Problematic consequences of the inclusion of capital goods inventory data in Environmental Product Declarations

Abstract Purpose A recent update to the Product Category Rules (PCRs) for Construction Products (of the International EPD System) has triggered a methodological issue for owners and users of Environmental Product Declarations (EPDs). The updated PCR has led to capital goods data being implicitly included in the Life Cycle Inventory (LCI) of EPDs. This paper critically examines the role of capital goods in EPDs and establishes major shortcomings in the current methodology, LCI datasets and interpretation. Methods To evaluate the role of capital goods in EPDs, this paper provides a discourse on the fundamentals of Life Cycle Assessment (LCA) methodology, scope, available LCI data and the impact of capital goods on EPD outcomes. Using the ecoinvent database, we analyse the impact of the inclusion and exclusion of capital goods in selected 38 construction products based on the EN 15804+A2 (2019) Standard. Finally, we estimate the relative contribution of capital goods to a suite of Life Cycle Impact Assessment (LCIA) indicators based on the archetypes of capital goods available in ecoinvent and apply Monte Carlo simulation to establish the range of uncertainties in the capital goods data for the selected construction products. Results and discussion Our research confirms that when capital goods are included based on currently available background LCI data, they mostly have a low effect (<10% increase) on climate change, but they can have an enormous effect (>100% increase) on abiotic depletion (minerals and metals), land use and/or human toxicity indicators. Interestingly, when looking further into the ecoinvent capital goods LCI datasets, it becomes clear that there are inaccuracies, inconsistencies, and possibly incorrect estimates of capital goods and infrastructure data. These findings raise questions about the suitability of the underlying LCI background data and whether non-attributable capital goods should be allowed to define EPD outcomes. Conclusion The requirement for the inclusion of capital goods leads to a major conundrum for LCA practitioners. It is suggested that capital goods be excluded until there is better refinement and improvement of the quality of LCI datasets and EPD programs provide clearer guidance on dealing with capital goods. Alternatively, EPDs could document transparently the inclusion or exclusion of capital goods, so that there is a clear separation of the effects of capital goods on LCIA indicators.