Superintendence of Industry and Commerce (SIC), Colombia - Integrated Sanction System 

Superintendence of Industry and Commerce (SIC), the Colombian consumer protection authority is in the finishing phases of development of a tool to aid in the application of administrative sanctions. The Sistema Integrado de Calculo y Application de Sanciones tool supports officers to calculate and record sanctions, a task which involves the consideration of a large number of variables.

To start, the relevant variables, including the size of the company at fault, its financial health and capacity to pay the fines, the type of wrongdoing and its severity, the impact on consumers (and the type of consumers who were victims of the wrongdoing), are documented. An investigating officer can also document other relevant factors such as: additional aggravating or mitigating factors, such as whether the agency is dealing with a repeat offender or a small business on a first offence. The tool also enables officers to weight the different factors included in the calculation. Once all the relevant fields and weightings are input, the tool calculates an appropriate level of fine. This calculation is not the final decision, instead, it is a suggested proportionate figure that can be used by an officer alongside their judgement as part of the process to set a fine.

As well as reducing discretionary margins and lessening the burden in decision-making, the tool will help ensure sanctions are more proportionate and thus make them easier to defend on appeal, adding legal certainty and reducing costs for the authority.

The more the tool is used, the more data will be captured and stored (in accessible excel cvc files). Such a database can then be a foundation for more functionality. In time the tool could be expanded to assess if the level of fines issued is effective as a dissuasive tool, through comparing the number of fines at particular levels with the amount of re-offending businesses.  As well as the analysis of past practices, the tool could be developed to make predictions for an appropriate, proportionate and dissuasive fine. At present the tool requires human input to populate fields and to make the final decision by applying discretion. This discretion, currently based on the experience of legal officers, could to some extent be ‘taught’ to a tool once it has enough data so that it can, not only make a more refined prognosis, but also make predictions.