The word “biostimulant” turns up more and more often on the shelves of agricultural and horticultural suppliers, and not only in viticulture. The descriptions on those shelves tend to run along the lines of invigorating the plant, or activating it. When the aim is to supply nutrients, there are fertilisers. When the aim is to prevent disease, there are plant protection products. What this third category, sitting between the two, actually does cannot be read off from that wording.
The reason it cannot be read off is not poor copywriting. It is that the regulatory framework does not require the explanation. The biostimulant category specifies, in a closed list, the kinds of effect a product may claim. It does not ask why the product works. The outline looks blurred because the part that is specified and the part that is exempted are discussed together, without being separated.
Effects Specified, Mechanism Not Asked
In the EU, biostimulants are a statutory product category. The Fertilising Products Regulation (Regulation (EU) 2019/1009, hereafter FPR) places them sixth among its Product Function Categories (PFC). The effects a PFC 6 product may claim are limited to four: improved nutrient use efficiency, improved tolerance to abiotic stress, improved quality traits, and improved availability of confined nutrients in the soil or rhizosphere. Abiotic stress here means loads that do not originate from living organisms, such as drought, heat, cold, salinity, waterlogging, ultraviolet radiation, and nutrient deficiency. Pests and diseases fall outside it.
The four are a limit, not an illustration. If a manufacturer observes a fifth effect and puts it on the label, the product cannot be placed on the market as a biostimulant.
What the framework does not ask is which constituent produces the effect, or through what action. There is no obligation to identify an active substance and none to present a mode of action. Suppose a seaweed extract is sprayed and improved tolerance to abiotic stress is confirmed. The manufacturer may sell that product as a biostimulant without answering which component of the extract acted, or how.
Only Pesticide Status Is Mutually Exclusive
Biostimulants are said to be neither fertilisers nor plant protection products. Half of that is right and half is not.
PFC 6 sits inside the FPR. A biostimulant is therefore placed, as a matter of law, as one kind of EU fertilising product. Japan’s guideline says the same thing in its own way: some biostimulants, it states, fall under fertilisers as defined by the Fertilizer Act or under soil improvement materials designated by the Soil Fertility Promotion Act. Where a product does fall under them, the guideline asks that it comply with those laws and then follow the guideline as well. The relationship with fertilisers and soil improvement materials is not exclusive. Obligations are added on top.
The only exclusive relationship is with plant protection. Control of pests, diseases, and weeds, and the promotion or suppression of physiological functions in crops, fall within the definition of an agricultural chemical under Japan’s Agricultural Chemicals Regulation Act. A product claiming those effects is therefore not a biostimulant. The necessary condition for being a biostimulant is not falling under plant protection.
In Japan, that determination does not end with the label. The guideline states that a product may fall under the Agricultural Chemicals Regulation Act on the basis of its raw materials, constituents, and pattern of use, even where no effect is claimed. It locates the substance of the test in an administrative notice on materials of uncertain status (Notice 19-Shoan-10394 of 22 November 2007), which sets out how borderline products are to be judged. Suppose a product contains an active substance for which a registered pesticide exists, at a concentration at or above that of the registered product, and is used for cultivation or crop management. That product incurs the obligation to register as an agricultural chemical even if it says nothing whatever about disease control. Restraint in labelling does not avoid pesticide status.
Four Headings in the EU, Three in Japan
The method of defining effects is the same in both. Each fixes the claimable effects by enumeration and asks nothing about why they occur. Japan’s Ministry of Agriculture, Forestry and Fisheries states in its own documents that the guideline definition was drawn up “with reference to definitions in Europe and North America.” The correspondence is not coincidental.
The difference shows up in how many items the enumeration holds.
The guideline names three targets of improvement: the availability of nutrients in the soil, the uptake and use efficiency of nutrients by the crop, and tolerance to abiotic stress. The first corresponds to the fourth EU heading, availability of nutrients in the soil and rhizosphere; the Ministry itself provides a table mapping the two definitions. Reading the texts against each other, Japan has three headings to the EU’s four, and the one missing is quality traits.
Quality and yield do appear in the Japanese definition all the same. It describes products that improve the three targets above and adds: “as a result of which the quality or yield of the crop improves.” Quality improvement can therefore be stated, but only as a downstream consequence routed through nutrient uptake and use efficiency or through abiotic stress tolerance. Quality traits cannot be made the object of an independent claim, as they can in the EU. That difference remains. Claiming quality is not, however, prohibited by the guideline.
Yield needs the same treatment. An increase in yield is not itself on the list of claimable effects. The guideline does, though, expressly require the use of quantitative indicators in evaluating effects, among them yield, fresh weight, and nutrient uptake. Yield can serve as an indicator supporting a claimed effect.
