Link4Skills Microsimulation Model

1) Purpose This package contains the code and parameter files to run the Link4Skills microsimulation model for the Reference scenario. Detailed documentation of the model, methodology, and analysis of results is available in: Marois, G., Potančoková, M., Bezat, A. et al. Projecting Labour Market Imbalances and Skill Mismatch Under Demographic Change in the EU. European Journal of Population, 42, 4 (2026). https://doi.org/10.1007/s10680-025-09758-2 Note (Update – 25.03.2026): This version includes an update introducing additional scenarios that combine multiple policy interventions, along with minor fixes in the modelling implementation. 2) Software requirements SAS 9.4 Disk space: ~8 GB free for outputs 3) Folder setup Create the following structure on your machine. The top-level scenario folder must be named exactly: Reference/ ├─ Outputs/ ← results will be written here ├─ Parameters/ ← all parameter go here └─ Population/ ← initial population inputs and population micro-data for each step of the projection Place files as follows: Put POP_2020.csv into Reference/Population/. Put all other CSV parameter files and occup_model.sas7bdat into Reference/Parameters/. Keep Model.sas at the same level as the Reference/ folder. Your tree should look like: .../YourProjectRoot/├─ Model.sas└─ Reference/ ├─ Outputs/ ├─ Parameters/ │ ├─ <all parameter files>.csv ├─ occup_model.sas7bdat └─ Population/ └─ POP_2020.csv 4) Running the model To execute the model, open Model.sas and replace the folder root path defined in the code with the path to your own project root. Once the root path is updated, run the full script. The model will automatically: Read inputs from Reference/Population/POP_2020.csv and all parameter files in Reference/Parameters/. Execute the simulation for the Reference scenario. Write results into Reference/Outputs/. 5) Included aggregated outcome datasets (all scenarios) This deposit includes two precomputed aggregated outcome files that cover several scenarios. These files are provided for convenience, are model outcomes, and are not required inputs to run the code. Output_Pop_count.csvPopulation counts by year, age group, sex, country of residence, region of birth, education, labor force status, and occupation, aggregated across all scenarios. Demand_occup.csvLabor market indicators by scenario, country, and year: labor demand by skill, occupational skills of workers, and vacancies by skill. You can use them directly for analysis or to reproduce figures and tables.