Greater than 4% Withdrawal Rate

[audreyh1]

In other words, resetting is not the problem -- the problem is in continuing to give yourself annual inflation raises during a period of high inflation and low market return, when you are in effect withdrawing at much higher-than-4% rates.

Bravo! You have hit the nail directly on the head! It's not the initial 4% that causes portfolio failure, it's the inflation adjusted raises even though the portfolio is underperforming.

Audrey

 

[Rich_by_the_Bay]

Bravo! You have hit the nail directly on the head! It's not the initial 4% that causes portfolio failure, it's the inflation adjusted raises even though the portfolio is underperforming.

Precisely.

If you only reset the SW amount when your portfolio is up, you are very vulnerable when you take that same withdrawal amount during down years. To be safe, either start with 4.x% and inflation-adjust annually, or use 4.x% of assets each year, whether the portfolio balance is up, down or the same.

 

[OkieTexan]

We're still going to be shooting at a moving target. Some years Peter gets robbed to pay Paul. Some years both Peter and Paul make a profit. That's just the risk in market investing. I hope to LBYM and stay well below the 4% SWR until I get a few years older.

FIRECalc is a great tool and it made me feel better the first few times I ran my numbers throught it. However, the timing the bulls and bears choose is out of my jurisdiction!

Add economics and politics to the mix and I am really in the dark. Will Congress respond to the Fed Chief's warning about "the consequences of baby boomer retirements" looming around the corner? Will economic disparities between those who have planned for retirement and have a $1M nest egg and those who have saved nothing lead the US to adopt a "wealth tax" to bail out future generations of consumers? These are only a few issues.

To make a solid plan for financial independence is indeed a wise thing to do. But I would probably think myself insane trying to finely predict the results over the next 30 years! An article in the current issue of the Wilson Quarterly magazine suggests that a sizable chunk of our accumulated wealth will probably end up in the hands of charities or the government anyway.

I wonder if Art Buchwald ever worried about SWR or Monte Carlo simulations?

 

[Bikerdude]

I wonder if Art Buchwald ever worried about SWR or Monte Carlo simulations?

I think not. I believe he was working until a few weeks ago. God bless a very funny man.

 

[Running_Man]

To make a solid plan for financial independence is indeed a wise thing to do. But I would probably think myself insane trying to finely predict the results over the next 30 years! An article in the current issue of the Wilson Quarterly magazine suggests that a sizable chunk of our accumulated wealth will probably end up in the hands of charities or the government anyway.

I know of no successful company that does not budget for both the short and the long term. Indeed it is a key control and process to almost every enterprise. To ignore the short term and only focus on a well made long term budget in essence is to ignore financial reality of current situation, whether it is improving or declining. The current situation has affected your long term budget - not reviewing is merely refusing to perceive the change.

If there are enough funds or fortune is on your side then ignoring the changes that are real have no consequence. To anyone who has enough money that a 4% or even 3% withdrawl is not a real necessity is truly fortunate. For many though, in many cases widows of the financial planners of a family end late in life with no portfolio as an all too real occurence. At the same time a safe increase in a portfolio with a reasonable (95%) chance of success may be worth adjusting to have a more enjoyable retirement. Both situations deserve review whether annually or biannually in my opinion.

Sometimes it may be best to steer your boat to avoid the icebergs or maybe into a port for some well-deserved R&R. For those with enough funds to build boats as needed, direction is irrelavent.

 

[samclem]

Actually, it is. It takes the data for S&P500, bond and CPI and looks for the worst case historical retirement survival sequence. Thus, it is by definition a worst case calculator. You can argue about whether fixed expense ratio, or any other data within it are appropriate, but the simulation portion of the calculator is a worst case engine.

Well, yes and no. As I understand it from rading the FAQ page, FIRECALC uses the actual historical data for these assets for each year and calculates the result for that year. Then the next year, (using the actual data), etc through whatever series length was requested (e.g. 30 years). Success or failure for that series is logged, then it does another 30 year window starting with the next year of actual historical data. FIRECALC uses all the years in the database, not just the bad ones. It is not a "worst case" engine, it is an "every case in the historical record, with emphasis on the US" engine. Now, if we recognize that the US economy has done exceedingly well over the last 50 years (esp relative to other world economies) for reasons that are unlikely to be repeated, we can appreciate the potential limitations of the predictions we base on the FIRECALC results. It's probably the very best tool out there, but it is limited by the data at hand. We all have to make guesses and take our chances, but nobody should use FIRECALC thinking he/she is just seeing worst-case data.

So, definitely, the 4% number should not be viewed as a rigid rule. I'm a fan of "take 4% of the available balance at the end of each year," with frequent checks to determine if your purchasing power is eroding/gaining.

