From our perspective, if one does not wish to actively trade securities and holds to the premise that it is very difficult to consistently beat the market over the long term, it becomes prudent to find broad investment vehicles that one can place money into without having to be concerned with when the money placed in such a vehicle.  Ideally, the vehicle has shown it can provide returns over the long term that, not only are positive on an annualized basis, but also beat inflation on an annualized basis ensuring the purchasing power of the invested money does not degrade.

​

This analysis started by using the ETF Screener at barchart.com (https://www.barchart.com/ca/etfs-funds/etfs-screener) to select their full data set.  At the time (as of May 15, 2025), their were 1,504 listed ETFs in the unfiltered screener results.  Historical, dividend adjusted, price data was then download for each symbol from the unfiltered list.  The data was then sorted based on the number of dividend adjusted end of day prices available in the data set.  Only ETFs with 2,500 or more dividend adjusted end of day prices were selected which narrowed the list of ETFs to consider to 210.

​

The analysis focused on generating histogram and scatter plot data to allow for a comparison of the 2, 3, 4, & 5 year annualized returns from the ETF compared to the 2, 3, 4, & 5 year annualized returns of the CPI index (https://www150.statcan.gc.ca/t1/tbl1/en/tv.action?pid=1810000601) for the sum of the growth in the Food, Shelter and Transportation (FST) subgroup values - this is a measure of inflation that focuses on the things most people _have_ to purchase.  These returns were calculated by ensuring each calendar day of the considered subset had an end of day value (if one did not, the previous value was copied into that day).  The 2, 3, 4, & 5 year annualized returns were calculated for each calendar day going back in time from April 30, 2025 to the respective number years after the ETF inception.  The histogram and scatter plot illustrations are rendered using the python library from seaborn (https://seaborn.pydata.org/) using the year of the start date (the day an investor might have purchased the ETF) of the annualized return as the grouping value.  A box is drawn around all the returns and vertical lines are drawn inside the box showing the statistical mean and first standard deviation for the ETF.  A red line is drawn as reference showing those returns to the right are greater than the respective annualized returns of the CPI FST subgroup sum.  As CPI values are provided once per month, the value was taken to be the last day of the month with a straight line step between the that and the previous month end to give each calendar day and end of day value.

​​

The ETF scatterplot data was then sorted based on the percentage of ETF 5 year annualized returns that beat the CPI FST (inflation) and how far the box was shifted to the right on the scatterplot - the further right the better.  This resulted in 23 ETFs that had their 5 year annualized returns contained in a box fully to the right of the CPI FST reference line.  This sort brought in QQU.TO, Betapro Nasdaq 100 2X Daily Bull ETF, which is significantly more volatile than the rest of the cohort and we have left it in this group to show what a more volatile return set of results looks like (it is also a long running ETF so has a large data set). The returns bounding box was shifted nicely to the right on the 5 year annualized returns scatter plot but those annualized returns have huge swings showing a simple sorting does require further review.

​​

Additionally, you will note that with exception of XWD.TO, XST.TO, HAZ.TO, and VXC.TO, they ETF cohort all provide returns based on the US stock market.  VXC.TO and XWD.TO are also heavily weighted in US stocks.

​

These sorted results are presented below.

​​​