Poll:What return do you assume in your model

[Gumby]

Based on a request by Graniac in this thread http://www.early-retirement.org/for...ould-you-be-happy-with-59805.html#post1155274 I thought it might be informative to take a poll to learn what rate of return on investments people use in their models. Please indicate the real return (i.e. above inflation) that you assume in your FIRE model. Because everyone's tax situation is different, it would be most useful for the number to be pre-tax. If you assume a change at some point, it may be helpful to indicate how and why in a comment.

 

[MichaelB]

I assume equities will return 6% and fixed income 1% real return before tax.

 

[Lsbcal]

My simple spreadsheet model has the following:

bond real return = 2.0%
stock real return = 7.5% for 2012
stock real return = 4.5% for 2013 and into future

I assumed higher stock returns this year because last year's was sub par. My return assumptions for stocks are set as variables so I can change them on a dime -- i.e. don't count on my numbers.

 

[Midpack]

Cut-n-paste from the other thread since I just answered there...

My plan is built on a real return of 2%. Asking about expected returns or expected inflation alone aren't very useful IMHO. I hope real returns are better and history would suggest they will be.

Another reason I use a low number is sequence of returns, maybe not mentioned here often enough. Once you begin to draw from your nest egg, sequence of returns is a far greater factor than during the accumulation years. The exact same long term real return, with different sequences of returns, can end with wildly different end of plan outcomes. Food for thought...YMMV

And being prepared for bad outcomes is what FIRECALC, 4% SWR, etc. are built on. Odds are the
actual outcome will be better if history holds. Seems like a good plan methodology.

What I plan on and what I expect are not the same. Never easy is it...

 

[Grainiac]

Based on a request by Graniac in this thread http://www.early-retirement.org/for...ould-you-be-happy-with-59805.html#post1155274 I thought it might be informative to take a poll to learn what rate of return on investments people use in their models. Please indicate the real return (i.e. above inflation) that you assume in your FIRE model. Because everyone's tax situation is different, it would be most useful for the number to be pre-tax. If you assume a change at some point, it may be helpful to indicate how and why in a comment.

Wow! You ask and you shall receive. Thanks Gumby.

 

[REWahoo]

Since my model is FIRECalc, my rate of return is that of the worst 30 year period over the past 139 years. Not sure what that is, so I'm not sure how to respond to the poll.

"Hope for the best, plan for the worst"

 

[scrabbler1]

I actually assign in my medium/long-term retirement spreadsheet a separate rate of return for each fund in my portfolio. I have 4 funds in my non-retirement portfolio and 2 in my IRA so it is not too unwieldy. for the purpose of answering the poll, I used as my assumed rate of return what I have assigned for the fund with the most money.

 

[Major Tom]

I don't assume any particular rate of return because, like REWahoo, I use Firecalc as my main predictor of portfolio solvency. Because I'm in the very early stages of retirement/semi-retirement (still not sure which) I use a WR that gives me a 100% success rate when run over several different time periods (I use at least a 30 year and a 40 year time period). I'm 48 years old, don't have a lot of money, and am a bit paranoid (Aaah, what were those noises?!)

However, Grainiac's question asked what rate of return I'd be happy with. Well, I'd be really happy with 15%

You did ask

 

[NW-Bound]

I assumed higher stock returns this year because last year's was sub par. My return assumptions for stocks are set as variables so I can change them on a dime -- i.e. don't count on my numbers.

Reversion to the mean, heh heh heh... I was about to brag about my return YTD in 2012, but afraid of offending the market god, heh heh heh...

Since my model is FIRECalc, my rate of return is that of the worst 30 year period over the past 139 years. Not sure what that is, so I'm not sure how to respond to the poll.

"Hope for the best, plan for the worst"

OMG! With your (S?)WR, you might have to go back to work when you are in your 80s.

I'd rather work now to build up more buffer while I still can, in my late 50s.

Just joking, of course.

