Why southern processing matters now
Türkiye still sets the tone for the hazelnut market. That concentration creates a real single-origin risk for EU industrial buyers, especially when you layer in the usual volatility drivers: weather shocks, TRY/EUR moves, labor availability, logistics disruptions, and quality variability between crop years.
Italy’s 2024 import data makes the point that Chile is already part of the industrial supply mix, not a trial origin. Italy imported about 54,163 tonnes of shelled hazelnuts in 2024, with roughly 36,678 tonnes from Türkiye and about 10,765 tonnes from Chile (HS 080222). If you are supplying pralines, spreads, or bakery inclusions, that is enough volume to matter in tenders and in day-to-day production planning.
Freshness is the practical reason Southern Hemisphere supply is getting more attention. Chile’s harvest and processing window can help cover EU demand when Northern Hemisphere lots are aging in storage. For many manufacturers, the goal is not to replace Northern Hemisphere origins, but to smooth the fresh-crop pipeline and reduce the need to carry extra safety stock.
Supply security is also getting stricter on the buyer side. Vendor-approval lists are tightening, and a second compliant origin helps when you need continuity after a Black Sea climate event or a quality downgrade in a dominant origin. Southern processing capacity only helps if it can deliver repeatable specs and documentation, but when it does, it supports a dual-hemisphere sourcing strategy that procurement teams can defend internally.
Three buyer questions tend to come up early. Will the lots match my roasting curve, or will I need to re-tune time and temperature? Can Chile deliver kernels year-round, not just a seasonal spot offer? And what is the minimum lot size that still makes sense for container utilization into EU plants?
What 2027 capacity changes
A concrete capacity step is already on the calendar. A new facility in Chile’s Los Lagos region was announced in 2026, integrating washing, drying, shelling, and sorting, with the expectation of reaching full operational capacity by 2027. In an April run, it operated at about 50% capacity and processed around 8.8 million pounds, roughly 4,000 tonnes.
For contracting, “full capacity by 2027” matters less as a headline and more as a change in how volume is offered. More processing headroom usually means more forward-contract volume offers, more stable monthly shipment programs, and less dependence on spot buying when the EU crop is tight. That fits better with annual tender cycles, where buyers want a clear volume and shipment cadence they can lock into production plans.
Shipment timing is where Chile can do real work for EU roasters. A common approach is to lock Q3 to Q4 arrivals from Chile to bridge into the Northern Hemisphere new crop. If that pipeline is reliable, you can reduce safety stock while still protecting service level KPIs, because you are not relying on older inventory to cover late-year demand.
Calendar mechanics also push buyers away from a single annual buy. A split-book strategy is often easier to manage operationally: keep the majority in the Northern Hemisphere, then allocate a meaningful share to the Southern Hemisphere, with option volumes you can trigger if Turkish prices spike or if incoming lots fail quality screens. The point is not to trade origins week to week, but to pre-build flexibility into the book.
Logistics details still decide whether a program works. Kernels typically move in full container loads in 20’ or 40’ containers, and quality depends on moisture and temperature management from packing through arrival. Added processing capacity can reduce lead times and widen the set of available grades, including whole kernels, diced, and meal, which helps plants that want one origin to cover multiple SKUs.
Chile’s role in diversification
Chile is not a niche producer anymore, but it is also not a scale substitute for Türkiye. FAOSTAT-reported production for Chile reached about 73,274 tonnes in 2024, placing it among the top producing countries. In global context, INC data frames Chile at roughly 4 to 5% of world production, while Türkiye remains dominant.
That scale is enough for industrial diversification. It can cover a meaningful slice of an EU buyer’s annual requirement, especially for manufacturers who want a second origin that can be qualified and repeated. It is not enough to assume unlimited spot availability at short notice, which is why capacity additions and forward programs matter.
Spec consistency is where centralized processing changes the conversation. EU buyers typically care about calibre sizing, kernel color, defect tolerances, blanching performance, moisture, and roast development. When shelling and sorting are done in a modern, integrated facility, you usually get tighter grading and more repeatable incoming quality than in fragmented supply chains with variable post-harvest handling.
The positioning versus Italy and Türkiye is fairly clear in practice. Italy often plays a premium role tied to specific styles and customer expectations. Türkiye plays the scale and price anchor role for many industrial recipes. Chile tends to win when counter-season freshness and industrial standardization matter more than terroir, especially for paste, inclusions, and high-throughput roasting lines.
