How hard is it to RE at the market top (2014)?

[robnplunder]

This raises another interesting point. The SWR studies are basically studies of investing in the USA...a more diversified portfolio of emerging markets and international markets should provide some decrease in correlation and thus better results with less volatility ---in theory.

... plus real estate (rental income, REIT), commodity, lottery tickets (), etc..

Much of the world was devastated after WWII. Yet, people survived.

If and when the going gets tough, retirees will stop traveling, ordering from Amazon, going out to eat, buying fancy cars, etc... It's not the end of the world.

... b b b but we want to survive in style .

 

[daylatedollarshort]

I think a bit of perspective is order. We have roughly 40 million American over the age 65, I am sure that the vast majority of them retired. I bet we have hundreds of million of people retire since WW2, and only tiny percentage of them used tools like FIRECalc.

The over 65 group has the lowest percentage of any age who live in poverty. So somehow all this people retired without running out of money. Kinda of hard to imagine when you think about it

A recent Vanguard study showed most current retirees still had pensions and many worked. The people who post here living off portfolio income appear to be in the minority -

COLUMN-Surprise: Even wealthy retirees live on Social Security and pensions | Reuters

"The findings are all the more striking because the big buzz in the retirement industry these days is about how to generate income from nest eggs. That includes creation of income-oriented portfolios, systematic drawdown plans and annuity products that act as do-it-yourself pensions. Yet few retirement account holders actually are tapping them for income. The Investment Company Institute reports that just 3.5 percent of all participants in 401(k) plans took withdrawals in 2013."

The Consumer Expenditure Survey also does not show dividends and interest to be a big portion of income for most senior households -

http://www.bls.gov/cex/2011/Standard/age.pdf

 

[NW-Bound]

... b b b but we want to survive in style .

You mean survivors like Robinson Crusoe do not have style?

Recently, I read the account of a young couple who moved to Alaska to scratch out a living. What impressed me most was that they got to Alaska on foot! Now, that's style, not sitting at home surfing Amazon and eBay.

 

[photoguy]

It's not that it's a double dip, it's that the present day market dip is being added to the front of every historical market dip (including doubles) during your FIRECalc testing. Making them all worse, and possibly affecting many scenarios.

Firecalc doesn't take into consideration the history leading up to the current date, but if you compute the conditional probability correctly (the subset of scenarios which match the user's trajectory) there shouldn't be any double counting.

Firecalc doesn't do any of this conditional probability, so the user has to do it themeselves or find another tool.

Yeah, I like mean reversion. I see it in every market recovery. Except bubbles. It's also a good reason to use FIRECalc. If there is mean reversion, it's in there.


We've just seen a really good example of it since 2008. At 10%/year we should only be up 61% from the bottom, not hitting new highs. I suspect emotion, uncertainty, and panic has quite a bit to do with prices being lower than they should be at the bottom and then recovering faster than 10%/year when things turn out OK. No, it's not all mean reversion, but I think it is a significant factor in many cases. Now, Japan or NASDAQ are certainly arguments that there are other factors at play.

You see mean reversion, but I just see this as a normal sequence in random returns. Equity returns have high standard deviation and are generally positive in most years. If you look at S&P 500 returns something like 3/4 are positive years. So a streak of a few good years is expected anyway.

I think this really needs a statistical test. Humans are predisposed to see patterns in randomness (e.g. hot hand fallacy) and constantly underestimate how frequently you might get a string of positive results (e.g., getting a long sequences of heads in coin flipping).

If you think it's OK to use FIRECalc with your March 2009 portfolio value, the interpretation of the results is up to you. It's only a tool that does what it does.

I'm not saying this at all. My main disagreement is with ERD50's statement that "you don't need to re-adjust after a market drop".

The additional information would indicate you are on a higher failure rate path. It's just that FIRECalc can't use that additional information in a meaningful way, nor do we really have enough historical data to be statistically relevant even if FIRECalc could do it.

I mostly agree with what you are saying here. I'll just note that urn2bfree asked about SWR theory which is broader than FIRECALC and hence I've been thinking more generally in my responses.

Yes FIRECALC is extremely limited. But we know in a qualitative sense which way the risk is going to move (and I think this is actionable). One might also argue that the MC methods can give a sense of the magnitude of the increased risk.

 

[ERD50]

...
My main disagreement is with ERD50's statement that "you don't need to re-adjust after a market drop". ...

I'm back, so let me explain it this way.

First, I think it is easier to discuss if we use a 100% historically safe WR, which based on defaults is a 3.59% WR.

The reason that "you don't need to re-adjust after a market drop", is because that value was historically safe in every single cycle. Every one. So no, you "you don't need to re-adjust after a market drop", because that market drop is just one of the squiggly lines on the output chart that dips and rises and passes with a 100% historically safe WR - without re-adjustment.

