Some six months after the International Monetary Fund frowned upon the way the Ministry of Finance and Development Planning does its national budgets, the African Development Bank has also expressed concern about the manner in which Statistics Botswana handles national accounts. On the basis of its assessment of Botswana’s processes, the bank worries that the country’s GDP may be “underestimated”.
Following revelation that seven of the ten fastest-growing economies in the world were in Africa, doubts were raised over the quality of the national accounts that measure the volume of economic growth. In response, AfDB established a programme of peer re-views for economic statistics, focusing on national accounts. This was in support of the bank’s longstanding objective to enhance statistical capabilities in its member countries. Working in collaboration with member countries, AfDB embarked on an audit of economic statistics and practices across the continent in 2013. Last week it published results for Botswana, Ghana, Zambia and The Gambia. Review teams were dispatched to host countries and in Botswana, this exercise was carried out by three national accounts experts from Ghana, Kenya and Sierra Leone over five days.
“None of the review teams reported on how their host country estimated imputed rents for owner-occupied dwellings. Many countries have difficulty applying the stratification approach or the user-cost approach advocated by the System of National Accounts 1993. They either use an alternative approach, such as asking owner-occupiers how much their dwelling would rent for, or they limit the imputation to urban areas, or they make no imputation at all. When properly measured, actual and imputed rents for dwellings together account for at least 5 percent of GDP in low-income countries; the percentage is nearly double in high-income economies. The GDP of countries that do not follow the SNA rules with regard to the estimation of imputed rents are likely to be underestimated,” AfDB says.
Own-account construc┬¼tion by households as well as informal sector activities and non-monetary activities (such as the collection of firewood and the fetching of water) are not included in the estimate of GDP. The review team found this to be anomalous.
“Informal and non-monetary activities are important and the failure of the four countries to make an explicit estimate of their gross value added is a serious omission and probably means that their GDPs are underestimated,” the review report says.
The reviewers found that Botswana ÔÇô like Zambia, was not making adequate use of administrative data (such as value added tax turnover data) to compile its national accounts. The two countries rely chiefly on annual and quarterly survey data collected from enterprises and establishments.
“… it is pointed out to Zambia and Botswana in the review reports that VAT data could be used to supplement the data collected by surveys ÔÇô this is an important consideration given the low response rate. None of the four countries is currently using supply and use tables to compile their production and expenditure accounts… A recommendation common to all review teams is that the host country should start compiling supply and use tables annually,” the bank says.
The reviewers also found that while Statistics Botswana maintains a business register that is updated on an annual basis using information from the Annual Economic Survey and the Quarterly Survey of Recent Trends, “the register is not exhaustive; it does not cover all establishments in the formal sector.”
In June last year, the IMF questioned the credibility of the Botswana national budget, noting in a report that the country’s revenues are persistently under-forecast, with an average error of 3.6 percent of GDP, whilst expenditures usually underperform, with an average error of 2.8 percent of GDP.
Ketso Makhumalo, who is Head of National Accounts and Prices Division at Statistics Botswana, was part of the 22-member review team, carrying out her own assignment in The Gambia.