Desparately Trying To Understand SWR?

[ShokWaveRider]

OK, I am not stupid but please bear with me.

We have no children and do not need to leave any of our cash to anyone, or do we want to.

When running calculators such as Firecalc, I always get a cash balance after my 30 year withdrawals. It shows say a 95% probability of success but shows a large cash balance after the withdrawal period.

Assuming we want to have say $100,000 left after 30 years as a buffer, how do I calculate how much we can withdraw from a $1m portfolio of after tax money? Our average rate of return is currently 3%

Is there a simple reliable calculator I can review. We have no pensions and when we do get them they will be small and more than likely only cover beer money.

I think this would be handy for discovering how much is needed to retire modestly. For those who are trying to figure it out.

SWR (Pun) "S"hok"W"ave"R"ider

 

[cute fuzzy bunny]

Two problems with this approach:

#1: trying for a small "end" portfolio size...there are some periods of time when we've had historically long downturns. If you didnt have enough towards the end, you might run out of money before you run out of life. If you're consistently getting 100% results, look to the "maximum withdrawal at 95%" figure and re-run with that. Noting that there are now 5% historic periods where you ran out of money. Those are probably the depression and the 1965-1985 periods when things were pretty tough. It is entirely likely we'll see another one of those in the next 30-40 years. Do a "detail" run and scroll over to look at the end portfolio size for each period. In your 95% runs, you're going to see a LOT of endings with very low portfolios...and some with humongous ones. Think about how you'd be feeling at 90 with one of those low portfolio balances staring you in the face.

#2: you dont know when you're gonna die, unless you plan on forcing the decision. Medical advances are pretty good. Its completely possible many of us will live to 90-100 and feel pretty good until we're in our early to mid 80's.

Almost nobody dies these days from the top killers of 100 and 50 years ago. Hence the things wiping us out now may not be factors 50 years from now.

You can take a shot at the complete consumption portfolio, but I prefer a self sustaining portfolio where I dont directly access the principal, just the dividends and growth minus inflation.

 

[wabmester]

Simply think of FIREcalc as a historical record. It'll tell you how certain investments did in the past. That terminal value it gives you is just an average of the terminal values for each historic sequence. It's useless as a predictor.

If you plan to leave your portfolio in CDs paying 3%, then you need to make some assumptions about inflation. If you assume inflation will average 3% per year, then you can assume that it's safe to take out 1/30 of your nest egg each year for 30 years with adjustments for inflation.

If you want to keep a 10% cushion, then take out 3% instead of 3.33%.

 

[Zipper]

Shock, $100 000 in 30 years will be chump change. Look back to 1974 and what $10 000 would buy. Unless your name is Ken Lay or Gary Winnick, I wouldn't worry too much. Those 2 bums made sure they looked after #1! Maybe you should too.

 

[John Galt]

Hello Zipper. I am!

John Galt

 

[John Galt]

And another thing...........it doesn't matter to me if
100K is "chump change" in 30 years as I don't plan to be around. "Die broke" is a good concept, tough to actually
achieve though.

John Galt

 

[Bob_Smith]

Shock, $100 000 in 30 years will be chump change. Look back to 1974 and what $10 000 would buy.

If the next 30 years were to match the last 30 (1973-2003), in 30 years you would need around $415,000 to buy what $100,000 buys now.

 

[charlie]

Hey Shok,

Check out Vanguard's immediate annuity. If you
are serious about dying broke, a joint life annuity
with a COLA , available in increments of 1% up to 5%,
might suit you fine. Depending on your age, you
should be able to get more than 3%. Don't shoot
the messenger, these are not for folks who don't
want to lose control of their money.

You could invest $100k in I-bonds for emergencies,
and invest the rest in an annuity.

OTOH, if you invested in something like Vanguard's
Target Retirement Income fund, you could reap 3.37%
in dividends and interest until the cows come home.
You might, just might, find a charity worthy of you
benevolence in the future.

Cheers,

Charlie

 

[Jazz]

What am I missing?

Treasurey securities are currently at multi-decade lows and STILL yield >4% for a 10 year note (which is more than the benchmark SWR)......so if you have enough principal for a 4% SWR, why do you need any stocks at all? Why not start a ladder of Treasuries including TIPS using a (zero cost ) Treasury Direct account?

Wouldn't this strategy provide protection during inflationary times (even if the rate increases lag the actual cost of living)? The fallback to this strategy is absolute security of principal and the ability tol

 

[hocus]

If you have enough principal for a 4% SWR, why do you need any stocks at all?

