New editorial guide, published 17 October 2026. The historical sources linked below are preserved separately and are not represented as current product or service information.
An inflation headline can describe slowing price growth while a reader sees higher prices in the shop. Those observations are not necessarily contradictory. A rate of change and the level being changed are different quantities. The historical inflation item in this domain’s catalogue is useful context for learning that distinction, but it is not a current economic release. This guide uses invented index values to explain the arithmetic. It does not forecast inflation, describe a household’s actual costs, recommend a financial product or claim that a simple classroom basket reproduces an official statistical methodology.
Begin with the level being compared
Imagine a hypothetical price index of one hundred in the first period and one hundred and ten in the second. The increase is ten divided by one hundred, or ten percent. Now suppose the index reaches one hundred and fifteen in the third period. It is higher again, but the increase from one hundred and ten is about 4.55 percent. The rate of increase has slowed while the level has continued to rise. Nothing in that arithmetic says that every product increased or that each household experienced the same change.
The European Central Bank’s inflation explainer describes a broad increase in goods and services prices and explains that items receive different weights in an aggregate measure. That supports a basic distinction between a general indicator and a single price. It does not mean any invented basket is an official measure. When reading a release, identify the named index and its coverage before interpreting the rate. A consumer-price index, a producer-price measure and a particular grocery bill are not interchangeable simply because all involve prices.
Keep monthly and annual comparisons apart
A monthly rate compares one month with the preceding month. An annual rate compares with the corresponding month a year earlier, under the series’ stated method. They answer different questions. A positive monthly movement can coexist with a lower annual rate if the earlier comparison base differs. Do not combine the two into a single narrative without checking the actual series. Record the periods beside the values in your working notes. A chart labelled inflation that omits the comparison period asks readers to supply a definition they may not share.
Use another invented index example. If last September was one hundred and this September is one hundred and two, the annual increase is two percent. If August immediately before this September was one hundred and one, the monthly comparison uses one hundred and one as its denominator instead. The two calculations are not competing estimates of the same rate. Show the index values when explaining the difference. Avoid multiplying a monthly rate by twelve and presenting the result as the observed annual rate; that would substitute a new calculation for the published comparison.
Distinguish a rate change from a price change
If an annual inflation rate moves from four percent to three percent, it falls by one percentage point. The relative reduction in the rate is one divided by four, or twenty-five percent. Neither statement says that prices fell by one percent or twenty-five percent. The object being compared is the rate itself. Use explicit wording in headlines and captions. A reader should not have to infer whether a percentage describes the price level, the change in that level or the change in the rate of change.
Check verbs as well as numbers. Prices eased can be ambiguous when the evidence only shows slower growth. A sentence saying price growth slowed is narrower and often clearer. Conversely, if a particular index actually declined over the stated period, say which index and period rather than implying every price fell. The arithmetic cannot choose the correct scope for you. It can only expose whether the words describe the quantities supplied. An editor should inspect the surrounding explanation before replacing a precise phrase with a familiar but less exact synonym.
Read contributions in their own unit
A component’s contribution to an aggregate rate is not automatically the component’s own inflation rate. In a simplified invented basket, a heavily weighted category with modest price growth can contribute more to the overall change than a lightly weighted category with rapid growth. The official method may involve more detail than this example, so consult the release notes before reproducing a contribution calculation. The practical reading habit is to keep contribution measured in percentage points distinct from the category’s rate measured in percent. Similar typography does not make them the same quantity.
Do not add percentages from unrelated components unless the method establishes that doing so is valid. A table may contain rates, weights and contributions in adjacent columns. Copying a value from the wrong column can produce a sentence that looks plausible while describing the wrong concept. Read the column heading and footnote together, including any rounding note. If rounded contributions do not exactly sum to a displayed total, investigate the stated precision before claiming an error. Equally, do not assume rounding explains a large discrepancy without checking the actual figures.
Avoid turning the average into everyone’s experience
An aggregate measure summarises a defined population and basket. A household that spends a different share on particular categories can experience a different pattern. That does not automatically invalidate the aggregate, and the aggregate does not invalidate the household’s observation. Keep the two questions separate: what does the published indicator measure, and what happened to this particular set of purchases? A personal example can illustrate variation, but one receipt cannot establish a national rate. A national rate cannot describe the exact change in every person’s cost of living either.
Be cautious when attaching a policy or welfare conclusion to a single number. A historical release can establish what the indicator reported for a period, but broader conclusions require additional evidence and an explicit question. This guide does not assess whether a particular institution should change policy. It teaches a prior step: make the statistical statement coherent before debating its implications. A report can be numerically accurate while its headline overstates what follows. Separate the measured result, the source’s interpretation and the writer’s judgement so readers can evaluate each in turn.
Write a compact, reproducible metric note
A useful note names the index, geography, reference period, comparison period, rate, release date and any revision status relevant to the story. Link to the original release rather than only another article describing it. If you calculate an additional comparison, label it as your calculation and show the inputs. Keep historical figures in historical tense. The old September release in this domain’s catalogue remains an old release even if the page was captured months later. A newer archive timestamp does not turn the reported rate into a current observation.
Before publication, ask someone to explain the claim back in plain language. If they say prices fell when the article only establishes slower price growth, revise the wording or the visual hierarchy. Do not rely on a footnote to reverse the main impression created by a large headline. The desired result is modest: a reader can identify what changed, relative to what, over which period and in which unit. Those four questions make an inflation story more intelligible without pretending that one index settles every economic or personal question attached to it.
Primary references
Related historical source records
- Inflation in eurozone climbs to 2.2 percent in September · captured 7 January 2026