Return on capital: the test of a good business

The question return on capital answers

Absolute profit is close to meaningless without knowing what was consumed to produce it. A company earning 50 million a year is impressive if it employs 200 million of capital and unremarkable if it employs 2 billion.

Return on capital puts those two facts into one ratio. In its simplest form:

Return on capital = operating profit after tax, divided by the capital employed to generate it.

The result is a percentage that answers a genuinely useful question: for every unit of money tied up in this business, how much does the business produce in a year?

This matters for three reasons.

1. **It is a test of the business, not the accounting.** Margins can be high because of the industry. Growth can be bought. A durable, high return on capital usually means the company is doing something competitors find hard to copy.
2. **It sets the value of growth.** Growth is only valuable if the capital used to fund it earns more than the capital costs. A company growing fast at low returns is destroying value while looking successful.
3. **It is comparable across very different business models,** in a way that margins are not, because it accounts for both profitability and capital intensity at once.

The balance sheet and note disclosures needed to compute it are in every company's periodic filings, which are publicly available in regulated marketsSource^source.

Defining the numerator and the denominator

There is no single official definition. Return on capital employed, return on invested capital, return on equity and return on assets are related measures that differ in what they include. That is not a flaw so much as a fact you have to manage: pick a definition, write it down, and apply it identically to every company and every year you compare.

The numerator

The most defensible numerator is operating profit after tax, sometimes called net operating profit after tax. You take operating profit and multiply by one minus the effective tax rate.

Why operating profit and not net profit? Because the denominator includes capital supplied by both lenders and shareholders, so the numerator should be the profit available to both, which is measured before interest. Mixing after-interest profit with total capital understates the return of leveraged companies for the wrong reason.

Why after tax? Because tax is a real cost and pre-tax returns overstate what is actually earned. If you use pre-tax profit, compare only with other pre-tax figures.

The denominator

Two common constructions, both defensible:

• **Capital employed** = total assets minus current liabilities. Simple, computable from the face of the balance sheet, and widely used.
• **Invested capital** = equity plus interest-bearing debt (including lease liabilities), minus surplus cash. Slightly more work, and it isolates the capital deliberately put into the business.

Two refinements worth applying consistently:

• **Use an average of opening and closing capital,** not the year-end figure, because the profit was earned across the year while the balance sheet is a snapshot at a point in time. A large acquisition in December otherwise crushes the ratio for no real reason.
• **Decide how to treat surplus cash.** A company holding a large cash pile it does not need for operations will show a depressed return if that cash is left in the denominator. Removing it isolates operating performance, but requires you to judge how much cash is genuinely surplus, which is a judgement, so state it.

Related ratios and when they are appropriate

• **Return on equity** uses net profit over shareholders' equity. It includes the effect of leverage, so it can be raised simply by borrowing more. Useful, but it mixes operating performance and financing policy into one number.
• **Return on assets** uses total assets in the denominator, including all liabilities, so it tends to sit lower and treats supplier credit as capital.
• **Return on tangible capital** excludes goodwill and acquired intangibles. It answers a different question, discussed below.

Margin times turnover: the decomposition that explains everything

Return on capital can be split into two components, and the split is where the insight lives.

Return on capital = operating margin after tax, multiplied by capital turnover.

Where capital turnover is revenue divided by capital employed.

This decomposition shows that there are two entirely different routes to a good return: earn a lot per sale, or make the same capital work more times a year.

Suppose two hypothetical companies each achieve a 20 percent return on capital. All figures are invented.

Company X, a speciality chemicals maker:

• Revenue 500, capital employed 500, so capital turnover is 1.0
• Operating profit after tax 100, so after-tax operating margin is 20 percent
• Return on capital = 20 percent multiplied by 1.0 = 20 percent

Company Y, a food distributor:

• Revenue 2,000, capital employed 500, so capital turnover is 4.0
• Operating profit after tax 100, so after-tax operating margin is 5 percent
• Return on capital = 5 percent multiplied by 4.0 = 20 percent

Identical returns, opposite economics. Company X depends on holding price and would be badly hurt by a competitor undercutting it, but its returns are not very sensitive to a modest slowdown in throughput. Company Y depends on volume and working capital discipline; a small delay in inventory turns or a slip in receivables collection hits it hard, but a competitor cannot easily undercut a 5 percent margin.

The decomposition also tells you where to look when returns change. If return on capital fell, was it the margin (pricing, cost) or the turnover (more capital tied up per unit of sales)? Those have different causes and different fixes.

Return on capital versus the cost of capital

A return on capital is only meaningful relative to what the capital costs. Both debt and equity have a cost; equity's cost is not observable and must be estimated, which is why any spread calculation is approximate.

Conceptually:

• If return on capital exceeds the cost of capital, each additional unit invested adds value.
• If it is below, growth consumes value, and the faster the company grows the more it consumes.
• If it is roughly equal, growth is financially neutral and the business is essentially a pass-through.

Suppose a hypothetical company earns 14 percent on capital while its blended cost of capital is estimated at 9 percent. The 5 percentage point spread is the value created per unit of capital per year. Now suppose a competitor earns 7 percent against the same 9 percent cost. Both companies may report growing profits. Only one of them is getting anywhere.

Cost of capital is an estimate with a wide honest range, and small changes in assumptions move it by percentage points. Treat a spread of one or two points as inside the noise. Sustained double-digit spreads across a full cycle are the ones that mean something, and even then they invite competition rather than guaranteeing continuation.

