Great British Railways will police its own conduct as referee of a ticket market it also competes in. The regulator calls its new rulebook 'enforceable safeguards', though the code never actually requires it to check.
Estimated reading time: 7 minutes
From spring 2027, Great British Railways will decide which independent ticket retailers get access to its systems, data and fares, at the same time as competing with every one of them through its own ticket website and app. On 9 October 2026 the Office of Rail and Road opened a consultation on the rulebook meant to stop that arrangement favouring GBR: a draft Retail Code of Practice, binding on GBR alone through its operating licence, open for comment until 11 December 2026 [4]. Announcing it, ORR's Director of Strategy, Policy and Reform said the draft "will provide enforceable safeguards to help maintain a fair and open online retail market" [2]. Read the 20-page Code itself, and those safeguards turn out to rest almost entirely on GBR marking its own homework.
A referee that also plays
GBR's two roles are stated plainly in the Code's own opening section: it is both "an online retailer" and "the provider of online retail market-wide management functions in Great Britain", meaning it will run the central systems, data and licensing arrangements that independent retailers, such as Trainline, need to sell tickets at all [1], Section 1.1. ORR's press release is specific about the scale of the switch: GBR's own online retailer "will replace the 14 websites and apps of DfT-contracted train operators" [2], consolidating most of England, Scotland and Wales's passenger rail retailing under one organisation that is also a competitor in that same market. The Code's six principles require GBR to manage that conflict of interest, give retailers fair and transparent access to products, systems and data, and not confer "undue preference or advantage" on itself [1], Section 4. None of that is controversial as a statement of intent. The question a teardown has to ask is what happens when GBR falls short of it.
Who checks the checker
The Code's compliance section answers that question, and the answer is: mostly GBR, checking itself. "GBR must establish and maintain arrangements to monitor its compliance with this Code on an ongoing and proactive basis" [1], Section 9.1, and must "be able to demonstrate" to ORR, on request, how it has handled issues its own monitoring throws up [1], Section 9.3. ORR's own role, by contrast, is framed as optional throughout the document. Five separate clauses give ORR the power to review GBR's governance, access arrangements, data handling and compliance records, and every one of them uses the same phrase: "ORR may at its discretion review" [1], Sections 1.4, 5.21, 6.27, 7.8 and 8.29. Nothing in the Code obliges ORR to check anything on a schedule; it is entitled to look, not required to.
Even when a non-GBR retailer formally raises a "Code compliance concern" with ORR, the consequences on offer are thin. Section 8.20 lists six possible outcomes, from "take no further action" through "continue to actively monitor" and "require GBR to undertake further review", to the only two with any bite: "issue a warning notice" and "consider whether licence enforcement is appropriate" [1], Section 8.20. Consider, not escalate to. And "licence enforcement" is not a power this Code creates: the Code's own glossary defines it as "licence enforcement by ORR under section 55 of the Railways Act 1993" [1], Section 3.8, a general power ORR already holds against any operator for any licence breach, Code or no Code. Section 9.12 drives the point home: "Nothing in this Code limits ORR's discretion to decide whether, when or how to take licence enforcement action" [1]. Strip away the self-monitoring duty GBR already owes itself and the pre-existing statutory power ORR already held, and what the Code actually adds is a formal complaints channel and the option of a warning letter.
There is also a live reminder of how provisional the whole structure still is. Section 2.1 ties GBR's duty to comply with this Code to a piece of legislation that has not yet passed: "The [Railways Bill] provides for GBR's compliance with this Code to be secured through its licence" [1], brackets in the original. Checked directly against Parliament's own record on 9 November 2026, the Railways Bill is at report stage in the House of Lords, with further sittings scheduled for 18 November, 8 December and 15 December 2026, and still needs third reading, and any Commons consideration of Lords amendments, before Royal Assent [5]. ORR's consultation on how GBR's conduct will be enforced closes 11 December, in the middle of that unfinished passage, for a body that does not yet exist in law to hold the licence the whole Code depends on.
The case for it is its own admission
ORR published a draft impact assessment alongside the consultation, and it diagnoses the underlying risk accurately. GBR's combined roles, it says, "create potential risks relating to conflicts of interest, self-preferencing, restricted or less favourable access to products, systems and data, information handling, transparency and accountability" [3], paragraph 11. What follows that diagnosis is less assured. The document states plainly that the expected benefits of introducing the Code "have not been quantified" and have instead been "subject to a qualitative assessment" [3], paragraph 6. Its evidence base is a round of "detailed workshops held between March and May 2026 with a small number of representative stakeholders", discussions the document itself calls "exploratory" and warns are "not necessarily representative of the market as a whole" [3], paragraph 17.
