Stock splits are common. The widespread use of stock splits has historically kept the share price of publicly traded companies constant despite inflation and real growth. Reverse splits are used to maintain price points in much the same way.
One long-term study of stock splits for the period 1963 to 1982 showed that, on average, 6 percent of companies split their stock each year, at a median presplit price of 43.50 dollars.11 Another study showed that the average share price on the NYSE has fluctuated between $30 and $40 since the 1930s.12 This price range has been maintained even though consumer prices have increased by a factor of 10 and the S&P index increased by a factor of 16 since the 1930s. Of 7,726 companies, 21 percent split over the past 15 years, and 5 percent split over the past 5 years.
But since 1990, stock splits have not kept pace with the market and the average stock price has increased. Over the period 1990 to 2005, the average price of all shares outstanding rose from $18 in 1990 to $40 in 2005. Over the same period, the percentage of shares priced above $50 soared from about 3 percent to more than 9 percent.
The data do show, however, little increase in the number of companies that are willing to let their stock price go above $100, and this figure has held at about 1 percent. This seems to be primarily a marketing decision with $100 representing an important premium position price point in the psyche of investors.
The prevalence of splits has long been paradoxical because they have clear costs (e.g., listing fees, administrative costs, and brokerage commissions) yet no obvious economic benefits; the split, by itself, has no favorable impact on future cash flows.
The Cost of Stock Splits
Stock split costs include higher costs to the company for registration and administration and higher brokerage commissions borne by investors. For companies listed on the NYSE, new shares created by a split will incur an initial listing fee, up to a maximum of $250,000, and may lead to increased annual listing fees. Administrative costs include higher printing and mailing costs to service a larger shareholder population. One estimate puts the costs of a stock split for a Dow 30 company above $1 million.13 In addition to the costs borne directly by the company, a stock split increases the costs borne by investors, and brokerage commissions typically increase with the number of shares traded even when the dollar amount of the trade is the same.
Benefit of Stock Splits
Many reasons are cited to explain the prevalence of stock splits. In terms of value creation, these arguments may be categorized along one of four rationales: ownership, stock liquidity, buy-side signaling, and sell-side promotion.
Ownership
Stock splits do typically increase the number and diversification of shareholders, ostensibly by making it easier for retail investors to purchase round lots of shares and perhaps through the irrational appeal of a lower price point. This could lead to a more liquid and efficient market in that security, with value created through lower bid-ask spreads, less risk, and a lower cost of capital.
One survey found that 98 percent of managers cited lower round lot costs and 94 percent cited an increase in the number of shareholders as benefits of stock splits.14 A stock split reduces the cost of a round lot. A stock split also appears to increase the number of shareholders. A lower stock price is associated with a substantial increase in the number of shareholders when we compare different companies at the same point in time. We found that a higher split factor (i.e., the ratio of post-split to presplit shares outstanding) is associated with a substantial increase in the number of shareholders. One study found that company size (total market value of equity), stock volatility, and stock price explained 32 percent of the variation in the number of shareholders of Compustat companies. Holding size and volatility constant, each $1 reduction in stock price increased the number of shareholders by 1.4 percent. This model implies that a two for one split that reduced a stock price by $25 would increase the number of shareholders by 42 percent. Similar results were achieved with a time series model across 5 years, with split factor, beginning equity market value, change in volatility, and 5-year price explaining 7 percent of the change in the number of shareholders.15
The greatest benefit to stock splits may be a broadened and potentially more sympathetic shareholder base. Individual shareholders tend to be more likely to support incumbent management (e.g., option reserves and director elections) than institutional shareholders. Stock splits can reduce the threat of hostile takeovers and help gain access to a more sympathetic shareholder base for proxy votes.
Stock Liquidity
Prior research shows that stock splits can reduce trading volume.16 One study found that a two for one split reduced trading volume over a 5-year time series by 2 percent of shares outstanding of more than 20 percent of the ADTV.17
Of all the capital market tactics to increase stock liquidity that we evaluated, stock splits exhibited the smallest benefit. In fact, as the ADTV of illiquid stocks improves by a modest 20 percent, the ADTV of liquid stocks decreases by 5 percent. Only for 4 percent of the illiquid stocks is liquidity enhanced enough to warrant reclassification to a liquid stock. Stock splits do not generally improve liquidity in a meaningful way, most likely because stock splits do not address underlying root cause issues of float and volume. Furthermore, illiquid companies undergoing stock splits continue to be illiquid post stock split.
