Man0EUvRE CS3 Dataset: Renewable Pulls and Industry Relocation

Final Industrial Energy Demand under Renewable Energy Endowment Shocks – Simulation Results from Case Study 3 (Man0EUvRE Project) Description: This dataset contains simulation results on sector- and country-level final industrial energy demand generated by the agent-based macroeconomic model developed in Case Study 3 (CS3) of the Man0EUvRE project ("Energy System Modelling for Transition to a net-Zero 2050 for EU via REPowerEU", Grant Agreement No. 101069750, co-funded by the European Commission under the CETPartnership Joint Call 2022). Scientific context The transition to renewable energy reshapes industrial competitiveness because the distribution of renewable resources is geographically uneven. Regions endowed with abundant low-cost renewable electricity may develop new comparative advantages, potentially attracting industrial production – a mechanism referred to as the renewable pull effect (Samadi et al., 2023). CS3 investigates how such heterogeneous renewable energy endowments affect industrial relocation decisions and the resulting country-specific final energy demand across Europe. The underlying model is a discrete-time, agent-based, stock-flow consistent macroeconomic simulation framework built with the open-source sfctools library (DLR). It represents 30 industrial sectors across 11 European countries in a multi-regional input–output structure calibrated to EXIOBASE 3.9.5. Firms are heterogeneous agents that compare unit production costs across countries and may relocate probabilistically (multinomial logit rule with home bias and congestion frictions) or – in an extension scenario – switch products within a capability-constrained product space. Energy endowment shocks are derived from the renewable export cost index of Kan et al. (2025) and applied as permanent proportional changes to country-level energy endowments at the mid-point of each simulation run (T = 340 periods, 20 Monte Carlo repetitions per scenario). Dataset contents The dataset consists of two files reporting Monte Carlo summary statistics of final industrial energy demand: CS3_IAMC_2022_means.xlsx – Monte Carlo means across 20 simulation runs CS3_IAMC_2022_medians.xlsx – Monte Carlo medians across 20 simulation runs Both files follow the IAMC data format (long format: Model / Scenario / Region / Variable / Unit / 2022) and report final energy demand in EJ/yr for the post-shock equilibrium state. Variables include sector-level demand for 30 explicitly modelled industries (e.g. Final Energy|Industry|C_STEL for steel, Final Energy|Industry|C_CHEM for chemicals) as well as aggregate categories (Final Energy|Industry, Final Energy|Industry|Other, Final Energy|Industry|FossilFeedstock). Scenarios Five scenarios are included, varying behavioral and adjustment parameters while holding all other calibration targets and endowment shocks constant: Scenario β_C κ τ Product switching Reference No-Shock 8.0 0.02 2.0 Off Base Shock 8.0 0.02 2.0 Off Beta_High Shock 16.0 0.02 2.0 Off HB_Low Shock 8.0 0.00 2.0 Off Temp_Low Shock 8.0 0.02 1.5 Off With_Prodswitch Shock 8.0 0.02 2.0 On The Reference No-Shock scenario provides the counterfactual baseline without any energy endowment modification. The remaining scenarios apply regional renewable energy endowment shocks (δ_r) derived from Kan et al. (2025) and differ only in relocation friction and cost-sensitivity parameters, enabling robustness analysis. Geographic and sectoral scope Regions: Denmark, Finland, France, Germany, Greece, Italy, Netherlands, Norway, Poland, Spain, Sweden. Explicitly modelled industries (30): aluminium (C_ALUM), chemicals (C_CHEM), cement (C_CMNT), copper (C_COPP), ceramics (C_CRMC), electrical machinery (C_ELMA), fabricated metals (C_FABM), furniture (C_FURN), garments (C_GARM), glass (C_GLAS), leather (C_LETH), lead/zinc/tin products (C_LZTP), machinery and equipment (C_MACH), media (C_MDIA), medical instruments (C_MEIN), motor vehicles (C_MOTO), nitrogen fertilisers (C_NFER), office machinery (C_OFMA), other non-ferrous metals (C_ONFM), other non-metallic minerals (C_ONMM), other transport equipment (C_OTRE), paper (C_PAPE), phosphate fertilisers (C_PFER), plastics (C_PLAS), radio/TV equipment (C_RATV), rubber products (C_RUBP), steel (C_STEL), textiles (C_TEXT), wood products (C_WOOD), other manufacturing (C_OFMA). Important scope note: Quantitative findings apply exclusively to the 30 explicitly modelled industries. The aggregate variable Final Energy|Industry|Other carries no endogenous relocation dynamics and represents sectors not individually resolved in the model. Households and the service sector are not covered. Intended use These results have been handed over to TU Berlin for integration into the GENeSYS-MOD energy system model as part of the Man0EUvRE project's European scenario analysis. References Kan, X., Reichenberg, L., Hedenus, F., & Daniels, D. (2025). Renewable export cost index as an indicator of global renewable energy trade potential. Communications Earth & Environment, 6, 112. https://doi.org/10.1038/s43247-025-02094-7 Samadi, S., Fischer, A., & Lechtenbohmer, S. (2023). The renewables pull effect: How regional differences in renewable energy costs could influence where industrial production is located in the future. Energy Research & Social Science, 104, 103257. Stadler, K., Wood, R., Bulavskaya, T., et al. (2025). EXIOBASE 3 (3.9.5) [Data set]. Zenodo. https://doi.org/10.5281/zenodo.14869924 Baldauf, T. (2023). sfctools [Software]. DLR Institute for Networked Energy Systems. https://gitlab.com/dlr-ve/esy/sfctools Funding This research was funded by CETPartnership under the European Partnership Joint Call 2022, co-funded by the European Commission (Grant Agreement No. 101069750).