The Four Percent Rule

[old woman]

Not everyone expects to live exactly 30 years. The rule is a guide line only over 30 years so if you retire at 45 and plan to live to 95 take less. I retired at 66 so might not live 30 years, I took 5% this year because I had expenses that aren't annual. I might take too much until social security at 70 then cut back to 2-3%. I might take 2% one year then 6% the next always thinking about 4% but not letting a rule take over my choices.

 

[NW-Bound]

My expenses during working years were never steady. I did not buy a new car and paid cash every year, or had a major home repair, etc... So, now in retirement, whatever withdrawal amount I choose, how am I going to stick to that every year?

Some people withdraw the planned amount each year, but put aside what they do not spend to build a surplus. I do not separate money into physical piles like that. What I do is to look back at past annual expenses, and see what the average looks, and how each year compares to that average.

If I spot an uptrend in my spending, then it is alarming and something must be done to curb it. If I can reassure myself that the high expenses in one year were one-time events such as daughter's wedding, major home repairs, vehicle purchase, etc..., then it is not so scary because they do not repeat. If different one-time events keep popping up, bringing up the average, then my lifestyle is more expensive than I thought and must be simplified.

So, it is the average over several years that matters.

 

[NW-Bound]

By the way, FIRECalc can also model the effect of SS coming online later. It will tell you that you can withdraw more than 4% until SS starts, at which point you will cut back the dollar amount that you withdraw. Of course it makes sense, but running FIRECalc gives you something a bit more concrete than just the idea that it is safe. It takes into account the size of the SS check relative to the size of your stash, and how many years before you get there.

Again, that number is still just a guideline, but a lot better than a wild guess.

 

[pb4uski]

Have only started the FIRE journey, thus only one withdrawal so far (4%). I plan to follow the method samclem mentioned, i.e. 4% of porfolio value, with no inflation adjustment. The equity portion of a portfolio is there to provide growth, and inflation protection. So that's my inflation adjustment. YMMV, plus it's not chiseled in stone, and can be changed...

I plan to use a rachet strategy. Let's say that I decide that 4% is a sustainable withdrawal rate. My withdrawal would be the greater of 4% of the current balance or last year's withdrawal plus inflation. So if the portfolio grows by more than inflation, I get a "real" raise... as if I retired anew using a 4% WR.... if the portfolio goes sideways or declines, then I get last year's withdrawal plus inflation which is still sustainable.

The reason I plan to do this is that I think the risk of withdrawing too little and ending up as a rich old man who denied myself more than necessary is much greater than the risk of running out of money.

This mindset allows us to feel free to splurge a bit when we are "ahead of the game" and enjoy our retirement while at the same time being sufficiently prudent to hopefully avoid running out of money.

It also avoids the problem associated with a constant percentage of the current balance of having to take a significant cut in withdrawals in the event of a downturn or dying rich if investment results are favorable.

 

[NW-Bound]

I plan to use a rachet strategy...

I thought about that too. But my earned income stopped only 4 years ago. And I have not consciously tracked my expenses for much longer before that.

I do not want to "ratchet up" too soon, remembering the infamous quote by Irving Fisher right before the Depression crash: "The market has reached a permanently high plateau".

Steady, steady... Slow, slow...

 

[Danmar]

The reason I plan to do this is that I think the risk of withdrawing too little and ending up as a rich old man who denied myself more than necessary is much greater than the risk of running out of money.

This mindset allows us to feel free to splurge a bit when we are "ahead of the game" and enjoy our retirement while at the same time being sufficiently prudent to hopefully avoid running out of money.

I agree. But so far(10 years) I have just been spending dividends which currently represent a 3.7% yield. These divs have grown nicely (7-10% per year) and my portfolio continues to grow quite a bit as well. Total return For 2016 now about 20% . So I am now planning on liquidating small amounts of stock on a regular quarterly basis. Perhaps about .7% per year. This will increase my available total spend by about 10% per year. In the short term this extra cash will just accumulate in our savings account. Not sure what we will eventually spend it on but no shortage of possibilities.

 

[GTFan]

I thought about that too. But my earned income stopped only 4 years ago. And I have not consciously tracked my expenses for much longer before that.

I do not want to "ratchet up" too soon, remembering the infamous quote by Irving Fisher right before the Depression crash: "The market has reached a permanently high plateau".

Steady, steady... Slow, slow...

And not only that but also given your health insurance situation (and potentially those of many others here on the ACA), might be best to keep a larger than usual reserve and/or plan on more of your WDs going towards HC.

 

[audreyh1]

It also avoids the problem associated with a constant percentage of the current balance of having to take a significant cut in withdrawals in the event of a downturn or dying rich if investment results are favorable.