Evidence also exists that stock volatility increases with stock splits though the long-term increase is modest.18 This model estimates that a two for one split increases stock volatility by 0.005, less than 2 percent of the average sock volatility. This gives rise to another manager benefit to stock splits: A higher stock volatility makes existing stock options more valuable; however, future option grants may be revised downward over time to offset this effect. Nor is there evidence to support a lower beta (cost of equity) due to splitting or among splitting companies. A lower beta is required to support the liquidity argument or the tax option argument; increased volatility adds a tax option value and, thus, lowers the stocks expected or required rate of return.19
Signaling
Stock splits can provide a way for managers to signal more positive future prospects, increasing the stock price through a higher attribution of growth value.
However, for liquid and illiquid stocks, signaling seems poor; stock splits are associated with negative excess returns, −1 percent and −9 percent, respectively.20 Furthermore, studies have shown the long-term excess returns are negative. One long-term study found 5-year post-split excess returns to be significantly negative.21 Though abnormal stock returns were positive for the first year after the ex-date of a split, they were negative for the second through fifth years after a split, with a mean 5-year matched buy-and-hold cumulative abnormal return of −16.2 percent.
Splits are often associated with other changes (e.g., profits and dividends), making it difficult to establish cause, versus mere association. Bid-ask spreads, as a percent of stock price, increase after a stock split proportionally to the decrease in stock price due to the split. Companies may use splits as a favorable signal effect.22 The signaling argument is only justification for a stock split if the split causes the market reaction and not other concurrent or expected events. But one study found that companies maintained their excess return only if the split was employed to affect a dividend increase.23 Companies that failed to increase their dividend lost part or all of their excess return. If the split, rather than the prospect of an increase, caused the excess return then a failure to increase would not erode the return. Research shows that the market reaction to a second stock split is reduced if the earnings improvement following the first split was below average.24 A share repurchase likely provides a better signaling mechanism than a stock split.25
Promotion
Some have proposed that because stock splits increase brokerage commissions and generate transaction fees, they may lead to increased research and sales coverage. Brokerage commissions are higher for smaller investors and smaller trades, and bid-ask spreads are larger for lower priced stocks. Some believe that these incentives improve the supply and demand situation and increase share price.
However, our own analysis as well as published research suggests that splits do not benefit trading volume or stock prices, making the sell-side promotion rationale a weak argument to justify a stock split.
For companies that do choose to split their stock, we recommend they revisit dividend policy and consider maintaining the same dividend per share, an effective dividend increase, particularly if signaling is the rationale for the split.26 We propose that stock options be adjusted for the split, and that grant targets be adjusted for the split by multiplying target grants by the split factor. The objective of fixed share grant guidelines is to align manager and owner interests, and the fixed percentage ownership interest can only be maintained with this adjustment.
1. Yakov Amihud and Haim Mendelson, “The Effects of Beta, Bid-Ask Spread, Residual Risk, and Size on Stock Returns,” Journal of Finance 44:2 (June 1989). Yakov Amihud and Haim Mendelson, “The Liquidity Route to a Lower Cost of Capital,” Journal of Applied Corporate Finance 12:4 (Winter 2000). Justin Pettit et al, “Rx for Stock Liquidity” (November 2005). Available at SSRN: http://ssrn.com/abstract=845544
2. Kate O’Sullivan, “Hey! Look at Me,” CFO Magazine (October 2005).
3. An informal poll of equity analysts indicates that CEOs (companies with $1.5 billion market cap and under) spend 20 to 25 percent of their time on investor relations, and CFOs spend an even higher proportion of their time. Harrison Hong and Ming Huang, “Talking up Liquidity: Insider Trading and Investor Relations,” Journal of Financial Intermediation 14 (2005).
4. Richard Roll, “A Simple Implicit Measure of the Effective Bid-Ask Spread in an Efficient Market,” Journal of Finance 39:4 (September 1984). This article showed that bid-ask spreads widen inversely to firm size. Najah Attig et al, “Effects of Large Shareholding on Information Asymmetry and Stock Liquidity,” SSRN Working Paper Series (August 2004). This article showed that stocks with a greater separation between control and ownership had wider bid-ask spreads.
5. Christine A. Botosan, “Evidence That Greater Disclosure Lowers the Cost of Equity Capital,” Journal of Applied Corporate Finance 12:4 (Winter 2000).