How can it be otherwise?



Now, maybe you are talking about taking a 4% WR with 95% success, and trying to identify when you are on one of the 5% failure paths, and then readjusting? OK, I don't think FIRECalc gives the fine granularity to look at this, other than trying to use the single year spreadsheet output (but that is only valid for 30 year cycles, IIRC), look at where the problems occur, and see if you can formulate a spending adjustment that will recover. I can tell you from previous studies that 1966 is one of the failure start years.

And I think you will find it takes some serious cuts for some serious time to recover - it isn't simply a matter of hamburger versus steak once a week, or dropping cable. From my earlier post:

I just tried this in FIRECalc -

FIRECalc: A different kind of retirement calculator << link to my data entries

After default 4% WR, 30 years, ~ 95% success (but with 40K/1M for easy calculation). Then try to get to 100% with spending adjustments. I found it took cutting spending in half, in the fourth year, and continuing this reduced spending for 6 more years to get back to 100%.

That might work for someone with lots of discretionary spending, but I would not care to live 6 of my first 10 years of retirement at a 50% spending cut level.

Personally, I'd rather go with a constant 3.59% WR which succeeds 100% historically for 30 years, with no adjustments.

You can try slicing that different ways, it would be interesting to see other cuts/times that would get back to 100%, so please report back if you run some.

I also suspect you'd have a tough time identifying the failures other than in the rear-view mirror. A scary market drop followed by a decent recovery and moderate inflation might be a success path, while a more modest drop followed by years of inflation might be a killer.

But, we might learn something if we could segregate the worst 10% of the starting years and study them. We might see some commonalities.

-ERD50

 

[urn2bfree]

I am skeptical that returns are as random as a coin toss or roulette wheel. Economic behavior alters at peaks and after crashes - unlike coin tosses and roulette wheels, markets DO have memory of what happened before so they are related or correlated and not random

...and I think if one were to study hundreds or thousands of years of markets you will find that barring complete destruction of a market, (hence the need for diversification) what goes up will come down but also what goes down will go up.

I don't know the answer to this but I would be interested to know the record number of consecutively negative years in the history of all markets...so far, What is it? 3? 4? As pointed out in that article about Japanese markets, even in the horrible lost 20+ years of the Nikkei, a rebalancing person with a 50/50 AA would have made money despite a two decades long downward trend...

 

[clifp]

A recent Vanguard study showed most current retirees still had pensions and many worked. The people who post here living off portfolio income appear to be in the minority -

COLUMN-Surprise: Even wealthy retirees live on Social Security and pensions | Reuters

"The findings are all the more striking because the big buzz in the retirement industry these days is about how to generate income from nest eggs. That includes creation of income-oriented portfolios, systematic drawdown plans and annuity products that act as do-it-yourself pensions. Yet few retirement account holders actually are tapping them for income. The Investment Company Institute reports that just 3.5 percent of all participants in 401(k) plans took withdrawals in 2013."

The Consumer Expenditure Survey also does not show dividends and interest to be a big portion of income for most senior households -

http://www.bls.gov/cex/2011/Standard/age.pdf

This isn't really surprising IRA were around until 1974 and 401K until 1978.
Neither were particularly popular until the late 1980s in largely due to cut back of pension by relatively small number of large companies with pension.

So most people retiring today have had most 35 years to contribute to a IRA/401K and often only been doing so for 15 to 20 years. Obviously a typical retired who retired in say 2000 had had even less time to contribute to saving plan.
The 85% of companies had pension statistic is misleading because it is only companies that offered a retirement plan. Up until the IRA, the majority of American had no saving vehicles available at all.

 

[pb4uski]

I wonder if there might be some overthinking going on here...

 

[REWahoo]

I wonder if there might be some overthinking going on here...

Now we'll have to discuss the various approaches to overthinking, with special emphasis on the historical view vs. technical analysis.

 

[springnr]

Firecalc is a data point, an input for ones consideration. In the case of Joe and Mary being caught in the twilight zone, their realities are a little different.

Joe using last year’s information is sitting in the first row of the roller coaster arms extended making funny noises.
Mary using this year's information is presented with data implying now might not be the best time to start withdrawing.

 

[NW-Bound]

...I think you will find it takes some serious cuts for some serious time to recover - it isn't simply a matter of hamburger versus steak once a week, or dropping cable...

I wholeheartedly agree with you that if one finds himself in a bad period as happened in the past, it takes a lot of expense cut back to make a difference. Following is some info I get from FIRECalc, using a starting portfolio of $1M with an AA of 50/50 which works out better in the worst case than the default 75/25.

