> For the complete documentation index, see [llms.txt](https://laurence-wilse-samson.gitbook.io/textbooks/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://laurence-wilse-samson.gitbook.io/textbooks/antitrust-research-methods/readme.md).

# Introduction

> **Course-tested, last updated November 2026.** One of 22 textbooks freely readable online ([all books](https://lwilsesamson.com/textbooks/)), all written with AI assistance. Corrections are welcome at <laurence@lwilsesamson.com>.

This is a working manual for antitrust and regulation analysis. It grew out of lecture notes and casework, and it puts empirical methods next to the qualitative evidence, case law, and agency practice that decide real matters.

* Audience: upper-level undergraduates, early graduate students, and practitioners.
* Approach: code-forward in R, with "Method," "Qualitative," "Case," and "Debate" boxes in each chapter.
* Jurisdictions: the United States, the European Union, and South Africa carried throughout, with notes on the UK and Asian regimes where they sharpen the point.
* Code, cached analysis files, and the render pipeline are available on request. Captions name the sources, and the public series can be fetched directly. Restricted datasets are not redistributed.

Start with the Preface, or use the navigation to jump to a chapter.

### Status of this edition

I keep a note like this one for each of my textbooks. Each book is in one of four states: *ready to teach*, *complete and under review*, *in development*, or *reference*.

**This book is complete and under review.** The chapters, the code boxes and the empirical appendix are in place, and the key claims in each chapter have been compared with their sources. It grew out of lecture notes and casework, and a nine-hour short course for practitioners is built from it, with slides, exercises and worked answers. It is not yet "ready to teach" in my sense, because no full-semester path through it has been built and taught, and no outside specialist has reviewed it.

* **Prerequisites:** intermediate microeconomics and basic econometrics (regression, hypothesis testing, panel data).
* **Last substantive review:** November 2026, a check of 222 selected claims, of which 4 could not be reached.
* **Teaching materials:** Method, Qualitative, Case and Debate boxes in each chapter; R code boxes; the empirical appendix and templates; and, on request, the short-course slides and exercises.
* **Known limitations:** the claim check covered selected claims and was not a line-by-line proofread; some examples use synthetic data where no open source exists, and say so; case law and agency guidelines change, so check the current text of any guideline before relying on it; nothing here is legal advice.
* **What would move it to "ready to teach":** a section-level core course taught for a full semester, one case worked end to end on real data, and a read by at least one outside specialist.

### About the author

I am an economist in New York. I am Associate Director at Oxford Economics, I teach as an adjunct at NYU Wagner and Columbia SIPA, and I am a Senior Research Fellow at the University of Johannesburg's Centre for Competition, Regulation and Economic Development. I took my PhD in economics at Columbia. More at [lwilsesamson.com](https://lwilsesamson.com/).

### How this book was made

I wrote this book with substantial help from AI language models, principally Claude (Anthropic), which I used for drafting, editing, R code and the data pipeline, worked examples, and figures, and consistency checking. The outline, the argument, the selection of material and the judgments throughout are mine, as is the responsibility for what remains wrong. Figures were generated from code using the sources identified in their captions; illustrative figures are labelled.

In October 2026 I revised the whole text in two further passes, also with AI help: one to remove the stock phrasing that AI drafts produce, and one to bring the prose closer to my own academic writing.

In November 2026 I had the key checkable claims in each chapter (222 in all: numbers, quotations, case holdings, dates and reported findings) compared with a primary source, using AI tools for the lookups and reviewing each change myself. Of those 222, 178 were verified, 29 corrected, 8 softened, 3 removed, and 4 could not be reached from the audit environment. This was a check of selected claims, not a line-by-line proofread, so please check numbers and quotations against the primary sources before citing them.

Please report errors to <laurence@lwilsesamson.com>. Corrections usually go live the same day.

*© Laurence Wilse-Samson 2026. All rights reserved.*

*Last updated: November 2026*