Fideres reflects on the new European guidelines on passing-on damages


Summary

On 2 July 2019, the European Commission (“EC”) published a revised set of guidelines for national courts on how to deal with pass-on in the context of competition claims. In this note, Fideres provides: 

§  A summary of the key aspects of pass-on as mentioned in the EC guidelines

§  An overview of the main challenges when estimating how much of an overcharge has been passed on, and how to deal with those in practice, especially when data availability is limited


The Commission’s guidelines: what’s new?

The EC previously published a set of draft guidelines on 5 July 2018. While the draft guidelines and the recent publication remain substantively the same, the following points have been expanded and are now explained in greater detail:

§  Importance of quantifying the loss of sales that is accompanied by an increase in prices

§  Additional examples of how to deal with fixed and variable costs when determining pass-on

§  Different techniques used to carry out pass-on analysis in practice

§  Simulation approach as an additional way to quantify pass-on.

Overview

Why is pass-on important?

Pass-on is important for the calculation of damages in competition claims. As explained in further detail below, pass-on is used by  defendants, to argue that claimants have passed on most or all of the overcharge and therefore are not damaged, as well as by claimants, to show that the overcharge has been passed on to them (in case of indirect claimants) or has not been passed on further downstream (in case of direct claimants).

As per the guidelines, national courts are required to rely on economic analysis and make reasonable assumptions regarding pass-on estimation to avoid both over-compensation and under-compensation to the parties involved.

Overcharge and pass-on

The difference between the price actually paid, in the presence of anti-competitive conduct, and the price that would have prevailed in the absence of the infringement is referred to as the “overcharge”. When direct purchasers increase their sale prices to indirect purchasers, this behavior constitutes pass-on of overcharges. The total value of sales lost as a result of the direct purchasers increasing prices is known as the “volume effect”.

Pass-on: theory and estimation of quantum

 Economic theory

According to economic theory, the most significant factors that affect the existence and magnitude of pass-on are:

§  Input costs: it is important to ascertain the nature of input costs for the direct purchaser, i.e., whether the direct purchaser’s costs vary with quantity (variable) or remain constant (fixed). For example, if the cartel overcharge affects the direct purchaser’s fixed costs, it is less likely to be passed on as these costs do not typically affect their price setting in the short run.

§  Nature of demand faced by the direct purchasers: the extent to which a direct purchaser can raise its own price while facing an overcharge depends on how demand responds. For example, if the direct purchaser faces sensitive or elastic demand, the overcharge is less likely to be passed on.

§  Strength and intensity of competition in downstream markets: the degree of passing-on depends on the level of competition between direct purchasers. For example, if there is fierce competition amongst direct purchasers, they would be less likely to increase their prices and pass-on overcharge, to remain competitive and maintain their market shares.   

The chart below lists the important markers highlighted by the EC in assessing the degree of pass-on:

 Estimation of pass-on quantum

There are broadly two scenarios in which national courts deal with pass-on: 

According to the EC guidelines, in order to estimate pass-on related price effects, national courts must consider the following two main approaches:

Comparator-based methods

Comparator-based methods involve analysing the prices set by the direct purchaser during the infringement period and the prices set in similar markets. Qualitative evidence such as internal documents that reveal key information about pricing decisions may also be relevant, but in terms of quantitative evidence, the EC has identified that regression analysis is the best technique for courts to rely on.

The different approaches to determine but-for prices under comparator-based methods are outlined below:

Passing-on rate

The passing-on rate approach involves analysing how previous changes in a firm’s costs have affected prices before or after the infringement period. For instance, the passing-on rate may be estimated by analysing how historical changes in the cost of copper have affected the price of wire harnesses. If an increase in the cost of copper by GBP 10 is followed by a price increase of wire harnesses by GBP 5, this constitutes a pass-on rate of 50% to the car manufacturer.

