Calculating withdrawal rate

[mickeyd]

I thought the idea of SWR was based on the portfolio balance at retirement with COLA increases?

For planning purposes it's great but when you get down to actually living on/off your retirement stash you may actually decide to spend it in a way that pleases you. That way may be different than the way that you planned it. So what?

 

[youbet]

That's why I called mine a WR not an SWR.

Yep. One of the biggest points of confusion here on the board is the difference between SWR (as in the Trinity Study) and WR.

You obviously "get it."

 

[Hopeful]

For planning purposes it's great but when you get down to actually living on/off your retirement stash you may actually decide to spend it in a way that pleases you. That way may be different than the way that you planned it. So what?

I have no problem with that. As someone new to the ER concept I am just trying to understand the different draw down concepts, especially as it pertains to ER. So much information exists regarding retirement at 65, not so much about at 45.

 

[W2R]

That's why I called mine a WR not an SWR as I don't follow the Trinity Study SWR model of an SWR at the time of retirement and monitoring the annual WR against that original balance.

I think you would do fine with your proposed 3% SWR at age 45 and sticking to the traditional SWR model, as long as you didn't spend more during the up years and were able to trim back a little during the down years.

+1

Also, I do compute mine according to the Trinity "SWR" method later on after I get around to obtaining the CPI, as a double check. Actually I am not sure if the previous year's official, final CPI is actually even available on January 1st although perhaps it is. Anyway, whether it is or not, I do this computation later on. Although the Trinity SWR method does not allow as much money as the simpler method based on percentage of one's end-of-year balance, so far they have been at least somewhat close and my spending is within allowable limits according to either of these methods.

I also compute WR based on my lowest portfolio value of the recent recession (which for me occurred on 3/9/2009), and have been OK by that method too although it has been the most demanding of the three.

Finally, as a 4th method for doublechecking purposes, I total my dividends for the previous year and compare with my spending. So far my spending has been completely covered by my dividends although just barely last year.

 

[Meadbh]

an swr is really not a fixed amount especially if it is a method based on worst case scenerios. it can really be far to low over time. the fact is a swr and not just an wr has it changing as the years go by.

why in my opinion?

except for the inclusion of 2 very very bad time frames the historical safe swr would be around 6.5%..

it is when we include the 2 worst time frames that things get pulled that low.

now throw in the fact studies show the first 5 years are the most critical and the first 15 years determine your entire swr for a 30 year plus time frame and you may just want to keep adjusting things upward as you clear the critical hurdles.

as these major milestones are passed and all is well an swr can start to rise and still be an swr. if you are well above the 2% real return kices calculates we need as an average per year return over the first 15 years to keep 4% from failing then there is no reason by year 16 swr can't be brought up more in line with historical averages of 5-6%

I am in my second year of ER. I tracked pre-ER spending and calculated my budget based on that, making sure that projected annual spending would be 3% or less of both net worth and investable assets. I also had a financial plan done in year 1. It projected my assets to drop for the first two years due to debt repayment (mostly a car loan), but to increase every year after that (assuming constant conservative returns and expenses adjusted for inflation). Being acutely aware of the importance of the sequence of returns, my goal for the first 3-5 years is to avoid a drop in investable assets in any year. At the end of year 1, assets had increased by a multiple of expenses. If the market were to tank in the next few years, I would scale back expenses, particularly travel. But if things are going well, by 2017 or so I should feel comfortable to loosen the purse strings. Meanwhile I budget monthly and calculate the cumulative variance for each budget category, but I plan my withdrawal amount annually.

 

[mathjak107]

i like bob clyatt's method and i will use that method of setting spending floors.

i intend to use the pfau/lkitces rising guide path allocation method as well.

i have everything reduced down to 37% or so eqyuities right now and intend to increase 1% a year when i start drawing.

i am going part time in july and hope to phase out a year later totally.

