Estimating global copper demand until 2100 with regression and stock dynamics

Branco W. Schipper*, Hsiu Chuan Lin, Marco A. Meloni, Kjell Wansleeben, Reinout Heijungs, Ester van der Voet

*Corresponding author for this work

    Research output: Contribution to JournalArticleAcademicpeer-review


    Future global copper demand is expected to keep rising due to copper's indispensable role in modern technologies. Unfortunately, increasing copper extraction and decreasing ore grades intensify energy use and generate higher environmental impact. A potential solution would be reaching a circular economy of copper, in which secondary production provides a large part of the demand. A necessary first step in this direction is to understand future copper demand. In this study, we estimated the copper demand until 2100 under different scenarios with regression and stock dynamics methods. For the stock dynamics method, a strong growth of copper demand is found in the scenarios with a high share of renewable energy, in which a much higher copper intensity for the electricity system and the transport sector is seen. The regression predicts a wider range of copper demand depending on the scenario. The regression method requires less data but lacks the ability to incorporate the expected decoupling of material use and GDP when the stock saturates, limiting its applicability for long-term estimations. Under all considered scenarios, the projected increase in demand for copper results in the exhaustion of the identified copper resources, unless high end-of-life recovery rates are achieved. These results highlight the urgency for a transition towards the circular economy of copper.

    Original languageEnglish
    Pages (from-to)28-36
    Number of pages9
    JournalResources, Conservation and Recycling
    Publication statusPublished - 1 May 2018


    • Circular economy
    • Copper applications
    • Copper recycling
    • Global copper demand


    Dive into the research topics of 'Estimating global copper demand until 2100 with regression and stock dynamics'. Together they form a unique fingerprint.

    Cite this