Apple has signed a 3- to 5-year long-term supply agreement for NAND flash memory with Japan's Kioxia, shifting its procurement strategy from the traditional “price negotiation + multi-sourcing” approach to “volume commitment + long-term contract,” potentially without a price cap.Article author and source: South Korea's Economic Forum and TrendForce report
When the world’s most powerful consumer electronics buyer begins to abandon price negotiation and instead signs long-term contracts with “no price cap,” the pricing power in the memory chip market is undergoing a fundamental shift.
On September 8, South Korea’s Economic Daily reported that Apple has negotiated a long-term supply agreement for NAND flash memory with Japan’s Kioxia, shifting its procurement strategy from the traditional “price negotiation + multi-sourcing” to “volume commitment + long-term contracts.”
This long-term agreement spans three to five years and may not include a fixed price cap. Apple has traditionally leveraged its massive purchasing volume to gain an advantage in component negotiations; this move is seen as a rare 180-degree shift in its procurement strategy.
Apple is facing intense competition for production capacity from data center customers, who are willing to sign long-term contracts and pay significant premiums. For Apple, securing NAND supply years in advance is becoming as important as minimizing prices.
Let go of the obsession with low pricing—shift your procurement logic.
Apple has traditionally adopted a short-term procurement model of price pressure and multi-source competition for NAND suppliers.
This long-term agreement with Kioxia, spanning three to five years and potentially without a price cap, completely breaks with convention. Multi-year supply agreements are more common in the AI high-bandwidth memory sector; Apple’s adoption of this model indicates a significant shift in the balance of bargaining power within the memory market.
The counterparty, volume, and pricing of the agreement between Apple and Kioxia have not been disclosed, nor has Apple confirmed it. However, market observers believe Kioxia is the most likely supplier. According to reports, Kioxia, along with SK Hynix, is among the few major suppliers that have adopted such agreements.
AI's displacement of production capacity is directly pushing cost pressures toward end pricing.
According to a TrendForce report, the memory cost of the 256GB iPhone 18 Pro may nearly quadruple compared to a year ago by Q3 2026; the memory cost’s share of the bill of materials (BOM) is projected to rise from approximately 10% in earlier 256GB Pro models to around 34%, and could exceed 40% in the first half of 2027.
According to supply chain reports, Apple’s first foldable iPhone is expected to feature a large amount of high-quality memory, with a starting price of approximately $2,099 to $2,299, and higher-capacity configurations potentially exceeding $3,000. This further explains why Apple has been securing NAND supply years in advance, even if it means accepting higher component costs.
Bargaining power shifts; long-term contracts may become the industry pricing benchmark
Apple's shift is not an isolated event.
TrendForce stated that Samsung Electronics, SK Hynix, and Micron may become candidates for long-term contracts covering LPDDR5X and LPDDR6 memory for upcoming iPhones. Samsung is reportedly planning to allocate 60% to 70% of its total capacity to long-term contracts.
NAND supply is more diversified, with SK Hynisi expected to account for approximately 30% of NAND supply for foldable devices and Samsung at 15%, totaling 45% for the two Korean companies; the remaining share is left for other suppliers, creating room for Kioxia and Apple to expand their collaboration.
For investors, there is an expectation gap here: the market still prices NAND based on the traditional "silicon cycle" framework, assuming that price increases will inevitably lead to overcapacity and a subsequent correction. However, AI-driven capacity constraints exhibit long-term and structural characteristics, suggesting that the peak height and duration of this storage cycle may exceed traditional market expectations.
