GUAN — Deck
Malaysia-based cocoa processor that buys beans, grinds them into butter, powder, cake and liquor, and sells to industrial chocolate makers like Mars, Hershey's and Nestlé across 60+ countries.
A commodity grinder the market is pricing off last cycle's peak earnings
What the bulls see: Trailing P/E of 8.8x sits 40% below the 18-year median, shares trade below tangible book, and FY2025 already printed RM 884M of free cash flow and RM 914M of debt repayment — the largest deleveraging year in company history.
What the bears see: FY2024's record RM 429M profit was an inventory re-pricing gain on a cocoa price that ran from $2,500 to $12,000 per tonne — not a structural earnings step-up. Through-cycle EPS is closer to RM 0.08–0.10, making the stock fairly valued, not cheap.
Why it matters now: Q1 FY2026 earnings drop around 1 June — the first quarter without realised hedging losses overhanging results. That single print settles whether the deleveraging is structural or a pause.
FY2025 swung cash flow by RM 2.8B — the mechanism is working
The accounting mechanism: when cocoa spikes, inventory absorbs every ringgit of paper profit plus fresh debt. When cocoa eases, working capital unwinds. FY2024 burned RM 1.68B of operating cash despite RM 429M of GAAP profit; FY2025 generated RM 1.17B as inventory sold down. Five-year cumulative net income is +RM 929M; cumulative free cash flow is −RM 4.1B.
Net debt / EBITDA is the only metric that matters
- Still 4.9x after the best deleveraging year on record. Well above the 2x bar commodity processors are held to. Interest expense of RM 337M consumed 53% of operating income in FY2025.
- Every 100bps of rate movement wipes RM 32M of pretax profit — roughly 11% of net income. The CFO also chairs the Risk Management Committee, a structural conflict in a company whose central risk is hedging and inventory valuation.
- The stress-case trigger: a single quarter of rising inventory days without a rising butter ratio. That combination in FY2023–24 forced short-term debt from RM 796M to RM 3.4B. The 2014 template — the only outright net loss in the 18-year record — followed the same mechanism.
Insiders and sovereigns bought the drawdown
- CEO Brandon Tay acquired 200,000 shares at RM 0.71 on 6 April 2026 — his first disclosed open-market buy, two months after the RM 0.65 low. All three executives bought during FY2024 as the stock ran from RM 1.60 to RM 4.40; none sold.
- KWAP, the Malaysian state pension, grew its position 734% through 2025. Norges Bank added 19M shares during the worst year for the stock. Institutional sponsorship strengthened through the decline.
- Insider ownership sits at ~64% via the Tay family holdco plus direct stakes. The CFO is the second-largest individual shareholder at 6.82% — a rare configuration that aligns him with minorities on the hedging book he runs.
A death cross in a base — bounce, not recovery
The second death cross in 18 months printed on 4 June 2025 — consistent with commodity whiplash, not a clean trend. The bounce off the February low is real but has not cleared the 200-day at RM 0.87. Price action is pricing net-debt anxiety, not fundamentals; a single cash-flow print flips the regime.
Management kept building while the balance sheet screamed
The narrative they dropped: The 2020–2021 pitch framed SCHOKINAG Germany as a consumer chocolate growth platform and Côte d'Ivoire as a 240,000 MT volume engine. Five years in, Côte d'Ivoire runs at ~27% of stated capacity and SCHOKINAG is described as a 'Made in Europe' label, not a growth story.
The narrative they adopted: FY2025 guidance opens with 'a cautious approach, prioritising working capital' — a quiet admission the growth era has ended. Yet in December 2025, the CEO told The Edge he would not slow expansion despite record gearing.
What this means for trust: Eleven tracked promises since FY2020: four delivered, two partial, two walked back, three missed. The FY2024 'record profit, ability to adapt' framing immediately preceded a share-price halving — the most damaging sequence possible for investor credibility.
Below book on a late-cycle processor — the Q1 print settles it
- The setup is dated and binary. Q1 FY2026 earnings around 1 June is the first clean quarter without hedging-loss drag. CFO/NI above 1.0x plus net debt / EBITDA under 4x converts the RM 1.25–1.50 base case from hope to arithmetic.
- The bull case has four of five legs already printed: RM 914M debt down, RM 884M FCF in, multiples at 10-year lows, CEO and sovereign pensions buying the drawdown. Missing leg: operating leverage on normalised cocoa.
- The bear case is arithmetic on through-cycle earnings. RM 0.10 normalised EPS at 11x = RM 1.10, not mispriced. Below-book on a thin-margin commodity converter with 4.9x net debt and no pricing power is priced correctly until the cash-flow quarter proves otherwise.
Watchlist to re-rate: ['Q1 FY2026 earnings around 1 June: CFO/NI ratio and net debt / EBITDA are the only two numbers that matter', 'Cocoa futures round-trip: Ivory Coast cut farmgate prices 57% in Feb 2026 — spread economics should recover as legacy hedges roll off', 'FY2025 annual report in late April: first pay-for-performance test after profit fell 47%; risk-factor rewrite also lands here']