 

[Alan]

an swr is really not a fixed amount especially if it is a method based on worst case scenerios. it can really be far to low over time. the fact is a swr and not just an wr has it changing as the years go by.

why in my opinion?

except for the inclusion of 2 very very bad time frames the historical safe swr would be around 6.5%..

it is when we include the 2 worst time frames that things get pulled that low.

now throw in the fact studies show the first 5 years are the most critical and the first 15 years determine your entire swr for a 30 year plus time frame and you may just want to keep adjusting things upward as you clear the critical hurdles.

as these major milestones are passed and all is well an swr can start to rise and still be an swr. if you are well above the 2% real return kices calculates we need as an average per year return over the first 15 years to keep 4% from failing then there is no reason by year 16 swr can't be brought up more in line with historical averages of 5-6%

Excellent post. Throw in the ages of achieving SS and or other income streams, and that SWR will continue to vary.

Since we are a couple retired at ages 55 and 54 we have 4 future streams of income (from UK and US SS, plus a UK private pension plan) due to come on stream between age 62 and 70.

For us the SWR method based on the Trinity Study would be difficult to apply, but what I will do, for fun, is recalculate the SWR at age 62, 63, 67 and 70 as those new income streams come on-line.

 

[mathjak107]

the rising glide path falls right into place for us. as i increase equities from 62-66 i will be increasing risk at the same time i will file for ss at 66 . that will reduce my dependency on selling from my portfolio at the same time we are raising equity levels and risk .

on the other hand if markets suck my first 5 years i will not have a full equity position to hurt me.

if markets are good the first 5 years i still have enough in equities to build up a nice cushion for downturns later on.

it seems win/win

 

[CaliforniaMan]

an swr is really not a fixed amount especially if it is a method based on worst case scenerios. it can really be far to low over time. the fact is a swr and not just an wr has it changing as the years go by.

why in my opinion?

except for the inclusion of 2 very very bad time frames the historical safe swr would be around 6.5%..

it is when we include the 2 worst time frames that things get pulled that low.

now throw in the fact studies show the first 5 years are the most critical and the first 15 years determine your entire swr for a 30 year plus time frame and you may just want to keep adjusting things upward as you clear the critical hurdles.

as these major milestones are passed and all is well an swr can start to rise and still be an swr. if you are well above the 2% real return kices calculates we need as an average per year return over the first 15 years to keep 4% from failing then there is no reason by year 16 swr can't be brought up more in line with historical averages of 5-6%

For later retirees, say with a 30 year planning horizon the critical times makes sense to me, however for early retirees it probably is not applicable. In fact for everyone this year is always year 1. So in effect the next 5 years are always the most critical, and the next 15 have the greatest effect on the following 15 years. This is an empirical and a statistical observation and makes obvious and intuitive sense for the 30 year long retirement. However, if you retire at 45 with say a 50 year planning horizon, I fail to see how taking into account only the first 5 or the first 15 years makes any sense. You really have to count each year as the first year, and when you do that, there is nothing helpful with you knowing that since the past 15 years were good, you now have less to worry about. You are still at year one, and still have 35 years. IMO there is too much importance given to these 5 and 15 year studies which are applicable to the later retiree only.

 

[mathjak107]

For later retirees, say with a 30 year planning horizon the critical times makes sense to me, however for early retirees it probably is not applicable. In fact for everyone this year is always year 1. So in effect the next 5 years are always the most critical, and the next 15 have the greatest effect on the following 15 years. This is an empirical and a statistical observation and makes obvious and intuitive sense for the 30 year long retirement. However, if you retire at 45 with say a 50 year planning horizon, I fail to see how taking into account only the first 5 or the first 15 years makes any sense. You really have to count each year as the first year, and when you do that, there is nothing helpful with you knowing that since the past 15 years were good, you now have less to worry about. You are still at year one, and still have 35 years. IMO there is too much importance given to these 5 and 15 year studies which are applicable to the later retiree only.

i couldn't even comment on longer than 30 years and what time frames are crutial since so far i have not seen studies look at that aspect.

