The decisions that look wrong when you make them

Good morning. In April 2024, General Electric finished the breakup that critics had been pushing for since the Immelt era. The old conglomerate split into three publicly traded companies: GE Aerospace, GE HealthCare, and GE Vernova. Within months of the separation, the combined market value of the three pieces was substantially larger than what the whole company had been worth together.

The standard reading of that outcome is that GE finally fixed a broken company. However, the more useful reading is that GE took longer than they should have and paid for the hesitancy of management.

Today's memo is about three decisions like that one. When to sell a business. When to buy back your own stock. When to acquire someone else's. Each question has a long academic literature behind it, spanning three decades and thousands of transactions. The findings converge on a single point: the version of each decision that creates the most long-term value almost always looks wrong at the moment it's made.

Which is why most companies get them wrong.

---

PART ONE

Selling the Business That Still Works

Open any financial publication on a given Monday and you will find a dozen stories about companies buying each other, and maybe one about a company selling off a division. This reporting ratio is backwards, and the academic evidence makes it clear why.

Emilie Feldman at the Wharton School has spent most of her career studying the sell side of corporate strategy. The broader divestiture literature, going back to John and Ofek in 1995 and Daley, Mehrotra, and Sivakumar in 1997, has consistently documented positive announcement returns when companies divest units that do not fit the parent's strategic focus. The parent gets sharper focus and freed capital. The divested unit gets independent decision-making under its own capital logic. Both stock prices tend to rise. Feldman's own more recent work, including her 2018 paper with Gartenberg and Wulf in the Strategic Management Journal, has continued to document the value created when corporate portfolios are simplified. The academic evidence is one-sided enough that the interesting question is no longer whether divestitures create value. It is why they remain so underutilized.

The barrier to acting on this evidence is psychological. Selling a profitable business looks like admitting that the original acquisition was a mistake, even when the thesis was sound and only the strategic context has changed. Selling a growing business means forgoing cash flows that are visible today for the possibility of better returns elsewhere. Any divestiture creates personnel consequences that boards prefer to avoid. So companies wait. They wait until the unit starts to stall, until the strategic rationale for holding becomes impossible to defend, until activist investors force the issue. By that point, the best price is gone.

The counterintuitive result: the best time to sell is when the business is still attractive to buyers. That is when price multiples are highest, when the transition is cleanest, and when the parent captures the most value. Companies that wait until a business is in visible decline end up selling to buyers who know they have leverage.

Danaher is the instructive case. Over the past decade, the Washington-based industrial conglomerate spun off Fortive in 2016, Envista in 2019, and Veralto in 2023. Each divestiture separated a healthy, growing, profitable subsidiary from the parent. Each announcement was second-guessed by commentators at the time. Each one proved to be the right call. Danaher's current business, focused on life sciences and diagnostics, now compounds at rates that would have been impossible if the parent had held onto everything. The stock prices of the spun-off companies, meanwhile, generally delivered strong returns as independent entities. Two successful companies grew out of soil that had held one sprawling conglomerate.

The discipline of a successful divestiture program: sell when the business is valuable, not after it has already declined. The mechanism that makes this hard is the same mechanism that makes it rare, and therefore valuable when executed.

---

PART TWO

The Two Kinds of Buybacks

A decade ago, a finance professor at the University of Michigan named Amy Dittmar got her hands on a dataset that corporate finance had never seen. It contained the actual prices at which every U.S. public company had executed share buybacks, month by month, from 2004 through 2011. Real transaction prices. Not announcements. Not quarterly averages. The numbers that accountants settle on after the trades clear.

She described it later as a peek inside a black box. Working with Laura Casares Field of the University of Delaware, she published the results in the Journal of Financial Economics in 2015. The paper asked a question that had been debated for decades without resolution: can managers actually time the market when they buy back their own stock?

The headline finding was modest. The average firm bought back shares at a price roughly 2 percent lower than the average market price of those shares over comparable windows. On the surface, that suggests corporate America is a slightly skilled buyer of its own stock.

