Decarbonisation Technology - August 2023 Issue

particularly strong position. For example, if several refineries wish to share a CO2 or H 2 pipeline, it may reduce the risk to make any such sharing on a FRAND (fair, reasonable and non-discriminatory) basis rather than limiting the sharing possibilities to some exclusive subset of companies within the cluster. In many situations, a more detailed assessment of the full context might be necessary to properly identify not only the full potential for more extensive synergies but also the potential for antitrust risks. Involvement of independent parties In some cases, it may be considered appropriate to have independent entities involved in a co-ordination role between cluster participants, such as the cluster landlord or some form of consultant. Their activities might be to collect intra-day measurement data at set frequencies from cluster sites into a unified data model hosted in a cloud or edge environment. Permitted users can then visualise more granular cluster- wide performance for improved situational awareness of a particular parameter, such as power supply or CO2 emissions. An initial step might involve scheduling planned events, such as shutdowns, into overall production plans. Deeper optimisation algorithms can be built to accommodate unplanned events, such as changeable weather affecting renewable power generation capacity. While the above arrangement might be appropriate for some scenarios of cluster participation and some parameters being co-ordinated among cluster participants, it may not be the case across all clusters. For example, the European Commission has a long history of investigating cartels that were effectively started, co-ordinated, or sustained by ‘industry consultants’. So, due consideration must be given where third-party ‘gatekeepers’ are involved and to the information necessary for an efficient third-party co-ordination role. Given the potential complexity of transfers between cluster participants, each scenario must be assessed on a case-by-case basis. The presence of problematic antitrust situations generally depends on a few factors:  The nature of information being shared. For

example, sharing of information about costs is generally more acceptable than sharing information about prices  Aggregated information is typically more acceptable than detailed, disaggregated information (such as prices for similar products)  Number and nature of cluster participants in question. For example, it is more acceptable for small companies in highly competitive markets, or even better in a supplier-customer relationship, but it is more problematic if firms are direct and strong competitors. Conclusion Industrial clustering offers significant and compelling triple bottom line performance benefits. The scale of benefits reaped will depend on the level of synchronisation between respective cluster participants. However, with greater integration between third-party operations and a reduction in ‘arms-length’ relationships, greater potential exists for antitrust issues to arise. As with most aspects of a legal nature involving multiple parties, the ‘devil is in the detail’, and each situation requires evaluation on a case-by-case basis. Significant clustering benefits may be realised with the least antitrust risk where:  Clear protocols have been established for the analysis and sharing of information  The nature of vertical and horizontal relationships between cluster participants is well understood  The parameter(s) of collaboration, for which information is being shared, pertains to non- core, commoditised ancillary transfers, such as excess power consumption, CO2 emissions, or another common parameter that is very difficult to accurately link back to activities of two or more strongly competing assets in the same cluster. This might include the production or supply volumes of a very specific type of product where the firms are major competitors.

VIEW REFERENCES Patrick Smith patrick.smith@rbbecon.com Duncan Micklem Duncan.Micklem@yokogawa.com