Abstract

Introduction
In an era when sample surveys are expensive and respondents are hard to find and extract information from, administrative data sources become more attractive. In addition, due to the process of digitalization, individuals leave both more and more frequent administrative footprints after their activities. Monthly statistics on population’s statuses become now possible for Statistics Sweden to produce. This unfolds opportunities to follow e.g. cyclical labor market phenomena on a granular level.

To obtain a comprehensive picture of the population's status of employment, it is necessary to develop methods which identify individuals' who are self-employed and classify their status of employment. Therefore, Statistics Sweden has developed two separate algorithms to identify: i) self-employed persons in sole proprietorships, trading or limited partnerships and ii) owners of limited companies. As far as we know, these algorithms are in an international context novel methods of identifying self-employed individuals on monthly basis.

Novel algorithms
The algorithm to determine self-employment is divided into several steps. In a first step, administrative data is obtained from the Swedish Tax Agency and the Swedish Companies Registration Office, which is delimited at the individual and company level with the help of the Register of Total Population (RTB) and the Business Register (FDB). Based on this set of data an algorithm determines who are self-employed.

The algorithm for identifying the owners of limited companies is more straight forward. It uses the information of which individuals are in the register of beneficial ownership at the Swedish Companies Registration Office and delimited at the individual and company level with the help of the Register of Total Population (RTB) and the Business Register (FDB).

Result
The results from these algorithms satisfy users with adequate statistics on the development of self-employed on monthly basis. When producing this new statistics, monthly development of self-employed, in its different status of employment, can hence follow the current development with a lag of two months.