Market differentiation in consumer goods: one size fits all?

Consumer goods have seen significant price rises lately, driving up general inflation levels. Some retailers point to territorial supply restrictions limiting their ability to secure lower wholesale prices—they consider these to be anti-competitive. This article considers two specific economic aspects of this discussion: first, it reflects on the effect of territorial supply restrictions on wholesale prices; second, it considers how economics can help to distinguish where product differentiation ends and territorial supply restrictions begin.

Consumer goods have seen significant price rises lately, driving up general inflation levels. A recent Agenda article asked what competition law can do to battle inflation;¹ on 6 January this year, the Dutch financial newspaper Het Financieele Dagblad published an article highlighting significant differences in wholesale prices of consumer goods across EU member states, and the limitations that retailers (and, ultimately, consumers) face in benefiting from these price differences by purchasing cross-border.²

The practice of packaging differentiation—using different packaging for what is essentially the same underlying product—is one example of a wider group of practices referred to as territorial supply restrictions.³

We see examples of product differentiation every day as suppliers aim to make their products and services better suited to the diverse preferences of end-users. Economic theory has established that some degree of price differentiation can be beneficial to generate wider usage of products.⁴

The contentious issue with territorial supply restrictions is whether these measures aimed at limiting international trade, and in particular intra-EU trade, can be deemed to be in breach of EU competition law.

This article aims to answer two questions:

1. How can economics help to establish the effect of territorial supply restrictions on retailers?
2. How can economics help to establish when product differentiation between countries becomes anti-competitive?  

Before each of these questions is discussed, we provide a short introduction to territorial supply restrictions and EU competition law.

Territorial supply restrictions under EU competition law

The European Commission has a history of investigating territorial supply restrictions. These cases span multiple industries, including consumer electronics (e.g. Consten-Grunding, 1966) and pharmaceuticals (GSK, 2001–08).

More recently, there have been two investigations focused on the food and beverage industry: the first is the Commission’s 2019 decision to fine AB InBev for abuse of dominance by limiting intra-community trade,⁵ and the second is the Commission’s ongoing investigation into Mondelez.⁶

For the latter, since the Commission formally opened an investigation in January 2021, there have been no subsequent official statements.⁷ In February 2023, it was stated that Mondelez had set aside an amount of €300m to resolve the antitrust probe.⁸ In this case, the Commission also explicitly refers to a possible infringement of Article 101 TFEU, a difference between this case and the investigation into AB InBev.

The effect of territorial supply restrictions

Assuming it was possible to show that a producer had abused its dominant position by limiting intra-community trade through excessive differentiation or other restrictions, the question then turns to what effect this behaviour has on competition.

Effects on wholesale prices

Both the Financieele Dagblad article and a 2020 European Commission Study on territorial restraints (‘The 2020 Commission Study’) suggest that reduced hurdles to intra-community trade would see retailers being able to profit from lower input prices.⁹ However, this static expectation is debatable and does not apply universally. Producers may also differentiate prices in different markets to reflect the differences in local market position, product preferences and ultimately willingness to pay.

The following example illustrates this point. Let’s assume a producer holds a strong market position in its home market where it commands relatively high wholesale prices. As part of its growth strategy, it seeks to enter a new market and to establish itself there. It offers retailers in this new market the same product at a lower unit wholesale price. Extensive trade between these markets would lead to retailers starting to source in the new market for resale in the home market.

The financial markets teach us that international trade will quickly eliminate differences between the prices of two identical goods in different locations (for instance the price of a company share listed on two different stock exchanges). The process of exploiting (and eliminating) price differences is referred to as arbitrage.

In our example, this would lead to the multinational producer seeing the product it supplies to the foreign market, coming back into its home market and driving down the wholesale price it commands in that market. The extent to which this occurs will be partially driven by practical considerations such logistics (i.e. can the foreign supply be repatriated quick enough?) and transportation costs (i.e. is the price difference significant enough to cover the additional transportation costs?).

Met with this prospect, the producer now faces a trade-off: do the proceeds of the increased revenue from sales in the new market outweigh the reduced revenues in its home market? If not, the conclusion from this simple case may well be that the producer stops supplying in the new market. Another outcome could be that the producer increases its prices for the new market to limit the margin from arbitrage/parallel trade.

The point here is that the static view that producers will maintain the same set of prices following a hypothetical infringement decision is difficult to sustain. The analogy with the waterbed is a useful one in this context: if the firm cannot maintain high prices in one market, this may affect the prices charged in other parts of the business.

Effects on retail prices

Looking beyond the effects on input prices for retailers to the effects on prices for end-users, the analysis becomes more complex as one needs to consider the question of the extent to which retailers would pass on the savings. The body of economic research on such pass-on questions is extensive, but in simple terms there are two important parameters. The first is whether the change in costs is firm specific or market-wide. The more likely it is that all retailers face the same reduction in wholesale prices or sourcing costs, the more likely it is that they will pass these on to end-users. The second parameter is the level of competition between retailers, the higher the level of rivalry between retailers and thus the lower their profit margins, the greater the extent to which a change in costs will be passed on.

