cFiresim SEVERELY overestimates success rates for Mustachians

[dragoncar]

Fortunately there is no equivalent of the efficient market hypothesis for the emergence of totalitarian regimes.

Since the market is efficient, it will tank long before you are personally aware of SHTF.   This will leave you with approximately bupkis to flee.  If index funds existed in 1933, you would have had a decent 10 years of negative returns to fund your escape.

[dragoncar]

Or just keep your money in a bank in Switzerland.

Yeah, sure.  Do you?

Not that the US couldn't touch it.  If Trump, in a fever dream,  told Switzerland "wire the money or bombs" I think they would comply.  If they didn't your money would still be gone.

After all, they caved to reporting rules due to no more than economic threats (http://www.cnbc.com/2017/01/02/swiss-banking-secrecy-nears-end-following-new-tax-rules.html)

[Tyson]

Or just keep your money in a bank in Switzerland.

Yeah, sure.  Do you?

Not that the US couldn't touch it.  If Trump, in a fever dream,  told Switzerland "wire the money or bombs" I think they would comply.  If they didn't your money would still be gone.

My point is that things like nationalization don't happen over night.  There's usually a build up to it.  And yes, plenty of people see it and get out before it happens.  That's part of why nationalization and seizure of assets gets implemented - too many people leaving and taking their money with them. 

So here's my tip - when you see massive numbers of wealthy people fleeing your country - join them ASAP. 

[maizefolk]

Fortunately there is no equivalent of the efficient market hypothesis for the emergence of totalitarian regimes.

Since the market is efficient, it will tank long before you are personally aware of SHTF.   This will leave you with approximately bupkis to flee.  If index funds existed in 1933, you would have had a decent 10 years of negative returns to fund your escape.

Kristallnacht happened in 1938. The german stock market didn't start to decline until 1944.

[dragoncar]

Since it hasn't been linked in this thread yet, William Bernstein's short treatment of how investment risk differs from other risks: http://www.efficientfrontier.com/ef/901/hell3.htm

Don't get me wrong, I don't hide my money in a mattress, think collapse of the US is likely, etc.  But I also don't kid myself that the number cFiresim spits out is the end-all and be-all of success rate, for broad definitions of success.  Once you get to the point that non-investment risks are on par with investment risks, it doesn't make sense to increase your savings.

Fortunately there is no equivalent of the efficient market hypothesis for the emergence of totalitarian regimes.

Since the market is efficient, it will tank long before you are personally aware of SHTF.   This will leave you with approximately bupkis to flee.  If index funds existed in 1933, you would have had a decent 10 years of negative returns to fund your escape.

Kristallnacht happened in 1938. The german stock market didn't start to decline until 1944.

I was hardly there, but according to Wikipedia, we are talking hyperinflation, then 30% unemployment rate, followed by closure of exchanges in 1935.  If that happened in the US, you would not have much left.

[dragoncar]

So here's my tip - when you see massive numbers of wealthy people fleeing your country - join them ASAP.

Yes, this is exactly why I always ask my rich neighbor for stock tips. 

[The Money Monk]

The important factor with mustachians is that behavior would definitely be not be static - even slight behavioral modifications during a downturn could easily turn that 40$ back up to near 100% success.

Seriously, what sane person (especially a mustachian) would not modify their spending at all or find a temporary income source in the event of a giant downturn? Nobody mustachian is just going to keep spending the $40k a year until it runs out.

Alternatively, getting a part time job bringing in just $10k or $15k a year for a year or two during these low times would also make you much safer. Doing both (temporarily reducing spending AND getting additional income) during these times would virtually guarantee success I imagine.

[Playing with Fire UK]

If something forces me to flee the country and I have enough assets to buy or bribe my way out with my family and shirt on my back, I call that a win. Similarly, if I or a family member catches an awful expensive disease that isn't covered by insurance or state health care and that wipes out all my money, I call that a win.

There are some things that money can't buy, for everything else, there's Mastercard money.

The purpose of the SWR isn't to protect me from aliens and Daleks, it is to make my no worse financially off than if I'd stayed at work.

[gerardc]

The trick is to be ready and willing to endure pain and sacrifice, a war, surviving on $0 eating insects, getting all sorts of diseases, seeing your family killed, having your limbs cut, etc. Once you accept that as a possibility, your success rate is 100%.

[dragoncar]

If something forces me to flee the country and I have enough assets to buy or bribe my way out with my family and shirt on my back, I call that a win. Similarly, if I or a family member catches an awful expensive disease that isn't covered by insurance or state health care and that wipes out all my money, I call that a win.

There are some things that money can't buy, for everything else, there's Mastercard money.

The purpose of the SWR isn't to protect me from aliens and Daleks, it is to make my no worse financially off than if I'd stayed at work.

Yup.  And that's why I stick with a 4% SWR.  Daleks don't GAF if you have a 1% SWR.

[Monkey Uncle]

Why continually worry about outlying statistical situations that can neither be accurately predicted or mitigated against with a passive investment allocation?  It's better to invest in adaptability skills in those potential situations.  For the 4% rule to fail there has to be both a precipitous drop in asset value soon after draw-down begins AND long-term very low real returns. Even if this happens, someone with only half of FI assets remaining is better suited to take advantage of a universally bad situation, if they adapt to it.

