Misclassification in an experimental poverty measure

Among the recommendations and proposals set forth in a 1995 National Research Council (NRC) panel report on measuring poverty, perhaps the most controversial was the treatment of medical needs. A test of poverty misclassification using data from the Consumer Expenditure Survey does not support the contention that medical needs must be treated differently from other needs in the measurement of poverty. Household out-of-pocket medical expenditures often do not occur uniformly over the year. Consequently, the distribution of such medical expenditures in quarterly data, the period used most by the Bureau of Labor Statistics in its publications of expenditure data, is much more skewed among households than is the distribution of annual medical expenditures. The NRC panel proposed to vary its new poverty thresholds by the number of adults and children in the family, as well as geographically.