Financial Modelling: A model-based approach to risk assessment
During planning and financing stage of a renewable energy project, it is required to determine projected revenues, costs and returns as well as quantify risks.
InputThe financial model of the project needs to take into account the fundamental assumptions about the project such as plant parameters (engineering / design), costs, tariffs, interest rates, production levels. In addition, where possible, risks to cash flows should be parameterized. For instance, construction risk may be reflected by a combination of cost overrun and delay in commissioning.
OutputThere could be many outputs. A basic model will provide forecasts of cash flows before financing in monthly, quarterly or annual intervals. More sophisticated models may take financing or tax options into consideration and provide pro-forma financial statements for future years. On the basis of those forecasts, ratios such as gross margin, interest cover or return on investment may be calculated.
In the base case, reasonable assumptions are made for all significant parameters, and agreed with all stakeholders. The probability of reaching, at least, base case-level is 50:50.
A sensitivity analysis measures changes in key financial outputs (e.g. IRR or interest cover) due to variations in significant input drivers.
It answer questions like: Which drivers are the outputs most sensitive to? If there is a price increase in X, what impact does it have on revenues?
This analysis is very Insightful and should precede any further risk assessment. However, it does not take into account variation of several parameters at the same time or likelihoods of changes occurring.
Apart from the base case, other scenarios may be created by varying multiple input parameters.
Typically, this should include a "worst case" and a "best case".The "worst case" is often used by investors to test the potential downside, whilst the "best case" reflects the entrepreneur's dream. Rather than constructing a "best case", it may be more instructive to work out a scenario ("target case") under which a target revenue is reached. I.e. this answers the question: "What do we have to do in order to have $X revenue by Year-5?"
Using the Monte Carlo simulation technique, multiple parameters can be varied at the same time, depending on certain assumptions of likely ranges and variations of those input parameters.
The analysis provides expected averages and distributions of each output parameter. As an example, the output could be: The expected average irr = 5%, but there is a 5% chance that it will be 12%.
Whilst incredibly powerful, due to its complexity, the results can lack in transparency. The fewer parameters the better!