What an index number tells you
An index number expresses a value relative to a chosen starting point, called the base period. The base is commonly assigned a value of 100. Later values show how much the measured quantity has changed compared with that base. If a store’s monthly sales index is 120, sales are 20% higher than in the base month. An index of 85 means sales are 15% lower than in the base month.
For a single quantity, the calculation is straightforward: divide the value in the period of interest by the value in the base period, then multiply by 100. If a shop sold 200 items in the base month and 250 in another month, its index for the later month is 125. Indexing makes the direction and relative size of a change easy to see, even when the original figures use different units or have very different starting levels.
The difference between a baseline change and a period-to-period change
An index always points back to its stated base, but a reader may want to compare two later periods instead. Suppose an index rises from 120 to 132. The second value is 32% above the base, not 32% above the first value. To find the change between those two periods, divide 132 by 120, subtract 1, and multiply by 100. The result is a 10% increase.
This distinction matters when an index falls, too. A decline from 120 to 100 is a drop of about 16.7% from the first period, even though the index moves down by 20 points. Index points and percentages answer different questions: points describe the difference between index values; a percentage change uses a particular period as its denominator. When someone says an index “rose by 10,” check whether they mean 10 points or 10%.
Why the choice of base matters
The base gives an index its reference point. Choosing a different base changes the displayed index values, but it does not change the underlying observations. Imagine values of 50, 60, and 75 across three periods. With the first period set to 100, the index reads 100, 120, and 150. With the second period set to 100, it reads about 83.3, 100, and 125. In both versions, the change from the second to the third period is 25%.
A base period should be clear and reasonably useful for the question being asked. A period with an unusually high or low value can make later movements look striking when compared with it. That does not make the index wrong, but it makes the context important. Before interpreting a chart, look for its base period and ask whether the comparison is meaningful for the decision at hand.
When an index combines several things
Some indexes track one quantity; others combine many quantities into a single measure. A price index, for example, may represent changes across a collection of goods and services. In a combined index, the method must decide how much influence each item has. An equal-weighted index treats each component alike, while a weighted index gives some components greater influence. Different choices can produce different results from the same set of item-level changes.
That is why a combined index may not match any one person’s experience. If the items you buy, use, or care about differ from those emphasized by the index, your own pattern may move differently. When comparing two indexes, check what each measures, which items it includes, how it weights them, and whether their definitions have stayed consistent over time.
A practical checklist for reading an index
Start with the label: what quantity is being indexed, and what period equals 100? Next, identify the comparison you want. To compare a value with the base, subtract 100 from the index value to get the percentage difference from the base. To compare any two periods, divide the later index value by the earlier one, subtract 1, and multiply by 100. Keep the sign: a negative result indicates a decline.
Finally, ask what the index leaves out. An index shows relative change, not the original size of a quantity. Two shops can both have a sales index of 120 while selling very different numbers of items. A combined index may also hide components moving in opposite directions. Use the index to understand the change it was designed to show, then seek the underlying values or component details when the size or cause of that change matters.
