loading
Whether it worked
Most rules come with a promise: in five years, go back and check whether this worked. The answer is published as a post-implementation review. This is what those checks found — and how rarely anything changes as a result.
We read every review published on legislation.gov.uk in 2026 — 38 catalogue entries covering 35 separate reviews — and coded each one the same way: what kind of measure, what kind of review, what it recommends, whether the objectives were met, whether the Regulatory Policy Committee saw it, and what it says the rules cost business. Not one recommended removing a regulation.
This is stage three of Whether it worked. The forecast came first and the RPC’s verdict on it second, both before the rule took effect. This page is what the department says afterwards — and the National Audit Office is the only stage where somebody outside the department asks the same question.
reviews recommended removing the regulation they examined. Twenty-five recommended keeping it exactly as it is, seven proposed amendments, one a replacement, and one split the call regulation by regulation.
The pattern
Twenty-five of 35 recommend keeping the regulation unchanged, and none recommends removal. Even the reviews that find real problems tend to land on Keep: the care-home visiting review records that objectives were only partly met and that further action is required, then recommends keeping the regulation as drafted.
Only 18 conclude the objectives were met. Fourteen say partly, two say no, and one — the onshore fracking protected-areas rules — cannot say at all, because the moratorium means the regulation has never operated. It was still recommended for keeping.
Response rates are routinely too low to support a conclusion: 8 of 26 for the honey regulations, 11 of 70 organisations for the Human Medicines Regulations, with no responses at all from patient organisations. Two reviews collected no new evidence, resting instead on the consultation done before the rules came in.
Only nine reviews state an equivalent annual net direct cost to business, and five of those repeat the original impact assessment's estimate rather than test it. The second review of the construction design and management regulations carried forward a 2018-price figure without any fresh cost–benefit analysis at all.
Seven reviews carry an RPC opinion, and every one is green. The rest record N/A, a de minimis exemption, or — in two cases — have no RPC box on the form. The stage of the cycle where a department assesses its own performance is the stage least likely to be checked by anyone else.
The Farming Rules for Water review was due in April 2021 and appeared in May 2026, roughly five years late, with COVID-19 given as the reason. The rules on compulsory basic training for motorcyclists received their first ever review this year, some 35 years after they came into force.
Every review
One row per catalogue entry, grouped by department. The first two columns say what was reviewed and under what duty; the last four are the findings. Hover or focus any cell for the wording behind it, and follow the title to the review itself.
Type of regulation and type of review are taken straight from the PIR front sheet, as are the recommendation and the RPC opinion. "Policy objectives met" has no box on the template — it is coded from each review's own conclusion, and the wording behind every call is in the hover text.
| What was reviewed | What the review concluded | |||||
|---|---|---|---|---|---|---|
| Review | Type of regulation |
Type of review |
Recommendation | Policy objectives met |
RPC opinion |
Cost to business |
Find it on this site
Say what you are after and I will link you straight to it
Support
Plans, signing in, what the data covers — and your own account