Traditionally, competition authorities start their assessment by identifying the competitive scene, i.e. the relevant market, which includes the firm’s product and potential substitutes. They then calculate market shares to decide whether the firm(s) under scrutiny has enough market power to benefit from the adoption of an anticompetitive strategy.
However, when market shares are not a strong indication of market power, market definition is least useful. Difficulties arise in products differentiated markets since market boundaries are problematic to draw; as well as in bidding markets, in markets with networks, and in highly dynamic and innovative market. This is so because competition might be “for the market” rather than “in the market.” Moreover, when market concentration has no strong links to the economic theory by which competition would allegedly be harmed, market definition is also of limited help.
From an economic point of view, the assessment of market power does not always require the identification of the relevant market. Economists prefer to focus on the level of competition in the market and to gather and assess evidence in the way that best sheds light on the competition question at hand. To better serve the purpose of competition law analysis, economists have developed techniques that allow the direct assessment of market power, some of which are discussed below.
Direct evidence of the effects of a firm’s behaviour might better answer the question at hand. Indeed, sometimes it might be enough to establish that the firm’s price is too high compared to its marginal cost, or that its economies of scales are so great that no one can replicate them, and hence compete effectively. Past events in the market, such as recent mergers, the entry, expansion or exit of market players and their effects on prices, are likely to be very informative regarding the potential competitive effects of future occurrences in the market.
Alternatively, market power can be assessed directly by calculating the Lerner Index, which predicts the price-marginal cost margin and reveals the firm’s ability to raise prices above short-run marginal costs. However, in many industries the competitive price is not equal to short-run marginal costs. Moreover, the Lerner Index is not applicable to firms producing multiple products – the majority – and it assumes that the firm is competing according to a static short-run non-cooperative Nash equilibrium – but this will often not be the case.
Another way to identify market power is to observe the way firms and industries react to variation in marginal costs. This can be done, for example, by focusing on the residual demand estimation, which measures the ability of a firm to raise prices by reducing output after taking into account the demand responses of buyers and the supply responses of its competitors. The residual demand curve is the horizontal difference between the market demand curve and the total supply of all other firms.
As the exercise of market power follows from demand inelasticity, a firm holds market power and can increase prices when buyers do not have good demand substitutes. If demand is not highly elastic, the exercise of market power can be directly observed by looking at the competitive price or competitive industry output. If costs have not changed, the most plausible explanation for any observed variation in the market price is that the demand elasticity has altered while firms were exercising market power.
A fourth technique to assess directly market power relies on the assumption that firms may behave differently when cooperating than when they compete. If data regarding multiple types of behaviours is available, econometricians will assess whether the specific shift in prices can be better explained by two types of behaviours rather than one.
Lastly, the effects that a merger can have in a market can be directly assessed by focusing on consequent price changes rather than increases in market shares. Merger simulations and pricing pressure tests are more suited to assess deals in products differentiated markets than the market definition/market shares approach.
Merger simulation directly calculates how much the price will increase post-merger by first modelling the nature of competition in the pre-merger scene and then combining this data to predict the post-merger scenario. Merger models can play an important role in the assessment of the effects post-merger marginal cost reductions have on prices, but they cannot represent the only basis for a competition authority’s final decision, as they omit important parts of the competitive effects analysis, such as entry, buyer power, product repositioning and changes in the mode of competition post-merger.
When closeness of competition is the central issue, market shares are not necessarily indicative, while diversion rations directly assess the issue. Diversion rations calculate the percentage of lost volumes that are captured by another product following a unilateral price increase. If combined with margin data, diversion ratio can perform so-called price-pressure tests, which can assess the change in pricing incentives post-merger – and, at instances, the magnitude of the price change. The firm’s net incentive to raise prices after the merger is identified by comparing its incentive to increase prices due to lost competition against the opposing incentive to decrease prices due to cost savings.
In conclusion, depending on the type of behaviour under scrutiny and the available data, it is not always necessary or useful to delineate the relevant market to establish if a firm is able to exercise market power. The choice towards one or the other method should be dictated by the structure of the industry under scrutiny and the legal question at hand, rather than by the certainty that market definition is the best tool available.