Withdrawals when Market is High?

FIRECALC models this - use the % remaining portfolio method to play with it and determine what you think your withdrawal rate should be.

Tried FireCalc (which is my general preference).

Went to "Percentage Remaining Porfolio" under "Spending Models".

Can't figure out where/how to adjust or change percentage. When I change the "0" to "95" there doesn't seem to be any appreciable change in the results.

I know, I know...I'm only through my first cup of coffee and I shouldn't try these things until noon, but....sometimes I'm still a dunce even then.
 
This is true when comparing constant withdrawal to % remaining portfolio withdrawal, but that makes a big difference.



My calcs show 4.35% is probably the safe withdrawal rate for 30 years with a portfolio of 50% total stock market and 50% 5-year treasuries, compared to 4% usually quoted for the constant inflation-adjusted $ withdrawal. And you won't run out of funds either, but you will have to do some serious belt tightening during a bad run of down markets as your income could drop to up to 60% in real terms - so it's best if your budget has a high % of discretionary spending. However, the average ending portfolio is 100% of what you started with in real terms after 30 years, and the lowest ending portfolio is 86% of what you started with real - not shabby at all. In fact you could almost think of this as a "perpetual" rate, and it does hold for 40 years (smaller data set though), but starts slipping at higher durations, so it's not quite perpetual.

FIRECALC models this - use the % remaining portfolio method to play with it and determine what you think your withdrawal rate should be.

Audrey, I believe this article backs you up

http://www.fa-mag.com/news/is-4-5---still-safe-27153.html?section=47
 
Tried FireCalc (which is my general preference).

Went to "Percentage Remaining Porfolio" under "Spending Models".

Can't figure out where/how to adjust or change percentage.

I know, I know...I'm only through my first cup of coffee and I shouldn't try these things until noon, but....sometimes I'm still a dunce even then.
How to adjust the percentage - that is done in the first tab by entering your spending (i.e. withdrawal) amount versus your portfolio size. It's not entered anywhere as a %. The default values given there of 30000 and 750000 give you a rate of 4%. It's a small box on the side so not easy to find even after coffee.

When I change the "0" to "95" there doesn't seem to be any appreciable change in the results.

That's because the 95 rule is almost identical to the straight %remaining portfolio in most cases. It's only when you compare some of the worst case runs that you see a divergence, and even then the difference is small. So yes - overall there is no appreciable change in the results.

You can conclude then, I think, that if the 95% rule appeals to you, it's as good as good as the straight % of remaining portfolio. In terms of drops in income and terminal value it is only very slightly worse.

The big reason is that the 95% rule does not adjust for inflation, so if inflation is the big headwind in terms of portfolio growth/maintenance, then the 95% rule isn't protecting you that much, your income is shrinking much more dramatically in real terms even though it's only dropping 5% per year in nominal terms. But it is more gradual than the straight % remaining portfolio case.
 
Last edited:
Since this author is comparing SAFEMAX I believe they are still modeling the constant $ withdrawal method which is inflation-adjusted from the initial portfolio value when you retire. This is very different from the % remaining portfolio method which does not adjust withdrawals to inflation but calculates the withdrawal based on a constant % of the current portfolio value each year.
 
Last edited:
I'm not a fan of extreme belt tightening so would rather spend moderately in order to smooth over rough sequences of years. If you look at the simulation I showed here: http://www.early-retirement.org/forums/f28/test-of-constant-spending-model-85514.html
you will see some up years in what turns out to have been the worst sequence in post-WW2 times (started in 1966). You might even see a sequence of up years like 1970, 1971, and 1972 before a nasty recession pulls you down again.

I guess for me the answer to the OP's question would be that if my portfolio has grown substantially and I had already set aside a reserves fund within that portfolio, then maybe I would modestly start spending more. But my planning would already cover most desired fun/discretionary spending so why would I want to increase spending much more?

Personally, we are up about 11% from our inflation adjusted 2003 portfolio at the start of my retirement. So I could see maybe a bit more spending. The simulation ideas I showed in the that link above show a similar pattern ... yes, I am not an investment genius.

But if I'd retired in the year 2000 then the picture would be different. That same simulation would show us down about 18% at the end of 2016 from the start portfolio in 2000. Let me know if you want to see that simulation here.
 
