As I get older, I find myself coming around more and more to Warren Buffett’s view that stock splits are not something to view favorably. Everyone knows that the mechanics of a stock split don’t themselves create wealth; they just take alter the number of shares outstanding to alter the share price while keeping the common stock economic interest the same.
Despite this, I maintained a lukewarm positive view towards stock splits because the desire to split a stock usually signals some type of management belief that earnings will be rising at a nice clip in the near-term future. It’s irrational, but it causes a lot of psychological tumult for investors to see a stock do a 2 for 1 split and then fall 50% in value (if Chevron had done a 2 for 1 split in 2014 when it crossed $100, then a 100 share position at $100 would have become a 200 share position at $50. When the oil slide hit, the price of the stock could have gone to the $20s, which wouldn’t have been in keeping up with appearances).
This intuition about favorable earnings following a stock split is supported by the studies of K.C. Chen, Sangphill Kim, and Peter Xu, whose work is easily googleable under the keyword term “Long-Term Performance after Stock Splits: A Closer Look.” This was their conclusion: “Firms that split their shares from 1931 through 1990 outperform size-matched firm by 5.10 percent and beta-matched firms by 4.62 percent in the first year following the split.”
But I have come to disfavor stock splits beyond liquidity considerations (e.g. it wouldn’t do the world good if Coca-Cola traded at $75,000 per share because it would lock out small and medium sized investors.) Mostly, this is because I have noticed that stock splits have become tools used by management to help ensure that overvalued stocks remain overvalued.
One of the gaping holes of corporate governance is there is no immediate self-correcting mechanism for stocks that have gotten expensive since the management and shareholders love to see the stock price go higher. The stodgy hold forever types that like low valuations because it provides opportunity for new investment, more effective share repurchases, and an accelerant effect on dividends are an overwhelming minority of the investment community. The average S&P 500 Index Fund, for example, is held for five days. For too many people, the stock market is viewed as a place where you can easily buy some abstract letters on a screen, hope it pops in price, collect some moderate proceeds, and then try to repeat the process. The fact that you’re a residual claimant on the company’s profits from now until eternity is not usually grasped.
Then, you have the fact that it is common for management teams to earn bonuses and even basic compensation tied to the performance of a stock. Heck, the common refrain for a corporate manager’s legacy is to say, “The stock appreciated by 1,211% over the length of his tenure as CEO.” What incentive exists to correct that if a chunk of the legacy is due to the stock’s valuation unfairly increasing from 17x earnings to 24x earnings over the measuring period?
It is difficult to envision any realistic scenarios when a corporate management team or the shareholder base would ever rally for the stock to come down in price. If the stock is undervalued, management can turn to stock buybacks, mergers and acquisitions, dividend hikes, efficiency initiatives, and so on, to attempt to redress the inefficiency. If the management team is a cause of the inefficiency, an activist investor can wage a proxy battle to secure new management.
When it comes to overvalued stocks, stock splits can be used as a strategy to perpetuate the overvaluation by making the sticker price of a stock seem more reasonable. Take a look at something like Starbucks. By a great majority of the traditional metrics, the stock is moderately overvalued. The stock trades at just shy of $60 per share. It made $1.58 last year, and ought to earn somewhere around $1.85 per share this year. The $59.70 current price suggests a trailing P/E ratio of 37, and a looking forward valuation of 32x earnings.
Is it possible that the Starbucks Board of Directors voted to execute a 2 for 1 stock split last year due to the belief that a $60 stock with $1.58 in earnings seems like a more attractive purchase than a $120 stock with $3.16 per share in profits? My own observations of how others invest suggests that people do respond to the stated per share price of the stock, rather than focusing on the relationship between the profits per share and the share price (which is the relationship that really matters).
If you read the biography of Warren Buffett by Andrew Kilpatrick titled “Of Permanent Value”, you will see many quips from advisors stating that they declined to purchase the stock as its price hit $400, $800, $1,200, $1,600, $5,000, and so on, due to the belief that the per share price seemed too high. That’s mathematical illiteracy–the term “share” is a nearly meaningless concept without a discussion of the extent of the profit claim that the share represents.
For Starbucks shareholders that hold for the next ten years, your best case scenario is that profits grow by 12% annually and trade at 22x earnings in 2026. The realistic worst case scenario is that earnings grow by 6% annually over that time frame, and the valuation comes down to 16x earnings. If we use the projected $1.85 per share as our base since we are measuring what happens at the end of 2026, we are talking about profits being as high as $6.11 per share in 2026 or as low as $3.37. That is a price as high as $134 or as low as $53.92. Under the most optimistic set of scenarios, you are looking at 8.42% capital appreciation plus an extra percentage point for dividends (returns somewhere around 9.5% annually).
Under a 6% growth scenario, you’d get negative capital appreciation of 0.88% annualized, but with dividends, you’d end up getting 0%. And that doesn’t adjust for inflation.
Long story short, when I run the numbers on Starbucks, your possible outcomes are these: (1) if the stock keeps growing at a solid double-digit pace in the neighborhood of 12%, you’ll get total returns that are close to the historical performance of the S&P 500; (2) if you get only 6% annual earnings per share growth over the next six years, you will earn no nominal gains; and (3) there are also the returns not contemplated by my premise. To beat the market with Starbucks, you will need at least 13% annual earnings per share growth if the terminal valuation is 22x earnings. You will sustain a paper loss if the earnings per share growth comes in at 5% or lower.
To quantify the degree of Starbucks’ overvaluation, I would say that the actual investor results will lag the business performance results of Starbucks by three to five percentage points annually over the next ten years. The reason why the overvaluation is calculated as a range band instead of a precise figure is because it is guesswork to determine how investors will alter the P/E ratio for the stock in response to particular earnings announcements.
The harm, then, of the Starbucks stock split is that it perpetuates the overvaluation by making a cosmetic change that enables it to continue. Again, it’s understandable why this happens–the management owes fiduciary duties to the current shareholders and not the prospective ones, and the current shareholders want to maximize paper wealth–but it also demonstrates that stock splits may create a marketplace downside by taking advantage of behavioral psychology in a way that makes it easier for an expensive stock to remain expensive by changing the packaging that makes it easier to continue self-delusion about a mature company with 13,000 stores trading at a valuation of over 30x earnings.
Starbucks ought to trade somewhere in the low to mid $40s, based on current earnings and depending on the degree of optimism about future earnings. The shareholders that have been around for 5+ years have captured a bit of windfall to the exuberance of the marketplace–namely, the fact that profits have “justified” share price growth from $10 in 2007 to the low $40s today. But the gain from, say, $43 to the current price of $59.70 is a product of the investor community being willing to pay a price for the stock right now that is a bit higher than what will come to seem rational in hindsight.
It is entirely possible that the rapid price increases at Starbucks will continue for the next few years. Overvaluation usually gets disturbed when there is a specific event that causes a pause in double-digit earnings growth that wakes investors from the reverie and forces them to scrutunize the firm with fresh eyes. Think of something like Chipotle which has come down from $750 in October to $455 now in response to health scares at the restaurant. Or it can be a slower thing like Nike from 1999-2010 in which capital appreciation gently trails the earnings per share growth of the company. I’m agnostic on the manner in which overvaluation corrects itself.
But stock splits due enable the overvaluation to continue. It’s a little bit easier to convince yourself that Starbucks is a good buy in spite of the 30+ P/E ratio when you see a price tag of $60 instead of $120, and for that reason, I have grown wary of their regular use–once you get beyond maintaining desired liquidity levels for a stock, you might just be making a superficial change to slow down the arrival of inevitable P/E compression.