Associated British Foods plc: Primark, FoodCo and the Value of the Split

July brought a further change, and this latest incremental downgrade really was concentrated in Sugar. ABF said that, aside from Sugar, the full-year outlook was unchanged. Sugar was newly expected to lose £25m–£60m for FY2026, with the range driven by European gas prices, African currency risk and operational uncertainty in Tanzania. Primark’s full-year adjusted operating margin remained guided at about 10%. The cumulative FY2026 earnings deterioration since November is broader than Sugar, however. Primark has moved from “slightly below” an underlying FY2025 margin of roughly 12% toward about 10%; Grocery and Ingredients had already weakened; Sugar has moved from a small expected profit to a material loss. The July cut is a Sugar event. The FY2026 downgrade cycle is a Primark-Europe, US-food and Sugar event.

Sugar itself mixes cyclical and structural risks. Low European sugar prices and the spike in gas costs linked by ABF to the Middle East conflict are commodity-cycle inputs outside management’s direct control. Tanzania’s production ramp is more temporary and execution-related. A potential Malawi kwacha devaluation is different: currency resets, inflation and convertibility are recurring features of operating across African markets, so the precise devaluation is episodic while the exposure is structural. ABF also warned that FY2027 Sugar losses could be worse than the upper end of the FY2026 range if high gas costs and weak pricing persist. Investors should therefore resist both extremes: neither capitalising FY2026’s Sugar loss forever nor dismissing it as a clean one-off.

The food restructuring has become concrete. Vivergo’s Hull bioethanol plant was put into orderly closure in August 2025, with production scheduled to stop by the end of that month after the UK government declined the support and regulatory certainty ABF said the business required. The changed UK-US ethanol trade environment was an important part of the economic pressure. The Hovis acquisition, agreed in 2025, received final Competition and Markets Authority clearance on 2026-06-16 and completed on 2026-07-08. The CMA’s work itself describes the structural difficulties and persistent losses in ABF’s existing bakery operation. ABF expects meaningful production and distribution benefits from combining Hovis with Allied Bakeries, but it has not publicly quantified a synergy target or a precise payback schedule. Published reports put the purchase consideration at about £75m.

Capital allocation has also changed the equity story. ABF returned cash through ordinary dividends, specials and several large buybacks as post-pandemic cash generation recovered. FY2024 free cash flow reached £1.355bn, before falling to £648m in FY2025. Net cash before lease liabilities declined from £1.044bn at FY2024 to £390m at FY2025 and only £3m at the February 2026 half year; total net debt including leases was £3.027bn at that interim point, although leverage remained a manageable 1.2 times. ABF’s specific £250m FY2026 buyback programme was completed on 2026-08-14. The cash-rich phase has faded. The balance sheet remains sound, but future buybacks increasingly need to compete with demerger costs, store expansion and FoodCo restructuring.

The controlling shareholder matters more as separation approaches. Wittington Investments held 58.8% at the FY2025 year-end, and the Garfield Weston Foundation owned 79.2% of Wittington. A January 2026 filing recorded Wittington with 421,243,985 voting rights, then 59.08%, of which 56.56% was held directly and 2.51% indirectly, which is why Wittington’s own published description of about 56% is the direct-only figure rather than a conflicting one. Holding that absolute share number constant against ABF’s lower 702,947,191 shares outstanding at 2026-08-28 implies about 59.93% today. That 59.93% is an inference, not a newly filed percentage. Buybacks naturally raise Wittington’s percentage without requiring it to purchase shares. Since Wittington has already said it intends to retain majority ownership of both post-demerger companies, minority shareholders should not capitalise a takeover premium into the separation thesis.

The price history supports a Primark-first interpretation. ABF fell roughly 8% on the April 2025 Sugar warning, but it fell about 12% on 2025-09-10 when Primark trading disappointed. The 2026-01-08 profit warning drove another roughly 10% fall as Primark Europe and US food weakness undermined the previous growth outlook. When ABF actually announced the Primark demerger on 2026-04-21, shares fell around 3%, because the earnings deterioration outweighed the structural value-unlock narrative. July’s Sugar warning cost another 3.7%. The stock’s £20.72 close on 2026-09-04 remained 12.2% below its £23.59 52-week high from 2025-11-04.

A useful hierarchy falls out of that. Primark execution has the largest effect on intrinsic value and the share-price multiple. Sugar can remove £50m–£100m from annual earnings surprisingly quickly, but should command a low cycle-normalised valuation anyway. Grocery and Ingredients are more valuable than the conglomerate narrative usually implies. The demerger can make those differences visible, but the announcement itself has already happened, so “break-up optionality” cannot be counted twice.

My qualitative portrait is a company in transition. ABF is moving from a family-controlled conglomerate whose diversification obscured very different returns on capital toward two family-controlled listed groups whose economics will be easier to price. The transition coincides with a weaker earnings year and occurs before the market has enough evidence to price Primark US as a proven high-return growth business. The bull/bear dispute comes down to whether FY2026 represents a recoverable trough in Europe and Sugar while US stores establish attractive economics, or the first evidence that Primark’s mature European engine deserves a structurally lower multiple.

Vertical history, financial review, and price narrative

ABF’s history is a story of patient family capital moving from basic food production into two forms of consumer scale: branded and processed food on one side, high-volume value retail on the other. The legal entity was incorporated in Britain in October 1934, while the operating history generally begins with W. Garfield Weston’s UK bakery expansion in 1935. The company later took the Associated British Foods name as the bakery estate widened into a diversified food group.

The historical institutional feature that survived every strategic change is Weston control. That gave ABF unusually patient capital compared with a conventional widely held public company. It could retain businesses through long restructuring cycles, operate with modest financial leverage and let Primark develop without forcing it into the dominant retail fashion of each decade. The same structure now creates a minority-shareholder question: patience can be an advantage in operating decisions while control can preserve a conglomerate or capital structure for longer than an outside owner would tolerate. The announced demerger reduces the first issue without eliminating the second because Wittington intends to control both descendants.

The earliest phase, from the 1930s through roughly 1960, was a bakery consolidation story. Scale mattered in flour, baking, logistics and purchasing, and the group grew from its bakery base into a broader food manufacturer. Then came branded tea through Twinings and, over later decades, sugar, ingredients and agriculture. One vertically concentrated bakery operation became a portfolio of food supply chains. ABF’s own historical materials document the expansion from Allied Bakeries into the broader Associated British Foods structure and later acquisitions in branded food and sugar.

A second and ultimately more consequential phase began in 1969 with the first Penneys store in Dublin, the business that became Primark outside Ireland. Retail initially sat inside a much larger food group rather than defining it. That ownership mattered. Primark’s formula depended on simple stores, high merchandise throughput, a low gross-margin philosophy relative to fashion specialists, tight sourcing and very low prices. It avoided expensive catalogue and, later, transactional e-commerce infrastructure. The model was structurally different from Zara’s fast-response, high-gross-margin system and from H&M’s eventual omnichannel model. Over time Primark’s scale made physical retail itself a cost advantage rather than an obsolete distribution channel.

The third phase was Primark’s UK and continental European scaling, particularly from the mid-2000s through 2019. Store growth increasingly changed the mix of ABF profits. Large stores in high-footfall locations gave the retailer strong productivity without needing last-mile delivery. Its customer proposition became easy to understand: fashionable or functional clothing at prices often below specialist rivals, purchased in person. This period also created the strategic weakness now being tested. A retailer built around store economics has less first-party online transaction data, lower digital convenience and less ability to transfer demand between stores and online during disruption.

