Capital Allocation Is the CEO's Real Job
Strategy, culture, and hiring matter, but the single decision that most determines whether a CEO creates or destroys value over a decade is narrower, deciding what to do with the cash the business generates.
The Decision That Matters More Than Strategy
Ask a room of business students what a CEO does and you will hear about strategy, culture, hiring, and vision. All of that matters, but the single decision that most determines whether a CEO creates or destroys shareholder value over a decade is narrower and less glamorous, capital allocation, the process of deciding what to do with the cash a business generates after it has funded its own operations. Warren Buffett has written for decades that capital allocation is the most important job of any CEO, and that most executives are promoted into the role having spent their careers in operations, sales, or engineering, with no real training in how to think about deploying capital. A company that grows revenue brilliantly but allocates its cash poorly can still destroy value for shareholders. A company with mediocre growth but disciplined capital allocation can compound value for decades.
The Menu of Options
Every dollar of free cash flow a company generates has to go somewhere, and there are really only five places it can go. Reinvest in the existing business, new stores, new factories, more salespeople, research and development. Acquire another company. Pay down debt. Pay a dividend to shareholders. Buy back the company's own stock. A CEO's job is to rank these five options by expected return and risk, every year, and allocate cash accordingly, rather than defaulting to whichever option is most habitual or most comfortable. The mistake many companies make is treating capital allocation as an afterthought, an annual routine of paying the same dividend as last year and reinvesting whatever is left, instead of an active decision made with the same rigor applied to any other major business choice.
Reinvestment, the Default and Its Limits
Reinvesting in the core business is usually the first and best option, when a company genuinely has high return projects available. A retailer opening new stores that generate a 25 percent return on the capital invested in them should keep opening stores as long as that return holds and there is market left to enter. The problem is that this opportunity does not last forever. Every market eventually saturates, every factory eventually hits its ceiling of demand, and the marginal new store or factory earns a lower return than the last one. A disciplined CEO recognizes when reinvestment has crossed from a great use of capital to a mediocre one, and that recognition is exactly where most of the interesting capital allocation decisions happen, because admitting the core business cannot absorb any more capital at a good return is an uncomfortable thing for a growth oriented management team to say out loud.
Dividends vs Buybacks
Once reinvestment opportunities run out, cash typically returns to shareholders in one of two forms. A dividend is a direct cash payment to every shareholder, a recurring commitment that the market expects a company to maintain or grow once it starts, which makes it a signal of confidence but also a rigid one, since cutting a dividend is read by the market as a serious warning sign. A buyback, formally a share repurchase, is when a company uses cash to buy its own stock on the open market and retire it, which reduces the number of shares outstanding and increases each remaining shareholder's percentage ownership of the company. Buybacks are more flexible than dividends, a company can pause them with no signal sent, and they are tax efficient for shareholders in many jurisdictions since a buyback is not a taxable event the way a dividend payment is. The catch is that a buyback only creates value if the stock is bought below its intrinsic worth. A company buying back overvalued stock is destroying value for the shareholders who do not sell, exactly the same mistake as overpaying for an acquisition, just executed one share at a time.
A buyback is not automatically good and a dividend is not automatically safe. Both are just ways of returning cash, and the only question that actually matters is whether the price paid, for a share or for an acquisition, is below what it is worth.
Debt Paydown and M&A
The remaining two options round out the menu. Paying down debt reduces financial risk and interest expense, and it is usually the right move when a company is more leveraged than its business can safely support, but paying down debt that is already cheap and manageable, instead of reinvesting or returning cash, can be a low return use of capital that shareholders would rather see deployed elsewhere. Acquisitions can create real value when a company buys a business that is worth more inside the acquirer, because of cost savings, distribution, or technology it could not build itself, than it is worth as a standalone company, and can pay a price below that combined value. The historical track record of corporate acquisitions is not encouraging, with a large share of major deals failing to earn back the premium paid, which is exactly why the best capital allocators treat acquisitions as one option among five, evaluated against the same required return as every other use of cash, rather than the default move whenever cash builds up on the balance sheet.
How the Menu Gets Ranked
| Use of cash | Best used when |
|---|---|
| Reinvest in the business | expected return clearly beats the cost of capital |
| Acquire another company | price paid is below the value the deal creates |
| Pay down debt | leverage is a real risk to the business |
| Pay a dividend | cash generation is durable and predictable |
| Buy back stock | the stock trades below its intrinsic value |
The Bottom Line
A CEO who cannot explain, in plain language, why the company's cash is going where it is going, reinvestment, acquisitions, debt paydown, dividends, or buybacks, and why that ranking beats the alternatives, is not really doing the most important part of the job. Everything else a CEO does eventually shows up in how well that decision gets made.