I avoid the significant cut problem by allowing unspent funds to accumulate outside our portfolio. These are available to draw on in the event of a significant "pay cut" by our portfolio shrinking and thus our income (withdrawal) dropping.

Our spending hasn't kept up with our withdrawals, so this has been easy to do.

No inflation adjusting either. I leave that out of the equation. Either the portfolio keeps up with inflation or it doesn't. If it doesn't, we live with less income. In the very long run, the portfolio should keep up with inflation. In shorter and intermediate timeframes, it might lag inflation.

 

[REattempt]

When people talk about being able to "safely" withdraw 4% of their invested assets each year (putting aside the many issues with this "rule" and the benefits of flexibility in post-retirement spending), they're talking about the 4% covering taxes as well as living expenses, right? And this includes capital gains taxes, correct? If so, isn't this "rule" somewhat useless because each person's tax basis in their investment assets will differ -- dramatically -- and so will the capital gains tax they need to pay when they're liquidating assets in retirement? How does capital gains tax play into the four percent rule?

Is the answer simply that, as a rough general guideline, based on historical results, a retiree can safely withdraw 4% of her investment portfolio each year, and if she has substantial capital gains then she will just have to pay a bigger chunk of that 4% withdrawal to the government that someone who does not have capital gain? So each of these people - the one with big capital gains in the one with no capital gains - can still withdraw 4%, but one of these people will have a lot less than 4% to live on, while the other will have most or all of her 4% to live on?

For a limited time, you can read the original here:

http://afcpe.org/assets/pdf/vol1014.pdf

 

[Sunset]

I plan to use a rachet strategy. .....

Is there not a considerable risk as you are taking a portion of every extra increase of the portfolio above the inflation amount. Like poking a tiny leak in the retirement ship.

Whereas, I think the 4% concept is that when the portfolio goes up an extra amount, that extra is left alone to compensate for when it is flat or goes down. And the retiree takes out the standard amount plus inflation.

 

[pb4uski]

No... not at all. Let's use an example. For discussion purposes, let's agree that 4% is a prudent WR in the circumstances. (The principle of ratcheting is the same whether the WR is 4% or 3% or whatever).

Say I retire on Jan 1 with $1 million and a 4% WR so on Jan 2nd I have $960k and it grows to $1.1m by the end of the year (a 14.6% return for the year... a good year).

Under the 4% rule and 2.5% inflation my second year withdrawal would be $41k. Under ratcheting my withdrawal would be $44k, a 7% real increase in spending so I can splurge on some things.

If someone else with $1.1 million retires a year after I did, is their withdrawing $44k from their $1.1m portfolio in their first year of retirement more risky than me withdrawing $44k from my $1.1m portfolio in my second year of retirement? If you think so, please explain why.

 

[REWahoo]

If someone else with $1.1 million retires a year after I did, is their withdrawing $44k from their $1.1m portfolio in their first year of retirement more risky than me withdrawing $44k from my $1.1m portfolio in my second year of retirement? If you think so, please explain why.

All things being equal, your $44k is less risky since you have one less year to live than does the new retiree.

 

[NW-Bound]

This dilemma has come up times and times again.

It seems like a paradox, because someone who retired at the peak of the market gets to withdraw much more than he could if he did at a market trough on the same portfolio.

Here are some actual numbers. On 10/9/2007, the S&P was at 1565. On 03/09/2009, the S&P was at 677! That's a huge drop of 57%.

Ignoring the bitty dividend, a $1M at the top of the market would become $430K later. So, if you retired at the bottom of the market, you could withdraw only 4% x $430K = $17K instead of $40K/yr.

So, how does one explain this paradox? It is actually quite simple.

If you look at all the squiggly lines that FIRECalc plots for any input parameter set, you will see that there's a wide range of outcome. Thirty (30) years from now, you may become a decamillionaire, or you may be totally broke.

If you take 4% at the top of the market, your portfolio may barely survive 30 years from now. But if you take 4% at the bottom of the market, you are most likely to die a lot richer than you are now. In either case, you've made it.

It's like a student passing a course with a grade A or a grade C. If you keep ratcheting your withdrawal when the portfolio goes up, you are increasing the chance that you will be barely solvent. You still pass the course, but instead of grade A, you will be sweating bullets barely making a C.

 

[pb4uski]

All things being equal, your $44k is less risky since you have one less year to live than does the new retiree.

Ugh... thanks for the reminder.

 

[pb4uski]

This dilemma has come up times and times again.

It seems like a paradox, because someone who retired at the peak of the market gets to withdraw much more than he could if he did at a market trough on the same portfolio.

Here are some actual numbers. On 10/9/2007, the S&P was at 1565. On 03/09/2009, the S&P was at 677! That's a huge drop of 57%.

