An Economy Grew 89 Percent Overnight Without Producing Anything

The difference was the measuring instrument.

The large percentage jump that made headlines was a level revision, not an annual growth rate. Output already existed. The old series, built around 1990 weights and a narrow set of surveys, simply failed to record it.

A base year is a price structure, not a starting line

The National Bureau of Statistics FAQ on the exercise confirms the current base year was 1990 and the new base year selected was 2010. Rebasing reconstructs constant-price GDP using a later year's price structure. The statistical office describes the exercise as periodic, recommended every five years. The Nigerian rebasing also moved the accounts onto the System of National Accounts 2008 version, ISIC Revision 4 and CPC version 2.

Eurostat's explanation of the System of National Accounts puts the consequence plainly. The SNA favours a moving base year, and changing the base year affects the real rate of growth. That sentence contains the whole confusion. A rebasing does not make the economy grow. It resets the relative prices and classifications used to calculate both the level and the year-to-year change.

The level rises because the coverage expands

The Nigerian National Bureau of Statistics listed what went into the 2014 exercise: methodological adjustments, better data collection techniques, inclusion of new economic activities, re-classification of some activities, an expanded business register, and expanded data sources. Nothing in that list is a new factory. Each item is a change in what the statistician can see.

Industries that had been operating outside the old survey frame entered the accounts. The expanded business register pulled in firms that previously had no statistical existence. The adoption of ISIC Revision 4 and CPC version 2 changed how activities and products were sorted. The rebased GDP was higher because the measured economy was broader. A telecommunications operator in Lagos was not created by the rebasing; it was finally counted.

Real growth moves when the base moves

The effect on growth comes through the price weights. Eurostat's national accounts guidance says data are first calculated with the previous year as the base year, then expressed with respect to a fixed reference year. Eurostat calls the preferred measure of year-to-year GDP volume movements a Fisher volume index. Longer-period changes come from chaining those year-to-year movements.

Fixed-base volume measures age badly. A price structure from 1990 weights goods and services in proportions that no longer reflect how people spend. Chain-linking replaces that with short-term links calculated between adjacent years, which are then stitched together. The result is not a single fixed-weighted series. It is a sequence of overlapping measurements that approximates constant-price GDP over the long run.

The Nigerian NBS said it was undertaking a back-casting exercise to re-compute GDP estimates for years prior to the new base year, 2010. Back-casting is the bridge that stops a researcher from comparing a 2009 figure on one base with a 2010 figure on another. The UK Office for National Statistics defines chain-linking in its October 2016 methodology note as linking together volume series using link factors to form a consistent chain volume measure time series. The same ONS document says volume series are updated so their reference and base years are moved forward, usually by one year.

International practice moved away from fixed bases

Japan's Statistics Bureau says it changed from a fixed base year method to chain-linking for real GDP figures and GDP deflators. The Australian Bureau of Statistics, in its 2022-23 national accounts methodology published in October 2023, describes chain-linked series as being linked together to form a continuous time series. The direction is consistent. Fixed-base measures age quickly, and chaining keeps annual movements comparable while the reference year is shifted.

Nigeria's 2014 exercise used a modern framework and then began the slower work of back-casting older years. That sequencing is common. The initial publication delivers a new level and a revised recent history. The older history arrives later, once the statisticians have re-estimated it.

What rebasing is not

Rebasing is not a claim that new output appeared overnight. It is a re-weighting and re-coverage exercise. The NBS revised nominal and real GDP estimates and growth rates for 2010 to 2014, changing the whole recent vintage, not one quarter. A reader who treats the new level as proof that Nigeria's economy expanded by 89 percent in a single year is making a category error.

The practical rule is to compare only like with like. A 2013 GDP figure from the old 1990-base series and a 2014 figure from the new 2010-base series cannot be spliced into one growth calculation. They are not denominated in the same unit. Before using two vintages of any series, check the revision note. The NBS note on GDP rebasing and re-benchmarking, and the FAQ paper, state the base year and the standards used. The statistical office's back-casting work closes the gap.

The question after any rebasing is not whether the economy changed. It is whether the series you are reading has been re-referenced to match the one sitting next to it.