 

[REWahoo]

OMG! With your (S?)WR, you might have to go back to work when you are in your 80s.

I look on the bright side - I'm only 15 years away from being in my 80's so my odds of running out of money before running out of me lessen with each passing year.

I'd rather work now to build up more buffer while I still can, in my late 50s.

Sounds good to me - since I'm not the one working any more.

Just joking, of course.

Of course. Me, too.

 

[FIREd]

I don't assume any particular rate of return either.

Let's say I assume a real rate of return of 5% for my model. Unless there is a way for me to get a guaranteed 5% real (and there is none I can think of at this time), my model will have to be constantly adjusted to reflect actual real returns. So why assume a real rate of return in the first place? I see my plan as a work in progress and we'll see where it takes us.

 

[NW-Bound]

"Hope for the best, plan for the worst"

But of course!

In my case, let's see...

Worst case: Boondocking in the 25' MH, living off SS. MH already paid for, parked in driveway. Checked. SS, let them bring on means testing (what means?). Checked.

Best case: Hmm... This one is a bit tougher. MH has cargo carrying capacity of 2,600 lbs. Subtract out 1,600 lbs for real cargo, that leaves 1,000 lbs to hide Krugerrands. At the price of $1,600/oz, that's $25.6M that would fit! No problem!

 

[Major Tom]

But of course!

In my case, let's see...

Worst case: Boondocking in the 25' MH, living off SS. MH already paid for, parked in driveway. Checked. SS, let them bring on means testing (what means?). Checked.

Best case: Hmm... This one is a bit tougher. MH has cargo carrying capacity of 2600 lbs. Subtract out 1,600 lbs for real cargo, that leaves 1,000 lbs to hide Krugerrands. At the price of $1,600/oz, that's $25.6M that would fit! No problem!

Both the worst and best case scenarios involve an MH. Now that's forward planning

 

[W2R]

I generally assumed 3.5% yield over inflation for very rough projections when I was in the accumulation phase. I don't do projections much if at all any more, though, since it'e not like I'm going to un-retire and since my withdrawal rate is pretty low.

 

[Grainiac]

I look on the bright side - I'm only 15 years away from being in my 80's so my odds of running out of money before running out of me lessen with each passing year.

Good point! One can also reduce the odds of running out of money by drinking lots of beer and gaining 10 lbs per year. I'm using both strategies.

 

[youbet]

Since my model is FIRECalc, my rate of return is that of the worst 30 year period over the past 139 years. Not sure what that is, so I'm not sure how to respond to the poll.

"Hope for the best, plan for the worst"

I agree with your technique! Assuming an annual return is likely to lead to huge discrepancies due to the importance of "sequence of returns."

You might assume 6% equity and 2% fixed, for example, and have that turn out to be an very close estimate for a 30 year period. Yet, you may never have a single year, or you may have very few years, close to those numbers. At best only the average will work out. The difference between having higher returns early on vs. low returns early on is huge even though both situations may result in the same average returns over time.

The same is true for inflation assumptions.

I use tools like FireCalc and Monte Carlo simulators to get a possible range of results with probabilities and accept that the likely outcome over time will be wild fluctuations in my FIRE portfolio performance vs any plans or expectations. It's just the way it is folks. You have to accept variability and limited control to play this game, otherwise you're just kidding yourself.

Through approximately six years of FIRE, the "wild fluctuations" assumption has certainly been right on! The recession lows where I bottomed out down 30% or so were something my models showed as possible but not probable. Still, something I needed to be prepared to deal with. My "simple spreadsheet" model where I used 2% real return was so far off, it was only entertaining to look at. At the recession bottom, the "simple speadsheet" showed me being up from my starting point and I was actually 30% + down.

 

[NW-Bound]

Both the worst and best case scenarios involve an MH. Now that's forward planning

Remember when you took those math classes, how often they said "that reduces the problem to one that has already been solved"?

Seriously, I was suggesting that when one looks at what is truly life-essential, we should feel so lucky to be living in a developed country.