The pandemic exposed that weakness brutally. With stores closed, Primark could not substitute online sales in the way Inditex, H&M or NEXT could. FY2021 Retail revenue was only £5.593bn and adjusted operating profit £321m, after a £94m repayment of job-retention support; Retail margin was 5.7%. The store estate nevertheless ended the year at 398 stores and 16.8m square feet. The pandemic proved that Primark’s no-transactional-e-commerce structure carried extreme fixed-cost exposure when physical traffic disappeared.

The recovery proved the other side. Retail revenue rebounded to £7.697bn in FY2022 and adjusted operating profit to £756m, a 9.8% margin. By FY2024 revenue had reached £9.448bn and profit £1.108bn, with an 11.7% margin and reported Retail ROACE of 18.7%. Inflation, freight and input shocks hit margins in FY2023 before recovery, but the store model retained enough customer appeal for Primark to rebuild profitability without becoming a conventional e-commerce retailer.

The fifth phase began during FY2025–FY2026. Earnings weakened again, but for a different reason from the pandemic. Physical stores were open. UK Primark was gaining share. The problem was continental European demand and value perception, while Sugar moved from profit to loss and parts of Grocery softened. At the same time ABF started cleaning up structurally weak assets: Vivergo closed, Hovis was acquired to consolidate UK bakery, Spanish Sugar was restructured, and management finally decided to separate Primark from the food portfolio.

The original listing history is less cleanly documented than the operating history. ABF has been publicly traded in London for decades, but the primary digital archive available for this research did not provide a sufficiently reliable original IPO price, proceeds figure and first-day valuation to state those as facts. I do not fabricate an IPO narrative. The investment-relevant ownership structure is clearer: Weston family influence predates the modern public-market ABF, and Wittington remains the controlling shareholder today.

The five-year segment reconstruction shows how dramatically the profit mix changed. These segment revenue rows are rebuilt from divisional disclosure and do not tie exactly to the group revenue line in FY2023 to FY2025, where roughly £97m to £175m sits outside the five named segments; the retrieved material does not identify that residual, so I leave it visible rather than force a reconciliation.

Source: ABF annual reports and results. Adjusted operating profit excludes amortisation of non-operating intangibles, transaction costs and exceptional items under ABF’s adjusted measure.

Several things stand out. Retail revenue grew 70% between FY2021’s pandemic-disrupted base and FY2025, while adjusted operating profit more than tripled. Grocery revenue grew much less, yet its margin stayed in a narrow 10.7%–12.1% band. Ingredients improved from 10.0% to 12.6%. Sugar went from a roughly 8%–9% margin business in FY2021–FY2022 to essentially break-even in FY2025. Agriculture never earned more than a 2.9% margin in this five-year period.

Primark’s share of group adjusted operating profit rose from 31.8% in FY2021 to 52.7% in FY2022, 48.6% in FY2023, 55.5% in FY2024 and 64.9% in FY2025. The FY2021 comparison is pandemic-distorted, but FY2024–FY2025 still shows the structural point: ABF’s equity value increasingly became a Primark call even before the demerger was announced.

The segment returns expose a second layer. In FY2024, the last cross-section for which the historical material retrieved here provides clean segment ROACE comparisons, Grocery produced 35.8%, Retail 18.7%, Ingredients 16.9%, Sugar 10.9% and Agriculture 8.0%. FY2022 showed Grocery at 29.3%, Ingredients 14.8%, and both Sugar and Agriculture at 10.3%. Grocery has the strongest disclosed capital-return characteristics in the group. The average is pulled down mainly by commodity-processing and agriculture assets.

A strict ten-year segment ROIC series cannot be reconstructed consistently from the public summaries without imposing changing definitions of segment capital employed, leases and acquisition assets. ABF itself reports ROACE, not an externally standardised segment ROIC, and segment free cash flow is not disclosed. I use reported ROACE where comparable instead of manufacturing a ten-year precision the statements do not support. The available record is still enough to reject the claim that all food operations are structurally thin-return businesses.

At group level, the post-pandemic financial recovery was strong but uneven.

ABF KPI series.

FY2024 was unusually cash-rich. Free cash flow of £1.355bn was followed by only £648m in FY2025 even though FY2025 adjusted operating profit remained £1.734bn. FY2022 was worse, with negative £84m free cash flow despite £1.435bn adjusted operating profit. Working capital, inventory, crop financing and high investment requirements make ABF’s cash conversion meaningfully less stable than a pure asset-light branded consumer company.

FY2025 earnings quality was also complicated by exceptional costs. Statutory operating profit was £1.483bn versus £1.734bn adjusted. ABF reported £188m of exceptional charges, including £154m of non-cash impairment and £34m of restructuring-related cash costs. That adjustment is large enough to matter, particularly because restructuring is no longer rare: Vivergo, Allied/Hovis, Sugar footprints and the coming demerger all create legitimate but recurring “transition” costs. I would value ABF on normalised cash-producing power, not on adjusted EBIT treated as fully distributable.

Primark’s estate expansion explains much of the reinvestment. The following density figures are derived from disclosed Retail revenue and year-end selling space, so they are chain-level revenue per square foot rather than true same-store productivity.

Company-reported estate and revenue figures; density is my calculation.

The post-pandemic density recovery was powerful: from roughly £333 per square foot in FY2021 to about £500 in FY2023–FY2024. FY2025 slipped to around £487 despite estate growth. That is consistent with the current problem. Primark can keep adding stores while like-for-like demand weakens, producing total sales growth without improving the productivity of the mature estate. New-store growth is valuable only if future cohorts preserve store economics.

ABF does not disclose a five-year geographic series for Primark operating margin, sales density, capex per store and store payback. It discloses geographic sales trends and estate data unevenly, but not the complete unit-level dataset this analysis would require. The missing information is especially important for the United States. The best public-source estimate from the July 2026 data is a £15m–£18m annualised sales run-rate per end-period US store; there is no defensible company-disclosed US EBIT per store or payback period to pair with it.

That uncertainty affects how the historical share price should be read. ABF has been assigned alternating labels over the past decade: Primark growth vehicle, pandemic casualty, inflation recovery, conglomerate value and now demerger/restructuring. The last two years show how quickly the narrative changes when operating data move.

Price-event data from Reuters and market data sources.

This sequence gives a clean price-attribution hierarchy. Sugar mattered in April 2025 and July 2026. Primark-related disappointments produced larger moves in September 2025 and January 2026. The April 2026 demerger announcement did not create an immediate re-rating because the market received weaker profit information at the same time. Vivergo’s closure and Hovis clearance have strategic significance for FoodCo, but neither generated a market move comparable with the major Primark or group-earnings updates. The stock is more company-specific than a simple UK-consumer beta and more Primark-sensitive than a Sugar proxy.

A 2026-07-06 RBC “Sell” recommendation is sometimes cited as a discrete catalyst. I could not independently verify a fresh July 6 rating initiation or downgrade from the available public record. Publicly indexed sources instead show RBC moving ABF to Underperform on 2026-04-13 with an £18.50 target, and later commentary around July’s warning identifying additional EPS downside. I treat “RBC negative” as a valid element of sentiment, but not the unverified July 6 action as a separate confirmed share-price event.

