Green Hydrogen R&D Tax Credit in the UAE: The 2026 Guide

Green hydrogen R&D tax credit UAE is the enacted tiered, non-refundable credit under Cabinet Decision No. 215 of 2025, implemented by Ministerial Decision No. 24 of 2026, applying to tax periods starting on or after 1 January 2026 — and it is the single most valuable incentive available to a hydrogen developer doing real research in the State. The tiers pair rates with average-R&D-staff tests: 15% on the first AED 1,000,000 of qualifying spend (staff of at least 2), 35% on the next AED 1,000,000 (staff of at least 6), 50% on spend above AED 2,000,000 up to the AED 5,000,000 cap (staff of at least 14) — for a maximum credit of AED 2,000,000 per qualifying entity or tax group per period. Pre-approval by the Emirates Research and Development Council is mandatory, and it comes before the spend, not after.

This page is the claim how-to for hydrogen developers: the instruments, the tier ladder, what hydrogen R&D qualifies, the pre-approval gate, the claim mechanics, utilisation and the clawback traps. One deliberate boundary: this page uses only the enacted regime — the 15/35/50% tiers, the AED 5 million cap, pre-approval and non-refundable carry-forward. Proposal-era figures from before enactment do not appear here and should not be relied on anywhere. The companion hub covers the hub's full tax picture for green hydrogen, including corporate tax, VAT and customs on plant equipment.

Two instruments make up the regime, and the dates matter. Cabinet Decision No. 215 of 2025 sets the framework — issued 31 December 2025, applying to tax periods and fiscal years commencing on or after 1 January 2026. Ministerial Decision No. 24 of 2026 implements it — issued 18 March 2026, effective on issuance, applying to the same periods. The credit is computed per qualifying entity or per tax group per tax period.

The framework targets UAE juridical persons subject to corporate tax or top-up tax, expressly including free zone persons, and foreign persons carrying on business through a UAE permanent establishment. Two exclusions are absolute and both bite hydrogen developers. First, an entity that elected Small Business Relief for the period cannot claim the credit — the two incentives are mutually exclusive, so the early-stage developer must choose. Second, an entity subject to neither corporate tax nor top-up tax has nothing for the credit to offset — though the carry-forward mechanics mean the position should be modelled across periods, not judged on a single year.

Confirm the current text of both decisions before structuring around them: this is a new regime, and implementing guidance continues to develop. The instruments as described here are the enacted position, not a summary of proposals.

Ministerial Decision No. 24 of 2026 tiers the credit by expenditure band, and every band pairs a rate with an average-R&D-staff threshold. Both must be met: if either the spend or the staff threshold for a tier is missed, the rate drops to the highest tier where both are satisfied. The ladder:

• Staff below 2: no credit at all — the regime requires a minimum research team.
• Staff of 2 to 5: 15% on all qualifying spend.
• Staff of 6 to 13: 15% on the first AED 1,000,000, plus 35% on qualifying spend above AED 1,000,000.
• Staff of 14 or more: the full tiering — 15% on the first AED 1,000,000, 35% on the next AED 1,000,000, and 50% on spend above AED 2,000,000 up to the AED 5,000,000 cap.

The cap arithmetic: 15% of AED 1,000,000 is AED 150,000; 35% of AED 1,000,000 is AED 350,000; 50% of the AED 3,000,000 between AED 2 million and the AED 5 million cap is AED 1,500,000. Total: 150,000 + 350,000 + 1,500,000 = AED 2,000,000 — the maximum credit. A developer with AED 4,000,000 of qualifying spend and 16 R&D staff computes 150,000 + 350,000 + 50% × 2,000,000 = AED 1,500,000, as the worked example below shows step by step.

The staff test is the binding constraint for most hydrogen developers: the 50% tier needs 14 R&D staff on average, and the uplift economics below assume that team exists. The tier ladder worked for an AI startup shows the same ladder applied to a different research profile.

Qualifying activity is tested under the decisions' definitions: the activity must take place in the State, as part of an R&D project, and it must be novel, creative, uncertain, systematic — evidenced by a plan and a budget — and transferable or reproducible. The assessment follows the OECD Frascati Manual, and the social sciences, humanities and arts are excluded outright. For a hydrogen developer, the classification exercise is project by project.

What can qualify: novel electrolyser stack architectures, new catalyst and membrane research, process innovations in hydrogen production, storage or transport that resolve genuine technical uncertainty. What generally does not: routine plant construction and commissioning, deployment engineering, product customisation for a specific offtaker, and feasibility studies that do not themselves resolve technical uncertainty. Building the plant is not researching it — the distinction the pre-approval Council will test first.

Each project needs at least AED 500,000 of qualifying expenditure per period — sub-scale workstreams should be consolidated into properly scoped projects rather than claimed as fragments. Document the Frascati case per project: the technical uncertainty, the systematic investigation plan, the budget and the transferability of the results.