Howver the real story is in the variation. Infrequent repurchasers, the companies that buy only when they believe shares are genuinely undervalued, achieved discounts of 2 to 6 percent on their purchases. They then earned positive risk-adjusted returns of roughly 0.3 percent per month for the three to thirty-six months following each repurchase. Frequent repurchasers, the companies that buy back stock quarter after quarter regardless of price, got no such benefit. They paid close to average market prices and earned no abnormal returns.

The divide maps onto a behavioral distinction that is easy to describe and hard to practice. Some companies repurchase stock opportunistically, with an explicit valuation framework, when internal analysis says the shares trade below intrinsic value. Other companies repurchase stock routinely, because cash has accumulated and returning it is easier than finding a use for it. The first group creates value. The second group converts cash into shares at prices that fail to beat holding the cash.

The cautionary tale is Intel. Between 2005 and 2020, the company spent roughly $108 billion on share repurchases at prices that, in hindsight, averaged well above the stock's trading range in the years that followed. The foundry strategy then required more capital than the operating business was generating. The company had to raise debt and cut its dividend. Buying back stock at elevated prices and later needing to fund capital at depressed prices is the worst possible outcome for shareholder capital. It happens when buybacks are treated as a tool for managing quarterly earnings per share rather than as a capital allocation decision.

The discipline, then: treat share repurchases with the same rigor as any other investment. If the expected return from retiring shares at today's price is lower than the opportunity cost of capital, do not buy. Let the cash accumulate. Wait. The research shows that companies willing to wait capture the returns. Companies unable to wait do not.

The hard part, again, is the waiting. Boards and investors grow restless when cash sits. The routine buyback becomes the path of least resistance. That is exactly the behavior the research identifies as destroying value.

---

PART THREE

Acquiring When No One Else Can

In 2005, Jarrad Harford, a finance professor then at the University of Washington, published a paper in the Journal of Financial Economics that has shaped how economists think about merger waves ever since. He looked at decades of M&A activity across industries and found that waves form when two conditions converge. A sector-specific shock, whether economic, regulatory, or technological, creates the opportunity or the urgency for deals. Abundant liquidity makes those deals financeable. Both conditions have to be present. Shocks create the opportunity. Liquidity enables execution. When they overlap, deal activity spikes reliably.

That was the descriptive finding. The practical implication came from the follow-up research. Ran Duchin and Breno Schmidt, writing in the Journal of Financial Economics in 2013, extended Harford's framework by studying long-run returns for acquirers. Companies that acquired during merger waves experienced annualized buy-and-hold abnormal returns roughly 4.65 to 6.25 percentage points lower than acquirers operating outside of waves. The effect showed up in operating performance as well, with two-year post-merger changes in return on assets that were 0.75 to 2.14 percentage points lower for in-wave acquirers. Duchin and Schmidt argue that merger waves are periods of reduced monitoring, where analyst forecasts get noisier, CEO turnover becomes less sensitive to performance, and managers face weaker consequences for bad deals.

The mechanism is well understood. During a wave, competition for targets drives up prices. The discipline imposed by scarce capital disappears. Marginal deals get done that would never have passed an investment committee in a tighter environment. Reputational pressure to participate, the fear of being seen as passive while competitors reshape the industry, pushes management teams into transactions they might otherwise have declined. Average deal quality falls. Returns follow.

The inverse proposition is less formalized in the academic literature but consistent with everything Harford and Duchin and Schmidt found. Deals completed in the years immediately following a period of capital tightening tend to be better than deals completed during the preceding expansion. The mechanism works in the opposite direction. Scarce financing filters out weaker acquirers. Only well-capitalized buyers with genuine strategic rationale can close transactions. Those are precisely the conditions that produce good acquisitions, even if the headline deal count is low enough that no one writes magazine articles about a boom.

