I was asked in a comment whether I am mistakenly over-stating Russell 3000 performance - which in turn makes MFI look worse versus benchmark. A fair question as I am just a guy with a spreadsheet and can certainly make mistakes.
Before you go further, this probably is mostly for number geeks. it isn't necessarily fun, though I found it interesting.
Here is my current table comparing Russell 3000 year by year (and inception to date - assuming start in January 2006).
Annual | Inception to Date | |||
Year | Russell | MFI | Russell ITD | MFI ITD |
2006 | 11.4% | 15.0% | 11.4% | 15.0% |
2007 | 4.1% | -6.7% | 16.0% | 7.3% |
2008 | -37.0% | -38.0% | -27.0% | -33.4% |
2009 | 32.5% | 45.2% | -3.3% | -3.3% |
2010 | 18.4% | 22.8% | 14.5% | 18.7% |
2011 | -0.6% | -10.5% | 13.9% | 6.2% |
2012 | 16.4% | 9.7% | 32.6% | 16.6% |
2013 | 33.0% | 51.7% | 76.3% | 76.8% |
2014 | 12.3% | 12.1% | 98.0% | 98.1% |
2015 | 0.4% | -8.9% | 98.7% | 80.4% |
2016 | 12.5% | 13.2% | 123.5% | 104.2% |
2017 | 20.8% | 4.7% | 170.1% | 113.9% |
2018 | 9.5% | 16.1% | 195.7% | 148.2% |
So I have Russell 3000 here up 195.7%.
Going into Yahoo Finance, here is what I get for Russell 3000 (meaning the IWV ETF in this event).
Year | Start | End | Dividends | Change |
2006 | $ 74.58 | $ 82.04 | $ 1.17 | 11.6% |
2007 | $ 82.04 | $ 84.40 | $ 1.40 | 4.6% |
2008 | $ 84.40 | $ 52.00 | $ 1.31 | -36.8% |
2009 | $ 52.00 | $ 65.28 | $ 1.20 | 27.8% |
2010 | $ 65.28 | $ 74.95 | $ 1.19 | 16.6% |
2011 | $ 74.95 | $ 74.18 | $ 1.35 | 0.8% |
2012 | $ 74.18 | $ 84.68 | $ 1.65 | 16.4% |
2013 | $ 84.68 | $ 110.65 | $ 1.78 | 32.8% |
2014 | $ 110.65 | $ 122.29 | $ 1.99 | 12.3% |
2015 | $ 122.29 | $ 120.31 | $ 2.39 | 0.3% |
2016 | $ 120.31 | $ 132.98 | $ 2.38 | 12.5% |
2017 | $ 132.98 | $ 158.17 | $ 2.49 | 20.8% |
2018 | $ 158.17 | $ 171.90 | $ 1.28 | 9.5% |
Interesting. So some modest differences in the past, though past few years look good. On a cumulative basis, the Yahoo table is +187.0%, so a bit worse than my 195.7% in my table I've kept over the years. Am I wrong (mistaken) or is Yahoo history not so good?
The comment was a 30 point differential (as opposed to my 8 point differential) - I don't think I'm off by that much. Perhaps the comment treated dividends a different way or used something different than IWV ETF.
I next went to Fidelity, which I find to be very reliable.
Year | Start | Dividends | End (+Div) | Change | Cumulative |
2006 | 72.35 | 0.82 | 82.86 | 14.5% | 14.5% |
2007 | 82.86 | 1.39 | 85.22 | 2.8% | 17.8% |
2008 | 85.22 | 1.3 | 54.21 | -36.4% | -25.1% |
2009 | 54.21 | 1.19 | 68.79 | 26.9% | -4.9% |
2010 | 68.79 | 1.34 | 79.65 | 15.8% | 10.1% |
2011 | 79.65 | 1.48 | 80.22 | 0.7% | 10.9% |
2012 | 80.22 | 1.65 | 92.37 | 15.1% | 27.7% |
2013 | 92.37 | 1.78 | 120.12 | 30.0% | 66.0% |
2014 | 120.12 | 1.98 | 133.74 | 11.3% | 84.9% |
2015 | 133.74 | 2.35 | 134.11 | 0.3% | 85.4% |
2016 | 134.11 | 2.38 | 149.16 | 11.2% | 106.2% |
2017 | 149.16 | 2.48 | 176.83 | 18.6% | 144.4% |
2018 | 176.83 | 1.28 | 192.94 | 9.1% | 166.7% |
Slight difference in timing as I have January of 2006. And we get a lower cumulative of 166.7%. I think the difference is in how you treat dividends. In this table, the cumulative dividends are purely additive in they add $21.42 to the final price. In my table, I am implicitly assuming you reinvest dividends. I think that is the appropriate approach and it is also the approach I am using on the MFI side. I guess I am surprised it makes a 20 point difference. But over time it compounds.
Let us look at a simple example. Suppose you have a $1,000 investment. It grows 8% a year in price plus you get a 2% dividend per year.
Ok, this simple example explains difference very well (whew). So if you don't reinvest the dividends, then the growth is from 1,000 to 2,518 over 12 years. You would also get $379 in dividends so you total gain (not reinvesting dividends) is +190%. But if you reinvest dividends, this simple example is just a 10% investment over 12 years, which is 214%. So over 12 or 13 years (amount of time I've been in MFI) reinvesting a 2% dividend is worth 20 to 30 points. Voila. Compounding.
Year | Start | Dividend | Cum Div |
2006 | 1,000.00 | 20.00 | 20.00 |
2007 | 1,080.00 | 21.60 | 41.60 |
2008 | 1,166.40 | 23.33 | 64.93 |
2009 | 1,259.71 | 25.19 | 90.12 |
2010 | 1,360.49 | 27.21 | 117.33 |
2011 | 1,469.33 | 29.39 | 146.72 |
2012 | 1,586.87 | 31.74 | 178.46 |
2013 | 1,713.82 | 34.28 | 212.73 |
2014 | 1,850.93 | 37.02 | 249.75 |
2015 | 1,999.00 | 39.98 | 289.73 |
2016 | 2,158.92 | 43.18 | 332.91 |
2017 | 2,331.64 | 46.63 | 379.54 |
2018 | 2,518.17 | ||
Begin | 1,000.00 | ||
End | 2,518.17 | ||
Change | 152% | ||
End + Div | 2,897.71 | ||
Change w Div | 190% | ||
12 Years @ 10% |
214% |
2 comments:
Dear Marsh,
I'm really glad to see you read my comment and made a post about it.
Well for my calculation I used Russell 2000 not 3000 and i used Yahoo Finance.
I did simple math: 1000 $ / 764 (which is Russell 2000 quotation on 05/01/2006) x 947 (1711 (which is the current quotation) - 764) = 1239$ so a gain of 123%.
Correct me if my calculation is wrong...
Fede - that approach does not work as it excludes dividends.
Marshall
Post a Comment