25 times income needs?

[Zipper]

If you are over 60 and can get CPP, and OAS at 65, or SS at 62 I think 20x's will do it.

 

[heyyou]

www.bylo.org/saferetr.html
www.members.tripod.com/DaveLeeMn/
www.bobsfiles.home.att.net/VariableWithdrawals.html
The Trinity and Bengen and Jaye Jarrett studies were my early exposure.  Since then, there have been refinements by studies of the retirement lengths, withdrawal sources, run frequency (Monte Carlo), and Guyton and other's withdrawal rules.  Scott Burns has reviewed Kotlikoff's software that optimizes withdrawals from various retirement savings accounts.  The Journal of Financial Planning published most of the best stuff.  I recently re-read Jaye Jarrett's early study.  He only ran one data set (history from 1929 to 1999) but he analyzed how a small difference in % at the start, changed the 30 year ending balance by millions.  That was illuminating to me, whereas the other studies just reported the numerical results.  Bottom line is that you need about 25 multiples (4% of portfolio with annual inflation adjustment) of your living expenses if you retire at the start of the worst historical investing periods, like 1928 (Great Depression)or 1968(annual inflation 10 to 14%).  For a variable retirement income, use 20 times (5% of portfolio) with no inflation adjustment.  Retire in 1949 and you can draw 8% annually if you can see 30 years into the future.

 

[brewer12345]

Heh, I wonder what a portfolio that contained TIPS and commodities would have done fr the 1966 retiree?

 

[Nords]

Heh, I wonder what a portfolio that contained TIPS and commodities would have done fr the 1966 retiree?

I really wish a Bernstein or some other researcher would get on that project. For bonus points it could consider the effects of mixing the portfolio with varying percentages of cash, too.

At a 2004 Schwab sales presentation customer-appreciation dinner I sat next to a guy who'd bought a huge slug of 9% 30-year bonds in 1974. He was not looking forward to his investment options but the Schwab commission-only sales financial advisors were drooling from their fangs...

 

[brewer12345]

The problem is that the DJ-AIG commodity futures index didn't exist back then, and neither did TIPS. I think raddr did a reasonable job of trying to cobble together data, and I have seen some other studies that use a created commodities index, but it is always a little loosey-goosey when researchers start creating the data they work on.

 

[Nords]

The problem is that the DJ-AIG commodity futures index didn't exist back then, and neither did TIPS.  I think raddr did a reasonable job of trying to cobble together data, and I have seen some other studies that use a created commodities index, but it is always a little loosey-goosey when researchers start creating the data they work on.

Yep, Raddr can take comfort in being the leading expert in that field.

Based on Raddr's numbers I think that a 1966 retiree would have been a lot better off.  Heck, considering that most studies of 1966-82 don't account for dividends (spent or reinvested), that retiree was a lot better off than most of us have been led to believe.

So, yeah, at the next fire sale we should add more commodities.  I guess we can say the same for beaver-cheeze futures, too...

Maybe it's because I'm updating our estate planning this month, but I have this nagging image of my bereaved spouse going through the "When Nords wakes up dead" folder and muttering to herself "Beaver cheeze? What the $%^& is that and why didn't he tell me about it?!?"

 

[Zipper]

Not to worry Nords.

The next guy will be able to figure it all out.

 

[Nords]

Not to worry Nords.
The next guy will be able to figure it all out.

Well, her motto is "If at first you don't succeed..."

 

[mathjak107]

Well, her motto is "If at first you don't succeed..."

isnt it "IF AT FIRST YOU DONT SUCCEED,TRY TRY A GUN"

 

[brewer12345]

isnt it "IF AT FIRST YOU DONT SUCCEED,TRY TRY A GUN"

Shades of JohnGalt again...

 

[NYCGuy]

It seems to me that many people here take the 4% SWR rate somewhat too seriously, although I may be reading too much into people's intentions from the tone of the present discussion. In my opinion, a long time horizon makes the results both more and less predictable. All of this long-term investment advice is admittedly based on an assumption that the future will essentially like the past, a controlling assumption that is usually forgottern as quickly as it is made. It's difficult even to talk about a future that is not mostly like the past. Now, if the future is indeed like the past, financial aspects like rates of reture or of inflation will vary within the bounds of the past. For short periods of time they may diverge meaningfully from historical measures, but as the period under consideration stretches further into the future they should fall tidily into the expected range. But all of this is only true under the original assumption that the future is like the past in the ways that matter. However, the longer a future we consider the more likely it is that the future will be impacted by substantial changes that no one has foreseen. I doubt if many Japanese financial planners in the mid-eighties factored in an upcoming fifteen-year period of deflation. Yet, the deflation is what has actually dominated their economic lives.

