The report nobody read

What RTS 28 promised, and what it delivered

From 2018, the European Union required investment firms to publish, once a year, the five execution venues they used most for each class of financial instrument, together with a summary of the execution quality obtained there — a disclosure known as RTS 28, the firm-level half of a pair with RTS 27, which required trading venues themselves to publish quarterly execution-quality data.¹ The premise was straightforward: if a firm's own conduct was visible, a client, a competitor, or a journalist could hold it to account.

The premise did not survive contact with practice. RTS 27 was suspended across the European Union in 2021 as part of a broader "quick fix" to MiFID II's reporting burden, with the legislation's own recitals acknowledging that the reports were seldom read and did not let investors make any meaningful comparison between venues.² RTS 28 stayed on the books longer, but industry submissions to European regulators' own consultation on a replacement noted that supervisory attention to RTS 28 compliance had already been deprioritized before its formal repeal, and made much the same criticism from the other side: the disclosures were too granular for most clients to use, and too generic to support any real comparison between brokers.³

The mechanism that has replaced it

On April 14, 2026, the European Commission adopted a new regulatory technical standard that repeals the delegated regulations underlying both RTS 27 and RTS 28 outright, replacing them with a single standard specifying the criteria for establishing and assessing the effectiveness of firms' own order execution policies.⁴ The standard is still moving through the European Union's ordinary review process — the Council and Parliament retain the right to object — and takes effect on the timeline the Commission itself has set out in the regulation; it is not yet binding on firms.⁵

What the new standard actually requires reads as more prescriptive on process than RTS 28 ever was on outcome. Firms must document the governance behind how they select execution venues, maintain a list of those venues and the circumstances under which each is used, monitor the policy's effectiveness at least quarterly, complete a full assessment at least annually or whenever something material changes, and — for any firm that routes to a single venue — periodically justify that choice against the alternatives.⁶ None of that is lighter-touch than an annual public disclosure. It is a different kind of scrutiny: continuous and internal, rather than periodic and public.

The American model, minus the report

The United States never had an RTS 27/28 equivalent to begin with. Under Financial Industry Regulatory Authority (FINRA) Rule 5310, a member firm must use "reasonable diligence" to find the best market for a security and execute so the resulting price is as favorable as possible under prevailing conditions, weighing factors that include the character of the market, the size and type of the order, and the accessibility of quotations.⁷ Where a firm does not review execution quality order by order, it must instead conduct a "regular and rigorous review" of that quality, at minimum quarterly, security by security and order type by order type — an internal obligation with no public report ever attached to it.⁸

Two adjacent US rules are easy to mistake for a domestic version of the European disclosures, and the distinction is worth holding onto. Rule 605 publishes exchange-level statistics on execution speed and price; Rule 606 requires brokers to disclose how they route client orders. Both are genuinely public and genuinely about execution quality, but neither certifies that best execution was achieved for any specific client — they are transparency about routing and market structure, not evidence the underlying duty was met.⁹ The country that never asked for a comparability report has been quietly running, for years, the model the European Union has only just adopted.

Is the duty more subjective now?

The United Kingdom offers a useful control, because it ran this experiment first. The Financial Conduct Authority (FCA) removed its own RTS 27 and RTS 28 publication requirements from its rulebook in December 2021, five years ahead of the European Union, leaving the underlying "all sufficient steps" standard resting on a firm's execution policy and the FCA's own direct supervisory testing rather than a published report.¹⁰

That direct testing produced a genuinely useful data point in December 2025, when the FCA published the findings of a multi-firm review of how wholesale banks deliver best execution in UK listed cash equities — its first such review since 2014. Practices had generally strengthened since then, but the area where the FCA found the least progress was governance and oversight: the discipline of building a monitoring process, giving it real authority, and acting on what it turns up.¹¹ That finding was only possible because a supervisor went and inspected the process directly. A published comparability report, however detailed, was never going to surface something as specific as which committees actually challenged a bad outcome.

Put the three regimes together and the honest answer to the subjectivity question resists the headline version. The underlying legal test has not loosened anywhere: "all sufficient steps" and "reasonable diligence" were always judgment calls, weighted differently by order, client, and instrument, and neither regulator ever tried to reduce that judgment to arithmetic. What has moved is the artifact used to check whether the judgment was exercised well — from a static, comparability-oriented public report to a continuous, internally run process a supervisor inspects directly. The discretion that remains sits in methodology: which comparator venues a firm chooses for its own transaction cost analysis, what counts as a deficiency worth escalating — and that discretion was never pinned down by the old disclosures either. What has genuinely changed is who gets to see the working, and a regulator with the power to inspect a live process arguably sees more than a client with a table of numbers ever could.

The shape of it

None of this evidencing — old report or new process — was ever really about the report itself. A published top-five list was only ever a proxy for something a client actually needs to know: was there a real, working process behind this trade. Removing the proxy does not make that process any easier to fake; if anything, it removes the option of letting a published table stand in for a process that was not really being run. What a firm owes now was arguably always this: a live monitoring practice, and a record underneath it complete enough to survive a regulator asking to see it work.