The Public Policy Institute of California and the Stanford Center on Poverty and Inequality have joined forces to develop a new poverty measure for the state of California. Limitations to the official poverty threshold are widely recognized; the Census research Supplemental Poverty Measure (SPM)  1. sets more realistic poverty thresholds based on actual household expenses across the nation, and varies geographically with the cost of living; 2. considers the role of expenses on things such as transportation to work, child care, and medical care; and 3. calculates resources including non-cash transfers, like SNAP and TANF, and refundable tax credits. The CPS-based Census estimates by state reveal that California has the highest poverty in the nation. However, this research measure does not allow for nuanced examination of variation within and across California—by demographic characteristics, or at the county level.

In this paper, we discuss the development of an SPM-style measure specifically for California using a variety of national data (ACS, CPS, CES) and administrative data for the state. Basing our analyses fundamentally on variables available in the ACS, we will report one year poverty estimates for California , for specific demographic groups, and for sub-state regions. We take a broadly similar approach as Census and state-level SPM research. However, our preliminary analyses reveal substantial variation across the state both in expenditures, like housing costs, and also on the in-kind SNAP program. Therefore, we focus attention on the sensitivity of our California Poverty Measure estimates to assumptions made about these factors.

On the resources side, we use administrative data on SNAP receipt and dollar values to adjust estimates in the ACS  given known underreporting in household surveys. We specifically consider the sensitivity of the SNAP estimates to the assumptions made about the roughly 2.5 million undocumented immigrants in California and the 1.2 million SSI recipients, both of which are categorically ineligible for SNAP, despite being likely eligible on income grounds . On the expenditures side, we begin with a discussion of thresholds based on national expenditure data, then adjust for California-specific housing costs in several ways that account for ownership and mortgage status. Other state-specific adjustments are made for medical out of pocket expenses, work expenses, child care expenses and additional safety net programs to obtain a detailed understanding of economic hardship in California.

We have committed to releasing  estimates for 2011 at the state and county level in October 2013. In advance of the APPAM conference, we will construct several scenarios that highlight the impact of different expenses and transfers on poverty and also demonstrate how potential changes in safety net programs might affect the poverty rate across counties and demographic groups.