How to Forecast China’s Lithography Leap

Why It Matters

Accurate forecasts of China’s lithography capabilities guide semiconductor supply‑chain risk management and inform export‑control policy. Knowing which forecasting method yields higher accuracy helps governments allocate analytical resources efficiently.

Key Takeaways

• Delphi panel beat crowd by 12% on 2026 lithography forecast
• Both methods adaptable to varied budgets and participant sizes
• Forecasting timeframe more critical than cost differences
• Training forecasters improves accuracy and reduces bias

Pulse Analysis

Lithography sits at the heart of semiconductor manufacturing, shaping the performance of everything from smartphones to AI accelerators. As China accelerates its drive to domestically produce deep‑ultraviolet (DUV) and extreme‑ultraviolet (EUV) machines, stakeholders—from chipmakers to national security agencies—face heightened uncertainty about when, or if, the nation will bridge the technology gap. Forecasting tools that translate expert judgment and crowd wisdom into probabilistic estimates become essential for anticipating supply‑chain disruptions and calibrating policy responses.

The RAND study pits two distinct forecasting frameworks against each other. The Delphi method, a structured expert elicitation process, delivered a 12% higher accuracy rate on the near‑term (2026) question, largely because participants weighted the compressed development timeline as a major obstacle. In contrast, the crowdsourced platform aggregated a broader set of perspectives, offering comparable insights but with slightly lower precision. Both methods proved flexible—scalable to different budgets and participant pools—yet the research highlighted that the active forecasting window, rather than cost, drove the most significant performance variance. This suggests that timing and iterative data collection can outweigh traditional resource considerations.

For policymakers and industry leaders, the findings underscore the value of hybrid forecasting strategies. Continuous crowdsourced monitoring can capture evolving sentiment, while periodic Delphi workshops refine expert consensus and correct systemic biases. Investing in forecaster training further sharpens accuracy, ensuring that strategic decisions—such as export‑control adjustments or R&D funding allocations—rest on robust, evidence‑based projections. As the global chip race intensifies, mastering these forecasting tools will be pivotal for navigating the next wave of technological competition.