Author Topic: Anyone have numbers for worst 10.20, 30 year S and P past returns? Also % risk?  (Read 5679 times)

andysandp

  • Stubble
  • **
  • Posts: 156
Anyone have the return numbers for the worst 10 year, 20 year, and 30 year S and P historical returns WITH Dividends?

Also anyone have % of Risk for 30 year rolling returns WITH Dividends?  I know past doesn't guarantee future but it's good to see the past numbers.

For example,  Based on past, 30 year rolling return WITH Dividends had:

100% of time at least 4% return not including inflation
90% of time at least 6% return not including inflation
80% of time at least 10% return not including inflation
50% of time least 12% return not including inflation
20% of time at least 15% return not including inflation
5% of time at least 18% return not including inflation

These are not real numbers, but anyone know the real numbers?

Thanks!

« Last Edit: April 06, 2017, 10:56:04 AM by andysandp »

Babybalrog

  • 5 O'Clock Shadow
  • *
  • Posts: 89
  • Location: Raleigh, NC
I remember something from PAUL A. MERRIMAN at Marketwatch about like this. I was looking around and couldnt find the article exactly but this was as close as I could find
http://www.marketwatch.com/story/understanding-performance-the-sp-500-in-2015-02-18
Quote
(Fine print: These figures don't take into account inflation, taxes or any management costs.)
However, because not everybody has a 40-year investment horizon, I analyzed all the 15-year periods from 1928 through 2014 (there were 73 of them).

The S&P 500's highest 15-year performance, a compound return of 18.3%, came in 1985 through 1999. The lowest, a barely positive return of just 0.6%, was from 1929 through 1943. Ouch!

About six of every 10 of these 15-year periods produced returns over 10%, and the rest fell below that mark.


From another article
Quote
Longer time periods make more dependable returns. For the Standard & Poor's 500 Index, one-year returns ranged from a high of 54% to a heart-stopping low of minus 43.3%. However, the range for 15-year periods was quite a bit narrower: a high of 18.9% to a barely profitable low of 0.6%.

Among all the 40-year periods, as we saw above, the range of compound returns was fairly narrow, from 15.9% to 10.7%.

http://www.marketwatch.com/story/8-lessons-from-80-years-of-market-history-2014-11-19


For a more reasearch based view
http://www.sciencedirect.com/science/article/pii/S2405473915000331

andysandp

  • Stubble
  • **
  • Posts: 156
I found all rolling 20 year return with Dividends here:

http://allfinancialmatters.com/2015/12/14/sp-500-20-year-rolling-returns/

Notice only one period had less then 4% return out of 70!!

Anyone have 10, 15, and 30 year rolling returns WITH Dividends?
« Last Edit: April 07, 2017, 10:54:48 AM by andysandp »

MDM

  • Senior Mustachian
  • ********
  • Posts: 11477

Maschinist

  • 5 O'Clock Shadow
  • *
  • Posts: 27
Problem is the table is not CPI adjusted.
For example the periods with low/mid sixties starting point are ~8% year without CPI adjustment but in reality where below 2%/year adjusted. The same you would find with starting years 1902-1907

I found all rolling 20 year return with Dividends here:

http://allfinancialmatters.com/2015/12/14/sp-500-20-year-rolling-returns/

Notice only one period had less then 4% return out of 70!!

Anyone have 10, 15, and 30 year rolling returns WITH Dividends?

frugal_c

  • Bristles
  • ***
  • Posts: 300
Absolutely, check out this link:

https://dqydj.com/sp-500-historical-return-calculator/

They have max returns, min returns and 10th thru 90th percentile returns for 10,20,30,40 year periods.  You can also select custom period lengths if you are interested.   The results are inflation adjusted.

MDM

  • Senior Mustachian
  • ********
  • Posts: 11477
Problem is the table is not CPI adjusted.
Yes, that would be a problem for some uses.

Not a problem if comparing S&P returns vs. some fixed interest rate....

maizefolk

  • Walrus Stache
  • *******
  • Posts: 7400
Here. Based on the shiller dataset for all possible start months with dividends reinvested.


andysandp

  • Stubble
  • **
  • Posts: 156
Absolutely, check out this link:

https://dqydj.com/sp-500-historical-return-calculator/

They have max returns, min returns and 10th thru 90th percentile returns for 10,20,30,40 year periods.  You can also select custom period lengths if you are interested.   The results are inflation adjusted.

Wow thanks for this Calculator!  There seems to be some discrepancies

According to the All Financial Matters Chart for 20 year rolling periods, there were only 3 times the return was less then 5.86%.  That means  95.71% the market returned 5.86% or more

Why does your chart only say only 90% of the time the market returned 5.755% or more?
In fact, According to the All Financial Matters Chart, 90% of the time the market returned 6.87% or more!

According to the All Financial Matter Chart, the worst 20 year return gave 2.86% return.  Your chart said the worst 20 year return was 2.048%.

Any thoughts?

http://allfinancialmatters.com/2015/12/14/sp-500-20-year-rolling-returns/

https://dqydj.com/sp-500-historical-return-calculator/




« Last Edit: April 08, 2017, 09:01:26 AM by andysandp »

maizefolk

  • Walrus Stache
  • *******
  • Posts: 7400
Well for starters, the dqydj calculator is using data from 1871-present and AFM is using data from 1926-present.

Likely the datasets are a bit different too even when the years line up (using different baskets of stocks). Similarly, there may be different measures of inflation employed.

Other things that can go wrong:
1) CAGR vs "average annual return" the first is correct, the second is misleading.
2) Dividend and price data are often stored as percentages. Do you multiply by the price increase for the year and then by the dividend yield to estimate dividend payments for the year? Or use the starting price for the year * the dividend yield? Or the average of the starting and ending price * the dividend yield? Little things like this can create small but systematic differences even when the underlying data is identical.

frugal_c

  • Bristles
  • ***
  • Posts: 300
That calculator is not mine, so I can only guess at how it works.  I have been able to verify a few of the data point so I think it is correct.  I am quite confident it is calculating CAGR, not some kind of average.

Some other things, I think it is calculating using every months data available. Maybe the other calculator uses data at start of year.  As maizeman identified it is using data back to 1871 which makes a difference.  There were some brutal periods in there, I think 1906-1921 had negative real returns for instance.

If you want to check their numbers go to their standard return calculator, if you look at some of the brutal start periods, 1929 or 1906 you should be able to find some of the numbers you see.

Also, the worst 20 year period it has is -.2%.  I don't know exactly when that was but I found a 19 year period from august 1901 to august 1920 that had -0.1% (yes negative one tenth of a percent CAGR).
« Last Edit: April 08, 2017, 09:34:41 AM by frugal_c »

Tyler

  • Handlebar Stache
  • *****
  • Posts: 1198
Try this: https://www.crestmontresearch.com/stock-matrix-options/

Note that there are versions with and without inflation. I personally find the real returns much more valuable.

 

Wow, a phone plan for fifteen bucks!