The financial model is built on our real-life knowledge of how things happen in the industry in Romania. This approach has its pros and cons. Of course, the pan-European reality is different. However, we think that Romanian case is “one of the worst” in the entire EU, so the reality check will only provide us with upsides. For example, Romania has low electricity prices and quite unstable RES support government policies. But! We can still make the fund fly, even in the theoretical extreme case that its assets all remain in Romania.
Overall, we figured that if we had our projection mounted over various third-party reports for the entirety of Europe, the quality of the document would be worse.
The model goes like this:
It starts with calculating the future cash flow of the wind farm we currently run. We use it to calculate nominal “for 1MW of nameplate power” figures so we can extrapolate the cash flows of the fund running many farms. In these calculations, there isn’t much one can question. Expenses are proven with actual documents, being not much different from industry averages across Europe. The only peculiarity is a norm for many businesses: one part of expenses heavily depends on the age of the infrastructure, while another part almost doesn’t.
* * *
The next step is modelling the “role-model” farm—the virtual entity to model typical 1MW of nominal power in a community of generators where various age turbines and even turbines of different technology generations are mixed.
To account for such variety, we assume that at any given point in time the fund consists of two parts: modern turbines and super-modern turbines. The latter class has the tendency to decrease costs. The logic is described in more detail here. Due to severe discounting factor, this technology factor is comparatively weak.
The model of the nominal farm is on a 20-year long horizon because that is most probably the shortest remaining lifetime of turbines we will buy. Such time horizon makes our estimation neutral or even pessimistic.
The next step is to model the core transactions in the fund during the next 5 years (typical business planning horizon). We do it based on the quarterly cycle. Two assumptions should be noted:
(1) We start with the farm that now belongs to us and that the fund will buy with 11% discount to the fair price (more on that discount below).
(2) To forecast the cash flow, we use the average of 11 virtual turbines of different ages with 1-year increments. We have taken 11 years instead of 20 and did NOT start from the youngest ones to avoid inputting extreme cases into the mean. To understand the fund’s logic, we don’t need to simulate the growing group of turbines and watch their ages separately, quarter by quarter. We consider the admitted simplification both financially legit and physically meaningful.
* * *
The final step is to conduct the sensitivity analysis and come up with the sustainable set of parameters.
- WACC is assumed 18% and this fact simply represents our strong “business philosophy” standpoint.
- The 7% fund administration fee includes ALL expenses, not only the management fee.
- Starting annual investments directly affects only the fund NAV. Common sense suggests that the scaling factor should be tethered to the size of the fund: the bigger the NAV, the more the discount we can buy new farms at and lower the costs. However, we decided to leave the two factors separate so manual tests of the model remain simpler. We believe the separate approach to costs and discounts estimation avoids any “hidden” qualities of the model.
- The electricity price is just what we can contractually expect, based on current fixed volume/price arrangements. It should be noted, this price includes the quite large grid imbalance part. In terms of a greater-region electricity grid, Romania is poorly balanced; there are not enough large hydro or nuclear plants to level the naturally synchronous wind energy supply oscillations. Effective price can deviate 25% or so, based on wind weather. Importantly, this deviation is exactly what makes the fund business attractive. We have reasons to believe the same class of causes opens up the opportunity to buy assets at great discounts.
- Annual growth of electricity price partly reflects the inflation expectation (RON/EUR) and partly the price growth itself.
Initial investments growth rate simply implies some non-zero leverage in the marketing reach.
Discount to fair price, with which we can buy additional generators, is probably the most important parameter in the model so it is set to the very MINIMUM.