Persistence of pasture in grazing systems has technical and economic dimensions. Profit from investment in pasture is maximised when the profit from the pasture is maximised over cycles of investments in pasture over the life of the farm business. The economic decision-rule is that an existing pasture should be replaced when the expected extra average addition to farm profit per year over the whole of the expected life of the next cycle of pasture investment exceeds the expected addition to farm profit from one more year of the existing pasture. This profit-maximising decision-rule means that the persistence of pasture is an economic phenomenon to be accounted for over several cycles of investment—a different concept to technical views that focus on the number of years of existence (i.e. persistence) of a pasture in one investment cycle. The number of years in which a pasture performs near peak potential annual dry matter (DM) production is a useful perspective on pasture persistence. The longer a pasture persists at peak level the more profitable. An empirical example was analysed of a pasture that had declined to carrying annually 6 dry sheep equivalents/ha (DSE/ha) and reinvestment occurred. The new pasture attained a peak of 11 000 kg/dry matter/ha in years 4–7, carrying an extra 15 DSE/ha.year, and declined to 50% of peak DM production by year 11, which was maintained until year 20. The modified internal rate of return for the base model of investing in pasture improvement was 12% real. The profit-maximising life of the pasture analysed was 8 years in repeated cycles over the life of the farm business. If this pasture produced at 65% of peak kg DM/ha for years 11–20, then the pasture was equally profitable whether the life of the pasture was any length from 8 to 20 years. If the pasture maintained production >65% of peak annual kg DM/ha, then longer pasture lives were more profitable than shorter lives.