An economic approach to distinguish between product differentiation and territorial supply restriction

One complaint voiced by retailers in the Financieele Dagblad article is that producers differentiate their offering beyond what is needed to meet the regulatory obligations and consumer tastes, in an effort to carve up the market and improve their market position. This element was part of the Commission’s investigation into AB InBev. It was accused, among other things, of adapting product packaging for the Dutch market to limit the ability of retailers to acquire these products for a relatively lower price in the Netherlands, and sell them in Belgium where wholesale and retail prices are higher. This implies that its producers may change their products in different local markets beyond what may be needed to meet the differences—and that this (excessive) differentiation acts as a hurdle for cross-border sourcing and is therefore a territorial supply restriction.

This article discusses two economic approaches to explore whether a case can be made for excessive differentiation as an infringement of Article 102. The first is based on regression analysis of a dataset comparing prices for the same product across multiple EU member states. The second approach is based on consumer research in one or more national markets into the product attributes most valued by end-users.

Regression analysis of international prices

The first approach uses a dataset of international prices to demonstrate that the price level in a specific country cannot be explained by a selection of market elements such as local taxes, purchase power and other conditions.

This approach was deployed as part of the 2020 European Commission study.¹⁰ The authors built a dataset of prices using publicly available data, and used this to construct a model that aims to explain the price difference between pairs of near-identical products in two EU member states. The researchers use three groups of variables to explain the price difference: the first set of variables looks at local market conditions, the second set looks at the competitive position of the products and retailers and, finally, a separate parameter is included labelled ‘Territorial Supply Restriction’.

The advantages of this approach are that the dataset required is relatively easy to obtain, and that the model can be used to look at multiple markets and products in one and the same model. The main disadvantage is that it will be difficult to create a dataset that is large enough to generate robust results for a single product or family of products. This makes the model well suited to quantify the overall effect, but less suited to individual cases.

Discrete choice analysis of consumer preferences

The second approach looks at dissecting the preferences of consumers in an effort to establish whether the elements on which products differ across markets are relevant to local consumers. This links to the definition of excessive differentiation introduced above: if products differ in ways that do not matter to consumers, this would support the presumption of excessive differentiation.

The objective is, therefore, to observe whether the element in which the products differ between countries is ‘a parameter of competition’. A recent example of such an analysis—albeit in a somewhat different context—is the ACM’s decision to allow an agreement between several firms to discontinue the plastic handles found on soft-drink and water multi-packs.¹¹ A key consideration for the ACM was that ‘the handles (or the ease thereof) do not play a role in the competitive process’.

This type of consumer research is well established in marketing and competition economics¹² and is referred to as discrete choice analysis (and often as conjoint analysis. In short: by repeatedly asking respondents to choose between a limited set of choices that differ in terms of product characteristics, one can infer which product characteristics matter most to consumers.

The main advantages of this approach are that it can be focused on one product or product family and that it is based on a type of research commonly used by marketeers, increasing the likelihood that such analysis is already available or can be used more broadly.

One size fits all?

The topic of territorial supply conditions restrictions continues to generate debate. The topic captures a large group of business practices. It is important to bear in mind that even where such restrictions exist, the economic effects of price differentiation can often be net positive rather than negative. This is a standard result in economics, but one which is somewhat counterintuitive and is often overlooked in mainstream discussions on this topic. This article also focuses on how economic analysis using real-world data can help to assess the extent to which product differentiation across markets is driven by local market conditions or territorial supply restrictions. Ultimately, the question of whether one size indeed fits all is often case-specific and fact-dependent.

¹ Oxera (2022), ‘Fighting inflation with competition: the right tool for the job?’, Agenda, December.

² Braaksma, J. and Hekking, H. D. (2023), ‘Europa snakt naar goedkopere boodschappen maar stuit op inkoopmuren’, Het Financieele Dagblad, 6 January.

³ For a useful introduction to this wider theme, see Huveneers, C. (2020), ‘Territoriale leveringsbeperkingen tussen de Benelux-landen: werkt de interne markt voor iedereen?’, Markt & Mededinging, 3.

⁴ See, for example, Niels, G., Jenkins, H. and Kavanagh, J. (2016) Economics for Competition Lawyers, second  edition, Oxford University Press, March, section 4.5.

⁵ European Commission (2019), ‘Commission Decision of 13.5.2019 relating to a proceeding under Article 102 of the Treaty on the Functioning of the European Union (the Treaty) AT.40134 – AB InBev beer trade restrictions’, 13 May.

⁶ European Commission case AT.40632 Mondelez trade restrictions.

⁷ Ibid.

⁸  Chee, F. Y (2023), ‘Mondelez sets aside 300 mln euro to resolve EU antitrust probe’, Reuters, 3 February.

⁹ Respondents to a survey overwhelmingly expect the removal of territorial supply restrictions to allow them to benefit from a cost reduction and/or source more from countries from which they could not import before. See European Commission (2020), ‘Directorate-General for Internal Market, Industry, Entrepreneurship and SMEs, Study on territorial supply constraints in the EU retail sector: final report’, Publications Office of the European Union, Table 35.

¹⁰ European Commission (2020), ‘Directorate-General for Internal Market, Industry, Entrepreneurship and SMEs, Study on territorial supply constraints in the EU retail sector: final report’, Publications Office of the European Union.

¹¹ See, Authority for Consumers & Markets (2022), ‘ACM is favorable to joint agreement between soft-drink suppliers about discontinuation of plastic handles’, 26 July.

¹² See, Oxera (2022), ‘The use of discrete choice experiments in applied economic analysis’, Agenda, 31 October.