MK-PENS – DYNAMIC MICROSIMULATION PENSION MODEL

Finance Think maintains and adapts the MK-Pens – Dynamic Microsimulation Pension Model. MK-Pens has a dynamic form and involves the movement of individuals in a time horizon as they age, takes into account the mutual (family) relationships of the individuals, their behavioral reactions and the effects of changing of their labor-market status on development indicators. In this way, MK-Pens creates a lifetime for any person, including the likelihood of death, a change in economic status, retirement age, earnings, and the like (Emmerson et al. 2004).

The first component of the dynamics of MK-Pens is the movement of the population over time. It can be determined by static and dynamic components. In the models of static aging of the population, the projections are exogenously given, and only reweighting of the groups is made, without changing the individual characteristics over time (age, change in economic status, mortality, etc.) (Merz, 1993; 1994) . Whereas in dynamic aging models, individual characteristics change endogenously over time (Caldwell, 1990), taking a certain likelihood of changing the characteristics. The number of variables that can be predicted in a dynamic manner depends on the availability of data, the risks and the ability to predict probabilities (Dekkers, 2003). In our model, fertility projections and total population are exogenously given, i.e. taken from the projections of UN-Population (static component), while total mortality is a residual and is calibrated to replicate the mortality rate published by the State Statistical Office of Macedonia. The mortality of the individuals in the sample is derived from a health function (dynamic component).

The second component of the dynamics is the change in the economic status of individuals taking into account their behavioral responses. The goal of behavioral microsystems is to evaluate the behavior of individuals as a function of variables that are directly dependent on the change of simulated policy (Spadaro, 2007). Pension reform has various effects and will cause different behaviors to different groups in the sample. Hence, the assessment of individual behaviors of policy change is significant for later to simulate taxes, contributions and expenses. Individuals in the model are classified into the following categories of economic status: unemployed, employed, inactive working-age persons, pensioners, children and persons without a pension over the age of 62/64 years. The main assumption for the transition from one to another economic status is the employment rate, which is an exogenous variable, and the assumption for employment growth is on average 4 percentage points over a period of five years, according to the average annual growth of the past years. Individuals receive employee status according to a derived function of probability of employment, regressed on a vector of variables.

Children, pupils and pensioners alter the economic status in a static way, by passage of time. Children transit into the status of pupils at age 6; pupils transit into the status of inactive working-age persons with the age of 15 years; employees transit into retirees with fulfillment of retirement conditions. After the change of status, the model reweights the persons in the sample according to their new status. The reweighting is conducted only for statically determined statuses (children, pupils, inactive working-age population, and people in retirement age without a pension).


The model is available to interested parties, only in the premises of Finance Think and under certain contractual relations. In case you want to use it, send a request to info@financethink.mk.