Untangling Buybacks Vs. Dividends

by: Peter Hofmann, CFA
Summary

No value is created when a company pays dividends or buys back shares.

Share repurchases are preferable to dividends only if done at a price below the intrinsic value of the company.

Other purported benefits of share repurchases aren’t very compelling.

Given today’s market structure and interest rate environment, investors should view share repurchases skeptically.

According to the Financial Times, corporations in America repurchased almost $800 billion worth of stock in 2018, fueling an ongoing debate over the merits of share repurchases. Coupled with $456 billion of dividends, a record $1.25 trillion of cash was paid to shareholders.[i]

Politicians have promoted the idea that money used for share repurchases would be better spent on reinvestment, including higher pay for workers. I’ll be up-front about my bias: such proposals place a lot of faith in corporate executives’ ability to deploy excess capital productively. I prefer to see companies pay their shareholders and let the market decide how to invest it. The focus of this piece is on how to best do that.

Both dividends and share repurchases return cash to shareholders. As long as certain conditions are met they are equivalent and leave shareholders and the company no better or worse off. A simple stylized example will illustrate.

The equivalence of dividends and share repurchases – a stylized example

We’ll assume a simple company with 10 shareholders whose intrinsic value is reflected in its market price. Each shareholder owns 100 shares, for a total of 1000 shares. The company earns $10 per share and trades at a price earnings multiple of 15, or $150 per share. Each shareholder thus has wealth of $15,000.

The company has $2,000 of cash it wants to return to shareholders. It can choose to pay a dividend of $2 per share. If it does this, the share price would drop by the amount of the dividend, to $148 per share. The P/E ratio is now 14.8. After the dividend, each shareholder owns $14,800 of stock and $200 of cash. Total wealth of each shareholder is still $15,000.

Alternatively, the company can choose to repurchase $2,000 of shares at $150 each, or a total of 13 1/3 shares. Among the shareholders, some must be willing to sell (let’s assume one of the ten shareholders sells.) After the repurchase, the company has 986 2/3 shares outstanding. With fewer shares, earnings per share increase to (rounded) $10.14. As before, the reduction in cash lowers the P/E ratio to 14.8. With the higher earnings per share, the share price is $150. Again, total wealth of each shareholder remains at $15,000. For the shareholders who stayed, their $15,000 of wealth now represents 10.13% of the company instead of 10%. The selling shareholder’s wealth consists of $13,000 in stock and $2,000 in cash.

The example illustrates the theoretical equivalence of dividends and share repurchases. In real life, the comparison is less straight forward. Above all, it occurs in the fray of day-to-day market activity. With a share repurchase, investors as a group face a zero-sum game: if the company can repurchase more shares (at a lower price than its intrinsic value of $150) remaining shareholders benefit at the expense of those who sell. Conversely, if the repurchase price is above intrinsic value, remaining shareholders receive less value than they would from a dividend, and the selling shareholders benefit.

The Real (and Only) Reason to Prefer Share Buybacks Over Dividends

The key issue for share repurchases is thus the valuation of the shares to be bought back. Share repurchases are equivalent to dividends, or even preferable to dividends, if the shares are bought at fair value or a valuation discount. This is why Warren Buffett has embraced share repurchases over dividends:

“For continuing shareholders, the advantage is obvious: If the market prices a departing [investor’s] … interest at, say, 90¢ on the dollar, continuing shareholders reap an increase in per-share intrinsic value with every repurchase by the company. Obviously, repurchases should be price-sensitive: Blindly buying an overpriced stock is value destructive, a fact lost on many promotional or ever-optimistic CEOs.”[ii]

The problem is, of course, that one must trust management’s ability to 1) correctly assess the fair value of shares and 2) time repurchases to capture the discount. If you believe markets are efficient, this is a non-starter since the market price should already reflect intrinsic value. If not, you might embrace the notion that management’s unique perspective as insiders gives them a leg-up in assessing the value of the company. Unfortunately, there is little evidence that this is the case.

Not-so-good Reasons to Like Buybacks

Proponents of buybacks offer several other purported benefits, none of which are very convincing.

