Figures › Maternity, paternity, adoption and other family pay

Statutory pay recovery: small employers' extra compensation

The top-up a small employer may reclaim on top of the statutory pay itself, set to cover the employer National Insurance on it.

9%
Primary source only Statutory Instrument Great Britain in force since 2026-04-06

Checked against its sources: statute verified 2026-09-21

Scope: NOT the 109% an employer reclaims. Regulation 4(b) gives a small employer "(i) an amount equal to such payment, and (ii) an additional amount under regulation 3" — the payment in full plus this 9% — and no instrument anywhere prints the 109% total. That total is the derived companion Statutory pay recovery: small employers (total), recomputed by the build from this record; the eligibility test is Small Employers' Relief: National Insurance threshold and the non-small employer's 92% is Statutory pay recovery: standard rate. Regulation 3 also says what the percentage IS — the estimated secondary Class 1 contributions on statutory maternity pay across all employers, as a percentage of that pay — so it is a compensation estimate that the Secretary of State re-states each year, not a fixed rate.

Current sources

Source A · Legislationhttps://www.legislation.gov.uk/uksi/2026/201/made (SI 2026/201 reg 5: "In regulation 3 of the Statutory Maternity Pay (Compensation of Employers) and Miscellaneous Amendment Regulations 1994 (determination of the amount of additional payment to which a small employer shall be entitled) … (b) for \"8.5 per cent.\" substitute \"9 per cent.\""; in force 6 April 2026, reg 1(3), which brings "regulations 5 and 6(a) and (b)" into force on that day)
retrieved 2026-09-14

History

ValueFromToStatusSources
9%2026-04-06—Primary source only

Receipt

Everything above is reproducible and independently checkable — no trust in this site required (how verification works).

Figure ID
employment.statutory_payments.small_employer_additional_rate
Name in the API
Small Employers' Relief: additional recovery percentage
Record fields (source type · unit · jurisdiction · status)
si·percent·GB·single_source_A
Record hash (canonical JSON, SHA-256)
sha256:b3571063289c0f051f71c916109f8f04560c93b07c4cfd538b4b2ce28124a004
Source snapshots (SHA-256)
Legislation: sha256:09ea408aea566325e5981ec483c97530b6282af24e5e87e146f5a81b22a2788c
Listed in the signed manifest
release 3.0.0 manifest · record_hashes["employment.statutory_payments.small_employer_additional_rate"]
Machine-readable record
/v1/figures/employment.statutory_payments.small_employer_additional_rate · current only: /v1/figures/employment.statutory_payments.small_employer_additional_rate/current
Compiler's notes

RE-REPRESENTED the day it was authored. On the morning of 2026-09-14 this row was written as an SI carry (ADR-0014) with an si_probe block, because the setter substitutes the SPELT percentage — "for \"8.5 per cent.\" substitute \"9 per cent.\"" — and the amendment parser read digits-then-% only, so no derivation existed to beat the carry. That afternoon D21(a) decided a spelt percentage is a rate, the parser read it, and the row was re-authored as the NF-1 derivation it should always have been. ADR-0012: a re-representation keeps the id, and it is no lineage event — nothing was renamed, split, merged or redefined; the value, the date and the definition are the ones the carry carried. The si_probe block is gone with the carry: the build now re-derives 9% from SI 2026/201's own substituted words, which is stronger evidence than a probe of the host's current view and does not expire. Single-source A on purpose: no archived GOV.UK page states this percentage. The recover-statutory-payments guide states the TOTALS an employer reclaims — 92% and 109% — and never the 9% that is the difference, and source_match.states_value answering True for "9%" against that page is right about the bytes and wrong about the law, because the only "9%" on it is the tail of "109%". That sentence belongs to the 109% record and is registered there. ONE value: the setter's for-text attests 8.5% to 5 April 2026, but the instrument that SET it — S.I. 2025/330, named in the 2026 Order's own footnote to S.I. 1994/1882 — is not in the snapshot store, so the earlier value has no start date read from bytes and is not published. Seed 2025/330 and it can be.