ABSTRACT
Recent publications have drawn attention to a key problem in the provision of macroeconomic statistics in countries in sub-Saharan Africa, namely that many GDP estimates are unreliable, out of date and likely to change dramatically upwards when revised (Jerven, 2013a, 2013b, 2013c), and that therefore current data may underestimate and misrepresent GDP for many countries. In response to such recent publications and the significant upward revision of GDP in Ghana (Jerven & Duncan, 2012), some commentators have gone as far as to declare the situation a statistical tragedy (Deverajan 2013). In the light of the many recent improvements made in the provision of most social statistics, it may
seem unfair or misleading to call the current state of affairs a tragedy. The importance of economic statistics in particular and the role of the statistical offices in general was neglected in the 1980s and 1990s. With the Millennium Development Goals, measurability came into focus, and with many African countries seeking to enter capital markets and attract investors. The data situation in African countries is changing rapidly. It may still be too early to declare that an ‘African statistical renaissance’ has fully occurred (Kiregyera, 2013). As a recent wave of GDP revisions have signalled, there is increased activity in the provision macroeconomic statistics from statistical offices in response to this growing demand, but the progress has been uneven. The upward revision in Ghana, almost doubling GDP, was caused by an update of the statistical
methods and the data basis for aggregating GDP. Until 2010, the Ghanaian economy had been accounted for using a 1993 base year, and data sources were growing increasingly out of date and were not capturing changes in the Ghanaian economy. When the base year was changed to 2006 in 2010 and new data sources were used to get a more accurate picture of the Ghanaian economy, GDP in Ghana almost doubled. As a result, in 2011 Ghana moved from the status of a low-income economy to
the status of a lower-middle-income economy. Preliminary figures from Nigeria indicate a similar large jump in GDP there (The Economist, 2014), while recent reports from Kenya and Zambia signal increases in the order of 25-30 per cent as benchmark years are being updated across the region (Financial Times, 2014; The Africa Report, 2014). Recent reports that have assessed trends in growth and poverty in Africa and have sought to portray
an accurate picture of economic conditions on the continent show an increasing awareness of the problems of outdated statistical systems and the fact that GDP estimates between African economies are not comparable (African Development Bank, 2013a; International Monetary Fund [IMF], 2013a). This report from the statistical offices in Nigeria, Liberia and Zimbabwe highlights that there is a growing demand for statistics, accompanied by increased attention to the importance of updating economic statistics, particularly the data and methods used in national accounts to calculate GDP and other important economic development metrics. However, a rebasing of GDP estimates for a country is a costly exercise that consumes much time and many resources. Currently, the work to update economic statistics in Nigeria and Zimbabwe is still ongoing, while
the last effort to generate an authoritative estimate of the Liberian economy ultimately proved unsuccessful. Currently, the base years for calculating GDP in Nigeria, Liberia and Zimbabwe are 1990, 1992 and 1994 respectively. The preliminary new base year estimates for Nigeria were announced in April 2014, and the change caused an upward revision of the current price estimates of 89 per cent (National Bureau of Statistics [NBS], 2014). While these reports speak mainly to the situation in the country concerned, it may be noted that on the Statistical Capacity Indicator published by the World Bank, Nigeria, Zimbabwe and Liberia scored 76, 53 and 43 on a scale from 0 to 100, where 100 is meeting all international standards on periodicity, methodology and source data in reporting (World Bank, n.d.). For 2012, the average overall score for 53 African countries considered was 59, while the average for the whole sample of countries was 66. As noted above, the picture of benchmark years for the continent changes rapidly. However, in 2013 the
IMF’s Regional Economic Outlook for sub-Saharan Africa surveyed 45 countries, and found that only four countries met the five-year rule (having a base year of 2007 or newer) – Cape Verde, Malawi, Mauritius and South Sudan – while 28 countries had base years more than 10 years old and 13 countries were still using base years more than 20 years old (IMF, 2013b). Thus the question we discuss here applies widely.
Nigeria’s National Bureau of Statistics and the Reset of the Nigerian Statistical System
