Ombudsman: 58% of environment recommendations ignored

TL;DR

Reporting on the Parliamentary Ombudsman’s 2025 annual report said that 58% of the Commissioner for Environment and Planning’s justified recommendations went unimplemented, the highest rate of any commissioner, and that 22 reports were escalated to Parliament [2]. We read the report itself [1] and recomputed the figures from its tables.

1. The number is the Ombudsman’s own. Table 1.3 of the 2025 report [1]: of 12 sustained cases closed by the Commissioner, 7 recommendations (58%) were not implemented. The 22 reports sent to Parliament match Table 1.22.
2. It is a small count, and it moves a lot. 7 of 12 is a 95% interval of roughly 32–81%. The same office had 6 of 13 (46%) in 2023 and 2 of 8 (25%) in 2024. Over 2023–2025 it is 15 of 33 (45%).
3. It is a snapshot at closure. The Commissioner’s own chapter says that five of the 12 were still unimplemented when the report was written; two more were implemented after being referred to the House. On that basis 42% remain open (counting the one “partly implemented” case of Table 1.3 as implemented).
4. “Highest” holds on the Ombudsman’s table, not on every denominator. Environment and Planning 58% against Education 53% of sustained cases. Counting only cases where a recommendation was made, Education is 67% (8 of 12). The Commissioner’s remit is also wider than “environment”: planning, building, transport licensing and heritage.
5. Verdict: largely supported (high confidence). The figures match the primary source; the caveats above do not change the substance.

1. Supported
2. Largely supported
3. Not substantiated
4. Misleading
5. Contradicted

IN THIS REPORT

1 The statement and the primary source

Our candidate record took the figures from a Newsbook report of 2 July 2026 [2], which attributes them to the Parliamentary Ombudsman’s Annual Report for 2025. We read the report [1], published on the Office’s website. The Ombudsman’s wording is below. The newspaper’s phrase “justified recommendations” is a paraphrase of “sustained cases”.

2 Method

Question. Do the Ombudsman’s tables support “58% unimplemented”, “highest among commissioners” and “22 reports escalated”, and how stable is the figure?

Data. We typed Table 1.3 (sustained cases closed, by office and outcome) from the 2023, 2024 and 2025 annual reports [1, 3, 4] and Table 1.22 (reports sent to Parliament) from the 2025 report into data/cc-102/ombudsman_sustained_cases.csv, and checked that every row adds up. tools/cc-102-report/calc.py recomputes the rates on two denominators and checks the 2025 totals (81 sustained cases, 22 not implemented, 22 reports). Intervals are Wilson 95%.

Grades. The Ombudsman’s annual reports are official institutional records (C). We have no access to the underlying case files; the case notes were not needed for the counts.

3 What the 58% counts

Table 1.3 counts sustained cases closed during the year, not complaints received and not individual recommendations: one case can carry several. “Not implemented” is the status when the case was closed. The Commissioner’s chapter [1, printed pp. 218–219; PDF pp. 219–220] gives the position later: of the 12 sustained cases, five recommendations were implemented, two more were implemented after referral to the House of Representatives, and five remained unimplemented (including the Planning Authority and the Transport Malta cases it lists). The two accounts reconcile only if the one “partly implemented” case is counted as implemented: Table 1.3 has 4 implemented + 1 partly = 5, which is the chapter’s five, and 7 not implemented at closure = the chapter’s 2 implemented later + 5 still open. The 42% (5 of 12) depends on that reading. The Commissioner’s remit covers planning, building control, transport licensing and heritage as well as environmental matters; the Planning Authority alone accounts for 37 of 86 complaints in 2025 [1, Table 3.4].

4 Results

Source: Table 1.3 [1]; recomputed in data/cc-102/rates.csv. Cases with a recommendation = sustained cases minus those sustained with no recommendation.

