Author Topic: Early Retirement Calculator Overstates Returns?  (Read 3899 times)

Realist35

  • 5 O'Clock Shadow
  • *
  • Posts: 33
Early Retirement Calculator Overstates Returns?
« on: July 14, 2017, 02:05:23 AM »
Hi guys,

In the attached spreadsheet I've tried to simulate my expected real life scenario and years to retirement, motivated by the early retirement calculator on https://networthify.com/calculator/earlyretirement?income=78000&initialBalance=100000&expenses=35100&annualPct=5&withdrawalRate=4.

I created my own compounding calculator (attached spreadsheet) to simulate the early retirement calculator. In column F I have calculated the returns after inflation using the real return formula I found at  http://financeformulas.net/Real_Rate_of_Return.html (could be approximated with just the real return in the early retirement calculator, that is nominal return - inflation = 5%). In column E I have simply adjusted the nominal returns obtained in column C for inflation using the formula found here http://www.buyupside.com/calculators/inflationjan08.htm. Columns E and F should be similar but they differ a lot. Any clue why? I'm interested in this because the early retirement calculator in my opinion over estimates the returns.

Thanks a lot:).

Monkey Uncle

  • Handlebar Stache
  • *****
  • Posts: 1740
  • Location: West-by-god-Virginia
Re: Early Retirement Calculator Overstates Returns?
« Reply #1 on: July 14, 2017, 04:40:44 AM »
In my opinion, calculators that apply a steady return of x% every year aren't worth much.  The real world doesn't work that way - you might make 22% one year and lose 15% the next.  And whatever you plug in for x% is just a guess.  You could get lucky and do most of your accumulating during a period that resembles the 1990s, or you could get unlucky and do your accumulating during a period that resembles the 1970s.

I'd put more faith in something that models real-world scenarios using historical data.  Although cFiresim was made for modeling draw-down scenarios instead of accumulation, I think you should be able to get some useful information out of it by plugging in a zero annual spend.  You could then evaluate the median and average amount of assets at the end of a simulation of a given length.  The range of values (high and low extremes) would probably be even more useful than the central tendency measures, as the extremes would tell you your range of likely outcomes.  You could easily run multiple simulations of different lengths to give you an idea of how long it would take to reach your FIRE number.  Again, you want to be looking at the range of outcomes.  No calculator can tell you with any confidence that you are going to hit your number in exactly n number of years.

Realist35

  • 5 O'Clock Shadow
  • *
  • Posts: 33
Re: Early Retirement Calculator Overstates Returns?
« Reply #2 on: July 14, 2017, 06:14:24 AM »
Thanks mate!

I just had a quick look at the calculator. Geez, it's pretty confusing:). I can't see the option of annual contributions to my portfolio. For example, I'd like to retire in 15 yrs, starting on 100k portfolio and buying Xk worth of shares every year. How do I find what the X would be if I assume the average annual total return will be 7%?

Cheers:).

John123

  • 5 O'Clock Shadow
  • *
  • Posts: 45
Re: Early Retirement Calculator Overstates Returns?
« Reply #3 on: July 14, 2017, 06:41:03 AM »
Thanks mate!

I just had a quick look at the calculator. Geez, it's pretty confusing:). I can't see the option of annual contributions to my portfolio. For example, I'd like to retire in 15 yrs, starting on 100k portfolio and buying Xk worth of shares every year. How do I find what the X would be if I assume the average annual total return will be 7%?

Cheers:).

Add it in the "Other Income" section.

MDM

  • Senior Mustachian
  • ********
  • Posts: 11477
Re: Early Retirement Calculator Overstates Returns?
« Reply #4 on: July 14, 2017, 10:55:11 AM »
See Getting Confused Trying to Calculate Savings Rate and links in that post for some math you could use to check your spreadsheet.

DarrellEW

  • 5 O'Clock Shadow
  • *
  • Posts: 14
Re: Early Retirement Calculator Overstates Returns?
« Reply #5 on: July 14, 2017, 12:47:41 PM »
Hi guys,

In the attached spreadsheet I've tried to simulate my expected real life scenario and years to retirement, motivated by the early retirement calculator on https://networthify.com/calculator/earlyretirement?income=78000&initialBalance=100000&expenses=35100&annualPct=5&withdrawalRate=4.

I created my own compounding calculator (attached spreadsheet) to simulate the early retirement calculator. In column F I have calculated the returns after inflation using the real return formula I found at  http://financeformulas.net/Real_Rate_of_Return.html (could be approximated with just the real return in the early retirement calculator, that is nominal return - inflation = 5%). In column E I have simply adjusted the nominal returns obtained in column C for inflation using the formula found here http://www.buyupside.com/calculators/inflationjan08.htm. Columns E and F should be similar but they differ a lot. Any clue why? I'm interested in this because the early retirement calculator in my opinion over estimates the returns.

Thanks a lot:).

Your Column F is missing an inflation adjustment for that year's savings. For example, cell F8 should be calculated as "=D8/(1+$D$3)^B8+F7*(1+$D$4)". With this adjustment, Columns E and F are equivalent.

