Valuation rangeEver since I started following this bullboard some two years ago, I’ve been looking at as much data as I’ve been able to locate in order to determine what this Santo Tomas project is worth. My goal has been to inject some data analysis into this discussion forum. This is where I think that valuation is headed. We don’t yet have a pre-feasibility report that lays out all the project costs for the eventual mine at Santo Tomas, a detailed mine plan showing the sequence of ore processed, and a cash flow model from which the NPV can be calculated. We do however have many good data points from which a preliminary NPV range can be developed: Thornton’s historical resource estimate from Figures 14 and 15 in the Santo Tomas Technical Report (“ST Report”), industry production cost data, 3D-IP survey data which suggests a significant additional inferred resource, and recent assay results that confirm a good CuEq lift from Au and Mo. I used this data to calculate a range of NPVs based on multiple grade and mill size scenarios using $3.50 and $3.75 long-term Cu price. It spans $2.65 B - $5.39 B. To learn how I arrived at this estimate, continue reading. Thornton’s historical resource estimate placed the North and South zones at a combined 332.9 tonnes of ore with an average grade of 0.437 % Cu. Co-metals will very likely carry the CuEq to 0.50% or better. The 3D-IP surveys clearly identify high chargeability within the historical drill locations of measured Cu mineralization and it extends to adjacent areas. Thornton’s estimate understates the size of this resource. My post on April 22 last year estimated that Santo Tomas holds 2 B tonnes ore in the North and South zones, with the caveat of uncertainty as to the grades and percentage of viable ore that could ultimately be converted into measured and indicated resource above the cut-off grade. This estimate was based on analysis of the historical resource drilling data combined with the more recent 3D-IP chargeability survey data, and did not include any potential ore contribution from Brasilles - which is a wildcard that represents potential significant upside. A mill size of 120 kilo-tonnes per day (ktpd) can process a billion tonnes of ore over 25 years and can manage Thornton’s 332.9 tonne volume in 7-8 years. A 150 ktpd mill can process that in 6 years. I constructed scenarios where the known higher grade historical resource would be processed for the first 6 or 7 years, depending upon the mill size option, and then additional areas of lower grade would be processed in subsequent years. I went with a 30-year production plan because a 150 ktpd mill can process 1.6 B tonnes ore over that timeframe and would run for a few more years should the 2 B tonne ore estimate be fully realized. A 180 ktpd mill can process 2 B tonnes over 30 years, but I want to research this further because I don’t yet have enough data to estimate capital costs for this size mill, and I recognize that a larger mill would be desirable to process this volume of ore in 20 – 25 years. For those of you unfamiliar with how NPV is calculated, a good example that I located is from Table 22-5 in the Panora Minerals’ Preliminary Economic Assessment (PEA) report for the Cotabambas project. It demonstrates how annual cash flow and costs are captured over a 20-year project life and rolled up into a NPV. The “extracted metal value” revenue stream is shown to be higher in the first four years of production on the order of $770 M - $879 M. This decreases over time to $280 M in three of the last four years of the project, and then to $229 M in the final year. This example shows how a mine plan prioritizes higher grades in the early years in order to maximize the effect of cash flow on NPV. The first two years of the project are the construction phase, during which there is no revenue. Revenue starts in year 3. I assumed CAPEX of $2B and $3B for the 120 ktpd and 150 ktpd mill options respectively. These figures are in line with the CAPEX estimates from the other project comparables reviewed in my January 10 and February 6 posts. Only Josemaria had a proposed mill size of 150 ktpd and its CAPEX was a cool $3 B. I ran a series of hypothetical 30-year NPV calculations while adjusting the following three key input variables: mill size, production cost by grade, and average grade. To keep the model simple, I assumed a higher constant uniform grade for the first 6 – 7 years of mine life followed by a stepped down lower grade for the remaining years. In reality, the mine plan will have more variability in grades over the life of mine, and perhaps even superior grades in just the first few years. It’s premature at this point to predict how that mine plan will play out. I went with 7 years for the 120 ktpd mill option and 6 years for the 150 ktpd mill option, as each would process Thornton’s North and South Zone higher grade estimates over those respective timelines. For the remaining period out to year 30, I estimated a lower grade supply from other areas. I used a production cost of $0.99 per pound for an average grade of 0.50% CuEq, derived from