Why does anyone invest in bonds long term?

[Lsbcal]

Lsbcal: Yes you did satisfy my original request of finding a scenario of a static AA on a > 10 year time scale that lead to better performance with bond allocation, so thank you much for sharing. I'm curious to see the rest of those time tables, how many years out is the break even point where 100/0 catches up? Picking 1929 as the starting point is definitely cherry picking the worst possible dataset, but it does prove your point, it could happen, touche. Am I willing to roll the dice that a crash the likes of 1929 doesn't happen the year I retire, we'll see I guess.

When I do such simulations I personally want to know the worst case historically. Those would be found at years 1929 (depression) and 1966 (2 recessions, inflation). Also the years right around those years can be pretty bad.

In looking at the 100/0 catchup point, I think that was for the simulation year 1951 (age 82). The important point is that a $1,000,000 portfolio went to $427,000 in 4 years during (by 1933). That would be too scary for me.

The VPW tool is a great resource for coming up with one's AA. If one is comfortable with spreadsheets, by fiddling with the sheets one can also simulate other types of withdrawals besides the suggested VPW algorithm.

 

[OldShooter]

... This seems to make good sense, but ...
I'm wondering if we have any backtesting data/tools to support the idea ...

I know of none. The way I come to this conclusions is starting from the premise that a mixture of equities and fixed-income is wise for retirement. (ref page 64 of the JP Morgan slide deck) I also do not believe the that "classical" formulas make much sense. (Like 110 minus age = your equity percentage. ) The reason is that a person's risk tolerance depends in large part on his/her needs and the size of his/her portfolio. A person with large assets relative to needs is really investing for his/her heirs and beneficiaries and, in addition, probably has a high risk tolerance. So a larger equity tranche is probably appropriate. A person with "just enough" may be better served by 100% fixed income.

So from that belief, how to figure the fixed-income portion? That's where I like the bucket concept. Granted, as someone pointed out, it is a mild form of market timing. Granted, too, there are going to be scenarios where it does not maximize portfolio size. There are also going to be scenarios where it saves the bacon. I like bacon, so that is the tradeoff for me. There are too many variables IMO for any spreadsheet or simulation to deal with, not the least of which is projected lifetime (as has been pointed out). Frankly, making this tradeoff heuristically is fairly easy for me because my wife and I have more money than we will ever need. So, if I don't get it quite right, it won't matter much. YMMV, however.

 

[Whisper66]

Here's what people keep missing when they talk about the 'fear of selling stocks in a down market': .......
A portfolio of stocks and fixed income will likely pay ~ 2.5% in divs......ERD50

+1
That is an excellent point I don't hear mentioned often. I took a look at this a couple months back. I found that our bonds and cash would cover our living expenses for ~ 5 yrs. With our relatively low dividends included, we was good for ~ 12 yrs. Ended up getting quite a bit more comfortable that we could outlast any downturn without selling off equities. Good point.

 

[Digger1000]

At age 51 I retired 6 weeks ago. I am 97% in stocks and 3% cash. My planned WR is 2.7-2.9%. Last week I did an interesting exercise. I acted as if I retired on Jan 1,2000 and had all of my assets in the S&P 500 Index. I then went year by year inserting the S&P 500 Index return and also inserted a 2.9% withdrawal rate to start. Now about the withdrawal rate, I did not change the dollar amount as things went south in the stock market therefore the WR rate actually was over 6% at 1 point in the exercise. I did increase the withdrawal dollar amount late in the process. At one point in this exercise my assets were down 51%. However as of today(last week) my assets had actually fully recovered and were up 4% from where I had begun, including all of the yearly withdrawals.

 

[travelover]

.......... I am 97% in stocks and 3% cash.........

A retired relative was 100% stocks in 2008. He ended up taking out a HELOC on his house to avoid selling selling stocks at bargain basement prices for living expenses. Prior to that, he scoffed at bonds and cash.

 

[Digger1000]

Just because the S&P 500 went down 37%? That's insane.

 

[audreyh1]

Just because the S&P 500 went down 37%? That's insane.

