SWR Question

[grumpy]

Now that I am fully retired (as of 9/30/04) I have been thinking about portfolio withdrawals. Can someone answer the following questions for me?

If I have already determined that a 4% SWR will meet my spending needs with a 95% probability of success, and my portfolio returns more than 4% plus inflation in the first year of retirement, can I safely withdraw all of the portfolio gain for the year?

My reasoning is that, if I simply maintain my portfolio value at the initial level it had when I did the SWR calculation, then that rate will continue to be safe in the future given that same portfolio value. Furthermore, I will have one less year of life expectancy at that point so my probabiliity of success should actually increase slightly.

Until I encounter a year where my portfolio fails to earn more than 4% plus inflation (or suffers a loss) I believe I can safely withdraw all portfolio growth above my initial starting value at retirement. These extra funds might be used to splurge on a nicer car, extra travel, etc.

Is there a flaw in this logic? Any comments will be appreciated.

Grumpy

I think your plan is sound.

Another way to look at it, is to run FireCalc with your portfolio balance after your withdrawal, less 1 year of life expectancy. You may find that it is now 100% Safe.

[sgeeeee]

. . . Is there a flaw in this logic? *Any comments will be appreciated.

* * * * *Grumpy

No. There's no flaw. The potential flaw to SWR as computed by a historical simulator is in the basic assumption that the future will be no worse than the worst case in the past. Provided that implicit assumption remains true, then you can legitimately re-run FIRECALC each year and increase your SWR every time returns are good enough to justify that increase.

Note also that if returns are poor and FIRECALC produces a lower SWR than the previous year, you do not need to reduce your target SWR. This fact makes a lot of people uncomfortable, but if you feel comfortable with the implicit assumption about the future being no worse that the worst case in the past, then it is true.

[Hyperborea]

I see one small flaw and that is ignoring inflation. *Assume that you have a $1M portfolio allowing you to withdraw $40K at the start of year 0. *With inflation of say 3% then after one year at the start of year 1 you would like to withdraw $41.2K to keep your buying power level.

If you had withdrawn all the growth in the portfolio over year 0 then you would have a $41.2K withdrawal coming from $1M portfolio. *That would be a 4.12% withdrawal rate (WR). *This is not the same as the WR possibly climbing while following the SWR study rules as these also assume that growth when it occurs beyond withdrawal needs is left to compound.

This WR would continue to climb if you wanted to counteract inflation. *The question becomes one of whether the increased withdrawal rate is counteracted by the shorter withdrawal period.

[Bob_Smith]

If I have already determined that a 4% SWR will meet my spending needs with a 95% probability of success, and my portfolio returns more than 4% plus inflation in the first year of retirement, can I safely withdraw all of the portfolio gain for the year?

I'd be very reluctant to spend all my gains until I hit a decline, especially at the start of a potential 40+ year period. A string of good years in the beginning may be a prelude to a severe decline. Holding on to the excess in the good years may save you if that turns out to be the case. If I were further along, I might start increasing my withdrawals, but not at the beginning. But that's just me.

I'd be very reluctant to spend all my gains until I hit a decline, especially at the start of a potential 40+ year period. A string of good years in the beginning may be a prelude to a severe decline. Holding on to the excess in the good years may save you if that turns out to be the case. If I were further along, I might start increasing my withdrawals, but not at the beginning. But that's just me.

While I may react the same way as you Bob, if you run FireCalc every year and took the SWR for that year you would be 'SAFE' from a Historical perspective.

Sometimes I feel we are so pessimistic here as group, that our heirs are going to end spending the bulk of our money. Remember the SWR is the WORST Case from a historical perspective.

[sgeeeee]

I'd be very reluctant to spend all my gains until I hit a decline, especially at the start of a potential 40+ year period. A string of good years in the beginning may be a prelude to a severe decline. Holding on to the excess in the good years may save you if that turns out to be the case. If I were further along, I might start increasing my withdrawals, but not at the beginning. But that's just me.

From the discussions that have taken place here in the past, I think a lot of retirees would be reluctant to even keep spending level if they faced negative returns for several of the first 5 or 10 years of retirement. A lot of people have indicated that they would probably tend to cut back if their returns were weak and not try to use their inflation adjusted SWR that they computed at the beginning of their retirement.

So if the original question was about how others would handle the situation, you will probably hear a lot of conservative spending suggestions. But if the question is about re-applying the FIRECALC analysis after experiencing high returns and increasing your SWR . . . then that is certainly just as valid as using the FIRECALC SWR from the beginning.

