Hybrid life cycle assessment (LCA) does not necessarily yield more accurate results than process-based LCA

Hybrid life cycle assessment (LCA), through combining input-output (IO) models and process-based LCA for a complete system boundary, is widely recognized as a more accurate approach than process-based LCA with an incomplete system boundary. Without a complete process model for verification, however, the performance of hybrid LCA remains unclear. Here, using a counterexample we show that hybrid LCA does not necessarily provide more accurate results than process-based LCA, simply because the aggregation of heterogeneous processes in IO models may introduce more errors. In so doing, we prove that only when IO-based LCA and process-based LCA have the same level of detail would they yield the same results. Whether hybrid LCA provides more accurate estimates depends on whether the IO model introduced serves as an adequate proxy for the missing products as opposed to if they were estimated by a complete process model. The use of a highly-aggregated IO model runs the risk of overestimation, and could result in a larger relative error than the truncation error resulting from an incomplete process model. Our study seeks to provide a balanced view of hybrid LCA, and our findings offer important insights for future hybrid LCA studies to improve the accuracy and realm of applicability of the approach.