OK, I took Affenhorsts valuation model and updated/changed it. I turned it into a discounted cash flow model (at least I believe I did, not an expert) and changed a lot of the assumptions. A short explanation to both show how I understood it so people can find errors as well as to teach people who dont know what that model is.
Basically it means that I take the expected profit from next year and apply a fraction to the valuation of the company to this year. The discounted cash flor is a way to calculate the "fair" market cap of a company.
Think of the company like a loan (wikipedia for sure has a better description). Say its worth X$ today but you dont know X and next year, you expect to get your money back and some interest. For the company the interest is the divident it gives on its profit. For the sake of argument, lets say the company issues all its profits as divident. So say the company would make 100$ profit, so you would expect to get back 100$. Now the question is, how much would you be willing to pay in order to get 100$ next year. This is where the treasury yealds come in. They sit right around 5% at moment. And these yealds are US treasury or some other major country. These are practically risk free. So you know if you pay 95$ today in bonds, you would get 100$ next year risk free. Obviously a company is more risky than bonds, so say you are not paying 95$ for a company with profit 100$ but say. Say the company has a risk of 15% of defaulting and you have a 85% chance of getting your money back and then some. So you would say.. ok, for this kind of company, if I value it at 85$, I have a 15% chance of losing it all and a 85% chance of gaining 15$. Sounds balanced, however you have to compare it to the risk free bonds. So its more like: I am willing to pay 80$ to gain 100$ becasue I subtract the bond yield as well.
So the chashflow discount rate is 20% in this case. But if your business is much more risky, say.. you have a chance of 50% of losing it all, you might say.. ok, in this case, I would only pay 45$ to get 100$ or lose it all. And you see how this ties in to the treasury yealds as the higher the yields are, the less you are willing to invest in a company. And since yields depend on inflation, high inflation drives yields up, which in turn drives company values down. Exactly what we observe in the current inflation cycle.
Back to our company. Say, I would give it a 20% dicount rate. Meaning, I would pay 80$ for a 100$ return next year. But I also expect the company to have aprofit the year after. So I say, the 100$ the year after is not worth the same as 100$ next year, but say.. only 80$, with 20% risk profile we already established. So I would actually pay $100 * (1-0.2) + $100 *(1-0.2)^2 today to get $100 each for the next 2 years. So in fact, I would pay $144 today to get $100 next year AND $100 the year after. And of course, I dont do that for only 2 years but for many years to come. And thats how you find the net present value of the company today.
For the Talga model, I simply take the net present value as the fair, risk adjusted market valuation. Indeed, I used the 20% discount rate in the example above, even though that is dramaticaly pessimistic. However, I do that because I think the market is not much more optimistic about it. Take it as.. say my expectation what other people will think Talga is worth. This also means I need to model the profits each year. And I start with much more pessimistic assumptions. I think Talnode-C is going to be sold for $8000 per t, not 12000 as the DFS suggests. And I am thinking Talnode-Si will be just 10% in volume and has the same profit margin. of 40% instead of 80% as in the DFS. So we are looking at a profit after taxes of about $2500/t, which seems realistic to me.
As for timelines.. I expect production to start in 2027 at 10t/year and full capacity 2028. Then, I expect the expansion to kick in 2031 and ramp up after that in a couple of years. From that I calculate the profit, which is then discounted back to have the net present value. And the valuation seems good and reasonable to me! Of course, if you change the discount rate to something reasonable like 10% and increase the profits on Talnode-C.. the valuation becomes totally nutty, but at moment, I am quite happy with this rather pessimistic but probably more realistic approach.
View attachment 72257
Here is the document in xlsx format:
link to doc
