The Swedish Tax Agency’s measures to tackle undeclared work

The audit shows that the Swedish Tax Agency has reasonable knowledge of the extent of undeclared work, the environments in which it occurs and of the methods used. The Agency has a well-designed model for the identification and analysis of risks, and a clear logic for how the risk assessment is to be implemented as effective measures. However, in some respects the model is not applied as intended.

The Swedish Tax Agency’s risk analyses compile and largely evaluate information about which the operations are already aware. The lack of usable external information means that risk analyses add limited value to the planning and prioritisation of measures to combat undeclared work. Rather, the previous year’s resource allocation and other aspects are greater governing factors. The weak link between risk analysis and measures is particularly manifest in the population of foreign companies, among which the Swedish Tax Agency has given measures against undeclared work lower priority despite a high risk rating.

The Swedish Tax Agency primarily counteracts undeclared work through risk-based checks or tax audits on companies. These are based on samples to identify areas with an elevated risk of error.

The audit shows that the quality of the Swedish Tax Agency’s samples for controls is sound, as the samples lead to checks that often result in adjusted tax or registration. However, there is a lot to indicate that the adjustments largely relate to matters other than undeclared work; in other words, the Swedish Tax Agency has addressed errors other than those for which the checks were aimed.

The current tools and powers of the Swedish Tax Agency are not adequate to effectively address complex and systematic forms of undeclared work. The Agency is therefore taking stock of the changes needed to bolster controls. It has also chosen to place greater emphasis on preventive controls to counteract arrangements that may enable undeclared work. However, there is no follow-up that could demonstrate the success of such controls.

The audit shows that reforms to combat undeclared work have so far not had the intended effect. The possibility of identifying undeclared work through checks on staff ledgers is limited. The requirement for employers to submit data at the individual level in the employers’ monthly tax return has not bolstered control activities either, and intra-agency cooperation to prevent crime in the workplace has been limited as a result of confidentiality barriers that were removed only in the summer of 2024.

The audit points to a number of shortcomings in the monitoring of the Swedish Tax Agency’s measures to tackle undeclared work. The reason is that current monitoring systems and procedures have not been designed to monitor specific risk areas, such as undeclared work. For example, it would be desirable to be able to measure the extent to which changes in employer contributions could be linked to undeclared work, and the extent to which demonstrated undeclared labour at a company has led to an increase in income tax for the natural persons concerned. There is also a need for knowledge about the impact of preventive measures against undeclared work.

The Swedish Tax Agency attaches great importance to risks that are harmful to society. While this may seem reasonable, an excessively narrow focus on serious errors can create control-free zones that can be exploited by unscrupulous actors. Also, because of the lack of follow-up, the Agency’s operations do not receive the continuous feedback that is essential to an effective risk-based approach. This poses a risk of the adjustment to altered risks of undeclared work not taking place fast enough, and of the Swedish Tax Agency missing relevant risk areas. Moreover, the lack of information on the results of the measures makes it difficult for the Swedish Tax Agency and the Government to determine whether resource allocation is balanced and whether tools and powers are adequate.