Are the reams of statistics emanating from the United Nations (UN), World Bank and International Monetary Fund (IMF) mere works of fiction?
Well, at best they are wildly inaccurate, according a growing band of economists.
As the cheerleaders and naysayers fight it out in the 'Africa rising' debate, a third party now says no one has the reliable data to prove the case.
Scepticism about statistics does not mean that Africa is not growing fast or even that it is not much richer than the UN says.
Nigeria has not reviewed the structure of its growth data since 1990. The most conservative estimates are that its GDP of some $270bn is in reality twice that size
It simply argues that these institutions should have the honesty to admit their figures are somewhat detached from reality.
It is more than an intellectual debate.
Statistics can determine the allocation of resources and investment decisions affecting millions of lives.
Then there is the politics. Governments want to report economic growth, especially in an election year.
Multilateral institutions do not want to be caught out by local politics, so they do their own calculations.
In Ethiopia, the government, IMF, African Development Bank (AfDB) and World Bank all reported differing growth figures during the past five years.
Naturally, the government's were by far the highest.
In mid-March, the IMF said that Mali's economic growth is "encouraging" and will pick up to 4.8% this year.
Really? Oh yes, says the IMF: GDP shrank by 1.2% in 2012 after a bad harvest in 2011.
Its data gatherers are an intrepid bunch, given there was a military putsch in Bamako last March and the northern half of the country was under the control of insurgents for most of 2012.
Shanta Devarajan, the World Bank's outgoing chief economist for Africa, reckons that two out of three African governments are using unreliable statistical bases that date back to the 1960s and take no account of the sweeping structural changes in the last half century.
What Devarajan calls "Africa's statistical tragedy" is illustrated by the updating of Ghana's GDP figures three years ago, adding another $13bn overnight.
"The tragedy is that we [the World Bank and IMF] were happily publishing growth figures for Ghana over the past decade, when in fact the national accounts were understating growth by 62%," he explains.
Implications for policy are huge, argues Devarajan: "Even the economists' celebratory estimate of poverty declining in Africa during a period of growth needs to be taken with a grain of salt. In reality, there are many countries for which we simply don't know."
Fellow economist Morten Jerven, from Simon Fraser University, writes in his new book* that "more than half the rankings of African economies up to 2009 may be pure guesswork."
Another correction is due for the Nigerian government, which has not reviewed the structure of its growth data since 1990 – before the development of the multibillion-dollar cellphone market,
the massive expansion of the service sector and the birth of another 80 million people.
The most conservative estimates are that its GDP of some $270bn is in reality twice that size, says Jerven.
That would amount to another 40 economies – each roughly the size of Malawi's – to be added to continental GDP.
Belatedly recognising the data crisis, international organisations are scrambling for solutions.
In December 2011, the AfDB launched the Open Data for Africa platform to make its statistics accessible, along with their sources.
Devarajan calls for a substantive boost to national statistical organisations, with frequent reviews at the highest level.
Jerven goes on to propose that "economic anthropologists" should go beyond official data collection to measure the spreading ownership of consumer goods such as washing machines, even using satellite imagery to capture changes in the intensity of artificial light.
And some economists in the Organisation for Economic Cooperation and Development want to replace the primacy of GDP with a much broader measure of social development.
It sounds like a good idea, but for the moment, as Devarajan says, we simply do not know●
*Poor Numbers: How We Are Misled by African Development Statistics and What To Do About It by Morton Jerven, Cornell University Press.