Forecasting statewide entitlement caseloads: Analysis of institutional change.

摘要

In the state budget process, lawmakers and budget writers rely on the accuracy of revenue and expenditure forecasts. The expenditure forecasting process is driven by caseload projections of the demand for services such as Medicaid, foster care, long-term care for the elderly and disabled, and public education. This research analyzes the development of the caseload forecasting process in Washington State. In 1997 the Caseload Forecast Council (CFC) was created by the legislature as an independent agency responsible for the production of statewide caseload forecasts.; This study examines the role of the CFC as a driver of institutional change and its impact on forecast processes and outcomes. This study has a quantitative and qualitative component. The quantitative analysis addresses a series of hypotheses on the institutional determinants of forecast accuracy and bias: creation of the CFC led to improved forecast accuracy and reduced forecast bias, improvements in accuracy occurred gradually over time, and forecast accuracy was improved by factors such as more frequent revisions, the technical workgroup, and program knowledge. This research found that forecast accuracy did gradually improve over time after the CFC was created, and that more frequent revisions, the technical workgroup, and program knowledge all contributed to improved forecast accuracy. The qualitative portion investigates the technical workgroup process in more detail via comparison of the different dynamics across workgroups. This analysis did not find any relationship between variations in the workgroup dynamics and forecast accuracy, but it did conclude that variations in workgroup dynamics such as group trust in the forecaster, consensus-building, and expertise of members can affect the legitimacy of the forecast process and product.