Inforum's State Employment Modelings (STEMS)

The Inforum State Employment Modeling System (STEMS) serves to estimate employment, with industry detail at the state level. It's data base is derived from the Department of Labor Bureau of Labor Statistics Employment and Earnings data. The model is made up of employment forecasting equations for individual industries in each state. The industries are divided into two groups: base and secondary. Estimates for the base group industries are dependent on national levels of employment and trends in state shares of national employment. Estimates for the secondary group industries are also dependent on national levles and state trends, as well as on estimates for the base industries in the same state. The base industries are those engaged in manufacturing, agriculture, and mining, along with federal government "industry". Secondary industries are those engaged in providing services, and the construction industry. Employment estimates in STEMS are not based on constant shares, but respond to trends in individual industries. The secondary employment levels depend on base employment levels as well.

The STEMS forecasts rely on employment estimates generated by Inforum's LIFT model. The STEMS forecasts are consistent with the LIFT forecasts, which serve as controls on national employment estimates. Procedures within the STEMS model insure that state-level estimates add up exactly to national estimates. The STEMS model is currently simulated within an IBM-compatible personal computer environment. The STEMS model takes just a few minutes to calculate results for a given simulation. A standard set of tables display the results of STEMS model simulations.

The most significant uses of STEMS to date include detailed analysis of employment trends in the State of Maryland  nd an analysis of the impacts of the North American Free Trade Agreement on the economies of individual states, particularly those which share a border with Mexico.

About Us | Site Map | Contact Us | Page last updated: November 5, 2020