R&D Stage Classification and Aid Intensity
Which of three stages an activity was classified into set the ceiling on how much public money it could carry. R10 records the ceilings, the top-ups that could be stacked on them, the caps that overrode the stacking, and the costs the percentages were applied to.
Why the stage a project sat in decided the money
R10 reviews a decade of case-handling practice under one R&D state-aid framework, written by two of the authority's own case handlers in a personal capacity. The assessment architecture — the two-layer assessment, the notification triggers, the incentive-effect test that actually decided outcomes — is treated on state aid assessment of large R&D projects. This page takes the part that produced numbers.
Four questions produced them. Which stage is this activity? What proportion of eligible costs may therefore be aided? What may be added on top, and what stops the additions compounding? And which costs was the proportion applied to?
That is the idea worth carrying out of this page. Stage classification was not a taxonomy exercise; it was the variable that set the price, and R10 calls classifying it precisely "a crucial point of the assessment".
The three stages, and what R10 gives you about them
- Fundamental research
- Industrial research
- Pre-competitive development
- Market introduction
The framework distinguished three stages of research, defined in an annex: fundamental research, industrial research and pre-competitive development. Technical feasibility studies sat alongside them as a separately treated category, priced by reference to the stage they prepared for.
What R10 does supply is a worked classification for a sector where the boundary is genuinely hard to draw. It shows the reasoning the authority applied rather than the definition it applied it to.
AN APPLIED STAGE CLASSIFICATION, AS RECORDED IN R10
| Activity | Classified as | Reasoning given |
|---|---|---|
| Pre-clinical testing — testing active principles on a series of animal species, increasingly akin to man | Industrial research | On the available evidence there was a very high rate of unsuccessful tests at this phase, and results were still a very long way from production and marketing. They could therefore not be considered technical certification or prototype validation, and so were not pre-competitive development |
| Phase I, phase II and phase III clinical testing | Pre-competitive development | The products used in these phases could be considered crude prototypes, tailor-made for the tests, and could not legally be commercialised |
A classification recorded in one decade of practice under one framework, drawn substantially from studies by an intergovernmental economic body and an international industry federation. Not a general definition of either stage.
Read the two rows against each other and the operative test appears. The question was not how scientific the work looked, but whether the thing tested was recognisable as a prototype and how far the results sat from a marketable product.
How R10 says stage classification was handled in practice
The base intensities
Intensity here means the aided proportion of eligible costs, expressed gross. The values below are the ceilings the framework permitted, not the amounts routinely granted.
BASE AID INTENSITIES — REGULATORY VALUES OF THE FRAMEWORK OF 1996, IN FORCE AT MOST UNTIL 31 DECEMBER 2006
| Category | Maximum intensity | Provision and notes |
|---|---|---|
| Fundamental research | 100% of eligible costs | Point 5.2, and only where the funding constitutes state aid at all |
| Industrial research | 50% | Point 5.3 |
| Pre-competitive development | 25% | Point 5.5 |
| Technical feasibility study preparatory to industrial research | 75% of the study costs | Point 5.4 |
| Technical feasibility study preparatory to pre-competitive development | 50% of the study costs | Point 5.4 |
| A project spanning industrial research and pre-competitive development | Must not exceed the weighted average of the two permissible intensities | Point 5.9 |
| Aid delivered as a repayable advance | Up to 60% for industrial research; up to 40% for pre-competitive development | Point 5.6 as applied in decision practice |
Regulatory values of one framework in one jurisdiction, for a period ending no later than 31 December 2006. Not current law.
Two rows are worth pausing on. Fundamental research at 100 percent is counterintuitive to anyone who expects public funding to require matched private money — but the further work sits from a saleable product, the less an aided firm can distort a market with it. And feasibility studies were priced above the stage they prepared for, 75 against 50 and 50 against 25.
The four bonuses, and the caps that overrode them
On top of a base intensity the framework allowed top-ups expressed in percentage points rather than as multipliers, each rewarding a stated behaviour or circumstance.
BONUSES — REGULATORY VALUES OF THE FRAMEWORK OF 1996, IN FORCE AT MOST UNTIL 31 DECEMBER 2006
| Bonus | Value | Conditions as R10 records them |
|---|---|---|
| Small and medium enterprise (point 5.10.1) | +10 percentage points | R10 notes it was rarely usable on large R&D projects, given their size |
| Regional (point 5.10.2) | +10 points in the more disadvantaged category of assisted area; +5 points in the other | Location-based, by reference to the two regional limbs of the treaty |
| Research-programme link (point 5.10.3) | +15 points, increasable to 25 | Required the project to accord with the objectives of a specific project or programme under the then-current multi-annual research framework programme, and a direct link between the two. Raised to 25 only where the project also involved effective cross-border cooperation — between firms and public research bodies, or between at least two independent partners in two states — and results were widely disseminated and published |
| Dissemination or cooperation (point 5.10.4) | +10 points | Required wide dissemination and publication, or "effective cooperation". R10 reads that wording as excluding any contract research from the bonus |
Regulatory values of one framework in one jurisdiction, for a period ending no later than 31 December 2006. Percentage points added to a base intensity, not multipliers.
