APV valuation in renewable energy is not without skeletons.
Geshie Naidoo, 20 November 2021
If you are a renewable energy investor, then consider this one of the most fundamental articles you will ever read.
The APV (Adjusted Present Value) method is not without closet skeletons. Academic literature is riddled with ambiguity on what discount rate to use when discounting for tax shields. A combination of sources point to at least 5 allowable rates that can be used in practice: risk free rate, unlevered cost of capital, cost of debt, appropriate level of tax shield risk, and the cost of tax for the government. In the latter option, the tax shields can also be calculated by taking the difference of the discounted taxes for the unlevered project, and the identical levered project.
It is therefore logical that applying either of these could result in as many as 5 different values of project cash flows. And all with a sound theoretical basis. Bringing this argument back home, there is a risk to investors that a loophole in the current APV method thus allows for project valuations to be inflated for vested interest, leaving each investor with a shorter end of the same stick.
The purpose of project valuation is to value the market value of equity. The levered project value at the start of each period less the market value of debt at the start of each period. The market value of equity is then the value at the beginning of each year, for the residual life of a project.
***********************************************
V(L,t) = D(t) + E(t) = V(U,t) + PVTS(L,t)
Therefore, E(t) = V(U,t) + PVTS(L,t) - D(t)
***********************************************
The APV method is commonly used by investment banks to value highly geared renewable energy projects. It is a preferred method in highly leveraged projects, where pre-determined debt is often between 60-75% of a project's total funding structure. The present value of tax shields within the APV method is a clear way of showing how valuation can fluctuate with increased financial risk for equity investors. Higher tax shields, higher financial risk, and the higher the required rate of return on equity. Tax shields is the smaller component of the APV composition, and often overlooked with dire consequences for investors during negotiation.
To demonstrate my point above, I have created a very short case study below.
*************
Short case study: Serious implications for investors
Situation of the case
Assume you are the Chief Investment Officer of a large bank, DFI, pension fund, or private equity firm. You are considering an investment in a renewable energy project. The promoter is a recent BW5 REIPPPP winner, with a 20 year BOT PPA agreement guaranteed by the National Treasury. The promoter is hosting an investor roadshow looking to sell equity to other prospective investors. He makes a presentation of the valuation for a new solar power IPP project to you.
The project has a 20 year life, and issues startup debt with an opening balance at the start of year 1 of ZAR150m. The debt issuance earns a blended IRR return of 800 basis points for the lenders. And the unlevered cost of capital for the existing equity holders is 1,200 basis points. Interest payments are paid annually, while the principal debt will be repaid via a bullet at the end of year 20. The operation of the power plant commences at the beginning of year 1. EPC takes place from start to end of year 0. O&M will commence from start of year 1 to end of year 20.
Glaring question you need to ask the project promoter
What discount rate was used to value the present value of the tax shields?
Why is this question so important
The higher the discount rate used, the lower the value of the tax shields, and vice versa. We also know that the discount rate used can vary between the risk free rate and the unlevered cost of capital (which is also the unlevered required rate of return on equity for the current investors). According to this example, the mispricing exposure could be a range of 800-1,200 basis points.
So why all the fuss
According to our model, at a discount rate of 800 basis points, a 20 year valuation of the tax shield component using the APV valuation method equates to ZAR47,1m at the start of year 1. However, if instead a discount rate of 1,200 basis points was used, then this 20 year valuation drops to ZAR35,8m at the start of year 1. This is a discount spread of ZAR11,3m (ZAR47,1m - ZAR35,8m) on the tax shield component of the APV method alone.
Source: All calculations performed using Ener Re Energy Finance model.
Conclusion of the case
Agency theory implies that there is moral hazard in relying solely upon the information from the project promoter in the transaction. In this case, the underlying financial model assumptions.
In this case, assuming you were considering a 50% investment at the beginning of year 1, then you could potentially be paying a premium of up to ZAR5,7m (50% times ZAR11,3m) for your investment. This amount rises to ZAR7,9m (80% times ZAR11,3m) if you plan to buy 80% of the project. These values would exclude any inefficiencies priced into the unlevered FCF valuation, as well as any other premium which the seller may be front-loading onto the final market value of the equity.
Hence the discount rate used to value tax shields is not as trivial as some industry practitioners will have you believe, especially if you're the prospective investor on the other end to this transaction. Remember, you have a fiduciary duty of care to buy high quality assets with matching asset-liability profiles. You must create deep and lasting value for your retirees and shareholders.
**************
The APV method relies on the derivation of unlevered FCF to determine the unlevered value component of the APV method. Another pitfall of relying solely on the APV method, is that the valuation practitioner does not have a reference point upon which to reconcile and validate the value obtained. Not without meticulously tracing back each step to ensure that all assumptions were consistently applied in the model, and that there are no underlying bugs in the programming algorithm.
Ener Re uses an approach to create multiple checks and balances in project valuation. This is achieved through discounting 4 types of cash flows - unlevered FCF, CFE, CFC, and the levered Project Value (obtained using APV) - using a combination of 5 discount rates - unlevered cost of capital, cost of debt, MMII (Miller and Modigliani's 'proposition II') adjusted required rate of return, WACC before taxes, and WACC after taxes. Each method must produce the same valuation result.
If the discount rate is altered to recalculate the present value of tax shields, a different value is obtained for the project value. However, all 4 methods of DCF above ensures that once again a consistent value is reached. So what does all this mean?
Firstly, using different DCFs allow the financial modeler to triangulate the same value using different rates and different cash flows. It provides ample assurance in the value calculated as the project value. And secondly, using a different discount rate for valuing tax shields will result in different valuations (higher rate, lower value, and vice versa).
In summary, using this approach will produce 2 different but theoretically accepted values for a project. Each approach is corroborated using 4 different DCF methods. That's 8 different outcomes, with 2 unique values. That's if your model is correct, and someday is required to defend itself in a court of law.
Our approach is what makes Ener Re unique in the project finance market. Let us be part of your next renewable energy investment.
Ener Re is a niche financial engineering brand, specialising in unlocking value for renewable energy investors. We help large investors and project promoters find common ground during price negotiations on large projects. If you enjoyed reading this article, please send me your thoughts on geshie@ener-re.com