See more episodes of Quant Concepts here
Emily Halverson-Duncan: Welcome to Quant Concepts, virtual office edition. Investing in growth companies can be a great way to increase your investments over the long term. While growth stocks can have their associated risks, they can also offer a lot of attractive upside in bull markets. While these types of stocks may not be suitable for a short-term investor, for someone with a long-time horizon, they could be quite valuable. Today we're going to look at a Canadian model searching for growth stocks in the CPMS 250 with the hopes of producing excess returns across the backtest timeframe. So, let's take a look at how to build that.
Jumping into our CPMS here. As mentioned, our universe for this model is the CPMS 250, which in other words is the largest 250 stocks by the market flow. At our first step we're going to go ahead and rank those 250 stocks and some of the factors we're going to look at here are both forward and trailing reinvestment rate, which is a measure of profitability for a company, and we want high values for both of those. Three-month earnings per share estimate revision, which looks at a company's earnings per share estimates, and how they've been revised across the last three months. So, if they were revised upward, that's an indication we'd expect earnings to increase, and again, high values are best for that. And then lastly, quarterly earnings surprise and quarterly earnings momentum. Quarterly earnings surprise is looking at a company's reported earnings and whether or not they beat or missed expectations and quarterly sales momentum is looking at their sales growth quarter-over-quarter.
On the screening side, once we've applied all of those ranking filters, we're going to apply a few screens. Some of the ones that we have here are three-month earnings per share estimate revision, we want that value to be greater than or equal to 0, so we want it to be at least the same estimate that we got three months ago or higher. Quarterly earnings surprise, we also want that to be greater than or equal to 0, so how a company meets or beat their earnings. And then lastly a couple of screens on trading volume as well as market flow just to make sure companies are liquid enough when you're purchasing them.
On the sell side, we're going to sell companies if their three-month earnings per share estimate revision falls below negative 10%, or if the quarterly earnings surprise metric falls below negative 2%. It's also worth noting that we've got a screen on the industry side that we can have no more than five stocks per economic sector. So now that these have all been applied, we can see how our back tested across the long term.
Looking at our back test here, we ran a back test from December 1991 to October 2020. The benchmark we used was the S&P/TSX Composite. And across that timeframe the model returned 15.1% annualized, which is an outperformance of about 7.2% across that timeframe. Turnover is a little higher than some of the models we've looked at about 96%. Turnover is looking at how often you're trading stocks for a model, so at 96% on a 20-stock portfolio you'd be placing roughly 20 trades per year.
A couple of metrics I always like to look at downside deviation, which is the volatility of negative returns. For the strategy is 11.6% compared to the benchmark at 10%, so it is a little bit higher, so you can see there is some additional risks associated with investing in growth stocks. And then also the green and blue chart here, which looks at how the model did in both up and down markets. In up markets, the model outperformed 63% of the time and in down markets outperformed 68% of the time. So still over half the time, it is outperforming in both up and down markets.
So, if you're interested in a little bit of extra growth or you've got an account or you have a long-time horizon, this might be a style that would be worthwhile considering.
For Morningstar, I'm Emily Halverson-Duncan.