Vanguard says that P/E is a reliable indicator of future returns.  No formula is going to give you the day or week when the market corrects or tumbles, but it appears there are warning signs and strategies that can help. 

The 4% rule failed after the great depression, and when enough history plays out, it might also fail during some of the severe downturns of the 2000's.  Guess what?  CAPE market valuations are back up in lime with the great depression started (although I would argue it's not quite as bad now as it was then).  Don't give yourself a false sense of security that the idea from a single research paper puts the question of the safety of a 4% withdrawal rate to rest, at least with these high market valuations.

The bolded statement is not true, and Sol and others have already debunked similar statements that you made in the 4% rule thread.  Go run cFiresim at either a 60/40 or 75/25 stock/bond allocation, and you will see that the only failures occurred for start years in the mid and late 1960's, due to the 1970s stagflation era.  The Great Depression did NOT cause a failure of the 4% rule.  And if you run a 17-year sim that captures the Y2k retiree, he/she is doing demonstrably better than any of the 1960s/70s failure trajectories were doing 17 years in.  I'm pretty sure this has been pointed out to you before also.

Please stop spreading misinformation.

[runewell]

The 4% rule failed after the great depression, and when enough history plays out, it might also fail during some of the severe downturns of the 2000's. 

The bolded statement is not true, and Sol and others have already debunked similar statements that you made in the 4% rule thread.  Go run cFiresim at either a 60/40 or 75/25 stock/bond allocation, and you will see that the only failures occurred for start years in the mid and late 1960's, due to the 1970s stagflation era.  The Great Depression did NOT cause a failure of the 4% rule.  And if you run a 17-year sim that captures the Y2k retiree, he/she is doing demonstrably better than any of the 1960s/70s failure trajectories were doing 17 years in.  I'm pretty sure this has been pointed out to you before also.

Please stop spreading misinformation.

All you have to do is look at Table 3 in the original Trinity Study and you'll see that for a 30-yr time frame with an adjustment for inflation between 1926-1995 the success rate was only 95%-98% for a portfolio that had at least 1/2 stocks, 71% success for 25% stocks, and only 20% success for 100% bonds.  Not sure why my statement is true!  Let's see what we think about you not worrying about the year-2000 retiree...

cFiresim is only one simulation tool, and it doesn't even agree with the results of the Trinity study.  As long as people are going to quote the 4% rule I'm going to stick with one of the sources of that very rule.  But let's use it to simulate a 30-yr outcome starting at 2000.  We don't need to stop at 17 years of simulation, we can do it in two stages to try and add on the other 13 years to get a proper 30-yr time frame, shall we?

I used the cFiresim default calculations with a 2017 retirement date and a 2034 retirement end year.  Someone starting with $1M in 2000 would only have about $629K of their portfolio left in a US market - with a lofty CAPE ratio of 30 - we can't even factor in how expensive the market is, but that doesn't sound very safe at all. 

Put that $629K back in the simulator and run it for another 13 years.  Adjusting expenses for inflation, $40,000 in 2000 is now $57,545 in 2017, so that's our starting expense number, and the simulations years are 2017-2030.  Now let's count how many times the portfolio fails over the 134 historical simulations.  The answer is 48!  That's the best way I can think to simulate a 30-yr horizon starting with 2000, and I get a 36% failure rate with this method.  If we assume that market returns will not be as good for the next 30 years (CAPE research shows this to be very likely) then you would probably get a significantly higher success rate.

Please stop spreading misinformation.

[GenXbiker]

If we assume that market returns will not be as good for the next 30 years (CAPE research shows this to be very likely) then you would probably get a significantly higher success rate.

Please stop spreading misinformation.

Market returns being not as good due to a current high CAPE would result in a "lower" success rate, higher "failure" rate.

[runewell]

If we assume that market returns will not be as good for the next 30 years (CAPE research shows this to be very likely) then you would probably get a significantly higher success rate.

Please stop spreading misinformation.

Market returns being not as good due to a current high CAPE would result in a "lower" success rate, higher "failure" rate.

Yeah thanks.  That's what I meant.

I know people aren't going to like my two cents on this board, but what can I say, I'm an actuary.  My job is to come up with the correct insurance prices and make sure that there are enough reserves on hand so that an insurance company can pay all of its claims.  I naturally need to take a pessimistic view of modeling to account for the unknown or the unexpected so that my company is strong.  (My company is Chubb so my company is pretty strong in any case) 

My concern is that not only should we be more pessimistic with current assumptions but that people immediately assume assumptions are safe and assume that moving away from them is also safe.  A 5% withdrawal rate instead of a 4% withdrawal rate?  Now instead of a 5%/2%/5%/29%/80% failure rate depending on a 30-yr allocation, we have a 15%/17%/24%/73%/83% failure rate.   

I hope you all have financially secure early retirements.  I'm hoping to quit around age 60 with almost $2M.  I would retire earlier than that but I am reluctant to take any chances with health insurance and costs which is the biggest unknown apart from stock returns.  Perhaps I will be able to shave a couple years off that if investment returns are good and/or markets are not highly valued, only time will tell.  I don't intend to constantly be a Debbie Downer about everything, I just worry that there is an excess of optimism on this board and a bit more pessimism would not be a bad thing.

