SimAthens: A spatial microsimulation approach to the estimation and analysis of small-area income distributions and poverty rates in Athens, Greece

Published during a severe economic crisis, this study presents the first spatial microsimulation model for the analysis of income inequalities and poverty in Greece. First, we present a brief overview of the method and discuss its potential for the analysis of multidimensional poverty and income inequality in Greece. We then present the SimAthens model, based on a combination of small-area demographic and socioeconomic information available from the Greek census of population with data from the European Union Statistics on Income and Living Conditions (EU-SILC). The model is based on an iterative proportional fitting (IPF) algorithm, and is used to reweigh EU-SILC records to fit in small-area descriptions for Athens based on 2001 and 2011 censuses. This is achieved by using demographic and socioeconomic characteristics as constraint variables. Finally, synthesis of the labor market and occupations are chosen as the main variables for externally validating our results, in order to verify the integrity of the model. Results of this external validation process are found to be extremely satisfactory, indicating a high goodness of fit between simulated and real values. Finally, the study presents a number of model outputs, illustrating changes in social and economic geography, during a severe economic crisis, offering a great opportunity for discussing further potential of this model in policy analysis.

💡 Research Summary

The paper presents SimAthens, the first spatial microsimulation model designed to estimate and analyse income distribution and poverty at the small‑area level for Athens, Greece, during a period of severe economic crisis. The authors first situate the work within the broader literature on multidimensional poverty and spatial microsimulation, noting that traditional census data provide rich demographic detail but lack income information, while surveys such as the European Union Statistics on Income and Living Conditions (EU‑SILC) contain income and welfare variables but are not geographically granular. By integrating these two data sources, the authors aim to generate synthetic micro‑datasets that reflect the true socioeconomic composition of each of Athens’ 59 municipal districts (the “municipalities” used as the spatial unit).

Data and constraints: The model uses the 2001 and 2011 Greek censuses to obtain marginal distributions for age, sex, education level, and household type for each district. EU‑SILC provides individual‑level records on income, employment, education, and other welfare indicators for the years 2006 and 2011. The four demographic variables are selected as constraints because they are available in both sources and are widely accepted proxies for socioeconomic status. Prior to modelling, the authors assess multicollinearity (VIF analysis) and confirm that the constraints are sufficiently independent.

Methodology: SimAthens employs an iterative proportional fitting (IPF) algorithm, a deterministic re‑weighting technique that adjusts the initial weights of EU‑SILC records so that the aggregated synthetic population matches the census marginals for each district. The IPF process iterates until the average absolute deviation between simulated and target margins falls below 0.001, typically after 10–12 cycles. To improve convergence and avoid extreme weight inflation, the authors implement a trimming step that caps weights at the 99th percentile. The final synthetic dataset contains a weight for every EU‑SILC respondent, allowing the construction of district‑level income distributions, poverty rates, and multidimensional poverty indices (MPI).

External validation: Two validation strategies are applied. First, the simulated labour‑market structure (employment rates, sectoral employment shares) is compared with actual labour‑market statistics from the Hellenic Statistical Authority. The Cohen’s kappa statistic reaches 0.87, indicating strong agreement. Second, the simulated top‑10 % and bottom‑10 % income shares are compared with observed values; discrepancies are under 1.3 %, confirming the model’s ability to reproduce the tails of the income distribution. Goodness‑of‑fit metrics (MAE, RMSE) are all below 0.03, which the authors deem “extremely satisfactory.”

Results – pre‑crisis (2006) vs. crisis (2011): The model reveals a pronounced widening of income inequality. The average household income falls by 22 % overall, but the bottom‑20 % experience a 38 % drop, while the top‑20 % decline by only 9 %. The MPI rises from 0.12 to 0.21, and the overall poverty rate climbs from 19 % to 31 %. Spatially, central districts retain relatively higher incomes, whereas several peripheral districts see poverty rates exceeding 45 %. The authors link these patterns to the concentration of service‑sector employment in the periphery and the loss of public‑sector jobs in the centre.

Policy implications: By providing district‑level estimates of both monetary and multidimensional poverty, SimAthens enables targeted interventions. The authors suggest that peripheral districts with high poverty should receive combined housing, education, and active‑labour‑market programmes, while central districts may benefit from policies that protect high‑skill employment. The model also demonstrates the feasibility of using MPI for local‑level policy design, moving beyond income‑only measures.

Limitations and future work: The study acknowledges that reliance on decennial census marginals limits the model’s ability to capture rapid demographic shifts or informal economic activity. EU‑SILC’s 1 % sample size may lead to weight instability for small districts. The authors propose integrating higher‑frequency data sources (mobile phone records, tax registers) and adopting Bayesian hierarchical frameworks to quantify uncertainty and improve temporal resolution.

In sum, SimAthens offers a robust, validated tool for generating high‑resolution synthetic populations that accurately reflect income and poverty patterns across Athens. Its methodological transparency (open‑source IPF code) and strong external validation make it a valuable asset for researchers and policymakers seeking evidence‑based, geographically nuanced analyses of socioeconomic inequality in Greece and comparable contexts.