What this means for the statement

• The figure is correct as the Ombudsman states it. Both the 58% and the 22 reports match the tables.
• “Highest” is true on the Ombudsman’s own basis (58% against Education’s 53%), but the margin is one case’s worth and reverses if cases with no recommendation are excluded (Education 67%).
• “Went unimplemented” is a status at closure. By the time the report was written two of the seven had been implemented, leaving five (42%).
• Newsbook’s “justified recommendations” reads as a count of recommendations; the table counts sustained cases.
• The 22 reports equals the total of not-implemented cases in Table 1.3, but the split by office differs in two places (Parliamentary Ombudsman 5 vs 4, Health 2 vs 3). Environment and Planning agrees at 7. We report this as a reconciliation point, not an error in the claim.

5 Testing the sub-claims

6 Verdict and requests for evidence

Why. The 58% and the 22 reports are in the Ombudsman’s own report and recompute exactly. We rate the statement Largely supported rather than Supported because it is a snapshot of a very small count, two of the seven were implemented later, and “highest” is not robust to the choice of denominator. None of these changes the substance: a majority of this Commissioner’s sustained cases closed in 2025 without the recommendation being implemented.

What this verdict does not say. It does not say the authorities were wrong not to act, or that the Commissioner’s findings were right in each case. It is not a judgement of any authority or person.

Evidence we are asking for

• From the Office of the Ombudsman: the list of the 12 sustained cases of 2025 with their status at closure.
• From the Office of the Ombudsman: how the per-office counts in Tables 1.3 and 1.22 reconcile.
• From the authorities concerned: the reasons given for not implementing the five open recommendations.

7 Limitations

• Figures are typed from the published tables; we did not see case files. Counts are of cases, not of recommendations.
• Table 1.3 uses status at closure; later implementation is reported only in the Commissioner’s narrative.
• Cases closed in 2025 may have been opened in earlier years, so rates are not a measure of 2025 conduct alone.
• A rate on 8–13 cases a year cannot support a trend claim. We tested no individual case on its merits.

References

1. [1] Office of the Ombudsman, Malta (2026). Annual Report 2025, Table 1.3 (pp. 15–16), Table 1.22 (p. 37), Commissioner for Environment and Planning chapter (pp. 214–220), Table 3.4. Page numbers are the printed ones (PDF page = printed page + 1). Presented to the Speaker of the House, June 2026. (Full text read.) https://ombudsman.org.mt/media/0ownfd4k/digital-annual-report-2025-en.pdf
2. [2] Newsbook (2 July 2026). Public entities ignore Ombudsman as defiance reaches record high. https://newsbook.com.mt/en/public-entities-ignore-ombudsman-as-defiance-reaches-record-high/
3. [3] Office of the Ombudsman, Malta (2025). Annual Report 2024, Table 1.3. https://ombudsman.org.mt/media/q4nbhpzv/annual-report-2024-ombudsman.pdf
4. [4] Office of the Ombudsman, Malta (2024). Annual Report 2023, Table 1.3. https://ombudsman.org.mt/media/fr1ndd4u/annual-report-2023-en.pdf
5. [5] Miżien. Analysis code and outputs: tools/cc-102-report/; data/cc-102/.

Appendix A: series standards and verdict scale

Every investigation in this series follows the same rules so that verdicts are comparable and challengeable.

Confidence

High: multiple independent lines of evidence agree. Moderate: evidence is relevant but incomplete or indirect. Low: evidence is thin, second-hand or conflicting.

Evidence grades

A experiment · B observational study with a control or gradient · C review, guidance or official statistics · D assertion or anecdote. The Ombudsman’s annual reports are official institutional records (C).

Standards

• Quote the speaker, not the headline, with outlet and date, and link the original where possible.
• Separate the claim from the evidence offered for it, and say who made which statement.
• Grade the evidence and show differing views side by side where the evidence is unsettled.
• State what we could not access or verify, and what would change the verdict either way.
• Ask for the missing evidence in specific, answerable requests.
• Offer a right of reply where a check finds a claim Not substantiated, Misleading or Contradicted (or a pledge Not measurable, Off track or Missed), before wider circulation, and append responses.
• Correct errors openly with a version number and change log.
• Whistle-blow only on evidence: a verdict of Misleading or Contradicted requires documents or data that can be shown, not inference.
• Check every side. Government, opposition, NGOs and developers are held to the same standard.

Appendix B: revision log