[dividendman]

I hope you all have financially secure early retirements.  I'm hoping to quit around age 60 with almost $2M.

I think few of us "hope" to have financially secure retirements. We know that we can do it with high probability because we have skills and assets that allow is to adapt to changing conditions.

It's fine that you require the security of being 60 with $2M. Everyone is different.

I think most people on this forum would rather enjoy doing what they want from 30, 40 etc to 60 and beyond and maybe have to earn more money later rather than wasting the prime of their lives doing things they don't want so the later years can have a higher probability of comfort.

That's the trade-off.

Why don't you work till you're 90 and have $5M? Isn't that even safer? You might live to 140, you never know.

[maizefolk]

All you have to do is look at Table 3 in the original Trinity Study and you'll see that for a 30-yr time frame with an adjustment for inflation between 1926-1995 the success rate was only 95%-98% for a portfolio that had at least 1/2 stocks, 71% success for 25% stocks, and only 20% success for 100% bonds.  Not sure why my statement is true!  Let's see what we think about you not worrying about the year-2000 retiree...

cFiresim is only one simulation tool, and it doesn't even agree with the results of the Trinity study.  As long as people are going to quote the 4% rule I'm going to stick with one of the sources of that very rule.

Runewell, you do realize you're not being very consistent here, right? You've been dismissing the trinity study on these threads for the last few days as a twenty year old out of date study, people have pointed out the wide range of updated research and simulation tools, and now you've turned around and are holding up the original 1995 paper as the gold standard.

One example

But let's use it to simulate a 30-yr outcome starting at 2000.  We don't need to stop at 17 years of simulation, we can do it in two stages to try and add on the other 13 years to get a proper 30-yr time frame, shall we?

I used the cFiresim default calculations with a 2017 retirement date and a 2034 retirement end year.  Someone starting with $1M in 2000 would only have about $629K of their portfolio left in a US market - with a lofty CAPE ratio of 30 - we can't even factor in how expensive the market is, but that doesn't sound very safe at all. 

Again, we've already discussed how the CAPE ratio is misleading, in fact you're the one who linked to the study showing that it is currently overestimating how overpriced the market is, perhaps by as much as 50%.

Put that $629K back in the simulator and run it for another 13 years.  Adjusting expenses for inflation, $40,000 in 2000 is now $57,545 in 2017, so that's our starting expense number, and the simulations years are 2017-2030.  Now let's count how many times the portfolio fails over the 134 historical simulations.  The answer is 48!  That's the best way I can think to simulate a 30-yr horizon starting with 2000, and I get a 36% failure rate with this method. 

I'm assuming you clicked the "adjust for inflation" box in cFireSim, so that $629K is in 2000 dollars. So you'll want to either keep spending constant at $40k/year which is probably the simplest approach, or also adjust $629k in remaining net worth up to the $912k it is in 2017 dollars.

If we assume that market returns will not be as good for the next 30 years (CAPE research shows this to be very likely) then you would probably get a significantly higher success rate.

Please stop spreading misinformation.

Again, you yourself pointed out that the CAPE is drastically overestimating how overvalued the market is right now.

[maizefolk]

Fortunately there is no equivalent of the efficient market hypothesis for the emergence of totalitarian regimes.

Since the market is efficient, it will tank long before you are personally aware of SHTF.   This will leave you with approximately bupkis to flee.  If index funds existed in 1933, you would have had a decent 10 years of negative returns to fund your escape.

Kristallnacht happened in 1938. The german stock market didn't start to decline until 1944.

I was hardly there, but according to Wikipedia, we are talking hyperinflation, then 30% unemployment rate, followed by closure of exchanges in 1935.  If that happened in the US, you would not have much left.

I'll just leave this chart here.

[dude]

Why continually worry about outlying statistical situations that can neither be accurately predicted or mitigated against with a passive investment allocation?  It's better to invest in adaptability skills in those potential situations.  For the 4% rule to fail there has to be both a precipitous drop in asset value soon after draw-down begins AND long-term very low real returns. Even if this happens, someone with only half of FI assets remaining is better suited to take advantage of a universally bad situation, if they adapt to it.

Vanguard says that P/E is a reliable indicator of future returns.  No formula is going to give you the day or week when the market corrects or tumbles, but it appears there are warning signs and strategies that can help. 

The 4% rule failed after the great depression, and when enough history plays out, it might also fail during some of the severe downturns of the 2000's.  Guess what?  CAPE market valuations are back up in lime with the great depression started (although I would argue it's not quite as bad now as it was then).  Don't give yourself a false sense of security that the idea from a single research paper puts the question of the safety of a 4% withdrawal rate to rest, at least with these high market valuations.

The bolded statement is not true, and Sol and others have already debunked similar statements that you made in the 4% rule thread.  Go run cFiresim at either a 60/40 or 75/25 stock/bond allocation, and you will see that the only failures occurred for start years in the mid and late 1960's, due to the 1970s stagflation era.  The Great Depression did NOT cause a failure of the 4% rule.  And if you run a 17-year sim that captures the Y2k retiree, he/she is doing demonstrably better than any of the 1960s/70s failure trajectories were doing 17 years in.  I'm pretty sure this has been pointed out to you before also.