 

[W2R]

For later retirees, say with a 30 year planning horizon the critical times makes sense to me, however for early retirees it probably is not applicable. In fact for everyone this year is always year 1. So in effect the next 5 years are always the most critical, and the next 15 have the greatest effect on the following 15 years. This is an empirical and a statistical observation and makes obvious and intuitive sense for the 30 year long retirement. However, if you retire at 45 with say a 50 year planning horizon, I fail to see how taking into account only the first 5 or the first 15 years makes any sense. You really have to count each year as the first year, and when you do that, there is nothing helpful with you knowing that since the past 15 years were good, you now have less to worry about. You are still at year one, and still have 35 years. IMO there is too much importance given to these 5 and 15 year studies which are applicable to the later retiree only.

What this brings to mind (for me, anyway), is that none of these methods, even the Trinity study, are exact predictors of future success to the extent that we could just put withdrawal on autopilot and forget about it. They are certainly not terrrible methods, but really I think that for all of us, our future success can be more greatly ensured by our adaptability to change as time passes, after choosing a satisfactory initial baseline from which to begin.

Should my confidence be reinforced because, looking back on it, I retired during what was called the worst crash since the Great Depression? Honestly I don't think so, even though I am in my 5th year of retirement and the market has done nothing but soar so far. But who knows; we might be on the verge of a market collapse that would make the Great Depression look like child's play (I don't have any reason to believe this is the case!!! But hypothetically anything is possible). My point is that we know about the past and the present, but we do not know very much about the future. As time passes and the future rolls into the present, we need to be nimble and adapt to whatever comes our way. Then we will be fine no matter what that may be.

 

[mathjak107]

the more i learn about these studies i realize more and more their value is in they are excellent for giving us guidance as to what failed more often than not holding the low water mark of 4%.

like building a house that can endure storms, if we at least build to withstand the worst of the past we are at least starting out on the right foot .

things can always be worse but you put the odds on your side by at leasting ruling out that which didn't work so well so far..

 

[braumeister]

But who knows; we might be on the verge of a market collapse that would make the Great Depression look like child's play

As long as you can stifle your urge to say "W#**!", we should be OK.

 

[CaliforniaMan]

the more i learn about these studies i realize more and more their value is in they are excellent for giving us guidance as to what failed more often than not holding the low water mark of 4%.

like building a house that can endure storms, if we at least build to withstand the worst of the past we are at least starting out on the right foot .

things can always be worse but you put the odds on your side by at leasting ruling out that which didn't work so well so far..

+1
Very good metaphor for what we are all trying to do in retirement!

 

[seraphim]

The guidelines mentioned are merely that - guidelines - and I employed the Trinity Study and FireCalc as quick reality checks to make sure my original plans were on on an acceptably realistic track. I had in mind a set amount of cash I wanted to withdraw annually, well below any guidelines, then would evaluate each year based on our spending habits. I had a chunk of spare cash the first year in savings, so we spent as we wished the first year and tracked our expenses, and found that our spending fell in line with our planned withdrawal. In January we evaluate for the upcoming year, and continue tracking expenses. We adjust our withdrawal each year to account for money left from the previous year's distribution.

We have an RMD from a non-spousal inheritance, which Is setup for automatic distribution from Vanguard on April 1, merely because spring is often a time when the indexes are higher. That distro can be changed at any time, so if all asset classes are down, I just delay the distribution. Or if the markets unusually high, take it early. The rest of the WD comes from a taxable account at Vanguard. The plan is make a lump withdrawal in January, but that is also flexible. The indexes dropped this January, so I waited until February's recovery. Isn't really necessary, but I sold fewer shares that way. Just economical, I think.

This July I plan to use some cash to buy a car, so may have to withdraw a few thousand more in the fall, or cut down down expenses. I have placed an arbitrary limit of never withdrawing more than 3% of the original portfolio amount in any given year. One goal is to leave a nice inheritance to Junior. With the current economic conditions and poor job conditions, he may need it. If our expenses start to hit or exceed the 3% figure, then it's time to modify our lifestyle.

I like the thought of flexibility in withdrawals, and conservative thinking - medical expenses may hit and wipe out someone who hasn't planned for a worst case scenario, and who hasn't devised alternative plans to deal with disaster. Even then, who knows what will occur, so I do not subscribe to the notion of a 'safe withdrawal rate', and I currently consider the 4% guideline (OK - 3% non-COLAd) as a ceiling, not a floor.