At £20.72, ABF remains below its November 2025 high despite an announced demerger. That is economically sensible. The value-unlock mechanism has improved; the earnings denominator has deteriorated.

Business model, moat, industry cycle, and horizontal peers

Primark makes money by removing costs from the value chain that many fashion retailers treat as necessary. It runs large physical stores, buys at enormous scale, keeps price points low and avoids a full transactional e-commerce operation. UK click-and-collect expands assortment access without turning Primark into a last-mile-delivery business. The choice is deliberate: home delivery and returns would add fulfilment expense that is difficult to absorb at very low average selling prices. ABF has instead invested in digital discovery, marketing and click-and-collect while keeping the economic transaction centred on stores.

The fixed-cost side of that choice is substantial. Stores carry rent, lease liabilities, labour, utilities and inventory whether footfall is strong or weak. Fashion sourcing, freight and product costs move more directly with volume, while currency affects landed merchandise costs. That produces considerable operating leverage. The pandemic showed the downside at 5.7% Retail margin in FY2021; reopening and normalising store productivity lifted the margin to 11.9% by FY2025. FY2026’s expected retreat to about 10% shows that a few points of like-for-like weakness, new-store dilution and operating investment can remove a meaningful portion of profit.

Primark’s strongest moat is its cost-and-price architecture. Large-store volumes, supplier scale and a model that avoids subsidising widespread home delivery let it sell garments at prices difficult for conventional omnichannel chains to match while retaining a double-digit operating margin in good years. A second moat is the physical-store destination effect: consumers accept the inconvenience of travelling to stores because the price-value proposition is strong enough. A third is purchasing scale. Those advantages survived pandemic recovery, inflation and increased competition, which makes them more than marketing claims. FY2024 and FY2025 margins above 11% after the FY2021 shock are the operating evidence.

The limitations are equally real. Primark has no network effect and little switching cost. A customer can choose Shein, H&M, Zara, Uniqlo, supermarket apparel or second-hand clothing on the next purchase. Transactional digital data is thinner than at an online-first competitor. Physical-store density becomes an advantage only after a market reaches enough stores and awareness. That makes the US expansion a different business problem from adding one more UK store.

The United Kingdom is now a mature core. Q3 FY2026 UK sales rose just 1%, like-for-like was broadly flat and the overall market was declining, yet Primark gained share. That is a respectable mature-market result, not a growth engine. Continental Europe is the present repair job: Q3 total sales fell about 1% and like-for-like fell 3.6%, with year-to-date like-for-like down 4.8%. Management pointed to weak consumer confidence and a need to improve value perception, then acted on it: from 2026-07-20 Primark cut prices permanently by up to 29% across hundreds of lines in all 19 markets under an “Iconic Value” programme, which puts a further call on the roughly 10% Retail margin guide. The United States is the growth engine: Q3 sales rose 16%, year-to-date 14%, but from only about 6% of total Primark sales.

The US strategy deserves option value, not mature-retailer value. At roughly £0.6bn–£0.7bn annualised sales using the July geographic mix, the operation is already commercially meaningful but remains too small to determine group earnings. A midpoint annualised sales figure of roughly £0.65bn against 41 end-period stores points to about £16m sales per store. That puts the £0.6bn base-case value for the US operation in my SOTP at roughly 0.9 times annualised sales. That deliberately avoids pretending an undisclosed EBIT margin exists. In the optimistic case, £1.2bn corresponds to about 1.8 times the current run-rate and assumes continued mid-teens expansion plus proof of attractive margins. The conservative £0.2bn value treats US as a low-value expansion programme whose economics remain unproven.

The first Manhattan store opened as Primark continued its US rollout in 2026, and the broader estate expansion has kept total US sales growing at double-digit rates. The most valuable new disclosure ABF could provide would be mature-market US sales density, store contribution margin and cash payback. Until then, “41 stores” is an expansion statistic, not a return-on-capital statistic. Two dated markers frame the runway. Primark opened its 500th store, in Naples, on 2026-09-02, five days before this report’s base date, taking the estate across 19 markets and about 20 million square feet. And the ambition of 60 US stores by 2026, which management publicly reaffirmed during 2025, ends the year unmet at 41.

FoodCo runs several distinct economic machines. Grocery contains genuinely branded, high-return categories such as Twinings, Ovaltine, Jordans and Patak’s alongside more commoditised edible oils and the structurally difficult UK bread operation. Its 11%–12% segment operating margin and historically very high ROACE imply meaningful brand and manufacturing economics, despite the weaker pieces. Ingredients is a global B2B operation in yeast, bakery ingredients and speciality ingredients. It has less consumer branding but benefits from technical know-how, customer relationships and scale; margins improved from 8.7% in FY2022 to 12.6% in FY2025.

Sugar is fundamentally different. British Sugar, Illovo and Azucarera transform agricultural commodities using energy-intensive assets, so profitability is exposed to beet/cane economics, regulated or traded sugar prices, gas, harvest conditions, FX and government policy. The move from £199m profit in FY2024 to a £2m loss in FY2025 and expected £25m–£60m loss in FY2026 shows the size of that operating leverage. Agriculture is lower-margin still, with AB Agri earning £25m on £1.616bn of FY2025 revenue.

The group spans at least four cycles. Primark participates in the discretionary consumer and apparel inventory cycle. Grocery is substantially more defensive but still exposed to commodity inputs and promotional intensity. Sugar is an outright agricultural, energy and commodity-price cycle. Agriculture is tied to farm economics. Currency runs through all four, from Primark’s merchandise sourcing to African Sugar translation and transaction exposure. The demerger will make the different betas more visible to investors without eliminating them.

Policy matters most where business models meet borders. The UK-US ethanol trade changes contributed to Vivergo becoming uneconomic without government intervention, and ABF subsequently closed the plant. Sugar is exposed to tariff regimes and import timing, as the July update’s South African discussion showed. Primark and Shein face the same apparel consumer but very different customs economics. Changes to low-value import treatment can reduce the relative advantage enjoyed by direct-from-China online sellers, potentially improving the competitive position of store-based importers, though tariffs can also raise Primark’s sourcing costs. Reuters has documented both Shein’s tariff/de-minimis exposure and Primark’s continued US expansion.

Horizontal comparison puts Primark between five useful reference points rather than beside one perfect peer.

Inditex has become the premium economic model in mass fashion: rapid design-to-store cycles, strong Zara brand equity, global stores plus a large online business, and unusually high gross margins. FY2025 Inditex sales reached about €39.9bn, net income about €6.2bn and gross margin 58.3%, while online sales exceeded €10bn. Customers pay more than at Primark but get rapid fashion refreshes, a broad digital channel and global brand consistency. Its superior margins and digital/physical integration justify a higher multiple than Primark.

H&M is the closest listed value-fashion reference. It has spent years repairing profitability after inventory, markdown and execution problems while operating a much fuller e-commerce model than Primark. More recent results showed operating-margin recovery, including a 10.7% fourth-quarter operating margin in its 2025 reporting period. H&M offers greater digital convenience; Primark’s counter-advantage is lower cost-to-serve and, in markets where its store density is mature, sharper headline pricing.

Fast Retailing’s Uniqlo became something different again: a global functional-basics and fabric-technology retailer, with greater exposure to Asia and less dependence on fast-fashion trend replication. Its capital-market multiple reflects expectations for sustained international expansion and a higher-quality growth profile than ABF currently receives.