Please stop spreading misinformation.

John Greaney has charted the progress of the Y2K retiree (scroll to bottom of link below). It's not all wine and roses, but all things considered -- particularly when you take into account his numbers account for 4% initial withdrawals with inflation-adjusted withdrawals every year since -- they are not doing too bad.  Consider how easy it would have been for them to adjust spending down a fraction, or pick up some easy part-time work, and the effect of Social Security, which most are surely collecting and probably have been collecting for some time. Their actual portfolios are probably far better off.

http://www.retireearlyhomepage.com/reallife17.html

[runewell]

Runewell, you do realize you're not being very consistent here, right? You've been dismissing the trinity study on these threads for the last few days as a twenty year old out of date study, people have pointed out the wide range of updated research and simulation tools, and now you've turned around and are holding up the original 1995 paper as the gold standard.

I noted that the Trinity study was 20 years old, and that it should be supplemented with newer ideas, but I don't believe i dismissed it.  My point was that with markets at high valuations, using any method with a historical lookback of investment returns would be too optimistic since research shows that higher valuations lead to lower than average returns going forward.

One example

Again, we've already discussed how the CAPE ratio is misleading, in fact you're the one who linked to the study showing that it is currently overestimating how overpriced the market is, perhaps by as much as 50%.

Yes I did, but the conclusion still shows that markets are quite expensive.  And Vanguard agrees that high P/Es lead to lower returns going forward.

I'm assuming you clicked the "adjust for inflation" box in cFireSim, so that $629K is in 2000 dollars. So you'll want to either keep spending constant at $40k/year which is probably the simplest approach, or also adjust $629k in remaining net worth up to the $912k it is in 2017 dollars.

OK I made a mistake there.  I looked at the column (new at this) and want this to be right and not exaggerate it.
882,830 is the portfolio value and i'll apply 1.5% inflation to the last spending number to get 56,983.
The failure rate is 5.2% that's much better.  I still claim that the true failure rate is higher than that with an expensive market, but as others have said predictions are tricky.  Thanks for noting my mistake, I sincerely don't want to spread false information.

"All models are wrong.  But some are useful."

[runewell]

Since the market is efficient

Robert Shiller in one of his lectures called the Efficient Market Hypothesis a "half-truth".

[maizefolk]

Thank you for acknowledging and correcting the error in your math for the 2000 retiree.

I sincerely don't want to spread false information.

The please please stop talking about the CAPE being at 30 and how that predicts terrible stock market returns, which you and I and basically everyone else agrees that comparing CAPE values today to those from before the 1990s is actively misleading. (Note: If you want to point out that the markets are overvalued, go for it! Just don't claim the CAPE value is an accurate representation of how overvalued they are.)

Talking about the CAPE being equal to what it was right before the great depression when it is clear that 1) those values are calculated based on underlying datasets that are not comparable, and 2) the effect of the difference in the way earnings are calculated is significant, is disingenuous and will discourage people from taking the time to discuss ideas with you.

Anyway, it's a free country, but that's my advice.

[runewell]

please please stop talking about the CAPE being at 30 and how that predicts terrible stock market returns, which you and I and basically everyone else agrees that comparing CAPE values today to those from before the 1990s is actively misleading. (Note: If you want to point out that the markets are overvalued, go for it! Just don't claim the CAPE value is an accurate representation of how overvalued they are.)

Anyway, it's a free country, but that's my advice.

I'll try to go easy on the CAPE sauce :)

How about P/B values?

[maizefolk]

The big gotcha with price to book ratios is going to be the rise of intangible assets, both "goodwill" from buying other companies for more than their book value and things like patents and brands and what have you. Something like 85% of the book value of the S&P 500 is now made up of intangible assets, while in 1975, only 17% of the book value of the S&P 500 was intangible. Since it is hard to assign accurate prices to intangible assets, I'd argue the accuracy of book values is likely decreasing over time.

But I don't know of any reported source of systematic bias in price/book value ratios like there is with CAPE.

[GenXbiker]

I hope you all have financially secure early retirements.  I'm hoping to quit around age 60 with almost $2M.  I would retire earlier than that but I am reluctant to take any chances with health insurance and costs which is the biggest unknown apart from stock returns.

That's true about the health insurance - I've got a couple years, maybe longer, to see how things play out.  Based on some increases over my current expenses that I expect at FIRE, I should be able to get by with a 2% WR over 35 years to cover increased expenses and still have a monthly cushion, and that's even without factoring in SS 10 years into retirement.  cFiresim gives me 100% with 5% WR equivalent when I factor in SS @62.  4% would be fine with me, but at least I know I have room to be flexible without necessarily having to return to work or cut to bare bones, although some part time or side gigs are not out of the question if I feel like doing that.

[Scandium]

Fortunately there is no equivalent of the efficient market hypothesis for the emergence of totalitarian regimes.

Since the market is efficient, it will tank long before you are personally aware of SHTF.   This will leave you with approximately bupkis to flee.  If index funds existed in 1933, you would have had a decent 10 years of negative returns to fund your escape.

Kristallnacht happened in 1938. The german stock market didn't start to decline until 1944.