Hyperborea brings up a detail in the calculation that is important. Let's say that you knew you were going to live 30 years and you ran FIRECALC to compute your SWR. After year 1, several things have changed: 1) your portfolio grew or shrank, 2) inflation affected your budget, and 3) your life expectancy is 1 year shorter. You have to look at all three changes to evaluate whether you end up with a higher SWR than the original one you calculated after considering inflation. Of course this is absurd for at least two reasons. You don't know how long you are going to live, and you don't know what your personal inflation rate is. So it is probably not advisable to take these numbers that literaly.

On the other hand, if you start your retirement and have 3 or 4 years of stellar returns with low inflation, it probably does make sense to re-evaluate your SWR and see if you can't get some value from an increased budget.

[cute fuzzy bunny]

You know what we need?

Some kind of magical, mystical tool that would do all of this for us. It could predict the future investment returns, calculate future inflation rates, determine any calamities that would befall us, and under ideal conditions, make a perfect cup of coffee for us every morning...

[Hyperborea]

You know what we need?

Some kind of magical, mystical tool that would do all of this for us. *It could predict the future investment returns, calculate future inflation rates, determine any calamities that would befall us, and under ideal conditions, make a perfect cup of coffee for us every morning...

Hey that sounds like the claims for the mythical "data based safe withdrawal" tool.

[Hyperborea]

On the other hand, if you start your retirement and have 3 or 4 years of stellar returns with low inflation, it probably does make sense to re-evaluate your SWR and see if you can't get some value from an increased budget.

In such a case I think it would be a reasonable thing to rerun the SWR calcs and increase your withdrawals unless you really want to fund a new building at your alma mater.

This is a big difference from eating the complete and total gains every year from your portfolio along with some base minimum guarantee of $40K. *If you removed the minimum guarantee then you would probably do fine but it's the combination of both all the gains and the min guarantee that is risky.

Gummy has a spreadsheet for calculating something like this. *It's almost a subset of his "Sensible Withdrawal Rates" calculations. *That, in general, assumes that you take a fixed base amount that is inflation adjusted and a fraction of the gains each year. You can find it at http://www.gummy-stuff.org/sensible_withdrawals.htm

[Bob_Smith]

Remember the SWR is the WORST Case from a historical perspective.

I'm not sure about this. I agree with the math, but my inflation rate is quite a bit higher than the official rate built into FIRECalc - primarily due to health care (and possibly some fudging at the BLS). I suspect that's the case for many ERs. So it really wouldn't have been a worst case for me. I don't really know what the worst case would have been under similar circumstances. So, I'd be inclined to bank any excesses early in the process and reassess after at least one full business cycle (maybe two). But that's just me, and I may be overly cautious.

[cute fuzzy bunny]

Yeah, considering "real" inflation for many of us may be 3-5% higher than CPI..."worst case" hasnt been breached. Not even close.

Of course you can take the CPI "substitution" approach and start eating hamburger instead of steak, enjoy the 20% lifestyle improvement your newer, faster computer brings you, cutting into your service levels and dropping services.

You hit a bottom on that at some point though, but the higher prices keep on a comin'...

Looking at my 70 year old dads perspective...his primary expenses are food, gas, utilities and health care. His healthcare costs have doubled in the last 3 years. Gas has gone from $1.65 a gallon to its current $2.37. Many common food items like beef, milk and eggs have gone up 30-40%.

He's very appreciative of his CPI mandated sub 3% social security boost

[ESRBob]

Grumpy,
Late to this post, but I agree with the general voice saying, "not so fast". The long term historical success rates are not writ in stone, and leaving some 'winnings' in during good years (2003 was a good year by any measure) are key to having portfolio survivability over the long run -- true in the historical data as well as good sense moving into an unknowable future.

What I do is slightly different and may be of use to you: I take 4% of the current year's portfolio (set in Jan) as my Safe WIthdrawal amount for the year. This gives you a bit of a 'raise' following a year like 2003, but is different from taking the entire extra portfolio earnings (above swr and inflation) as a bonus.

The every-year adjustment approach, from the few SWR studies I have seen on the topic, appears to be even safer and more conservative than the normal (set an initial withdrawal amount and tweak it every year for that year's inflation) SWR methodology. Your risk, of course, is reduced spending power after bad years, but the lighter touch on the portfolio helps portfolio survivability (you are not selling 'low' to such a degree) and the belt-tightening you do following a bad year is in synch with human nature. (Just as the 'raise' you want following a good year is human nature, which I suspect is the reason for your initial post).


ESRBob

[cute fuzzy bunny]

Real long term case studies?

The answer to this may be there and I just dont recall it, but has anyone published a good 20-30 year ER "what happened" story with their initial nestegg size, how they invested it, how much they took out, and what their lifestyles were like?

The only two I can think of are the Terhorsts, but they're not a typical ER scenario and Dominguez, which isnt a very pleasant story IMO.

