The National Insurance exemption on salary-sacrificed pension contributions is being capped at £2,000 a year from April 2029. The three-year run-up is when scheme reviews need to start, rather than a scramble in 2028.

In reading this article you will understand:

  • What the National Insurance Contributions (Employer Pensions Contributions) Act 2026 actually caps, when it takes effect, and who pays National Insurance on contributions above the £2,000 threshold.
  • Which employer clients are most exposed to the change, based on how HMRC and the IFS project it will land across income bands and sectors.
  • How to use the three-year run-in to review salary sacrifice scheme design with employer clients now, rather than waiting for the cap to become a live payroll problem.

From 6 April 2029, the National Insurance exemption on salary-sacrificed pension contributions will be capped at £2,000 a year per employee. The National Insurance Contributions (Employer Pensions Contributions) Act 2026, announced at the Autumn Budget on 26 November 2025 and given Royal Assent on 29 April 2026, is already law; only the commencement date is still three years out. For corporate advisers, that gap is the opportunity. A scheme design decision that would look reactive in 2028 is proactive work today, and the employer clients who move early will have more options open to them than those who wait for HMRC’s implementation guidance to force the issue.

The Current Position

Salary sacrifice works by an employee giving up part of their gross salary in exchange for an equivalent employer pension contribution, which keeps the whole amount out of both employee and employer National Insurance rather than only attracting income tax relief. It is now very widely used: HM Treasury has estimated that £32 billion of pension contributions ran through salary sacrifice arrangements in 2024, and 41 of the 51 businesses HMRC interviewed for its research into employer behaviour already operated pension salary sacrifice. The exemption currently has no cap. Every pound an employee sacrifices, however large, escapes employer National Insurance, 15% above the £5,000 secondary threshold for 2026/27, and employee National Insurance, 8% up to the upper earnings limit and 2% above it, in full.

What’s Changing, and When

From April 2029, that exemption stops at £2,000 of employee salary-sacrificed pension contributions a year. Anything sacrificed above the threshold will be treated as an ordinary employee pension contribution for National Insurance purposes, meaning it becomes subject to both employee and employer NICs at the standard rate; income tax relief on the full sacrificed amount is unaffected. Employer contributions, whether a standard matched contribution or an enhanced one funded by restructuring pay, stay entirely outside the cap. It is only the employee’s own salary-sacrificed amount that is limited, not the scheme’s total contribution rate. The £2,000 threshold carries no indexation, and the Act contains no automatic uprating mechanism, though the limit itself is set by regulations and could in principle be changed later. On the current design its real value falls year by year, and the number of members caught by it rises.

Deloitte’s modelling of an employee earning £50,000 who sacrifices 10% of salary, £5,000 a year, shows the scale involved. Once the cap applies, £3,000 of that sits above it, adding roughly £240 a year in employee NICs and £450 in employer NICs, plus £15 in Apprenticeship Levy for employers with a pay bill over £3 million. Multiply that across a scheme’s higher-sacrificing members and the employer’s NIC bill rises accordingly, before any decision is even made about passing the employee’s share of the increase back through pay.

Who This Actually Hits

The impact skews heavily towards higher earners, which is the justification HM Treasury gives for the change: the cost of the relief is concentrated among those on higher incomes. The IFS’s analysis of the reform found fewer than 1% of the bottom fifth of earners currently sacrifice above £2,000 a year, against 48% of the top 10%; across all employees, around 15% sacrifice above the threshold. It also varies sharply by sector: about 18% of private sector employees are affected against 7% in the public sector, rising to roughly 40% in finance and insurance and in information and communication, against under 2% in accommodation and food services. The OBR expects the cap to raise £4.7 billion in 2029/30, falling to £2.6 billion the following year. Part of that step down reflects timing effects in the first year, and part the expectation that employers and members will restructure rather than simply pay more.

How Employers Are Likely to Respond

HMRC’s pre-legislation research with employers is instructive here. Presented with several options for restricting the relief, businesses rated a £2,000 partial cap the most workable of those tested and preferable to removing the exemption altogether. Worth noting, though, that the version HMRC tested removed employee NIC relief above the threshold only, where the enacted cap removes employer relief as well. That doesn’t mean employers will simply absorb the cost. Some are expected to keep salary sacrifice for the exempt first £2,000 and move higher-sacrificing members onto conventional net pay contributions above it. Others may fund more of the total contribution directly from the employer, which stays uncapped, rather than through sacrifice. A few, put off by the payroll complexity of tracking every member against an individual threshold, may drop salary sacrifice pension arrangements altogether.

The Adviser’s Job Before 2029

None of that should wait for 2028. Model each employer client’s salary sacrifice population against the £2,000 threshold now, using real contribution data rather than an assumed average. The distribution across a scheme’s membership matters as much as the total: a scheme concentrated among higher earners, as many are in finance and professional services, is exposed in a way a lower-paid workforce isn’t. Then bring the employer a real choice rather than a warning: absorb the additional NIC as a cost of the current design, pass the employee’s increase through payroll as scheme rules already allow, or restructure toward direct employer contributions for the portion above £2,000. For some clients, whether salary sacrifice is worth running at all becomes a fair question rather than a given: where few members sit near the threshold, the administrative cost of maintaining it may start to outweigh a shrinking saving. What advisers shouldn’t do is treat the three-year runway as permission to leave scheme design alone. Better to have the review conversation while the cap is still a modelling exercise than once it is a live payroll problem.

Things to reflect on for CPD

  • How many of your employer clients would have members sacrificing above £2,000 a year today, and do you have real contribution data to check, rather than an assumed average?
  • Where a client’s workforce skews toward higher earners, how would you present the choice between absorbing the extra National Insurance cost, passing it through payroll, and restructuring toward direct employer contributions?
  • At what point, for a given client, does the administrative cost of maintaining salary sacrifice start to outweigh the National Insurance saving it still delivers above £2,000?
  • What’s your timetable for raising the 2029 change with clients, and how does it fit around their next scheduled scheme review?