Once revenue is forecast, costs determine margin, which determines operating income, which is the next driver of free cash flow. The temptation to assume gentle margin expansion every year is almost universal — and almost always wrong unless defended by a specific operational story.
Three levels of margin
- Gross margin — direct cost of producing the good or service. Set by competitive intensity, input costs, and pricing power.
- Operating margin — gross margin minus SG&A and R&D. Set by operating leverage and scale.
- EBITDA margin — operating margin plus D&A. Useful for cross-company comparison when capital intensity differs.
The operating-leverage trap
Operating leverage means fixed costs are a larger share of total costs, so revenue growth flows disproportionately to operating profit. Analysts often assume operating margins expand mechanically with revenue, modelling operating leverage as if every dollar of incremental revenue drops to EBIT at a 60-70% rate. In practice, real operating leverage is much smaller because: (1) growth requires real incremental investment in sales, marketing, and infrastructure; (2) wage costs creep in line with inflation; (3) competition forces price reinvestment.
Cost categories that move with revenue, and ones that don't
Variable costs (raw materials, sales commissions, shipping) move proportionally with revenue. Semi-variable costs (R&D, marketing, partial labour) move with revenue but with a lag and at a slower pace. Fixed costs (rent, senior management, core IT) do not move with revenue in the short run but do scale up in step changes (a new building, a new region, a new product line).
The margin-expansion default is suspicious
Look at any sell-side DCF and you will see margins expand 20-40 basis points a year through the forecast. This is the consensus default. Most of the time it is unsupported. If you assume margin expansion, write a sentence explaining the operational lever that delivers it: 'Mix shifts toward higher-margin product X.' 'Negotiated supplier price reduction at scale.' 'SG&A grows at half the pace of revenue.' If you can't write the sentence, the margin should be flat.
Tax rate
Use the marginal tax rate for forecasting, not the historical effective rate. Effective rates are distorted by one-time items, deferred tax movements, jurisdictional mix, and tax holidays. The marginal rate is the rate the business will pay on the next shilling of profit — that is what a forward-looking DCF needs. For Kenya, the marginal corporate rate is 30% for resident companies, 37.5% for branches of foreign companies, with some sector-specific exceptions.
Exercise
A Kenyan FMCG company has gross margin 38%, operating margin 12%, revenue KES 20bn. The sell-side consensus forecast has revenue growing 11% per year and operating margin expanding from 12% to 14.5% over 5 years — a 250 bps lift. (1) Default position: should you trust the margin expansion? (2) What three operational levers could justify it, and what evidence would you look for to confirm each? (3) If you cannot find evidence, what should you do to the forecast? (4) Walk through how a 250 bps margin error compounds in the DCF answer.