Editorial Article
The Traceability Deadline Moved Thirty Months. The Recall Count Went Up Anyway.
On 20 July 2028, the most prescriptive food traceability obligation in the world is due to begin enforcement — thirty months later than originally written, and after a Congressional directive that FDA not enforce it any earlier. That is the date now attached to the US Food Traceability Rule. It is worth asking what happened to the market while the deadline moved, because the answer is uncomfortable for anyone who built a business case around it.
The recall count went up.
What the rule actually requires
The Food Traceability Rule — formally, Requirements for Additional Traceability Records for Certain Foods — was published on 21 November 2022 and implements section 204(d) of the Food Safety Modernization Act. It applies to anyone who manufactures, processes, packs or holds a food on the Food Traceability List, and to foods containing a listed food as an ingredient where that ingredient stays in the same form. For those foods it requires records containing Key Data Elements at seven Critical Tracking Events, produced to FDA within 24 hours, or a reasonable time the agency agrees.
The seven events are harvest, cooling, initial packing, transformation, shipping, receiving and first land-based receiver. The list was built from 47 commodity categories, drawing on the Reportable Food Registry and on industry codes in FDA’s facility-registration system; its centre of gravity is fresh produce, soft and unpasteurised cheese, shell eggs, nut butters and seafood. The original compliance date for all persons subject to the recordkeeping requirements was 20 January 2026. FDA proposed a 30-month extension to 20 July 2028, on the reasoning that because the rule requires firms to share information with others in their supply chains, the only workable implementation is one where everybody arrives at the same time. The Continuing Appropriations, Agriculture, Legislative Branch, Military Construction and Veterans Affairs, and Extensions Act of 2026 then directed the agency not to enforce before that date, and FDA has said it intends to comply.
The number the deferral did not change
What also did not change is the regulator’s estimate of what all this costs. FDA’s Preliminary Regulatory Impact Analysis covers approximately 422,144 firms operating 566,448 establishments and puts annualised compliance cost at a primary estimate of USD 411 million a year, on a range from roughly USD 34 million to USD 2.4 billion. Per covered establishment, that is USD 725.57 a year. The alternative option in the co-proposal — exempting small retail establishments from the electronic sortable spreadsheet requirement — carries a primary estimate of USD 535 million.
Two things follow, and they are the reason this rule is a weaker demand driver than its reputation suggests. First, a range spanning USD 34 million to USD 2.4 billion is not a rounding interval; it is an admission that the cost depends almost entirely on how much of an existing system a firm can reuse, which is a property of the firm rather than of the regulation. Second, and more important, enforcement against 566,448 establishments cannot be inspection-led. It has to be document-request-led. The operative question for a buyer is therefore not whether a record exists but whether it can be assembled inside 24 hours of a request that arrives without warning — a systems question, but one that attracts a much smaller cheque than a transformation programme.
What the recall data show
Count the recall events from the agencies’ own published datasets and the sequencing is clear. Food recall events fell 9 per cent in 2024 to 516. They then rose to 641 in 2025 — a 24 per cent increase — in the year the compliance date was deferred, and stood at 483 by late November 2026.
| Year | FDA food enforcement events | USDA-FSIS actions | Combined |
|---|---|---|---|
| 2023 | 480 | 89 | 569 |
| 2024 | 461 | 55 | 516 |
| 2025 | 574 | 67 | 641 |
| 2026 (to late November) | 437 | 46 | 483 |
Recall incidence is not a function of the traceability rule. It moved up sharply in the year the rule was pushed back, which tells you that the thing generating demand for recall management is the food system’s own failure rate and not a Federal Register notice. That is good news for the response layer of the market and bad news for the compliance narrative, because it means the growth case has to be built on operational reality rather than on a deadline.
Europe tells a similar story from a different angle. The Alert and Cooperation Network recorded 9,460 notifications in 2024, up 8 per cent, with RASFF accounting for 5,250 — a 12 per cent increase. Around a third of RASFF notifications were border rejections, dominated by pesticide residues in fruit and vegetable consignments; Germany, the Netherlands and Italy were the largest notifying countries. Agri-food fraud notifications, by contrast, fell by one third to 546, and 1,959 RASFF and 1,335 administrative-assistance notifications were flagged as possibly resulting from an intention to deceive. The operational load in Europe is rising on safety and border control, not on fraud.
