“An academic finance fairytale, but sometimes it works out.”

Making the rounds as a dividend investor in a stock market, I am frequently—as in everyday, as in within 5 minutes of every conversation—challenged about buybacks and their supposed superiority to dividends. So I wanted to lay out a typology of buybacks—a quick, simple matrix to save the time and energy of those asking and those answering (me) the question. As a historian, however, I need to explain briefly how the question came about it. It is not so much twisted logic as it is the logic of a twisted road.

Theoretical equivalence

Let’s start with a single assumption. The academics in the 1950s and 1960s did so, as they set out the basic tenets of academic finance. They assumed that an investor would be indifferent between a capital gain and a dividend payment. In their simplified black-board math, they had to. In the measurement of total return, the two forms of reward have equal standing. Period.

This is not the place to point out that a dividend is an outcome directly related to business ownership and that a capital loss (or gain) is a stock market result that can often be utterly unrelated to the affairs of the business. We also won’t point out at this phase that a capital gain is of no use in funding consumption, reallocating capital, or any other practical purpose. It is simply a number on a screen or a brokerage statement. In order to have real-world utility for the investor, it needs to be a harvested capital gain. That is, the asset needs to be sold (at a loss or a gain) in order to fund consumption or to be invested elsewhere. The academics didn’t bother with that consideration in their effort to theorize, but as a real-world investor, you have to.

Practical equivalence

So we have the supposed investor indifference between capital gains and dividend payments. Now we layer in buybacks. They didn’t exist as a mass phenomenon in the 1950s and 1960s when modern finance was being worked out. Buybacks became a more practical option for corporate America starting in 1982 with an SEC rule change. Within a few years, the S&P 500 Index companies were spending billions on their own shares. Since the mid 1990s, buyback dollars have outnumbered dividends paid in all but a handful of years. (This year or next, if current trends continue, publicly traded US corporations will spend $1 trillion, yes trillion dollars on buying back their own shares. Let that sink in.)

Corporate managers and many investors now conflate buybacks and capital gains. They assume that one leads to the other. Despite no definitive evidence of causation and correlation beyond markets moving up over time, there is little theoretical reason not to make the assumption. The key literature of Modigliani & Miller 1958 and can be updated easily enough to slot in buybacks and consider them to be a driving force for capital gains. (The single nit to pick there is whether the buyback is financed from debt or from free cashflow. For the sake of convenience, we will assume that the buybacks come operating cash flow. If they come from debt, they are a violation of M&M 1958.)

After those twists and turns in the academic and regulatory road, we now have a sort of vaguely theoretical and possibly practical equivalence of buybacks and dividends. They are lumped together in the phrase of “cash returned to shareholders.”  I put that phrase in quotes, because the semantics are wrong: one goes to shareholders; the other goes to sharesellers. But like many troubling moments in the buyback-dividend equivalence narrative, we will set it aside for now.

Let’s get to the typology. These are listed roughly in terms of relative innocence for the dividend investor in a stock market. What I don’t have, and perhaps someone will contribute, is the actual buyback dollars allocated by type. Surely a junior sell-side associate has run those numbers? I have no doubt that the big bucks are spent on the later scenarios, but that many attractive dividend-paying companies find themselves quietly and regularly in the first several scenarios.

Hedge fund devotees and short-term traders will want to reverse the order and the judgements that I have made.  They are free to do so. To be clear, … “I come not to bury buybacks, but also not to praise them… The folly that men do lives after them… Yet Fama says buybacks are misunderstood. And Fama is an honorable man…” (JC, Act III, scene II)

Time will tell which of these views is correct. My intention is to shine some basic light on the fairy tale buyback narrative that is highly profitable on Wall Street, but not much use on Main Street. But you will make your own judgments. And feel free to chime in with other variants that I may have overlooked.

The buyback matrix.

Actual Purpose
Stated Purpose
Source of Buyback
Practical Implication
Employee compensation
Offset share issuance to employees
Share count remains flat over time; buybacks in line with shares issued to employees
Operating cashflow
Overly complicated, but common form of employee compensation
Neutral, if overall comp levels are reasonable.
A large number of corporations engage in this practice. The dividend investor in a stock market would unnaturally & unnecessarily limit their opportunity set by excluding all such companies.
Maintaining control of cash
Maintain a fixed absolute dollar dividend amount
Goal is manage aggregate dollars paid out in dividends by having share count decline by the amount of desired per share div growth.
Operating cashflow
Seen in cyclical companies such as energy
Reasonable capital allocation strategy
Designed to sustain dividends in a cyclical environment, to avoid future cuts
Fixing a dividend leak
Reduce cash dividend obligation
Share count reduction  leads to reduced dollar dividend obligation.
Operating cashflow
It has its place.
Supportive of the dividend, but cannot be a permanent solution for companies in distress.
Seen opportunistically in mature/declining industries with strong cashflows but declining or no growth prospects.
Acquisition offset
Offset share issuance used for acquisitions.
Share count returns to the prior level.
Operating cashflow.
Common in large acquisitions
Neutral, if share count returns to prior level.
In this instance, the buyback is not the issue. It is the acquisition. Acquisitions large enough to require share issuance (followed by buybacks) have a very mixed record.
Tax clientele effect
Give investors a choice between a dividend and a harvested capital gain, for tax reasons
Assumes buybacks lead to capital appreciation; assumes investor avoids harvesting a capital loss
Operating cashflow
Turning over investor tax management to investment company executives.
Too clever by half.
Clientele effect widely seen to justify buybacks.
A modern finance fairy tale
“Return cash to shareholders”
Share count reduced by  similar % as the $ value of the buyback divided by the market cap outstanding at the time of the announcement.
Operating cashflow
Shareholders receive a larger portion of the corporation; all other factors being held equal, their dividends may benefit, as the same dividend dollars are being divided among fewer shareholders.
A modern finance fairytale, but sometimes it works out.
There are selected instances where regular buybacks contribute to intermediate-term per share dividend growth.
Leveraging the balance sheet
“Return cash to shareholders”
Reduce share count
Simple violation of M&M 1958; leveraging up does not change the productivity of the underlying assets.
Usually ends badly.
Companies financing buybacks with debt rather than cashflow are rarely in the forefront of the long-term dividend paying universe. They can’t afford to be.
Share price speculation
“Return cash to shareholders”
In the case of companies with no or de minimus dividends, it doesn’t matter.
A popular space for large tech companies and other companies with no history of paying a dividend.