Pan American Silver — Bigger scale, stronger profitability

With the recently completed Yamana transaction, PAAS continues its track record of value accretive acquisitions that add scale, in terms of both production and resources, and significantly improve profitability and cash flow generation. The company now operates four producing silver and seven gold projects in the Americas with combined FY22 pro-forma revenues of US$2.7bn and mine EBITDA of US$1.0bn. In addition, PAAS has a healthy pipeline of organic growth opportunities focused on silver, continues to advance its world-class Escobal silver project through the ILO 169 consultation and now holds a 56.25% stake in the large-scale, brownfield polymetallic MARA project in Argentina. On our revised estimates, the stock trades at an FY24e EV/EBITDA of just 5.8x, with the commodity prices remaining supportive (albeit volatile) and underpinned by the tighter monetary policy, increased geopolitical risks and growing industrial demand for silver.

The acquisition of the selected Yamana assets solidified PAAS’s position as the second-largest global primary silver producer, with pro forma FY22 output of c 28Moz. This compares to Polymetal at 21Moz and industry leader Fresnillo at 51Moz. While FY23 guidance of 21–23Moz of silver only takes into account nine months of production from the Yamana assets, adjusting FY22 numbers for the closure of Manantial Espejo (3.5Moz) and adding Escobal at c 20Moz suggests PAAS is capable of producing in excess of 40Moz of silver per year. Further, while we estimate that the share of gold in the overall output is set to increase to c 78% in FY23 (71% in FY22), putting Escobal back into production (FY25e) should lower gold’s contribution back below 70% (before the impact of the upcoming Dolores gold mine closure). As such, the company continues to maintain a strong exposure to silver, which we believe is beneficial against the backdrop of ongoing global decarbonisation.

The majority of the Yamana assets are low-cost operations and we expect them to visibly reduce PAAS’s consolidated costs, and improve profitability and cash flow generation. In FY22, the acquired Yamana assets generated US$676m in mine EBITDA at a margin of 56% compared to US$365m and 24% for PAAS. Based on FY23 guidance, we estimate that the Yamana projects lower the cash cost and all-in sustaining costs (AISC) by c 10% and 5% in the silver segment and 11% in the gold segment. A significant boost to production and improved profitability will also lead to a step change in the cash flow generation. Assuming no reduction in tax liability on the balance sheet, we expect PAAS’s net operating cash flow to increase from just US$32m in FY22 to US$494m in FY23, with a further improvement in FY24 thanks to the annualised contribution from the Yamana assets. At end-Q1, PAAS had net debt of US$674m; we expect it to reduce to US$344m in FY24 (including short-term investments).

We have reinstated our estimates and now expect PAAS to generate EBITDA of US$734m in FY23 and US$1.1bn in FY24. This puts the stock on an FY23e EV/EBITDA multiple of 8.6x, falling to just 5.8x in FY24e, as we expect the company to enjoy the full effect of the Yamana consolidation and gradual easing of inflationary pressures. As we noted previously, the primary silver producers appear to be trading at somewhat higher multiples compared to the gold universe; however, at the current point in the commodity cycle this valuation disparity is compensated by the higher margins generated in gold, with PAAS not being an exception. On our estimates, PAAS is currently valued at a c 15% average discount to both peer groups on FY24e EV/EBITDA. Our sum-of-the-parts valuation, which takes into account PAAS operations, Yamana assets and MARA, implies a value of US$22.5/share, a c 40% upside to the current share price. The key risks include the commodity price performance, consolidation of the Yamana assets as well as regional legislative changes and political uncertainty in Central and Latin America.