No Statutory Category in Japan
Alongside the scope of effects, the second difference is standing. The EU biostimulant sits inside a regulation, which is legislation. Japan’s guideline was issued as a Director-General’s notice from a bureau of the Ministry (Notice 7-Shoan-1353 of 30 May 2025), an administrative instrument that creates no statutory category. As the accompanying explanatory material says of itself, it is there “to encourage voluntary efforts by business operators.” The document carries no regulatory authority of its own. What functions as regulation in Japan is the three existing statutes: the Agricultural Chemicals Regulation Act, the Fertilizer Act, and the Soil Fertility Promotion Act.
There is accordingly no prior procedure in Japanese law for calling a product a biostimulant. The sole condition is that it fall under none of the three statutes, and a company that meets it is free to describe its product as a biostimulant. In the EU, by contrast, placing a PFC 6 product on the market with CE marking requires third-party conformity assessment.
Only the Microbial Side Is Restricted
This does not mean that any content will do so long as an effect appears. There is a restriction on content, but only on part of it. PFC 6 divides into microbial 6(A) and non-microbial 6(B), and only 6(A) carries a restriction by named list. As of 2026, 6(A) is limited to four taxa: Azotobacter, Azospirillum, Rhizobium, and mycorrhizal fungi. Bacillus and Trichoderma are not included.
Non-microbial 6(B) carries no such restriction. It takes in seaweed extracts, protein hydrolysates, humic acids, chitosan, and plant and animal extracts, among others, and this is the side where most of the research sits.
What deserves attention is that the four-taxon restriction is an EU provision with no counterpart in the Japanese guideline. The Japanese explanatory material goes no further than listing humic substances, seaweed extracts, amino acids, minerals, and microorganisms among raw materials. Nothing there prevents a Bacillus-based product from being distributed in Japan as a biostimulant.
Why these four taxa cannot be traced back beyond the fact that the text names four. For the four effect headings, a functional definition was the option adopted. A competing proposal placed at its core those properties that cannot be explained by known nutrients or plant growth regulators (PGR), and it was dropped for want of demonstration. That history is on the record in the literature. For the list on the materials side, no comparable record is to be found.
When One Material Falls Under Both Categories
Because the biostimulant provisions ask nothing about constituents, the same material can fall under either of two categories. Chitosan is an example.
Chitosan counts among the non-microbial biostimulants. It is also used to induce plant defence responses and thereby suppress disease. Research has shown that these two actions run through shared pathways on the plant side. The regulatory framework separates them regardless: tolerance to abiotic stress falls within the biostimulant scope, and effects on disease fall within plant protection.
A manufacturer selling chitosan as a biostimulant therefore cannot state a disease control effect. Claiming control requires market authorisation as a plant protection product in the EU, and pesticide registration in Japan. For one and the same material, which effect is stated determines which procedure must be followed. Because the two differ in how much has to be submitted, there is a reason to lean towards the biostimulant route.
The determination can nonetheless be settled on the constituent side before it reaches the label. The test is whether the product contains an active substance at a concentration at or above that of a registered pesticide and is used for cultivation or crop management. The test is not a threshold of efficacy; it is a comparison of concentration against an already registered product. It applies only where a registered pesticide with that active substance exists.
The EU has a requirement pointing the same way. Using a substance with known pesticidal activity at a low dose as a biostimulant is permitted in itself, but the European Commission asks the manufacturer for documentary substantiation. The manufacturer must show either that the dose is significantly below the dose that is pesticidally effective, or, with experimental data, that there is no pesticidal function.
Which category a material falls into is set by the effect stated and the concentration of the constituent together. For the same material, a change in either the stated effect or the concentration changes the category.
The Growth Comes from Regulation and Market Design
Figures reporting growth in the biostimulant market circulate widely. Estimates from private research firms place the 2025 global market between roughly 3.1 and 4.5 billion US dollars. The numbers diverge from firm to firm. On continued growth, several estimates agree.
The reason for the growth does not lie in the materials. Regulation and market design are pushing biostimulants forward from two separate directions.
One is regulation that reduces the available means. The EU set targets to halve both the use and the risk of chemical pesticides by 2030 and to bring 25 percent of agricultural land under organic farming. Japan set targets of its own in the Strategy for Sustainable Food Systems (MeaDRI) of May 2021: a 50 percent reduction in chemical pesticide use on a risk-weighted basis and a 30 percent reduction in chemical fertiliser by 2050, with organic farming extended to 25 percent of cultivated area. The Sustainable Use Regulation (SUR), which was to have turned the EU targets into obligations, was withdrawn in 2024. The published figures have not taken binding form.
