81
Posts: 17355
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Sequence of returns risk while accumulating

Post by willthrill81 » Tue Feb 20, 2018 12:18 pm

Random Walker wrote:
Tue Feb 20, 2018 10:40 am
Reference was made above I think to “Murphy’s Law Of Retirement”. Late accumulators benefit from a bull market, see their portfolios swell, and then decide to retire. They are retiring after a long bull, equity valuations high, future expected returns low, whole potential distribution of returns shifts left, potential good outcomes less good, potential bad outcomes more bad. These retiring beneficiaries of recent bull market are at increased risk of retiring into an adverse sequence of returns.
Dave
The one advantage that many in that situation would have is that they benefited from a bull market at the end of their accumulation period. But if this just barely helps them reach their 'number', you're absolutely right that they could be in for a rough go of it. Those who retired in the year 2000, for instance, had a huge boost in the 1990s, but the 2000s were tough on retirees (but good for accumulators).
“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

Random Walker
Posts: 4399
Joined: Fri Feb 23, 2007 8:21 pm

Re: Sequence of returns risk while accumulating

Post by Random Walker » Tue Feb 20, 2018 12:37 pm

Willthrill,
This is the reason I am a fan of customizing one’s own glide path rather than blindly following an age based rule that decreases a percent or two per year. The market doesn’t move in 1-2% yearly adjustments. Like Bernstein writes in his Life Cycle Investing book, when the market has been extraordinarily generous those close (within 5-10 years) to retirement may well want to make more aggressive changes to derisk the portfolio or lock in the liability matching portfolio.

Dave

User avatar
Topic Author
willthrill81
Posts: 17355
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Sequence of returns risk while accumulating

Post by willthrill81 » Tue Feb 20, 2018 12:59 pm

CurlyDave wrote:
Mon Feb 19, 2018 11:50 pm
1. I think that in accumulation SR Risk is the wrong term to use. It should be SR Effect. As I approach retirement I know exactly what my portfolio is, and have a very good idea of what it will be at retirement. I must adjust my lifestyle choices so that my portfolio accommodates them.
Perhaps, but it's not a risk if you wind up benefiting from a great SR.
CurlyDave wrote:
Mon Feb 19, 2018 11:50 pm
2. It would be really interesting to see a plot of final portfolio vs. starting year. I expect it would be a fairly smooth curve, but maybe not.
Below is a graph of a portfolio comprised of 60% U.S. TSM / 40% intermediate term Treasuries beginning with 1972-1991 and ending with 1998-2017. As in the OP, the starting portfolio value is $1,000 with $1,000 invested monthly for 20 years.

Image

The best period was 1980-1999 with an ending inflation-adjusted portfolio of $67,753, which illustrates that having great returns at the end of accumulation is ideal. Conversely, the worst period was 1989-2008 with an ending inflation-adjusted portfolio of $31,663, less than half as much as the period beginning nine years prior. Accumulators are just as subject to SRR as retirees, albeit the consequences may not be as severe.

Now it's true that this conflates both SRR and the compounded annual growth rate of the specified period, but they're both very real risks that cannot be separated in the real world.
Last edited by willthrill81 on Tue Feb 20, 2018 2:40 pm, edited 1 time in total.
“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

MnD
Posts: 4578
Joined: Mon Jan 14, 2008 12:41 pm

Re: Sequence of returns risk while accumulating

Post by MnD » Tue Feb 20, 2018 2:33 pm

Grt2bOutdoors wrote:
Mon Feb 19, 2018 5:18 pm

Cliff Notes Version:

1) Save more.
2) Expect less.
3) There is no free lunch.
4) Keep working.
5) Earn more from work

I made a desired wealth accumulation spreadsheet in 1987 for 1987 through planned retirement for 2018 (in Lotus 123!) with forecast salary progression, savings rate and investment returns. Actual savings rate % was pretty close, actual investment return rate was fairly poor compared to "rosy scenario" rate in forecast and actual salary ended up quite a bit higher than estimated. Overall outcome was reaching desired wealth just about spot on with forecast, but by making more from work and earning less from investment returns versus plan.
70/30 AA for life, Global market cap equity. Rebalance if fixed income <25% or >35%. Weighted ER< .10%. 5% of annual portfolio balance SWR, Proportional (to AA) withdrawals.