By March 2009, the S&P 500 was down 56% from the 2007 peek.

 

[GravitySucks]

Well an under 3% WR is probably going to survive at any AA.
I was north of 85% equities the second year of ER when we had a 12% or so drop and it caused me some stress. I slowly moved to a 60/40 AA and slept well through the next 12+% dip. But my WR is much higher than yours.
The theoretical comfort zone did not match the actual for me.

 

[ERD50]

A retired relative was 100% stocks in 2008. He ended up taking out a HELOC on his house to avoid selling selling stocks at bargain basement prices for living expenses. Prior to that, he scoffed at bonds and cash.

Not that I'm recommending 100% equities (but I wouldn't argue against it either), but what did the HELOC cost him? I'm pretty confident that 100% stocks minus the cost of the HELOC put him way ahead of a more conservative AA in the long run.

-ERD50

 

[2017ish]

A retired relative was 100% stocks in 2008. He ended up taking out a HELOC on his house to avoid selling selling stocks at bargain basement prices for living expenses. Prior to that, he scoffed at bonds and cash.

Not that I'm recommending 100% equities (but I wouldn't argue against it either), but what did the HELOC cost him? I'm pretty confident that 100% stocks minus the cost of the HELOC put him way ahead of a more conservative AA in the long run.

-ERD50

We've opened a HELOC for 100% luxurious [annual] E.R. spending (200% decent E.R. [annual] spending) to compensate for heavy equities and to spread cost of Roth Conversions.

BUT, a real risk is that in 2008-2009 type conditions, the bank could shut the line down.... I figure if we draw fast, we should be ok with the conversion option, but it is not a safety valve that is guaranteed to be there.

 

[travelover]

......... I'm pretty confident that 100% stocks minus the cost of the HELOC put him way ahead of a more conservative AA in the long run.

-ERD50

You may be right, but that is not a strategy I would employ.

 

[ERD50]

We've opened a HELOC for 100% luxurious [annual] E.R. spending (200% decent E.R. [annual] spending) to compensate for heavy equities and to spread cost of Roth Conversions.

BUT, a real risk is that in 2008-2009 type conditions, the bank could shut the line down.... I figure if we draw fast, we should be ok with the conversion option, but it is not a safety valve that is guaranteed to be there.

You may be right, but that is not a strategy I would employ.

I would not recommend relying on a HELOC either. I think we also saw some posts of people not being able to access them in 2008-2009.

My only point was the end result in the case presented was probably positive. And I don't think it is unreasonable to be 100% equities (not recc it either). Some (very few) posters here have said they are 100%, and they seem well aware that they will need to sell some during dips, and are OK with weighing that against the likely higher growth overall and mentally accept the volatility.

-ERD50

 

[OldShooter]

Good thread. Thanks, all. I continue to believe in a bucket approach but I have been exposed here to options and arguments that will be of long-term value to my thinking.

 

[audreyh1]

I think it's really important to point something out in this discussion of why not 100% stocks:

When you choose the "Constant (inflation adjusted) Spending" where withdrawals are based on the initial portfolio value and adjusted by inflation each year so that withdrawals are constant in real dollar terms, the asset allocation has no impact on the amount you withdraw each year. The amount withdrawn is exactly the same, no matter what the allocation is set to. So why would you choose a given allocation? Well - portfolio survival would be the main criteria, followed by consideration of your heirs inheriting the remaining portfolio.

Comparing Portfolio Success Rates (where 100 means you don't run out of money during the period) for the 4% withdrawal rate case for 100% stocks, 75% stocks, and 50% stocks, I think it's rather obvious why a retiree might choose to have some allocation to bonds:

Allocation and Duration	Withdrawal Rate
	4%
100% stocks
15 years	100
20 years	100
25 years	98
30 years	93
35 years	91
40 years	88
75% stocks
15 years	100
20 years	100
25 years	100
30 years	98
35 years	93
40 years	92
50% stocks
15 years	100
20 years	100
25 years	100
30 years	100
35 years	96
40 years	86

Up to 35 years, the portfolio with 50% allocated to bonds out survives the 100% stocks case, and at all durations the 75% stocks allocation out survives the 100% stock allocation.