Share Repurchases Have Tax Advantages

One common argument is that repurchases are preferable because of how they are taxed. This was true when dividends were taxed at substantially higher marginal tax rates than realized capital gains and the majority of stocks were owned in taxable accounts. It’s far less compelling today.

Shareholders in the U.S. are taxed when dividends are received at federal tax rates up to 23.8%. Repurchases allow gains to accumulate and be taxed only once the shares are sold. Since long-term capital gains rates also top out at 23.8%, many shareholders should be indifferent unless they expect their tax rate on capital gains in the future to be lower than the rate on dividends today. [iii]

Moreover, the difference is moot for shares owned by tax-exempt investors such as pension funds and by individuals in retirement accounts. According to one estimate the proportion of U.S. stocks owned in taxable accounts had declined to 24% in 2015 from 84% in 1965.[iv] When combined with dividend and capital gains tax rates that are equivalent at many income levels, the tax advantages of share repurchases are a lot less compelling than they used to be.

Debt-Financed Repurchases Create Value

A lot of share repurchases are paid for with debt.[v] (This is supposed to make us feel good.) Therefore it’s sometimes claimed that share repurchases create value because companies get to deduct the interest. I have two concerns with this argument. One, while it may be true that higher leverage decreases a company’s cost of capital and thus increases its value, this is separate from any decision to repurchase shares. Swapping equity capital for debt capital can just as easily be accomplished through dividends. Two, it’s not surprising given the low-interest-rate environment and easy credit conditions that many companies have “optimized” their capital structures (though the long-term effects of this are as yet unknown.) It’s also true, though, that equity valuations have been high, increasing the odds that remaining shareholders will ultimately be on the losing side of the equation.

Share Repurchases “Signal” Good Things Ahead

Research has found that stocks react favorably to share repurchase announcements.[vi] It’s then often stated as an article of faith that the reaction is because management is “signaling” their belief that the company’s shares are undervalued. We are to believe that the market, heretofore apparently catatonic, suddenly awakens and closes the valuation gap. I don’t buy it. Every management team will claim their stock is undervalued, but the notion that shouting it from the rooftops (or in an 8-K) causes valuation to increase is speculative at best. I will counter with several equally speculative, but more plausible reasons: First is the aforementioned leverage effect. While stock repurchases don’t create value per se, the decrease in the cost of capital from higher leverage may. Second, one might reasonably expect to see an “ex repurchase” trade – once investors know repurchases are coming, they may bid up the stock. What do I mean by this? Consider that for dividends, the “ex” date is public knowledge, but for open market share repurchases it is known only to insiders. This means investors must buy quickly following announcement in order to capture any benefit from the return of capital. Third, a repurchase program could result in a liquidity benefit from having an additional buyer in the market.

Share Repurchases Can Be Used More Opportunistically

Many companies combine dividends with opportunistic share repurchases, arguing that a dividend should be predicable and repeatable. It’s true that in the U.S. investors are conditioned to expect stable or increasing dividends and harshly penalize companies that reduce their dividends. But there is nothing to prevent companies from setting more flexible dividend policies or declaring special dividends instead of share repurchases. An excellent example is The Progressive Corporation (NYSE:PGR), which has a clearly articulated variable, yet flexible dividend policy.

Share Repurchases “Offset Dilution” from Stock-Based Compensation

This is a popular rationale when companies announce share repurchases. It’s smoke and mirrors. When a company issues equity, to its employees or otherwise, it gives a percentage of existing shareholders’ ownership to the new shareholders. The choice to pay employees in shares rather than cash may well have merit, since it conserves cash and may align employee interests with those of shareholders. But, as the stylized example above illustrates, spending additional cash to reduce the number of outstanding shares does nothing to increase the value of existing shareholders’ diluted wealth.