The reinvestment test

Level of return is only half the story. What matters over time is how much capital the company can deploy at that return. This is the framework to apply, and it takes three inputs, all computable from published statements.

1. **The level.** Compute return on capital for each of the last five to ten years, using one consistent definition. Note the average and the volatility, not just the latest figure.
2. **The increment.** Compute incremental return on capital: the change in operating profit after tax over a multi-year period, divided by the change in capital employed over the same period. This tells you what the newly deployed money is earning, which can be very different from the average. Use at least three years, because single-year increments are dominated by timing noise and can be wildly negative or absurdly high for trivial reasons.
3. **The reinvestment rate.** Estimate what proportion of profit is being put back into the business rather than paid out or left idle. Roughly, retained earnings plus net new capital spending as a share of operating profit after tax.

Then read the three together:

• **High level, high increment, high reinvestment.** The rarest and most powerful combination. The company is compounding capital at attractive rates. Ask what protects it, and what happens when the reinvestment opportunities run out.
• **High level, low increment, high reinvestment.** A warning. The existing business is good; the new money is not earning the same. Averages hide this for years.
• **High level, high increment, low reinvestment.** A good business with limited places to put money. Cash accumulates or is returned. Nothing wrong with it, but do not expect compounding.
• **Low level, any increment, high reinvestment.** Growth is being funded at returns that may not cover the cost of the capital. The income statement can still show rising profits throughout.

The value of the framework is that it separates "this is a good business" from "this business can get bigger while staying good", which are different claims that people routinely merge.

Where the measure breaks down

Goodwill and acquisitions

When one company buys another above book value, the excess is recorded as goodwill and sits in the denominator. Including it measures the return on all money actually spent, including the price paid for acquisitions. Excluding it measures how good the underlying operations are, ignoring what was paid.

Both views are legitimate and they answer different questions. Include goodwill when asking, "has management deployed shareholders' money well?" Exclude it when asking, "is the operating business itself strong?" A large gap between the two figures is itself informative: it means acquisitions were expensive relative to the assets acquired.

Goodwill also behaves oddly over time. It is not amortised but is tested for impairment, so it can sit unchanged for years and then drop sharply in one periodSource^source. An impairment mechanically improves return on capital in later years by shrinking the denominator, which is the opposite of what the underlying event implies. Always check whether a jump in returns followed a write-down.

Leases

Lease accounting brings right-of-use assets and lease liabilities onto the balance sheet for lesseesSource^source. That materially increases measured capital employed for retailers, restaurant groups, airlines and logistics companies. It also moves the cost from operating expense to depreciation plus interest, which raises operating profit.

Two consequences: comparisons across the transition to the current standard are not like for like, and comparisons between a company that owns its premises and one that leases them are now closer to comparable than they used to be, though not identical.

Intangible-heavy businesses

Spending on research, brand building, software and training often cannot be capitalised and is expensed as incurred. For a company whose real assets are of that kind, the balance sheet understates capital employed, and return on capital comes out flattered, sometimes absurdly so. A software or consumer brand company showing a 90 percent return on capital is usually revealing an accounting artefact rather than an economic miracle.

You can adjust by capitalising some historic research or marketing spend over an assumed life and adding it back to the denominator, but every step of that is an assumption. At minimum, note that the reported figure is inflated and avoid comparing it directly with an asset-heavy peer.

Financial institutions

For banks and insurers, capital is the raw material and is set largely by regulation, so return on capital employed as defined above does not carry the same meaning. Return on equity alongside regulatory capital ratios is the conventional framing, and the analysis is a different discipline.

Cyclicals and long project cycles

A mining, shipping or property developer's return on capital swings enormously across a cycle, and peak-year figures are not repeatable. Average across a full cycle, and be aware that assets carried at old cost can make returns look better than the economics of replacing those assets would suggest.

What return on capital does not tell you

• **It is not a valuation.** A business earning excellent returns can be priced so highly that the returns are already reflected. Return on capital is a statement about the business; price is a separate question.
• **It does not predict durability.** High returns attract competition. The ratio tells you the returns exist today, not what defends them.
• **It is backward-looking.** It measures capital deployed in the past against profit earned recently.
• **It can be gamed by shrinking.** Selling assets, outsourcing production or running down the asset base raises the ratio while the business may be weakening.
• **It says nothing about risk or solvency.** A highly leveraged company can post a strong return on capital right up to the point at which refinancing becomes a problem.
• **It is definition-sensitive.** Two analysts computing return on capital for the same company can differ by several percentage points entirely because of definitional choices.

A practical procedure

1. Choose one definition of numerator and denominator, write it down, and use it for every company and year in the comparison.
2. Compute the ratio for at least five years, using average capital employed.
3. Decompose each year into after-tax operating margin and capital turnover, and note which one moved.
4. Compute the ratio twice, once including goodwill and once excluding it, and note the gap.
5. Check the lease liability balance and whether the comparison period spans an accounting change.
6. Compute incremental return on capital over three to five years.
7. Estimate the reinvestment rate and read the three results together using the reinvestment test above.
8. Sanity check against the cash flow statement, because a high accounting return with persistently weak cash conversion is a contradiction that needs resolving before you rely on either figure.

Done consistently, this is one of the highest-value routines in company research, mostly because it forces you to hold profitability and capital intensity in view at the same time instead of praising one and ignoring the other.

This article is general financial education. It is not investment advice, not a recommendation about any company or security, and not a substitute for reading a company's own filings or seeking advice from a licensed professional about your circumstances.