The portfolio reaches the lowest point in Year 5, then recovers towards Year 10 if the WR is sufficiently low. Note how one can lose 40% even when not making any withdrawal.

Worst Case at 5 Yr 10 Yr
0% WR: $626K $715K
1% WR: $590K $632K
2% WR: $544K $550K
3% WR: $498K $468K
4% WR: $452K $386K

See how a bad stretch in the past lasted more than 10 years, and the portfolio could not recover after 10 years even without any WR.

Have I scared everybody yet?

So, can you wait 10 years to retire?

 

[haha]

WOW. Did the 3% and 4% wr runs go on to fail?

Ha

 

[NW-Bound]

The 4% WR went under at Year 25. The 3% WR stabilized at $500K until Year 30.

Again, these were historical worst case that happened during the 60s or the Vietnam war era. Inflation was the real killer.

 

[CaliforniaMan]

I wholeheartedly agree with you that if one finds himself in a bad period as happened in the past, it takes a lot of expense cut back to make a difference. Following is some info I get from FIRECalc, using a starting portfolio of $1M with an AA of 50/50 which works out better in the worst case than the default 75/25.

The portfolio reaches the lowest point in Year 5, then recovers towards Year 10 if the WR is sufficiently low. Note how one can lose 40% even when not making any withdrawal.

Worst Case at 5 Yr 10 Yr
0% WR: $626K $715K
1% WR: $590K $632K
2% WR: $544K $550K
3% WR: $498K $468K
4% WR: $452K $386K

See how a bad stretch in the past lasted more than 10 years, and the portfolio could not recover after 10 years even without any WR.

Have I scared everybody yet?

So, can you wait 10 years to retire?

Ugh, makes me want to go back to work. No not really! At the worst it means I have 10 years to party!

 

[robnplunder]

So, can you wait 10 years to retire?

Nope. I will take my chances even if I have to live in Alaska, hunting for food, wearing animal hides, and live in a one room cave. It's not my style but, heck, it's better than going back to work after FIRE backfires.

 

[NW-Bound]

And for us pre-geezers, remember that towards the middle of the 30-year FIRECalc terrible run, SS will come online and save our tush. In fact, for many of us, SS becomes available a lot sooner than that. Let the youngster ERs sweat, we geezers keep on surfing Amazon.

 

[REWahoo]

And for us pre-geezers, remember that towards the middle of the 30-year FIRECalc terrible run, SS will come online and save our tush.

+1

Been there.
Done that.
Got the T-shirt (at Goodwill).

 

[NW-Bound]

Speaking of SS, the other day when I finally sat down and figured out how much SS I will be getting at 62, 66-1/2 (full retirement age or FRA), and 70, I entered each scenario into FIRECalc along with my stash, then raised WR until the 30-year run failed.

What I saw was that claiming SS at FRA for me would support the highest WR before failure. However, the difference between the 3 SS cases was less than $2K/yr. They all showed that I could spend a whole lot more money than I do now. It was enough for me to buy a 3rd home (but I would be crazy to).

The above is just academic for me at this point, as I still have a few years to ponder this, and besides I am not planning to push my WR to such high values.

PS. I included my wife's SS in the above runs. Her SS is less than mine, but more than the 1/2 spousal freebie.

 

[daylatedollarshort]

The 4% WR went under at Year 25. The 3% WR stabilized at $500K until Year 30.

Again, these were historical worst case that happened during the 60s or the Vietnam war era. Inflation was the real killer.

Thanks for the stats, NW-bound. I think there are many here who get this and are okay with the risk and for some this may be a surprise.

As a total coward, my strategy was to look for ways to cut expenses to live on < (pensions + SS + (0 - 1 real returns in conservative investments)) without cramping or maybe even improving our overall lifestyle. That has been our main hobby these past few years since DH retired.

Good for you guys who are better risk takers. I would be freaking out if the market tanked like this or even thinking about the possibility of surviving a drop like this in our retirement years.

 

[NW-Bound]

FIRECalc shows that even with zero withdrawal, a 50/50 portfolio lost 40% after 5 years in the worst case. However, that was with inflation adjustment. In nomimal values, the portfolio might look constant or might even increase, but when you have inflation around 8% a year like that bad period, after 5 years your stash's buying power gets decimated.

 

[ERD50]

Thanks for the stats, NW-bound. I think there are many here who get this and are okay with the risk and for some this may be a surprise.

As a total coward, my strategy was to look for ways to cut expenses to live on < (pensions + SS + (0 - 1 real returns in conservative investments)) without cramping or maybe even improving our overall lifestyle. That has been our main hobby these past few years since DH retired.

Good for you guys who are better risk takers. I would be freaking out if the market tanked like this or even thinking about the possibility of surviving a drop like this in our retirement years.

