Elasticity Explained: How Strongly Buyers React to Prices and Income
- ▸Price elasticity of demand is the percentage change in quantity demanded divided by the percentage change in price; when its absolute value is above 1, demand is elastic and a price rise lowers total spending.
- ▸HMRC's 2014 estimates for UK alcohol range from −0.08 (wine bought in shops) to −1.25 (spirits in pubs and bars), so a 10% price rise cuts quantity by under 1% for the first and by about 11% for the second, assuming a constant elasticity and full pass-through to prices.
- ▸Elasticities decide who pays a tax: in an illustrative market with supply elasticity 0.5 and demand elasticity −1.5, buyers bear a quarter of a per-unit tax and sellers three quarters; if supply elasticity were 1.5, the split would be even.
Understanding elasticity — simply put
Elasticity measures how strongly one quantity responds when another changes, in percentage terms. Price elasticity of demand answers a question every seller and every finance ministry asks: if the price goes up by 1%, by what percentage does the quantity people buy go down? Because it compares percentages, the answer does not depend on whether prices are in euros or rupees, or quantities in liters or crates. That makes elasticity the standard way to compare how sensitive buyers are across goods, countries and time.
How does it actually help?
Take a café that raises the price of a coffee from €4 to €5 and finds that daily sales fall from 100 cups to 80. The price elasticity of demand is
where is the price elasticity of demand, is the percentage change in quantity demanded and is the percentage change in price. Measuring each change against the average of the old and new values (the midpoint method), sales fell by 22.2% and the price rose by 22.2%, so over that range. The minus sign says quantity moved opposite to price; the size, 1, says the two moved in equal proportion. Revenue was €400 before and €400 after, which is exactly what a unit elasticity implies.
Why does this matter?
Elasticity tells a firm whether a price rise will raise or lower its revenue, and it tells a government how much a tax will raise and who will end up paying it. HM Revenue and Customs uses estimated elasticities to forecast how alcohol duty receipts respond to a change in duty rates, because a higher duty shrinks the tax base. Central banks watch the elasticity of imports to the exchange rate, and health ministries use the elasticity of tobacco and sugary-drink demand to predict how far a tax will cut consumption.
What makes an elasticity estimate reliable?
An estimate is only as good as the price variation behind it. If prices rose because demand was rising, a naive comparison mixes the demand curve with the supply curve and understates how sensitive buyers really are. Reliable estimates use price changes that come from the supply side or from policy, such as a tax change, and hold income and the prices of other goods constant. The conditions are set out under "The assumptions" below.
The takeaway
An elasticity is a ratio of two percentage changes. Computing it by hand, reading real estimates, and linking it to revenue and to who pays a tax all build on that one idea.
The four elasticities you will meet
Price elasticity of demand compares the change in quantity demanded with the change in the good's own price. It is negative for almost every good. Economists often quote its absolute value, , and classify demand as:
- elastic when (quantity reacts more than proportionally),
- unit elastic when ,
- inelastic when (quantity reacts less than proportionally),
- perfectly inelastic at 0 and perfectly elastic as .
Income elasticity of demand is , where is income. A good is normal when , a necessity when , a luxury when and inferior when . With , a 4% rise in income raises the quantity bought by about 6%.
Cross-price elasticity of demand is , the response of good to the price of good . It is positive for substitutes (a higher train fare raises bus travel) and negative for complements (pricier printers reduce ink sales). If is train journeys and is the bus fare, means a 10% rise in the bus fare raises train journeys by about 3%.
Price elasticity of supply is , where is quantity supplied. It is usually positive and larger in the long run, when firms can build capacity.
A hand-worked example
The café numbers above are illustrative. Price rises from to and quantity falls from to .
Step 1 — the problem with simple percentages. Measured from the starting point, quantity falls by and price rises by , giving . Measured from the end point, quantity rises by when price falls by , giving . The same two points give two answers, depending on direction.
Step 2 — the midpoint (arc) formula. Divide each change by the average of the two values:
Step 3 — divide. . The arc elasticity gives the same answer in both directions.
Step 4 — check with revenue. Total revenue is : before and after. Revenue is unchanged, which is what unit elasticity predicts: the fall in quantity exactly offsets the rise in price.
Step 5 — point elasticity on a demand curve. Suppose the café's demand curve is the straight line , which passes through both points. The slope means each extra euro on the price loses 20 cups a day. Point elasticity is :
- at , : (inelastic);
- at , : (elastic).
The slope is the same everywhere on a straight line, but elasticity is not: it grows in absolute value as price rises and quantity falls. Slope and elasticity are different things. The two point values match the two one-sided answers from Step 1 because the line was drawn through the observed points, so they are not an independent check. The arc value of −1 describes the whole move from 4 to 5; at the single price of 4, demand on this line is inelastic.