Qualifying expenditure is a closed list with one valuable uplift. Staff costs — the R&D team's salaries and related costs — are uplifted by 30%: AED 2,000,000 of researcher payroll counts as AED 2,600,000 of qualifying spend. Consumables: test materials, pilot-plant inputs, prototype components consumed in the research. Subcontracting fees: research contracted to universities, labs or specialist firms. And the arm's-length share of a cost contribution agreement for joint R&D.

For a hydrogen developer the cost pool has a characteristic shape: heavy on specialist staff (the uplift does real work), heavy on consumables (catalysts, membranes, test rigs), and often significant subcontracting to research institutions for characterisation and testing that the developer cannot do in-house. Cloud and compute costs for modelling sit in the consumables analysis — evidence the spend against the project's technical plan.

What does not count: capital expenditure on the production plant itself, routine operations and maintenance, marketing, and any spend outside the approved project's scope and period. The expenditure must be in-State — research performed outside the UAE does not qualify, however central to the project.

Pre-approval is not optional and it is not an afterthought. The Cabinet Decision makes approval by the Emirates Research and Development Council a condition of the credit, and proof of that approval is part of the claim file. The Council — not the decisions — sets the submission form, manner and timeline for pre-approval. That is the detail to check on the Council's channel before planning around the credit, because the channel is where the current process lives.

The practical consequence is sequencing: the claim file starts before the spend. Identify the qualifying projects, prepare the technical case against the Frascati criteria, submit for pre-approval, and only then incur the expenditure with the credit budgeted. Expenditure incurred before approval is expenditure at risk — the developer that spends AED 4,000,000 and then discovers the project was never going to be approved has a research programme, not a tax credit.

For hydrogen developers the pre-approval submission should lead with the technical uncertainty: what is novel about the electrolyser design, what the systematic investigation will resolve, and why the outcome is transferable. Pre-approval mechanics for a research organisation show the same gate from a different sector's perspective.

The claim is made as part of the tax return — or the top-up tax return — for the period in which the expenditure was incurred. Claims submitted after the return due date are not considered, except in exceptional circumstances accepted by the Authority — a narrow escape, not a fallback. The return deadline is therefore the claim deadline: miss the nine-month filing window and the credit for the period is gone.

The claim file has three layers. The approval proof: the Council's pre-approval for each project claimed. The computation: qualifying expenditure by project, the 30% staff-cost uplift applied, the band test with the average-staff figure evidenced, and the tier arithmetic shown step by step. The technical documentation: the project plans, budgets, activity records and expenditure evidence supporting the claim — kept for seven years.

Build the file as the spend happens, not at year-end: timesheets tying staff to projects, procurement records tying consumables to the technical plan, subcontractor agreements and deliverables. The corporate tax return checklist covers the filing mechanics the claim rides on.

The credit is non-refundable in Phase 1: it offsets corporate tax first, then top-up tax where applicable, and any excess carries forward. It is not a cash refund — a developer with no tax liability in the period does not receive a payment; the credit waits for future liabilities. This shapes the planning: the credit's value is realised against the developer's tax profile over time, not in the year of spend.

The ordering matters for groups: the credit is computed per qualifying entity or per tax group per period, and the utilisation follows the entity's or group's liability stack. A developer in a tax group should model where the credit lands — the R&D entity's spend generating the group's credit is normal, but the documentation must support it.

Early-stage developers with no current tax liability sometimes discount the credit; that is usually a mistake. The carry-forward preserves the value, the pre-approval locks the project's eligibility, and the tier arithmetic rewards scale — the developer that builds the R&D function now claims the 50% tier later. The credit is a multi-period asset, not a single-year deduction.

Two clawback-adjacent rules need to be understood before the first claim. Becoming a Qualifying Free Zone Person within five years of the last claim triggers a clawback of the credit — the decisions recapture the benefit where the entity moves into the 0% free zone regime shortly after claiming. A free zone hydrogen developer planning a QFZP election must sequence it around the R&D programme: claim first and elect later without advice, and the credit comes back.

The interaction with the free zone position more broadly needs care: the credit is available to free zone persons (they are expressly within the framework's scope), but the QFZP election and the credit interact through the clawback, not through simultaneous benefit. Model the five-year window before electing.

The QFZP position for hydrogen projects on the hub covers the free zone analysis the clawback rule sits inside — read the two together before deciding the entity's status for any period in which a claim is made or contemplated.

An early-stage hydrogen developer faces a genuine fork: Small Business Relief or the R&D credit — the two cannot be combined for the same period. SBR treats the entity as having nil taxable income (revenue of AED 3 million or less, through 31 December 2029); the credit needs taxable income (or top-up tax) to offset, with the excess carried forward. Electing SBR excludes the credit entirely for the period.