The other is the building of a route to market. Before the FPR, harmonisation in the EU reached little beyond mineral fertilisers, and biostimulants sat under national law and mutual recognition. A manufacturer had to go through a separate procedure at each border. The FPR extended harmonisation to biostimulants and created a route to the whole internal market at once. What industry sought reads less as proof of efficacy than as a single point of entry to the market.
Over the period in which the market widened, there is no sign that demonstration of mechanism advanced. The proposal built around properties that known components cannot explain was recorded in 2017 as having “no clear demonstration,” and a 2022 meta-analysis confirmed the same gap. Neither the regulation nor the market design presupposes that how biostimulants act has been established.
Do Biostimulants “Work”?
Whether biostimulants in fact produce an effect breaks in different directions depending on conditions. Change the type of material, the method of application, the indicator measured, or the crop, and the reported result changes with it. There is no position from which they work, or do not work, as a single class.
Meta-analyses across crops return positive values. What is measured there is yield. The indicator the regulatory framework does not admit as an object of claims is the one at the centre of testing whether the products work. A 2022 analysis pooled 1,087 paired comparisons drawn from 180 studies, a paired comparison being one treatment plot set against one control plot. It reported +17.9 percent for marketable yield, meaning yield in saleable condition. By material category, seaweed extracts came in at roughly +16.5 to 18.0 percent, and the band that includes humic acids and protein hydrolysates at +14.8 to 17.1 percent. Across the analysis as a whole, commercial products returned +14.4 percent and non-commercial ones +21.8 percent.
The reach of these figures is limited. The analysis covers non-microbial products, so effect sizes for mycorrhizal fungi and bacterial formulations are not in it. The indicator is marketable yield alone, with quality traits out of scope. The breakdown of the included studies by crop is not given in the text. After pooling, variation between studies remains large, and no value is reported for the range within which the result of the next single trial would fall. The result for an individual vineyard cannot be predicted from the mean.
No Difference Detected Is Not No Effect
A result in which no significant difference was found is not proof that there is no effect. This is where misreading occurs.
A trial returns an estimate of the size of the effect together with the range within which that value may lie. A range that stays wide and straddles zero means that both a large positive effect and a large negative one could sit inside it. That is an answer of “unknown.” Saying that two treatments are the same requires a different design: deciding in advance how large a difference still counts as practically the same, securing enough replicates to detect that difference, and testing for equivalence. Field trials of biostimulants seldom follow this procedure.
The misreading is made on both sides. Those promoting a material use the absence of a significant difference as grounds for safety or equivalence. Those criticising it use the same absence as grounds for no effect. It is one and the same error.
There is a further problem upstream of this. Even where a difference is detected, it cannot necessarily be attributed to biostimulant action. Seaweed extracts and protein hydrolysates contain nutrients, nitrogen among them. Without a nutrient-equivalent control in the trial, an increase in yield cannot be separated into biostimulation and added nitrogen. For microbial formulations, a killed-inoculum control is needed to remove the contribution of dead cells and medium components, and a carrier-only control to remove effects arising from the formulation. Trials carrying controls of this kind are not common. One trial included a plot in which nitrogen, phosphorus, and potassium were matched to the amounts in the seaweed extract. In that trial, the matched plot itself raised vine indicators significantly against the untreated control. The magnitude that could be reported as a biostimulant effect appeared from adding nutrients alone.
What This Means in the Vineyard
Field trials in grapevine show conspicuous variation in their results. Change either the material or the indicator measured and the result that comes out changes. On application method, no trial applying the material to the soil in grapevine could be found, so there is nothing to compare against.
For foliar application of a seaweed extract, one report found no significant difference in yield, juice sugar, or acidity at harvest. In the same trial, leaf area did differ, so a change did appear on the vine side. The photosynthetic indicators moved in one of the two seasons. Change the variety, the region, or the season, and the same material applied the same way gives a different result.
Yield has moved in other reports. In a trial on a wine grape variety with foliar applications of humic acid and amino acids, yield rose from 11.22 to 13.20 kg per vine. The control in that trial was water only. No plot received nitrogen equivalent to the nutrients contained in the material, so the increase cannot be attributed to biostimulant action.
Differences have appeared in fruit composition in another trial. Skin anthocyanins increased significantly in all three wine grape varieties treated with a seaweed extract. Here too the control was a water spray only, and the nutrient contribution is not separated out.
Against this, the same treatment has pointed in opposite directions across traits. In the trial where yield rose, the plot with the largest yield gain fell into the lowest group for total soluble solids among the treated plots. A separate review has collected cases in which the dose that maximises yield and the dose that maximises pigment do not coincide. A plot in which one trait improves can be a plot in which another declines.
Results from applying biostimulants to grapevine are not stable. Some traits move and some do not, and the direction divides by trait. Even with the same material used the same way, a change in site, season, or the trait being observed moves the result. From the user’s side, there is no way to gauge what a given input will return.