 

[sgeeeee]

. . .It is not a "worst case" engine, it is an "every case in the historical record, with emphasis on the US" engine. . . .

No. . . Firecalc tests every case in its data base in order to find the worst case. Only that worst case defines SWR. The other data is ignored for that result. That is what FIRECalc does, and that is a worst case simulator.

 

[wabmester]

No. . . Firecalc tests every case in its data base in order to find the worst case. Only that worst case defines SWR. The other data is ignored for that result. That is what FIRECalc does, and that is a worst case simulator.

I've mentioned this before, but I think it's also useful to know the historical best case. Ask FC to find the WR with a 1% success rate, and that will give you the best historical WR for your portfolio and period.

I think it was something like a 10% WR for the default case. Something to hope for, eh?

 

[perinova]

I've mentioned this before, but I think it's also useful to know the historical best case. Ask FC to find the WR with a 1% success rate, and that will give you the best historical WR for your portfolio and period.

I think it was something like a 10% WR for the default case. Something to hope for, eh?

Humm good to know. That means 10% WR pretty much guaranties failure...

 

[donheff]

and poor Joe, who had managed to accumulate a much higher net worth before retiring and takes the same conservative 4% withdrawal rate... goes down in flames along with Jane, victim of that remaining 5%

Nah, because Joe couldn't spend all the excess he was pulling when he adjusted upwards so he had a spare pile of cash to cover the down times.

 

[Robert the Red]

No. . . Firecalc tests every case in its data base in order to find the worst case. Only that worst case defines SWR. The other data is ignored for that result. That is what FIRECalc does, and that is a worst case simulator.

It is only a worst case simulator if you look at the 100% SWR. For any lesser percentage, it will be looking at a number of bad cases, only one of which will be the worst.

If you are not looking at the worst case (100% SWR), then the "resetting withdrawal amount up after a good year" (RWAUAAGY) has at least one hidden assumption in it. The probabilities in Firecalc are like assuming you take your money and are transported back in time to a random starting point -- the probabilities are the average over this collection of starting points. However, in the RWAUAAGY approach, you are only being time traveled back to start at time points that are just after a good year in the market. IF the financial time series (inflation and markets) that go into Firecalc are statistically independent from year to year, then the choice of totally random starting points or just the collection of starting points after good years is unimportant. But if the time series are NOT independent from year to year (that is, there is correlation), then the choice of which collection of starting points makes a difference.

In an extreme correlation case (just for illustration), if every strongly good year was followed by a strongly bad year, you might be in a situation where retiring at a fixed SWR+inflation method gives 90% success when considering all sequences of years. But retiring just after a bad year might be 100% and just after a good year 80%. Averaged out, we get 90%. The extra information (the market's performance last year) has predictive value if there is "serial correlation" in the financial time series.

Of course, real time series (anti-) correlations are not so strong, but correlations do exist. This may be one source of the intuition expressed earlier that RWAUAAGY is OK after several good years in a row -- that whatever correlations in the financial fluctuations that may be there have died out -- that this year isn't that correlated with 5 years ago (say).

I'll add that the use of historical time series in Firecalc is implicitly a rejection of the idea of statistical independence between years. Otherwise, a Monte Carlo approach, choosing a random sequence of years (rather than the actual sequence of years), could be used to give more statistical power by averaging over more sample cases -- because there are a lot more random sequences of years than actual sequences.

 

[rmark]

A retiree is in their own personal bear market, as investments must be sold to support their lifestyle, resulting in lower lows and lower highs than investors who are saving. In essence they have reduced return due to market volatility. The studies look for a high degree of safety over a typical retirement period, looking for the “bad runs” that cause portfolio failure. Due to the lack of good worldwide historical market data, most research has focused on raw U.S. data, unadjusted for typical investor actions. This is logical in the sense of academic research, where reproducible results are important, but limits its “real world” usefulness where badly timed allocation changes, non-average portfolios, and hidden investing costs prevail. And historically these did prevail.

So the studies generally result in

U.S. past 60/40 return 6% - 1% volatility = 5% gross return
High safety 60/40 return 6% - 2% volatility = 4% gross return

While generally ignoring this

Lower international returns 5% - 2% volatility = 3% gross return
More conservative 40/60 retiree 4% - 2% volatility = 2% gross return
Minimal 1% historic investing costs 3% - 2% volatility = 1% net return
Badly timed allocation changes? Taxed reinvested withdrawals? = 0% return

In Guyton’s case, if I recall correctly, he tested against a period with a high starting expected return that became an even higher realized return due to capital gains. Coming up with a higher withdrawal rate would be expected. Bengens “layer cake” looks more like a marketing tool, and the temptation to pretend that the retiree will be a ‘superinvestor’ to make the numbers work out must be tremendous.