Step 6 — income and cross-price elasticities work the same way. Monthly income rises from 40,000 to 44,000 and restaurant meals rise from 10 to 12 a month. Midpoint changes: for meals and for income, so : restaurant meals are a luxury for this household. In another case the bus fare () rises from 2.00 to 2.50 and train journeys () rise from 100 to 110. Midpoint changes: and , so : the two are substitutes. All numbers are illustrative.
Worked example: UK alcohol and the duty base
HM Revenue and Customs publishes the price elasticities it uses to cost alcohol duty changes. Working Paper 16 (Sousa, December 2014) estimates them from household spending data in the Living Costs and Food Survey for 2007–2012, separately for the on-trade (pubs, bars and restaurants) and the off-trade (shops and supermarkets). The paper states that own- and cross-price elasticities "form the basis of behavioural costings."
The table applies each estimate to a 10% rise in the shelf price, assuming the elasticity stays the same across the whole 10% move (constant-elasticity demand): the new quantity is times the old one, and spending is times the old spending. The simple approximation gives slightly larger falls, for example −12.5% instead of −11.2% for spirits in pubs and bars.
| Drink and channel | Own-price elasticity | Change in quantity after a 10% price rise | Change in consumer spending |
|---|---|---|---|
| Spirits, on-trade | -1.25 | -11.2% | -2.3% |
| Beer, off-trade | -0.74 | -6.8% | +2.5% |
| Cider, off-trade | -0.74 | -6.8% | +2.5% |
| Ready-to-drinks, off-trade | -0.52 | -4.8% | +4.7% |
| Cider, on-trade | -0.49 | -4.6% | +5.0% |
| Spirits, off-trade | -0.45 | -4.2% | +5.4% |
| Beer, on-trade | -0.34 | -3.2% | +6.5% |
| Wine, on-trade | -0.24 | -2.3% | +7.5% |
| Ready-to-drinks, on-trade | -0.24 | -2.3% | +7.5% |
| Wine, off-trade | -0.08 | -0.8% | +9.2% |
How to read it. Only one category, spirits in pubs and bars, is elastic: there a 10% pricier drink cuts the quantity bought by about 11% and total spending falls. Every other category is inelastic, so spending rises when the price rises. Wine bought in shops barely responds at all: a 10% price rise cuts quantity by less than 1%. For beer, cider and ready-to-drinks, the shop version is more price-sensitive than the pub version, which fits the idea that shoppers can switch brands and stores more easily than someone already at the bar. Wine and spirits go the other way, which shows that such patterns have to be measured, not assumed.
From prices to tax. A duty rise is not the same as a 10% price rise, and the model above is not the same as HMRC's costing. UK alcohol duty is charged per unit of alcohol, with VAT added on top, so the same duty rise is a larger percentage of the price of a cheap supermarket can than of a pub pint. Retailers may pass on less than the duty rise, or more of it where competition is weak. HMRC's costings also use cross-price elasticities, because a pricier beer pushes some drinkers toward cider or wine. The table shows the size of the first-round response to a given change in the price consumers pay, which is the input those costings start from.
How sensitive are the numbers? A different study of the same survey covering 2001/2–2009 (Meng and coauthors, Journal of Health Economics, 2014) follows groups of households defined by birth year, gender and socioeconomic status over time (a pseudo-panel). It finds off-trade beer at −0.98 and off-trade cider at −1.27, both more elastic than HMRC's −0.74, and off-trade spirits at −0.08, far less elastic than HMRC's −0.45. With Meng's figures, a 10% price rise cuts off-trade cider by 11.4% instead of 6.8%, and spending on it falls by 2.5% instead of rising by 2.5%. For off-trade beer, spending still rises, but by about 0.2% instead of 2.5%, close to unit elasticity. Even the ordering changes: off-trade spirits is mid-table in one study and the least elastic category in the other. Treat any single estimate as a central value with a wide range around it.
Elasticity and revenue
For a firm or a tax authority the key link is between elasticity and total revenue . Marginal revenue, the extra revenue from selling one more unit, is
where is negative. Applied to the café's straight-line demand at , where the point elasticity is : euro per extra cup, so selling one more cup a day by cutting the price would lower revenue by about a euro. The rules that follow:
- when demand is elastic, a price cut raises revenue and a price rise lowers it;
- when demand is inelastic, a price rise raises revenue;
- revenue is at its maximum where .
This is why a profit-maximizing monopolist never sets its price where demand is inelastic: there, a higher price would both raise revenue and save on production costs.