NEXT is a particularly useful UK economics comparator because it combined stores with a sophisticated online and third-party platform model. Product and price make it less directly comparable. It is also much more digitally developed, with a long record of disciplined capital returns. The comparison highlights what Primark gives up to preserve low selling prices: digital convenience, marketplace revenue and detailed customer-data economics.

Shein is the disruptive edge. It built an app-first, high-frequency, highly responsive assortment engine with extreme price transparency and little reliance on stores. Its advantage is selection, speed and digital discovery. Its vulnerability is regulation and cross-border parcel economics. Reuters reported by September 2026 that Shein’s shares had fallen more than 20% from its Hong Kong IPO amid changing regulatory and competitive conditions; earlier reporting also showed pressure on growth and profitability as tariff treatment changed. Primark’s store model looks technologically old beside Shein but can become relatively more attractive when small-parcel import subsidies disappear.

A market-multiple snapshot illustrates the different capital-market narratives. These are market-data vendor measures as of the 2026 research period, and fiscal years differ across companies, so the figures are valuation anchors rather than accounting-perfect comparables.

† ABF uses FY2025 adjusted EPS of £1.749 and the £20.72 2026-09-04 closing price; this is my calculation. ‡ I do not use a vendor ABF EV/EBITDA in the SOTP because leases, central costs and the coming split make segment EV/EBIT more transparent.

ABF’s 11.8 times trailing adjusted earnings multiple is low next to the pure fashion names. That discount is partly deserved. ABF has Sugar losses, low-return Agriculture, stranded central costs, a controlled-shareholder structure and a Primark business whose current like-for-like growth is negative. Inditex and Fast Retailing offer cleaner growth. H&M and NEXT offer public-market investors clearer pure-play economics.

The discount also conceals quality. Grocery’s ROACE and Ingredients’ margin profile compare well with many industrial or consumer businesses, while Primark’s mature European estate has historically generated double-digit operating margins and attractive returns. The demerger should help investors stop averaging fundamentally different businesses into one multiple. The remaining discount after separation should reflect operating quality and control rather than conglomerate complexity.

For SOTP purposes I value Primark’s UK-plus-European core before crediting any US growth at roughly £10.0bn–£13.8bn enterprise value, with £11.5bn as the base case. That corresponds broadly to a 10–14 times normalised operating-profit framework for a mature but differentiated physical value retailer, below the premium assigned to Inditex and Fast Retailing and closer to the H&M/NEXT reference range. The US is then valued separately at £0.2bn–£1.2bn. This is important discipline: the core stands on its own; the valuation does not require the US to become another UK.

Management and governance sit between operating quality and valuation. George Weston has led ABF through a long period in which Primark became the dominant profit source while the group retained a conservative financial structure. Under the announced split, George Weston is intended to lead FoodCo, while Eoin Tonge is intended to lead Primark. Tonge’s position has a history worth stating: Primark’s chief executive of 16 years resigned with immediate effect on 2025-03-31 after an external investigation, and Tonge, then ABF’s finance director, stepped in. Michael McLintock is expected to remain chairman through completion. That allocation keeps substantial ABF institutional memory on both sides.

Capital allocation deserves a mixed grade. Management resisted levering the balance sheet to manufacture EPS, continued investing through cycles and has returned large amounts of capital. It also retained low-return or loss-making assets such as Allied Bakeries and Vivergo for a long time. The current restructuring is evidence of action, but it follows years in which stronger assets subsidised weaker ones. The demerger will make that cross-subsidy much harder to hide.

Current fundamentals, guidance drift, and capital allocation

FY2026 has unfolded as a sequence of expectation reductions. The starting point was optimistic. At FY2025 results, ABF expected adjusted operating profit and EPS to grow in FY2026. Primark’s rollout was expected at roughly 4%, its margin only slightly below the underlying FY2025 level, Grocery approximately flat in profit, Ingredients broadly flat, Agriculture broadly unchanged and Sugar profitable, albeit only modestly.

January broke that framework. ABF said FY2026 adjusted operating profit and EPS would be below FY2025, citing weaker-than-expected Primark trading in continental Europe and weaker US demand affecting food operations. The shares fell about 10%. The significance was broader than one soft Christmas period: a business initially expected to grow earnings had become an earnings-decline story only weeks into the new financial year.

The 24 weeks to 2026-02-28 quantified the weakness. Group revenue was £9.470bn, down about 2% at constant currency. Adjusted operating profit fell to £691m from £835m, a 17% decline at actual exchange rates and 18% at constant currency. Adjusted PBT was £663m and adjusted EPS £0.707. Free cash flow was only £71m, gross investment £534m and net cash before leases had nearly disappeared at £3m. Total net debt including leases was £3.027bn and leverage 1.2 times.

Retail sales were still up about 2%, helped by approximately 4% new space growth, but the margin fell to 10.1%. UK like-for-like sales were up 1.3%; continental Europe was down 5.6%. Grocery adjusted operating profit fell about 20%, led principally by US oils, while Ingredients profit fell 7% amid softer US bakery-ingredients demand. Sugar lost £27m. Agriculture contributed only £6m. Primark’s FY2026 full-year margin was guided around 10%, turning the margin reset from an interim fluctuation into a full-year issue.

July showed stabilisation in some areas but another Sugar deterioration. Q3 group revenue was £5.304bn, up 3% at actual currency and flat at constant currency. Year-to-date revenue was £14.774bn, up 1% actual and down 1% constant currency.

ABF Q3 trading update.

Retail’s Q3 total-sales growth was respectable because new stores contributed around five percentage points. Like-for-like sales remained down 2.2%. The UK held broadly flat like-for-like while gaining market share; continental Europe remained weak at -3.6%; US total sales rose 16%. That is a better mix than H1 in Europe, but not yet a clean recovery.

I reconstruct current FY2026 adjusted operating-profit capacity as follows. This is my estimate, not formal segment guidance.

† FY2025 central/unallocated is the difference between summed segment AOP and reported group AOP.

The reconstruction uses FY2025 reported segment profit, the H1 data and the July statement that only Sugar’s outlook changed from the immediately preceding outlook. It is broadly consistent with market commentary around approximately £1.5bn group adjusted operating profit after the warnings.

The guidance bridge matters because investors can otherwise misattribute the cut. From November to today, Retail plausibly accounts for around £150m of the profit deterioration versus an FY2025 baseline if a roughly 10% margin is earned on sales around the current run rate. Grocery and Ingredients together contribute several tens of millions. Sugar contributes at least roughly £25m–£60m against the FY2025 result and more versus November’s expectation of a small profit. The July-only revision, however, came overwhelmingly from Sugar because the weaker Retail/Grocery/Ingredients framework had already been communicated.

The lower end of the £25m–£60m Sugar-loss range assumes relatively benign outcomes: no Malawi currency devaluation and adequate production recovery in Tanzania. The worse end incorporates higher gas prices, a Malawi currency adjustment and a slower operational ramp. ABF also highlighted high gas prices for the 2026/27 beet campaign and the potential for onerous-contract accounting if sugar selling prices remain inadequate against energy costs.