 

[sgeeeee]

It is only a worst case simulator if you look at the 100% SWR. For any lesser percentage, it will be looking at a number of bad cases, only one of which will be the worst. . .

I'm not sure why there is so much resistance to the fact that FIRCalcs SWR simulation is a "worst case" simulator. But it is. If you look for a 95% probability of success, FIRECalc simply simulates all possible historical cases and selects the 5% worst cases. It is still searching for worst case results. How many of the bottom it selects does not change the fact that it is looking for worst case(s).

As far as resetting SWR after a good year goes, there are plenty of reasons to be conservative. Some of the explicit and implicit assumptions behind using a historical simulator to determine SWR are:
1) The future will be no worse than the worst case (or cases) of the past.
2) The length of the retiree's retirement is known (30 years or whatever you set in the simulation)
3) The investments of the retiree are well represented by the S&P 500 index fund, a bond fund, and other asset classes that have been tabulated (with appropriate fees).
4) Your annual spending pattern is well described by the inflation driven spending model in FIRECalc.

The first assumption is merely that. We have 120+ years of history to look at, but no one really knows that we've seen the worst. The second assumption is never true. The third and fourth assumptions are probably not completely true for most of us. These assumptions alone leave enough room for error to consider treating the ~4% result with some skepticism. You can run perturbation simulations that address the last 3 assumptions. What happens to SWR if you use 33 years? 36 years? 39 years? etc. What happens as you adjust expense ratio? your asset allocations? etc. What happens if you simulate periodic increases or decreases in spending? These perturbations alone will cause SWR to fluctuate by several tenths of a percent.

If you play with these simulations for long enough, you will develop some level of confidence about the results. (It may be a very low level.) Let's say you simulate your retirement picture as closely as possible to what you consider your most probable situation is and will be. Further imagine that you come up with a 4.2% SWR ($42,000 annual spending per $1M portfolio). Now you run perturbations on all the variables over ranges you feel are reasonable and you get SWRs from 3.9% to 4.5% ($39,000 to $45,000 per $1M). If you start making <$3000 per year adjustments to your withdrawal rate because of superior market performance, you are probably only fooling yourself. You are already living with +/- $3000 per year risk. You are making refinements inside the error bars. You can make the adjustments and live with the same risk you started with, or you can maintain your level of spending and reduce some of that risk. Either response can certainly be justified based on your own risk profile.

But if you have a single sudden windfall year and FIRECalc places your newly calculated withdrawal rate at $50,000 per year, then that result is as valid as your original result. The assumptions have not changed. You do not take on additional risk by using FIRECalc's predictions after a single windfall year over the risk of using FIRECalc to start off.

 

[REWahoo]

I'm not sure why there is so much resistance to the fact that FIRCalcs SWR simulation is a "worst case" simulator. But it is.

The guy who wrote the program appears to agree with you:

... Firecalc is a worst case planning tool...

 

[Cut-Throat]

Well said Sgeeeeeeeeeeee!

 

[cute fuzzy bunny]

Hmmm, that working wife thing really helps out. Running firecalc with her income factored in and going for a 95% success factor says I can withdraw 7.32% per year.

Woo hoo!

 

[sgeeeee]

Hmmm, that working wife thing really helps out. Running firecalc with her income factored in and going for a 95% success factor says I can withdraw 7.32% per year.

Woo hoo!

As long as you keep her working you can spend like a drunken sailor.

 

[cute fuzzy bunny]

Shoot, i'm gonna go tell her to put in for some overtime!

Heck, she's only working 2 days a week now.

 

[WRBT]

Yes that perfect panacea... spouse works and your withdraw a pretty safe 0%

 

[Nords]

I think not. I believe he was working until a few weeks ago. God bless a very funny man.

Heck, he wrote & published a book while he was waiting to die at hospice.

How many people make a profit out of being hospitalized at death's door?!?

 

[cute fuzzy bunny]

Nah, she gets to drive the lexus. I only get to drive it when its out of gas. I'm stuck with either the Pilot with the baby seat in the back (at least its not a minivan) or the rav4 that smells like dog.

Which means I pretty much have her captive. She can only get about 380 miles away before she'll have to figure out where the gas goes in. But since she intubates people, it might not take her long to figure it out.

 

[Nords]

But since she intubates people, it might not take her long to figure it out.

I hope that doesn't involve enemas, or she'll ruin the muffler & catalytic converter...

 

[cute fuzzy bunny]

Nah, intubation is at the other end...

 

[Nords]

Nah, intubation is at the other end...

Oh.

Well in that case I'd be very concerned about your radiator, your oil, and your windshield washer!

 

[ESRBob]

You mean I shouldn't have spent all those years selling my wife on the benefits of ER? Now ya tell me. She liked the pitch so much she quit work a year before I could.