Building Community and Capacity for an Internet-Accessible and State-Level Microsimulation Model of Tax and Benefit Policies

Urban Institute, Washington DC

Investigators

Abstract

Social and economic policy increasing involves changes in both benefit programs and the tax systems, yet methods to assess program impacts and interactions are limited, and the available approaches provide primarily federal-level information. This project will develop a complete substantive and technical plan for a new tool for assessing government policies-an Internet-accessible microsimulation model of tax and benefit policies able to estimate the impacts of policy changes at the state level. A microsimulation model is a complex computer program that mimics the operations of government programs using either actual program rules or hypothetical rules. Changes in federal tax and benefit policies are often assessed using microsimulation models developed by or for federal government agencies; this project will lay the groundwork for a tool that will allow microsimulation analysis at the state level, and that will be accessible to anyone with Internet access. With the envisioned model, state government analysts could estimate the cost or caseload impact of a legislatively-proposed change in a tax or benefit program, an academic researcher could assess the impact of a package of changes on state poverty, students could conduct analyses to better understand how programs operate, non-profit policy organizations with competing philosophies could discuss a state policy proposal using quantitative estimates from the same source, and an individual could see how changing programs would affect different families. The design for the new model will be based on an existing simulation model-the TRIM3 microsimulation model, which is funded and copyrighted by the Department of Health and Human Services, Office of the Assistant Secretary for Planning and Evaluation (HHS/ASPE) and developed and maintained by the Urban Institute. The data source will be the American Community Survey, conducted by the Bureau of the Census, which provides large samples for state-level analysis. TRIM3 already includes detailed state-level simulations, but it primarily uses a different data source, and its computer architecture and user interface do not allow extensive public access. The planning for the new model will include three inter-related activities. First, the researchers will hold meetings and webinars with prospective users, including state government staff, academics, and other individuals interested in state-level policies. Participants will be asked what modeling capabilities are most important, the type of user interface that is needed, and the types of modeling results that would be most helpful. Second, the research team will develop prototype designs for the user interface and for results in various formats. These designs will be shared with the potential community to obtain their reactions and suggested improvements. Finally, the research team will determine what computer architecture will best support this type of model, allowing multiple users to simultaneously set up and run simulations, with very fast run speeds. The final written product will be a complete plan for the implementation of the new model. The work will also result in the development of a community of researchers and policymakers who have applied or who want to apply microsimulation techniques to the analysis of policy options in their states.