the production cost comparison in my February 6 post (eight project comparison), and scaled cost to the other grade scenarios. I took into account a Cu mineral recovery rate of 90%. Not all metal in the ore is recovered by the mill and sold in the concentrate. Some amount is lost in the various flotation, gravity circuits, and other areas as waste. The average industry Cu recovery rate from the dataset of eight comparable projects cited in my February 6 post was 87%. The preliminary test cited on page 10 of the ST Report suggested 95%. We won’t know for sure until after comprehensive metallurgical tests are carried out over a statistically valid sample set from the different zones. So for now, I am selecting 90% as a midpoint between the two. Mill and grade scenarios: 120 ktpd mill 0.55% for 7 years and 0.40% for 23 years 0.55% for 7 years and 0.35% for 23 years 0.55% for 7 years and 0.30% for 23 years 0.50% for 7 years and 0.40% for 23 years 0.50% for 7 years and 0.35% for 23 years 0.50% for 7 years and 0.30% for 23 years 150 ktpd mill 0.55% for 6 years and 0.40% for 24 years 0.55% for 6 years and 0.35% for 24 years 0.55% for 6 years and 0.30% for 24 years 0.50% for 6 years and 0.40% for 24 years 0.50% for 6 years and 0.35% for 24 years 0.50% for 6 years and 0.30% for 24 years * These grade scenarios will change as additional exploration data becomes available. * CAPEX: $2B for 120 ktpd option / $3B for 150 ktpd Price of Cu: $3.50 / $3.75 Production cost was scaled depending upon the grade in order to better approximate the life-of-mine average grade and average cost per pound. I used these values: 0.55% Cu - $0.90 / pound 0.50% Cu - $0.99 / pound 0.40% Cu - $1.2375 / pound 0.35% Cu - $1.4143 / pound 0.30% Cu - $1.65 / pound NPV is based on after tax annual cash flow over a number of years, less CAPEX, using a discount rate. Most projects apply an 8% discount rate, which is what I used in these calculations. To calculate annual after tax cash flow for a given year: (Daily mill rate)*(grade)*(Days per year) = Pounds CuEx produced - (mill loss assumed 10%) = Annual amount CuEx recovered (Annual amount CuEx recovered)*(price per pound Cu) = Annual revenue (Annual revenue) - Production costs - Annual depreciation = Gross profit - Royalty to Mexican government (7.5%) - Environmental tax (0.5%) = Net Profit - Taxes (30%) + Annual depreciation = Annual after-tax cash flow I used a 20-year depreciation schedule for both the $2B and $3B CAPEX mill size options, which is appropriate for buildings and infrastructure in Mexico. I assumed the full CAPEX investment in year 1. Tax benefit from depreciation drops off after 20 years and that impacts annual after-tax cash flow in operating production years thereafter. A few caveats are worth highlighting. We neither know the Brasilles grade contribution nor do we have a volume estimate from across the river. That has the potential to skew the NPV calculation in a positive trajectory. Thornton estimated a modest ore volume for the South Zone and assumed that the majority was waste, which we can now see from the 3D-IP surveys grossly understates the potential volume. Expansion of the North and South zone resources could easily add additional volume to the 0.50% or better grades, in which case, the initial 6-7 year higher grade period in the NPV calculation could extend for additional years. This would boost cash flow in the early years and the NPV as well. The future price of copper is also a wild card in our favor. I initially went with long-term Cu at $3.50 / pound, and that may be a bit conservative. So I included a second table to demonstrate how just a $0.25 increase in the price of Cu adds another $500 - 800 million to NPV. Table 1 – NPV range of estimates for long-term Cu at $3.50 / pound | Grades | 120 ktpd mill | 150 ktpd mill |
| 0.55% & 0.40% | $3.92 B | $4.55 B |
| 0.55% & 0.35% | $3.12 B | $3.80 B |
| 0.55% & 0.30% | $2.82 B | $3.42 B |
| 0.50% & 0.40% | $3.74 B | $4.03 B |
| 0.50% & 0.35% | $2.95 B | $3.27 B |
| 0.50% & 0.30% | $2.65 B | $2.89 B |
Table 2 – NPV range of estimates for long-term Cu at $3.75 / pound
| Grades | 120 ktpd mill | 150 ktpd mill |
| 0.55% & 0.40% | $4.60 B | $5.39 B |
| 0.55% & 0.35% | $3.77 B | $4.59 B |
| 0.55% & 0.30% | $3.43 B | $4.15 B |
| 0.50% & 0.40% | $4.39 B | $4.83 B |
| 0.50% & 0.35% | $3.56 B | $4.02 B |
| 0.50% & 0.30% | $3.22 B | $3.59 B |
* I anticipate that these estimates will change as additional exploration data becomes available.*
The recent data releases and the availability of industry cost data is informative in approximating the value of this porphyry deposit at a time when significant copper porphyry deposits are hard to come by. As investors, we are in a rather enviable spot with the available historical data from other open-pit copper mines and can see that there’s considerable upside from the current market cap.
Let me know what you think. Am I missing anything? Do you have other relevant industry cost data to share? What long-term Cu price do you think is appropriate for calculating a NPV? Do you need any help locating specific data points referenced?