I would classify the Sugar drivers as follows in economic terms. European sugar selling prices are cyclical. Gas is also cyclical/exogenous, while Sugar’s energy intensity is structural. Malawi devaluation is an episodic manifestation of recurring African FX risk. Tanzania’s production ramp is temporary execution risk. South African tariff and import timing is policy/timing risk. Onerous-contract charges would be accounting recognition of a real commodity-cost mismatch, not an independent source of economic loss.

That distinction prevents a common valuation error. A £60m Sugar loss should not be capitalised at a 15-times multiple as though it were permanent branded-consumer impairment. The energy and commodity cycle can reverse. Yet a zero value for the risk is also wrong because operating sugar plants in multiple jurisdictions inherently brings recurring energy, agricultural, FX and policy volatility.

Vivergo is a different case because management has removed the exposure. ABF announced in August 2025 that the Hull plant would close in an orderly process, with bioethanol and animal-feed production ceasing by the end of the month, after government support and regulatory certainty failed to materialise. The direct lesson for valuation is that Vivergo’s losses should not be extrapolated indefinitely. The governance lesson is less flattering: ABF can tolerate weak assets for a long time before finally exiting them.

A clean Vivergo-only closure cost is not separately disclosed in the primary summary material. FY2025 group exceptional items were £188m, with £154m non-cash impairment and £34m restructuring-related cash costs across actions, but assigning those amounts specifically to Vivergo would overstate the available disclosure.

The Hovis transaction attacks a similar problem by consolidating instead of exiting. The CMA gave final clearance on 2026-06-16, and ABF completed the acquisition on 2026-07-08. The regulator’s evidence describes difficult structural conditions and longstanding losses in ABF’s existing bakery operations. ABF expects significant benefits from combining manufacturing and distribution, but it has not disclosed a quantified annual synergy target or exact timing. Published transaction reports put consideration at roughly £75m.

The industrial logic is credible. Bread distribution is route-dense and fixed-cost-heavy; combining two networks and raising plant utilisation can produce large unit-cost benefits even in a mature category. The investment case should still assign no large synergy value until management specifies savings, restructuring spend and timing. A plausible FY2027–FY2028 benefit is an inference, not guidance.

The demerger is the largest restructuring action. ABF’s April announcement aims to create two listed businesses before the end of 2027. Shareholders are expected to receive both securities. Wittington supports the plan and expects to hold a majority of each. George Weston is intended to lead FoodCo and Eoin Tonge Primark. The announcement effectively acknowledged that the capital market can value a pure Primark and a food portfolio more efficiently than a blended conglomerate multiple.

The market’s immediate reaction was caution, not celebration: ABF shares fell around 3% on the announcement day because H1 earnings weakness arrived alongside the separation news. This is useful evidence against treating the full conglomerate discount as free upside. Investors had already anticipated portfolio action after ABF’s strategic review, and a weak earnings base can overwhelm a cleaner corporate structure.

Distribution history also needs to be read against cash flow instead of EPS.

Company KPI and capital-return disclosures.

FY2024 supported unusually generous shareholder distributions because free cash flow was exceptional. FY2025 did not. Buybacks plus dividends materially exceeded the £648m of that year’s free cash flow, drawing down accumulated cash. Net cash before leases fell from £1.044bn at FY2024 to £390m at FY2025 and £3m at the February 2026 interim. That does not make the balance sheet stretched, but it changes the marginal capital-allocation equation. A recurring £500m-plus buyback cannot be assumed while free cash flow is weak and separation costs are coming.

The latest specific £250m programme has already been fully executed; ABF announced completion on 2026-08-14. No amount remains under that particular programme to count as future demand for the shares. AGM authorities may permit future repurchases. That is different from an announced funded programme.

Buybacks interact unusually with control. Wittington’s 421.24m voting rights disclosed in January 2026 represented 59.08% then. If those shares remain unchanged, the lower August 2026 group share count of 702.95m implies roughly 59.93%. This is a textbook mechanical accretion of control through company repurchases.

The Garfield Weston Foundation’s 79.2% ownership of Wittington gives the control structure an unusually long horizon, but it does not justify claiming that ABF’s dividend policy is legally designed to feed the Foundation. The narrower inference is defensible: Wittington has little reason to pursue a short-term sale of control, and the announced intention to retain majorities in both Primark and FoodCo reduces takeover optionality. The same structure can support patient investment while keeping a persistent minority discount.

Valuation, risks, catalysts, and tracking dashboard

At the 2026-09-04 close of £20.72 and 702.95m shares, ABF’s equity value was £14.57bn. Using February 2026 total net debt including leases of £3.027bn as the most recent fully reported debt measure gives an indicative enterprise value of about £17.59bn. Against FY2025 adjusted EPS of £1.749, the stock trades at about 11.8 times trailing adjusted earnings. Against FY2025 free cash flow of £648m, the equity free-cash-flow yield is only about 4.45%.

Those two multiples tell different stories. Eleven to twelve times adjusted earnings looks inexpensive next to most listed fashion peers. A 4.45% trailing FCF yield looks less compelling when the UK 10-year government bond yielded roughly 5.14% on 2026-09-04. FY2025 free cash flow was depressed relative to FY2024, so neither figure should be used alone.

A precise historical valuation percentile would require a consistently constructed decade-long forward-consensus series. I do not have a primary-source series reliable enough to call the stock, for example, “12th percentile” without false precision. The defensible observation is narrower: the current earnings multiple is well below the valuation accorded to pure listed global apparel leaders, while the discount coincides with weaker FY2026 earnings, negative Primark like-for-like sales, a Sugar loss and a pending corporate split. The low multiple therefore contains both conglomerate discount and genuine earnings-quality discount.

The cash-flow passthrough check comes before SOTP. ABF’s disclosed five-year free cash flow was £419m, -£84m, £269m, £1.355bn and £648m from FY2021 through FY2025. That volatility tells more than a single operating-cash-flow/net-income ratio because ABF’s working capital and capital investment swing substantially. The public summary material retrieved for this report does not support a consistently restated five-year operating-cash-flow/net-income ratio under identical definitions, so I do not manufacture one.

Maintenance versus growth capex is also not separately disclosed by ABF. My valuation assumes roughly £0.75bn–£0.85bn of the £1.244bn FY2025 gross investment was maintenance/replacement and £0.40bn–£0.50bn was growth investment, principally store expansion and productive capacity. That is a research assumption, not company guidance. Adding estimated growth capex back to reported free cash flow gives indicative FY2025 “owner earnings” of roughly £1.05bn–£1.15bn, equivalent to a 7.2%–7.9% yield on the current £14.57bn market cap, or about 12.7–13.9 times owner earnings.

The difference between that owner-earnings multiple and the 11.8 times adjusted EPS multiple is under 30%, so owner earnings do not displace accounting earnings as the primary valuation lens here. The exercise says the same thing as the financial history: ABF is investable on normalised cash generation, but reported free cash flow from any one year is too volatile to serve as a terminal value.

SOTP is the natural method because the company has already decided to split.

My Primark core valuation uses enterprise value. I separate the UK-plus-Europe operation from the US expansion because the former has proven mature economics while the latter has not disclosed store-level margins or payback. Grocery, Ingredients, Sugar and Agriculture are also valued on enterprise-value bases. Central costs are capitalised as a negative enterprise value. Net debt including leases is then deducted once at the group level. I give no separate value to any accounting pension surplus because its post-demerger allocation, funding constraints and distributability are less certain than operating cash. A final modest discount reflects family control, separation execution and stranded-cost risk.