Share Repurchases Increase Earnings Per Share

Occasionally one hears the argument that decreasing the shares outstanding will increase earnings per share, and that this is a good thing. But this math is incomplete because it fails to take into account the reduction in earnings due to the cash (or debt) used to buy back the shares. Whether repurchases are accretive or dilutive to earnings per share depends on the multiple at which shares are bought.[vii] As Credit’s Suisse’s Michael Mauboussin points out, they have been highly accretive to earnings per share at least in part because companies have been able to borrow at record low interest rates. But, he notes, “Since accretion says nothing about the economic merit of a buyback, you can’t say whether it is good or bad.”[viii] Of course, if management repurchases shares at prices above intrinsic value to hit accounting-based earnings per share compensation targets, shareholders that stick with the company are left holding the bag.

Share Repurchases Let You “Roll Your Own” Dividend

Buyback proponents often argue that you can “roll your own” dividends by simply selling some stock. They argue this provides greater flexibility and allows each investor to tailor distributions to their circumstances. But how realistic is this recommendation in practice? With a growing number of investors owning shares through funds, especially index funds and ETFs, the “roll you own” option does not actually exist. If companies don’t pay dividends, fund investors seeking distributions need to liquidate the entire index, which will include selling shares of companies with good investment opportunities and no excess distributable cash.

Moreover, on a practical level, it’s hard to imagine many individual investors hold large enough positions to create their own dividends efficiently. Consider an investor who owns 1000 shares of Microsoft with a total value of $120,000[ix]. To replicate Microsoft’s 1.5% annual dividend yield he would need to sell 3¾ shares of the company every quarter.[x] Standard discount brokerage commissions represent annualized transaction costs of well over 1% of a trade of this size.

Why We Should Mistrust Buybacks

Misaligned Management Incentives

I argued that the tax differences between dividends and share repurchases are increasingly irrelevant. There are two groups of shareholders for whom this is not the case. One consists of the very corporate boards and managers who make the payout decision. These and other wealthy investors who can afford to hold shares indefinitely will have a strong preference for share repurchases because the basis in shares “steps up” upon death, allowing them to avoid taxation altogether. Even during their lifetimes they are able to fund charitable contributions with appreciated shares, securing a deduction against other income and effectively avoiding taxation of gains. Case in point: Mark Zuckerberg’s pledge to donate 99% of his Facebook stock “to charity,” which will may well mean that no taxes will ever be paid on $45 billion of capital gains.[xi]

This may or may not bother you from a societal point of view, but it’s an un-level playing field that should bother you as a shareholder. It means that the very executives/shareholders making the repurchase decision have a “break-even” price for share repurchases up to 31% above intrinsic value and still end up with wealth equivalent to a taxable dividend distribution. Meanwhile, for investors who rely on their portfolios to support retirement spending, this translates into value-destroying prices relative to a dividend. (See below for more on why many individual investors can’t simply sell when the company is buying.)

Whether or not executives consciously pursue uneconomic repurchases is difficult to ascertain, though there are certainly many dramatic anecdotal examples[xii] of adverse outcomes for remaining shareholders.

(The other group for whom there is differential taxation is foreign investors, whom the U.S. taxes on dividends but not capital gains. Since foreign investors by some estimates account for as much as 30% of shareholders of U.S. corporations, this is not a trivial issue in the face of record budget deficits, but it gets us into another policy debate, which is not the focus of this article.[xiii])

In dismissing the “signaling” effect of share repurchase announcements, I also left out the impact on executives. It turns out that there is a tendency by executives to sell shares after a repurchase announcement, allowing them to benefit from the run-up in share prices. For firms that saw high insider selling following share repurchase announcements,

“after about 60 days the stock actually begins to under-perform. A year later, the group with above median insider selling under-performs relative to the group with below [or] no insider selling by over 3%—wiping away any gains from the larger initial positive buyback announcement effect.”[xiv]

Share Repurchases Usurp Capital Allocation by the Shareholders

Share repurchases represent a forced reinvestment of distributions into the company for many shareholders. In our example above, each remaining shareholder owns $15,000 of stock in the company, whereas with a dividend they would own $14,800 of stock and $200 of cash, giving them the option to reinvest in the company or allocate their capital elsewhere. With a growing number of shareholders owning shares through index funds or ETFs, many are effectively forced to reinvest in the company – management usurps the market’s role in judging whether shares are worth reinvesting in.

