Proactive evaluation of social policies

SAS Provides a robust basis for reliable microsimulations

What will the effect be of changing the pension levels? What would it cost citizens to upgrade certain minimum social incomes? What would be the impact on poverty of changing the social payments system? The Belgian Federal Public Service (FPS) for Social Security — the former Ministry of Social Affairs — is able to answer these questions proactively. Their MIcrosimulation Model for Belgian Social Insurance Systems (mimosis) evaluates the effectiveness and cost of any social policy. The speed and robustness of SAS allow for fast simulations on a large, representative set of data.


The SAS-based micro-simulation model evaluates social policies within fifteen minutes, enabling us to fine-tune each policy to an optimal cost/benefit ratio.

SAS provided the robust basis necessary to build a complex model that can proactively and reliably simulate the consequences of a social policy.

Koen Vleminckx
Head of research at the Directorate General (DG) Strategy, International Relations, and Research of the Belgian FPS for Social Security

From reactive to proactive policy evaluation

In collaboration with three academic teams and with the support of Federal Science, we have built a microsimulation software model. Mimosis is able to simulate certain social policy reforms beforehand,” proudly states Koen Vleminckx, Head of research at the Directorate General (DG) Strategy, International Relations, and Research of the Belgian FPS for Social Security. “Who gains? Who loses? What is the impact on the Social Security budget? Mimosis gives us reliable insight into the impact that a proposed social policy will have. SAS has played a key role in making mimosis possible.

Accurate and representative evaluation of social policy

Ensuring the reliability of mimosis proved challenging. First of all, the micro-simulation model needed a large enough dataset in order to generate an outcome that is representative for the entire Belgian population. “We draw our data from the Labor Market and Social Policy data warehouse, which is governed by the Crossroads Bank for Social Security. It unites administrative data from all of the social security agencies within Belgium. From this data warehouse, we extract a sample of 100,000 individuals. We complement this sample with data from all other household members for a total of more than 305,000 individuals,” notes Koen Vleminckx. “We chose SAS for managing and organizing this vast and rich dataset because of its proven track record in effectively handling large databases.

Secondly, the micro-simulation model needed to take into account the spillover effects of social policies. Koen Vleminckx observes that a social policy can have unanticipated effects in many different domains. For instance, adapting the pension level can have an impact on personal income taxes, the contributions which have to be paid on pensions, or the family charges for an unemployed person. The combination of SAS tools and FORTRAN source code enables the incorporation of these spillovers into mimosis.

Fast processing enables fine-tuning

Another major challenge was to enable the fine-tuning of social policies. “We are not only able to assess given policy reforms, we can also explore alternatives that either reduce the cost of a reform or improve its benefits,” explains Koen Vleminckx. “To make this possible, we needed to maximize the processing speed so that we could run various simulations without losing too much time.

mimosis is able to assess certain social policy reforms within fifteen minutes. “SAS has enabled us to gather, manage, and process huge amounts of data efficiently. This gives mimosis the necessary processing power to run simulations with a very large dataset in a very short time frame.

Improved budget management

The mimosis tool enables the FPS for Social Security to have greater control over its budget. Koen Vleminckx: “The former Federal Minister of Pensions asked us to evaluate one of the measures that the Federal Government had taken to deal with the ageing problem, the so-called Generation Pact. Instead of running a time-consuming and expensive survey, we evaluated the measure using mimosis. This was both costeffective and fast.

The speed of the micro-simulation model is appreciated by various government bodies. Koen Vleminckx illustrates with an example. “The Belgian Court of Audit asked us to evaluate the cost of a number of scenarios for upgrading certain minimum social incomes. We were able to offer them a full report within a week. They were amazed by such a quick reaction to their request.

Aiming for user-friendly reporting

The SAS-based micro-simulation model has already achieved major benefits for the FPS for Social Security. And it sees even more opportunities for the future. Currently, they are using eight-year old social security data. Therefore, they are investigating how SAS can assist them to update this dataset. “We are developing SAS programs that enable us to retrieve data from the Datawarehouse on a more regular basis, without the original programmer having to intervene every time,” says Koen Vleminckx. “In short, we have a window of opportunities to enhance mimos s even further. We are investigating how SAS can assist us in doing just that.


Federal Public Service Social Security


Build a micro-simulation model for social policy in Belgium


SAS® Analytics Pro


  • Improved social policies: SAS is able to manage a vast set of data and the relations between these data, which makes it possible to reliably determine the impact of each policy under consideration.
  • Effective budget management: thanks to the ability of SAS to efficiently handle very large volumes of data, mimosis is able to evaluate policies on short notice. The short run time makes it possible to fine-tune certain policy reforms.
The results illustrated in this article are specific to the particular situations, business models, data input, and computing environments described herein. Each SAS customer’s experience is unique based on business and technical variables and all statements must be considered non-typical. Actual savings, results, and performance characteristics will vary depending on individual customer configurations and conditions. SAS does not guarantee or represent that every customer will achieve similar results. The only warranties for SAS products and services are those that are set forth in the express warranty statements in the written agreement for such products and services. Nothing herein should be construed as constituting an additional warranty. Customers have shared their successes with SAS as part of an agreed-upon contractual exchange or project success summarization following a successful implementation of SAS software. Brand and product names are trademarks of their respective companies.

Back to Top