The Uses Hoped For and the Frame the Rules Provide
The uses hoped for from biostimulants in viticulture mesh poorly with the four EU headings. The most pressing use falls outside the definition. Two of the four headings sit where this crop has little to gain.
Downy mildew control in organic viticulture depends on copper. Regulation of copper is tightening not only in the ceiling on the amount used but on the side of the conditions of use. In France in 2025, the marketing authorisations of 34 copper-based fungicides were re-examined individually, and two came through with their grapevine use intact. Those two carry conditions: a dose per application, an interval between sprays, and a prohibition on spraying during flowering. The ceiling itself has not moved. A change is nonetheless under way in which the conditions of use narrow the situations in which copper can be applied.
Tighter conditions make control difficult in a way that differs from a cut in the total. Even with headroom left in the annual allowance, a grower who cannot spray the required number of times at the required moment will not hold the disease in a year of high pressure. The outcome of control is decided by whether infection takes hold, so a period in which spraying is not possible carries the same meaning as an insufficient total.
As the conditions for control narrow, attention turns to methods other than attacking the pathogen with a product. If resistance on the vine side can be raised enough to suppress infection, a grower can offset the constraint on spray numbers. Part of the expectation attached to biostimulants points here.
This use, however, falls within the biostimulant scope in neither the EU nor Japan. Effects on disease sit on the biotic stress side, outside the definition. Speaking of disease control with a material bought as a biostimulant is not possible as a regulatory matter. Before the question of whether it works, the claim cannot be made.
Improved nutrient use efficiency and improved nutrient availability in the soil account for two of the four EU headings. In wine grape growing, though, fertiliser inputs are not large to begin with. Where the input is small, raising the efficiency with which it is used leaves only a narrow margin to move. Half of the four headings thus point to territory where this crop has little to gain.
What fits the definition directly is tolerance to drought and heat. Improvement in fruit composition can be claimed in the EU as a quality trait, but in Japan it can only be written as a result routed through nutrient use efficiency or abiotic stress tolerance. Interest in wine grapes runs to fruit composition, to pigments and aroma precursors, so the overlap with the Japanese definition is small.
Seen from wine grape growing, two of the four EU headings are usable head-on: abiotic stress tolerance and quality traits. In Japan, quality traits do not stand as an independent claim, which leaves abiotic stress tolerance alone. The uses that attract expectation and the frame the rules provide do not overlap. Nor is there any prospect at present of that gap closing soon.
What Reaches the Wine Is Unknown
Most cultivation trials in grapevine stop at analysis of the fruit. Cases in which that fruit was made into wine and the wine analysed are few. One study vinified fruit from plots treated with chitosan and reported significant differences in wine anthocyanins and phenolics and in sensory evaluation. Another evaluated a seaweed extract as far as the wine and its microbial community.
There is a regulatory reason as well for trials stopping at the fruit. The effects a biostimulant may claim are effects on the vine and the fruit. The composition and aroma of the wine do not fall within the defined scope. Going on to vinification adds nothing to what may be stated on the label, so fruit is far enough to substantiate a claim.
What biostimulants do to wine is, at present, largely unknown. There is no material on which to judge how a product used in the vineyard shows up in fermentation behaviour or wine composition, or, if it does, whether the direction is good or bad. With only a handful of reports carried through to wine, unknown is where the matter stands.
Material for a Decision
A biostimulant is a category of input that, on condition of not falling under plant protection, may claim effects of kinds specified in advance. In the EU it operates as a statutory category inside the Fertilising Products Regulation. In Japan it operates as a designation placed outside the three existing statutes. The kinds of effect that may be claimed are fixed, while no explanation is required of why those effects occur.
Brought to viticulture, the position sorts into three points. First, results from application vary, and the direction divides by trait. Second, most of the uses that attract expectation lie outside the scope of claimable effects, or in a position where this crop has little to gain. Third, how a material used in the vineyard shows up in the wine rests on only a handful of reports.
Even so, this is not a situation in which the class can be left out of consideration. The direction of reducing pesticide and chemical fertiliser use has not disappeared from national policy, and constraints on the conditions of use are being added to existing means, copper among them. As the available means diminish, the search for replacements continues.
Whether to use them, and what to verify if they are used. The decision rests on the conditions of the individual site and business. That said, several things can be settled before the decision. Which of the effect types specified by the rules does the effect the product claims correspond to. What did the trials behind that claim measure. Were the trials presented run under conditions close to one’s own site. And what would have to change on one’s own site for the introduction to be judged a success or a failure. The last of these has to be decided before the material is chosen. Since the rules do not guarantee that an effect will be reproduced, the means of verification are for the user to provide.
