Every price change tests price elasticity, whether the seller measures it or not. A loyalty program runs more of these tests than most businesses. A member price, a double points week and a new award chart all change what something costs a member, and each one shows how hard members push back.

What is price elasticity of demand?

Price elasticity of demand shows how much buyers react when a price moves. The OpenStax text puts it as “the percentage change in the quantity demanded of a good or service divided by the percentage change in the price” (OpenStax). Price and the amount bought move in opposite ways, so the raw result is negative. By habit, economists drop the sign. A result of 0.5 means a 1% price rise cuts sales by half a percent.

The result falls into one of three groups. Above 1, demand is elastic, and a 10% price rise cuts sales by more than 10%. At exactly 1, demand is unit elastic. Below 1, demand is inelastic, and sales fall by less than the price rose (OpenStax). Sellers have a mirror measure, the price elasticity of supply. It is built the same way from the amount they offer for sale.

Two extremes sit at the edges. At zero, the amount bought does not move at all when price changes. At the other end, buyers take any amount at one price and none at all above it. The text says goods with many substitutes, and goods that eat a large share of income, are likely to have highly elastic demand. Its examples are Caribbean cruises and sports vehicles (OpenStax).

How do you calculate price elasticity?

Divide the percent change in the amount sold by the percent change in price. Most economists use the midpoint method. It measures each change against the average of the old and new values. The text gives the reason: “one obtains the same elasticity between two price points whether there is a price increase or decrease” (OpenStax).

OpenStax works one example. A price falls from $70 to $60, and sales rise from 2,800 to 3,000 units. Quantity rises 6.9% and price falls 15.4%, so elasticity is 0.45. Farther up the same straight demand curve, the answer is 1.47. The slope stays fixed while elasticity changes along the curve (OpenStax).

ElasticityNameWhat a price rise does to revenueEstimates from studies
Above 1ElasticRevenue falls, because sales drop by a larger percent than price roseRestaurant meals 2.27, premium cable TV 1.77, computers 1.44
Exactly 1Unit elasticRevenue stays the sameOff-peak rail transit 1.00
Below 1InelasticRevenue rises, because sales drop by a smaller percent than price roseHousing 0.12, electricity 0.20, gasoline 0.35, beer 0.80

The figures in the last column come from studies that OpenStax gathers in its Table 5.2 (OpenStax). The revenue column is the total revenue test, which follows from the math: revenue is price times units sold.

What makes demand elastic or inelastic?

Need and choice decide most of it. OpenStax notes that demand for necessities such as housing and electricity is inelastic, while restaurant meals are more price sensitive. By its figures, a 10% rise in the price of a restaurant meal cuts the amount bought by 22.7%. The same rise for housing cuts it by 1.2% (OpenStax).

Time matters too. “Elasticities are often lower in the short run than in the long run,” the text says (OpenStax). A driver facing dearer fuel can carpool next week. Buying a car that uses less fuel, or a home closer to work, takes years.

The level you measure at matters most. A shopper who balks at the price of gas still buys gas. A shopper who balks at one brand of cola buys another. A study pooled 1,851 price elasticities from 81 marketing studies and found an average of -2.62 (Bijmolt, van Heerde and Pieters, 2005). Those studies track sales, market share and brand choice. Next to the category figures above, the gap shows how much more a single brand is exposed to its rivals.

How does price elasticity affect revenue?

Elasticity tells a seller which way to move. If demand is elastic, a price cut raises total revenue, because sales rise by a larger percent. If demand is inelastic, a price rise raises revenue (OpenStax). Profit brings in cost as well. A price cut that lifts revenue can still lose money on a thin margin.

Paid loyalty programs run this test on their fees. Costco raised its U.S. and Canada Gold Star fee by $5 to $65, effective September 1, 2024. The Executive fee rose from $120 to $130 (Costco, July 2024). Its U.S. and Canada renewal rate was 92.9% at the end of fiscal 2024 (Costco 10-K, 2024). A year later it was 92.3%.

Costco tied the dip to members who joined online, who renew at a slightly lower rate. The fee increase drove about 40% of membership income growth in fiscal 2025 (Costco 10-K, 2025). The rate trails, counting renewals from seven to 18 months earlier, so later reports carry more of the new fee. So far, an 8% rise in the base fee has barely moved renewals. Our guide to paid loyalty programs treats renewal as the test of any fee.

How do loyalty programs measure price elasticity?

Every member offer changes a price. A member price cuts it directly. A double points week cuts it by the value of the extra points. A higher award price raises the cost of what points buy, which is why members read a devaluation as a price rise.

Program data ties each purchase to a member, so the team can measure response by segment. The clean method is a random holdout that does not get the offer. A control group shows what sales would have been anyway, so the elasticity reflects the offer alone.

Price rules also create natural tests. Uber’s surge price jumps at set points. A team of economists used those jumps, and almost 50 million records, to measure how demand reacts at several points on the curve (NBER, 2016). A program with price breaks or point thresholds has the same kind of edges to study.

For illustration, a coffee program tests a $4 member price on a $5 drink. In the test group, weekly drinks rise from 10,000 to 12,000. The holdout group also rises, from 10,000 to 10,500, because of the weather. Without the holdout, the midpoint math gives an elasticity of 0.82. Against the holdout, it is 0.60. Either way demand is inelastic, and the test group’s $48,000 trails the $52,500 it would have paid at full price.

Run the same test on each segment. If top-tier members buy at nearly the same rate with or without the discount, the discount is a gift to them. If lapsed members respond strongly, that is where the same budget earns its cost.

How GRAVTY handles price elasticity

GRAVTY puts the member side of a price test in the loyalty team’s hands. Offers, member prices and bonus rules are written in GRAVTY Visual Rules, a patented no-code rules engine. AgenticTest runs a live A/B simulation before launch. Once a test is live, it watches the lift and moves budget away from weak test arms on its own. Lift is read by segment while the campaign runs (GRAVTY for retail).

Agentic AI Compass, the AI suite in GRAVTY, also runs offer tests. It plays out the likely result before a test goes live, and it shows the reasoning behind each call it makes. Every test ends with a measured response for each group of members. That is the raw input any price elasticity estimate needs.

Frequently asked questions

Is price elasticity of demand always negative?

Yes. Price and quantity demanded move in opposite directions, so the raw number is negative (OpenStax). Economists drop the sign and report the absolute value. An elasticity of 0.45 means a 1% price rise cuts the quantity bought by 0.45%.

What is cross-price elasticity?

Cross-price elasticity is the percent change in demand for one good divided by the percent change in the price of another. Substitutes, such as coffee and tea, have positive values. Complements, such as coffee and sugar, have negative values (OpenStax). A discount on one item in a program can lift sales of its complement.

What is the difference between elasticity and the slope of a demand curve?

Slope counts the change in units, and elasticity counts the change in percent. On a straight demand curve the slope stays the same at every point. Elasticity does not. In the OpenStax example, one straight line gives 0.45 at low prices and 1.47 at high ones (OpenStax).

Why is demand more elastic in the long run?

Buyers need time to find substitutes and change habits. In the short run, a driver facing higher fuel prices can carpool or turn down the heat at home. Over years, the same person can buy a car that uses less fuel or add insulation (OpenStax).

What is an example of inelastic demand?

Cigarettes are the textbook case. The text cites research that a 10% rise in the price of cigarettes cuts how much adults smoke by about 3%. That is an elasticity of 0.3 (OpenStax). Housing, at 0.12 in its table, reacts even less.