The study published at RetireEarlyHomePage.com claims that a portfolio with 74 percent S&P stocks is "100 percent safe" for 30 years for investors with a 4 percent annual take-out. The author of the study (intercst) says that it is "irrational" for any aspiring early retiree to have less than a 74 percent stock allocation. The argument is two-pronged: (1) you cannot get a SWR of much higher than 4 percent with other investment classes; and (2) stocks offer greater long-term growth potential.

This conclusion does indeed generally follow from the REHP study's findings. The problem is that the study uses a methodology that is analytically invalid for purposes of determining SWRs. JWR1945's research shows that the SWR for an 80 percent stock portfolio in January 2000 was 1.6 percent. The SWR for TIPS when they were paying a 4.1 percent real return was 5.8 percent. There are many circumstances in which it is entirely rational to be invested in asset classes other than stocks.

What the historical data really shows is just what common sense would lead you to expect. There is a trade off between safety and growth. If you have lots of slack in your plan, you can afford to take on the risks associated with investing in a volatile asset class like stocks. If you want the highest SWR possible from a modest nest egg, you need to limit your allocation to stocks at times of high valuation.

SWR analysis is a powerful tool for informing investment stratagies. But it must be used in reasonable ways to deliver useful results. The REHP study assumes that changes in valuation levels have zero effect on SWRs (no downward adjustment is made in the SWR in times of high valuation). So it produces the nonsensical conclusion that stocks are always the best asset class from both a safety and growth perspective. The reality is that stocks are great at doing what stocks do, but that there are often portfolio allocations that get you to a safe retirement a lot sooner than the 74 percent stock allocation described in the REHP study as "optimal" at all times and for all investors.

 

[unclemick]

There is no sure thing as in a reliable calculator or SWR. The calculators are there to offer insight and possiblities - nobody controls the future. The raging debate over calculators/SWR is improve their capability to capture historical data and meaningful varibles to offer better insight.

In 1993 - what passed for data and a income take - out was 6-8% depending on whose retirement table you read. Memory says - I could have used 7% had I done a 72t.

Real money is what your portfolio gives you. Pros and cons of various calculators are a hoot if you wish to join the debate. Pay attention to the weaknesses and strength's of the methods being debated/discussed/ analyzed and use them 'hand grenade wise' to keep you in the ballpark. Again - what the market gives you and a little common sense go a long way.

Not to worry - if you live - history will be what it was. Party on. OR as they say in Nawln's -

BON TEMPS ROLLERE - Let the Good Tims Roll!

 

[hocus]

The raging debate over calculators/SWR is improve their capability to capture historical data and meaningful varibles to offer better insight.

That's a highly constructive comment. Thank you for saying that, UncleMick.

 

[Hyperborea]

What am I missing?

Treasurey securities are currently at multi-decade lows and STILL yield >4% for a 10 year note (which is more than the benchmark SWR)......so if you have enough principal for a 4% SWR, why do you need any stocks at all?  Why not start a ladder of Treasuries including TIPS using a (zero cost ) Treasury Direct account?

You're missing inflation. If you run FireCalc with such a portfolio you'll find that historically you'd have a fairly good chance of surviving the 30 years - somewhere in the mid-80% region depending on the specific numbers you use.

The problem is that at the end of the 30 years the mean portfolio value was about 80% of the original value. That's 80% of the original value without taking inflation into account. With even a mild 3% inflation over the years the median ending balance is 33% of the original amount. If it were 4% you're only going to have 25%.

A strategy like this might be ok if you're 70 but for any early retirees it's likely to leave you living in group homes and doing your own dentistry with a pair of pliers. In fact, that was pretty much the fate of the most (in)famous advocate of such a plan, Joe Dominguez, the author of "Your Money or Your Life".

Even a mild injection of equities increases both the survival rate and the median final value. For an early retiree in their 30's, 40's, or even 50's this is going to be vital to prevent poverty.

As for alternate SWR non-calculators that involve market timing I would suggest that you consider the math skills or lack thereof of the main proponent and for second and third criteria look at their lucidity and coherence.

 

[hocus]

As for alternate SWR non-calculators that involve market timing I would suggest that you consider the math skills or lack thereof of the main proponent and for second and third criteria look at their lucidity and coherence.

William Bernstein's math skills are just fine. He is also lucid and coherent. There's a good reason why Bernstein describes the conventional SWR methodology as "highly misleading" at times of high valuation. He looked at the historical data. Any reasonably informed analyst who looks at the historical data will come to the same conclusion.

