Thursday 10 June 2010

Cap-Weight vs Fundamental: Live Realistic Portfolio Showdown

In recent months I've converted my passive index portfolio strategy from one based on capitalization-weighted ETFs to fundamentally-weighted ETFs in the expectation that this will pay off with higher returns and lower volatility. The theory and the back-tested data notwithstanding, the proof is in the pudding so I've constructed two parallel portfolios which are permanently posted at the bottom of this blog. I'll hopefully see how my new portfolio fares in comparison to what might have been - it's either public embarrassment or triumph that is in store for me down the road.

Since pudding is something you can actually eat, the portfolio will be as realistic as possible, what an actual investor will experience, as opposed to so-called index returns one typically sees in the financial press, which exclude various MERs, commissions, tracking errors, currency exchange fees, taxes etc.

Here is how these portfolios will operate:
  • $100,000 Initial Capital - though most people must gradually build up a portfolio, I've started with a lump sum to invest; to convert my own portfolio I've actually had to pay an extra 7 trading commissions ($70) to sell off the cap-weight ETFs that no longer fit but I have ignored this cost.
  • Trading Commission - $10 each trade, so the initial total value of the portfolio has lost $120 for the 12 trades to establish each portfolio
  • Asset Allocation - both portfolios have the same basic percentages allocated by geography and asset class (see the breakdown in the tab AssetAllocation-ETFs) with one prime difference - the cap-weight portfolio includes Value ETFs for USA Small-Cap equity (VBR) and for Global Developed equity (EFV) following the cap-weight view of the world that one adds Value stocks as a tilt. Meanwhile, the Fundamental portfolio simply includes Smaller company ETFs, according to the fundamental metrics NOT cap-weight, for the same USA and Global geographies. In the Canadian REIT class, I have chosen the brand new BMO ETF (ZRE) which equally weights its holdings, since equal weighting also breaks the over-investment in growth stocks that corrupts cap-weighting. In several asset classes no fundamental ETFs are available so we are restricted to using cap-weight ETFs, like XMD (Canadian Small), RWX (Global REIT) and DJP (Commodities). This asset allocation difference is the essence of the divergent approaches.
  • Real Prices - I used actual market quotes during the day yesterday June 9th as my buy prices. Note how the real investor cannot buy exactly the number of shares to place the exact amount allocated to each asset class. Through the magic of GoogleFinance and Google Docs, I have created a spreadsheet that automatically and continually retrieves current market prices so that a very realistic picture of the portfolios can be seen at any time.
  • Rebalancing - will be reviewed once a year in mid July after semi-annual distributions have been received and rebalanced if holdings are more than 1/4 from their target value e.g. for RWX whose allocation is 2%, that is a 0.5% up or down deviation. Even with a fairly big $100k portfolio, it is not desirable to rebalance too often with too small buy-sell amounts - even 0.5% of the initial $2000 allocation is $500, so a $10 trade is a 2% cost. For 12 annual rebalancing trades or $120, the cost to the $100k portfolio is a 120/100000 = 0.1% extra annual cost. Such seemingly small differences do matter over the long run.
  • Taxes - I am assuming the portfolios are within registered accounts that qualify as retirement accounts under US rules (RRSP, RIF, LRIF, LIRA but not TFSA or RESP) so that there is no 15% withholding tax deducted from distributions received from US ETFs
  • Distributions - I will add cash distributions to the portfolios as they are received. To keep things a bit simpler I will assume that USD cash will remain as USD and not be converted into CAD (thus avoiding the attendant built-in currency exchange fee). This is in keeping with the slow trend by discount brokers (Questrade, RBC and some others do so today) to enable USD to be kept as USD in registered accounts.
  • Foreign Currency - the value in Canadian dollars (CAD) is what counts to me and to most Canadians so the net value of USD-traded ETFs is converted back into CAD automatically through the use of the ETF CurrencyShares Canadian Dollar Trust (FXC), which tracks the value of CAD in USD pretty closely. None of the foreign holdings in either portfolio are hedged since I believe the costs of hedging and the tracking error of hedged ETFs outweigh the benefits in the long run. Conversion of CAD with USD is assumed to cost 0.9% (about what I seem to pay with my broker).
  • DRIP - CRQ, ZRE and ZRB offer automatic free reinvestment of distributions so I will calculate that; for the others, the cash balance will accumulate for a year until rebalancing is done. Since I cannot figure out how much interest the cash would collect - a minimal amount if any these days - I won't include any interest for now but if interest rates start to shoot up, I'll try to do an estimate based on rates I see in my own account.
  • Tracking Through the Months and Years - to get an idea of the relative volatility of the two portfolios (I don't expect too much difference since the fundamental indexers themselves have figured out that there is a high correlation between the ups and downs of the funds ... but we shall see), I'll take a month-end snapshot of the portfolio totals and begin graphing them. In ten years, it should be interesting! (If that seems too long, maybe we can take comfort in the fact that Charles Darwin took twenty years to continue his research before publishing his book after he had developed the theory of natural selection).
Over time, the Fundamental Weight ETFs should all sooner or later establish a lead over the Cap-weight ETFs - shown in Red numbers in the middle column of Fund-vs-CapWt-MktValue spreadsheet - that will eventually be large enough never to be overcome. That is my expectation, hope and prayer! Go Reds, go!


Michael James said...

This sounds like it will be an interesting experiment. My guess is that it will take quite a while to get statistically significant results, but that's no reason not to start.

Jordan said...

Have you considered using WisdomTree International Real Estate Fund (DRW) instead of RWX for your fundamental protfolio because this ETF also claims to be "fundamental" but instead of using 4 factors WisdomTree/Siegel bases it's allocations on dividends, which would still break the cap-weighted over investment.

CanadianInvestor said...

Jordan, had not been aware of DRW, thanks for the tip will have a look at it.

Wikinvest Wire

Economic Calendar

 Powered by Forex Pros - The Forex Trading Portal.