Flirting with my number

[growing_older]

So I have done all kinds of FIRECalc calculations and have backed into a number for my desired minimum FI portfolio size. If I have more than that, my chances of success go up, which I like. If I have less, my chances of success drop below what I'm willing to take, so I'll keep working.

But in the last month, my portfolio has reached this number. Twice. But only for a single day each time before dropping back below it. With all the talk of war in the Middle East, it has now fallen well below my target by almost an entire year worth of expenses. So much for the elation of reaching my goal.

I'm starting to think that reaching my target is not enough for FI. I also need to reach it sometime other than on a market surge or else it will just drop back down again. Which means I really need to overshoot the goal in order to be anywhere comfortable that fluctuations won't push me back under it. Right now, I'm thinking I'll choose another life event in the next year or two and evaluate then. If I'm comfortably over, then hooray. If not, then back to One More Year, or the next appropriate life event.

Has anyone else faced this "toe over the line" issue with portfolio value and how do I make sure I'm not self selecting for RE at the end of a long run up, just in time for a long drop down?

 

[LOL!]

That's great considering August 2013 was the worst month in the market since May 2012. I'm envious because it reads like you went up while all I see is down. Or was your post delayed a month or two?

But yes, one should not retire at the high point and expect no problems going forward. However, retiring after a significant pullback while one's numbers look good is strangely satisfying.

 

[growing_older]

Yes, the two high points were early in the month, and it's been gut wrenchingly downhill from there. I've seen larger declines, but this one seems to be more emotional because it's happening just at and around the target number I set. At first I was hopeful that dips over and under were just daily fluctuations and the rising trend would continue, but after 4 weeks of down I seem to be resigned that a nice run up is not likely to endure, so I better have more margin of error in my numbers. Painful to watch, but probably less painful than making a move to ER while underfunded and regretting it.

 

[Cooked]

This is really good news. You have hit your 'number'. There is comfort in that. Now that 'number' has slipped away. I would ask two questions. First your biggest concern is whether your 'number' is correct. For example, maybe you have conservative estimates that have made you set your number too high. Second, if it it correct and you hit it and now the market is on sale, why not rejoice?
What I did at this point was to adjust my thinking. Once I went above then below my number I began the shift to the final AA. This meant accumulating more 'cash' reserves (another number) . Seven short months later I hit my 'number' again and have never been below is since AND my final reserves strategy was almost completely in place.

You just crossed third base. You can worry and hang on third or run for home plate. The difference in this game is if you tag home and reflect, you can always mulligan your way back to third base. Smile and run for home you have finished all the heavy lifting.

 

[W2R]

Growing_older, it sounds to me like your assessment of your dilemma is absolutely correct and realistic.

The best thing you can do, in my opinion, is to keep that in mind and exercise your own very best judgment about when to retire. Maybe having a cash buffer to rely upon for a few years might provide a little cushion in case of market decline, to use until the market goes back up.

By the way, congratulations!

 

[photoguy]

Has anyone else faced this "toe over the line" issue with portfolio value and how do I make sure I'm not self selecting for RE at the end of a long run up, just in time for a long drop down?

I'm struggling with this myself as we hit our number in march.

The way I see it is that you can put in a factor of safety in multiple places during your SWR calculations. E.g. increase the number, lower the % withdrawal rate, discount SS, etc. I'd try to look at all the places you may have been more conservative than normal before deciding.

 

[explanade]

How did you calculate the number? I think ING advertises that it can come up with your number.

Did you plug in different numbers until you got 100%?

 

[growing_older]

I'm trying not to over think any particular number. I've played with assumptions and confidence levels and I can make the number go up and down, but in this case I was just noticing that my portfolio had reached a nice plain 25x (4%SWR) my projected retirement expenses. There have been plenty of times my savings have reached (or retrenched from) nice round numbers, but this one seems to have added impact since it arguably allows FIRE. Will still be a lot of number crunching to feel safe, but it was interesting how different (and bad!) dropping below it felt, compared to other past fluctuations. Maybe I need to rethink my risk tolerance. Or maybe I just need to get 10% or more over and then I won't feel so nervous about exactly this number.

 

[Animorph]

If your number is conservative, I wouldn't worry too much about dipping below it once you hit it. But, make sure you have the right number with your retirement AA in place before pulling the plug.

The FIRECalc technical answer is that you have to hit your number on Dec 31/Jan 1. It uses annual returns by calendar year, so it doesn't hit exact market peaks and dips. Retiring on a market peak to the day is a little outside FIRECalc's assumptions.

 

[palomalou]

Ah yes, the one that got away. OP, I totally feel your pain!

 

[ERD50]

A 'number" that survives FIRECalc is a number that was reached at a peak (or at least a year end peak, as Animorph points out).

