In this article, I will show how I plan to implement stock buybacks into the scores of companies included in my analyses. This variable was pointed out to me in the comments section of one of my previous articles, and I now think it is something that could be very useful to include in this project. Most people on here probably know what a stock buyback is, but I'll include a broad definition of it and explain why I think accounting for them fits the thesis of my research. A stock buyback is essentially a way for a company to invest in itself through purchasing its own shares and removing them from the marketplace. By buying its own shares, the company decreases the number of shares outstanding, usually resulting in a rising earnings per share. Now we come to the reason why I believe accounting for this will strengthen my research.
Let's say, for example, there is a technology stock with a price per share of $40.00 and earnings per share of $2.00. Dividing EPS by price per share leaves us with a P/E ratio of 20 (40 / 2). Now, assume this tech company initiates a heavy share buyback program. By taking some of the shares out of circulation, EPS would increase (considering earnings didn't decrease), since the denominator in the ratio decreased. For the sake of simplicity, let's assume the share buyback program decreases the share count enough to make the new EPS $2.50. By recalculating price per share using this number and our original P/E ratio of 20, we find the price per share increases to $50 (2.5 x 20).
The purpose of that exercise was to display how share repurchases can contribute to stock price appreciation. Therefore, all else equal, companies that buy back more shares than others will see a greater appreciation in their stock price. Each company, if it even has a buyback strategy, repurchases different amounts of shares than others. So, if I account for the magnitude of share buybacks in my research, I think I'll be able to more accurately identify stocks set to outperform their competitors in the long run.
It is worth noting that buybacks don't have as strong of an effect on price as my example due to the fact that there are so many shares on the market. A 50 cent increase in EPS in one year solely attributed to share repurchases would cost most companies their entire cash flow.
It is worth noting that all data about share buybacks was gathered using E*Trade. Now I'll get in to explaining how I'll be implementing this concept into my research.
The Variable of Choice
There are a couple different ways to look at a company's share buyback strategy. I plan on using a pretty simple methodology to capture repurchases, and that's looking at the balance sheet and identifying the number of common shares outstanding. A company deciding to finance some of its operations through equity is one example of what could make this number increase. On the other side of the equation, shares outstanding would decrease if a company bought back some of its shares.
Analyzing the trend of common shares outstanding year over year allows an investor to see the net effect of a company's decisions. If a company implemented stock buybacks but issued more shares than it repurchased, the total number of shares outstanding would still increase. The opposite would also hold true - a company that bought back stock at a greater clip than it issued would see its outstanding shares decrease.
This decrease is what I want to focus on, because when it occurs, it directly causes EPS to increase (unless earnings take a hit). As mentioned in the introduction, this bump in EPS leads to stock price appreciation on most occasions. Hence, I believe it's an efficient variable to account for, since the goal of my research is to identify stocks that possess the potential for long-term price appreciation in relation to some of their competitors.
The Method and Example
The first step in my method is to find the percentage of shares outstanding that the company added or removed since the previous year. If I see a company retiring a higher percent of its total shares than another, it leads me to believe that company has a more aggressive buyback strategy, thus creating an advantage for price appreciation. Here's the data for this metric from my most recent analysis: the airline industry.
|Common Shares Outstanding '17 (million)||Common Shares Outstanding '18 (million)||% Change in Shares|
|Alaskan Air (ALK)||123||123||0.00%|
In its most previous complete fiscal year, Southwest retired a greater percentage of its total common outstanding shares. Therefore, it will see the most improvement in its score after adjusting for buybacks. And Alaska Air's total shares outstanding did not change at all, meaning the company will benefit the least when accommodating for share repurchases.
Initially, I was just going to convert these percentages to decimals and then subtract them from each company's score, but after thinking about it for a few days, I decided that this would not properly show the strength of some buyback programs versus some of the weaker ones. For example, Southwest retired nearly twice the percentage of shares as American did, but the net effect of its score in relation to American would only be about .03 - an incremental improvement that I feel doesn't capture the significant difference in buyback strategies for each company. So, the actual weighting that is to be implemented will look to encapsulate these different strategies as well as accurately represent these strategies through its impact on all of the stocks' scores.
No matter what the percentages of shares retired are, the possible effect on a company's score will be standardized to be between .10 and -.10. The company with the lowest percentage of shares retired will always see its cumulative score worsen by .10. The top company will see its score being enhanced by .10. The effect seen on the scores of the other 3 companies in each analysis will depend on their relations to each extremum. Here is how it looks with the airline stocks:
|% of common shares retired||Standardized effect on scores||Original Score||Adjusted Score|
Using this method, I feel a solid middle was struck, neither overstating or understating the effect buybacks have on the final score. For those wondering where I got these "standardized" numbers, here was my formula, using American as the example:
(.2 x 3.15) / 6.11 = .103
Next, to find the net effect on American's score, we must add .103 to -.1, which equals .003. Here we can see since American's rate of share reduction is close to the halfway point between 0 (Alaska Air) and 6.11 (Southwest), the effect is practically nothing (since we are standardizing between the values of -.1 and .1, which means the halfway point would have an effect of 0).
The effects seen on the final scores in this article are unique to the airline analysis - meaning they will vary across groups of stocks as rates of common shares outstanding reduction will also vary. In my industry analyses, this method will be added on to the end and be calculated once the final score is weight-adjusted.
Going forward, I'm going to be hesitant to try to account for more variables. Now, I think there is a solid base, and if too many factors are added into my project, I think there will be too many moving parts. With that being said, if I do come across something I feel is essential to account for, I will try to add it quickly as possible in order to maximize the time that all the data is unified and measured in the same way.
This method to account for buybacks is by no means perfect. It assumes that previous year's buyback rate will be similar in the current year, which isn't always the case. However, there is no sure way to know ahead of time how many shares a company will retire, so I settled on my current method. As always, feel free to voice any insights or concerns in the comments. Every opinion is much appreciated!
Disclosure: I am/we are long DAL. I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.