The core Primark range corresponds roughly to 10–14 times normalised operating profit. The premium end requires renewed positive European like-for-like growth and continued roughly 10% or better margins. The current listed-peer context ranges from H&M around 8.5 times EV/EBITDA and NEXT around 11.7 times to Inditex around 14.4 times and Fast Retailing above 20 times on the cited market snapshot. Primark should not receive Inditex’s multiple today because it lacks Inditex’s margin quality, digital integration and current growth. It deserves more than a distressed retailer multiple if the mature core can sustain double-digit operating margins.

Grocery’s £4.4bn–£6.1bn range uses roughly 10–13 times normalised operating profit, reflecting its mix of high-return brands and weak bread/oils assets. Ingredients receives a similar but slightly wider quality range. Sugar is valued explicitly on cycle-normalised asset earnings rather than FY2026 losses; Agriculture receives a low multiple because its margins and returns are low. The £0.8bn–£1.15bn negative central value reflects FY2025’s roughly £150m difference between segment and group adjusted operating profit, adjusted for the possibility that some costs disappear or move with the separated entities.

These are research valuations, not company guidance. The demerger creates an important potential double count: using pure-play segment multiples already assumes better valuation visibility. I therefore do not add a separate “break-up uplift” after calculating SOTP.

The corresponding scenario framework is:

† Simplified calculation assumes terminal fair value in three years and a flat £0.63 annual dividend, with no explicit credit for future buybacks.

Permanent-loss risk differs by scenario. In the conservative case, the trigger is evidence that 9%–9.5% Primark margins are themselves too high and US expansion destroys capital. In the base case, permanent loss emerges if continental Europe remains structurally negative while Sugar and bakery consume enough cash to prevent demerger-led capital returns. In the optimistic case, the biggest risk is paying today for a recovery that never produces the assumed double-digit Primark margin and US proof point.

This is valuation-scenario analysis within a research framework, not investment advice.

The expectation gap going into the 2026-09-10 update is unusually narrow and measurable. The market already knows FY2026 earnings are below FY2025, knows Sugar may lose £25m–£60m, knows Primark’s margin is expected around 10%, and knows the demerger is coming. The next surprise has to come from the slope of the operating data, not from the existence of those problems.

A Primark European like-for-like result moving materially toward zero would challenge the bear case quickly. A further deterioration below roughly -5%, especially combined with a margin guide below 10%, would support the idea that the European value proposition needs a deeper reset. The United States matters less to FY2026 earnings but more to the five-year multiple. Continued 15%–20% sales growth without evidence of profitability is encouraging but incomplete; store-level returns would be a much larger information event.

Sugar’s expectation gap is asymmetric. A FY2026 result near -£25m and a FY2027 outlook showing losses narrowing would remove a substantial source of fear because my valuation already applies only a modest value to Sugar. A result below -£60m or a FY2027 loss moving toward or through £100m would challenge the assumption that current losses are predominantly cyclical.

The independent margin-of-safety test is less generous than the base SOTP. Current £20.72 is above the conservative fair value of roughly £18, so there is no discount to conservative intrinsic value. The margin-of-safety sufficiency verdict is therefore: none.

The most fragile base assumption is Primark’s value. The base SOTP assigns £12.1bn of EV to Primark core plus US. Cutting that assumption to 70% removes about £3.63bn of value, or roughly £5.16 per current share before minor discount interactions. The £24 base fair value falls to approximately £18.8. This tells us where the valuation risk actually lives: in the multiple and earnings assigned to Primark, not in whether Sugar is worth £0.4bn more or less.

A flat-earnings stress test is also sobering. At £20.72, the FY2025 ordinary dividend of £0.63 represents about a 3.0% cash yield. Even assuming a hypothetical £250m annual buyback maintained indefinitely adds only about 1.7% of current market capitalisation, bringing simplified shareholder yield toward 4.7%. The UK 10-year gilt yielded roughly 5.14% on 2026-09-04. Under three years of flat earnings with no multiple expansion, there is no margin of safety at this buy price.

The risks capable of creating permanent rather than temporary loss are specific.

Primark Europe is the highest-probability operating risk and has high impact. Q3 continental European like-for-like sales were already -3.6%, after year-to-date -4.8%. Two further updates around -5% or worse would suggest more than weather or weak confidence. The transmission mechanism would run from lower sales density into markdowns and store deleverage, then into a Retail margin below the current roughly 10% guide, followed by a pure-play Primark multiple below my base assumption.

US economics are a medium-probability, high-long-term-impact risk. The observable variable should be revenue per mature store and ultimately cash payback, rather than headline store count. If annualised sales per mature US store settle below roughly £14m while openings keep consuming capital and geographic profitability remains undisclosed, the US deserves little option value. A poor expansion can also lower the pure-play Primark multiple because investors will see the next growth leg as capital consumption rather than compounding.

Sugar energy and FX exposure has high near-term probability and medium-to-high earnings impact. The company itself has framed FY2026 around a £25m–£60m loss and warned FY2027 may be worse. The indicators are European gas prices, sugar pricing, Malawi FX and Tanzania production. The path to equity value is partly direct earnings loss and partly indirect: persistent losses absorb FoodCo cash, raise leverage and make the post-split dividend less attractive.

Demerger execution is a medium-probability, medium-to-high-impact risk. The target is completion before end-2027. A material delay, unattractive debt allocation or duplicated central infrastructure could erase much of the pure-play re-rating. ABF guided in April 2026 to one-off separation and transaction costs in the region of £75m and aggregate dis-synergies below £45m. My SOTP already capitalises £0.8bn–£1.15bn of central/separation burden, and the £150m stranded-cost watch level sits deliberately above that guidance; evidence that stranded annual costs actually approach £150m after separation would warrant a larger deduction.

Control is a low-probability acute risk but a persistent valuation constraint. Wittington is around 60% on the latest inferred share base and plans to retain control of both companies. The likely transmission is a lower pure-play multiple rather than a sudden earnings shock: investors cannot reasonably underwrite a takeover premium or assume activist-led capital changes.

Capital returns create the final risk. H1 free cash flow was only £71m, net cash before leases had fallen to £3m and the £250m buyback has been completed. If cash conversion stays weak while demerger and Hovis integration spending rises, buybacks will have to slow or the balance sheet will move away from ABF’s historically conservative posture.

The positive catalyst list is correspondingly concentrated: a 2026-09-10 update showing European Primark like-for-like moving toward flat; preservation of the roughly 10% FY2026 Retail margin; FY2026 group adjusted operating profit above the reconstructed £1.50bn–£1.57bn range; a FY2027 Sugar outlook materially better than the current warning; quantified Hovis savings; and detailed demerger capital structures that show limited stranded cost. The annual results on 2026-11-03 should answer several of those questions.

Negative catalysts are the mirror image but not symmetrical in valuation. Primark Europe below -5% like-for-like or a Retail margin below 9.5% damages the highest-multiple asset. Sugar below -£60m damages a lower-multiple asset. A demerger delay beyond 2027 damages the expected multiple re-rating. That hierarchy is why a 1 percentage-point Primark margin miss matters more to equity value than a similar absolute earnings surprise inside Agriculture.

Operating data are from ABF’s latest updates; yield and calendar references are from the cited market and company sources.

