Swedish National Audit Office (RiR)
1 More knowledge of the risks is necessary
Deficiencies in design of quality work

Many different types of cases are dealt with in the population registration organisation and the work is carried out in sections that are spread out over the country. If quality work is to be effective there must be well-functioning governance and appropriate allocation of resources. The Swedish NAO notes that there have been difficulties in the governance of population registration, but that the reorganisation implemented in the Swedish Tax Agency in summer 2017 creates conditions for simpler and clearer governance.

As regards resources, the Swedish NAO notes that the dimensioning of quality work as a whole and allocation of resources as part of that work is governed by other factors than identified risks and the opportunities the Swedish Tax Agency has to manage these risks. This is partly because the handling of incoming cases, such as change of address forms and immigration, is given higher priority than the controls and partly because of the risk analysis problems mentioned above. The Swedish NAO considers that more knowledge of the risks is necessary and that it must be possible to safeguard resources for quality work to ensure its mix of measures is entirely appropriate.

2 IT tools old and inadequate
Finding errors is more difficult

The audit finds that the IT tools in several parts of the population registration organisation are old and inadequate. When processing cases within the control system a processing tool is used that was developed when the Swedish Tax Agency took over population registration from the Church of Sweden in 1991. A more appropriate tool has been under development for the past ten years, but it has not as yet been fully implemented. This reduces the case officers' ability to carry out their work effectively.

Deficient IT tools also make the work of finding errors in the register difficult. For example, the population registration organisation does not have access to the sampling tools used in the tax operations, which severely restricts the possibility of producing appropriate control samples.

At present the largest source of information on errors in population registration are the notifications sent to the Swedish Tax Agency by private individuals, agencies and other organisations. In 2009 the Swedish Tax Agency was tasked by the Government to develop an IT service to enable other agencies to report the errors electronically. The Swedish Tax Agency has produced a specification for how reporting is to take place, but since the agencies themselves have to bear the development costs, only one agency has signed up. Good system support for prioritising between notifications received is also lacking.

3 No tools for investigating and dealing with some errors
Low share of error notification handled

An important factor for effective quality work is the existence of adequate tools for investigating and dealing with the errors found. The Swedish NAO notes, however, that even here there are deficiencies.

The Swedish Tax Agency currently handles about 40 per cent of the error notifications received by the Agency. The explanation for the low percentage is not poor accuracy but is mainly due to the fact that all notifications are investigated in the same time consuming way. Work is in progress to simplify the handling of some notifications, which the Swedish NAO sees as positive. The Swedish NAO considers that further improvement is needed in the work on notifications and that more knowledge is needed about the extent to which the Swedish Tax Agency's own control samples should be given priority over the notifications. It would not be appropriate to consider extending the notification obligation to more than the seven agencies that are currently subject to it until these developmental steps are taken. Such an extension has been proposed by both the Swedish Tax Agency and Lars-Erik Lövdén's inquiry.

The audit shows that there are several situations that are difficult to deal with. For example, there are some address-related errors that are difficult to investigate, but the Swedish NAO would like to draw attention in particular to the controls in connection with population registration for EEA immigration. Here there are difficulties both in checking ID documents and the employment certificates used to prove that a right of residence exists. It is important to have effective controls in connection with EEA immigration, since that type of immigration has been identified as a way of creating false identities that can then be used for criminal activities. As the situation is today, when a false identity has been entered in the population register it is also difficult to remove, even if the Tax Agency discovers it.

4 Low knowledge of the register's quality
Risk of insufficient reference data

The audit shows that there is knowledge of several different types of error that may exist in the population register. However, to a great extent there is a lack of knowledge of how common the errors are and of the consequences that arise in society due to these errors. This has meant that the risk analysis for population registration is of a general nature and does not provide sufficient reference data to enable well-balanced priorities between different types of error. In the present situation it cannot be guaranteed that controls are directed at the errors that should be given highest priority.

Moreover, the low level of knowledge about the quality of the population register means that there is no picture of the level of quality in the register, and in what direction it is progressing. Thus, this picture is lacking despite the fact that the Government wishes the Swedish Tax Agency to make an assessment of the quality of population registration. The consequence is that it is difficult to determine the extent to which the objectives concerning high quality in population registration are achieved and the size of the resources the service needs.


The aim of population registration is to meet national needs for basic population data. The data in the register is used for instance to show where the individual should be taxed as a basis for payments from the welfare system and social planning. Incorrect data in the population register may have a number of negative consequences for society. The Swedish Tax Agency's work to maintain quality in the register thus plays a very important role.

The data in the population register is continually updated and the quality work is associated with several challenges. One example is that the address data in the register is dependent on citizens' own reporting of changes, at the same time as problems in the housing market and a reduced need for physical mail may diminish the incentive to contribute to high quality levels in the population register. The increasing number of moves to and from Sweden also imply increased risk of errors in the population register. Apart from this, there are incentives to report in incorrect data to the population register to be able to commit various types of crime.


In view of the important role population registration plays in society and the challenge of maintaining quality in the register, the purpose of this audit is to investigate whether the Swedish Tax Agency conducts the work of maintaining quality in population registration effectively. In this context quality refers to correctness of the data in the register.

When designing the audit several delimitations were made. To start with, the definition above of quality means that processing time, consistent assessments and other questions related to processing cases are not included. The question of coordination numbers is not addressed in the audit, since the responsibility for ensuring that these numbers are correct is spread among several agencies. Nor are the management of protected identities and issuing identity cards included in the audit.

The items above were selected and named by the e-Government Subgroup of the EUROSAI IT Working Group on the basis of publicly available report of the author Supreme Audit Institutions (SAI). In the same way, the Subgroup prepared the analytical assumptions and headings. All readers are encouraged to consult the original texts by the author SAIs (linked).