Buyers still need to ask the hard questions early. Which cultivars dominate the supply base, and do they behave predictably in roasting and blanching? How stable are yields and in-shell to kernel ratios across crop years? And what is the track record in EU confectionery use cases such as paste, gianduja-style applications, and inclusions where defect tolerance and flavor consistency show up immediately?
Traceability and verification
A “new origin” is only usable at scale if the audit stack is clear. EU buyers typically start with a supplier approval audit, then a HACCP and food safety system review, then trace-back from kernel lots to orchard or collection points. The last piece is verifying whether the program runs segregation or mass balance, depending on the claim and the customer.
Segregation is the lower-risk model for premium and private label customers. It relies on dedicated lines or bins where possible, validated clean-downs, container sealing, and clear labeling that prevents mixing. Mass balance can be acceptable for some sustainability claims, but only if reconciliation is tight and stock accounting is audit-ready.
Digital lot identity is what makes traceability usable during a real incident, not just during an audit. A practical B2B kernel program usually needs orchard block ID, harvest date, drying batch, shelling line, sorter settings, a COA link, and the container number. If those fields are consistent, you can run faster recalls and also do continuous improvement on defects and roast outcomes.
Verification should not stop at paperwork. Pre-shipment COAs and retain samples are standard, but arrival inspection in the EU is where you confirm defects, moisture, and early rancidity indicators. Periodic unannounced audits during peak processing months help validate that the system holds under throughput pressure, which is when origin mixing and re-bagging risks tend to rise.
Quality and safety checkpoints
Moisture control is the foundation of shelf life and roasting performance. Hazelnuts live or die on post-harvest drying and storage conditions, so buyers should set acceptance specs for moisture and water activity targets and require documented drying curves and calibrated meters at origin. If you cannot verify drying discipline, you will pay for it later in oxidation, mold risk, and inconsistent roasting.
Storage and oxidation management should be treated as part of the spec, not a nice-to-have. Temperature-managed warehousing, FIFO discipline, and oxidation monitoring such as peroxide value or free fatty acids, where used, help protect flavor and reduce surprises on the line. Packaging choices matter too, including liners and moisture barriers that keep kernels stable through ocean transit and EU warehousing.
Aflatoxin compliance needs to be explicit in contracts and in testing plans. Under Regulation (EU) 2023/915, maximum levels for hazelnuts placed on the market for the final consumer or used as an ingredient are 5.0 µg/kg for aflatoxin B1 and 10.0 µg/kg for total aflatoxins. Buyers should align internal limits, supplier limits, and release rules so there is no ambiguity at intake.
Sampling and testing protocols are where many programs fail quietly. Specify the sampling plan and aggregate sample size, require accredited lab methods, and define actions on non-conformance, including hold, rework or recondition options, sorting evidence, and re-test rules. If you do not define the decision tree, you will end up negotiating it under time pressure.
Roasting and blanching KPIs should be part of qualification, not just sensory notes. Skin removal percentage, breakage after blanching, color in Lab*, and lot-to-lot flavor consistency are the metrics that protect throughput and finished product appearance. Improved sorting and more uniform drying in Chile can support these outcomes, but only if you measure them and feed results back into supplier grading.
Buyer playbook and KPIs
Start with a controlled pilot that is big enough to be meaningful on your line. One to three containers, split across two lot codes, lets you test within-lot and between-lot variability. Run it side-by-side against current Türkiye or Italy lots in the same production lines, including paste, diced inclusion, and whole roasted, so the comparison reflects real operating conditions.
Protect the downside with contract clauses that match how factories actually work. A spec annex should define calibre, defects, mold and foreign matter tolerances, and moisture at loading. Add an aflatoxin compliance clause aligned to EU limits, COA requirements, right-to-audit language, and clear incoterms with responsibility for condition at loading spelled out.
KPIs should be simple, repeatable, and tied to cost. Track OTIF, percentage of lots within calibre, moisture conformity, defect rate, blanch yield and skin removal, roast color variance, complaint rate, and net kernel yield including breakage and dust. Those metrics translate directly into rework, downtime, and finished product consistency.
Commercial benchmarking should be total-cost, not just €/kg. Compare landed cost against Türkiye and Italy, then add quality yield losses, extra sorting costs, and line downtime. The diversification value is also real, even if it is hard to price precisely, because a second qualified origin reduces exposure when a dominant origin has a price shock or a quality issue.
Scaling to 2027 should be trigger-based. Expand volume after three consecutive compliant lots, stable lead times, an audit pass, and KPI thresholds met on roasting and blanching. If those triggers are hit, pre-booking becomes rational as the Los Lagos facility moves toward full capacity by 2027, because capacity access is often decided before the market feels tight.