In this series, I (Jin Choi) talk about my goal of reaching $1 million in my TFSA account by 2033. If you want to know what a TFSA is, I recommend you read my free book.
September Results: Down 1.2%
At the end of September, I had $56,630 in my TFSA account, which was down by 1.2% since the start of the month. By comparison, the Canadian stock market went down by 0.9% while the U.S. stock market went down by 0.1% in Canadian dollar terms. Therefore, my portfolio underperformed the broader market.
The majority of my portfolio consists of oil and gas stocks. Oil prices rose from $69.84/bbl to $73.16/bbl (in US dollars) in September. However, my stocks remained flat because of the widening differentials between US and Canadian oil prices. The Western Canadian Select (WCS), which serves as the oil price benchmark for the Canadian oil sands, sunk by $8/bbl in September.
As I’ve explained in my last month’s TFSA update, there’s currently a lack of pipeline capacity to flow all the oil out of Alberta. Some oil producers signed deals with railways to transport their oil, which should help alleviate part of this problem, but train cars are not built overnight and it may take some time for WCS to recover.
I will remain heavily invested in oil stocks for the time being, because I think they are unreasonably cheap. The value of these companies depends to a great extent on what happens 5+ years out, so short term challenges like the ones we’re seeing don’t deter me. But once my oil stocks reach fair value, I plan on selling them and employing a quantitative value investing algorithm instead.
I was able to make some headway into this research in recent months, and I even gave a talk about it at a conference early this month. The strategy of the quantitative value investing algorithm is explained below, along with the slides that were used during my talk.
Research shows that one way to identify cheap stocks is to use the ratio of EBIT (earnings before interest and taxes) to EV (enterprise value = value of a company’s stocks + value of bonds - cash). Companies with high EBIT to EV ratios are considered cheap, and you’d have outperformed the market if you consistently invested in high EBIT/EV companies. But the problem with using the EBIT/EV ratio is that it is backwards looking. In other words, it uses already released financial reports.
It turns out that if you could predict future EBIT and were able to use that as the basis for picking stocks, you would do much better. So I created a machine learning model that does an OK job of predicting EBIT, and showed that picking cheap stocks based on those predictions would have outperformed the traditional backwards looking EBIT/EV strategy.