The failures come in starting years with a peak - that's how it works out. So it should be OK (historically) to retire at a peak with that WR.

But you mention 4%. That failed 5% of the time, not making it to 30 years. Are you comfortable with that? Is 30 years really an 'outside' number for you?

-ERD50

 

[growing_older]

But you mention 4%. That failed 5% of the time, not making it to 30 years. Are you comfortable with that? Is 30 years really an 'outside' number for you?

No, this is not my belts and suspenders, super safe, I will actually quit my jib number. It's just the first significant number that COULD mean FI and it's tantalizing to duck over it for a day, then fall back by a year's expenses. My actual focus on the number and sensitivity to dropping below it surprised me, as I have ducked over and under lots of round numbers on the way here. Never felt as strongly, perhaps because they were just way stations on the trip to ultimate FIRE. As indeed this number probably should be too.

 

[explanade]

Is there really a rule of thumb for x times annual expenses to arrive at some number?

Some have cited 36x for instance.

 

[Animorph]

Is there really a rule of thumb for x times annual expenses to arrive at some number?

Some have cited 36x for instance.

You can run all the calculators and come up with your own target portfolio number. It's different for everyone, but 25x annual expenses would be a decent early target. However, "The Number" as a concept did come from some source (I never paid attention to it) that maybe had their own calculation if you wanted to be super official.

I had a rough idea of my target number, which increased with inflation, and was prepared to retire whenever I hit it.

 

[explanade]

ING used to run TV ads:

Retirement Calculator - ING Your Number

 

[Katsmeow]

Is there really a rule of thumb for x times annual expenses to arrive at some number?

Some have cited 36x for instance.

Wee 33x is basically a 3% withdrawal rate. 25x is basically a 4% rate.

That said - for most people X is not the annual expenses. X is the amount of portfolio that they would be planning to withdraw each year.

And, for most people, that is not their annual expenses. That is, most people will have some amount of SS. Some people will have a pension. The withdrawal from the portfolio each year is the amount of expenses left after receiving SS or pension.

For us, when the dust clears (kids are out of school and out of the house), SS will cover roughly 2/3 of our annual expenses. So we really don't need our portfolio to cover 100% of our annual expenses. We really only need to have the portfolio to cover 1/3 of our expenses so I might think of x as being 1/3 of annual expenses not all of them.

Also bear in mind that for many people - particularly early retirees - withdrawal rates early in retirement may be well above 4% and withdrawal rates later may be very low, well under 3%. This is because as more sources of retirement income come online then there is less need to withdraw from the portfolio. Someone retiring before taking SS or pension will need to withdraw a higher percentage of portfolio than that same person might withdraw after SS and/or pension come on line.

 

[explanade]

Right but there are also taxes. So lets say they want to spend $50k after taxes. But they'd have to withdraw more than $50k to account for the taxes.

 

[Katsmeow]

Right but there are also taxes. So lets say they want to spend $50k after taxes. But they'd have to withdraw more than $50k to account for the taxes.

I always include projected taxes in my spending number.

 

[siamond]

A 'number" that survives FIRECalc is a number that was reached at a peak (or at least a year end peak, as Animorph points out).

The failures come in starting years with a peak - that's how it works out. So it should be OK (historically) to retire at a peak with that WR.

But you mention 4%. That failed 5% of the time, not making it to 30 years. Are you comfortable with that? Is 30 years really an 'outside' number for you?

-ERD50

Let me re-inforce ERD50's point. Say you picked a safe withdrawal % (SWR) that you feel is... indeed very safe (e.g. 3.5% with a well diversified portfolio).

At a given point in time, the portfolio target is reached, and person A retires (with an annual spend X = portfolio value times SWR). While person B (with the same portfolio) was distracted, and let one more year go by. Then person B considers retirement again, but darn, the portfolio value dropped 10%, so person B does NOT retire.

But wait... Person A's portfolio dropped as well. So how come Person B can't retire with the same spend X, in constant dollars. This makes no sense, right?

Fact is it is perfectly fine for Person B to retire with such spend. Now the reasoning critically depends on the fact that the SWR is VERY safe... Not 90% safe, but 98% safe or something like that. And to pay attention in the first few years that you're not in a super unlucky situation (e.g. like 1937 or 1965/66).

PS. There is a broader strategy behind what ERD50 and myself are saying. But this will be for a separate thread...

 

[RobLJ]

A 'number" that survives FIRECalc is a number that was reached at a peak (or at least a year end peak, as Animorph points out).

The failures come in starting years with a peak - that's how it works out. So it should be OK (historically) to retire at a peak with that WR.

But you mention 4%. That failed 5% of the time, not making it to 30 years. Are you comfortable with that? Is 30 years really an 'outside' number for you?

