Managing Assumptions During Financial Modelling

Making assumptions is an integral part of every financial calculation. It is a known fact that if the assumptions are modified even slightly, the numbers on the model tend to change dramatically. The problem is that financial modeler is forced to make several assumptions while creating the model. When several of these assumptions are being made, it is important to create a mechanism which allows these assumptions to be managed in a coherent and easy to understand manner.

Financial modelers often do not pay attention to the management of assumptions. Since it does not involve any calculations, this is often thought to be an administrative task and is often delegated to the newest member of the team. However, the reality is that managing the assumptions is probably the most crucial task in the financial modeling process. In this article, we will explain why this task is important and also explain the mechanism which is used to manage this process.

Why is it Important to Manage Assumptions during the Financial Modelling Process?

There are several assumptions which have to be made during the process of financial modeling. If these assumptions are not properly documented, then they will remain in the mind of the modeler. As a result, people using the model and interpreting its results will have no idea where the results came from.

Several things can go wrong while documenting the assumptions related to a financial modeling project.

• The assumptions may not be available in a written and easily comprehensible format

• The assumptions may be scattered, and hence, the user of the financial model may not be able to find all the relevant information at the same time.

Best Practices for Managing These Assumptions

• All assumptions related to a financial modeling project should be stored and tracked centrally. An assumptions database should be created for this purpose.

• The assumption database should be a part of the model itself. If the model and database are two different files, the user may not download the assumption database, and hence, the entire purpose may be defeated. Also, the assumptions file may be lost or may not be updated as time progresses.

• Also, financial models are often created in several different versions. Each version is an improvement over the previous version. In many cases, the underlying assumptions in the model also change. If the assumptions are attached to the model itself, the user need not be trained about the assumptions that have been changed. Instead, they can simply refer the document which explains the basis for various calculations which are taking place in the model.

• It is important to ensure that the assumption document is also version controlled. This means that the initial set of assumptions should be given a number. Then when changes are made in the assumptions, the number should be changed to reflect a version change. The correct assumption document should then be mapped to the financial template. This will create a system wherein the assumptions of the model will be self-explanatory. This also makes it possible to quickly revert back to an earlier set of assumptions if the newer ones are found to be unrealistic.

Procedure for Creating an Assumptions Database

• There are several users who may be involved in using the financial model on a day to day basis. Each of these users may have a different opinion about what should be the assumptions which form the basis of the financial model. The financial modeler needs to first identify important stakeholders whose opinions should actually be considered during the financial modeling process. These stakeholders should then be consulted to finalize a list of assumptions.

• The financial modelers also need to include an entry date and due date in the assumptions database. This will help understand the date at which the assumptions were made and the information which the company had on hand on that given date. For instance, a lot of assumptions will be made during the initial few days. However, a lot of them may overlap with other assumptions and may hence be discarded.

• The final set of assumptions should be given some sort of numerical ID. This ID must be updated as per the version control norms mentioned above. Also, this ID must be attached to the specific worksheets where they affect any calculation. As a result, whenever a user views any calculation, they can simply refer to the assumption ID and cross-reference it with the database to find out the exact nature of assumptions which have been made.

• The financial modeler also has to make a decision regarding the level of detail which needs to be included in the assumption database. For instance, if the modeler assumes a 5% discount rate, he/she should explain how this rate was derived. The end-user may or may not agree with the opinion of the financial modeler. However, they will at least understand their point of view and as a result, may be able to make adjustments to their own calculations if they don’t agree with the modeler.

The reality is that financial modeling is about predicting the future. Everyone’s beliefs about the future are bound to be different. These beliefs are presented to the end-user using the assumptions database. If the user does not agree with the assumptions, they can change the calculation themselves.