R&D tax credit jewellery repair UAE: The Complete Guide

Before touching the mechanics of the R&D tax credit, a jewellery repair business needs a clear-eyed eligibility assessment — because for most workshops, the answer is no. The UAE R&D tax credit is designed to reward systematic investigative work that resolves genuine scientific or technological uncertainty: developing something new, not doing existing work well. A repair shop's daily trade — ring resizing, chain soldering, prong re-tipping, pearl re-stringing, rhodium plating, stone tightening, ultrasonic cleaning, engraving and routine restoration — is skilled craftsmanship, but it applies established techniques to known problems. That is routine operation, and routine operations are explicitly outside R&D. No amount of skill, artistry or customer satisfaction converts a standard repair into qualifying research. The credit is not a reward for craftsmanship; it is an incentive for innovation. This matters because claiming routine work as R&D exposes the business to denied claims, interest and penalties. An honest triage at the start saves the cost of a failed application: if your workshop's activity is best described as excellent execution of known techniques, the R&D credit is almost certainly not for you, and your tax planning time is better spent on Small Business Relief, correct VAT recovery and the other positions covered in this guide. The remainder of this guide is written for the minority of repairers who may genuinely cross the line — and for everyone else who wants to understand exactly where that line sits.

The UAE research and development tax credit provides a tax credit calculated as a percentage of qualifying R&D expenditure, with three tiers — 15%, 35% and 50% — and an overall cap of AED 5 million per claimant. The tier is not chosen by the claimant: it is assigned at pre-approval, which means the business applies before or as the R&D programme runs, and the administering body determines which tier the project merits based on the nature of the work. Phase 1 of the regime is non-refundable, which is a critical design feature to understand: a non-refundable credit can reduce a corporate tax liability to zero, but it does not generate a cash refund where the credit exceeds the tax due. A workshop with little or no taxable profit therefore gets limited immediate value from the credit, though the mechanics of carry-forward, if available, should be confirmed in current guidance. The credit sits inside the corporate tax framework — it offsets the 9% corporate tax that applies to taxable income above AED 375,000 — rather than operating as a standalone grant. Because the regime is relatively new and its administrative details continue to develop, every rate, cap and procedural point in this section should be confirmed against current Federal Tax Authority and Ministry of Finance guidance before being relied on; this guide is general information as understood on 28 September 2026, not a substitute for advice on a live claim.

The three tiers — 15%, 35% and 50% — are meant to differentiate levels of R&D intensity or strategic value, with the highest tier reserved for the most significant qualifying programmes. The claimant does not self-select: the tier is assigned only at pre-approval by the Emirates R&D Council, which assesses the project description, the technological uncertainty involved and the planned investigative work. This makes the pre-approval application the single most important document in the process — a vague application risks a low tier or rejection, while a well-evidenced one gives the Council what it needs to assign the appropriate rate. The AED 5 million cap applies to the credit amount, not to the expenditure: it limits the maximum relief any claimant can receive. The arithmetic is straightforward but worth working through. Suppose a workshop has AED 2,000,000 of qualifying R&D expenditure and is assigned the 35% tier: 2,000,000 × 35% = 700,000, giving a credit of AED 700,000 — well within the cap. At the 50% tier, AED 12,000,000 of qualifying expenditure would produce 12,000,000 × 50% = 6,000,000, which exceeds the cap, so the credit is limited to AED 5,000,000. At the 15% tier, reaching the cap would require qualifying expenditure of about AED 33,333,333 (5,000,000 ÷ 15% = 33,333,333). Because the credit is non-refundable in Phase 1, the practical value also depends on the claimant's corporate tax liability: a credit of AED 700,000 against a tax bill of AED 146,250 eliminates the bill but leaves AED 553,750 of credit that cannot be refunded in cash under Phase 1 rules. Confirm carry-forward and utilisation mechanics in current guidance before modelling.

