Margin squeeze modelling can be simple or complicated, but never trivial

27 July 2012 | Regulation and policy

Ian Streule


Margin squeeze modelling can be simple or complicated, but never trivial

Regulatory authorities, competition bodies and operators rely on margin squeeze models when testing, challenging or defending an allegation of abuse of dominance in which a vertically integrated player sets prices that squeeze its competitors. This article outlines the aspects of margin squeeze modelling based on Analysys Mason's extensive skills and experience in this field.

The legal work surrounding margin (or price) squeeze allegations is extensive and complex. However, in these cases some kind of margin squeeze model will be used, which will form the foundation to prove or disprove the legal case. The margin squeeze model will present the underlying numerical comparison of costs, prices (revenue) and profit margins relevant to the business situation being tested.

Margin squeeze models can be simple or complicated, and the choices made will always be argued to fit the case in question. These choices (which we structure as scale, scope, financial test and cost/price principles) strongly affect the complexity of the margin model, its inputs, assumptions, calculations and outputs (see Figure 1).

Figure 1: Aspects of a margin squeeze model [Source: Analysys Mason, 2012]

Figure 1: Aspects of a margin squeeze model [Source: Analysys Mason, 2012]

Equally efficient operations are particularly relevant in ex-post cases, but a similarly efficient operator (that is, similar but not identical) can be established. We find that defining a reasonably efficient operator can be the hardest, because it requires judgement on what is reasonable scale.

The scope of the model depends on the case in question, but can range from easy individual product tests, to complex bundles of services across multiple markets. It is well recognised that testing service margins in bundled offers is challenging on both the cost and revenue sides, because bundled products share significant cost and revenue pools (such as network infrastructure and monthly revenue).

Single period profit models are used in specific situations where dynamic, time-varying effects are limited (for example, mature product markets); building a discounted cash-flow margin model is more complex, particularly where new services might be experiencing strong but highly uncertain growth. Period-by-period models can also be built and can be more complex than discounted cash flow models because they require accurate amortisation of costs in each period.

Establishing cost and price principles for the model and margin test requires a detailed investigation into the costs of delivering the product(s) in question, and identifying what revenue might be obtained. Numerous questions will arise, such as the following.

  • Should the model apply the long-run incremental costs (LRIC, LRAIC) of the retail arm, or fully allocated costs (FAC) including a share of common costs?
  • Does the entrant's proposition include a price discount for lower standards of customer support, or conversely is there a premium price for a premium brand?
  • Is there only one wholesale supply option, or are there multiple possible offers? The definition of wholesale inputs to be tested becomes particularly difficult when there is non-homogeneous supply in a geographic market (for example, islands of NGA).
  • Can the revenue be uniquely identified? What about non-regulated services in the same bundle? How should customer acquisition costs and revenue be treated?
  • How should on/off promotional periods be incorporated?
  • What is a reasonable level of profit?

These issues are not trivial because each one will accumulate and add/remove overall margin (positive or negative) to the end result.

Analysys Mason has recent experience in developing margin calculations for stakeholders in countries across Europe in a wide range of areas in the telecoms, media and technology sectors, including:

  • fixed broadband (unbundling and wholesale broadband access)
  • fixed voice (residential and business offers)
  • access to media content
  • mobile voice and broadband
  • bundled fixed network products
  • wholesale fibre unbundling and fibre-based bitstream access
  • MVNO and national roaming access.

One of Analysys Mason's margin squeeze test models has recently been published by the Norwegian telecoms regulator as part of its investigations into the mobile market.1

Contact for further information.