Why concentration isn’t a good measure of competition

By Alexander Baker. He is Managing Director at Fingleton. Fingleton provides strategic regulatory advice. Excerpts:

"The focus on concentration implies a strong and simple link between the structure of a market and consumer outcomes. In other words, that measures to limit or reduce market concentration will inevitably stimulate competition. However, there are good reasons to believe that the opposite may be true.

First, measuring concentration effectively relies on clearly identifying the boundaries of the markets firms operate in. This is easier said than done.

Take Starbucks, for example: what market is it in? Would you base market shares on coffee income? Or all drinks? What about the cakes, biscuits, salads and sandwiches it sells? If all food and drinks, should fast food outlets — like McDonalds, which sells some of the same products — be included or not? Should supermarkets be included, given they offer such food products (and increasingly coffee) in store?

Making the wrong judgement as to where the boundary of the market falls risks drawing incorrect inferences as to how effectively competition is working: make the boundary too narrow, and you invite intervention where none might be needed; make the boundary too wide, and you risk incorrectly concluding that no action is needed where it might be.

The practical challenge of accurately defining markets is one reason why competition regulators have moved away from rigid, structuralist approaches to competition enforcement over the last decade or so, focusing instead on market outcomes.

Second, concentration tells us less about whether competition is working than is often assumed.

While there is some basis in economic theory to believe that the number of firms in a market tells us something about consumer outcomes, it is more complex in real world markets. For example, a small number of large firms can often serve customers more efficiently than a large number of small ones and consumers in concentrated markets can still be well served where entry is easy.

How many competitors to WD-40 can you name? Does it matter? Are two airlines competing on any given route pair materially worse for consumers than three or four? Don’t people generally consider the market for mobile phone contracts, with four big players, more competitive than energy markets, with six big players?

The judgement as to whether a market works effectively varies depending on factors such as the nature of the product or service, its cost to produce, its value to consumers, and, how easy it is for a new firm to provide that product or service.

Concentration is just one indicator and does not contain all relevant information. It is better to focus attention and policy interventions on the factors which make markets open and competitive, recognising these may vary between markets.

It is notable that many of the markets the SMF focuses on — energy, banking, telecoms — have been subject to numerous recent regulatory analyses which have determined that markets could work better for consumers without relying on simple measures of market structure. The fact that these sectors are highly regulated means we should exercise caution in generalising to other consumer markets, and the fact they have dedicated sector regulators invites questions as to why competition isn’t working more effectively in these markets.

Third, an excessive focus on concentration risks unintended consequences which are equally damaging to consumers. It is a false panacea.

Reducing the question of whether markets function effectively down to a single measure invites false precision, and — where the link between concentration and consumer outcomes is weak — invites policymakers to focus on the wrong thing.

Industry concentration targets risk disincentivising incumbent firms to compete in some markets, because even small changes in market shares could trigger regulatory action (a problem exacerbated by the mechanics of HHI, the SMF’s favoured measure of market concentration). Where entry doesn’t occur, for example because entry barriers are high, a focus on concentration measures could encourage collusion between existing market participants, resulting in higher prices to consumers than would occur otherwise.

Automatic switching of inactive energy and telecoms consumers to ‘challenger companies’ — another SMF proposal — may result in less concentrated markets, but it disempowers consumers and encourages them to rely on government intervention to secure the best deal. It also dulls the incentive for competitor firms to become more efficient in order to win business.

Finally, there are likely to be more effective competition-enhancing policy measures which go with the grain of consumer behaviour."