Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
FUTURE COMPUTING ENVIRONMENTS FOR MICROSIMULATION MODELING 200 need for simple, understandable model rules in the policy arena encouraged the construction of static models, the inner workings of which could be explained and interpreted more easily than for dynamic models. The primary demand for socioeconomic microanalytic simulation models has therefore been substantially limited to individuals and organizations that are directly involved in or that support the formation and evaluation of policy alternatives. With the exception of pure income tax simulations, such activity has occurred mostly at the national level, owing to the level of resources currently required to implement detailed models and the lack of sufficiently dense microdata sets for producing statistically meaningful results for areas not having large populations.53 Despite the promise of socioeconomic microanalytic simulation techniques, the use of such models is rare in university instruction. Principal causes have been the brief duration of the semester or quarter unit of course duration, the amount of time to learn and use microanalytic simulation systems, the significant computing resources required to support the use of such models in a college or university environment, and the current difficulty of defining realistic policy experiments. This situation is particularly unfortunate because it tends to limit knowledge of microanalytic simulation modeling mainly to those already engaged in it. The net result is that the current population of individuals concerned with microanalytic simulation modeling, either as a methodology or as a projection tool, is very limited and does not appear to be growing, at least in the United States.54 If such models have a substantial payoff in the national policy arena, they are not achieving their potential contribution. Nor are social scientists being introduced to such models during the formative period of their professional training. Assuming that the technique has more potential than is currently being realized, investments that lower cost and time barriers to learning about and using such models have a double payoff. Factors Affecting the Availability of Microsimulation Models The primary factors that affect the production, availability, and usefulness of socioeconomic microanalytic simulation models are (1) substantive knowledge in the social sciences required to build operating characteristics for such models 53 This situation is in contrast to state simulation of alternative federal and state income lax policies, where microdata for the tax-paying units under current law are captured routinely through the tax-filing process. Nevertheless, the simulation of any tax policy that increases the tax base must account for those units that were not in the original tax-paying unit population but that will enter it as a result of a change in the law. This situation is analogous to simulating policies that increase eligibility for benefits, causing new benefit recipient units to enter the relevant population. 54 Both the demand for microanalytic simulation tools for the analysis of tax and transfer policies and the availability of SPSD/M have resulted in a recent burst of activity in this area in Canada.