What are persons in employment?

Depending on the survey, persons in employment are shown partly in different delimitations.

Persons in employment according to the concept of the International Labour Organization
The EU term of persons in employment is based on the definition of the International Labour Organization (ILO). It is largely identical to the term used in national accounts. According to that definition, any person aged 15 years or over is considered as employed if he/she worked for remuneration for at least one hour per week.
To ensure a standardised survey in all EU member states, however, Eurostat excludes persons performing compulsory military or alternative service and persons in employment living in institutions, which differs from the ILO definition. Persons who interrupted their work although their employment contract continues to be valid are considered as persons in employment by Eurostat even if the interruption exceeds three months.

Why are there differences in the number of persons in employment between employment accounts and the labour force survey of the European Union (EU)?
Both figures are based on the definition of the International Labour Organization (ILO). In practical work, however, Eurostat applies different delimitations for some groups of persons (cf. keyword “persons in employment according to ILO”). In addition, there are some other differences in results.
The number of persons in employment from the employment accounts are based on a multitude of statistical sources, so that all data sources available are included in accounting. Therefore, when examining employment in the context of short-term economic trends, it is the results of employment accounts that are mainly used. The figures of the labour force survey are based on a harmonised survey conducted in all EU member states. This is to ensure the international comparability of the employment indicators from the survey aspect. In Germany, the labour force survey is conducted as part of the microcensus.
Despite the small definitional differences, there are considerable differences between the results of the labour force survey and the employment accounts. The main reasons are methodical differences between the two statistics: (1) The definition of the International Labour Organization (ILO) on employment differs considerably from everyday understanding, insofar as remunerated activities from one hour per week are considered as employment. So coverage may be problematic in household surveys where respondents consider themselves mainly as, for example, pensioners, unemployed, housewives or students and, consequently, do not indicate minor secondary activities to the interviewer. Employment accounts, however, apply a different methodical approach: For small-scale activities, they mainly use data from the legally required reports on minor employment. (2) Also, as a sample survey, the labour force survey involves a statistical inaccuracy, that is the random sampling error. As the sample of the microcensus is very large, however, that inaccuracy is rather small for the number of persons in employment.
Conclusion: The two figures differ from each other because the underlying data sources are different. When examining short-term and overall economic trends of employment, the figure of employment accounts is used. For European comparisons and for detailed analyses of subpopulations, the labour force survey is used.