Skip to main content
No. of Recommendations: 12
WSJ has another long article on retirement calculators. Jesus people it's 4%.

http://online.wsj.com/news/articles/SB1000142405270230374990...

No matter how much money we may be saving for retirement, one nagging question lingers: Will it be enough?

Online retirement calculators aim to help solve that mystery. They can be useful tools, but don't rely on them for bankable forecasts, experts say.

</snip>


intercst
Print the post Back To Top
No. of Recommendations: 1
SWR??? Yeeccchhhhh. I get so sick and tired of that. It is a CRUTCH, people.

Seems to me the people need to look harder at what they are spending in retirement and maybe have to adjust accordingly.

As for the famous 4 percent stuff? Hell, I've been taking more than that every single year for several years, and my IRA is tiny vs these big ones people talk about! Ah, but I manage mine (as you all are no doubt sick and tired of hearing), so I keep it afloat anyway.

Vermonter
Print the post Back To Top
No. of Recommendations: 1
As for the famous 4 percent stuff? Hell, I've been taking more than that every single year

Of course!

The 4% rule is for something like a 97% or 95% success rate. So the odds are about 20 to 1 that you can take more.
Print the post Back To Top
No. of Recommendations: 1
/Has anybody done simulations which start at a set dollar amount in the portfolio, and then withdraw to cash a set amount with a rule based system? Perhaps sweep to cash anything in excess of 110% of initial port
value, reinvest from cash if port reduces to 80% of value?

Just curious JIm k
Print the post Back To Top
No. of Recommendations: 0
JimKredux asks,

/Has anybody done simulations which start at a set dollar amount in the portfolio, and then withdraw to cash a set amount with a rule based system? Perhaps sweep to cash anything in excess of 110% of initial port
value, reinvest from cash if port reduces to 80% of value?

</snip>


I'm not aware of a study done under those specific parameters. However, there have been all kinds of variations on the 4% rule. The Bogelheads website has a link to some of the studies.

http://www.bogleheads.org/wiki/Safe_withdrawal_rates

intercst
Print the post Back To Top
No. of Recommendations: 4
Has anybody done simulations which start at a set dollar amount in the portfolio, and then withdraw to cash a set amount with a rule based system? Perhaps sweep to cash anything in excess of 110% of initial port
value, reinvest from cash if port reduces to 80% of value?


I haven't done this directly, but I did do a quite comprehensive spreadsheet of the "cash bucket" strategy using the standard X% + CPI withdrawals. I suspect my findings would apply to your proposal.

Basically, it doesn't work out very well. At all.
In good times your allocation will shift toward cash, thereby reducing the gains you would be making in stocks. In bad times, you would be moving from cash to stocks while stocks were tanking, thereby losing money.

Numerous studies -- more rigorous than the naive web/magazine articles -- have show that the best strategy is to maintain your asset allocation and *not* sway back and forth depending on the vagaries of the market.
Print the post Back To Top
No. of Recommendations: 4
Has anybody done simulations...

Go here and you can run your own simulations.

http://www.firecalc.com/

JLC
Print the post Back To Top
No. of Recommendations: 3
Has anybody done simulations which start at a set dollar amount in the portfolio, and then withdraw to cash a set amount with a rule based system?

Jim - unless I am totally missing your question, this is what the "4%" system is and there have been hundreds of simulations.

The "4%" rule, as originally composed by William Bengen in the 1990s says If

#1 You put your portfolio is this mix of S&P500 and Total Bond Fund
and
#2 Withdraw 4% in the first year, increasing each year for the CPI (not your personal cost of living index)
and
#3 Make your withdraws from both sides of the portfolio in proportion to the portfolio mix (i.e. if you actual portfolio in December 2014 is 57% stocks and 43% bonds, your December 2014 withdraw has $0.57 of cash from stocks for each $0.43 from bonds)
and
#4 You re-invest all dividends/income back into their source (i.e. all stock dividends go back into stocks & bond interest back into bonds).

You have a high likelihood (>95%) of not running out of money over the course of 30 years.

There are a few other things baked into the cake - one of which is do not formally rebalance the portfolio annually. (Data shows using the above approach you actual return increases without rebalancing - well actually the return rate increases without rebalancing until one hits about 7 years. I take rebalancing every 7 years as approximating do not rebalance.)

The withdraw method suggested (#3 above) actually does an effective job of rebalancing the portfolio without added tax consequences and this process functionally is withdrawing funds from the side whose current value is highest.

Bengen has published a lot of work dealing with many issues - including various diversifications (small caps, foreign, IRA sheltered vs non-sheltered, etc.) approaches to dealing with the 5% situation when funds will not last 30 years and, in my view, the most important item - what situations are most likely to result in funds running out before 30 years.

