Undeclared Danish Labor: Using the labor input method with linked individual-level tax data to estimate undeclared work in Denmark.

This paper shows the advantages of an individual-level approach when estimating undeclared work using the labor input method. It shows the shortcomings of an aggregated approach, namely assuming that all workers with missing administrative data are working fully undeclared, being unable to adequately account for overtime, extreme values and not being able to detect and correct for errors in the administrative data. The paper illustrates how these shortcomings can be overcome to a large extent by using an individual-level linked dataset and yield results that are useful both for researchers and for tax authorities. It shows that the method can estimate undeclared work for the self-employed, as well as show seasonal and industry differences in undeclared labor. Denmark is used as a case study, and unlike other papers utilizing individual-level data, this paper provides detailed instruction for how a similar approach can be applied in other countries with Labor Force Survey data. The paper shows how an individual-level approach can yield results that are useful for example for tax administrations' monitoring of undeclared work across sectors. The study uses Danish Labor Force Survey data linked with individual-level tax data, yielding estimates of undeclared work that are in line with past Danish studies of related aspects of undeclared work, namely that approximately 29% of workers have undeclared hours, 25% of wage earners and 37–39% of non-wage earners, and that the value of these hours is close to 2% of the Danish GDP. [ABSTRACT FROM AUTHOR]