KlangFool
Posts: 15454
Joined: Sat Oct 11, 2008 12:35 pm

Re: Sequence of returns risk while accumulating

Post by KlangFool » Tue Feb 20, 2018 3:40 pm

randomguy wrote:
Tue Feb 20, 2018 10:04 am
KlangFool wrote:
Tue Feb 20, 2018 8:16 am
randomguy wrote:
Mon Feb 19, 2018 11:42 pm

I am very suspect that difference between success and failure was the house. A 500k mortgage, taxes and so on is going to be on the order 4k payment. That isn't going to be enough to be the difference between success and failure over 10 years (and if it is over 10 years, they making money on their house). Seems far more likley they had additional spending issues.
randomguy,

Come on. Which part do you not understand? It never ends with the house. It only starts with the house,

1) In this area, the 500K to 600K house is a 3,000 square feet SFH.

2) The median income is 150K. So, you have the dual professional (Engineers) making 80K each inter-mix with single income director (200K) and so on in the neighborhood.

3) You are surrounded by brand new SUVs. Could you drive an old used car in this neighborhood?

4) Pre-school childcare is about $1,200 per month per kid.

5) Your neighbors went on European vacation this summer.

6) Your neighbors bought a new car for their teenager.

KlangFool
So what your saving is an expensive house was a smart decision for these people. Instead of spending money on things with no value, they were "saving" it by paying off the mortgage. If you drop these peoples housing payments from 4k to 2k, the added money isn't going to end up in a 401(k). It is getting spent. Again this isn't a house poor situation. This is a spending issue.
randomguy,

<< Instead of spending money on things with no value, they were "saving" it by paying off the mortgage.>>

Which part don't you get it? In addition to the expensive house, they spend more money on things with no value. They never pay off the mortgage. They could not. In fact, many of them have to re-mortgage their house just to keep up with the Joneses.

<< If you drop these peoples housing payments from 4k to 2k, the added money isn't going to end up in a 401(k). It is getting spent. Again this isn't a house poor situation. This is a spending issue.>>

If they live in a cheaper house, they have to spend less to keep up with Joneses. The spending problem is tied up with the house and the Joneses that you live with.

KlangFool

CurlyDave
Posts: 1550
Joined: Thu Jul 28, 2016 11:37 am

Re: Sequence of returns risk while accumulating

Post by CurlyDave » Tue Feb 20, 2018 3:48 pm

willthrill81 wrote:
Tue Feb 20, 2018 12:59 pm
CurlyDave wrote:
Mon Feb 19, 2018 11:50 pm
2. It would be really interesting to see a plot of final portfolio vs. starting year. I expect it would be a fairly smooth curve, but maybe not.
Below is a graph of a portfolio comprised of 60% U.S. TSM / 40% intermediate term Treasuries beginning with 1972-1991 and ending with 1998-2017. As in the OP, the starting portfolio value is $1,000 with $1,000 invested monthly for 20 years.

Image

The best period was 1980-1999 with an ending inflation-adjusted portfolio of $67,753, which illustrates that having great returns at the end of accumulation is ideal. Conversely, the worst period was 1989-2008 with an ending inflation-adjusted portfolio of $31,663, less than half as much as the period beginning nine years prior. Accumulators are just as subject to SRR as retirees, albeit the consequences may not be as severe.

Now it's true that this conflates both SRR and the compounded annual growth rate of the specified period, but they're both very real risks that cannot be separated in the real world.
Somehow, this curve does not pass my "sanity check".

If I invest $1000 per month for 20 years, that is $240,000. A "best" ending value of $67,753 means I should have just stuffed it under my mattress for those 20 years and done 3.5 times better.

