Many observers perceive the accuracy and reliability of African economic statistics as far less than ideal. Morten Jerven’s 2013 book “Poor Numbers: How We Are Misled by African Development Statistics and What to Do about It” asserted that the statistical capacity of most African countries was weak and that a significant portion of the official data churned out involved “a great deal of guesswork”. This judgement was not entirely a surprise to well-informed stakeholders.
Statistics can, like any powerful tool, be used or misused. American author Mark Twain observed that “There are three kinds of lies: lies, damned lies, and statistics.” The selection of a base year for economic analysis can be an opportunity for statistical manipulation. A base year is the first of a series of years in an economic index that reflects the value of specific activity, such as foreign direct investment (FDI) or gross domestic product (GDP). It is typically set to an arbitrary level of 100. Variations in the index are expressed relative to the base year.
Base years may be replaced from time to time so data in a particular index reflects current trends. Ideally, the new base year reflects a recent yet reasonably stable period. The 2008-2010 financial crisis reveals why a base-year that reflects structural change is problematic. In response to sharp declines in housing values, many U.S. banks accepted government support and changed accounting methods (for example, suspension of market-to-market accounting) during that period. The significant market disruptions and other changes during that era will distort fiscal analysis using 2009 as a base-year.
Previous African statistical rebasing exercises reported information that, with the benefit of hindsight, may have led businesses and investors to accept greater risk. That is, after the initial scepticism about the abrupt and largely upward reviews. Kenya, Nigeria, Tanzania, Uganda and Zambia, which rebased their GDPs in 2014, had their outputs either doubled, up by a third or quarter, for instance.
There are costs from delayed reviews. That is apart from the fact that if they were more frequent, every five years being the preference of the United Nations Statistical Commission (UNSC), they would not be met with as much scepticism. There is not now as much disbelief about the economic output numbers as in the past, though; as there are now more relatively frequent reviews. Understandably, however, the sentiment about what could be missing, unaccounted for, or wrongly estimated in African numbers continues to endure.
| Recent Africa GDP rebasing exercises | ||||
| Year of rebasing | Previous
Base year |
New
Base year |
Increase in nominal
GDP post-rebasing |
|
| Ghana | 2018 | 2006 | 2013 | 32% |
| Senegal | 2018 | 1999 | 2014 | 29% |
| Ivory Coast | 2020 | 1996 | 2015 | 40% |
| Source: World Bank, Reuters | ||||
Inflation could be lower than an older stale base year suggests, for instance, with implications for interest rates and the like. And GDP rebasing exercises have consistently revealed, in the African case, at least, that economies were bigger than the old data suggested. So not only are there costs to businesses and investors from issues of timeliness and accuracy of African statistics, there are losses to the economy itself. Governments also lose tax revenues from what could have been greater economic activity consequently. Thus, the need for African statistical agencies to take more seriously the refreshment of the data they produce cannot be overemphasised.
This article highlights and analyses recent developments in African statistics to arrive at an informed conclusion on the current quality of Africa’s numbers. Have they improved? Ultimately, businesses and investors on the continent need to be much better informed about the current state of African statistics and how to use them to make decisions, given the available data.
African statistics are improving but challenges remain
The Mo Ibrahim Foundation African Governance Report (MIFAGR) measures and monitors governance performance in African countries. Its latest “Agendas 2063 & 2030: Is Africa on track?,” published in October 2019, reports that statistical capacity in Africa, which is “a nation’s ability to collect, analyse, and disseminate high-quality data about its population and economy,” is improving but remains low. The annual average trend score of the World Bank’s governmental statistical capacity IIAG sub-indicator for Africa increased by 0.60 in 2014-2017, higher than the longer and older range score increase of 0.43 in 2008-2017, pointing to improvements (MIFAGR, 2019). On challenges, Africa’s statistical capacity underwhelms on data coverage & openness and funding.
| Statistical Capacity Assessment | |||
| Africa | World | ||
| Planning | Statistical Society Presence | 50.9 | 42.5 |
| Statistical plan implemented | 76.5 | 71.9 | |
| Production | Data coverage & openness | 33.4 | 45.2 |
| Use | Statistical capacity indicator | 57.3 | 62.3 |
| Statistical literacy indicator | 13.3 | 13.2 | |
| Use of statistics index | 27.0 | 25.8 | |
| Investment | Total commitments | $4.9M | $2.5M |
| Statistical plan fully funded | 31.4% | 60.9% | |
| Source: Partnership in statistics for development in the 21st century (PARIS21) | |||
Still, even as there have been marked improvements in African statistics over the years, since Jerven (2013) at least, there have lately been matters arising. And while it would be erroneous to generalise, as there are significant distinctions between countries on the continent, exemplars like South Africa have lately given cause for some concern. For instance, the state slashed the Statistics South Africa (Stats SA) budget by 160 million rand in 2015, with no new staff since then. In mid-February 2020, the South African Statistics Council reported that the country’s statistics agency required at least $13 million to continue publishing accurate data. The implications of such long-term budget cuts include smaller survey sample sizes, leading to wider margins of error.
Funding gaps for statistics are not unique to South Africa. Underfunding is a recurring theme for many African statistics agencies. For example, Nigeria’s statistics agency constantly complains about underfunding, and received only about 40 per cent of its data production budget in 2019.
To improve the accuracy of its statistics and cut costs, Nigeria’s National Bureau of Statistics (NBS) automated most of its data production process and now publishes more of its previously paper-based reports electronically. Nigeria and South Africa are the two largest economies in Africa. They are largely representative of the continent’s two extremes of development; the former being its most advanced, and the latter being its poorest by size.
There are some positive trends. More than 70 per cent of African countries either have a national statistics strategy in place or are implementing one. Still, many issues continue to weigh on the development of African statistical capabilities.
Edited version of the article was first published by the NTU-SBF Centre for African Studies of Nanyang Business School, Singapore. References are in the original article.