I was hardly there, but according to Wikipedia, we are talking hyperinflation, then 30% unemployment rate, followed by closure of exchanges in 1935.  If that happened in the US, you would not have much left.

I'll just leave this chart here.

Noted. As they say: "Always sell right after the landwar in Asia"

[VoteCthulu]

Don't get me wrong, I don't hide my money in a mattress, think collapse of the US is likely, etc.  But I also don't kid myself that the number cFiresim spits out is the end-all and be-all of success rate, for broad definitions of success.  Once you get to the point that non-investment risks are on par with investment risks, it doesn't make sense to increase your savings.

Out of curiosity, what do you think the probability of non-investment risk over the next say 50 years is?

Some risks are local and have standard methods of assessing, like earthquakes, floods, etc. but I don't know how to price in global financial collapse, aliens, or Republicans winning the next 10 elections.

[maizefolk]

Noted. As they say: "Always sell right after the landwar in Asia"

And into my investment policy statement it goes. ;-)

[DavidAnnArbor]

Funding the State department to provide foreign aid to help with droughts, famine, provide new farming methods, using diplomacy to solve ethnic conflicts, and overall helping people would go a long way toward lessening chances of war.

[runewell]

but I don't know how to price in global financial collapse, aliens, or Republicans winning the next 10 elections.

Financial collapse cannot hold office, so I would narrow it down to aliens or Republicans. 

[dragoncar]

I'll just leave this chart here.

Thanks, that's interesting and I hadn't found a chart as good as it.  However, is it inflation adjusted?  Either way, I'm more concerned with what was going before the "hitler comes to power" point.  I'm suspecting that even at the "worries the hitler has gone too far" portion, you are still in negative territory.

Since the market is efficient

Robert Shiller in one of his lectures called the Efficient Market Hypothesis a "half-truth".

Yeah, that was tongue in cheek because someone brought up the humerous idea of EMH for dictatorships.

Don't get me wrong, I don't hide my money in a mattress, think collapse of the US is likely, etc.  But I also don't kid myself that the number cFiresim spits out is the end-all and be-all of success rate, for broad definitions of success.  Once you get to the point that non-investment risks are on par with investment risks, it doesn't make sense to increase your savings.

Out of curiosity, what do you think the probability of non-investment risk over the next say 50 years is?

Some risks are local and have standard methods of assessing, like earthquakes, floods, etc. but I don't know how to price in global financial collapse, aliens, or Republicans winning the next 10 elections.

Of course we can't predict the future, so anything we come up with is just a guess.  The Bernstein link I posted mentions a 20% risk, but that is also very back of envelope.  I'm fine using this as a rule of thumb, however, since it jives nicely with the 80/20 rule, and the idea of diminishing benefits after you reach an 80% cFireSim success rate. 

The same logic also applies to minimizing these non-investment risks.  There's only so much you can do to mitigate (move to another country, climate, natural disaster area, etc.) but for all you know you will just worry too much and develop hypertensive heart disease, then move to Canada which is unexpectedly taken over by a totalitarian regime.  Sometimes it's better to just enjoy your life and keep a dispassionate eye out for low-hanging fruit (e.g. typical disaster preparedness)

[Tyson]

It's funny, people that talk about the future being different than the past - it's ALWAYS negative.  The past is always held up as a better time than the future.  It boggles my mind.  It's NEVER, "Oh, here's some things that will make things even better in the future than in the past".  Even though it's just as likely that things will be permanently better than it will be that they will be permanently worse.

And not only that, if there's a recent trend toward things being better (or even a medium/long term trend), it's never "Oh, things are trending better, yay us".  Nope, it's always "Oh no, things are overpriced!"  "Oh no, things are gonna crash!"  "Oh no high CAPE will kill my retirement". 

It's just amazing to watch.

[Classical_Liberal]

@Pizzasteve
A well articulated couple of posts, they are in line with my thoughts regarding risk. I'm glad you stuck around for a couple more, even if I don't always agree with you.

@Dragoncar
I certainly sympathize with your thoughts about being able to react before crisis. My worries with socio-political events are more alone the lines of overreacting.  IOW, I'm more worried I'll see red flags and bail when there is no crisis imminent.  As a result, I make several different very small, but potentially lucrative bets in the event of "problems".  Even if those lottery numbers hit, I'd probably be out most of my stash.  Having some is better than none; emotionally it prevents me from overreacting.

I'm fine using this as a rule of thumb, however, since it jives nicely with the 80/20 rule, and the idea of diminishing benefits after you reach an 80% cFireSim success rate. 

This is really the key.  Massive concern about small percentages of failures in back-tested portfolios is wasted energy.  There is easily a greater risk of death or other unknowns.  It's much better to follow Pizzasteve's advice; remain smart and flexible, don't waste time saving to 100% back-test success and then resting on laurels.

[Radagast]

Your data runs are very helpful. Would you mind re-running the glide path with different initial allocations? ie - initial 40, 50, 60, 70, 80, 90% stocks. The withdrawal rates at 4, 4.5 and 5% are probably most interesting.

Here you go.

