Price elasticity

Price elasticity of demand measures how strongly quantity demanded changes when price changes, all else equal. It is one of the most useful pricing concepts—and one of the easiest to misuse. A measured response reflects a product, customer segment, market, time period and competitive context; it is not a permanent property of the product.

Demand is elastic when quantity responds proportionally more than price and inelastic when it responds proportionally less. Necessities, availability of alternatives, switching cost, budget share and time to adapt all influence the result. The elasticity of a generic category can differ sharply from the elasticity of one brand: fuel demand may change slowly at category level while drivers switch quickly between neighbouring stations.

When to use it

• Estimate how price changes may affect volume and revenue.
• Compare response across segments, channels, products or time horizons.
• Design price tests, promotion policy and capacity plans.
• Apply the model in the business situations described here.

Origins

Alfred Marshall formalised elasticity of demand in his 1890 Principles of Economics. His formulation described the responsiveness of quantity demanded to a change in price and became part of modern microeconomics. Current practice distinguishes own-price elasticity from cross-price and income elasticity and uses percentage changes so results can be compared across units.

What it is

A lower price may increase category consumption, attract customers from competitors or both. Those mechanisms should be separated. A manufacturer may not use more nuts and bolts merely because they are cheaper, yet it may switch supplier. Similarly, cheaper butter can gain share from margarine before household consumption reaches practical limits.

Elasticity usually describes a local relationship within observed conditions. Large changes can alter customer perception, competitor behaviour, distribution, capacity and the market itself. Extraordinary cases can also violate the usual downward-sloping demand pattern. The Irish potato-famine story is often presented as a 19th-century Giffen-good example: households facing poverty may buy more of a staple whose price rises because costlier foods become even less affordable. The historical evidence and attribution are debated, so it is safer to treat the Giffen effect as a theoretical possibility rather than a settled case fact.

Price is the only one of the 4Ps that directly collects value; the other 3Ps normally require expenditure. A price that is too low may leave value unclaimed, while one that is too high may destroy demand or trust. There is not one context-free “optimum” price: profit, growth, capacity, fairness, positioning and customer lifetime value can imply different objectives.

The standard calculation is:

percentage change in quantity demanded

Price elasticity =

percentage change in price

Because demand commonly falls when price rises, own-price elasticity is usually negative. Analysts often discuss its absolute magnitude, so state the convention explicitly.

At −1, the percentage changes in price and quantity are equal in magnitude; demand is unit elastic. A 10 per cent decrease in price accompanied by a 10 per cent increase in volume is an illustration, although revenue effects depend on the exact calculation method and starting point. When the magnitude is greater than −1, demand is elastic; when it is smaller than −1, demand is inelastic.

Historic examples in this article cite leisure air travel at −1.5 and Coca-Cola at −3.8. Treat these as context-dependent illustrations, not reusable constants. Industrial buying can be less responsive where specifications, qualification costs, reliability and contracts make switching difficult. The inherited “−0.” figure is incomplete and should not be used; −0.8 is a plausible illustrative value only when supported by a defined study.

Collect prices, realised net revenue, units, promotions, customer mix, competitor activity, availability and timing. A simple before-and-after comparison is weak because other conditions change. Customer awareness may be delayed, contracts may defer response and competitors may react. Use randomised or staged tests where ethical and feasible, or apply causal methods that control for confounders.

Developments of the model

When transaction history is unavailable, market research can estimate response. Gabor–Granger questions present prices and ask purchase likelihood; see the New product pricing article. The method is vulnerable to hypothetical bias and strategic answering because respondents are not making a real purchase.

Conjoint analysis can estimate trade-offs among price and product attributes, but its validity depends on design, sampling and model assumptions. Neither survey technique should be presented as observed elasticity without validation against behaviour.

Qualitative criteria can still guide hypotheses:

Low switching costs and commodity status generally increase price sensitivity. Patents, a strong brand, written specifications, scarce availability, third-party payment and valued personal service may reduce it. Each can also work differently: insurance can make the end user less price-sensitive while making a purchasing intermediary more demanding. Use the checklist to decide where evidence is needed, not to assign a number.

How to use it

A regional newspaper group facing declining profitability considered raising advertising prices. Its sales team expected any increase to destroy demand. Management surveyed retailers, entertainment businesses and other advertisers about the importance of newspaper advertising and their likelihood of buying different formats at different prices.

Three-quarters regarded advertising as important and placed newspapers high among available media. Around 25–35 per cent said they would not buy newspaper space at any price and were classified as outside the addressable segment. Among potential buyers, the estimated elasticity was −0.8, suggesting relatively inelastic demand.

The analysis implied that a 10 per cent price increase would reduce volume proportionally less, potentially maintaining revenue while improving contribution. The company raised prices, reinforced the value proposition and introduced points-based rewards for loyal, high-spending advertisers.

This is a survey-based case, not a controlled proof. Respondents’ stated intentions may differ from purchases, and profitability also depends on sales cost, capacity, advertiser outcomes and long-term retention. A stronger implementation would stage the change, define guardrails, compare suitable groups and monitor net revenue, volume, churn and customer value.

Some things to think about

• Estimate elasticity at the level where the decision is made: segment, offer, channel and time horizon. Aggregate averages can hide customers who respond in opposite ways.

• Communicating differentiated value can reduce price sensitivity, but it does not make demand permanently inelastic. Verify willingness to pay and monitor competitive response.

Top practical tip

Start with a precise price, customer, product, channel and time window. Use realised net price and causal evidence where possible, then report uncertainty rather than a universal elasticity constant.

Top pitfall

Do not infer elasticity from one uncontrolled price change or from stated willingness alone. Promotions, mix, competitors, contracts, stockouts and delayed response can all produce a misleading estimate.

Further reading

• Marshall, A. (eighteen ninety). Principles of Economics. Macmillan.
• Nagle, T.T., Hogan, J.E. and Zale, J. (twenty sixteen). The Strategy and Tactics of Pricing. Routledge.