Wednesday, January 7, 2015

Portfolio performance 2009-2014

It's time to compare the performance of our portfolio in the year 2014 against passive index investing.
Our "benchmark investment" is an imaginary passive ETF that closely tracks the performance of MSCI all country world (ACWI) index in euros.

In 2012 we significantly underperformed, in 2013 we significantly overperformed and last year we lost a bit to the benchmark. Overall, we are a bit ahead in terms of performance.

Our performance during 2009-2014 has been as follows:

I have made changes to how "benchmark investment" is calculated and this reflects in performance of all the years (in case you compare to my previous "portfolio performace" articles). Over the years I have also corrected some minor errors in the spreadsheets.

Our portfolio has been more defensive (less risky) and less volatile than the ACWI index. Therefore, it has underperformed in bull market and overperformed in bear market - except for 2013 where our big turnaround bet to Nokia paid off and we also got above market gains from Western Digital and UPM.

During 2014 it looked like we may beat our benchmark by wide margin. Especially our U.S picks from the IT sector delivered above average gains (Western Digital, Intel, Microsoft, Cisco Systems) and overall things were progressing very nicely - until last quarter.

Euro declined from 1,3791 to 1,2141 U.S. dollars (-11,96%) during 2014. Our portfolio is much more tilted to Europe and Eurozone than the ACWI index where U.S dollar exposure is currently 52% (source: XTF). This was likely the most serious headwind for us in 2014. Also, our oil sector positions - especially Fred Olsen Energy - did very badly in the latter part of 2014. Mining sector wasn't doing that good either.

Generally, our european positions - outside of the sectors that did very badly - didn't help the race by producing above average gains - apart from Orion. However, most of our european positions pay out above average dividends compared to what ACWI index funds pay out. Long term that will matter and limits the downside risks.

 Our benchmark index 2009-2014 (based on MSCI All Country World Index; Net; Euros)

I have not calculated our performance during 2007-2008 because we were off the market. During that time and also before that our benchmark index would have been different because the investment objective was different and risk level was much lower.

About the "benchmark investment"

I have chosen to construct my own imaginary benchmark index fund out of MSCI ACWI index instead of choosing one particular index fund that tracks the index. The main reason for this is that I would never invest all our money in any particular fund. I would rather choose several funds managed by several companies that as a whole would track the index close enough. My estimate for average cost level for the benchmark investment is 15EUR + 0.5% per transaction and 0,48% tracking difference per year. The latter is estimated to be 1,6 times Total Expsense Ratio that in turn is estimated to be 0,30%.

Also, I am taking into account paid taxes. The ones that we have paid for real and the ones we would have paid if we would just buy and hold the benchmark investment instead. In Finland, gains of ETFs and stocks are treated the same. There is, however, difference in how payouts are treated. Tax rate for dividends from stocks is lower. That is actually one reason, why we have much more direct investments to stocks than money parked in passive ETFs. Over time, even small differences matter a great deal.

To take into account taxes that would have been paid by us, our benchmark investment gets penalty of 0,60% per year. This is assuming that in reality it would pay out roughly 2% yearly dividend which would be taxed at 30% in Finland before we could invest money back into the fund.

The index data itself is available via MSCI Barra web site as excel-file at least at the time of writing this post. I use a version of ACWI index which has large and mid cap companies in it. I use it with the following parameters:
- EUR (as in euros)
- Daily (as in daily quotations of the index)
- “Net” (as in “With Net Dividends” that takes into account taxes that an insitutional investor or fund would have to pay to various countries where they have holdings)

The benchmark investment is always fully invested into the passive index. Starting balance was invested at end of 2008 to the index. By dividing the money with the value of the index, you get “shares in index”.

About calculating the yearly returns

The yearly return of the index and the "benchmark investment" will be different if more money is added into the brokerage accounts during the year. To take that into account, the yearly returns are calculated as [balance at end of year N] - [(balance at end of year N-1) + (additions to brokerage accounts during the year N)] / [(balance at end of year N-1) + (additions to brokerage accounts during the year N)]

Background materials:

For the changes done this time around, I actually browsed through quite a bit of materials. Best are linked here.

About MSCI index calculation methodology

About tracking difference and "hidden costs" that are not explained by Total Expense Ratio (TER):

No comments:

Post a Comment