An evaluation of the scale of undeclared work in the European Union and its structural determinants: estimates using the Labour Input Methodas
The analysts at the Grey Economy Information Unit of the Finnish Tax Administration have studied the report published by the European Commission, the method used and the data concerning Finland. We wish to state the following with regard to the data used in the evaluation:
In the report undeclared work is measured as the discrepancy between the demand for and supply of labour. The difference between the statistics is interpreted as the scale of undeclared work, which means that the reliability of the method is based on the comparability of the sets of data.
The supply of labour shown in the measurement is obtained from the Labour Force Survey. In Finland this is done by Statistics Finland by interviewing 12,000 private individuals on a monthly basis. Based on the survey, the supply of labour (hours worked, labour input) can be divided between employees and self-employed persons.
The demand for labour comes from the structural business statistics. The survey gives the number of employees in terms of full-time equivalents. For companies not included in the surveys the data as full-time equivalents is estimated. For employees the data is estimated using the annual declarations to the Tax Administration. For self-employed persons the data is estimated using sets of data such as those of the Tax Administration and Social Insurance Institution of Finland Kela. The surveys do not directly give the number of hours worked but an estimate of the number of employees. In Finland the staff numbers of companies are to a large extent based on estimates. Estimating the numbers concerning self-employed persons, in particular, has proven difficult due to insufficient data. This obviously creates uncertainty with regard to the results of the estimates.
The reasons why there may be errors in the results of the measurements include at least the following:
The statistics to be compared – demand for and supply of labour – have been compiled in different ways, any discrepancies may be due to the statistical methods.
The statistical quantities – number of jobs vs. hours worked – must be converted into commensurate ones, which may also cause errors in the measurement.
In the report the estimated share of undeclared work in Finland in total is 9.3%, for those in employment relationship it is 3.6% and for self-employed persons it is 45.3% (the results concern work in the private sector only).
The result for Finland seems unbelievably high – almost half of the labour input by self-employed persons would be undeclared. The reliability of the result suffers from the uncertainty of the data source concerning demand. In terms of self-employed persons, in particular, the coverage of the survey for the structural business statistics is weak as it is not sent to companies with less than 5 employees. Thus the numbers concerning self-employed persons are largely based on estimates.
The Finnish and Swedish societies, economies and business life can in most respects be considered very similar. According to these results, however, the share of undeclared work among self-employed persons is much lower in Sweden than in Finland. This is a further indication that the results of the evaluation are not reliable.
From the perspective of the fiscal grey economy measuring undeclared work among self-employed persons as such is not that relevant, especially if this is based on the hours worked. Self-employed persons may not pay themselves any hourly wages but their earnings are composed of the company’s profit. This means that a measurable variable other that the hours worked should be used, e.g. turnover, business profit or taxes paid.
Besides the public evaluation report, the analysts of the Grey Economy Information Unit had access to a longer, unpublished version of the report. Neither one of these reports presented any arithmetical operations concerning the scale of undeclared work presented as the results. More detailed information on how the figures for the scale of undeclared work have been obtained would be needed for a more in-depth analysis of the evaluation.