Despite some volatility, commodity prices remain supported by higher inflation and heightened geopolitical risks, as well as strong industrial demand for silver. The Silver Institute expects industrial demand to grow 4% in 2023 (+5% in 2022), driven by the anticipated 15% increase in photovoltaics (+28% in 2022), while the overall silver demand is seen falling by 6% (+18% in 2022). As a result, the share of photovoltaics (PV) in the overall industrial demand is forecast to reach 28% in 2023 versus 20% in 2020 and 12% in 2015. We expect PV to remain the key driver behind silver demand as the global green transition continues to pick up its pace. Our commodity price forecasts assume a silver price of US$23.7/oz in FY23 and US$24.3/oz in FY24, while our nominal gold price assumptions are US$1,960/oz and US$1,928/oz, respectively. We therefore model a gradual normalisation of the gold/silver ratio from 83x in FY23 to 80x in FY24 and further to 73x in the longer term.

On 31 March, PAAS completed the all-share acquisition of the selected Yamana assets, paying US$2.4bn at the time of the announcement. It continues the company’s track record of well-timed and value accretive deals following its purchase of Tahoe Resources in 2019. Based on FY22 numbers reported by Yamana, the acquisition implies an EV/EBITDA multiple of 7.0x at announcement (8.1x at completion) on a combined PAAS/Agnico basis. This compares to PAAS’s multiple at the time of the announcement of 10.5x. As with Tahoe Resources, the Yamana deal was announced against a backdrop of weaker gold prices, which have subsequently recovered from c US$1,650/oz levels seen in November 2022 to around US$2,000/oz in Q223 and US$1,980/oz spot. For more detailed analysis of the transaction, see our note published in November 2022.

As part of the transaction, PAAS acquired four producing gold and silver operations in Latin America and a 56.25% interest in the pre-production polymetallic MARA project in Argentina. The company’s silver segment is now comprised of four operations: La Colorada (Mexico), Huaron (Peru), Cerro Moro (Argentina; ex-Yamana, 100% ownership) and San Vicente (Bolivia). The gold segment consists of seven projects: Jacobina (Brazil; ex-Yamana, 100%), El Penon (Chile; ex-Yamana, 100%), Timmins (Canada), Shahuindo (Peru), La Arena (Peru), Minera Florida (Chile; ex-Yamana, 100%) and Dolores (Mexico). On a pro forma FY22 basis, c 73% of revenues would come from Latin America, 18% from Mexico and 9% from Canada, with any single jurisdiction not exceeding 23% of revenues.

In FY22 Yamana’s Latin American operations produced 9.2Moz of silver and 565koz of gold. On a pro forma basis this implies combined PAAS/Yamana production of 28Moz for silver and 1,091koz of gold, elevating PAAS to the position of the second-largest global primary silver producer, ahead of Polymetal (Exhibit 3). While the company’s FY23 guidance of 21–23Moz in silver output only takes into account a nine-month contribution from the Yamana assets, adjusting FY22 numbers for the closure of Manantial Espejo (3.5Moz) would translate into pro forma silver production of c 24Moz. Adding Escobal at c 20Moz pa of historical production suggests PAAS could produce in excess of 40Moz of silver per year. Since the majority of the acquired assets are predominantly gold operations, the share of gold in the production mix is inevitably set to increase. We estimate it to reach 78% in FY23 versus 71% in FY22 (Exhibit 2). However, with the anticipated recommissioning of the Escobal silver project (which we now model in FY25), the share of gold should fall back to below 70% on an Au equivalent basis. Taking into account the upcoming winding down of Dolores, the share of silver could normalise further over the medium term.

In addition to raising the production profile, the Yamana acquisition has also brought a significant increase in mineral reserves and resources as the majority of new mines are long-life operations, in particular Jacobina and El Penon. Unsurprisingly, the biggest increase has come in gold, with proven and probable (P&P) reserves rising by 132% to 8,412koz of contained Au, while the silver P&P reserves went up by 11% to 570.2Moz (Exhibit 4).