Instead of models, historic data, guesses on what inflation might or might not have been, etc...a real life long term "How did it go"?

Anyone here almost reaching these sorts of time frames?

A 30 chapter play by play isnt really needed...just a "started with $1M, invested 50/50 in index a and index b, fixed withdrawal of x% or a variable withdrawal based on abc, lifestyle about the same at start and finish or differences noted.

[Hyperborea]

The answer to this may be there and I just dont recall it, but has anyone published a good 20-30 year ER "what happened" story with their initial nestegg size, how they invested it, how much they took out, and what their lifestyles were like?

The only problem with this is that single instances tell us little. *They may make one feel warm and fuzzy but statistically they are worthless. *They are also worthless from a financial planning standpoint. *It's much like the stories people have about their old gran who smoked a pack a day, drank a bottle of sherry a week, and ate nothing but bacon and eggs who lived to be 103. *They then conclude that the path to living a long life is just what their gran did.

While there may be people with just the right combination of genes and luck that would survive to 103 on such a regimen the odds aren't good. *We know that not smoking, drinking in moderation (particularly red wine - maybe some kinds of beer), eating a balanced diet, and at least a minimum of exercise will bring better odds but no guarantees of long life. *However, there will be outliers even doing the right things who die early (i.e. James Fixx).

The best that we can do is to take the path with the highest likelihood of success, build in some safety fallbacks, and adjust to changing conditions. *With our current data that probably means a diversified portfolio built of something between 50% and 80% equities spread amongst several sub-classes (including international and perhaps commodities), some assets in reserve (don't withdraw at the edge or have a house that can be downsized), and a variable withdrawal of some sort (either algorithmically or "by the seat of the pants").

The only two I can think of are the Terhorsts, but they're not a typical ER scenario and Dominguez, which isnt a very pleasant story IMO.

Dominguez took the path with a very low historical success rate but one that "appeared" safe because he foolishly ignored the effects of inflation. *He ended up living in a group home and likely doing his own dentistry. *I think it says a lot that he is the claimed "hero" of our troll.

[hocus]

Dominguez took the path with a very low historical success rate but one that "appeared" safe because he foolishly ignored the effects of inflation. *He ended up living in a group home and likely doing his own dentistry. *I think it says a lot that he is the claimed "hero" of our troll.

I'm not a troll, but Dominguez is indeed a hero of mine. He lived a life of great accomplishment. I have read a lot of books, and I can't think of one that had a greater positive influence on my life than "Your Money or Your Life."

Dominguez did indeed live in a group home. My understanding is that that was his plan from the start. So I don't see that as a bad thing. He decided that he was willing to live in a group home to be able to do what he wanted with his life sooner than would otherwise have been possible. I do not recall ever hearing him complain about it. If he was happy with the arrangement, I see no reason why I should be unhappy with him for making it.

I agree with you on the inflation point. I think that Dominguez completely dropped the ball on that one. I think that his advice on inflation is dangerous.

[cute fuzzy bunny]

Hyperborea -

I guess what I'm looking for is something more than the analytical measurements produced by firecalc. It "knows" what "reported inflation" has been, which I think is a mis-stated number. It knows mechanically what happened if someone did x, y and z infallibly.

I'm more interested in what really happened...not so much as a model to follow but for confirmation or denial of the "pure firecalc experience". IE...inflation was worse than expected, had to lower quality of life and withdrawal rate, tried variable withdrawals based on annual performance and that worked good/didnt work...etc. I have this experience, but only over a four year period and my lifestyle has changed so much in that four years that its simply not a good sampling.

By the way, as long as the grandma didnt smoke the bacon, I think thats a fine lifestyle!

[unclemick]

Not here either - But memory says researching 72t in late 1992 - 7% take out was available and heh,heh ah er conservative was to have 16-18 dollars saved for each dollar of income in retirement.

Even with the post 2000 period - returns were 'above the historical trendline' so our last twelve years ?probably? won't be necessarily reflective of the future.

Once we quit work - the get out of Dodge travel urge died after a couple years. Nobody got sick anymore from colds, flu. Our finances wouldn't fit well on a spreadsheet. Our cost of living - taxes, price of cars, food, etc. has creaped up - but our lifestye has creaped up - and staring at SS and maybe taking out some IRA money. I went overboard on CHEAP the first couple years.

I'd be surprised if someone posted twenty years of retirement neatly following a preplanned spreadsheet.

I'd be surprised if someone posted twenty years of retirement neatly following a preplanned spreadsheet.

Yup, Unclemick, I think we all roll with the punches!

If I started with a 4% SWR. Then I endured a 3 year bear market with a 40% decline, I would be less than a 2% actual withdrawal. - Human Nature.