6. Clauido Loderer and Lukas Roth, “The Pricing Discount for Limited Liquidity: Evidence from the SWX Swiss Exchange and the NASDAQ” (September 2003). Available at SSRN: http://ssrn.com/abstract=288965 found least liquid stocks suffered discounts of up to 30%. Aswath Damodaran, in The Dark Side of Valuation (New York: Prentice Hall, 2001) pp. 245–249. This article indicates that liquidity discounts fall in a continuum, with private companies typically valued at a 25 to 30 percent discount.
7. Douglas Diamond and Robert Verrecchia disclosure, “Liquidity, and the Cost of Capital” The Journal of Finance 46:4 (September 1991). This article shows that increased disclosure can improve liquidity and reduce the cost of capital.
8. Michael Brennan and Claudia Tamarowski, “Relations, Liquidity, and Stock Prices” Journal of Applied Corporate Finance 12:4 (Winter 2000). The authors found that more disclosure of forward-looking and key nonfinancial information led to a lower cost of equity and higher stock prices in seasoned equity offerings with greater benefits for companies with less research coverage.
9. Lisa K Meulbroek, “The Efficiency of Equity-Linked Compensation: Understanding the Full Cost of Awarding Executive Stock Options” Financial Management 30 (2001). Lisa Meulbroek developed a model to estimate employee discounts based on share price volatility, which ranges for most cases from 30 to 50 percent.
10. Excess return equals aggregate movement in market capitalization, in excess of the movement in the S&P 500, over the same 45-day time period, starting with each respective date of file. Time from file is typically 20 days.
11. Marlin R.H. Jensen, Claire E. Crutchley, and Carl D. Hudson, “Market Reaction to Equity Offer Reasons: What Information Do Managers Reveal?” Journal of Economics and Finance 18:3 (Fall 1994), pp. 313–329. The authors employ a piecewise linear model to provide evidence that managers signal the quality of the new investment when issuing equity. File to offer period not examined.
12. Josef Lakonishok and Baruch Lev, “Stock Splits and Stock Dividends: Why, Who, and When,” Journal of Finance 42:4 (1987).
13. James J. Angel, “Picking Your Tick: Toward A New Theory of Stock Splits,” Journal of Applied Corporate Finance (Fall 1997).
14. Ibid.
15. Baker and Gallagher, “Management’s View of Stock Splits,” Financial Management 9 (Summer 1980).
16. Justin Pettit and Stephen F. O’Byrne, “Stock Splits: What Good Are They?” Shareholder Value (May/June 2002).
17. Thomas E. Copeland “Liquidity Changes Following Stock Splits,” Journal of Finance 34:1 (March 1979). The author found that trading volume declined after a split. Josef Lakonishok and Baruch Lev, “Stock Splits and Stock Dividends: Why, Who, and When,” Journal of Finance 42:4 (September 1987). Lakonishok and Lev concluded that trading volumes are abnormally high before a split announcement and then return to a normal level after the announcement.
18. Justin Pettit and Stephen F. O’Byrne, “Stock Splits: What Good Are They?” Shareholder Value (May/June 2002).
19. Ibid.
20. Christopher G. Lamoureux and Percy Poon, “The Market Reaction to Stock Splits” Journal of Finance, 42:5 (1987).
21. The difference between the means of liquid and illiquid returns is significant at a 95 percent confidence level.
22. Craig Dunbar, Chuan-Yang Hwang, and Gershon Mandelker, “Long-Run Common Stock Performance After Stock Splits: Anomalous Evidence From 1929 to 1988,” Unpublished Working Paper, University of Western Ontario (July 1998). These results were pervasive over various subperiods of history as well as for a wide variety of methodologies.
23. Robert Conroy, Robert Harris, and Bruce Benet, “The Effects of Splits on Bid-Ask Spreads,” Journal of Finance 45:4 (September 1990).
24. Eugene Fama, Lawrence Fisher, Michael Jensen, and Richard Roll, “The Adjustment of Stock Prices to New Information,” International Economic Review 10 (February 1969).
25. Eugene Pilotte and Timothy Manuel, “The Markets Response to Recurring Events: The Case of Stock Splits,” Journal of Financial Economics 41 (1996).
26. Justin Pettit, “Is A Share Buyback Right For Your Company?” Harvard Business Review (April 2001).