The choice turns on the numbers. A pre-revenue developer with AED 1,000,000 of qualifying R&D spend and no tax liability gains little from SBR (there is no income to shelter) and banks a credit that carries forward — the credit usually wins. A developer with modest R&D spend and healthy commercial income may prefer SBR's simplicity. The trap is electing SBR by default at incorporation and discovering two years later that the R&D programme's early spend can never be claimed for those periods.

Decide per period, in writing, before the return: the election is the return's, and the credit foregone for an SBR period is not recoverable later. How the credit interacts with Small Business Relief works through the same fork for a research-services business.

Hydrogen developers often hold the IP the R&D creates — electrolyser designs, process patents, control software — and the question arises whether IP income can sit under the free zone 0% rate while the R&D that created it claims the credit. This is the point on this page that must be confirmed against the current text: the QFZP qualifying-income boundaries, the nexus requirements for IP income and the interaction with the five-year clawback rule are technical, fact-specific and subject to developing guidance.

The safe sequence: establish the R&D credit position first (pre-approval, qualifying spend, claim mechanics), then analyse the IP income position separately against the current Cabinet Decisions on free zone taxation, then check the clawback window before any QFZP election. Do not assume the credit and the 0% IP rate stack neatly — the regimes were designed with the clawback precisely to prevent casual combination.

Reported here as a get-advice point: the facts (where the R&D is performed, where the IP is held, the entity's free zone status per period) decide the answer, and this page does not supply it.

The R&D credit is the headline, but a hydrogen developer's incentive picture has other lines — each needing confirmation against its current Cabinet Decision text. Industrial and manufacturing incentives, customs-duty considerations on plant equipment, and any emirate-level programmes all sit outside this page's scope and change independently of the R&D regime. The hub's incentives section tracks the broader picture.

The discipline that applies to all of them: read the current decision text, not a summary. This regime's own history is the warning — proposal-era figures circulated widely before enactment, and anyone who planned around them had to rebuild when the 15/35/50% tiers, the AED 5 million cap and the staff tests were enacted. Treat every incentive the same way: instrument, date, current text, then plan.

For the R&D credit specifically, the texts are Cabinet Decision No. 215 of 2025 and Ministerial Decision No. 24 of 2026, for periods starting on or after 1 January 2026 — and the Emirates Research and Development Council's current pre-approval process, which is where the regime meets reality.

Take a hydrogen developer with AED 4,000,000 of qualifying R&D expenditure in the period and an average of 16 R&D staff. The cost pool: staff costs of AED 2,000,000 uplifted 30% = AED 2,600,000; consumables (catalysts, membranes, test-rig inputs) AED 800,000; subcontracted university research AED 600,000. Total qualifying: 2,600,000 + 800,000 + 600,000 = AED 4,000,000 — above the AED 500,000 per-project minimum, and each project pre-approved by the Council before the spend.

Staff of 16 clears the ≥14 test, so the full tiering applies: 15% on the first AED 1,000,000 = AED 150,000; 35% on the next AED 1,000,000 = AED 350,000; 50% on the AED 2,000,000 above AED 2,000,000 (4,000,000 − 2,000,000) = AED 1,000,000. Total credit: 150,000 + 350,000 + 1,000,000 = AED 1,500,000 — inside the AED 5,000,000 cap and below the AED 2,000,000 maximum.

Utilisation: the developer's corporate tax liability for the period is AED 900,000. The credit offsets it in full — tax payable AED 0 — and the remaining AED 600,000 offsets top-up tax if any, otherwise carries forward to future periods. Non-refundable, but not lost: the AED 600,000 waits for the next liability. Had the developer elected Small Business Relief for this period, the entire AED 1,500,000 credit would have been foregone — the election and the credit cannot be combined.

Four points on this page carry explicit limits. First, the regime is new: Cabinet Decision No. 215 of 2025 and Ministerial Decision No. 24 of 2026 are the enacted instruments, but implementing guidance continues to develop — confirm the current text of both decisions and the Council's current pre-approval process before structuring around the credit.

Second, the pre-approval process: the submission form, manner and timeline are set by the Emirates Research and Development Council, not by the decisions — this page cannot state a process the Council owns. Third, the QFZP/IP interaction: the qualifying-income boundaries, nexus requirements and the five-year clawback application are fact-specific and reported here as a get-advice point, not a concluded rule.

Fourth, proposal-era figures: any 30–50% framing from before enactment does not appear on this page and should not be relied on anywhere — the enacted tiers are 15%, 35% and 50% with the staff tests and the AED 5,000,000 cap stated above. Before acting on any figure in this guide, check it against the current Ministry of Finance or Council publication linked above.