Routine statistical analysis of investment returns assumes that they are normally distributed. Only they are not. They are too fat-tailed. But standard deviations and Sharpe ratios, which are only defined on normal distributions, don't have meaning. Some folks think investment returns are better described by a power law distribution, like earthquakes. Hard to plan for earthquakes. As the mathematician, Benoit Mandelbrot, points out, while a daily change in the stock market of eight standard deviations should occur once in a trillion years, the change on Oct. 19, 1987 was twenty standard deviations out. Should never have happened. So, it could happen again without necessarily being followed by a recovery in three months.

It is entirely possible that the world is less stable now than in the past, even though much of it is more prosperous. And if, like me, you do not subscribe to the reassuring conviction of American exceptionalism, then we may very well be subject to profound changes in the future from wars, nuclear accidents, nuclear on-purposes, military coup, shift in wealth or power to another region, economic depression, bird flu pandemic, global warming, currency collapse, or some other singularity that is now completely unforseen. My own expectation is that our actual future will probably be dominated by one or more of these changes. And the actual future is the one I care about.

So, to me watching for and trying to assess novel risks is more to the point that extrapolating well-understood results from the past, although that is also important. And integrating assessments of those risks into decision-making is not going to be by formula.

 

[Nords]

It seems to me that many people here take the 4% SWR rate somewhat too seriously, although I may be reading too much into people's intentions from the tone of the present discussion.

Here we go again...

http://www.efficientfrontier.com/ef/998/hell.htm
http://www.efficientfrontier.com/ef/101/hell101.htm
http://www.efficientfrontier.com/ef/901/hell3.htm
http://www.efficientfrontier.com/ef/103/hell4.htm
http://www.efficientfrontier.com/ef/403/hell5.htm .

 

[unclemick]

Hmmm

You know - what really ticks me off about these guys with their computers/programs/spreadsheets/etc etc - they seem to totally miss the dividends part.

In 1948, before all this MPT whiz bang junk, the only thing Mrs Wright(aka Andersen) needed on some days when I got off the Kindergarten Bus in the the old PacNW was a rain slicker or umbrella so the dividend checks wouldn't get wet.

That great investment genius, Yogi Berra, could have been referring to div.'s when he discovered - 'cash is the same as real money.'

To paraphrase Bernstein's buddy - Angus Madison - if your spot in the world isn't paying the 'right' dividend's - buy somewhere else.

heh heh heh heh heh heh heh heh - computers!, we don't need no stinking computers. Actually I'm a little suspicious about just plain electricity - long term that is.

 

[Gone4Good]

So, to me watching for and trying to assess novel risks is more to the point that extrapolating well-understood results from the past, although that is also important.  And integrating assessments of those risks into decision-making is not going to be by formula. 

So your point is to plan for the things that, by definition, can't be planned for?  

I agree with you that the future may not be like the past and that the "SWR" concept is fatally flawed in its dependence on fairly short and narrowly defined historic data.  But what is the alternative?  The relevant question we're all trying to answer is "should I quit my job or continue working?"  The threat of global pandemic, nuclear war, etc. doesn't really have a bearing on that decision.  In fact, some might argue that they'd rather have a few years of retirement under their belt before Armageddon strikes than die while fruitlessly trying to accumulate enough resources to withstand the worst possibilities imaginable.  

 

[Texas Proud]

It is entirely possible that the world is less stable now than in the past, even though much of it is more prosperous. And if, like me, you do not subscribe to the reassuring conviction of American exceptionalism, then we may very well be subject to profound changes in the future from wars, nuclear accidents, nuclear on-purposes, military coup, shift in wealth or power to another region, economic depression, bird flu pandemic, global warming, currency collapse, or some other singularity that is now completely unforseen. My own expectation is that our actual future will probably be dominated by one or more of these changes. And the actual future is the one I care about.

So, to me watching for and trying to assess novel risks is more to the point that extrapolating well-understood results from the past, although that is also important. And integrating assessments of those risks into decision-making is not going to be by formula.

Hmmm... there are very few risks that will affect the economy in a major way... take a look at 9-11.. it hurt for awhile, but then it went by... take a look at almost any war... it again goes away.. the last time a major health issue (early 1900 when a lot of people died of the flu.. and then add on top of that WWI).. the economy kept on a going...

People put to much fear into a catastrophy.... looking at histry there has been nothing that has permanently kept mankind from growing...

Lets say someone goes to the middle of Manhatten and blows up a suitcase nuke... well, it will be tragic, but the rest of the country will keep growing corn and wheat... keep producing and consuming.. and in a few years the economy will be back to its long term trend line... yes, we will have major phone taps and other govmt intrusions so it hopefully will not happen again.. but it will not stop the growth..

