Owner earnings: separating the spending that keeps you competitive

What free cash flow leaves out

Free cash flow is operating cash minus capital expenditure, and the subtraction is deliberately blunt: every dollar of capex is treated as money the business had no choice about. Most of the time that is close enough. For a company in the middle of building something, it is badly wrong.

Consider a business with 300 of operating cash flow and 220 of capital expenditure. Reported free cash flow is 80. Now split the capex: 110 replaces worn equipment and keeps the existing operation competitive, and 110 builds a second facility that does not yet earn anything. The existing business is producing 190, and the owner has chosen to reinvest 110 of it.

80 and 190 are both correct answers to two different questions. Only one of them is the earning power you should value.

The direction of the error matters: this convention penalises exactly the companies that are reinvesting into a durable advantage, and flatters the ones that have quietly stopped spending. A business that defers maintenance shows strong free cash flow for several years and then an unpleasant catch-up.

The formula

Three of the four terms are read straight off the statements. Net income and depreciation are given. Working capital change is given. Only maintenance capital expenditure has to be estimated, which is the honest statement of where the uncertainty is: this is a cleaner concept than free cash flow and a less precise number.

Maintaining operations vs. maintaining competitiveness

The word "maintenance" is where most of the mistakes live, because it has a narrow reading and a correct one.

• • Maintaining operations: replace what breaks. A truck wears out, you buy a truck. This is the reading that depreciation roughly captures.
• • Maintaining competitiveness: stay as good as the alternatives. If competitors have moved to equipment that costs 40% less per unit, replacing your old truck with the same old truck leaves you with the same operation and a worse business.

Competitiveness is the one that counts, and it is why maintenance capex is usually higher than depreciation rather than equal to it. Depreciation is a historical cost spread over time; staying competitive is priced at today’s cost of today’s standard. In an industry where the standard moves fast, the gap between the two is the whole story.

Three ways to estimate maintenance capex

None of these is right. Pick one, state which you used, and keep using the same one so your figures stay comparable across companies.

1. Use depreciation as the proxy. Fast, available for every company, and systematically too low in anything where technology or scale moves. Best treated as a floor rather than an estimate.
2. Take the lowest capex year of the last decade. The logic is that in the year a company spent least, it spent close to the minimum it could get away with. Clean and crude: it can land on a year where maintenance was simply deferred.
3. Scale capex to revenue. Compute capex as a percentage of sales across ten years and take the lower end of the range as the maintenance rate. This is the most defensible for a business that has grown a lot, because it keeps the capex figure proportional to the size of the thing being maintained.

If management discusses maintenance and growth spending separately in the annual report, use their split and say so. It is the only figure in this whole exercise that comes from someone who can see the capital budget.