Data science is a broad term that covers a broad set of both classic and novel tools, and techniques. The Compass Lexecon Data Science team was created to bring the latest developments in programming, machine learning and data analysis to economic consulting.
Data is a critical part of every question in economics and business. By applying data science techniques to merger proceedings, antitrust investigations and arbitration matters, we provide a unique value proposition to our clients.
First, data science helps us to be more innovative, developing new types of evidence in response to complex economic questions. In the context of a professional services merger (EC, 2021), we conducted an analysis of closeness of competition by looking at input substitution. To do so, we analysed news articles containing staff moves, and used Natural Language Processing (NLP) analysis to categorise and analyse the movements. In the Nvidia/Arm merger (EC, 2022), we conducted a sentiment analysis to systematically identify how third-party attitudes towards a potential merger changed following announcement of the merger. This involved deploying state-of-the-art deep learning language models.
Second, it helps us be more compelling, i.e., data science improves the quality and robustness of arguments and how we communicate results. In merger proceedings requesting the analysis of competition at the local level, we can quickly build bespoke, sophisticated and interactive maps to visualize the potential impact of the merger and the extent of competitive pressure exerted by the remaining competitors.
Third, data science allows us to be more efficient by automating manual or slow processes such as data collection from web sources or extracting information from large numbers of documents. Examples include a DOJ streaming service merger (US, 2021), on which we conducted a scraping exercise to develop a dataset containing the full set of films hosted on relevant streaming platforms. Similarly, Compass Lexecon has developed a “fuzzy matching” tool, to save critical time and reduce the risk of human error when matching string variables (e.g., customer names) in large, distinct datasets, for example, to assess the closeness of competition between merging firms, or to calculate market shares.
Finally, we conduct research to stay ahead of cutting-edge techniques but also develop data science solutions to boost the use of data for our clients and partners. For instance, we have developed the Compass Lexecon Decision Search Tool , an interactive web-app that allows users to search through competition authority decisions (on mergers, antitrust, state aid, etc) for specific terms or concepts, and efficiently explore results.
By ensuring robust coverage and smart search technology, the Compass Lexecon Decision Search Tool enables competition practitioners to save time by having a single reliable place to find information and avoid either manual review or cumbersome indirect review through general web search engines. In addition to providing significant value to our consulting work, the Compass Lexecon Decision Search Tool has enabled us to look at decisional practice in novel ways, for example by assessing the tools used by competition authorities to conduct market definition exercises.
We are providing this tool complimentary to our clients and partners. If you are interested in obtaining access for your organisation, please contact [email protected].
 Further information available here: https://www.compasslexecon.com/measuring-of-competition-using-natural-language-processing/.
 Further information available here: https://www.compasslexecon.com/providing-new-evidence-using-ai-in-merger-proceedings/.
 Further information available here: https://www.compasslexecon.com/game-set-and-fuzzy-match/.
 Further information available here: https://www.compasslexecon.com/ecmergerdecisions/.
 See Perkins (2021) Market definition in principle and practice, available at: https://www.compasslexecon.com/the-analysis/market-definition-in-principle-and-in-practice/11-15-2021/.