Who pays a tax: incidence
Elasticities decide how the burden of a per-unit tax is split between buyers and sellers, whoever legally hands the money to the government. For small taxes, the share paid by buyers through a higher price is
and sellers bear the rest, , with both elasticities measured at the market price before the tax. "Small" means a tax low enough that the elasticities barely change over the price range it covers; for straight-line curves the formula is exact. The side of the market that is less able to walk away pays more. If demand is perfectly inelastic, buyers pay the whole tax; if supply is perfectly inelastic, as with land, sellers pay it all. With the café, if supply elasticity were 0.8 at the starting point, buyers and sellers would split a small tax equally, because .
A full worked case. Take demand and supply (illustrative). Without a tax, gives and . At that point and , so buyers should bear of a tax. Add a tax of 5 per unit, so buyers pay . Then gives , and . Buyers pay 3 more (60% of the tax) and sellers keep 2 less (40%). The government collects . Because 6 units a day are no longer traded, there is a deadweight loss of : value that neither side nor the government receives. The more elastic demand or supply is, the more trade a tax destroys, which is why taxes raise money most efficiently on goods with inelastic demand, such as alcohol and tobacco.
What makes demand more or less elastic
- Substitutes. The more close substitutes a good has, the more elastic its demand. One brand of beer is far more elastic than beer as a whole.
- Share of the budget. Goods that take a large share of income draw a stronger response than small purchases.
- Necessity or luxury. Essentials such as basic food and heating fuel are inelastic.
- Time. Demand is more elastic in the long run, when households can change habits, appliances or cars.
- Definition of the market. Narrowly defined goods have more substitutes and so higher elasticities.
The assumptions
Other things held equal. The elasticity measures the response to price alone, with income, other prices and tastes fixed. Why it matters: if incomes rose at the same time as the price, part of the change in quantity is an income effect, and the measured "price" elasticity is biased toward zero for a normal good (and away from zero for an inferior good).
The price change is not caused by demand. The price variation used must come from supply shifts or policy, not from changes in demand. Why it matters: when demand and price rise together, a simple regression of quantity on price traces a mix of the supply and demand curves and can even produce a positive "demand" elasticity. This is the identification problem that instrumental variables are designed to solve.
Small changes for point elasticity. The point formula describes the response to a small price change at one point on the curve. Why it matters: for large changes on a curved or straight demand curve the elasticity itself changes along the way, as the café example showed; use the arc formula or a constant-elasticity model instead.
A stable demand relationship. Estimates from one period are applied to another. Why it matters: HMRC's elasticities come from 2007–2012 data; if habits or the range of drinks on offer have since changed, the estimates may no longer describe today's buyers.
The formula
The derivations below are optional on a first reading.
Point elasticity. For a demand function , define
The second form follows because and . It shows why a log-log regression estimates an elasticity directly: the coefficient is . Applied to the café's straight-line demand at : .
Elasticity along a straight line. For with ,
which is 0 at , equals at the midpoint and tends to as . For the café, and , so unit elasticity occurs at , where revenue is largest at .
Constant-elasticity demand. The function , with and , has the same elasticity at every price: . A change of price by the factor multiplies quantity by . This is the form used for the alcohol table: with and , quantity falls to of its former level, an 11.2% fall.
Marginal revenue. With and inverse demand ,
Tax incidence. Let demand be and supply , where a per-unit tax drives a wedge between the price buyers pay and the price sellers receive. Write and for the slopes and , evaluated at the no-tax equilibrium, where , and . Differentiating the market-clearing condition with respect to gives
Multiplying the numerator and denominator by at the initial equilibrium turns slopes into elasticities: , which equals because .
Convention. EconoLens keeps the minus sign on price elasticity of demand and writes when the size is meant. Many textbooks define demand elasticity as a positive number by adding a minus sign to the formula; under that convention the marginal revenue formula becomes .
Aggregation rules. Because a household's spending must add up to its income, the budget-share-weighted income elasticities satisfy (Engel aggregation), where is the share of the budget spent on good . Not every good can be a luxury.
When this isn't the right tool
- Large or permanent changes. Elasticities describe responses near current prices. For a large tax or a structural change, model the whole demand curve or use general equilibrium reasoning.
- Prices set together with demand. When price and quantity are both driven by demand shifts, a simple comparison is biased; use instrumental variables or a policy change as a natural experiment, analyzed with difference-in-differences.
- Responses that build up over time. Short-run and long-run elasticities differ. Dynamic models from time series analysis estimate both.
- Many related goods. When substitution between goods matters, as with the alcohol duty, single elasticities are not enough; a full demand system with cross-price terms, grounded in consumer theory, is needed.
Practice problems
Problem 1. A bus company raises the single fare from 8 to 12 and daily journeys fall from 50 to 30 (thousands). Using the midpoint formula, find the price elasticity of demand. Did fare revenue rise or fall?