The exclusion of routine operations is the provision that decides most jewellery repair claims, so it deserves careful attention. Routine operations means work that applies established knowledge, standard techniques and commercially available tools to problems whose solutions are already known to a competent professional in the field. For a jewellery repair shop, this covers essentially the entire standard service menu. Resizing a ring is a solved engineering problem. Re-tipping prongs, replacing a clasp, re-stringing pearls, tightening stones, rhodium plating, laser engraving a monogram and polishing out scratches are all applications of known techniques — the fact that each piece is unique does not make the work investigative, because the uncertainty is about the condition of the item, not about whether the technique will work. Even difficult repairs — a shattered shank, a heat-sensitive stone, a complex antique mechanism — are generally still routine if the repairer is selecting among known methods rather than developing a new one. The test is technological uncertainty: is there a genuine question, at the outset, about whether the technical objective can be achieved at all, and is systematic investigation required to answer it? If the honest answer is that a skilled bench jeweller would know how to approach the job, the work is routine. Administrative, marketing and purely aesthetic design work are also outside R&D. Mischaracterising routine repairs as R&D is the fastest route to a denied claim, so document the uncertainty — not the difficulty — of any project you intend to claim.

A small number of jewellery repair activities can genuinely cross into R&D, and they share a common feature: systematic investigation to resolve technological uncertainty. Consider a workshop developing a novel low-heat joining technique for repairing heat-sensitive gemstone settings — for example, a method that bonds metal around opal or emerald without the thermal damage that conventional soldering causes. If the technique does not exist in the trade, and the workshop runs controlled trials varying alloy composition, temperature profiles and shielding methods to discover what works, that is investigative work aimed at resolving genuine uncertainty. Similarly, developing a new restoration chemistry — say, a tarnish-removal or patination process for antique silver that preserves original surface detail where existing methods destroy it — could qualify if it involves systematic experimentation rather than trial-and-error on customer pieces. Another plausible edge case is proprietary process technology: a repairer building and refining an in-house laser-welding protocol with documented parameter matrices, or developing software-driven stone-matching algorithms that go materially beyond off-the-shelf CAD tools. Note what these examples have in common: a defined technical objective, genuine uncertainty about feasibility, systematic experimentation, and contemporaneous records. Occasional tinkering between customer jobs does not meet the bar. If your workshop has a genuine programme of this kind, the next sections on pre-approval and documentation apply directly; if not, the honest conclusion is that the credit is not available, and that conclusion itself has value.

Pre-approval is mandatory — the tier is assigned only at pre-approval by the Emirates R&D Council, so there is no route to claiming the higher tiers retrospectively on work already completed without approval. The practical consequence is timing: the application must precede or accompany the R&D programme, not follow it. A workshop considering a claim should start by writing a project definition before any qualifying work begins: the technical objective, the specific technological uncertainty, why existing techniques are inadequate, the planned investigative steps, the personnel involved and their qualifications, and the expected expenditure categories. This document does double duty — it forms the core of the pre-approval application and it establishes the contemporaneous baseline that auditors look for. The application will need to distinguish qualifying R&D expenditure from routine business costs: bench time spent on the investigative programme may qualify, while the same jeweller's time on customer repairs does not, so time allocation records matter from day one. Because the Council assigns the tier, the application should make the strongest honest case for the programme's significance without overstating it — inflating claims at pre-approval poisons the well for the eventual tax filing. Procedural details, forms, deadlines and any fees sit with the administering bodies and should be confirmed in current FTA and Ministry of Finance guidance, as this is exactly the kind of administrative detail that evolves. Do not begin significant expenditure on an R&D programme in reliance on this guide alone; get the pre-approval position confirmed first.

The R&D tax credit operates within the UAE corporate tax system, so its value depends on the claimant's overall tax position. A jewellery repair company that is a taxable person pays 0% corporate tax on the first AED 375,000 of taxable income and 9% on everything above that threshold, under Federal Decree-Law No. 47 of 2022. The credit offsets that liability. Work through the arithmetic: suppose the company's taxable income for the period is AED 2,000,000. Step 1: subtract the 0% band — 2,000,000 − 375,000 = 1,625,000. Step 2: apply 9% — 1,625,000 × 9% = 146,250. The corporate tax before the credit is AED 146,250. If the company holds an approved R&D credit of AED 700,000 from the earlier example, the credit wipes out the AED 146,250 liability entirely — but because Phase 1 is non-refundable, the remaining AED 553,750 (700,000 − 146,250 = 553,750) does not come back as a cash refund. Whether unused credit can be carried forward to later periods must be confirmed in current guidance; do not assume it. Small Business Relief, available through 31 December 2029 for eligible businesses, interacts too: a workshop electing the relief may already face little or no corporate tax, which further reduces the immediate value of a non-refundable credit. The planning point is sequencing — model the corporate tax position first, then value the credit against it, rather than treating the headline tier percentage as cash in hand.