The dashboard should be read causally. Primark LFL and margin tell whether store productivity supports the Retail valuation. US growth should eventually be paired with sales-per-store and return disclosure; growth without returns is incomplete. Sugar indicators establish whether the current loss is mean-reverting. Leverage and FCF tell whether ABF can fund restructuring while maintaining shareholder distributions. The gilt yield is the opportunity-cost hurdle: a low-growth controlled equity needs a compelling earnings or cash-flow yield when risk-free sterling yields are around 5%.

Cross-synthesis, key data, uncertainties, and sources

Vertically, ABF has proved one capability better than anything else: it can allocate patient capital across long periods without relying on financial leverage to force returns. That patience helped incubate Primark from a small Irish retailer into the group’s dominant profit source. It also allowed Grocery brands and global Ingredients businesses to compound across many cycles. The same patience has a shadow. Weak assets can survive inside the portfolio for years because stronger assets subsidise them. Vivergo’s eventual closure and the Hovis-led consolidation of Allied Bakeries illustrate both sides of the governance model.

Primark’s success rested on an operating model that created a genuine cost position, not on luck or merely a favourable consumer era. A physical-only retailer survived the pandemic shock and rebuilt Retail margins from 5.7% in FY2021 to 11.9% in FY2025. That is unusually strong stress evidence. The model’s advantage is clearest in mature markets where store density, awareness and sourcing volume reinforce one another. The UK Q3 result, broadly flat like-for-like while gaining share in a declining market, is consistent with that moat still working.

Continental Europe shows where the moat is less automatic. Consumers still need to perceive enough value to travel to physical stores rather than buy online or use local competitors. A -3.6% Q3 European like-for-like decline after -5.6% in H1 indicates that scale alone does not protect demand. The next phase of the investment case depends on management adapting assortment, marketing and value communication without importing a costly online model that undermines Primark’s cost structure.

The US is strategically more important than its present earnings. Six percent of Primark sales cannot rescue a weak European core in FY2026. Sustained mid-teens growth could make it material over five years. The key unknown is capital productivity. Based on the company’s rounded geographic sales mix, current US sales run at roughly £15m–£18m per end-period store, but ABF has not disclosed the store contribution margins and paybacks needed to prove that the rollout is creating value. The market should price the operation as an option until that evidence arrives.

Horizontally, Primark owns an unusual niche. Inditex wins through speed, vertical responsiveness and high-margin omnichannel fashion. H&M combines mass fashion with a larger digital channel. Uniqlo wins through functional basics, materials and Asia-led globalisation. NEXT built a UK omnichannel and platform machine. Shein built an algorithmic, parcel-based assortment engine. Primark wins by making the physical store cheap enough, productive enough and distinctive enough that the absence of transactional e-commerce becomes an economic choice, not an obvious deficiency.

That strategy should be rated on the durability of roughly 10% operating margins, deserving neither Inditex’s 20-plus-times earnings quality premium nor a structurally distressed retailer multiple. A Primark earning 10%–11% with stable European LFL and profitable US expansion can justify a low-to-mid-teens operating-profit multiple as a separately listed company. A Primark drifting toward 7%–8% while expansion merely masks negative mature-store productivity deserves a single-digit multiple.

FoodCo is similarly more nuanced than the conglomerate discount suggests. Grocery’s historical ROACE above 30% and Ingredients’ double-digit margin are valuable economics. Sugar and Agriculture depress the portfolio’s quality, and bakery has required structural intervention. The separation can expose high-return food assets, but FoodCo will still be a mixed portfolio rather than a Unilever-like branded-consumer pure play.

The April 2026 demerger decision changed the equity calculus. Before the announcement, an investor might have paid for ABF while treating separation as a free option. That option has now been exercised. The relevant variables are timing, costs, debt allocation, stranded overhead and the multiples the two controlled companies receive. The stock’s fall on announcement day is instructive: corporate architecture did not override earnings.

I think the market is most likely misjudging the composition of quality, not the direction of FY2026 earnings. The earnings downgrade is obvious. Primark Europe is weak and Sugar is losing money. The less obvious point is that Grocery is a higher-return business than the generic “food conglomerate” label implies, while Primark US is a less proven business than the generic “growth engine” label implies. A better SOTP should raise the value assigned to Grocery and discount the certainty assigned to US expansion.

Over the next 12 months the critical variables are Primark’s European like-for-like trajectory, the Retail margin and the depth of FY2027 Sugar losses. The September 10 update and November 3 results are unusually important. A return toward flat continental-European LFL with the 10% Retail margin intact would support the view that FY2026 is an earnings trough. Another guide-down would make the coming pure-play Primark less valuable even though the separation itself remains intact.

Over three years, US store economics and FoodCo restructuring dominate. Primark needs to prove that US stores can earn enough per unit of capital to justify sustained expansion. Hovis must convert manufacturing and distribution consolidation into real cash savings, not just adjusted-profit restructuring charges. Sugar needs to revert toward normalised profitability or shrink its claim on capital. The demerger needs to complete without duplicating a large central-cost base.

Over five years, the key question is whether Primark can become a genuinely transatlantic retailer while preserving its differentiated operating model. If it does, today’s ABF multiple will look like a poor representation of the asset. If the US remains a small, capital-intensive appendage and Europe stagnates, Primark will look increasingly like a mature UK-led cash generator. Both outcomes are plausible enough that a large margin of safety matters.

The biggest bull points are concise:

• Primark generated £1.126bn, or 64.9%, of FY2025 group adjusted operating profit at an 11.9% margin, establishing a sizeable profitable core before US expansion is valued.
• UK Primark gained market share in a declining Q3 FY2026 market even with broadly flat like-for-like sales, evidence that the value proposition remains competitive in its most mature geography.
• Grocery’s FY2024 ROACE of 35.8% and Ingredients’ improving margins show that meaningful FoodCo value sits outside the low-return Sugar/Agriculture businesses.
• The Primark/FoodCo demerger is formally announced for completion before end-2027, making segment transparency and pure-play valuation a visible catalyst rather than speculative activism.

The core bear points are equally concrete:

• Continental European Primark like-for-like sales were still down 3.6% in Q3 and 4.8% year to date, so new stores are currently masking weak productivity in an important mature region.
• Primark’s FY2026 margin is guided near 10% after 11.9% in FY2025, implying roughly £150m of Retail profit pressure under plausible full-year sales assumptions.
• Sugar has moved from a small FY2026 profit expectation to a £25m–£60m loss and management says FY2027 could be worse, exposing a recurring combination of energy, commodity and African FX risk.
• At £20.72 the stock sits above my roughly £18 conservative value, while the roughly 3% ordinary-dividend yield is below the 5.14% UK 10-year gilt yield; a cheap-looking headline P/E therefore does not provide a conservative margin of safety.

The first pre-mortem is a Primark failure. Imagine FY2027–FY2028 continental European like-for-like sales stay around -4% to -5% despite marketing and assortment changes. US stores keep opening, but mature-unit annual revenue settles below £12m–£14m and contribution margins are poor. Primark’s group operating margin falls from the current roughly 10% guide toward 7.5%. Once separately listed, investors stop valuing Primark as a growth retailer and apply about eight times operating profit. Even if FoodCo remains solvent and profitable, combined ABF-descendant value could fall toward £10–£12 per current ABF share, roughly 40%–50% below £20.72.