More striking than what the assessment admits is what it declines to touch. Government's decision to let GBR run both the market-wide retail infrastructure and its own competing retail business, the single choice that creates the conflict of interest the whole Code exists to manage, is ruled out of scope in the assessment's own words: "This impact assessment... does not assess, review or comment on the Government's settled policy position on this matter" [3], paragraph 3. The one document meant to test whether the Retail Code actually works is built on an unquantified, self-described non-representative evidence base, and it is barred from questioning the structural decision that made a Code necessary at all.
The claims, tested
| The document's own words | What the record shows | Verdict |
|---|---|---|
| ORR's press release: the draft Code "will provide enforceable safeguards to help maintain a fair and open online retail market, supporting passenger choice" [2] | The Code puts GBR in charge of monitoring its own compliance (Section 9.1), makes every one of ORR's review powers discretionary ("ORR may at its discretion review", Sections 1.4, 5.21, 6.27, 7.8, 8.29), and names only a warning notice and "consider[ing] whether licence enforcement is appropriate" as consequences for a confirmed breach (Section 8.20), with licence enforcement itself an ORR power that already existed under section 55 of the Railways Act 1993 (Section 3.8) [1] | "Enforceable" describes a power ORR already had and may or may not use; what the Code adds is a complaints process and a warning letter, with GBR grading its own homework in between |
| The draft impact assessment correctly diagnoses "potential risks relating to conflicts of interest, self-preferencing, restricted or less favourable access to products, systems and data" from GBR's dual role (paragraph 11) [3] | The same document admits its benefits case "have not been quantified" (paragraph 6), rests on "exploratory" workshops with "a small number of representative stakeholders" that are "not necessarily representative of the market as a whole" (paragraph 17), and states it "does not assess, review or comment on" the decision to let GBR hold both roles at once (paragraph 3) [3] | The document meant to test whether the fix works rules the underlying design decision out of scope, then scores the fix against a sample too small to call representative |
Credit where due
Parts of the Code are specific rather than vague. Retail licence applicants must be told, up front, the information required, the costs involved, the decision criteria and "the indicative timing of key stages... and when a final outcome will be reached" [1], Section 6.7. Product trials GBR runs itself must have "a clear and reasonable end date" rather than running indefinitely [1], Section 6.14. A non-GBR retailer can raise a formal Code compliance concern with ORR even where a dispute also falls under a separate licence or contract, rather than being funnelled exclusively into private arbitration [1], Section 8.3. The consultation is also genuinely open to challenge on its own weakest point: ORR says responses "can include views on this impact assessment" itself, not just the Code's drafting [4], and the scope section is honest that the Code "does not replace or limit the application of legislation, competition law or other regulatory requirements" [1], Section 2.7, rather than overselling what one code of practice can do on its own.
Verdict
Four stars, from two documented mistakes. ORR has correctly identified a genuine structural problem: a single organisation that will set the rules of the online rail ticket market and compete in it at the same time. The draft Code's access and data provisions are detailed enough to be checkable, and the consultation invites scrutiny of its own impact assessment rather than hiding it. But a regulator that calls its own proposal "enforceable safeguards" while writing a Code in which it is never obliged to check, only entitled to, and in which GBR is handed the job of policing itself, is overselling the one word that matters most. And the impact assessment meant to justify the whole exercise admits its evidence is thin and declines to examine the decision that created the risk in the first place.
The star score counts two documented mistakes. First, ORR's press release calls the draft Retail Code "enforceable safeguards", but the Code text requires GBR to monitor its own compliance, makes every one of ORR's review and intervention powers discretionary rather than mandatory ("ORR may at its discretion review", appearing five times), and names only a warning notice and consideration of an already-existing licence-enforcement power as consequences for a confirmed breach. Second, the Code's own draft impact assessment states that its expected benefits "have not been quantified", that its evidence base is exploratory workshops with "a small number of representative stakeholders" it describes as "not necessarily representative of the market as a whole", and that it does not assess the Government's decision to give GBR both roles at once, the decision that created the risk the Code exists to manage. Two falls in the 1 to 3 band: four stars; the bands are on the ratings page. This piece makes no finding about the conduct of any named individual; the drafting choices examined are institutional, and the one named official quoted is quoted making a public statement in her professional role, not criticised. All five sources below were fetched directly on 9 November 2026; the Parliament bills.parliament.uk page returns a bot-blocking response to automated fetches but is reachable in a browser, so its current stage was independently confirmed via Parliament's own public Bills API.
Sources
- Retail Code of Practice: draft for consultation (PDF), Office of Rail and Road, 9 October 2026
- ORR seeks views on new safeguards for fair rail ticket market under Great British Railways, Office of Rail and Road, 9 October 2026
- Draft Impact Assessment: Retail Code of Practice (PDF), Office of Rail and Road, November 2026
- Draft Retail Code of Practice for GBR (consultation page), Office of Rail and Road
- Railways Bill, UK Parliament