 

[intercst]

As for alternate SWR non-calculators that involve market timing I would suggest that you consider the math skills or lack thereof of the main proponent and for second and third criteria look at their lucidity and coherence.

I second that note of caution.

I'm unaware of any responsible investigator who supports the ***** "stock-switching" SWR non-calculator.

As many successful early retirees have pointed out, using *****'s so-called "wonderful tool" would have left them completely out of stocks during the market run-up in 1995-2000 and again in 2003.

There's much more support in the research for maintaining a balanced asset allocation with both equities and fixed income securities.

intercst

 

[hocus]

using *****'s so-called "wonderful tool" would have left them completely out of stocks during the market run-up in 1995-2000 and again in 2003.

The Data-Based SWR Tool is a descriptive tool, not a prescriptive tool. It does not leave anyone in stocks or out of stocks. It reports accurately the SWR for stocks and for other asset classes so that investors can make informed asset-allocation decisions.

The core assumption of the REHP study, that changes in valuation levels have zero effect on long-term returns (and thus zero effect on SWRs), has been rejected by the followed experts: (1) William Bernstein; (2) Rob Arnott; (3) Andrew Smithers; (4) Peter Bernstein; and (5) Ed Easterling.

 

[intercst]

As for alternate SWR non-calculators that involve market timing I would suggest that you consider the math skills or lack thereof of the main proponent and for second and third criteria look at their lucidity and coherence.

William Bernstein's math skills are just fine. He is also lucid and coherent. There's a good reason why Bernstein describes the conventional SWR methodology as "highly misleading" at times of high valuation. He looked at the historical data. Any reasonably informed analyst who looks at the historical data will come to the same conclusion.

Here's what William Bernstein says about market timing -- the basis for *****'s "wonderful SWR tool"

http://www.efficientfrontier.com/ef/703/timer.htm

Note that you won't find anything in the article supporting *****'s theory that one should avoid stocks during what he describes as times of "high valuation". Bernstien might reduce his allocation to stocks by 5% or so during the bubble, but he in no way endorses the Hoco-Mania of a 100% fixed income portfolio. The final quote from the article is most telling.

Varying allocations—"timing," if you will—is similar to the consumption of alcohol. It can either enhance or degrade portfolio health; it all depends upon the circumstances and the quantity. When partaken in small, infrequent amounts from a concave vessel, its benefits are small but perceptible. When chugged indiscriminately, it is deadly.


intercst

 

[unclemick]

A good understnding of 'the horse you rode in on' is recommended before you dismount and try to tool off into the sunset in a Hummer.

Recent data runs over at the 'other forum' - albiet with a specific data set - are making the Norwegian widow smile - 100% dividend stocks, no fixed income - are looking competative.

Am I going to suddenly shift my portfolio mix (balanced index with DRIPS). No - but I may begin to shift to the dividend side more over time.

 

[hocus]

Here's what William Bernstein says about market timing

The article set forth at the link you provide is a good one, intercst. I hope that a good number of community members take the time to read it.

The description of my views set forth in your post are highly misleading. I have not put forward a "theory" that investors should avoid stocks at any particular times. I myself have not been invested in stocks since 1996, but I have said many times that that choice is not necessarily the right one for investors in other sorts of circumstances.

The "theory" that I have put forward is that, when one publishes a study of what the historical data says re SWRs, one incurs any obligation to report what the historical data says accurately. The historical data shows that changes in valuation levels affect long-term returns, and, thus, SWRs. Thus, I say that an analytically valid SWR methodology must show the effects of changes in valuation levels. The REHP study does not do that. My "theory" is that the number reported in the REHP study as the SWR for a high stock portfolio is wrong at the valuation levels that apply today.

My recommendation is that the study be corrected. There are two ways that could be done. An adjustment could be made for changes in valuation levels. If you added such an adjustment to the study's methodology, your study would report numbers in accord with those that have been reported by Bernstein and by JWR1945. The other way to correct the study would be to change the terminology, to make clear that your study was desgined to determine the historical surviving withdrawal rate (HSWR), and not the safe withdrawal rate (SWR).

 

[hocus]

Am I going to suddenly shift my portfolio mix (balanced index with DRIPS). No - but I may begin to shift to the dividend side more over time.

That makes sense, UncleMick.

The SWR Research Group board has been focused in the first year of its existence on data analysis. I hope that in time the focus will turn to questions of strategy. Ultimately, people need to know not only what the data says, but how to make use of what the data says.