The second pre-mortem is a bad split into two mediocre earnings stories. European sugar prices remain weak while gas remains elevated, Malawi devalues materially and Tanzania underperforms. Sugar losses exceed £100m. Hovis/Allied integration fails to deliver enough savings, while demerger-related duplication leaves about £150m of annual central costs across the two companies. Combined lease-inclusive net debt rises above £4bn and neither company receives a re-rating because FoodCo looks cyclical and Primark looks mature. In that stress case, a roughly 50% capital loss is plausible. This is a stress script, not my forecast.

The investment becomes materially better under one of two conditions. The first is price: a fall into the mid-teens while Primark’s operating evidence remains intact. The second is evidence: Europe returns toward flat or positive like-for-like sales, Retail sustains at least a 10% margin, US unit economics become visible and attractive, and the demerger reveals low stranded costs. At today’s price, investors are being paid for neither the worst-case outcome nor a fully proven recovery.

The key data reconciliation is straightforward. FY2025 adjusted operating profit was £1.734bn, 13% below FY2024 at actual exchange rates, with Primark providing £1.126bn. My current FY2026 range is £1.50bn–£1.57bn, built from company segment commentary rather than presented as company guidance. Current enterprise value is approximately £17.59bn using the latest reported lease-inclusive net debt. That is roughly 10.1 times FY2025 adjusted operating profit and around 11.2–11.7 times my FY2026 group profit range before depreciation distinctions. The SOTP produces approximately £18, £24 and £31 per share in conservative, base and optimistic cases.

The 12-month return case is dominated by earnings revisions and separation detail, not long-duration growth. Over 3–5 years the US option and pure-play multiples become more important. At £20.72, base-case upside to £24 is only about 16% before dividends. That is useful, but too small to compensate for the downside to conservative value when sterling risk-free yields exceed 5%.

The final research conclusion follows from that asymmetry. ABF owns one excellent mature value-retail franchise, one potentially important but unproven US growth option, a surprisingly high-return Grocery business, a good Ingredients business and two low-return/cyclical segments. The announced split should improve transparency and can narrow the conglomerate discount, but it does not repair negative European Primark like-for-like sales or make Sugar’s energy and FX exposure disappear. The current price already sits inside a reasonable base-value zone.

I would own ABF more readily after a material price reset than chase the demerger ahead of the FY2026 print. A £13.50–£14.40 entry price would put the equity at least 20% below my £18 conservative value while still allowing the investor to benefit from a successful split. Alternatively, fundamentals strong enough to raise the conservative valuation could justify a higher entry price, but that would require evidence rather than the mere passage of time. The September 10 update and November 3 results are the nearest tests.

【Company-profile scores】

• Fundamental quality: medium
• Growth: medium
• Moat: medium
• Financial soundness: strong
• Management credibility: medium
• Valuation attractiveness: medium
• Risk level: medium
• Suitable investor type: value / event-driven

【Investment rating】

• Rating: Hold
• One-line thesis: Primark’s 10% margin and demerger support value, but Europe weakness and Sugar losses leave little conservative margin of safety at £20.72.
• Current-price classification: acceptable hold
• Whether to wait for a better price: yes
• Target holding horizon: 3–5 years
• Expected annualized return: about -1.4% conservative, 7.7% base and 16.7% optimistic over three years under the stated dividend/terminal-value assumptions
• Max-loss risk: approximately 45%–50% in a stress case where Primark margin falls toward 7.5%, US economics disappoint, FoodCo losses persist and pure-play multiples compress
• Reassessment triggers: continental-European Primark like-for-like ≤-5% for two consecutive updates; Retail margin below 9.5%; FY2026 Sugar worse than -£60m or FY2027 loss approaches/exceeds £100m; net leverage above 1.5 times without a temporary separation explanation; demerger slipping beyond end-2027 or annualised stranded central costs approaching £150m against company guidance of dis-synergies below £45m.

【Ideal Buy Price】13.50–14.40 GBP

Basis: at most 80% of the approximately £18 conservative SOTP value. I would require Primark’s margin to remain at least around 9.5%, evidence that European like-for-like deterioration is stabilising, and the demerger to remain on track. Waiting carries the opportunity cost that a strong FY2026/FY2027 recovery or clearer split terms could re-rate the shares before this price is reached.

The acceptable hold range is £20.40–£27.60, corresponding to ±15% around the £24 base scenario. The clearly-overvalued threshold begins at £34.10, 10% above the approximately £31 optimistic scenario.

【Valuation Range】

• current: 20.72 GBP (close as of 2026-09-04)
• bear (conservative · ideal buy zone): [13.50, 14.40] GBP
• base (fair · acceptable hold zone): [20.40, 27.60] GBP
• bull (optimistic · above the clearly-overvalued line): [34.10, 37.20] GBP

Research uncertainties remain material. ABF does not disclose mature US Primark store contribution margin, capex per opening or cash payback, so the US option cannot be valued with genuine four-wall economics. It also does not provide a consistently restated ten-year segment ROIC series, and the available ROACE disclosures are the honest substitute. Maintenance versus growth capex is not separately disclosed, making owner earnings an estimate. ABF has quantified the separation itself: the April 2026 announcement put one-off separation and transaction costs in the region of £75m and expected aggregate dis-synergies below £45m. What remains incomplete ahead of detailed implementation disclosures is the demerger debt allocation and the split of ongoing central costs between the two companies. Finally, the existing in-house notes on H&M, NEXT, JD Sports, Tesco and Unilever were not retrievable in this research environment; I did not inherit or challenge their ratings or valuation ranges, and rebuilt peer comparisons from public-source evidence.

Primary sources carrying most of the analysis are ABF’s FY2025 annual results and annual report, which establish the segment baseline, adjusted-profit definition, exceptional costs, cash flow and outlook. Earlier ABF reports provide the five-year segment and return-on-capital reconstruction.

The February 2026 interim results and July 2026 trading update establish the latest operating trends, FY2026 margin outlook and Sugar range. The April demerger announcement establishes the end-2027 separation target and intended ownership/management framework. ABF’s voting-rights and buyback notices establish the current share count and completed FY2026 repurchase programme.

The CMA materials establish Hovis clearance and provide independent evidence on UK bakery economics. ABF’s Vivergo announcement establishes the closure decision and timing. Reuters reporting is used for dated market reactions and cross-company developments rather than as a substitute for ABF financial disclosure. The £20.72 reference price is the 2026-09-04 completed close, cross-checked against published market data.

Other tickers mentioned

• ITX.MC: Inditex is the global fashion-retail quality benchmark for Primark’s margin, speed, omnichannel model and valuation.
• HM-B.ST: H&M is the closest listed value-fashion comparator and the clearest peer for margin recovery versus Primark’s store-led model.
• 9983.TSE: Fast Retailing provides the Uniqlo benchmark for international expansion, functional apparel and premium growth valuation.
• NXT.LSE: NEXT provides a UK retail benchmark for omnichannel economics, online fulfilment and disciplined capital returns.
• JD.LSE: JD Sports provides UK apparel and consumer-demand read-across for discretionary retail.
• TSCO.LSE: Tesco provides UK consumer and grocery-demand context.
• ULVR.LSE: Unilever is a branded-consumer reference for the quality and valuation of ABF Grocery relative to a purer brand portfolio.

This report is based on public information and does not constitute investment advice. Markets carry risk; invest with caution.