The gap between macro and micro economic statistics: Estimation of the misreporting model using micro-data sets derived from the Consumer Expenditure Survey

In many countries we observe a gap between macroeconomic and microeconomic statistics. In order to explain the underreporting observed in microeconomic statistics, the present paper tests the misreporting hypothesis through the double hurdle model. The misreporting hypothesis is based on some key assumptions: It is assumed that there are two categories of 'zero expenditure households' in the Survey - one category involves households that did not purchase consumer durables during the survey period and reported zero expenditure correctly on the Survey while the other category of 'zero expenditure households' involves households that, although they purchased consumer durables during the survey period, reported zero expenditure on the Survey. This is the source of misreporting. We also assume that there are 'positive expenditure households' for consumer durables in the Survey. These positive expenditure households for consumer durables are assumed, in the misreporting hypothesis, to have reported correctly their expenditure on consumer durables in the Survey. This model enables us to correct for over-reporting of zero expenditure households in the micro data. The data used for estimation involves thirteen clusters of consumer durables from the Consumer Expenditure Survey complied by the Bureau of Labor Statistics. The empirical results are satisfactory in supporting the conclusion that misreporting plays an important role in the underreporting in microeconomic statistics compiled by the Bureau of Labor Statistics compared with macroeconomic statistics compiled by the Bureau of Economic Analysis, the US Department of Commerce