Do you agree with Bengen's "Layer Cake" withdrawal rate

[MikeK]

I was intrigued with Bengen's "Layer Cake" model in August's Issue of Journal of Financial Planning (it was mentioned in this week's WSJ).

http://www.fpanet.org/journal/articl...p0806-art6.cfm

He proposes that the withdrawal rate should consider someone's "special situations". This intrigued me because it's too often assumed that the withdrawal rate should be the same for everyone. I believe that everyone has a different situation -- although many factors are consistent. Bengen outlines the key "special situations" and how they impact your withdrawal rate plan.

Overall I think Bergen has added great value to the withdrawal rate conversation. My biggest concern is that some of these "special situations" are just taking on more risk for a higher return. This makes mathematical sense but most people can't stick to the plan.

For instance, he says you can have a higher withdrawal rate if you put 20% into smaller stocks. This has historical returns to back-up the theory. However, I find most people invest in smaller stocks because of the higher returns -- yet they quickly get out after the first significant downturn (especially a severe downturn). This would be more likely for someone in retirement.

One of the most intriguing aspects was his graph showing suggested withdrawal rate versus rebalancing timeframe. This surprised me. I do rebalancing 2x/year because it forces me to sell my winners and invest in my losers. I believe I could only rebalance 1x/year but I have significant cash to invest 2x/year and it works well with my current situation. It has kept me honest by not invoking emotions. He suggests that the highest returns can come from rebalancing every 4 years. Any thoughts?

My personal situation is that I'm considering retirement at 45 (I'm 40). I've used the FIRE model but am still concerned (maybe say "scared") because a) my money has to last for 45+ years, b) my kids are still 10+ years away from college and c) paying for health insurance could become unaffordable. Bengen's model assumes that most people are in the traditional 60 to 70 age range.

I'd be interested in anyone's thoughts.

Mike K.

[astroboy]

I also read the article a few days ago and it seemed to make sense to me. Seem to have read somewhere that Bengen was one of the original 4% SWR gurus - though technically he says it is 4.15%. As for someone with a long time frame, 40 years or so, he specifically says that is one of the scenarios where you would subtract .20% from your SWR to allow for a longer than normal 30 yr time which his 4.15% is based on.

[rs0460a]

I've always had a "problem" with the 4% withdrawl rate, since it does not allow for any variation in your retirement income. The article seems to address this, in some manner.

In my case, I am retiring at age 59. I do not plan on drawing SS till I'm 66. That being the case, my year-by-year withdraw rate is as follows:

59 - 4.47%
60 - 5.55%
61 - 6.23%
62 - 6.58%
63 - 7.19%
64 - 8.31%
65 - 9.41%
66 - 2.16% (drop!)

I don't pass the 4% withdrawl rate till age 83 (4.13%), and don't pass 5% till age 86 (5.39%).

My plan continues to have a "healthy estate value" through age 95+ (both DW/me).

In this example the 4% may be a "target", but there is great year-to-year varience based upon "outside influences"...

- Ron

[brewer12345]

I think it is an idiotic data mining exercise.

[preben]

Mike; a well diversfied portfolio has proven (historically) to have less volatility and higher returns - same applies when including small caps and other higher volatility assets in portfolio.

If one fears to "chicken out" in a down turn it makes more sense to re-evaluate overall risk tolerance and increase the fixed income/cash component.

As to re-balancing you misunderstand; if you twice a year have NEW money to invest ofcourse you should "re-balance" with them(as otherwise you would either be holding too much cash in portfolio or you would be buying the "winners" again). The studies however focus on people in retirement (no NEW money) and then nos seem to indicate re-balancing every 2 years or so to be optimal.

Personally I do it yearly or even more often if big gains/losses as I am more focused on maintaining my asset-allocation/risk profile than sqeezing out that extra % gain coming from momemtum. Cheers!

[rmark]

Torture the data until it screams.

Small cap stocks extra returns appears to be driven more by value factors, as poorly run large companies with low market caps recover.
(Arnott, "Disentangling size and value", Financial Analyst Journal, 2005)

Further, small cap stock returns appear to suffer when using dollar weighted returns. Most investors are too late to the party.
(Dichev, "What are investors actual historical returns")

The japanese market has had lower returns and higher volatility.
(Dimson, Marsh, Staunton, "Irrational Optimism")\

This looks more like an article to help financial advisors with their sales pitch.