My personal view is that the average investor should hold roughly a 50 percent allocation to stocks at times of moderate valuation, move down to about 30 percent at times of high valuation, and move up to about 70 percent at times of low valuation. Never have I endorsed any of the wild viewpoints ascribed to me by defenders of the conventional methodology who have given up on any hope of putting forward reasonsed arguments for their postions.

The purpose of SWR analysis is not to tell you what to do. The purpose is to inform you about what has worked in the past, so that you can make better determinations of what to do for yourself. The fact that the SWR for an 80 percent S&P portfolio is so low today (it is now at about 2.5 percent) does not translate into a recommendation to get entirely out of stocks. It is an important information bit that should be put to use by those seeking to construct successful plans for early retirement.

Your approach of learning what you can from each of a number of analytical tools, and permitting what you learn to gradually over time influence your decisions, strikes me as level-headed. In contrast, I find the unyielding dogmatism of those who ignore altogether the overwhelming evidence that changes in valuation affect long-term returns a big-time turn-off.

 

[intercst]

Here's what William Bernstein says about market timing

The article set forth at the link you provide is a good one, intercst. I hope that a good number of community members take the time to read it.

The description of my views set forth in your post are highly misleading. I have not put forward a "theory" that investors should avoid stocks at any particular times. I myself have not been invested in stocks since 1996, but I have said many times that that choice is not necessarily the right one for investors in other sorts of circumstances.

The "theory" that I have put forward is that, when one publishes a study of what the historical data says re SWRs, one incurs any obligation to report what the historical data says accurately. The historical data shows that changes in valuation levels affect long-term returns, and, thus, SWRs. Thus, I say that an analytically valid SWR methodology must show the effects of changes in valuation levels. The REHP study does not do that. My "theory" is that the number reported in the REHP study as the SWR for a high stock portfolio is wrong at the valuation levels that apply today.

My recommendation is that the study be corrected. There are two ways that could be done. An adjustment could be made for changes in valuation levels. If you added such an adjustment to the study's methodology, your study would report numbers in accord with those that have been reported by Bernstein and by JWR1945. The other way to correct the study would be to change the terminology, to make clear that your study was desgined to determine the historical surviving withdrawal rate (HSWR), and not the safe withdrawal rate (SWR).

Unfortunately ***** you continue to show a complete misunderstanding of the REHP study. The study accurately calculates portfolio withdrawals from 1871-2003 despite your claims of "missing data". It has been quoted in national newspapers and magazines from the Wall Street Journal to Newsweek.

I enjoy Hoco-Mania as much as the next person, but don't you find it the least bit strange that you're the only one who doesn't understand the study?

Do you ever wonder why your three-year long jihad against arithmetic has garnered so few converts?

intercst

 

[Workwayless]

***** says ]The purpose of SWR analysis is not to tell you what to do. The purpose is to inform you about what has worked in the past, so that you can make better determinations of what to do for yourself.



The REHP study is SWR analysis of what would have worked in the past. It's like the Fox News channel--it reports, but you need to decide what to do with the data. So, tell me again how the REHP study is misleading?

 

[hocus]

Do you ever wonder why your three-year long jihad against arithmetic has garnered so few converts?

Anyone who checks the Post Archives of the SWR Discussions will find that there were over 100 community members at the Motley Fool board who expressed an interest in reasoned debate of the flaws of the REHP study. There were scores of posters who said that it was the best on-topic discussion we had had in many months, prior to the time at which abusive posting practices were employed to shut it down. There have been a good number at this forum who have said similar things.

I put up the "What Bernstein Says" post on August 27, 2002. If you had any reasoned rebuttal to Bernstein's findings, you would have put it forward by now. The fact that you have not tells community members what they need to know about your willingness to shoot straight on this question.

Bernstein understands what the historical data says a whole bunch better than you do, intercst. When he says that your methodology is "highly misleading," he's saying that for a good reason. When he says that the SWR for a high-stock portfolio at the top of the bubble was 2 percent, he says that because that is what the historical data tells him it was. The SWR is whatever the historical data says it is.

 

[hocus]

So tell me again how  the REHP study is misleading?

William Bernstein devoted an entire chapter of his book "The Four Pillars of Investing" addressing these issues. My advice to any community member seeking to understand why the REHP study reports a number so far off the mark of the one obtained by looking in an informed way at the historical data is to read Chapter Two of that book.

 

[arrete]

William Bernstein devoted an entire chapter of his book "The Four Pillars of Investing" addressing these issues. My advice to any community member seeking to understand why the REHP study reports a number so far off the mark of the one obtained by looking in an informed way at the historical data is to read Chapter Two of that book.

Nice non-answer.

arrete