An R&D claim lives or dies on contemporaneous records — documents created as the work happens, not reconstructed when the return is prepared. For a jewellery repair workshop, the evidence file should start with the pre-approval application and the Council's tier determination, then build outward. Project records need to show the technological uncertainty and the systematic investigation: dated lab or bench notebooks recording each trial, the parameters varied, the results observed and the conclusions drawn; photographs of test pieces at each stage; and records of failed approaches, which are actually strong evidence of genuine investigation. Time records must separate R&D hours from routine repair hours for each person involved — a single timesheet category labelled 'workshop' will not survive scrutiny. Expenditure records need the same separation: materials consumed in trials, apportioned bench and equipment time, and any contractor or specialist costs directly tied to the project. Financial records should trace qualifying expenditure from the general ledger to the claim computation, with a clear audit trail. Keep customer job cards and repair logs too, because they demonstrate the boundary: they show what the business treated as routine, which strengthens the claim that the R&D programme was something distinct. Retain everything for the statutory record-keeping period. The discipline that matters most is contemporaneity — a bench notebook written during the trials is evidence; a summary written for the auditor is advocacy.

The same errors recur in R&D claims from craft and trade businesses, and jewellery repairers should treat this list as a pre-flight check. First, claiming routine repairs as R&D — the most common and most fatal mistake, usually discovered when the FTA asks what technological uncertainty the work resolved and the answer is 'none, we just did the repair well.' Second, starting the investigative work before pre-approval and then trying to claim retrospectively; the tier-assignment-at-pre-approval design means late applications face an uphill battle. Third, inflating qualifying expenditure by sweeping general workshop overheads into the claim — rent, utilities and the owner's time need a defensible allocation basis, not a hopeful one. Fourth, weak time records that cannot separate R&D hours from customer-job hours; without that separation, the expenditure computation collapses. Fifth, overvaluing the credit by ignoring non-refundability — modelling a 50% tier as cash rather than as an offset against actual tax liability. Sixth, neglecting the interaction with Small Business Relief: a workshop already sheltered by the relief through 31 December 2029 may find the credit adds little. Seventh, treating the pre-approval tier as the end of the process rather than the beginning — the eventual tax filing still needs the full evidence file. Each of these mistakes is avoidable with honest triage at the start, pre-approval before expenditure, and records kept as the work happens.

Several positions in this guide are subject to administrative detail that evolves, and they should be confirmed with the FTA, the Ministry of Finance or a qualified adviser before being relied on for a live claim. First, the pre-approval mechanics — application forms, timelines, the precise role of the Emirates R&D Council in tier assignment, and any fees — are administrative matters that this guide describes as understood on 28 September 2026; confirm the current process before committing expenditure. Second, the availability and mechanics of carrying forward unused non-refundable credit under Phase 1 should be verified in current guidance rather than assumed. Third, the boundary between routine operations and qualifying R&D is inherently fact-specific; the edge cases described here are illustrative patterns, not rulings, and a binding view on a specific programme requires professional assessment. Fourth, the interaction between the credit, Small Business Relief through 31 December 2029, and any free-zone positions must be modelled on the business's actual facts. Fifth, if the FTA disputes a claim, a reconsideration request must be filed within 40 business days of notification under Article 29 of Federal Decree-Law No. 28 of 2022 — diary that deadline immediately on receiving any adverse decision. Finally, this guide is general information, not tax advice; legislation and FTA guidance evolve, and the worked figures are illustrative. Where a position carries a do-not-rely flag, treat that flag as the most important sentence in the section.