So, if you aren't getting more income to spend each year, and have a slightly higher chance of failure, why would someone choose 100% stocks instead of 75% or 50%? Especially considering that the 100% stocks case is far more volatile, and you have no opportunity to rebalance?

In the worst case, you are slightly more likely to run out of money, but in the best, and probably also the average case, the 100% stocks version may result in a much larger ending portfolio. But that does the retiree no good during their lifetime, since that terminal portfolio goes to the heirs. So thinking of heirs (and being able to ignore volatility) may be a motivation to be more aggressive in terms of stocks. But I think it's clear that not being 100% stocks confers a benefit during the retirees lifetime.

I think in practical terms few here are running the straight constant (inflation adjusted) withdrawal method which is not particularly efficient, and instead look for ways such that portfolio growth during their lifetime does benefit them by increasing their annual income. Methods such as ratcheting up, % remaining portfolio, VPW, Bob Clyatt 95% rule, etc., all take the current portfolio value into account and allow the retiree to take advantage by increasing income if they are on a good portfolio growth path, or pulling back prudently if faced with a bad run.

In such scenarios you are more likely to increase income faster if you have a higher % allocated to stocks, yet you will be hurt more during a bad run. So it becomes a tradeoff between long-term growth of the portfolio and everyday and annual volatility. Each retiree has to figure out what they can live with and be able to stick to their allocation without panicking and throwing out their allocation and jumping out of stocks at a bad time and not being able to get back in. This is a common problem during market crashes, and we know from folks posting here during bad times that many do freak and realize their stock allocation is too high for their personal comfort. It's a pretty critical thing for an investor to figure out. Careful examination of goals is key.

 

[ERD50]

...

So, if you aren't getting more income to spend each year, and have a slightly higher chance of failure, why would someone choose 100% stocks instead of 75% or 50%? Especially considering that the 100% stocks case is far more volatile, and you have no opportunity to rebalance? ...

Again, not recommending it, but just trying to explain it - your analysis is interesting, but it is based on a 4% WR rate. What if someone looked at a lower WR, one that was 100% historically safe with 100% equities? If that WR was sufficient for them, they might like the idea of having the very significant chance of having a LOT of money later.

For example, $1M, 3% initial WR, 30 years, 100% equities:

The lowest and highest portfolio balance at the end of your retirement was $443,031 to $10,345,067, with an average at the end of $3,682,503

With same #'s but 50% equities:

The lowest and highest portfolio balance at the end of your retirement was $554,953 to $5,063,705, with an average at the end of $1,795,570

Downside risk of 100% equities is ~ $100K, upside benefit is ~ $5M, almost $2M on average.

At 45 years, it becomes more dramatic:

100% EQ: $285,352 to $44,173,166, with an average at the end of $8,096,124

50% EQ: $268,129 to $9,265,891, with an average at the end of $2,447,189.

No downside historically, the upside is huge. Once again, not recommending it, just sayin'

-ERD50

 

[Fedup]

In theory I should be 100% in stocks after reading this thread but I'm in the bucket for chickens or the chicken bucket as I call it, I can't bring myself to commit being 100% in stocks. I did adjust my AA on Vanguard to be higher than what was originally and I'm going to take my time getting there.

 

[ERD50]

In theory I should be 100% in stocks after reading this thread but I'm in the bucket for chickens or the chicken bucket as I call it, I can't bring myself to commit being 100% in stocks. I did adjust my AA on Vanguard to be higher than what was originally and I'm going to take my time getting there.

That's not a problem at all. If you look at success rates on FIRECalc versus AA, it's pretty flat from about 35/65 to 95/5. It drops pretty sharply below 35% equities though, I would be concerned for anyone going that low, unless they had such a large relative stash (low WR%) that AA just doesn't matter.

I also figure that if history 'rhymes' rather than repeats, that 35% is on the edge, and maybe that edge would shift one way or the other in the future? So I'd rather be a bit in from that edge, say 40-45% minimum EQ? Maybe 90% on the top end?

