Static versus Dynamic Microsimulation

There are two broad classes of microsimulation models-"static" and "dynamic." The terms refer primarily to the methods used to "age" a microdata file to create a synthetic file for a future year. Static microsimulation models such as TRIM3 typically use static aging techniques, changing certain variables on the original microdata file to produce a file with the demographic and economic characteristics expected in the future year. Person weights are modified to change the total population and the weighted characteristics of the population; labor force status may be changed to alter the unemployment rate; and incomes are adjusted for price changes. Simulations can then be run on the aged microdata file to estimate the impact of a change to be implemented in the future year. However, the model is not designed to estimate the effect of a program change on the future economic and demographic characteristics of the population.

"Dynamic" microsimulation models age each person in the microdata file from one year to the next by probabilistically deciding whether or not that person will get married, get divorced, have a child, drop out of school, get a job, change jobs, become unemployed, retire, or die. Simulations of government programs can be run in the current year, the final year of the aging process, or any interim year. The simulation of a government program in one year may affect a person's characteristics in the subsequent year. For example, whether or not someone will drop out of school could be programmed to depend partly on family income, which could, in turn, be affected by government transfer payments.

Static and dynamic models each have their own strengths. Dynamic models feature more detailed and realistic population aging. Although dynamic models have been used to age microdata files 50 or more years into the future, the simpler aging procedures in static models are generally only applied to estimates for the near future. Dynamic models are often viewed as better able to produce realistic long-range estimates, which account for interim shifts in economic and demographic trends. The advantage of static models such as TRIM3 is their very detailed program simulations. The more simplified program simulations in dynamic models may not incorporate the detailed program rules that are often the basis of policy changes. Future models might include both detailed dynamic aging and detailed program simulations.

TRIM3 does not currently have a module to allow point-and-click access to aging methods. However, programmers can create modified input files that are adjusted to hit targets for population, income levels, and employment levels.