If you want to learn about this work, get Bengen's book.
Print the post Back To Top
No. of Recommendations: 0
I think the key difference he is asking is about the idea of lopping off cash when the portfolio is way up, and moving that cash (assuming there is any) into stocks when the portfolio is way down.

I don't believe Firecalc can model this, and would be surprized if it or any other retirement modelling program actually does model this. It's just a weird thing to do.

Although I'm not quite sure exactly what he means by "withdraw to cash a set amount with a rule based system". Certainly the standard 4% SWR meets this criteria. The set amount being 4% of initial value, the rule being increased annually by CPI inflation. But nobody says "withdraw to cash", they just say "withdraw".
So either he is asking about a very different technique or he is totally unfamiliar with the 4% SWR method. Which is a pretty basic thing to be ignorant about.

The question reminds me a little bit of my problems understanding the initial Guyton withdrawal rules. The way he (Guyton) stated it, it could not actually be done in practice, since one of the rules unintentionally amounted to "wave hands here". This was corrected later, when Klinger joined him and they came out with the Guyton-Klinger rules.
Print the post Back To Top
No. of Recommendations: 0
Hi Rayyt,

I am aware of the 4%swr rule,what I was asking is after x years of cash or a short term bond ladder is set aside, does it make any sense to
sell stocks when the portfolio increases above a certain amount, and buy back in when the portfolio decreases below a certain threshold. I know that there are proponents of dollar cost averaging into the market over a long period of time, I was wondering if anyone has modeled " dollar value averaging" out of the market.

I was unclear about the question perhaps.


Thanks

Jim K
Print the post Back To Top
No. of Recommendations: 0
does it make any sense to sell stocks when the portfolio increases above a certain amount, and buy back in when the portfolio decreases below a certain threshold.

I doubt it.

I can't even really make sense of the question. What is meant by the "threshold"? What does that even mean? We're talking about retirement, so we're talking about a 30+ year period. Threshold can't be an absolute dollar amount or even a certain percentage of the initial value -- because a dollar today and a dollar 20 years from now have little relation to one another.
Threshold can't be a constant percentage growth, because that wouldn't make sense--it would be essentially trying to declare that the market grows according to the way you think it should.


there are proponents of dollar cost averaging into the market over a long period of time, I was wondering if anyone has modeled "dollar value averaging" out of the market.

Yes, and no.
Yes if the question is about withdrawal/decumulation strategies. That's the 4% SWR rule, for example. And also many other strategies, such as Guyton-Klinger (my personal favorite), "cash-bucket" strategy, etc. The topic has been studied and modeled extensively.

Unless you are an absolute newbie, you can't be unaware of this -- it would be the equivalent of someone asking about if studies had ever been made of the conjecture that 1+1=2.

Therefore I have to assume that you are asking some other question. But I have no idea of what that question might be. I suspect that you only have a hazy idea yourself. We all have had ideas that seemed promising until we sat down and tried to write the steps to implement the idea--and found that we couldn't.


The "no" relates to the topic of "dollar value averaging". If you are using the standard terminology and not a phrase that you just made up, then DVA is a mirage accumulation strategy which only works in people's imagination, because they only consider a very small subset of possible occurances and think that this restricted subset encompasses all possibilities.

Annyway...all the studies I've read conclude that the best way to run your portfolio is to stick to your planned asset allocation (60/40 or whatever) and take withdrawals & rebalances so as to stick to that allocation.


If you want to investigate a strategy of your own, you could start with this spreadsheet:
https://www.dropbox.com/s/xf4ma5blug27aws/SPY_Withdraw_by_Ca...

It was built for modelling a "cash bucket" withdrawal strategy, but could probably be modified to model the approach that I am guessing you are probably trying to do.
Print the post Back To Top
No. of Recommendations: 0
does it make any sense to
sell stocks when the portfolio increases above a certain amount, and buy back in when the portfolio decreases below a certain threshold


That sounds like market timing to me.
Print the post Back To Top
No. of Recommendations: 0
Thanks I'll look into guyton/klinger

JK
Print the post Back To Top
No. of Recommendations: 3
I am aware of the 4%swr rule,what I was asking is after x years of cash or a short term bond ladder is set aside, does it make any sense to
sell stocks when the portfolio increases above a certain amount, and buy back in when the portfolio decreases below a certain threshold.


If I understand your question correctly, the answer is no. The reason is the market doesn't care about the amount in your portfolio, it does what it is going to do regardless.

There are timing strategies that recommending modifying your portfolio balance based on certain factors. But none of those factors include how much money is in your port.
Print the post Back To Top