Annex 4 analytical methods

In the tradition of computable general equilibrium (CGE) models, RHOMOLO relies on an Arrow-Debreu equilibrium framework, in which supply and demand depend on the price system. Policies are introduced as shocks. Following a shock, the system moves towards a new equilibrium, with adjustments driven by optimal supply and demand behaviours. Like all CGE models, RHOMOLO therefore provides an evaluation of the interaction effects between all agents through markets, imposing full system consistency. This type of analysis is known as scenario analysis, whereby the results of simulations including policy shocks are compared to a baseline scenario with no shocks.

Given RHOMOLO's regional focus, particular attention is devoted to explicitly modelling spatial linkages, interactions and spillovers between regional economies. For this reason, models such as RHOMOLO are referred to as spatial CGE models.

Each region is inhabited by households that are aggregated into a representative agent whose preferences are characterised by a love of variety. Households derive income from labour (in the form of wages), physical capital (profits and rents), other financial assets and government transfers. Labour mobility can be switched on or off depending on the needs of the analysis to be carried out. Household income is spent on savings, consumption and taxes.

Firms in each region produce goods that are sold and consumed in all regions by households and governments. Other firms, either in the same sector or others, can also use these goods as inputs in their production processes. Transport costs for trade between and within regions are assumed to be of the 'iceberg' type, varying according to sector and region pair. The market structures of industrial sectors in each region can be modelled as either perfectly or imperfectly competitive. The latter can be characterised as monopolistic competition, Cournot oligopoly or Bertrand oligopoly. The number of firms in each sector and region is estimated using national Herfindahl indices, under the assumption that all firms in one region use the same technology. Firms with higher market shares are able to extract higher mark-ups from consumers than their competitors because of their higher weight in the price index. Since market shares vary by destination market, mark-ups also vary by destination market.

The article by García-Rodríguez et al. (2025) illustrates the dataset underlying RHOMOLO V4. Furthermore, the model requires several calibrated inputs and exogenous parameters to function. For instance, the interest rate is set at 0.04%, while the rate of depreciation of private capital is set at 0.15%. More generally, parameters related to elasticity of substitution on both the consumer and producer sides are based on similar models or derived from econometric literature.

Further details on model parameterisation can be found in Section 3 of the article by Salotti et al. (2025).

All RHOMOLO output variables are available on a yearly basis, with the level of detail broken down by region and sector. Regional variables can be aggregated by country, EU or other geographical area. 

• Households-related output variables:
  • Factor supply by household; Income; Taxes paid on income; Savings; Consumption; Price of consumption; Net disposable income.
• Firms-related output variables:
  • Price of exports; Lerner index of monopoly power; Market share; Average sales price; Average production cost; Profits; Fixed cost of production; Marginal cost of production; Aggregate intermediate input; Aggregate input of primary factor; Price of aggregate intermediate input; Intermediate demand for each good; Total factor productivity; Aggregate labour-factor demand; Price of aggregate labour-factor demand; Price of aggregate input of primary factor; Demand of each factor; Taxes paid on demand of each factor; Taxes paid on sales.
• Investment-related output variables:
  • Investment; User cost of capital; Price of investment.
• Government-related output variables:
  • Factor supply by Government; Income of Government; Current expenditure; Public investment; Price of consumption of Government; Transfers from Government to household; Savings.
• Import-related output variables:
  • Demand for composite of each good; Price of each composite good’s demand; Exports; Net exports.
• Other variables:
  • Price of each factor; Unemployment rates; Number of firms in each sector.

The RHOMOLO model is designed for policy impact assessment. The explicitly modelled spatial dimension at the NUTS2 regional level makes it a unique tool for territorial impact assessment. Spatial interactions between regions are captured through trade of goods and services (which is subject to trade costs), income flows, factor mobility and knowledge spillovers, making RHOMOLO particularly well suited for simulating human capital, transport infrastructure, R&D and innovation policies.

RHOMOLO has been used for the impact assessment of the European Regional Development Fund (ERDF), the Cohesion Fund, Horizon Europe, the European Social Fund (ESF),the European Competitiveness Fund, the portfolio of the European Investment Bank, and the European Defence Fund, among others.

Moreover, RHOMOLO is used for the evaluation of specific investment projects and other reforms depending on the requests made by Member States and regional authorities.

• Education and training 
• Economy, finance and the euro 
• Taxation 
• Employment and social affairs 
• Regional policy 
• Transport 
• Digital economy and society 
• Business and industry 
• Research and innovation 
• Single market 
• Trade