-ERD50

 

[audreyh1]

Again, not recommending it, but just trying to explain it - your analysis is interesting, but it is based on a 4% WR rate. What if someone looked at a lower WR, one that was 100% historically safe with 100% equities? If that WR was sufficient for them, they might like the idea of having the very significant chance of having a LOT of money later.

For example, $1M, 3% initial WR, 30 years, 100% equities:

The lowest and highest portfolio balance at the end of your retirement was $443,031 to $10,345,067, with an average at the end of $3,682,503

With same #'s but 50% equities:

The lowest and highest portfolio balance at the end of your retirement was $554,953 to $5,063,705, with an average at the end of $1,795,570

Downside risk of 100% equities is ~ $100K, upside benefit is ~ $5M, almost $2M on average.

At 45 years, it becomes more dramatic:

100% EQ: $285,352 to $44,173,166, with an average at the end of $8,096,124

50% EQ: $268,129 to $9,265,891, with an average at the end of $2,447,189.

No downside historically, the upside is huge. Once again, not recommending it, just sayin'

-ERD50

A lot of money later because they will ramp up their spending when they are a whole lot older? Or leave more to their heirs?

Sure - if someone wants to keep income low during their lifetime and leave the max for their heirs then a very high allocation to equities makes sense as I already mentioned. Especially if they arrange to just live off the dividends.

Personally, I'm interested in spending more during my lifetime and not having a really large terminal value portfolio which is why I prefer the % remaining portfolio approach. Since my annual income is based on my portfolio value at the end of each year I prefer to have less volatility, so I am closer to a 50% allocation. I can be more aggressive with withdrawal rate and lower the average ending portfolio if I choose - although that also will increase income volatility, but with higher average income it probably doesn't matter.

 

[ERD50]

A lot of money later because they will ramp up their spending when they are a whole lot older? Or leave more to their heirs? ...

I wasn't attempting to get into the reasons why someone might want to do that. There could be many. I was just explaining the possibilities.

But here's one reason an old codger might want to have a large stash near end-of-life:

With a really large stash, that old man could have two beautiful young women!

Personally, I'm interested in spending more during my lifetime and not having a really large terminal value portfolio which is why I prefer the % remaining portfolio approach.

Consider the "retire again & again" approach. From what I've seen, that is the most efficient method for leaving less on the table.

-ERD50

 

[audreyh1]

Consider the "retire again & again" approach. From what I've seen, that is the most efficient method for leaving less on the table.

-ERD50

The % remaining portfolio is close, just that it backs off during the down times instead of ignoring a rapidly dropping portfolio and going full steam ahead with inflation adjusted income. It thus recovers more quickly after a bad run, and never runs out of money.

It will tolerate higher withdrawal rates. You can go above 4.3% before the portfolio starts to shrink more rapidly in the later years, but you still won't run out of money. It's a wilder ride, but you get to spend more in early years.

Psychologically I don't see myself flying blind with withdrawal increases while the portfolio is shrinking fast using the constant method after I've ratcheted up (retire over and over) to a bad sequence of returns.

 

[Spanky]

With a really large stash, that old man could have two beautiful young women!

Maybe but why?

 

[Mulligan]

In theory I should be 100% in stocks after reading this thread but I'm in the bucket for chickens or the chicken bucket as I call it, I can't bring myself to commit being 100% in stocks. I did adjust my AA on Vanguard to be higher than what was originally and I'm going to take my time getting there.

Dont feel all by yourself Fedup. The US Bond market is twice the size in value of the entire US equity market, so some people (and entities) must think they serve a purpose still.

 

[Lsbcal]

Personally my focus includes the worst case minimum a portfolio will suffer. For instance, a portfolio of $1M that goes down to $200k before ending much higher in 30 years would not be at all attractive. Just survival rates don't give a full enough analysis.

 

[audreyh1]

Personally my focus includes the worst case minimum a portfolio will suffer. For instance, a portfolio of $1M that goes down to $200k before ending much higher in 30 years would not be at all attractive. Just survival rates don't give a full enough analysis.