The stacking was then stopped by two absolute ceilings at point 5.10.6, which R10 says had to be respected in all cases: a combination of bonuses could not exceed 75 percent gross intensity for industrial research or 50 percent gross intensity for pre-competitive development.
The same two numbers appeared twice more. A matching clause at point 5.13 allowed 75 and 50 percent where competitors outside the jurisdiction had received, or were going to receive, aid of equivalent intensity in the last three years. And where a case ran under the important-project limb of the treaty rather than the general compatibility limb, point 5.11 kept the same ceilings.
What counted as an eligible cost
An intensity is only meaningful against a defined base. R10 sets out the eligible-cost treatment recorded in a second annex, reinforced by a 2004 amending regulation which — although it did not apply to large individual R&D grants — was the newer legal text and was used as interpretive guidance.
The eligible-cost rules R10 records
Project-related costs only
Unlike other aid types, notably regional aid, eligible costs were merely the project-related costs of an R&D activity. The cost had to belong to the research, not to the firm.
Instruments, equipment and buildings, pro rata
Eligible solely to the extent and for the duration used for the research project. Where an asset was not used for its full life on the project, only the depreciation corresponding to the life of the project was eligible, on the basis of good accounting practice.
Transfer or capital cost
Costs of commercial transfer, or capital costs actually incurred, were eligible.
The asymmetry is worth looking for in any funding rule you apply: a precise ceiling on an undefined base puts the real discretion in cost admission, not in the headline percentage.
Turning a non-grant instrument into a percentage
An intensity ceiling only works if every form of support can be expressed as a proportion of eligible costs. R10 records the conversion machinery for the three instruments that mattered.
CONVERTING AN INSTRUMENT TO GROSS GRANT EQUIVALENT, AS RECORDED IN R10
| Instrument | Aid element | Practice notes |
|---|---|---|
| Soft loan | The difference between the rate the firm should pay and the rate actually paid | Market rates deemed to be the published reference rates, provided the loan was backed by normal security and did not involve abnormal risk. Where it did — "which is not unlikely in research projects" — the premium may amount to 400 basis points or more if no private bank would have lent |
| Repayable advance | Treated as effectively equating to a grant payment for threshold purposes, and evaluated as such | Defined as a loan repayable only in the event of a successful outcome of the research. The stated rationale is the extreme risk of such projects and the likelihood the beneficiary could not obtain private-market finance. Repayment amounts and procedures had to be disclosed at notification |
| Guarantee | Calculated under the authority's separate notice on aid in the form of guarantees | Normal rules applied, under a notice dating from 2000 |
Regulatory practice of one framework in one jurisdiction, for a period ending no later than 31 December 2006. The 400 basis point figure is a stated premium, not a measured market spread.
R10 records two tests applied to a proposed repayable advance: whether the conditions described a reasonable scenario, and whether reimbursement took place in a reasonable timeframe and at the reference rate. The design principle recorded alongside them is the interesting one — on complete success the advance was repaid at market rate plus a premium, and the premium increased if success was greater than expected, in which case "the State acts like a market lender".
What travels out of this, and what does not
You are unlikely to be applying this framework. You may well be applying an internal capital rule or a customer-funded development agreement that works the same way — a classification that sets a ceiling, adjustments that reward behaviours, and a cost base the ceiling applies to.
The structural ideas worth keeping
- A stage classification is a pricing decision. Decide what it changes before you argue about which box the work sits in.
- Price proximity to market: the nearer the work to a saleable product, the less external subsidy it should carry.
- Make the boundary case the worked example — the reasoning is more useful than either definition.
- Cap the compounding. An absolute ceiling is the only thing that stops individually reasonable adjustments producing an unreasonable total.
- Define the cost base as tightly as the ceiling, and set a conversion rule for every instrument you fund with.
For where these values sit among the other ten papers in this set, see the R&D project management quick reference, and for what a regulatory analysis can evidence beside empirical and modelling work, eleven R&D management papers compared. The private-sector counterpart of the cost-base question is on the financial frame for R&D management.
What to carry forward
- Three stages, one gradient: the closer to market, the lower the ceiling.
- 100 / 50 / 25 percent of eligible costs across the three stages; 75 / 50 percent for the corresponding feasibility studies. Regulatory values of a framework in force at most until 31 December 2006.