A related strand of research identifies what Golubov, Yawson, and Zhang, writing in the Journal of Financial Economics in 2015, called "extraordinary acquirers." Their finding is that firm-level fixed effects explain as much of the variation in acquirer returns as all deal-specific and firm characteristics combined. Some companies are just persistently good at acquisitions, and the effect persists across CEOs and different deal structures, suggesting the skill is embedded in the organization rather than the person running it. The pattern that characterizes the top quintile is serial execution of small deals in fragmented industries, often below the scale that makes the financial press. Constellation Software is the modern template. Judges Scientific in the United Kingdom. Halma. TransDigm in its niche defense business. None of them ride waves. All of them compound through disciplined, repeated capital deployment at prices their peers cannot match.

The discipline for strategic acquirers follows from the evidence. Acquire when capital is scarce and attention is elsewhere. When the financial press runs segments on record megadeal volume, the signal is to hold fire. The deals that close in the quiet years, funded by companies that retained capacity while peers were tapped out, are the ones that compound.

The 2026 situation sits in the middle of the cycle. Q1 produced 12 completed megadeals above $10 billion, according to WTW's Quarterly Deal Performance Monitor, the highest quarterly total since the firm began tracking in 2008. That is a wave-formation signal. But the shocks cutting through the economy, tariff uncertainty, elevated funding costs, unresolved questions about what AI will do to mid-market cash flows, are discouraging rather than permissive. Capital is available in pockets and scarce in others. The acquirers with the cleanest balance sheets and the most patient shareholders should be accumulating optionality for the year when the wave clearly breaks.

---

THE PATTERN

Three Decisions, One Discipline

Three capital allocation decisions. Three separate bodies of academic research, built on different data, drawn from different time periods. One consistent finding that runs through all of them.

Divest when the business is still valuable, before it starts to decline. Buy back stock when the price is cheap and restraint feels unnatural. Acquire when capital is scarce and no one else is bidding. Each of these actions, executed correctly, looks wrong at the moment it is taken. Selling a growing business looks like a mistake. Sitting on cash rather than deploying it looks like weakness. Acquiring into a difficult macro environment looks reckless.

The discipline required is the discipline to tolerate looking wrong. For quarters. Sometimes for years. Long enough for the outcomes to validate the decision.

Benjamin Graham and Warren Buffett wrote about variations of this idea seventy years ago. The reason it keeps surfacing in new academic research is that it remains rare in practice. Most boards cannot sustain the patience required. The quarterly reporting cycle, the peer comparisons, the board members who want to see activity, the analysts who ask what you are doing about M&A, the activists who push for buybacks. All of it works against the discipline the research identifies as most valuable.

The companies that compound capital over decades have institutionalized patience. They have boards that refuse to be embarrassed into action. They have long-horizon investors. They have compensation structures that do not reward short-term stock price movement. They have leaders who understand that doing nothing is sometimes the most consequential capital allocation decision of the year.

---

Next week I will look at one specific company that has practiced all three of these disciplines well over the past decade. You have heard of the name. You probably do not think of it as a capital allocation case study. The argument will be that you should.

Until then, watch for the companies that do hard things when no one is watching. Those are the capital allocators worth owning, and worth imitating.

Have a good Friday. Have a good weekend.

— Marques

---

Marques Blank is the founder of Blank Capital (fractional CFO and FP&A advisory) and Blank Capital Partners Former Northrop Grumman and Citibank. CMA, MBA, Series 65.

This newsletter is for informational purposes only and does not constitute investment advice. Research cited: Dittmar, A.K., & Field, L.C. (2015). Can managers time the market? Evidence using repurchase price data. Journal of Financial Economics, 115(2), 261-282. · John, K., & Ofek, E. (1995). Asset sales and increase in focus. Journal of Financial Economics, 37(1), 105-126. · Harford, J. (2005). What drives merger waves? Journal of Financial Economics, 77(3), 529-560. · Duchin, R., & Schmidt, B. (2013). Riding the merger wave: Uncertainty, reduced monitoring, and bad acquisitions. Journal of Financial Economics, 107(1), 69-88. · Golubov, A., Yawson, A., & Zhang, H. (2015). Extraordinary acquirers. Journal of Financial Economics, 116(2), 314-330.