[bosco]

Originally Posted by Ron'Da

In my case, I am retiring at age 59. I do not plan on drawing SS till I'm 66. That being the case, my year-by-year withdraw rate is as follows:

59 - 4.47%
60 - 5.55%
61 - 6.23%
62 - 6.58%
63 - 7.19%
64 - 8.31%
65 - 9.41%
66 - 2.16% (drop!)

I don't pass the 4% withdrawl rate till age 83 (4.13%), and don't pass 5% till age 86 (5.39%).

My plan continues to have a "healthy estate value" through age 95+ (both DW/me).

In this example the 4% may be a "target", but there is great year-to-year varience based upon "outside influences"...

- Ron

wow--3 significant figure precision. I'm impressed--please share with us what you know about how the market will perform over the next 8 years. If I knew this, I could lay those kind of precision plans also. Also, what inflation rates we will have over that period. Or are you going to take a variable withdrawal? If variable, then why the precision?

[poyet]

Mikek,
Selling the winners and adding to the losers is the exact opposite to what successful traders do. I agree we do not talk of trading, but if done systematically it must lead to measurable underperformance. Traders add to the winners and cut rapidly the losers.

[rs0460a]

Originally Posted by bosco

wow--3 significant figure precision. I'm impressed--please share with us what you know about how the market will perform over the next 8 years. If I knew this, I could lay those kind of precision plans also. Also, what inflation rates we will have over that period. Or are you going to take a variable withdrawal? If variable, then why the precision?

Use the Fidelity Retirement Income Plan tool, and you too can have this forecasting data .

- Ron

[rmark]

"Selling the winners and adding to the losers is the exact opposite to what successful traders do."

We should differeniate between individual securities and broad asset classes.

With the latter, selling the overpriced and buying the underpriced seems much more practical.

[kaudrey]

Ron, I'm with you. I don't use 4% because there are steps when my retirement income will change.

I plan to retire at 52. From 52-57, based on my assumptions of 6% returns, 3% inflation for expenses (5% in healthcare), no mortgage anymore etc), my withdrawal will be between 5-6%. When my pension kicks in at 57, that will drop to about 3%. With SS at 65 or so, it will drop below 2%.

It doesn't make sense for me to use 4% as a rule.

Karen

[MikeK]

Originally Posted by poyet

Selling the winners and adding to the losers is the exact opposite to what successful traders do. I agree we do not talk of trading, but if done systematically it must lead to measurable underperformance. Traders add to the winners and cut rapidly the losers.

This seems to be in direct conflict with my experience and standard asset allocation practice. Perhaps you are talking about individual securities? I don't consider myself a trader. I do ride my individual securities for longer periods of time -- but still look at overall asset allocation. If they significantly put my allocation out-of-whack then I will sell a portion.

My experience and research with individual investors is that too often people try to ride their winners. The problem isn't that they do this, but that they don't set a strategy for when to sell a stock. The best investors have a plan and strategy. The worst investors keep their stocks because "they're a great stock" or "I'm waiting for it to recover". I speak from some painful, personal experiences

Mike

[bosco]

Originally Posted by Ron'Da

Use the Fidelity Retirement Income Plan tool, and you too can have this forecasting data .

thanks, I'll check it out. Just out of curiousity as to what figures they think will happen in the future.

Anyone here tried the retirement calculator offered at

http://www.analyzenow.com/Dynamic%20...ing_pro_IV.htm

?

[Nords]

Originally Posted by bosco

Anyone here tried the retirement calculator offered at

No, but have you read Hebeler's book? It makes 4% seem wildly hedonistic and he'd probably have a h0cus conniption fit with the SWRs quoted here.

It seems ironic that the former president of an aerospace corporation doesn't do online sales, let alone PayPal...

[poyet]

Originally Posted by MikeK

This seems to be in direct conflict with my experience and standard asset allocation practice. Perhaps you are talking about individual securities?
My experience and research with individual investors is that too often people try to ride their winners. The problem isn't that they do this, but that they don't set a strategy for when to sell a stock. The best investors have a plan and strategy. The worst investors keep their stocks because "they're a great stock" or "I'm waiting for it to recover". I speak from some painful, personal experiences

Mike

The idea was to foster some discussion. I know you're not trading in designing an asset allocation strategy. But one must acknowledge that adding to the losers and cutting the winners will lower the returns. Matter of facts.

