Prices across the whole economy are 5% above the base year
The GDP deflator is nominal output divided by real output, multiplied by 100: 21,000 over 20,000 gives 105. The base year, by construction, has a deflator of exactly 100. So the 5 points above it say that producing the same real output costs 5% more than it did in the base year.
It also works in reverse, which is its main practical use. Real GDP is nominal GDP divided by the deflator and multiplied by 100, so a deflator lets you strip inflation out of any nominal series. That conversion is what turns a headline growth figure into something meaningful about actual production.
One covers everything produced; the other covers what consumers buy
The consumer price index tracks a fixed basket of household purchases, including imports. The deflator covers every good and service produced domestically — consumer goods, capital equipment, government services, exports — and excludes imports entirely. They are answering different questions and routinely differ by a point or more.
The difference is informative rather than a discrepancy. A country whose import prices spike sees the CPI rise while the deflator barely moves, because imports are not domestic production. An investment boom that raises equipment prices shows in the deflator and not in the CPI. When the two diverge sharply, the gap usually names which part of the economy the price pressure is coming from.
A fixed basket drifts out of date; the deflator never does
The CPI weights a basket that is updated only periodically, so it can overstate inflation when consumers substitute away from goods whose prices rose. The deflator uses current-period quantities by construction — it is the ratio of two measures of the same actual output — so its weights update automatically every period.
That strength is also a subtlety: because the weights move, a change in the deflator mixes price changes with changes in what the economy produces. Modern national accounts use chain-weighted indices to limit the distortion, but comparing deflators across distant years is still less reliable than comparing adjacent ones.
Both figures in the same currency, and a base year you know
Nominal and real GDP must cover the same period and the same geography, and the real figure must be stated in some base year\u2019s prices. The deflator is only interpretable if you know which base year that is: an index of 105 means nothing without it, and comparing indices built on different base years is meaningless.
Published real GDP series are usually chain-weighted rather than fixed-base, which makes the implied deflator a chain-type price index rather than a simple ratio. For most purposes the distinction is immaterial; for precise historical comparison over decades, it is not.
Sources & References
Figures on this page are checked against primary, authoritative sources. Links open in a new tab.