From Singapore: A once-manageable chip shortage has transformed into a structural crisis that will reshape consumer electronics pricing and availability well into 2027. The International Data Corporation has progressively downgraded its 2026 PC market forecast, now predicting global shipments will fall 11.6 per cent, a significant worsening from its earlier 8.9 per cent warning issued just weeks earlier.
The culprit is neither pandemic-driven logistics nor geopolitical friction, but a deliberate reallocation of silicon capacity. Memory demand from hyperscalers has surged so aggressively that DRAM and NAND production has been structurally redirected away from consumer devices and toward high-margin enterprise components like high-bandwidth memory and dense DDR5. The voracious demand for HBM by hyperscalers, such as Microsoft, Google, Meta and Amazon, has forced the three biggest memory manufacturers—Samsung Electronics, SK Hynix, and Micron Technology—to pivot their limited cleanroom space and capital expenditure towards higher margin enterprise-grade components.
For manufacturers of personal computers, the equation is ruthless. PC average selling prices would likely rise, increasing by 4 per cent to 6 per cent in a moderate scenario, and by 6 per cent to 8 per cent in a pessimistic scenario. Yet price increases alone cannot offset shrinking supplies. Dell and Lenovo, some of the biggest makers of enterprise and consumer PCs, have said that they will adjust their prices by as much as 15 per cent.
Smaller manufacturers face worse circumstances. Manufacturers whose business is mainly in the low end of the market are likely to suffer significantly. The business models of vendors such as TCL, Transsion, Realme, Xiaomi, Lenovo, Oppo, Vivo, or Honour are based on thin margins. This increase in cost will hit their margins substantially, and they will have no other option but to pass the cost (or part) to end users.
The timing could not be worse. The memory shortage is colliding with two major industry forces at once: the Windows 10 end-of-life refresh cycle and the much-touted push toward "AI PCs." These devices tend to have more RAM (Microsoft's Copilot+ PCs require a minimum of 16GB). As more small language models and large language models move on device, memory becomes even more important, with many higher-end systems shifting toward 32GB or higher. Just as the industry is seeing a need to add more RAM, it has become prohibitively expensive to do so, even if they can get supply. This will result in higher prices, lower margins, or a potential downmix in the amount of RAM in new systems at the worst possible time for this to occur.
The shortage extends beyond consumer devices. Samsung, SK Hynix, and Micron Technology produce 88 per cent of the world's automotive DRAM. SK Hynix has said its entire 2026 production is already sold out. Even Australia's tech supply chains cannot escape the effect. Global shortage means local retailers and integrators face the same constraint every manufacturer encounters: paying more or receiving less.
When will relief arrive? IDC expects the memory supply challenges to persist throughout 2026 and likely well into 2027. While we do anticipate that the rate of memory price acceleration will slow in the second half of this year, prices will continue to rise and remain elevated. Based on current assumptions, our model does not point to a reversion to 2025 pricing levels within the forecast horizon.
Some manufacturers are already taking action. Framework and some pre-built manufacturers have given customers the option to buy a system without memory modules. Others are evaluating alternative suppliers. ASUS expands its supply chain through partnerships with Chinese companies that include Changxin Memory and Yangtze Memory. The company is exploring procurement opportunities with Chinese manufacturers including Changxin Memory (CXMT) and Yangtze Memory (YMTC). The industry shift is reflected in major competitors such as Dell and HP and Acer who have started quality verification processes while obtaining supplies from Chinese memory makers.
The fundamental issue remains: This is not a typical boom-and-bust cycle; it's a strategic reallocation of silicon capacity that could persist for years, not quarters. Until new fabrication facilities become fully operational only in 2027 to 2028, the constraints will persist. For consumers planning device purchases, the arithmetic is unfavourable; for manufacturers betting on consumer demand recovery, 2026 will test institutional balance sheets and patience.