A conservative WR is great if you can fit your lifestyle into it, few really want to take on too much risk.

But there may not be as much difference in these scenarios as you think (or you may even have it backwards?). Even a zero real return portfolio will allow for a 3.33% WR adjusted for inflation each year (simply portfolio/30 years, since both portfolio and expenses are zero real return). That ignores volatility, which may lower this a bit - but follow along for a reference point.

For a historically 100% safe, 30 year period, EQ/Fixed AA:

100/0 AA supports a 3.42% WR
75/25 AA supports a 3.59% WR
50/50 AA supports a 3.71% WR
25/75 AA supports a 3.49% WR
0/100 AA supports a 2.55% WR

Remember, these survived the worst cycles in the FIRECalc database.

So the more 'conservative' investments did not provide the retiree as much safety as those with higher stock allocations. And the most 'conservative' does not appear to have provided even 0% real returns over that period. So maybe 0-1% real returns on 'conservative' investments is no slam-dunk, and maybe cannot be counted on? A 1% real return would support 3.84% WR (again, straight line calculations for simplicity). A 3.84% WR is higher than ANY of the above AA's.

I think the combination of inflation and lower growth opportunities (relative to equities) is what makes 'conservative' investments such poor performers in a stress test such as 1966-1995 30 year period.

If my observations are close, we 'risk takers' may be taking a more conservative withdraw method than you are in counting on at least 0% real return from conservative investments?

The 4% WR went under at Year 25. The 3% WR stabilized at $500K until Year 30.

Again, these were historical worst case that happened during the 60s or the Vietnam war era. Inflation was the real killer.

What year specifically did you use, I may run some variations on that? 1966 seems to keep coming up as the killer year.

Speaking of SS, the other day when I finally sat down and figured out how much SS I will be getting at 62, 66-1/2 (full retirement age or FRA), and 70, I entered each scenario into FIRECalc along with my stash, then raised WR until the 30-year run failed. ....

PS. I included my wife's SS in the above runs. Her SS is less than mine, but more than the 1/2 spousal freebie.

I don't think you will see any great benefit to delaying SS on a 30 year run, and it depends at what age you start the run - the younger you start it, the fewer 'payoff' years get included.

It really is 'what if' longevity insurance, and maybe a spousal beneficiary benefit, depending on specifics.

-ERD50

 

[NW-Bound]

... the most 'conservative' does not appear to have provided even 0% real returns over that period. So maybe 0-1% real returns on 'conservative' investments is no slam-dunk, and maybe cannot be counted on? A 1% real return would support 3.84% WR (again, straight line calculations for simplicity). A 3.84% WR is higher than ANY of the above AA's...

In a past post, daylatedollarshort talked about being heavily in TIPS. I do not follow it nor have any, so do not know what the yield is, or how it has been doing recently.

 

[ERD50]

In a past post, daylatedollarshort talked about being heavily in TIPS. I do not follow it nor have any, so do not know what the yield is, or how it has been doing recently.

Yes, TIPS will provide whatever you purchased them at (certainly above zero, IIRC they were over 3% real for a while), but I'm not sure many of us have been able to buy enough to make them a really significant part of our portfolio.

Perhaps DLDS has?

They probably do make good sense for diversification, I have not followed them much, I was thinking you could only buy $X/year?

-ERD50

 

[haha]

I think people who argue for higher allocations to price-variable assets, usually where at least part of the return is expected to come from selling appreciated securities, are those who are comfortable with statistical thinking. Statistical thinking is very nice, however we may forget that this derives from the law of large numbers, while each of us one person living his one life. And even Firecalc is based on a very flimsy data set. Just running through a 60 year sequence in 30 different trips with 30 different beginning and ending periods, in no way creates 30 independent trials. ( same principle holds with the longer history accessed by Firecalc)

Much more of this is luck than we care to admit. This is one reason why I am puzzled when I hear people reject taking SS at 70, because Firecalc gives them a "better" result when starting at some earlier age, and therefore getting a smaller monthly check.

I have invested assets =~50x my spending needs. This does not include much travel or unusual medical expenses, but neither does it involve clipping coupons or doing without anything that I want. If I do certain things it will cost me more. But I own my home in the city and neighborhood where I want to be, with weather I find good to very good, eat as well as I want, etc. Nevertheless, feeling unsinkable I believe would be as misguided as it was for those who boarded the Titanic in April of 1912.

Ha

 

[haha]

Yes, TIPS will provide whatever you purchased them at (certainly above zero, IIRC they were over 3% real for a while), but I'm not sure many of us have been able to buy enough to make them a really significant part of our portfolio.

Perhaps DLDS has?

They probably do make good sense for diversification, I have not followed them much, I was thinking you could only buy $X/year? -ERD50

That is I-bonds.