Bond ETF Commentary

Model Portfolio Monthly Report. 07/31/08

I have updated my US model portfolio spreadsheets to include dividends and naturally that has increased the CAGR's for all the models. Whether or not these models capture returns in the future as they have since the beginning of the test period (08/02/06) is anyone's guess but I would be surprised if they did perform as well. Given the performance of the QLD / TIP model and its associated UI and maximum drawdown, my US account is currently fully invested in TIP.

 

 

Jason Zweig on Financials

If you are wondering whether Benjamin Graham would buy financials right now, watch the interview with Jason Zweig to get his opinion.

Comprehensive Summary of Model Portfolios. 07/27/08

In the tables below I have included the Maximum Drawdown and the Worst Signal-to-Signal Performance for each portfolio.  In the Canadian table, I have changed the benchmark from the XIU ETF to the index itself (i.e. S&P/TSX 60).

 

 

For most people, the Maximum Drawdown numbers are going to jump out as too high to stomach.  However, bear in mind that the most money lost is that presented in the Worst Signal-To-Signal column.  For example, had you invested as per the QLD/TIP model starting on 08/02/06, the most money you would have lost from the time you purchased QID or TIPS to when you sold the ETF is 5%.  However, QLD, being a double exposure ETF, can shoot up in price exponentially and then come back down.  The Maximum Drawdown is a measure of the decline in portfolio value from a peak to the next trough.  So if you buy $100,000 of QLD and it increases in value to $150,000 but then drops back to $110,000, the portfolio would have incurred a 26.7% drawdown (i.e. the drop from $150,000 to $100,000) even though it is up 10% from the initial purchase.

As they say, an equity curve is worth a thousand words!  Below are the equity curves for all the model portfolios.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

For anyone considering following these models please bear in mind that you have to be mentally prepared for the high fluctuations in the price of double exposure ETF's (in this case QID, QLD and HXU). The wild price swings can be difficult to live with and therefore these models are certainly not for all investors.

As always, past performance is no guarantee of future performance!

Updated Tables

I have been wondering how well my model portfolios would work if they went into bond ETF's when the timers indicate that stocks should no longer be held. I have therefore added a line to each of my tables to include a double exposure ETF when my timer is long and a bond ETF when it isn't. For the US market I chose TIP (iShares Lehman TIPS Fund) and for the Canadian market I chose XSB (iShares Canadian Short Bond Index Fund).



I'll include the Ulcer Index with the next weekly update.

Timer and ETF Strategy Update. 07/25/08

And so ends another volatile week. My timers are still cash/short. I have attempted to improve on my performance by jumping ahead of the timers. On the US side, I bought QLD on 07/21/08 and now have a miniscule 1% gain. On the Canadian side, I bought HXU on 07/14/08 and now have a 6% loss.

Timer and ETF Strategy Update. 07/18/08

My Canadian and US timers remain short/cash. I am expecting at least a short rally in both markets. On the Canadian side, I have take a small position in HXU and I'll take a position in QLD when I think the US rally has started.

Timer and ETF Strategy Update. 07/11/08

Both my Canadian and US timers remain short/cash. The US strategies maintained their improvement in CAGR versus buy-and-hold on average. The Canadian strategy improved its outperformance over a buy-and-hold strategy.

Starting with this post, I am including the Ulcer Index. Given the high UI of the QLD/QID strategy, I don't see much reason to use it unless you believe that the inverse relationship between the US stock market (as far as swing trading is concerned) and gold isn't going to continue. If you are comfortable with the idea that the price of gold will move inversely in relationship to the S&P500, then the QLD/IAU strategy may provide excellent returns with less volatility. I am more comfortable with the QLD/RWM strategy and for that reason my current US holdings are 87% RWM and 13% IAU.

As with all model portfolios, bear in mind that past performance is absolutely no guarantee of future performance!

Timer and ETF Strategy Update. 07/04/08

As both the Canadian and US markets fell this week while both my timers were short/cash, the ETF strategies improved their performance relative to a buy-and-hold- the-market strategy.