That of course does not keep me from wanting more, but that's because I am scroogey! A guy's got to be honest with himself, ya know?

 

[Midpack]

You might assume 6% equity and 2% fixed, for example, and have that turn out to be an very close estimate for a 30 year period. Yet, you may never have a single year, or you may have very few years, close to those numbers. At best only the average will work out. The difference between having higher returns early on vs. low returns early on is huge even though both situations may result in the same average returns over time.

Agreed, aka 'sequence of returns'

 

[Alan]

I've never planned on a particular ROI, just plugged the numbers in FIRECALC, FIDO Income Planner and Financial Engines and looked at the probabilities of success.

Thinking about it now that I am distribution phase, I would be happy with a real return of 2%, although I recognize that the sequence of the gains and losses is a lot more important than it was during the contribution phase.

 

[Lsbcal]

In my comments above I forgot to include the results of a detailed model that I've used with our investment approach. The results for 1969 to present were very good indeed (1.045 => 4.5% compound annual return):

So in summary after 4% spending would still get a real 4.5% return. We'll see.

 

[ERD50]

Since my model is FIRECalc, my rate of return is that of the worst 30 year period over the past 139 years. Not sure what that is, so I'm not sure how to respond to the poll. ...

I don't assume any particular rate of return because, like REWahoo, I use Firecalc ...

I don't assume any particular rate of return either.
...

I agree with your technique! Assuming an annual return is likely to lead to huge discrepancies due to the importance of "sequence of returns." ...

Agreed, aka 'sequence of returns'

I've never planned on a particular ROI, just plugged the numbers in FIRECALC, ...

Another poll, another incomplete choice of responses

It's one of the reasons I have not done a poll, even with careful consideration, I'd probably screw it up.


But one way to think of this is, if you accept a (say) 3.5% WR from using FIRECALC, you are roughly expecting a 3.5% real return. Yes, roughly, as the portfolio end point will often be significantly higher or lower than the starting point. Ballpark.

-ERD50

 

[FIREd]

Another poll, another incomplete choice of responses

It's one of the reasons I have not done a poll, even with careful consideration, I'd probably screw it up.


-ERD50

Yes, trying to get us to provide a straight answer to a poll is like trying to herd cats. But we don't bite - usually.

 

[youbet]

But one way to think of this is, if you accept a (say) 3.5% WR from using FIRECALC, you are roughly expecting a 3.5% real return. Yes, roughly, as the portfolio end point will often be significantly higher or lower than the starting point. Ballpark.

-ERD50

I see where you're going but a bit of a stretch since the end point will be higher or lower than the starting point in 100%, or near 100%, of the cases. And if you accept a 3.5% WR using FireCalc, you're actually expecting less than a 3.5% real return because with FireCalc surviviability achieved by depleting capital is OK.

I really DON'T assume any average rate of return or compounded rate of return when planning. Those are handy for looking at historic portfolio performance but not very useful for projecting portfolio survivability. The distribution of possible outcomes of WR vs annual returns and inflation rates, whether using historical back-testing (FireCalc) or a Monte Carlo type simulation is the only way to go.

 

[Ken11]

With Firecalc I'm at 95% or 100% success until age 90.

However, I also use a flexible speadsheet that I constructed.
My mainline case assumes 3% inflation and 5.5% nominal returns on a total portfolio.
It assumes 1% inflation on SS- much lower than past experience.

As a 2010 retiree I am most worried about the "sequence of returns" risk in the current environment. Then again, Im usually wrong. :>)

 

[Alan]

As a 2010 retiree I am most worried about the "sequence of returns" risk in the current environment. Then again, Im usually wrong. :>)

Whichever year you retire, a few bad returns during the first few years can have a dramatic affect on your portfolio. I'm also a 2010 retiree and the first 2 years have been quite reasonable. For me, if we can get through the first 7 years without a string of bad returns then I'll feel much safer, as SS is then available as another income stream.