4.0% WR

ramp up over (years) \ initial stock allocation (%)	40	50	60	70	80	90	100
3	95.7	96.8	97.9	97.9	96.8	95.7	94.7
5	93.6	95.7	97.9	97.9	96.8	95.7	94.7
7	94.7	96.8	96.8	97.9	96.8	95.7	94.7
10	93.6	96.8	96.8	97.9	96.8	95.7	94.7
15	94.7	94.7	97.9	97.9	96.8	95.7	94.7
inf	62.8	79.8	86.2	90.4	92.6	95.7	94.7

Sweet, great post, thanks. My intuition tells my that starting at 40% bonds, following a 2% glide path per year to 20%, and then a 1% per year to 10% is about right. I don't think it is smart to get below 10% bonds because the market can always crash 90%,, which could put you in a tight spot even after 20 years. Plus it matches my narrative the optimal range is 10%-40% bonds, and allows me to check "all of the above"!

[Radagast]

Besides, if a country does nationalize (like Germany or China), gold won't help - it'll be taken from you like everything else.  Whether you stay or go, physical assets can be seized just as easily as any other asset.

Assuming you wait to leave until after capital controls are put into place. Otherwise just transfer your money out.

If you cannot do that, how likely are you to be able to get your physical gold out of the country without having it seized? An average sized stash (say $600k) would be about 35 lbs of gold at current prices, which would be hard to hide in your carry on luggage and in a situation with capital controls I imagine you wouldn't just be able to declare it and take it with you no questions asked.

Technically, the PP calls for most of your gold to be stored out of the country and far overseas, with just enough in a safe deposit box to make bribes.

Since it hasn't been linked in this thread yet, William Bernstein's short treatment of how investment risk differs from other risks: http://www.efficientfrontier.com/ef/901/hell3.htm

Don't get me wrong, I don't hide my money in a mattress, think collapse of the US is likely, etc.  But I also don't kid myself that the number cFiresim spits out is the end-all and be-all of success rate, for broad definitions of success.  Once you get to the point that non-investment risks are on par with investment risks, it doesn't make sense to increase your savings.

Fortunately there is no equivalent of the efficient market hypothesis for the emergence of totalitarian regimes.

Since the market is efficient, it will tank long before you are personally aware of SHTF.   This will leave you with approximately bupkis to flee.  If index funds existed in 1933, you would have had a decent 10 years of negative returns to fund your escape.

Kristallnacht happened in 1938. The german stock market didn't start to decline until 1944.

I was hardly there, but according to Wikipedia, we are talking hyperinflation, then 30% unemployment rate, followed by closure of exchanges in 1935.  If that happened in the US, you would not have much left.

You should read more Bernstein. In "Deep Risk" he analyzes stocks during the German hyper inflation and finds that after a lag of a year or two they come out very well.

[Radagast]

I promised to stop posting, but this topic is very interesting to me (complex systems, and the impact of disruptive forces on economic valuations of equities). So perhaps a rare appearance to comment.

That's stupid, why would you stop posting? Because Boarder42 had a hissy fit? You've been one of the better and more knowledgeable posters here, with a slightly different perspective. Keep posting.

[gerardc]

Sweet, great post, thanks. My intuition tells my that starting at 40% bonds, following a 2% glide path per year to 20%, and then a 1% per year to 10% is about right. I don't think it is smart to get below 10% bonds because the market can always crash 90%,, which could put you in a tight spot even after 20 years. Plus it matches my narrative the optimal range is 10%-40% bonds, and allows me to check "all of the above"!

You're right, I just backtested your strategy and I get 97.9% success (only 2 failed starting years), which is on par with the best strategy. Pretty much anything that starts with 60-70% stock and ramps up to 90-100% in 3-10 years gets around this success rate. Looking at higher WRs, similar strategies work well too, but it becomes more important to ramp up to high stock quickly (3-5 years) and/or start at higher stock (80%) and not linger too long in bonds.

[Radagast]

Sweet, great post, thanks. My intuition tells my that starting at 40% bonds, following a 2% glide path per year to 20%, and then a 1% per year to 10% is about right. I don't think it is smart to get below 10% bonds because the market can always crash 90%,, which could put you in a tight spot even after 20 years. Plus it matches my narrative the optimal range is 10%-40% bonds, and allows me to check "all of the above"!

You're right, I just backtested your strategy and I get 97.9% success (only 2 failed starting years), which is on par with the best strategy. Pretty much anything that starts with 60-70% stock and ramps up to 90-100% in 3-10 years gets around this success rate. Looking at higher WRs, similar strategies work well too, but it becomes more important to ramp up to high stock quickly (3-5 years) and/or start at higher stock (80%) and not linger too long in bonds.

Hmmm so similar results to similar options but more complicated :).

[Monkey Uncle]

All you have to do is look at Table 3 in the original Trinity Study and you'll see that for a 30-yr time frame with an adjustment for inflation between 1926-1995 the success rate was only 95%-98% for a portfolio that had at least 1/2 stocks, 71% success for 25% stocks, and only 20% success for 100% bonds.  Not sure why my statement is true!  Let's see what we think about you not worrying about the year-2000 retiree...

cFiresim is only one simulation tool, and it doesn't even agree with the results of the Trinity study.  As long as people are going to quote the 4% rule I'm going to stick with one of the sources of that very rule.

