Association for Public Policy Analysis & Management

A number of studies examine the concept of household financial security or fragility using measures of liquidity constraints, including the collection of detailed credit and savings data. Other studies rely on more subjective assessments of a household’s ability to pay a given sum of money within a defined time period. These types of questions have been used across a number of national surveys, including the Survey of Consumer Finances, Consumer Expenditure Survey, and Financial Capability Study, among ad hoc surveys and evaluations. Yet, the reliability and validity of these questions remains untested. Moreover, there is not a standard form of this question with variation in both the amount and the time involved. Some forms of questions are limited to personal financial accounts, while others incorporate the ability to access funds via social networks and family.  Without a standardized method of measuring access to liquidity in the case of a shock, research across studies or populations is limited. 

This study examines how variations in the form of questions used to measure household savings or reserve funds may generate consistent or divergent findings.  Using an online study sample, this paper tests a series of items that measure small dollar emergency savings and the capacity to cope with financial shocks. We include liquidity-type questions from established surveys, along with validated scales of food security and financial status to serve as comparison measures.   We randomly assign respondents to report on their self-rated ability to acquire between $500 and $2,000. We also vary the time period from five days to 30 days. We correlate these measures across respondents, as well as within respondents to other measures that proxy for financial status. We provide a summary of various approaches to develop an ideal set of questions that can be used to reliably measure the general concept of “emergency savings” and offer insights into how this set of measures can be used in future research and policy evaluations.