KIMOD consists of three separate, although highly interrelated, models. Simplifying somewhat, we could say that the three models are valid at different time horizons. In the (very) long run, when the effects of nominal and real rigidities have faded out, the values of the variables are the same across the three models.
The steady-state model is a static model based on the neoclassical theory of consumption, investment, production, and the labour market. In steady state, all real variables, such as GDP, consumption, and investments grow at the same rate, which implies that, for example, the consumption-to-GDP ratio is constant.
Moreover, the steady state model is a pure real model, in that there are no nominal prices, nominal interest rates or nominal exchange rates, only relative prices, real interest rates, and a real exchange rate. These are all constant in steady state, which implies that all nominal prices increase at the same rate.
The second part of KIMOD is the flex-price model. Prices and wages are fully flexible and all agents are fully rational and have perfect foresight. Thus the steady-state model is the static version of the flex-price model. The dynamics are mostly due to the introduction of two real rigidities, one in the capital market and one in the labour market. In the capital market, there are adjustment costs to investments.
This implies that the capital stock adjusts only gradually to a changed economic environment, i.e. a change in the steady state. The labour market is modelled as in Pissarides (2000). Vacancy costs imply that employers hoard labour if a fall in demand is not permanent. Equilibrium unemployment is determined by replacement ratios, separation rates, vacancy costs, matching efficiency and other variables incorporated in search theory.
In practical applications, KIMOD shares the same equilibrium values for the likes of output, unemployment and the real exchange rate as used elsewhere at the NIER. It is not likely, though, that all the variables in the economy will start at their flex-price equilibrium values. The time it takes for them to reach the flex-price counterparts is, however, less than it takes to reach the steady state. In absence of shocks, the economy should normally reach the flex-price equilibrium within 5—10 years. We emphasise that the flex-price equilibrium reaches the steady state only after the effects of the real rigidities in the markets for capital and labour have faded out.
The third and final part of KIMOD is the sticky-price model. Here, nominal rigidities in prices and wages are introduced which lead to sluggish volume adjustment. The nominal rigidities stem from the presence of imperfect information and bounded rationality. This implies, for example, that consumers, investors, and wage-setters are partially backward-looking; expectations are partly adaptive.
As time goes by after a shock, however, the agents learn and become more informed. As a result, variables adjust gradually towards their flex-price equilibrium values. The nominal rigidities give room for stabilization policies, such as monetary and fiscal policies, which aim to move the economy in the direction of the flex-price equilibrium. It is important to stress, though, that stabilization policies can not alter any equilibrium values. For projection purposes, the Sticky-price model variables start in actual data, i.e. the data used elsewhere at the NIER.