AES - Case Analysis & Write up
How would you evaluate the capital budgeting method used historically by AES? What is good and bad about it? Previously, the economics of a given project were evaluated at an equity discount rate for the dividends from any project and as it was mostly financed through local debt, which was non-recourse to the parent company, AES use to evaluate the dividend cash flows at a standard 12%. It is a simple approach with portions of sound reasoning. One could argue that all of these are being judged on a level playing field with a standardized discount rate, not to mention it saves time and effort. However, this methodology obviously fails to consider the factors of currency risk as seen in Argentina & Venezuela, regulatory risk, commodity risk and other risks involving the assumption that they will be able to receive the dividends fully denominated in US Dollars without any loss in value.
Does this make sense as a way to do capital budgeting?
The following reasons make Venerus’ model superior to the traditional model of capital budgeting used by AES. 1.
Venerus has used 3 main parts to measuring his cost of capital in the new methodology. Broadly they are, the overall risk (systematic), the country specific risk and the project /(cash-flow) specific risk. This new way of making capital budgeting decisions is not only more comprehensive but also more flexible. For example, if an additional rating or factor has to be added to the business-specific risk score, it can be done with ease. In this case, I have assumed the weights for the different categories for risk (counterparty credit, currency, regulatory etc.) are the same across countries. It is essential to understand that depending on the economy that AES operates in, these would have to be altered before calculating the risk adjustment. 2.
This method also pushes the cost of capital in several countries much above the previously used 12%. From a case and logical perspective, several of the countries where the new model has yielded high discount rates are known to have intensified level of currency or regulatory risks. 3.
In several of these countries the lack or minimal regulation of efficient markets causes the correlation between the local indices and American markets to be low, causing beta to be abnormally low. However, because Venerus’ model uses several additional factors to calculate the discount rate, it minimizes the effect of this anomaly. Nevertheless, there are a few limitations that the model is yet to accommodate. 1.
Despite the risks being considered, the probability of enforcement of laws differs between each legal system and country. There is no measure to incorporate that into this. 2.
Even if the cash flow is generated from the project as expected, the expropriate rate and the risk associated with it should be added to accurately value projects. 3.
Several variables used in the model are stationary and are not forward looking. This makes the forecasting technique equally unpredictable. 4.
It doesn’t have the provision to analyze real-events that might occur in the future like war or geo-political tensions. 5.
The currently used sovereign spread reflects only a debt-risk spread and assumes the same can be used for equity. Also, the sovereign risk cannot be measured in economies without a public debt market.
If Venerus implements the suggested methodology, what would be the range of discount rates that AES would use around the world? If the new methodology is implemented the projects would be discounted using a large range of discount rates as low as 10% to as high as over 30%, a total range of over 20 percentage points or 2000 bps. This obviously suggests that the variables used to measure various risk premiums across countries are different and tailored. Below is the spreadsheet with the various adjusted WACCs calculated for different projects across 15 countries....
Please join StudyMode to read the full document