Runewell, you do realize you're not being very consistent here, right? You've been dismissing the trinity study on these threads for the last few days as a twenty year old out of date study, people have pointed out the wide range of updated research and simulation tools, and now you've turned around and are holding up the original 1995 paper as the gold standard.

One example

But let's use it to simulate a 30-yr outcome starting at 2000.  We don't need to stop at 17 years of simulation, we can do it in two stages to try and add on the other 13 years to get a proper 30-yr time frame, shall we?

I used the cFiresim default calculations with a 2017 retirement date and a 2034 retirement end year.  Someone starting with $1M in 2000 would only have about $629K of their portfolio left in a US market - with a lofty CAPE ratio of 30 - we can't even factor in how expensive the market is, but that doesn't sound very safe at all. 

Again, we've already discussed how the CAPE ratio is misleading, in fact you're the one who linked to the study showing that it is currently overestimating how overpriced the market is, perhaps by as much as 50%.

Put that $629K back in the simulator and run it for another 13 years.  Adjusting expenses for inflation, $40,000 in 2000 is now $57,545 in 2017, so that's our starting expense number, and the simulations years are 2017-2030.  Now let's count how many times the portfolio fails over the 134 historical simulations.  The answer is 48!  That's the best way I can think to simulate a 30-yr horizon starting with 2000, and I get a 36% failure rate with this method. 

I'm assuming you clicked the "adjust for inflation" box in cFireSim, so that $629K is in 2000 dollars. So you'll want to either keep spending constant at $40k/year which is probably the simplest approach, or also adjust $629k in remaining net worth up to the $912k it is in 2017 dollars.

If we assume that market returns will not be as good for the next 30 years (CAPE research shows this to be very likely) then you would probably get a significantly higher success rate.

Please stop spreading misinformation.

Again, you yourself pointed out that the CAPE is drastically overestimating how overvalued the market is right now.

Thanks for the rebuttal, maizeman.

[Monkey Uncle]

Runewell, you do realize you're not being very consistent here, right? You've been dismissing the trinity study on these threads for the last few days as a twenty year old out of date study, people have pointed out the wide range of updated research and simulation tools, and now you've turned around and are holding up the original 1995 paper as the gold standard.

I noted that the Trinity study was 20 years old, and that it should be supplemented with newer ideas, but I don't believe i dismissed it.  My point was that with markets at high valuations, using any method with a historical lookback of investment returns would be too optimistic since research shows that higher valuations lead to lower than average returns going forward.

One example

Again, we've already discussed how the CAPE ratio is misleading, in fact you're the one who linked to the study showing that it is currently overestimating how overpriced the market is, perhaps by as much as 50%.

Yes I did, but the conclusion still shows that markets are quite expensive.  And Vanguard agrees that high P/Es lead to lower returns going forward.

I'm assuming you clicked the "adjust for inflation" box in cFireSim, so that $629K is in 2000 dollars. So you'll want to either keep spending constant at $40k/year which is probably the simplest approach, or also adjust $629k in remaining net worth up to the $912k it is in 2017 dollars.

OK I made a mistake there.  I looked at the column (new at this) and want this to be right and not exaggerate it.
882,830 is the portfolio value and i'll apply 1.5% inflation to the last spending number to get 56,983.
The failure rate is 5.2% that's much better.  I still claim that the true failure rate is higher than that with an expensive market, but as others have said predictions are tricky.  Thanks for noting my mistake, I sincerely don't want to spread false information.

"All models are wrong.  But some are useful."

Not quite sure what you're doing with the inflation adjustment there.  It would be simpler to do as maizeman said and just plug in the ending value from the 17 year run (629k) for the stash and keep the spending at 40k.  cFiresim handles the inflation adjustment for you.  When I do that with either a 60/40 or 75/25 asset breakdown, I get a failure rate of less than 4%.  All the failure runs in the modern era occurred during the stagflation period, which followed a period of garden-variety high CAPE (low 20s) in the late 60s.  I say "modern era" because under the 60/40 scenario, start year 1912 also failed.  Notably, the 13 year period beginning in 2000 did not fail (although it came close with the 75/25 breakdown).  So the 4% rule survived a 30 year period that experienced the highest CAPE in recorded history twice.

[Tyson]

Runewell, you do realize you're not being very consistent here, right? You've been dismissing the trinity study on these threads for the last few days as a twenty year old out of date study, people have pointed out the wide range of updated research and simulation tools, and now you've turned around and are holding up the original 1995 paper as the gold standard.

I noted that the Trinity study was 20 years old, and that it should be supplemented with newer ideas, but I don't believe i dismissed it.  My point was that with markets at high valuations, using any method with a historical lookback of investment returns would be too optimistic since research shows that higher valuations lead to lower than average returns going forward.

One example

Again, we've already discussed how the CAPE ratio is misleading, in fact you're the one who linked to the study showing that it is currently overestimating how overpriced the market is, perhaps by as much as 50%.

Yes I did, but the conclusion still shows that markets are quite expensive.  And Vanguard agrees that high P/Es lead to lower returns going forward.

I'm assuming you clicked the "adjust for inflation" box in cFireSim, so that $629K is in 2000 dollars. So you'll want to either keep spending constant at $40k/year which is probably the simplest approach, or also adjust $629k in remaining net worth up to the $912k it is in 2017 dollars.

