# A different way to look at the 4% SWR

#### Sam

##### Thinks s/he gets paid by the post
This might already be common knowledge to the number crunchers.

Assuming a perfect financial world, where inflation is constant at 3%. The 4% SWR would be equivalent to an ROI of 7.3%.

In other words, if your ROI is constant at 7.3%, inflation is constant at 3%, you can use the 4% SWR forever, and your principal would also remain constant forever, in inflation adjusted dollars.

In a perfect world where your nestegg earned 4 % above the inflation rate you could draw the 4% SWR forever.

However that's not quite how the 4% number was arrived at. Remember that the 4% SWR comes from studies over 30 year periods. So rather than a perpetual 4 percent rake off the stash, you only need it to last 30 years. Therefore you can eat into the principal as the years roll on. And therefore the rate of return that you would need is less than you have quoted.

The stock market has averaged (depending on the index used) maybe 8-9 % above the inflation rate when measured over long periods of time. However It isn't a perfect world, and you need to take precautions from getting wiped out during severe and prolonged market corrections such as the 70's malaise. It just turns out that a 4 percent SWR should see you through many such rough periods.

Bernstein discusses this in his series on the retirement calculator from hell...

http://www.efficientfrontier.com/ef/998/hell.htm
http://www.efficientfrontier.com/ef/101/hell101.htm
http://www.efficientfrontier.com/ef/901/hell3.htm
http://www.efficientfrontier.com/ef/103/hell4.htm

#### Attachments

• SWR.jpg
31.7 KB · Views: 219
Last edited:
This might already be common knowledge to the number crunchers.

Assuming a perfect financial world, where inflation is constant at 3%. The 4% SWR would be equivalent to an ROI of 7.3%.

In other words, if your ROI is constant at 7.3%, inflation is constant at 3%, you can use the 4% SWR forever, and your principal would also remain constant forever, in inflation adjusted dollars.

Except that nothing is constant in the financial world - See - Volatility. (i.e. - A 10 year span of bad markets despite a 7.3% average return over 50 years, could bankrupt you)

SWR does not try to keep principal intact - A common misunderstanding.

FireCalc uses actual past history to run scenarios against. - This implies that the future won't be any worse than the past.

SWR does not try to keep principal intact - A common misunderstanding.

It was never a misunderstanding for me. But at 4%, the final average principal is ALWAYS greater or equal to the starting principal.

It was never a misunderstanding for me. But at 4%, the final average principal is ALWAYS greater or equal to the starting principal.

At 4% SWR with 100% success, you never go broke...

In a perfect world where your nestegg earned 4 % above the inflation rate you could draw the 4% SWR forever.

Not even true for the first year.
Start out 100, take away 4, left with 96. 96 * 1.04 = 99.84.

It was never a misunderstanding for me. But at 4%, the final average principal is ALWAYS greater or equal to the starting principal.

No it isn't! - Depends on your time horizon. Specify a 2 year retirement period starting in 1929 and at the end of 2 years your Average Principal will be a lot less than what you started with.

Too many variables to make a statement like that!

I don't think anyone plan a 2 year retirement period. I personally don't plan for anything shorter than 15 years.

I don't think anyone plan a 2 year retirement period. I personally don't plan for anything shorter than 15 years.

I used that for simplicity sake. Take your 15 years starting in 1966. You'll see your average principal is far lower than what you started with. Since the market had a flat return from 1966-1982, you would expect that, wouldn't you?

Are you sure you understand how FireCalc works?

Not even true for the first year.
Start out 100, take away 4, left with 96. 96 * 1.04 = 99.84.

You must be a lottery millionaire or something.

Some people live off of the earnings as they come in throughout the year. Most people don't go Whole hog one day and then impatiently wait for the next installment.

I used that for simplicity sake. Take your 15 years starting in 1966. You'll see your average principal is far lower than what you started with. Since the market had a flat return from 1966-1982, you would expect that, wouldn't you?
Edit: I just tried that scenario, and the average ending principal is \$1,105,048. The starting principal is \$750,000 (using all default values, except 15 years instead of 30).

Are you sure you understand how FireCalc works?
Yes, more than you think, I think.

Last edited:
You must be a lottery millionaire or something.

Some people live off of the earnings as they come in throughout the year. Most people don't go Whole hog one day and then impatiently wait for the next installment.

And you must be... Nah, you're worth it.

Edit: I just tried that scenario, and the average ending principal is \$1,105,048. The starting principal is \$750,000 (using all default values, except 15 years instead of 30).

I thought you were talking about the average ending principal starting in 1966 and ending each year until 1981. The average ending principal for each year in one scenario. And remember if someone retired in 1966, that the only scenario that you get.
Here are the ending balances (taking the defaults) from 1966 on - Since none of them exceeds 750,000 again the average ending principal is well below.
1966 -

693,294 691,116 589,921 567,541 580,609 606,918 468,863 334,362 369,750 353,459 285,213 262,340 229,028 212,024

Are you talking about the average ending principal of all starting portfoilo years from the late 1800's going forward? Then I agree with you. But It's not a useful figure. Remember FireCalc is a 'worst-case tool'.

Last edited:
Understood. Yes, 1966 is a bad year to start.

Not even true for the first year.
Start out 100, take away 4, left with 96. 96 * 1.04 = 99.84.

I REALLY urge you to start messing around with Excel and make
of this stuff is to run the numbers yourself. A simple spreadsheet
would have one column for income, one column for remaining
portfolio, one column for ROI, and one column for inflation. Each
row represents one year. The formulas are easy. If you don't know
how to do it, you should learn - it'll be fun and worthwhile. Then,
once you have it working, you can mess around for yourself and see
what happens when you have years of low ROI and high inflation
(aka. negative "real" return).

I REALLY urge you to start messing around with Excel and make your own spreadsheets.

Me?

Not even true for the first year.
Start out 100, take away 4, left with 96. 96 * 1.04 = 99.84.

Withdraw after the 4% gain.

Withdraw after the 4% gain.

Right, but then the withdrawal rate would be 3.85% and not 4%.

I REALLY urge you to start messing around with Excel and make
of this stuff is to run the numbers yourself.
I would also add to make a SS for last year's taxes,
verify it's correct and then you use it to do quick and dirty whatifs,
updating it for 2007/2008 changes in Cap. Gains rate for example, allows
you to do some tax planning.
TJ

It was never a misunderstanding for me. But at 4%, the final average principal is ALWAYS greater or equal to the starting principal.

An astounding insight!

Withdrawal studies center on
- 60/40 market weight portfolio, 25 year withdrawals, U.S. market returns and volatilities
- arriving at past 4% of initial portfolio value + annual inflation withdrawals, no resetting
- using lower & higher starting withdrawal rates for longer & shorter withdrawal periods
- more activities early with more healthcare later, for nearly level spending until later life

Average retiree divides average portfolio over initial life expectancy + annual inflation.

While they’ve usually omitted
- lower returns, higher volatility, and longer bad runs exhibited by other capital markets
- lower dollar weighted return, past investing costs, more conservative retiree portfolios

Average retiree divides an average portfolio annually over remaining life expectancy.

With some authors suggesting
- changing allocations 1% yearly toward more bonds has little effect on withdrawal rates
- retirees have declining expenses during retirement, reducing the need to offset inflation
- overweighting past higher return assets for higher withdrawal rate or insufficient assets

Retiree divides an age weighted portfolio annually over remaining life expectancy.

Replies
21
Views
1K
Replies
144
Views
9K
Replies
74
Views
5K
Replies
19
Views
2K
Replies
117
Views
10K