KIMOD - A Dynamic General Equilibrium Model of the Swedish Economy

KIMOD - A Macroeconomic Model for Medium-Term Analysis

KIMOD is a dynamic general equilibrium model of the Swedish economy to be used for aggregate economic-policy analysis and medium-term macroeconomic scenarios. The unit for research and medium-term scenarios at the NIER is responsible for developing and maintaining the model, which was used for the first time in the NIER's work on the Wage Formation Report for 2003.

KIMOD is dynamic, meaning that investment and saving during a period affect future possibilities for output and consumption, respectively, and that all decision-makers take this into account. Time is divided into discrete periods of one year. The projections generated by the model are thus time-series with an annual frequency, and the econometric equations of the model are estimated on the basis of annual data. The national accounts are the preferred data source for initial data and for estimating parameters.

Thus, KIMOD is intended for use in macroeconomic analysis and in medium-term scenarios, medium term being a time horizon of two to six years. For other time horizons, the lower limit of usability for the model results from the fact that the frequency is set at one year. This means that there are no seasonal dynamics at all, and that data for parts of the current year cannot be used as the initial state for the model. For long-term purposes, the usability of the model is limited by the fact that demographic and other structural developments over time have not been modelled.

If you would like to know more about such aspects as the documentation for the model, please contact the staff members below.


KIMOD can be viewed as a synthesis of neoclassical and neo-Keynesian models. The neoclassical features dominate in the longer run (say 7-15 years) and constitute the core theory of the model. In the long run, prices are fully flexible and agents are fully rational. This implies that the economic variables based on the core theory generally respond very quickly to shocks. Such responses are not intended to represent the actual behaviour of the Swedish economy in the short run. Rather, the core theory serves to generate realistic and consistent long-run behaviour of the economy.

In the short run (say 1-6 years), neo-Keynesian nominal rigidities stemming from imperfect information, bounded rationality and adjustment costs come into play. The nominal rigidities imply that volume variables such as production and employment respond sluggishly to shocks so that the model can realistically mimic the business cycle dynamics of the Swedish economy.