THE AIM AND AMBIT OF THE COMMISSION'S WORK: Our objective at the Competition Commission is to maximise consumer welfare, the welfare of the man in the street, not that of producers. We seek to promote competition in two ways: by examining mergers and acquisitions for anti-competitive consequences, and by protecting consumers against restrictive business practices.

MERGERS: Mergers above stipulated thresholds (for example, where the merged entity would have combined assets above R200m) have to be notified to the Competition Commission. Previously, competition law allowed for review only after mergers had been effected. But this was ineffective. Trying to unravel a merged entity is like trying to unscramble an egg.

In assessing mergers, our focus is on whether the market structure that will emerge will be conducive to competition. Take the proposed Nedcor/Stanbic merger some four years ago. Already, the banking sector was dominated by four major players (though at that point we also had Saambou and some smaller banks). The key question was whether it would be competitive to have three retail banks instead of four. It is always difficult to assess future consequences as this involves guesswork and some speculation. But it is nevertheless important to try to look ahead - and here we felt that reducing the number of players would be anti-competitive. The finance minister agreed and he halted the merger.

RESTRICTIVE BUSINESS PRACTICES: Here, our main aim is to prevent price-fixing because this is bad for consumers. Our primary target is the cartel. We all know the example of Opec. The price of crude oil drops and they meet in Vienna and agree to cut production. The price goes too high, and they meet in Vienna and agree to boost production. No tribunal has jurisdiction over Opec because it is made up of countries, rather than companies. But this might change if the World Trade Organisation were to adopt competition law.

The danger of price-fixing is particularly acute in South Africa because market concentration is frequently very high. Often three or four players dominate 80% of the market. They do not necessarily have to collude in order to affect prices. They only need to follow one another. In the cement industry, there are three players who operate in three different geographical areas. The cost of transport is very high and they don't try to enter each other's markets. And the price of cement is the same throughout the country. Hence, we want to create a situation where there are no dominant firms.

Price fixing can also occur via resale price maintenance between manufacturers and distributors. The manufacturer may offer the distributor a 5% discount, but warn that the discount will be forfeit if the distributor sells at below the recommended price. Take Toyota. A customer complained that all the Toyota dealers around the country were offering the same discount of 3%. We sent staff out as investigators, posing as customers, to see if this was true. They found the same throughout South Africa. They also heard salesmen telling them, 'I'll get in trouble if I go lower.'

We fined Toyota R12m on the basis of an agreement. We had wanted R40m. We are now looking at resale price maintenance in the motor vehicle industry as a whole and will soon announce the results of our investigation. We subpoenaed information from the industry and, by mistake, were sent a huge document on the discount system. This provides all the evidence we need of resale price maintenance. People argue that there is no resale price maintenance in this, but only an indication of the maximum discount available. But a maximum discount results in a minimum price.

In looking at restrictive business practices, we also assess whether companies are refusing to give access to an essential facility to a competitor when it is economically feasible for them to do so. This is the basis on which we acted against GlaxoSmithKline and Boehringer
Ingelheim. They were denying access to an essential facility - the formula for AIDS drugs. This should be made available to all generic manufacturers in return for a reasonable royalty. We are not infringing their patent. The companies wanted to maximise their profit, but a patent can be used abusively.

A STATISTICAL OVERVIEW: The Competition Act of 1998 came into effect on 1st September 1999. Since then, we have dealt with over 1 000 mergers: about 150 a year on average, or 20 a month. In keeping with world-wide trends, we approve 90% to 98% of them. Hence, we have barred mergers only five or so times in each year of our operation.

Regarding restrictive business practices, we have dealt with 200 cases since we started and have referred 26 cases to the Competition Tribunal for prosecution. We do this if we find that there is a prima facie case to be answered. Between six and eight cases have fallen away, either because market conditions have changed or because the players have altered their stance. In such instances, we withdraw our case. Other cases are ongoing and, in some instances, we have already had decisions.

We have not pursued the remainder of the cases (the great bulk of them) because we perceive no substantial lessening of competition. If the effect on the economy is not significant, we do not intervene. The commission receives three to four complaints per week, mostly from the manufacturing sector alleging abuse of market dominance. This can result from exclusive supplier agreements that prevent customers from sourcing elsewhere. But it is also necessary to distinguish between market domination due to uncompetitive practices and the domination that results from efficiency.

EFFICIENCY AND INNOVATION: We recognise that companies can become dominant, not through anti-competitive behaviour, but through innovation and efficiency. This is how Microsoft initially came to dominate the PC software market (though there may be concerns now about the company acting in an anti-competitive way).

The object of competition regulation is to provide for efficient markets and to encourage innovation, because older goods become cheaper. In 1978, my family bought our first TV set. It was a German product and expensive. Now the market is full of Japanese products which have become much cheaper. Now even university students can buy TVs with their pocket money, plus VCRs and digital cameras. Competition has been good for the electronic goods industry. And the upshot has been good for consumers too, for there are now more satellite dishes in Alexandra than in many of the suburbs!

EMPOWERMENT AND THE PUBLIC INTEREST: Under South African competition law, we also have the power to help advance the previously disadvantaged. But the importance of promoting black economic empowerment (BEE) does not necessarily change our assessment of whether a merger should go through. Recently, we were asked to approve a merger between two companies which between them had 100% of the market in a particular sector. The target firm was a BEE company, but after the merger there would have been one company and not two. The effect would have been be to create a monopoly.

The public interest in BEE did not outweigh this. Consumer welfare is not served by a monopoly - and we would not approve a merger on public interest grounds in such circumstances. Creating a monopoly, even if it is a BEE company, would not be acceptable. A black monopoly is no better than any other monopoly.

But allowing a BEE company to be swallowed up by a large established player may not be in the public interest either. Take the proposed Tepco/Shell merger some years back. Tepco was the largest BEE-owned oil company in the country, and Shell proposed to buy it. Tepco had grown to having 3% of market share and had the potential to grow more. At the same time, it was not really an effective competitor at 3%. But was it proper to have the single symbol of black involvement removed from the market? In our view, the public interest did not justify a BEE firm being removed in this way. This decision was not made on competition grounds but on public interest grounds.

But circumstances may change too, as happened in this instance. By the time the matter went before the Competition Tribunal, Tepco was in financial difficulty. Its parent company (also a BEE company) had to keep on putting money into it to keep it afloat. BEE could not be served by a BEE firm constantly having to pump money into a failing BEE concern. It would be better for the BEE parent to have a stake in Shell and to get dividends on that, the tribunal considered.

BEE introduces many other complex variables. A BEE deal could have the effect of improving strained relations between the private sector and government in a given sector, and this could be beneficial all round. BEE players could introduce a new sensitivity to consumer interests, suggesting that BEE deals are to be welcomed. We must also be careful of what we rule anti-competitive. If we were to say that a merger which gives one company a 25% stake in another is anti-competitive, this could have major implications for BEE deals aimed at giving empowerment partners 26% of equity. We have to ask ourselves if the public would accept an anti-competitive transaction on BEE grounds. Perhaps it would if the BEE partners have sufficient management and control to act as a constraint on any potential abuse of dominance.