Monkey Uncle

  • Handlebar Stache
  • *****
  • Posts: 1740
  • Location: West-by-god-Virginia
Re: Early Retirement Calculator Overstates Returns?
« Reply #6 on: July 14, 2017, 07:10:46 PM »
Thanks mate!

I just had a quick look at the calculator. Geez, it's pretty confusing:). I can't see the option of annual contributions to my portfolio. For example, I'd like to retire in 15 yrs, starting on 100k portfolio and buying Xk worth of shares every year. How do I find what the X would be if I assume the average annual total return will be 7%?

Cheers:).

The beauty of cFiresim is that you don't have to assume an average annual return.  It runs simulations using actual historical returns.  So you would not be plugging 7% in anywhere.

If you are saving X dollars every year and adding that to your portfolio, do as John123 said and plug that into the "other income" section.

Here's an example:

Since you are running an accumulation scenario starting right now, leave the "retirement year" set at 2017.  You want to accumulate for 15 years, so set the "retirement end year" to 2032.

Leave "data to use" set at "historical data - all."

You are starting with $100,000, so set "portfolio value" to 100,000.

What is your asset allocation?  Since I don't know how your assets are invested, I'm going to leave this at the default 75% equities and 25% bonds, and also leave the "fees/drag" and "growth of cash" set to the defaults.

Most people rebalance annually so their asset allocation doesn't get out of whack, so we'll leave that set to the default, and we'll also default to a constant asset allocation. 

Set "initial yearly spending" to zero.  Doesn't matter if you leave the spending plan on "inflation adjusted" or not, because x% times zero will still be zero.

Skip the "social security" and "pensions" sections.

In the "other income" section, enter the amount that you plan to save each year and add to your portfolio.  I don't know what that number is, so I'm going to enter a guess of $35,000.  You should enter whatever you think you will save each year.  Leave recurring set to "true," and put in 2017 for the start year and 2032 for the end year.  I'm going to assume that your savings will increase with inflation, so I'll leave the inflation adjustment set to "true."  For simplicity, we'll leave the inflation adjustment set to "CPI," to avoid having to make a guess about what future inflation might be.

If you are planning on increasing your savings beyond the inflation rate, you would shorten the first savings row to however long you might save at that rate, then enter another savings row with the increased amount below it.  Repeat as necessary.  For simplicity, and because I don't know how much you are actually planning to save, I'm not going to do that.  I'll just assume you plan to save $35k per year, increasing with inflation, for the entire 15 year period.

Leave the "extra spending" section blank.

Hit "run simulation" and watch the output pop up.

Ignore the graphs and the success rate table.  Those apply to draw-down scenarios, but you're running an accumulation scenario.

Here's the meat of what you're looking for:  the average and median ending portfolio values are about $1.2 million.  The highest ending value is about $2.4 million, and the lowest ending value is about $0.5 million.

Ignore the rest of the tables on the output - they apply only to draw-down scenarios.

What does this mean?  It means that if you start with $100k, add $35k per year and adjust that amount to account for inflation every year, keep your portfolio invested 75% in US stocks and 25% in US bonds, with fees comparable to an index fund, and stock market/bond market returns are no better and no worse than they were in the past, then after 15 years you will have somewhere between $500,000 and $2.4 million dollars.

Any calculator that gives you a more precise answer than that is making unwarranted assumptions about the rate of return that you are likely to get.  No one knows if future returns will be 7%, 4%, 5.6%, or whatever, and they certainly won't be that amount year after year for 15 years.  The best you can do is assume that future returns will be somewhere within the range of all rates of return that have occurred in the history of the financial markets.  And that's why you can't predict today whether your portfolio 15 years from now will be worth $500,000, $2.4 million, or somewhere in between.

Oh, and it's also possible that something unprecedented could happen in the financial markets, and the actual result won't even be within that range.  But what is the likelihood of that happening?  Hopefully very remote.




Realist35

  • 5 O'Clock Shadow
  • *
  • Posts: 33
Re: Early Retirement Calculator Overstates Returns?
« Reply #7 on: July 14, 2017, 09:21:43 PM »
You guys absolutely rock! Thank you so much for your thoughts.

I spent some more time trying to put my real life scenario into a spreadsheet. Will post it in the case study section just to get your thoughts if possible, in case I stuffed up something:).

Realist35

  • 5 O'Clock Shadow
  • *
  • Posts: 33
Re: Early Retirement Calculator Overstates Returns?
« Reply #8 on: July 14, 2017, 09:47:02 PM »
And here is the link to the case study I just posted.

https://forum.mrmoneymustache.com/case-studies/my-fire-case-study/

Any thoughts would be much appreciated:).

Monkey Uncle

  • Handlebar Stache
  • *****
  • Posts: 1740
  • Location: West-by-god-Virginia
Re: Early Retirement Calculator Overstates Returns?
« Reply #9 on: July 15, 2017, 04:40:06 AM »
Ah, I see from your case study that you're in Australia.  So unless you're planning do all your investing across the pond in the U.S., cFiresim doesn't really work for you.  I don't know if there is a calculator available that does the same kind of historical simulations using Australian or other countries' markets.

 

Wow, a phone plan for fifteen bucks!