Bogleheads has a good page on withdrawal methods:
https://www.bogleheads.org/wiki/Withdrawal_methods

The 4% SWR method is referred to as Constant Dollar because you adjust your withdrawal by inflation, meaning in real dollars your withdrawal does not change regardless of market results. The drawback for some is it is designed to withstand the worst historical returns, so most retirees will have significantly more money at the end of your life.

I like Variable Percentage, since it can adjust withdrawals in response to good or bad markets, but you do need the ability to cut back on discretionary expenditures in bear markets.
 
There is no shame in taking a profit when markets are bumping up against new highs.

If that allows one to pay off some bills, or plan a fancier vacation, great! Or just stash the money away for the day when the market is banging against new lows.

Agree. After 10 years of retirement when I generally just spent divs, I have started to systematically start liquidating a small percentage each year. The very strong recent market has helped me to finally get over my reluctance to encroach on principal. Will spend a little more but mostly just increase cash balances for now. I'm sure I can eventually find somewhere to spend/gift it.
 
So how does one determine what % to use? I'm curious.

You can search on lots of threads about withdrawal rate (WR).

I will only need a WR of 1.5% so I do see the rising market as justification for a one time pop in spending. I also take some $ out of equities to keep my AA even.
 
My calcs show 4.35% is probably the safe withdrawal rate for 30 years with a portfolio of 50% total stock market and 50% 5-year treasuries, compared to 4% usually quoted for the constant inflation-adjusted $ withdrawal. And you won't run out of funds either, but you will have to do some serious belt tightening during a bad run of down markets as your income could drop to up to 60% in real terms - so it's best if your budget has a high % of discretionary spending. However, the average ending portfolio is 100% of what you started with in real terms after 30 years, and the lowest ending portfolio is [-]86[/-]% of what you started with real - not shabby at all. In fact you could almost think of this as a "perpetual" rate, and it does hold for 40 years (smaller data set though), but starts slipping at higher durations, so it's not quite perpetual.

FIRECALC models this - use the % remaining portfolio method to play with it and determine what you think your withdrawal rate should be.
Correction: Oops - can't read my own spreadsheet. The lowest ending portfolio for the 4.35% remaining portfolio withdrawal method is 51%, not 86%.

86% was the ending portfolio for the case that resulted in the lowest annual income at some point during the 30 year period. So even though the portfolio and income dropped to 40% real of the starting portfolio and income, it had recovered to 86% real of the starting value.

Here is the overview of some of the results - all in real terms. I have only tested the 50% total stock market, 50% 5 year treasury allocation. Other allocations likely have different results.

% withdrawal remaining portfolio$1M starting portfolio income$1M starting portfolio lowest incomeEnding portfolio for lowest income caseaverage ending portfoliolowest ending portfolioincome from lowest ending portfolio
6.00%$60,000$16,248$308,050$593,418$298,371$17,902.28
5.00%$50,000$17,294$698,231$815,142$409,854$20,492.72
4.50%$45,000$17,470$817,320$954,171$479,758$21,589.12
4.35%$43,500$17,481$856,722$1,000,171$512,306$22,285.33
4.25%$42,500$17,476$884,004$1,032,020$518,901$22,053.31
4.00%$40,000$17,419$955,934$1,115,994$561,123$22,444.93
3.50%$35,000$16,593$1,176,616$1,304,200$655,754$22,951.38
3.33%$33,300$16,210$1,240,415$1,374,917$691,310$23,020.63
3.25%$32,500$16,018$1,271,583$1,409,464$708,681$23,032.12
3.00%$30,000$15,369$1,373,939$1,522,919$765,726$22,971.78

The starting years resulting in lowest real during the 30 year period were: 1892 for the 6% case, 1899 for cases 4% through 5%, and 1906 for cases 3 through 3.5%

As others have pointed out. For the higher percentage cases, even though they suffer the lowest income during the 30 year period, they start out at much higher income such that in relative terms, you are often still better off even though your income drops more. More efficient use of that retirement portfolio while living! It illustrates that you had better have a lot of discretionary spending and/or use good years to set aside part of the excess. You don't want to go into this scheme with a high withdrawal rate and mostly fixed expenses.
 
Last edited:
Back
Top Bottom