It is funny to observe that what we do with money is the exact opposite to what we do in our everyday lives (cf behavioral finance). For example, if you try to win customers, ladies, etc, and try a strategy which fails, you "do not add to the losing strategy" by trying to repeat it as is, on the contrary you wonder what went wrong and adapt by favoring what did work (if something different did). Basically you cut the losing part of the equation and add to the wining side. Nature does the same claimed Darwin, though the experience is chaotic (strategies, i.e. genes appear as the result of unpredictable mutations), but favors winning designs.

Be it ETFs, funds, real estate, stocks whatever you have chosen, if it is a loser no need to worsen the situation by adding (averaging is just a means of losing more - traders say do not try to catch a falling knife...). If it is a winner (whatever we talk about) should lead you to increase the exposure.

Of course, any strategy must include entry and exit points. Exit points should enable you to decide when to cut a loser or end a winner as at the end, when it bends, the trend is not any longer your friend.

Food for thoughts.

[rs0460a]

Originally Posted by Nords

he'd probably have a h0cus conniption fit with the SWRs quoted here.

He can direct his portfolio/$$$...
I can direct mine...

(Where does Ho*us fit into this? Other than his plan for 2% withdrawls)

As Conan says "do you want to live forever? )

- Ron

PS: Cute what you do with the "H-word!"

[48Fire]

Originally Posted by kaudrey

Ron, I'm with you. I don't use 4% because there are steps when my retirement income will change.

I plan to retire at 52. From 52-57, based on my assumptions of 6% returns, 3% inflation for expenses (5% in healthcare), no mortgage anymore etc), my withdrawal will be between 5-6%. When my pension kicks in at 57, that will drop to about 3%. With SS at 65 or so, it will drop below 2%.

It doesn't make sense for me to use 4% as a rule.

Karen

I agree, too, that the 4% is a good rule of thumb, but I am in a similar situation where a pension will kick a few years after I ER. Do you have any model that predicts the 6% now, 3% later, or are you just creating a model from existing ones that represents the different phases?

[rs0460a]

Originally Posted by 48Fire

I agree, too, that the 4% is a good rule of thumb, but I am in a similar situation where a pension will kick a few years after I ER. Do you have any model that predicts the 6% now, 3% later, or are you just creating a model from existing ones that represents the different phases?

I'll assume you are directing the question to me, so I'll attempt to answer it (in my own words). To start with, I use several "retirement calculators" in an effort to not get the "best one", but rather to ensure that they all show a similar conclusion to the facts that I am able to provide (each of them have their own input criteria).

Currently, I use FireCalc, Quicken, Financial Engines (through Vanguard), and Fidelity (Retirement Income Planner).

Of the available "products", Fidelity is the one that provides the best "breakdown" year by year, but it also requires a lot of input data to result in a good "finite result". Yes you can just put in your retirement year and required "net income" (it calculates your taxes, down to the state level) or you can go the "full route" by filling out a detail explanation of your expenditures. It looks at "certain items" in a different way. For instance, while the tool currently builds upon a 2.7% inflation rate (which can be overridden), it looks at your stated expenditures for medical insurance premiums and adjusts them at a 7% level due to the variance against the standard inflation rate.

You can also detail your forecast expenditures into the future, if you feel your "outlook" will change. For instance, DW/me currently spend about 20% of our current net income on travel on a yearly basis (which I know is high, but it's my wife's passion <my passion is my wife >). Anyway, we have our retirement net income budget for travel still at 20%, but we reduced it several times during our retirement due to the assumption that we will reduce the frequency and type of travel (rather than Europe, a bus trip to a New Jersey casino!).

The output report of the Fidelity tool will give you a detail (year by year) of your expected starting balance, your withdraw (also shows expected total taxes on the gross - calculates SS under the current 85% rules) and can display this information on a 90, 75, and 50% confidence interval (CI). Being that I'm very conservative, I use the option to specify my own level and take it to 95% CI.

Anyway, that's the story about the tool. I may not be an "expert" in its use, but I do find it to be more detailed, and a good "partner" in use with the other available tools (such as FireCalc, which looks at the historic data a bit differently).

I look at using multiple tools in this manner - If your mother thinks you're a terrific person, you may believe it. If your DW tells you the same thing, it's probably the truth (if your friends agree, you probably should be a politician ).

Hope that gives you the info you need - if not, just ask again....

- Ron