OK I made a mistake there.  I looked at the column (new at this) and want this to be right and not exaggerate it.
882,830 is the portfolio value and i'll apply 1.5% inflation to the last spending number to get 56,983.
The failure rate is 5.2% that's much better.  I still claim that the true failure rate is higher than that with an expensive market, but as others have said predictions are tricky.  Thanks for noting my mistake, I sincerely don't want to spread false information.

"All models are wrong.  But some are useful."

Not quite sure what you're doing with the inflation adjustment there.  It would be simpler to do as maizeman said and just plug in the ending value from the 17 year run (629k) for the stash and keep the spending at 40k.  cFiresim handles the inflation adjustment for you.  When I do that with either a 60/40 or 75/25 asset breakdown, I get a failure rate of less than 4%.  All the failure runs in the modern era occurred during the stagflation period, which followed a period of garden-variety high CAPE (low 20s) in the late 60s.  I say "modern era" because under the 60/40 scenario, start year 1912 also failed.  Notably, the 13 year period beginning in 2000 did not fail (although it came close with the 75/25 breakdown).  So the 4% rule survived a 30 year period that experienced the highest CAPE in recorded history twice.

Darnit, now that you've shown that people shouldn't worry about CAPE, they'll have to go off and find something else to worry about!

Haha, I'm half joking, but it's true.  Some people are just worriers.  You allay one of their fears and they come back with another.  Watch and see.

[EscapeVelocity2020]

Post to follow - maybe runewell retorts?

[ender]

I think the fundamental thing to keep in mind with all the "indicators" is that they are mathematical ideas, intended to reflect and model things which are very non-mathematical.

Things like war are quite difficult to "model" into that approach.

[runewell]

Darnit, now that you've shown that people shouldn't worry about CAPE

When did this happen?  I think CAPE is still a worry.

[Tyson]

Darnit, now that you've shown that people shouldn't worry about CAPE

When did this happen?  I think CAPE is still a worry.

lol, of course you do.  And if it wasn't CAPE it would be sonething else.  As I said before, some people are just worriers.  Hell, most people are.  Its a big reason people "get out" of stocks, or even worse, try to time the market by moving into and out of cash. 

[GenXbiker]

I don't worry about it, but I'm just prepared for the next 10 to 15 years to provide much lower "real" gains than what we've been getting since 2009, and less than the historical average.  So I've built my stache up well above 25x because I can always use the extra dough, and it gives me more flexibility to cut back in a bear market.

[Classical_Liberal]

I don't worry about it, but I'm just prepared for the next 10 to 15 years to provide much lower "real" gains than what we've been getting since 2009, and less than the historical average.

I was "preparing" myself for this since 2013's gangbuster.  I've had double digit returns (counting 2017 YTD) three of the last four years since then.  Should I be more or less worried now?

[GenXbiker]

I don't worry about it, but I'm just prepared for the next 10 to 15 years to provide much lower "real" gains than what we've been getting since 2009, and less than the historical average.

I was "preparing" myself for this since 2013's gangbuster.  I've had double digit returns (counting 2017 YTD) three of the last four years since then.  Should I be more or less worried now?

I think this depend on your details.  If you've been preparing that long, you shouldn't be worried now, just prepared.  Speaking of 2013, that's when I reached the magic 25x.  I'm well above that now and will FIRE in 22 months as long as the market doesn't tank too badly.  I'm just not going to worry about it.

[farfromfire]

It astounds me that anyone could read through the hundreds of pages of careful mathematical analysis this forum has done about safe withdrawal rates, and then accuse us of blind faith because of a thread title summarizing our findings.

If the there is no reason to worry about the 4% withdrawal rule (as per the thread title), why is hundreds of pages of careful mathematical analysis going on then?

Congratulations, this wins dumbest post on MMM forums 2017! Your prize? You get to continue regurgitating the same tired arguments ad nauseam.

[clarkfan1979]

If I have 1 million in the stock market and I plan to spend "on average" $40,000/year, I'm not going to spend $45,000 if the stock market has a good year. I'm still going to spend $40,000.

If the stock market has a bad year, I'm ok spending $35,000-$38,000.

I don't see myself blindly spending 4% and paying zero attention to the market.

[GenXbiker]

If I have 1 million in the stock market and I plan to spend "on average" $40,000/year, I'm not going to spend $45,000 if the stock market has a good year. I'm still going to spend $40,000.

If the stock market has a bad year, I'm ok spending $35,000-$38,000.

I don't see myself blindly spending 4% and paying zero attention to the market.

It astounds me that anyone could read through the hundreds of pages of careful mathematical analysis this forum has done about safe withdrawal rates, and then accuse us of blind faith because of a thread title summarizing our findings.

If the there is no reason to worry about the 4% withdrawal rule (as per the thread title), why is hundreds of pages of careful mathematical analysis going on then?

Congratulations, this wins dumbest post on MMM forums 2017! Your prize? You get to continue regurgitating the same tired arguments ad nauseam.

That's been beat to death in these other threads:
http://forum.mrmoneymustache.com/investor-alley/stop-worrying-about-the-4-rule/
https://forum.mrmoneymustache.com/investor-alley/start-worrying-about-the-4-rule/