-ERD50

The 5% fear is an issue that is very much personal. I personally find some of those who insist on 99+% FireCalc (or two times FireCalc) to resemble obsessive compulsives, although I have no doubt they are doing what is appropriate for them.

That said, the number I'm planning for includes about 20% in travel, which is complete budget slack. And we once lived happily as poor grad students and can do so again as a modest middle-class; the number planned, based on a scale back of our current lifestyle, is still a life that I never expected for myself, nor my wife.

Like the OP, I recently hit my "original number" (using MIcrosoft Money), about 11 years ahead of plan, although that number assumed Social Security. And the warnings about retiring at a peak are sound; of course, the market could go up another 50-70% as well, given all the usual caveats about market timing.

 

[2B]

"The Number" is an illusion. Hitting it in a peak and then falling back is pretty normal. It's not unusual to have 2 or 3 10% equity drops in a year. A 25% drop should be considered likely every 7 or 8 years. I've seen several ~50% equity drops in my investing lifetime. Still with all this, the equity market averaged around +10% per year for the last 100 years or so.

It all comes down to your spending and other income. I don't think any retiree blindly follows the 4%+ inflation "rule" despite it being widely discussed. If you retired in October 2007, could you survive on what you would need to do to conserve your assets? Most of the retirees I know have hunkered down in light of the 2008 meltdown and the continuing poor yields in the fixed income market.

Being able to adjust means you have areas you can reduce spending. If your "Number" meets this requirement, I wouldn't worry about the little pullback we've just had.

 

[steelyman]

Yes, 2B. I remember hitting my "number" back in 2011, while still working. I remember posting about it and someone pointed out you may hit it again... and again... and...

They were right!

 

[ERD50]

The 5% fear is an issue that is very much personal. I personally find some of those who insist on 99+% FireCalc (or two times FireCalc) to resemble obsessive compulsives, although I have no doubt they are doing what is appropriate for them.

Yes, to each their own and all, but I really can't see 99+ compared to the default 95% being described as anything near 'obsessive compulsive'. Analogy: If you hit the last gas station before crossing a desert, would you shrug off a calculation that said you had enough gas to cross the desert 95% of the time? Would it be 'obsessive compulsive' to concern yourself with a 5% failure rate? I don't think so, I think it would be prudent.

That said, the number I'm planning for includes about 20% in travel, which is complete budget slack. And we once lived happily as poor grad students and can do so again as a modest middle-class; the number planned, based on a scale back of our current lifestyle, is still a life that I never expected for myself, nor my wife.

Here's the issue for me - I don't think the vast majority of those who say 'I'll just cut back spending if things go badly', 'We'll eat hamburger instead of steak', 'We won't eat out so often', etc have run the numbers to understand what that means. 20% or 30% won't get you there, based on historical numbers.

Go to FIRECalc and run the typical $40,000/$1,000,000 30 year profile for 95% success. Then try the spending models. You'll find it takes cuts of more than 50% (about 58% worst case, ~ $17K), and cuts below $25K go on for a decade or more. Not $8,000 (20%).

Alternatively, a 3.57% WR has been historically 100% safe for 30 years. That's a cut in spending of under 11% from a 4% spend rate.

Again, to each their own, but I'd rather have the higher degree of confidence that is associated with spending a straight, inflation adjusted 3.57% from the start, than to start with 4% and have to drop below 2% a few years later. For me that would involve a substantial downsize/move - and that would very likely be a bad time for real estate in a poor economy. Would I ever recover that loss? That would just leave me deeper in the hole.

-ERD50

 

[nash031]

Is there really a rule of thumb for x times annual expenses to arrive at some number?

Some have cited 36x for instance.

I doubt there can be much in terms of this type of thumbrule unless you're also specific as to whether someone has a pension or not.

For me, there's no way I will gun for 36x expenses. Now, what I COULD do is take my pension out of my expenses (creating "expenses after pension applied") and figure out my number based on THAT.

A thumbrule based on my preliminary "number" and "expenses after pension applied", I come up with 26x expenses, so that seems about right.

 

[Katsmeow]

Go to FIRECalc and run the typical $40,000/$1,000,000 30 year profile for 95% success. Then try the spending models. You'll find it takes cuts of more than 50% (about 58% worst case, ~ $17K), and cuts below $25K go on for a decade or more. Not $8,000 (20%).

I use the manual spending model on Firecalc where you can change spending per year. I've found that for my portfolio the difference between 95% and 100% is a difference of about $5000 a year. I don't make any changes for the first 6 years and then in year 7 change it to reduce it $5000 a year. I think that is something very doable. I could get to 100% with fewer cuts if I instead made changes in the first 6 years.