Talking about the %remaining withdrawal method:

Well even in the 6% withdrawal rate case you would drop down to $270K during the worst year, but end with a portfolio a little over $300K after 30 years. (50% total stock market, 50% 5-year treasuries). The average ending portfolio was almost $600K (real). So on average, you still had over half the inflation adjusted portfolio remaining, and the worst case ending portfolio was around $300K real. Historical data from 1870.

I consider the 4.35% case to be the "break even" case with the above portfolio, where you end up where you started on average in real terms, with a $1M portfolio. And in this case you can see that the portfolio scenarios are staying fairly steady rather than growing or shrinking over the long run. Lowest ending portfolio is a little above $500K, and a drop to $400K during the worst year. Since these ending portfolios are inflation adjusted, it might not seem like as much of a drop to a retiree looking at the nominal numbers.

The thing is, even if you start out with a higher withdrawal rate where income can drop faster during bad market runs, you still have higher income overall compared to several of the lower withdrawal rate scenarios even in the worst historical case. Mainly you've drawn down your portfolio more aggressively, and it will recover more slowly, and you might be stuck with somewhat lower incomes late during the period.

6% - you're pulling $60K, 4.35% you're pulling $43.5K, 3.5% you're pulling $35K initially. Those are huge differences in starting income, so not to be taken lightly.

Here is an illustration for the 30 year period:

% withdrawal remaining portfolio	$1M starting portfolio income	$1M starting portfolio lowest income	average ending portfolio	income from average ending portfolio	lowest ending portfolio	income from lowest ending portfolio
6.00%	$60,000	$16,248	$593,418	$35,605	$298,371	$17,902
5.00%	$50,000	$17,294	$815,142	$40,757	$409,854	$20,493
4.50%	$45,000	$17,470	$954,171	$42,938	$479,758	$21,589
4.35%	$43,500	$17,481	$1,000,171	$43,507	$512,306	$22,285
4.25%	$42,500	$17,476	$1,032,020	$43,861	$518,901	$22,053
4.00%	$40,000	$17,419	$1,115,994	$44,640	$561,123	$22,445
3.50%	$35,000	$16,593	$1,304,200	$45,647	$655,754	$22,951
3.33%	$33,300	$16,210	$1,374,917	$45,785	$691,310	$23,021
3.25%	$32,500	$16,018	$1,409,464	$45,808	$708,681	$23,032
3.00%	$30,000	$15,369	$1,522,919	$45,688	$765,726	$22,972

My instincts are not to really push it above 4.5% withdrawal rate (for 30 years), even though I'm leaving more on the table initially.

 

[Lsbcal]

...
I consider the 4.35% case to be the "break even" case with the above portfolio, where you end up where you started on average in real terms, with a $1M portfolio. And in this case you can see that the portfolio scenarios are staying fairly steady rather than growing or shrinking over the long run. Lowest ending portfolio is a little above $500K, and a drop to $400K during the worst year. Since these ending portfolios are inflation adjusted, it might not seem like as much of a drop to a retiree looking at the nominal numbers.
...

Nice table and I'm guessing it is for a 50/50 AA. It should have a column with the worst year drop (blue in your above comment Audrey). And we should remember the end of that year was not necessarily the worst portfolio value for that year.

I just don't think any retiree should count on their ability to withstand the stress of too much portfolio decline, especially in the early years (maybe age 50 to 70). I know it was a long time ago and in a hugely different world economy, but the numbers I showed from VPW for a 60/40 AA could, I think, really happen (hope not!). That shows a 27% portfolio fall over 4 years from 1929 to 1933. Note that the nominal values were actually lower (45% drop) because of deflation! One could only imagine the headlines if we were to have a real extended depression like back then. The 2008-2009 decline was a cake walk compared to that.

And a quibble, nobody should be fooling themselves with the nominal portfolio value. Inflation adjusted value is what counts. The nominal columns in the VPW tool are worthwhile to remind us of this fact.

Do I sound like I'm preaching? Hope not but I can scare myself with this stuff. Sweet dreams.