This paper considers what role in-home barcode scanner data could play in collecting household expenditure information as part of national budget surveys. One role is as a source of validation. We make detailed micro-level comparisons of food and drink expenditures in two British datasets: the Living Costs and Food Survey (the main budget survey) and Kantar Worldpanel scanner data. We find that levels of spending are significantly lower in scanner data. A large part (but not all) of the gap is explained by weeks in which no spending at all is recorded in scanner data. Demographic differences between the surveys accentuate rather than close the gap. We also demonstrate that patterns of expenditure across the surveys are much more similar, as are Engel curves relating food commodity budget shares to total food expenditures. A key finding is that the period over which we observe households in the scanner data significantly alters the distribution, but not the average, of weekly food expenditures and budget shares, which has important implications for whether two-week spending diaries common to budget surveys are giving a truly accurate reflection of a household's typical spending patterns. A second, more involved use of scanner data would be to impute detailed commodity-level expenditure patterns given only information on total expenditures, as a way of reducing respondent burden in budget surveys. We find that observable demographics in the scanner data explain little of the variation in store-specific expenditure patterns, and so caution against relying too heavily on imputation.
Scanner data expenditure inflation food measurement