In this study we investigate the external validity of the food insecurity and insufficiency measures—specifically, how these measures correlate with food expenditures—using the December 2003 Food Security Supplement of the Current Population Survey. We focus on a special segment of the population—households with low incomes and low food expenditures. If reports of food problems are based on a lack of food, reports should be nearly ubiquitous at the bottom of the expenditure distribution. We find, however, that this is not the case. We define and scale food expenditures several different ways and find that the reported incidence of food insecurity never rises above one-half anywhere along the corresponding expenditure distributions, leading to concerns about the external validity of the measure.