Welcome to Mathematica's Microsimulation Model

What is Microsimulation?

This section will take you through a series of screens that demonstrate how microsimulation works. You need to click at the end of each screen to continue.

Simply put, microsimulation is a computer program that simulates how a government program would operate under proposed changes and how participants would be affected. A microcomputer houses information about the low-income population, program rules and operations, and how low-income households participate under different benefit levels. The population is represented by a micro database of administrative records, survey records, or a combination of both.

In this database, a single record represents an individual person or household. The computer, acting as an electronic caseworker, applies the program rules for eligibility to the household and computes the household's benefit. The computer then decides whether the household will participate. Earlier research and other databases provide the basis for predicting whether a household will participate with a change in benefit level.

The computer then processes each remaining household in the database, counting the households that are eligible and that participate, to produce the caseload estimate. It also adds up each household's benefits, producing the cost estimate.

Mathematica's microsimulation models run through a web-based interface that allows users to change program rules governing eligibility, participation, or benefit calculation. By comparing caseload and cost totals under proposed program rules with those under existing program rules, the user can see the projected impact of a proposed policy change. A policy analyst can perform any number of "what if" experiments and view the cost and caseload estimates on the PC screen within a few seconds.

What are the MATH Models?

MATH—Micro Analysis of Transfers to Households—is the name for a family of models that let users analyze the impact of proposed changes to the Food Stamp program (FSP) and other government programs. MATH models perform "what if" experiments with many Food Stamp policy options and can run on administrative records or survey data. These options allow users to look at (1) impacts of proposed reforms on the caseload, (2) impacts of proposed reforms on benefit costs, (3) impacts of proposed reforms on particular subgroups like earners and elderly, and (4) the net impact of proposed reforms that have a positive effect on some groups and a negative effect on others. For example, MATH models have recently helped estimate the cost impact of expanding food stamp eligibility by allowing certain groups of immigrants to participate and by allowing households to not count the value of a car when their assets are calculated. MATH models can also be used to estimate FSP participation rates (the percentage of eligible persons that apply for and receive food stamp benefits) and the extent to which recipients participate in other government programs, such as TANF, SSI, GA, and Social Security.

The models also let users see how subgroups of families, such as those with children, elderly persons, or earnings, will fare when a policy changes. Users can identify who will gain and lose benefits across population groups, across different regions of the country, or across states. The next few screens describe the models and show the results of a hypothetical policy reform.

Click here to continue.

Back to Top