How to use online stock screening tools to help you in your search for undervalued stocks.
The Stock Screener
In the first part of this “How To Find Winning Stocks” series, I covered my favourite method of uncovering new investment ideas: the brute force method. I start with a list of all the stocks on the Canadian markets, weed out the resource production and exploration companies and then focus in on specific market cap segments. Doing this, I can generate a starting list of a few hundred names that I then go through one by one, using online sites like Morningstar’s to get a quick bird’s eye view of the company, then downloading the most recent financial reports from SEDAR (also conveniently available through links on the Morningstar page) to confirm the numbers and dig deeper into the more promising names.
While this works well for the relatively small Canadian market, a similar approach would be a much more herculean task in the much larger US market where you are faced with thousands of prospective companies to choose from instead of hundreds. For this market, and for those who don’t have hours to spend going through a comprehensive list of Canadian stocks, there are a wide variety of stock screening tools available on the internet that let you narrow down your search.
It’s fun to play around with various stock screeners. They all have their pros and cons. Some offer a comprehensive list of criteria to choose from but present the results in an awkward and cumbersome manner. Others have their best features hidden behind a paywall. Some offer some of the criteria you’re looking for but not others. Some let you specify exact cutoff points, others provide you only with broad pre-specified ranges to work with.
I’ve yet to find the perfect stock screener, especially since I’m unwilling to pay for any of them, but two that I keep coming back to are the screeners available at Finviz.com and the one you can find buried deep within the Globe and Mail’s website. Many discount brokerages also come with their own built-in screeners, so it is worth checking your broker’s website to see what they offer, if you haven’t already done so.
It’s easy to get carried away with stock screeners. Some of them offer a bewilderingly comprehensive set of data points that you can use to sift and sort the stock universe. You could ask a stock screener to screen for stocks that are trading near their 52 week highs or their 52 week lows. You could look for companies selling at low price to book ratios or low p/e ratios or that had above average growth rates. You could identify stocks with high returns on equity or maybe you’d want to use return on assets instead. You could screen for excessive debt levels using a bevy of different indicators. The possibilities are quite literally endless. And it is very easy to get lost in the weeds with this stuff.
After playing around with these screeners many times over the years, I have come to realize that the best approach is often the simplest. Don’t try to use a screener to actually pick your stocks for you. They can’t do that. The data the screen is based on is often flawed or only tells part of the story. For example, the big sale of a division might boost the earnings at a company one year, resulting in a seemingly low p/e and a high growth rate when you compare the artificially elevated current earnings to the more normal earnings of years previous. But the next year when that sale has worked its way through the system, earnings will fall back to earth and the company won’t look like such a bargain after all.
A screening tool is not mean to replace the careful, individual analysis of a company. You still need to roll up your sleeves and download the financial reports to actually make an intelligent decision about a stock. What the screeners are good for is generating a manageable list of ideas that you can work your way through one by one, looking for undervalued opportunities.
You might want to generate a list of stocks trading at less than their book value. Here, you’d be looking for potential asset plays or turnaround opportunities. Or you might want to go looking for high-flying growth stocks, in which case you’d set some basic parameters for 1, 3 or 5 year sales or earnings growth. Perhaps you’re interested in adding some dividend paying stocks to your portfolio in which case you’d use a dividend yield criteria. The goal should be to generate a list of a few dozen to a few hundred stocks which you can then go through one by one, looking for mispriced gems.
Using a Stock Screener to Explore the US Markets
Recently, I decided to take a look at the US market. I was curious to see how Trump’s tax reform package was affecting corporate earnings. The changes that were enacted at the end of last year were quite significant, dropping the official tax rate from 35% all the way down to 21%. I wanted to see if this fairly massive windfall had produced any potential bargains south of the border.
There are close to 10 000 stocks listed on the US markets. While this seems like a truly overwhelming number to deal with, as with the Canadian markets which technically have over 3000 listings, once you start screening and sorting, the list of investible candidates is more modest. In the US markets, though, it still runs into the thousands. I wanted a list of a couple of hundred at most that I could work my way through, so I headed over to the screening tool at Finviz.com to get a more manageable list to work with.
