More than a decade ago, economists Gelb and Black asserted that there has been a (too) narrow concern in South Africa with whether the country is receiving sufficient investment. The primary source of FDI statistics is the South African Reserve Bank (SARB), which provides annual reports on aggregate financial value of stocks and flows.
However, recently the South African government has shown a much greater interest in the quality of foreign investment coming into the country. With the draft Promotion and Protection of Investment Bill, the government aims to enact new regulations to ostensibly ensure that foreign investment contributes more to economic development. Are South African FDI statistics adequate for informing policy? Does the new focus on FDI quality change the data requirements?
Issues with relying on aggregate FDI data in SA
Since the end of apartheid, a macro-economic perspective has dominated the South African government’s approach to foreign investment. This sees the primary purpose of FDI as filling the country’s domestic savings gap. As such there has been much focus on the aggregate financial value of stocks and flows as the main measuring stick of FDI performance. However, even if FDI volumes were the main concern, there are several issues with relying on this aggregated data.
FDI inflows into South Africa generally vary widely over time. Fluctuations tend to be associated with large equity purchases in South African firms. If one or two deals happen to occur within any given year this can lead to a considerable upswing in FDI inflows, even though this might have very little to do with the investment climate per se.
In fact some of the largest increases in FDI inflows in South Africa over the last two decades have been associated with redomiciliation of South African transnational corporations (TNCs) to the UK. When this happens domestic investments get reclassified as foreign investments. Though this might result in a spike in FDI inflows, few would see this as a sign of a foreign investor attracted by sound economic fundamentals.
In addition, as pointed out by economists over the years, the SARB FDI stock data may not be very precise. The SARB collects information on stocks using the standard method, which involves a survey of companies’ balance sheets. However, there are questions about the representivity of the sample used. Furthermore, companies are allowed to report their balance sheet data at either book value or market value, but since market valuations are subject to extrinsic factors, measurement inconsistencies can occur.
Another major problem is that the SARB does not provide disaggregated information on FDI flows. It only reports information on the financial value of inflows to, and stocks held in, the country as a whole, with no information at the provincial level. This presents a challenge, since most investment promotion in South Africa is done by the provinces, with provincial economic development agencies operating relatively independently to try to bring in FDI.
The lack of more comprehensive FDI data makes it very difficult to gauge the investment climate in South Africa. Although there have been attempts, it is also difficult to use the data to analyse the determinants or inhibitors of inward FDI; for example, the relative importance of labour regulation, infrastructure, and property rights.
Moreover, without provincial FDI data, the government cannot formulate regional policy, nor can the effectiveness of policies be compared across provinces.
Especially relevant, given the South African government’s new attitude towards foreign investment, is that aggregate financial data cannot be used to assess the impact of FDI on economic development and growth. These data cannot be used to determine what types of firms operating in which sectors in which parts of the county have the greatest impact on development indicators.
As argued by Gelb, there is need for more accurate aggregate financial data on stocks and flows of FDI. The recording of stock data needs to be standardised across all companies, while flow data should be disaggregated geographically.
There is also a critical need for an official FDI database in South Africa at the firm level. Some independent firm-level surveys have been carried out over the years and have produced useful results. However, these surveys have been very irregular and there have been challenges incentivising companies to participate, resulting in low survey response rates.
A good firm-level database could be used to record detailed information about multinational companies’ entry modes and motivations. At subsequent stages of development, it could record information on wages, training expenses, reinvestments and patents. All of this data could be used to measure the impact of FDI on productivity, skills and technological development. The survey could also be used to get feedback directly from multinational companies about new policies.
Many developing countries have already adopted such a system. Vietnam, for example, requires all foreign investors to complete information on their FDI projects on an online portal before receiving an investment certificate and has also worked with UNIDO on large-scale surveys of investors. India maintains an extensive database of foreign (as well as local) companies based on information contained in their annual reports and China conducts an annual enterprise survey of all firms with annual sales of more than RMB5 million (about R10 million).
Assembling such a firm-level database will not be cheap, but it could be extremely valuable. Not only could this kind of intelligence inform research and policy-making in South Africa, but the availability of in-depth information on FDI could facilitate the process of consensus building between government and the business community.
This article draws on the findings of an upcoming occasional paper by the same author: ‘Policy Framework for FDI Promotion in South Africa: Operations, Effectiveness and Sustainability’.