Market concentration is a measure of the structure of the market and reflects the number and size distribution of firms within the market. A market where only a few firms hold most of the market share is considered to be concentrated.
In the first version of the Competition Amendment Bill, released in December last year, the explanatory memorandum notes that “the South African economy is characterised by unusually high levels of concentration, in part due to strategic barriers to entry created by incumbents as well as low rates of business formation and as a result of mergers and acquisitions”. It went on to state that “[r]esearch conducted by the Commission supports a widely-held view that markets in South Africa remain highly-concentrated, some twenty-three years after the end of apartheid”.
Although the explanatory memorandum stated that “[m]ost” studies find that economic concentration is relatively high in many markets in South Africa, it appears that reliance was placed on a single Commission study which “found there were 294 dominant firms in defined markets identified in the 31 sectors considered. 70% of the sectors have dominant firms in some of their defined product markets.”
Using this data as a basis to conclude the extent of concentration, and therefore the need to advance the de-concentration of the market, the Bill provides for an extension of the mandate of the competition authorities and the National Executive, particularly in the case of market inquiries.
Given the significance of the amendments proposed by the Amendment Bill, Tim Cohen, in the Business Day, challenged the premise of the amendments. In an article entitled “Business ‘Concentration’ – Competition bill is based on some dodgy numbers”, Tim Cohen argues that the HHI measurement of concentration is flawed in that, (1) its results depend on how the market is defined (ie a narrower definition of the market will result in inflated market shares, in turn optically resulting in higher levels of concentration), and (2) using the US’s measure of competitiveness (ie the “HHI” measure of concentration) is improper, as one cannot compare the South African and US economies (as they differ in strength and size – the South African economy will inevitably be more concentrated because it is much smaller and therefore there is less “space” for more players). By way of example, Tim Cohen questions the level of concentration in the financial services sector. He notes that in his experience the financial services sector is highly competitive, demonstrated by the recent entry of new players in the space.
The Commission responded to this challenge by publishing the research that the explanatory memorandum to the Amendment Bill referred to, being a study entitled a “comprehensive empirical account of the extent of concentration in product markets in South Africa”.
What is notable about the Commission’s research is that it is based on merger reports, which, in many cases do not need to propose or rely on a market definition because the transaction does not impact the market on any definition. Nevertheless, market share estimates are provided, often without extensive consideration of the relevant market or bases for the definition. This seems a shaky basis to assert conclusively that “70% of the sectors have dominant firms in some of their defined product markets”.
The South African economy appears to be concentrated based on a general impression about the presence of large successful companies and relatively few experiences of new entities growing to a comparable size of success. But whether such perceptions are sufficient to provide a data-based (as opposed to political or industrial policy) justification for new legislation is less clear.
Now that the proposed amendments are near to ending their journey to promulgated legislation, investors will need to be aware that mergers in sectors with few market players or with relatively few recent entrants may require more evidence to counter perceptions of concentration or insufficient market dynamism in order to get such transactions approved.