You cannot say whether a firm has "market power" without first answering an unglamorous but decisive question: power over what market, exactly? M4.03's own worked material already showed you, by example, that a market can look highly concentrated on paper and still have no pricing power at all if a competing supply (imports across a border, in that case) sits just outside the measured share table (M4.03 Worked Example 5, cited here, not repeated). This section supplies the formal versions of both instruments Phase 4 used informally.
How is the HHI calculated?
Concentration measures summarize how output or revenue is divided among a market's sellers. The four-firm concentration ratio, CR4, is simply the summed market share of the four largest firms. The Herfindahl-Hirschman Index, HHI, is the sum of squared market shares (shares expressed as whole numbers, so a monopoly scores 100² = 10,000 and a market of ten equal firms at 10% each scores 10 × 10² = 1,000), squaring rewards size disproportionately, so HHI is far more sensitive than CR4 to whether the top few firms are roughly equal or one firm dominates the rest. US antitrust guidance (the DOJ/FTC Horizontal Merger Guidelines) treats HHI below 1,500 as unconcentrated, 1,500–2,500 as moderately concentrated, and above 2,500 as highly concentrated, bands that are themselves perishable policy choices (check the current guideline before citing a specific threshold in live work). The two measures can tell noticeably different stories on the same market: a market with a leading firm's share table [35, 25, 15, 10] plus five small competitors at 3% each has CR4 = 85 (a number that on its own sounds like a tight oligopoly) but an HHI of only 2,220, squarely in the moderately concentrated band, not the highly concentrated one, precisely because no single firm is close to dominant. Concentration on its own answers a structural question, never a pricing-power one, exactly the discipline M4.03's import-competition example already demonstrated with real numbers, and exactly why this module treats CR4/HHI as an input to a judgment, never the verdict itself.
Advertisement
What is the SSNIP test?
The rigorous version of Porter's own informal buyer-power question, "could the industry raise price 5% and keep its customers?" (M4.03, cited, not re-taught), is the antitrust economist's SSNIP test (a Small but Significant, Non-transitory Increase in Price, conventionally 5%), used to decide whether a candidate market is correctly defined or must be widened to include a close substitute. The mechanics: a hypothetical monopolist over the candidate market considers raising price by t% (typically 5%). The critical loss, the maximum percentage of sales the monopolist could afford to lose and still come out ahead, given a gross margin m (as a fraction, (P−MC)/P), is CL = t / (t + m). If the actual expected loss in sales (approximated by |E| × t, using the segment's own price elasticity) is smaller than the critical loss, the price increase is profitable, and the candidate market is correctly (or too broadly) defined; if the actual expected loss exceeds the critical loss, the SSNIP would not be profitable (buyers would defect to a substitute in large enough numbers to make it a bad idea) and the relevant market must be widened to include that substitute. With a proposed 5% increase and a 40% margin, the critical loss is 5/(5+40) = 11.11%: a segment with elasticity −1.6 loses only 1.6 × 5 = 8% of sales, comfortably under the 11.11% threshold, so the price increase sticks and the market is correctly defined as-is; a segment with elasticity −3.0 would lose 15%, over the threshold, meaning the true market is broader than the candidate and includes whatever substitute those defecting buyers are running to. This is the exact formal machinery (nothing more, nothing hidden) behind every regulator's market-definition exercise and every analyst's informal "would customers actually switch?" gut check, now reduced to two numbers you can compute from a margin and an elasticity.