The Yamana assets are highly profitable and cash-generative operations that should significantly improve PAAS’s margins in both the silver and gold segments. In FY22, the four projects generated US$1.2bn in revenues and US$676m in mine EBITDA at an impressive margin of 56%. This compares to PAAS’s revenues of US$1.5bn and mine EBITDA of US$365m, a 24% margin. The key assets in the Yamana portfolio are Jacobina in Brazil and El Penon in Chile, which on a combined basis accounted for 62% of revenues and 69% of mine EBITDA. In our updated estimates for the combined company, we expect the mine EBITDA margin in the silver segment to improve from 23% in FY22 (PAAS standalone) to 28% in FY23. In the gold segment, we model an improvement from 25% to 48%.

With the exception of Minera Florida, the acquired Yamana operations are positioned towards the bottom of the cost curve. In the silver segment, Cerro Moro’s costs are comparable to those of Huaron, while in the gold segment, both Jacobina and El Penon are at the bottom of the guided cost ranges. While the historical cost numbers for Yamana and PAAS are not comparable due to different accounting approaches, we look at the combined FY23 production and cost guidance to assess the impact on costs from the addition of new mines.

Based on the mid-point of cost ranges for individual projects, the company guides FY23 silver segment cash cost and AISC of US$11.0/oz and US$15.1/oz on a combined basis. Stripping out Cerro Moro gives a WACC of US$12.3/oz and AISC of US$15.8/oz. This implies reductions of 10% and 5%, respectively. For comparison, in FY22 PAAS’s silver segment cash cost and AISC were US$12.7/oz and US$16.5/oz. The closure of Manantial Espejo is another factor that should result in lower silver segment costs, as the project reported a cash cost of US$19.7/oz in FY22.

Similarly, in the gold segment, the inclusion of three new mines is estimated to lower the combined cash cost and AISC by c 11% to US$1,033/oz (from US$1,159/oz) and US$1,355/oz (US$1,517/oz), respectively. This is due to Jacobina and El Penon being the lowest cost gold operations in the PAAS portfolio, with impressive mid-range cash cost guidance of US$780/oz and US$685/oz for FY23, while Minera Florida is expected to have costs broadly similar to the Timmins project at US$1,385/oz cash cost and US$1,775/oz AISC. In FY22, the gold segment reported cash cost and AISC of US$1,113/oz and US$1,649/oz.

The company has a pipeline of potentially large-scale silver developments that could bring a significant boost in production in the medium to long term.

Escobal is a world-class silver operation that was put on care and maintenance in 2017 under the previous owner Tahoe Resources. Following the acquisition of Tahoe by PAAS, the Constitutional Court of Guatemala determined that an ILO 169 consultation process with the Xinka indigenous people had to take place before the project’s mining licence could be reinstated. The process is managed by the Ministry of Energy and Mines of Guatemala. The consultation is the second phase of the process, followed by verification by the Supreme Court.

Historically, Escobal produced in excess of 20Moz of silver per annum at a cash cost net of by-product credits of about US$6/oz. The project has a long life underpinned by the P&P reserves of 264.5Moz of contained silver at a grade of 334g/t. Once the mining licence is reinstated, we understand the project can resume production within a short timeframe of about six months.

La Colorada Skarn is a 100% owned underground polymetallic deposit that sits under the currently producing La Colorada project in Mexico. As of September 2022, the project is estimated to have an indicated and inferred mineral resource of 227.3Moz of contained silver, plus a large volume of zinc and lead. PAAS is currently advancing the project through a preliminary economic assessment (PEA), expected to be complete in H223. As part of the study, the company has recently reported additional exploration and infill drill results that returned high-grade intersections at 391g/t, 233g/t and 98g/t of Ag with mineralisation exceeding 64m.

