|
Abstract:
|
The Bank of Canada, as the sole issuer of banknotes in Canada, is interested in tracking cash usage over the course of the pandemic. We have used a 3-day payment diary to estimate cash volume shares at the point-of-sale. Diary recording methods are useful for reducing measurement error due to recall, which are known to result in under-reporting of small-value expenditure. Given that cash is most often used for small-value transactions, cash volume may therefore be under-estimated when using recall-based methods. However, diary methods come with added cost, complexity and respondent burden. Thus, diaries may not be feasible to undertake on a frequent basis and may suffer from sample selection bias due to lower response rates, compared to the recalled questionnaire. This study uses data from the November 2020 Cash Alternative Survey, which asks respondents to complete both a recall questionnaire and a 3-day payment diary, to study differences in measurement and selection errors between the two approaches. We use the results to predict diary shares over the course of 2020, and by comparing to pre-COVID diaries paint a picture of both the long- and short-term trends in cash usage.
|
