Inflation is not “things are expensive”. It is a sustained rise in the average price level, and measuring it means building an imaginary shopping trolley, filling it with what a typical household buys, and pricing that same trolley month after month. Once you can see the trolley, the calculation stops being scary.
📚 What you need to know
Inflation is a sustained rise in the average price level. Deflation is a sustained fall. Disinflation is prices still rising, but more slowly.
The consumer price index (CPI) tracks the price of a fixed basket of goods and services bought by a typical household.
Items are weighted by how much of household spending they account for, so a 10% rise in rent matters far more than a 10% rise in shoelaces.
The index is set to 100 in the base year. An index of 106.25 means prices are 6.25% above base-year prices.
Inflation rate = percentage change in the CPI between two periods, not the index number itself.
Most central banks target around 2% — low, positive and predictable.
The CPI is an average and has real weaknesses: it ignores regional and household differences, quality changes and substitution.
Building the basket
Statisticians run a household expenditure survey to find out what people actually buy. Each category then gets a weight equal to its share of typical spending. Every month, prices are collected for hundreds of specific items in shops across the country and averaged.
Housing carries almost three times the weight of clothing, so a small rise in rent moves the CPI more than a large fall in the price of jeans.
The two steps of the calculation
CPI = Σ( weight × price index ) ÷ Σ weights
Inflation rate = ( CPInew − CPIold) ÷ CPIold × 100
WORKED EXAMPLE
Using the basket above, calculate the CPI for the current year and the rate of inflation since the base year. [3]
Step 1: multiply each weight by its price index35 × 108 = 3,78020 × 112 = 2,24018 × 103 = 1,85415 × 105 = 1,57512 × 98 = 1,176Step 2: add them and divide by total weightTotal = 10,625; weights add to 100CPI = 10,625 ÷ 100 = 106.25Step 3: turn the index into a rateBase year index = 100, so inflation = 106.25 − 100 = 6.25%CPI = 106.25, inflation = 6.25%The shortcut of subtracting 100 only works when you are comparing with the base year. Any other pair of years needs the full percentage change.
WORKED EXAMPLE
The next year the CPI is 110.5. Calculate that year’s inflation rate and comment. [3]
Step 1: percentage change, not subtraction(110.5 − 106.25) ÷ 106.25 × 100= 4.25 ÷ 106.25 × 100 = 4.00%Step 2: compare with last year6.25% then 4.00%Inflation = 4.00%: this is disinflationPrices are still higher than ever — the index rose from 106.25 to 110.5. Only the speed of the rise has slowed.
The index goes up. The rate goes down. Both are true.
This is where most students lose marks, so it is worth staring at. The index tells you the level of prices. The inflation rate tells you how fast that level is changing. An economy can have a rising index and a falling inflation rate at the same time — that is exactly what disinflation is.
Deflation would need the left-hand line to slope downwards. A falling bar chart on the right is only disinflation.
If an extract says inflation fell from 8% to 1.5%, do not write that prices fell. Write that prices continued to rise, but at a slower rate, so the cost of living was still higher at the end of the period than at the start.
Why 2% and not 0%?
A zero target sounds tidier, so why do central banks aim slightly above it?
Room to cut. With some inflation in the system, a central bank can push the real interest rate below zero when it needs to stimulate demand.
A safety margin against deflation. Aiming at zero means half the misses land in deflation, which is much harder to escape.
Wages adjust more easily. Firms find it very hard to cut cash wages, but with 2% inflation a frozen wage quietly falls in real terms, which helps labour markets clear.
Mild inflation signals healthy demand. Prices creeping up usually means firms are selling and hiring.
What the CPI cannot see
Every evaluation question about inflation data wants you to question the number. Here are the arguments that carry weight.
Limitation
What goes wrong
It is an average household
The basket reflects typical spending. A pensioner spending 40% of income on heating faces a very different inflation rate from a student with no heating bill.
It ignores regional differences
Housing costs can rise far faster in a capital city than in a rural region, but only one national figure is published.
Quality changes are hard to price
If a laptop costs the same as last year but is twice as fast, its real price has fallen. Adjusting for this is difficult and imprecise.
The basket lags behind behaviour
Weights are usually revised once a year, so a sudden switch in what households buy shows up late.
Substitution bias
When beef gets dearer, people buy chicken. A fixed basket keeps pricing the beef, so it overstates the rise in the cost of living.
Sampling and honesty
The expenditure survey covers thousands of households out of millions, and respondents have little incentive to record spending carefully.
Different countries, different indices
Some use a retail price index that includes housing costs differently, which makes international comparison less reliable.
Core inflation strips out food and energy, whose prices swing wildly for reasons outside the country’s control. Central banks watch it because it shows the underlying trend rather than one bad harvest or a spike in oil. Mentioning it is a quick way to show depth.
🧩 Reading an index question safely
Find the base year — the year with an index of exactly 100.
Decide what is being asked: the price level (index) or the rate of change (inflation)?
Comparing with the base year? Just subtract 100.
Comparing any other two years? Use the full percentage-change formula.
Check the sign. Positive change = inflation. Negative change = deflation. Positive but smaller than last year = disinflation.
Round only at the end, usually to two decimal places unless told otherwise.
💡 Exam tip
Show the weighted multiplication line by line. Method marks are available even if the final division goes wrong.
Check the weights sum to 100 before you divide. If they sum to something else, divide by that total instead.
State the base year in your answer. It anchors everything else you say about the index.
When describing a trend from a chart, quote at least two figures and a direction: “inflation fell from 8% in Y2 to 1.5% in Y5”.
For evaluation, name a specific group the average hides — low-income households spend a bigger share on food and fuel, so they usually face higher effective inflation.
Keep two decimal places in percentage answers unless the question says otherwise.
⚠ Common mix-up
Index number and inflation rate. A CPI of 106.25 is not 106.25% inflation.
Subtracting index numbers when the earlier year is not the base year. 110.5 − 106.25 is 4.25 index points, not 4.25%.
Falling inflation and falling prices. The first is disinflation; only the second is deflation.
Forgetting the weights. Averaging the price indices without weighting them gives the wrong CPI and no marks.
Claiming the CPI measures the cost of living exactly. It approximates it, with substitution and quality bias built in.
Saying “inflation is bad”. Low, stable, predictable inflation is a target, not a problem. It is high, volatile or negative inflation that causes harm.
Up next: Causes and Costs of Inflation — where the pressure on prices comes from, and why the source of it decides which policy will work.
Want this explained one-to-one?
Book a free session with an experienced IB Economics tutor and get your trickiest topics made simple.