Dividends, however, preserve the shareholders’ role in capital allocation. Dividends are distributed to investors regardless of whether the share are held directly or indirectly through funds.

Where does this leave investors? Should we favor dividend-paying companies over those that repurchase share? Can we differentiate between those companies that repurchase prudently and those that do not? These are fundamental investment questions without easy or generic answers. Companies with clearly articulated dividend policies and share repurchase guidelines make the task easier. Clearly not all share repurchases are bad, but there are a number of reasons to view them with skepticism. On balance, companies that pay dividends are likely to treat shareholders more equitably that those who rely primarily on share repurchases. Unfortunately, when it comes to investing fairness does not equate to out-performance.


[i] Financial Times, March 5, 2019. US investor cash return bonanza breaks records

[ii] Warren Buffett 2018 letter to shareholders of Berkshire Hathaway, p. 7. http://www.berkshirehathaway.com/letters/2018ltr.pdf Note that Mr. Buffett is speaking as a remaining shareholder. If he were a selling shareholder, I’m sure he’d be writing about the joys of selling at 110 cents on the dollar. This underscores the zero-sum aspect of share repurchases – they ultimately represent a transfer of value from one set of shareholders (sellers) to the others (remaining shareholders.) This has led some to argue that the price at which shares are repurchased doesn’t matter from an overall shareholder perspective.

[iii] To see why, return to our example. At a tax rate of 23.8% shareholders who receive a dividend pay tax of $200 x .238 or $47.60 after taxes. If the value of the firm increases by 5% over the following year, their stock will be worth $14,800 x 1.05 = $15,540. Their total wealth after one year (assuming they also invest their dividend proceeds at a 5% return) will be $15,540 + 152.40 x 1.05 = $15,700.02. In a share repurchase, remaining shareholders don’t owe current tax. But the value of their stock will be $15,000 x 1.05 = $15,750. They have an additional accrued gain of $210 compared with the dividend case. If they sell they will owe tax of $210 x .238 or $49.98 on that gain. Once again, their total after-tax wealth will be $15,750 – 49.98 = $15,700.02.

[iv] Steven M. Rosenthal and Lydia S. Austin, “The Dwindling Taxable Share of U.S. Corporate Stock”, Taxnotes, May 16, 2016, The Dwindling Taxable Share of U.S. Corporate Stock

[v] Clifford Asness, Todd Hazelkorn and Scott Richardson, “The Buyback Derangement Syndrome”, Journal of Portfolio Management, Spring 2018, 44 (5) 50-57; DOI: Buyback Derangement Syndrome

[vi] Office of SEC Commissioner Robert J. Jackson Jr., “Data Appendix to ‘Stock Buybacks and Corporate Cashouts: An Update’”, March 7, 2019. https://www.sec.gov/files/data-appendix-030619-letter.pdf

[vii] Generally, repurchases will accretive if the earnings yield (reciprocal of price earnings multiple) is higher than the after-tax cost of debt. With the S&P 500 earnings yield north of 4% and after-tax interest rates under 2.5%, repurchases have been highly accretive.

[viii] Michael Mauboussin and Dan Callahan, “Disbursing Cash to Shareholders – Frequently Asked Questions About Buybacks and Dividends”, Credit Suisse, May 6, 2014. http://www.shareholderforum.com/wag/Library/20140506_CreditSuisse.pdf

[ix] MSFT share price of $119.92 as of 4/8/2019

[x] After taxes the trades might be even smaller since part of the sale would return of basis.

[xi] How Mark Zuckerberg's Altruism Helps Himself

[xii] See for example Barry Ritholz, “Bad Buybacks”, The Big Picture, November 29, 2018 Bad Buybacks - The Big Picture

[xiii] Daniel J. Hemel and Gregg D. Polsky, “There’s a Problem With Buybacks But It’s Not What Senators Think”, Taxnotes, February 19, 2019, There's a Problem With Buybacks, but It's Not What Senators Think and Rosenthal and Austin (2016)

[xiv] Office of SEC Commissioner Jackson, 2019

Disclosure: I/we have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours. I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it. I have no business relationship with any company whose stock is mentioned in this article.