Estimation of the volume effect

The volume effect represents the lost volume due to an increase in price set by the direct purchaser. Volume effect is estimated by considering:

§  The change in quantity due to increased prices

§  The counterfactual margin that the purchaser would have achieved in the absence of the infringement without any pass-on.

In Cheminova (2015), the expert estimated the volume effect as a product of the counterfactual margin and the lost quantity of products sold due to pass-on. Sales in this case were calculated based on a market price elasticity estimate along with the observed prices and quantities.

Pass-on in practice

In general, many of the methods mentioned above are feasible only if they are “proportionate” to implement. When assessing proportionality of an order to disclose information, the court must consider choice of economic method and approach as per the guidelines. For example, what may be proportionate for a EUR 20m claim may not be the same for a EUR 200,000 claim. One of the main challenges that experts face is the lack of reliable data, especially at an early stage (i.e. pre-disclosure).

To understand the future scope of a claim, practitioners must use publicly available information, which is often limited, to arrive at a reasonable overcharge or pass-on estimate, before decided whether it is worthwhile to pursue a claim. 

We discuss two possible alternative approaches in the following section.

Alternative approaches

Producer versus consumer prices analysis

According to the Bureau of Labor Statistics (“BLS”), the Consumer Price Index (“CPI”) measures the changes in prices paid by consumers. The Producer Price Index (“PPI”) measures the average change in selling prices received by producers. Using national CPI and PPI data, we can estimate but-for prices and pass-on rates for retail or producer markets. Naturally, producer prices are a good indicator of retail prices within the same market. Similarly, producer raw material costs are a good indicator of producer prices.

By mapping within-nation CPI and PPI data, we use regression analysis to predict but-for prices that would have otherwise been paid by indirect consumers to the direct purchaser (“But-For” CPI) by accounting for the cartelised input cost (PPI). We can then estimate these model parameters to ascertain pass-on in different sectors and sub-sectors of product categories, according to levels of the supply chain that were affected. To bolster the standardized regression model, we can also add market-specific demand, supply and structural factors to create a sophisticated but-for model. Examples include quantities demanded and supplied; imports and exports; exchange rates; raw material costs; prices of substitutes and inflation.

On 31 October 2017, the Canadian competition authorities publicly announced that they were investigating seven bread suppliers and retailers for price fixing. A separate application for class certification and damages, on behalf of consumers for bread and related products, was also lodged in an Ontario court on 21 December 2017. Using this ongoing case as an example, we illustrate how our standardised model can be used to estimate but-for prices:

VAT and CPI

We can track changes in CPI due to changes in indirect taxes such as Value Added Tax (“VAT”). Economic literature suggest that firms generally increase prices in response to VAT increases, in order to pass on the incidence of the tax to the consumers.

In order to estimate pass-on in a given product category, Fideres has assessed the following key variables:

§  Historical VAT changes on sector and sub-sector CPIs

§  Control variables including the corresponding sector PPIs to adjust for any non-VAT cost shocks

Certain pitfalls, however, make this type of analysis more complex that it might appear at first glance. These include:

§  Imports/exports: typically, CPIs include import prices and PPIs include export prices – one must consider this when evaluating a national cartel

§  PPI-CPI mapping: not all categories of CPI have an exact corresponding match with PPI

§  Pass-through: the percentage price increase passed down the supply chain is likely to vary and is dependent on demand and supply elasticities, thus some divergence may be expected

§  Time lags: CPI data may lag PPI data – the time lag may considerably vary depending on the market in question

§  Sales and excise taxes: these are accounted for in CPI, but not PPI calculations

Conclusion

Pass-on can be difficult to quantify, mainly due to high data requirements. The alternative approaches discussed in this note can be useful because:

§  They rely on publicly available data, which is reliable and cost-effective

§  They can be used across a variety of markets and the breakdowns in product categories allow for analysing different levels in the supply chain

They are based on regression techniques, which, according to the EC, are best suited to provide reliable quantitative but-for prices, in the context of antitrust legal disputes.