So, 4% takes into account all you worries...

 

[sgeeeee]

. . . All of this long-term investment advice is admittedly based on an assumption that the future will essentially like the past, a controlling assumption that is usually forgottern as quickly as it is made.  . . . 

Not really. The assumption is that investing broadly in stocks and bonds in the future will not produce worse long-term results than the worst case that has happened over the past 130 years.

This is significantly different than assuming "the future will essentially be like the past". We are not counting on high or average historical returns -- only worst case. We are not counting on next year being typical of any year in the past or even next decade being typical of any decade in the past. We are hoping that the fundamental market pressures of a capitalist economy that produce stock returns, bond returns, and inflation are still similar to those of the past (ie. fear and greed). There are complex relationships between financial returns and inflation -- and between financial returns of the recent past and those of today. All of these complex correlations are implicitly included in the historical simulator. No other simulation approach offers this advantage.

Of course there is still no guarantee that the next 30 years won't be even worse than the 30 years immediately following the Great Depression or the decades following the mid-60's. But how do you plan for something outside the range of experience? Some people tout Monte Carlo simulation as an answer, but Monte Carlo programs depend on historical data to establish assumed returns distributions and fail completely to capture many of the correlations.

Historical simulation is not a predictor of the future. It is not a perfect guide into our retirement. It's just the best tool we have at the moment.

 

[astromeria]

One could hide in a fallout shelter with a pile of gold, MREs, and a chemical potty and wait till WWWIII is over I prefer to invest in a moderately conservative portfolio, withdraw 3.8%, and take my chances. I bet I have gold miners, chemical co's, potty manufacturers, and even MRE suppliers in my mutual funds.

 

[greg]

So, to me watching for and trying to assess novel risks is more to the point than extrapolating well-understood results from the past, although that is also important. And integrating assessments of those risks into decision-making is not going to be by formula.

NYCGuy:

So, over the next 5-10 years, what do you see as the risks?

I've sort of mellowed over the past few months. But . . . on the international front I still see an escalating war or boom or boom-boom-boom event as an extreme event. I currently give that a low percentage possibility. Even if it happens say after 2008, for speculation purposes, I suspect we will just bunker up as a country and protect our borders more--maybe fine tune the defense system in event of further problems.

On the domestic front I still see a looming huge debt problem. I'm anticipating a muddle thru though at this point--no huge depression (low percentage probability), no huge inflation (low percentage probability). I still see a huge amount of fixing needed starting in the next few years. I'm hoping for/anticipating gradual and incremental change for the better. (But it will still probably get worse before getting better)

 

[brewer12345]

NYCGuy: 

So, over the next 5-10 years, what do you see as the risks?

I've sort of mellowed over the past few months.  But . . . on the international front I still see an escalating war or boom or boom-boom-boom event as an extreme event.  I currently give that a low percentage possibility.  Even if it happens say after 2008, for speculation purposes, I suspect we will just bunker up as a country and protect our borders more--maybe fine tune the defense system in event of further problems. 

On the domestic front I still see a looming huge debt problem.  I'm anticipating a muddle thru though at this point--no huge depression (low percentage probability), no huge inflation (low percentage probability).  I still see a huge amount of fixing needed starting in the next few years.  I'm hoping for/anticipating gradual and incremental change for the better.  (But it will still probably get worse before getting better)   

Um, housing bubble fixed in your mind?

 

[mb]

Heh, I wonder what a portfolio that contained TIPS and commodities would have done fr the 1966 retiree?

Much better than it would have done for the 1982 retiree I suspect.

 

[haha]

Well NY City Guy, I guess the middle of our lineup knocked you right out of the box.

Ha

 

[greg]

Um, housing bubble fixed in your mind?

Hardly, it's the $79 Trillion in US debt as reported in the Financial Times a while back.

 

[BunsGettingFirm]

So your point is to plan for the things that, by definition, can't be planned for?

I agree with you that the future may not be like the past and that the "SWR" concept is fatally flawed in its dependence on fairly short and narrowly defined historic data. But what is the alternative? The relevant question we're all trying to answer is "should I quit my job or continue working?" The threat of global pandemic, nuclear war, etc. doesn't really have a bearing on that decision. In fact, some might argue that they'd rather have a few years of retirement under their belt before Armageddon strikes than die while fruitlessly trying to accumulate enough resources to withstand the worst possibilities imaginable.

Oh, oh, one I banker going after another. This should serve as a (rather mild) preview of my future working condition if I'm good enough to get in.

Don't forget to have lots and lots of good but pointless sex for those few years.

Regarding Apoc's comment, I don't see how we can just shore up our borders and wait out any wars especially if the wars are in the Middle East. On what energy source are we going to power our economy if the flow of oil is disrupted?