Solution. . . , so demand is elastic. Fare revenue was thousand a day and is now thousand, a fall of 40 thousand, as elastic demand predicts. Check: recompute from the other direction, a fall from 12 to 8, and the midpoint formula gives the same −1.25.
Problem 2. A regression on monthly data gives . What is the price elasticity of demand? By how much does quantity change if price rises by 5%? Compare the linear approximation with the exact constant-elasticity answer.
Solution. In a log-log model the coefficient on is the elasticity, provided the price variation behind it is not driven by demand: . The linear approximation gives . The exact answer is . The two differ by 0.4 percentage points; the approximation is close for small changes and worsens as the change grows.
Problem 3. In a market, the price elasticity of supply is 0.5 and the price elasticity of demand is −1.5. The government adds a tax of 2 per unit. Roughly how much of the tax do buyers pay through a higher price, and how much do sellers pay through a lower net price? Explain the result in one sentence.
Solution. Buyers' share , so the price buyers pay rises by about . Sellers bear the remaining 75%: the price they keep falls by about 1.5. Check with the sellers' formula: , and . Buyers pay less because their demand is more elastic: they can switch away more easily than sellers can cut back supply.
Further reading
- Supply and demand explained — internal
- Consumer theory and duality explained — internal
- OLS regression explained — internal
- Instrumental variables explained — internal
- Difference-in-differences explained — internal
- Time series analysis explained — internal
- General equilibrium explained — internal
- HMRC Working Paper 16: Estimation of price elasticities of demand for alcohol in the United Kingdom (December 2014) — external reference
- Meng et al. (2014), Estimation of own and cross price elasticities of alcohol demand in the UK, Journal of Health Economics 34 — external reference
Sources
- HM Revenue and Customs, Estimation of price elasticities of demand for alcohol in the United Kingdom, Working Paper 16, J. Sousa, December 2014, Table 5A (own-price elasticities, Living Costs and Food Survey 2007–2012). https://assets.publishing.service.gov.uk/media/5a7db32b40f0b65d88633e1c/HMRC_WorkingPaper_16_Alcohol_elasticities_final.pdf
- Meng, Y., Brennan, A., Purshouse, R., Hill-McManus, D., Angus, C., Holmes, J. and Meier, P. S. (2014), "Estimation of own and cross price elasticities of alcohol demand in the UK — A pseudo-panel approach using the Living Costs and Food Survey 2001–2009", Journal of Health Economics 34, 96–103. https://eprints.whiterose.ac.uk/78871/
- The café, bus, income, cross-price, tax and regression numbers are illustrative. Changes in quantity and spending in the alcohol table are EconoLens calculations from the published elasticities.
Frequently Asked Questions
What is price elasticity of demand?
Price elasticity of demand is the percentage change in the quantity of a good that buyers demand divided by the percentage change in its price. A value of −2 means a 1% price rise reduces quantity demanded by about 2%. It is negative for almost all goods because people buy less when prices rise.
How do you calculate price elasticity of demand?
Divide the percentage change in quantity by the percentage change in price. For two observed points, use the midpoint method: divide each change by the average of the old and new values, so the answer is the same whichever direction you measure. For a demand function, multiply the slope dQ/dP by P/Q at the point of interest.
What does it mean if demand is inelastic?
Demand is inelastic when the absolute value of the price elasticity is below 1, so quantity changes proportionally less than price. A price rise then increases total spending on the good. Goods with few close substitutes, such as basic fuel, are typical; wine bought in UK shops, estimated by HMRC at −0.08, is an extreme case.
What is the difference between slope and elasticity?
Slope measures the change in quantity per unit change in price, so it depends on the units used. Elasticity measures the percentage change in quantity per percentage change in price, so it has no units. Along a straight-line demand curve the slope is constant, but elasticity varies from zero at the quantity axis to very large near the price axis.
What is income elasticity of demand and how is it used?
Income elasticity of demand is the percentage change in quantity demanded divided by the percentage change in income. Goods with income elasticity above 1 are luxuries, between 0 and 1 are necessities, and below 0 are inferior goods. Businesses and governments use it to predict how demand will grow as incomes rise.
What does a positive cross-price elasticity mean?
A positive cross-price elasticity means that when the price of one good rises, demand for the other good rises too, so the two goods are substitutes, such as train and bus travel. A negative value means they are complements, used together, such as printers and ink.
How does elasticity determine who pays a tax?
The burden of a per-unit tax falls more heavily on the side of the market that responds less to price. The share paid by buyers is roughly the supply elasticity divided by the sum of the supply elasticity and the absolute demand elasticity. If demand is very inelastic, buyers pay most of the tax through higher prices.
Primary Sources
Cite This Article
EconoLens. (2026, November 16). Elasticity Explained: How Strongly Buyers React to Prices and Income. EconoLens. https://www.econolens.co.in/news/study-elasticity-explained