E17 - General Aggregative Models: Forecasting and Simulation: Models and ApplicationsReturn

We present a set of practical applications of the household sector stress testing approach used at the Czech National Bank. The aim is to test households' ability to repay their loans in the event of extremely adverse economic developments. Besides this, the household sector stress test methodology has so far been used for two other purposes: (i) to construct a simple reverse stress test and explore the sensitivity of Czech households to a rise in loan interest rates and a decline in income, and (ii) to provide some supportive evidence for the calibration of debt service-to-income limits.

Our paper focuses on the analysis of supply and demand shocks and on the estimation of expected costs of introducing Euro currency into Czech Republic (CR). The analysis is based on the theory of optimal currency areas by Mundell (1961, 1973) and uses a macroeconomic approach formalized by Bayoumi (1994). VAR models and the Blanchard-Quah decomposition (Blanchard and Quah, 1989) are used in order to simulate aggregated macroeconomic impulse response dynamics and to isolate supply and demand shocks for further inspection. Based on the analysis performed we conclude that given current circumstances and persistent differences in symmetry of economic shocks in CR and selected Euro zone countries, the costs from introducing the Euro to CR (as measured by fluctuations of real macroeconomic variables) would be nonzero, however presumably not significantly different from equivalent costs experienced in Austria or Slovakia (i.e. not prohibitive or signifi cantly damaging).

The Czech Republic is going to face ageing of its population. It will affect the economy in many ways. The pension system is one of them. This paper provides a view on possibilities how to insure long-term stability of the pension system in the Czech Republic using a mix of pay-as-you-go and fully funded system. Simulations are based on OLG model, long-term demographic forecast and limits of fiscal policy stemming from the necessity to fulfill Maastricht criteria and The Stability and Growth Pact. Those obligations creates a frontier for plausible solutions. Results suggest that it is possible to find a solution for mixed system providing more favorable conditions than purely parametric changes of PAYG. Taking into account fiscal limits, the contribution rate to the FF pillar would be similar to the rates in other CEE countries where pension reform already started.

This contribution undertakes to outline the state space models and the recursive technique called the Kalman filter to a wider audience of the Czech economics readers. One can find both terms in a whole range of empirical studies from recent years. Following this approach also allows us among other things to model unobservable variables such as the potential output or the natural rate of unemployment. The primary motivation of this article is to bring a basic introduction to the reader who does not deal with the time series analysis on an everyday basis and to sketch the roots of state space models and the Kalman filter. Those who are interested can find more proper and rigorous description in the original literature.