- Bonuses of +10, +10 or +5, +15 rising to 25, and +10 percentage points — all overridden by caps of 75 and 50 percent gross.
- R10 names the three stages but does not reproduce their definitions. Take the applied classification reasoning, not a definition.
- The framework never said how costs should be split between categories, so the real discretion sat in cost admission.
- None of these numbers is current law, and none should appear in a live funding case without the period attached.
Frequently asked questions
Can I use these aid intensities to size a funding application?
No. They are regulatory values of one framework, in one jurisdiction, in force at most until 31 December 2006, and a replacement was already announced when R10 was written. They are recorded here as historical practice and as a worked example of how a classification-plus-ceiling scheme is built. Any live application needs the instrument currently in force in the relevant jurisdiction.
What actually distinguished industrial research from pre-competitive development?
R10 does not reproduce the definitions, and concedes the distinction may in practice be difficult to draw. What it does record is the reasoning applied in one hard sector: activity was treated as pre-competitive development where the item being tested could be considered a crude prototype tailor-made for the test, and as industrial research where the failure rate was very high and the results were still a very long way from production and marketing.
Why could fundamental research be aided at 100 percent when development was capped at 25?
Because the framework priced proximity to market. The closer R&D sits to market introduction, the more easily aid to it can distort competition and trade, so the permitted intensity falls as the work approaches a saleable product. Fundamental research attracted the highest ceiling precisely because it was furthest from one — and only where it constituted state aid at all.
Could the bonuses be stacked to get past the base intensity?
Yes, in percentage points, but only up to two absolute ceilings — 75 percent gross for industrial research and 50 percent gross for pre-competitive development — which had to be respected in all cases. On a large project those caps were usually reached before the bonus list was exhausted, which made the cap rather than the bonus the operative constraint.
What counted as an eligible cost?
Project-related costs of the R&D activity only. Instruments, equipment and buildings counted solely to the extent and for the duration used on the project, with only the depreciation corresponding to the project's life eligible where the asset was not used for its full life. Land counted at commercial transfer cost or capital cost actually incurred. The framework did not stipulate how costs should be distributed between categories.
How was a loan or a guarantee turned into a percentage?
A soft loan's aid element was the difference between the rate the firm should pay and the rate actually paid, with reference rates deemed to be market rates provided the loan carried normal security and no abnormal risk — and a premium of 400 basis points or more where no private bank would have lent. A repayable advance was treated as equating to a grant for threshold purposes. Guarantees were valued under a separate notice of 2000.
References and source attribution
- R10 — large R&D projects: a decade of regulatory practice under an R&D state-aid framework. Legal and regulatory practice review in a specialist state-aid law quarterly, 2006, covering practice from 1996 to 2006; 8 printed pages; 57 footnotes; no tables and no figures, so the tables on this page are a reconstruction from prose and quoted framework provisions. Written by two case handlers of the authority in a personal capacity, with an explicit statement that the views are not an official position. Sections used here: compliance with the R&D-stage definitions, compliance with the eligible-cost definitions, aid intensities and bonuses, and the calculation of gross grant equivalents.
- The framework of 1996 on state aid for research and development, prolonged several times and in force at most until 31 December 2006, including its points 5.1 to 5.13 on intensities and bonuses and its annexes on stage definitions and eligible costs; together with the 2000 notice on aid in the form of guarantees and the 2004 amending regulation clarifying the treatment of equipment, buildings and land. All are instruments quoted within R10 and were not supplied to this library. Every value taken from them is a historical regulatory value of its period.
- Studies by an intergovernmental economic body and an international industry federation, cited within R10 as the substantial basis for the stage classification applied to pre-clinical and clinical testing. Not supplied to this library and not independently examined.
- Eleven copyrighted journal articles on R&D project management, supplied as a reading set assembled by a student for a literature review and profiled for this library. Front matter, abstracts, framework sections, tables and figures were read; article bodies were not reproduced, and all content here is paraphrase. R10 is the only regulatory analysis in that set, and the set is not a systematic or representative survey of the field.
- Supplied teaching source for this library (research methods and research process materials). Used here for page conventions and voice; it does not treat public funding, stage classification or aid intensity.
Suggested questions for Ask KEVOS
- Explain how a classification-plus-ceiling funding rule works, using the three R&D stages as the worked example.
- Draft a stage-classification argument for a project sitting between applied research and development, modelled on the pre-clinical versus clinical reasoning.
- What would a hard cap on stacked adjustments look like in our internal capital approval rules?
- Show me how to define an eligible-cost base tightly enough that a percentage ceiling is a real control.
- Compare the treatment of a repayable advance against a grant, and explain why the two tests point in opposite directions.
- Which numbers on this page are period-bound regulatory values, and how should I cite them?