And, even saying a zero got dropped in the graph is not going to correct things. The years ending in 2012-2017 are only going to have ~$360k ending value. This is very weak tea for a $240k investment. I suspect a problem with Portfolio Visualizer, but maybe that is not the case.

Just as an experiment, what happens if someone puts the $1000/month into I-bonds. What does that look like for 20 years ending in 2017?

User avatar
Topic Author
willthrill81
Posts: 17355
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Sequence of returns risk while accumulating

Post by willthrill81 » Tue Feb 20, 2018 3:55 pm

CurlyDave wrote:
Tue Feb 20, 2018 3:48 pm
willthrill81 wrote:
Tue Feb 20, 2018 12:59 pm
CurlyDave wrote:
Mon Feb 19, 2018 11:50 pm
2. It would be really interesting to see a plot of final portfolio vs. starting year. I expect it would be a fairly smooth curve, but maybe not.
Below is a graph of a portfolio comprised of 60% U.S. TSM / 40% intermediate term Treasuries beginning with 1972-1991 and ending with 1998-2017. As in the OP, the starting portfolio value is $1,000 with $1,000 invested monthly for 20 years.

Image

The best period was 1980-1999 with an ending inflation-adjusted portfolio of $67,753, which illustrates that having great returns at the end of accumulation is ideal. Conversely, the worst period was 1989-2008 with an ending inflation-adjusted portfolio of $31,663, less than half as much as the period beginning nine years prior. Accumulators are just as subject to SRR as retirees, albeit the consequences may not be as severe.

Now it's true that this conflates both SRR and the compounded annual growth rate of the specified period, but they're both very real risks that cannot be separated in the real world.
Somehow, this curve does not pass my "sanity check".

If I invest $1000 per month for 20 years, that is $240,000. A "best" ending value of $67,753 means I should have just stuffed it under my mattress for those 20 years and done 3.5 times better.

And, even saying a zero got dropped in the graph is not going to correct things. The years ending in 2012-2017 are only going to have ~$360k ending value. This is very weak tea for a $240k investment. I suspect a problem with Portfolio Visualizer, but maybe that is not the case.

Just as an experiment, what happens if someone puts the $1000/month into I-bonds. What does that look like for 20 years ending in 2017?
Sorry, that was $1,000 invested annually. It would make a small difference by looking at monthly contributions, but not much.
“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
Topic Author
willthrill81
Posts: 17355
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Sequence of returns risk while accumulating

Post by willthrill81 » Tue Feb 20, 2018 5:04 pm

n00b590 wrote:
Tue Feb 20, 2018 12:02 am
Unless I'm missing something, it seems your Monte Carlo simulations show the effects of variability in the actual time-weighted returns as well, not just in the sequence of returns. They are both risks to keep in mind, but I don't see how you can draw any conclusions on sequence risk specifically when your simulations conflate the two.
Here's an example from The Retirement Cafe blog.
Here's an example. Looking again at real S&P 500 market returns from 1871 to 2008 provides 108 rolling 30-year scenarios. If we assume a retiree started each of those periods with a million dollars and withdrew $45,000 every year, he or she would go broke in less than thirty years 9 times (8.33% failure rate).

If we graph annualized market returns for those 108 periods against terminal portfolio values (TPV), we find a correlation of only 0.8. (I say "only" because intuition might tell you that average market returns would explain all of the outcome.)

Image

At the bottom left of the chart, you will see that three periods successfully funded 30 years of retirement while averaging only about 3% market return per year. You will also see a portfolio for the period beginning in 1974 that generated a 6.8% annualized market return and failed. (Both circled in red.)

You can win with a 3% average return and lose with a nearly a 7% average. There's no magic here, it's just that the compound growth rate doesn't contain all the information you need to determine if a sequence of returns will lead to successfully funding retirement.
Over this time frame, the average return accounted for 64% (.8^2) of the variance in final portfolio value, definitely a big factor. The other 36% is due to the particular sequence of returns. The same is true of accumulation.
“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

Post Reply