Ignoring the bitty dividend, a $1M at the top of the market would become $430K later. So, if you retired at the bottom of the market, you could withdraw only 4% x $430K = $17K instead of $40K/yr.

So, how does one explain this paradox? It is actually quite simple.

If you look at all the squiggly lines that FIRECalc plots for any input parameter set, you will see that there's a wide range of outcome. Thirty (30) years from now, you may become a decamillionaire, or you may be totally broke.

If you take 4% at the top of the market, your portfolio may barely survive 30 years from now. But if you take 4% at the bottom of the market, you are most likely to die a lot richer than you are now. In either case, you've made it.

It's like a student passing a course with a grade A or a grade C. If you keep ratcheting your withdrawal when the portfolio goes up, you are increasing the chance that you will be barely solvent. You still pass the course, but instead of grade A, you will be sweating bullets barely making a C.

One thing that you are not factoring in is what REWahoo points out... that as ratcheting happens there are fewer years that the portfolio has to support. I concede that the risk of running out of money is higher.... I'm just trying to balance the conservatism embedded in a SWR... by definition a SWR is principally based on bad scenarios... but there are typically many more scenarios where the retiree dies with a boatload of money and ratcheting mitigates that outcome.

So if I'm 10 years into a 40 year retirement and things have gone well and I have much more money than I started with and have 30 years left, then it would be foolish of me to stick to a 10 year old estimate of what a SWR would be and not adjust.

I concede in an extreme scenario like you describe during the great recession that having previously ratcheted up would have made me nervous, but as I said before, the extra money is use for splurging so in that extreme event there would be plenty of room for belt tightening.

Like I said before, I'm more concerned about dying with a boatload of money than dying broke.

 

[NW-Bound]

The shorter duration indeed should be accounted for, as REWahoo pointed out. If one has been retired for 10 years, yes, it makes a difference. But the market swing can still dominate over that effect.

Let's take a retiree who ran FIRECalc for a 30-year duration in 2007, but decided to do OMY. In 2009, when he wanted to retire, what do you think FIRECalc would tell him when he ran his incredible shrinking portfolio with a 28-year duration?

 

[NW-Bound]

I forgot to add that I do not see a problem with ratcheting. I plan to do that myself.

If one's portfolio has doubled, so that his initial WR of 4% becomes 2% of current portfolio, there's no reason not to withdraw more. But if he doubles up his WR to match his portfolio exactly, then he is working himself down from a grade A to a lower grade.

On the other hand, if he's 1/2 way towards the end of his years, doubling his WR means going up from the current 2% WR to 4% WR but with 1/2 the years left. And that is certainly very safe (FIRECalc run of 15 years vs. the standard 30 years). FIRECalc will tell him to do WR of even higher than 4% with a 15-year duration.

 

[RobbieB]

I sure am glad I'm not the only one looking at it that way.

Yeah Baby, Blow that Dough -

 

[NW-Bound]

It's way too early for me to tell, being only 4 years into retirement. But can I see myself looking at a 6-figure portfolio instead of 7 figures?

Worse, do I want to see the leading digit decrementing? Not even that!

I have called myself a Scrooge, and now you know why.

 

[RobbieB]

But you have 2 houses and an RV.

Maybe one less house?

 

[NW-Bound]

With healthcare costs going up, I have been prepared to let go of both houses and just keep the 25' motorhome in the worst case.

 

[RobbieB]

You get rid of those rats yet?

 

[NW-Bound]

Thought would sell house with roof rats included.

One camera inside garage attic. One camera in upstairs attic. Both with motion detect and auto recording. No movement detected for the last 2 weeks, ever since the last rat caught in trap.

Fingers crossed.

 

[Looking4Ward]

.....So if I'm 10 years into a 40 year retirement and things have gone well and I have much more money than I started with and have 30 years left, then it would be foolish of me to stick to a 10 year old estimate of what a SWR would be and not adjust....

That's an interesting take and something that I've thought about as well. Into my third year of ER and while I've been living off my investments for almost 30 months now the value of my portfolio continues to be higher at year-end than when I started.

So it's hard not to do a mental "reset" and run the higher portfolio value with Firecalc but with less years - in other words, I've been calculating (for fun) a withdrawal rate based on 27 years rather than 30 using a portfolio value higher than what I started with three years ago.

The results are obviously much more than they were when I ran the same calculations three or four years ago with a lower portfolio and 30 year time span.

 

[FIREmenow]

So if I'm 10 years into a 40 year retirement and things have gone well and I have much more money than I started with and have 30 years left, then it would be foolish of me to stick to a 10 year old estimate of what a SWR would be and not adjust.

However, one could argue that your original 90+% success-rate predictions took into account those really good years (into the future) with the SWR staying the same.