Return on capital
Profit tells you how much a business made. Return on capital tells you how much it had to tie up to make it — which is the number that decides whether growth is worth anything.
Two companies each earn $100m a year. The first needed $200m of factories, inventory and cash to do it. The second needed $2bn. They are not remotely the same business, and the income statement alone will never tell you so.
The general shape
Every return measure is a profit figure divided by the capital that produced it. The versions differ in which profit and which capital.
- Return on equity (ROE) — net profit ÷ shareholders' equity. What owners earned on what they own. Flattered by debt.
- Return on assets (ROA) — net profit ÷ total assets. Ignores how the assets were financed, which is sometimes what you want.
- Return on invested capital (ROIC) — operating profit after tax ÷ (debt + equity − surplus cash). The most useful of the three, because it asks what the operating business earns on all the money committed to it, however that money was raised.
Operating profit $250m, tax rate 21%, so after-tax operating profit is $197.5m. Debt $600m, equity $900m, cash held beyond operating needs $200m. Invested capital = 600 + 900 − 200 = $1,300m. ROIC = 197.5 ÷ 1,300 = 15.2%. The business earns just over fifteen cents a year on every dollar committed to it.
Why the hurdle matters more than the level
A 15% return means nothing on its own. It has to be compared with what the money costs — the blended cost of the company's debt and the return equity holders require for the risk. That blend is the cost of capital.
A company earning 15% on capital that costs 8% creates value with every dollar it reinvests, and should reinvest as much as it can. A company earning 5% on capital that costs 8% destroys value with every dollar it reinvests, and its growth is actively making shareholders poorer. The second company will still report rising revenue and rising profit, and analysts will still describe it as growing.
Why high returns usually do not last
A business earning far above its cost of capital is advertising an opportunity. Competitors enter, prices fall, returns compress toward the cost of capital. The interesting question about any high-return company is what stops that happening — a brand, a network, a patent, a regulatory licence, a switching cost. If you cannot name the mechanism, assume the returns are temporary.
Where the measure misleads
- Old assets. Capital is recorded at cost less depreciation. A company with fully depreciated factories shows a small denominator and a spectacular return, while a competitor that just built the same factories looks poor.
- Acquisitions. Goodwill sits in invested capital, so an acquisitive company's returns look worse than an identical company that built the same assets. That is usually the correct answer, because the acquirer really did spend the money.
- Intangible-heavy businesses. Research and brand-building are expensed, not capitalised, so the assets they create never appear. Reported capital is tiny and reported returns are enormous. Software and consumer brands routinely show ROIC figures that are arithmetically correct and economically overstated.
- Leverage on ROE. Borrowing raises return on equity whenever the borrowing cost is below the asset return. This is why ROE is the weakest of the three and the most quoted.
The one sentence worth remembering
A business that compounds capital at a high rate for a long time is worth a great deal; a business that compounds capital at a low rate is worth roughly its assets, however fast revenue grows. Almost every long-term valuation argument is a disagreement about which one you are looking at. See valuation methods.