Two regimes, deliberately opposite
The American rule and the European duty are close to mirror images, and the difference shapes who buys what.
| United States — FSMA 204 | European Union — Regulation (EC) No 178/2002 | |
|---|---|---|
| Scope | Foods on the Food Traceability List, built from 47 commodity categories | Every food business operator, all food |
| What is specified | Key Data Elements at seven Critical Tracking Events | One step back and one step forward (Article 18); withdrawal and recall (Article 19) |
| Deadline to produce records | 24 hours from an FDA request | None; on request by competent authorities |
| Format | Prescribed data elements; no prescribed system | No prescribed format, data set or technology |
| In force | 20 July 2028 (originally 20 January 2026) | Since 2002 |
| Estimated cost of complying | USD 411 million a year — USD 725.57 per covered establishment | Not quantified by the Commission |
The commercial consequence is precise. A firm selling into both markets must hold the more demanding dataset across the wider perimeter: the American rule defines what the record must contain, the European duty defines how far back and forward it must reach. And because neither jurisdiction certifies the system that holds the record — GS1’s own mapping guidance for the American rule states that its standards are voluntary — the legal minimum is cheap and universally satisfiable, while almost all the spend is voluntary.
That is the paradox at the centre of this market. The European law has required traceability for twenty-four years and specifies almost nothing, and Europe has a large, functioning commercial market for traceability. The American law specifies a great deal and has not yet taken effect, and its compliance cost is 4.17 per cent of the global market the rule is credited with creating. In Europe, the retailer’s supplier-approval function does the work a regulator does in the United States — and it does it on response time rather than on legal exposure, which is a service-level purchase dressed as a compliance purchase.
Our interviews with supplier-approval and technical staff returned this consistently: the threshold is not legal compliance, which is assumed, but the speed of answer to a speculative query. That is the buyer who pays for the middle layer of this market, and it is why the Pheonix Research report on the food traceability and recall management market models software, data exchange and managed services growing at 11.42 per cent while the physical layer grows at 5.48.
The counter-thesis, and what to watch
It is entirely possible that the 2028 date is a coincidence rather than a cause, and that the American market will behave like the European one — driven by retailer specification, brand protection and the ordinary replacement cycle of coding and labelling equipment, with the rule as a convenient justification attached afterwards. A reader who believes that should treat the model’s back-loading as too cautious: demand would be smoother and stronger early, because a retailer’s supplier-approval cycle does not wait for a Federal Register notice.
Three things would settle it. A second deferral past 2030 would tell you the compliance-driven portion is smaller than anyone claims and that the market’s real drivers are commercial. A European implementing act prescribing a data format would do the opposite, converting a universal but unspecified duty into required spend across a far wider perimeter than the American list. And an expansion of the Food Traceability List beyond its current 47 categories would be the cheapest possible stimulus to the middle layer, because it would multiply the number of firms that suddenly need a supplier network rather than a filing cabinet.
None of the three has happened. What has happened is a thirty-month deferral, a recall count that rose 24 per cent anyway, and a regulator that has told the market, in published arithmetic, that it thinks compliance is worth about seven hundred and twenty-six dollars a site per year. The market is worth far more than that, which is the subject of the companion article on why traceability prices rise only 1.16 per cent a year.
Sources: FDA, Food Traceability Rule and Food Traceability List; FDA, Preliminary Regulatory Impact Analysis for the Food Traceability Rule; the text of the rule published 21 November 2022 and the compliance-date extension record; Regulation (EC) No 178/2002, Articles 18, 19 and 50; European Commission, 2024 Annual Report of the Alert and Cooperation Network; FDA and USDA-FSIS published recall datasets; GS1 US, EPCIS Recommendations for FSMA 204 Critical Tracking Events, Release 2.0, 15 May 2025. No third-party market research is cited.