If you fire up a new screen on the Finviz site, you’ll see a row of tabs near the top labelled “descriptive”, “fundamental”, “technical” and “all”. If you click the “all” tab, it will display a smorgasbord of all the criteria that are available to screen for with this screener. There are a lot of potentially interesting things I could screen for here. Sales growth, profit margins and insider transactions could all be interesting to look at. I wanted a broad screen, though, something that would generate the greatest number of stocks that potentially fell within my target areas of interest. My core investment philosophy has always centered around low p/e, low debt stocks, so these were the two main criteria I set.
Before selecting any screening criteria, the Finviz website tells me there are 7417 stocks in its database. (The Finviz site deals only with US listed stocks. For Canadian stocks, check out the Globe and Mail screening tool.) I decided to limit my search to those companies that were not only listed in the US but also domiciled there so I set the “country” criteria to “USA”. This got rid of almost 1000 stocks. Next, I could have set various market cap cutoffs to narrow my search, but I wanted to get a sense of the overall market, so I left the market cap selector untouched.
Instead, I set a low p/e cutoff and a low debt cutoff. By combining these two criteria I thought I could snare a good cross section of potentially undervalued companies. I set the p/e ratio to “under 15” and boom! Suddenly the list of matching stocks dropped from the thousands to only 675 potential candidates. (When limiting your universe to only those stocks trading at a p/e of less than 15 eliminates 90% of stocks you know the market is getting frothy!) By setting the p/e ratio to below 15, I not only limited my search to companies with low p/e ratios, but the screening tool also automatically weeded out all the companies with negative p/e ratios (ie those that are losing money). At any given point in time, I’ll often find that about half the companies in a market are unprofitable. There can certainly be good turnaround opportunities in this group but it takes a lot more effort to find them. For this review, I decided to focus only on the profitable companies and only on those that were selling for a reasonably low price relative to those earnings.
I could have set the p/e cutoff even lower at, say “under 10” but I didn’t. The earnings that the screening tools rely on to generate their p/e ratios are typically the earnings as reported by the company. As you’ll find, these are often not necessarily the earnings you want to look at. They’ll frequently include one time items like currency exchange gains and tax rebates that will temporarily boost or lower reported profits. If I set my p/e cutoff too low, I could miss a lot of promising possibilities and would likely end up with a list of companies whose earnings have simply been inflated by one time gains.
But 675 stocks is still more than I really wanted to wade through, so I set a low debt criteria as well. I set a LT Debt/ Equity criteria of under 0.5. There is no option for a debt to earnings ratio so I chose to compare a company’s debt to its shareholder’s equity or book value instead. By eliminating the companies whose debt is more than half of their book value, I managed to eliminate the more heavily leveraged stocks in the list and cut my list in half again to a more comfortable 310 names.
From here, I cut and pasted the results into an excel spreadsheet and then did some more fiddling, eliminating those without enough trading volume to buy (trading volume is one of the pieces of information that is helpfully given in the results section) and using the “sector” category also found in the results section to eliminate industries like resources (ie “basic materials”) and financials that I wasn’t that interested in. This narrowed the list even further to less than 200 stocks that I could review in more depth.
Using the Morningstar website, I took a look at each of these in the same way that I did with Johnson Controls in the second part of this series and then used EDGAR to download the financials for any that looked like they were worthy of further investigation.
As a general rule of thumb, I often find that out of every list of 100 stocks, I’ll typically find one (maybe two if I’m lucky) that I feel are worth investing in. My experience with this group was similar. In the end, I only found one that I was sufficiently enthused enough about to add to my portfolio. You can read about it here in my spotlight on Big Lots. But I did also come up with a list of 10 “runners up” that were very interesting but not quite cheap enough at the moment to draw me in. You can read about those stocks here.
A Stock Screener Can’t Do Your Work For You
In short, a stock screener can be a valuable tool in your hunt for undervalued stocks but be careful to use it in the right way. Don’t try to get it to do your stock analysis for you. For example, a company’s long term expected rate of growth is a very important variable to assess when valuing a company, but a stock screener can’t tell you what that rate of growth is. You can screen for stocks that have grown their sales by 15% or more over the last 3 years but that is just a starting point, not the end of the conversation. Were sales unusually depressed at the start of that 3 year run because the industry the company was in, was in a slump? Was there a big contract with a Fortune 500 company that boosted sales in the last few quarters? Were sales pumped up by a big acquisition that ended up hurting profits even though it may have goosed the top line? A screen can be a great tool for generating a useful list of stocks to look at, but from there, it is up to you to do the heavy lifting!
Full disclosure: I own shares in Big Lots.