The MARA project was acquired as part of the Yamana transaction. It is a large-scale, open-pit, brownfield (currently under care and maintenance) polymetallic deposit in Argentina, in which PAAS owns a 56.25% stake (and is currently an operator), with the remainder controlled by Glencore. The project is in the feasibility study stage and has P&P mineral reserves of 11.8bn pounds of contained copper (at 0.49% Cu) and 7.4Moz of contained gold on a 100% basis. Based on the pre-feasibility study completed in October 2019, the project has an estimated mine life of 28 years, initial capex of US$2.4bn and a net present value of US$2.0bn at an 8% discount rate. In September 2022, Glencore announced the acquisition of an 18.75% interest in the project from Newmont, paying US$125m upon closing and US$30m in deferred consideration.

Other projects include Navidad, a large-scale advanced silver project in Argentina, and Morococha, a producing silver mine in Peru that was put on care and maintenance in 2022 following decommissioning of the processing plant.

PAAS delivered a strong set of Q123 results, despite visibly lower silver and gold production and sales. Production numbers were affected by decommissioning of the Manantial Espejo project in Argentina, which produced 1Moz of silver and 9koz of gold in Q422 and 0.9Moz silver and 6koz of gold in Q122 on a 100% ownership basis. In Q123, the project sold 0.6Moz of silver, a reduction of 8% versus Q422 and 28% versus Q122. Sequentially lower sales were partly offset by the higher realised silver and gold prices achieved in Q123. However, the main highlight of the results was an impressive cost performance, with unit cash costs falling 15% q-o-q in the silver segment and remaining broadly flat in the gold segment. At the same time, the gold segment AISC was 20% lower quarteron-quarter, with the silver segment AISC dropping 21%. At the project level, the biggest reduction in costs came from Huaron and San Vicente in the silver segment, which saw 39% and 37% q-o-q falls in unit cash costs to US$5.7/oz and US$10.8/oz, respectively. In the gold segment, costs were only marginally higher at all operations.

As a result, the company reported a 4% q-o-q increase in revenues to US$390m and a 27% q-o-q increase in EBITDA to US$117m in Q123. This is despite a significant increase in mine, care and maintenance costs and higher G&A costs. A 9% reduction in EBITDA versus Q122 was mainly a result of higher care and maintenance costs, which doubled year-on-year due to the inclusion of costs associated with the MARA project and the decommissioning of Morococha and Manantial Espejo. At the bottom line, net profit was US$16.4m, with Q422 performance distorted by one-offs related to the Yamana acquisition. Earlier the company recommended a quarterly dividend of US$0.10/share.

While the Q1 P&L and cash flows were reported on a standalone base, the company provided a consolidated balance sheet following the completion of the Yamana transaction on 31 March. On a combined basis, the company had net debt of US$674m, which included short-term investments of US$102m and long-term debt of US$1,058m. The gross cash position (including restricted cash at MARA) was US$411m.

With its Q123 results, the company also provided consolidated production and cost guidance for FY23, which takes into account PAAS’s existing operations and a nine-month contribution from the Yamana assets. On a combined basis, the company guides total silver production of 21–23Moz, with the existing operations contributing 14.3–15.3Moz and the Yamana projects adding another 6.7–7.7Moz. Total gold production is expected to reach 870–970koz on the same basis, with the acquired mines forecast to add 405–465koz of gold. In FY22, PAAS produced 18.5Moz of silver and 553koz of gold. As is typically the case, both gold and silver production are expected to be second-half weighted, with Q4 seen as the strongest quarter for both metals (Exhibit 9).

On the cost side, the silver segment cash cost and AISC for the full year are guided at US$10.0–12.0/oz and US$14.0–16.0/oz versus US$12.2/oz and US$14.1/oz in Q123. Costs are expected to be the lowest in Q3 on the back of the higher production and the addition of the low-cost Cerro Moro project, whose cash cost is forecast at only US$5.5–8.8/oz thanks to gold by-products. In the gold segment, full-year cash cost and AISC are guided at US$975–1,110/oz (Q1: US$1,120/oz; FY22: US$1,113/oz) and US$1,275–1,425/oz (Q1: US$1,196/oz; FY22: US$1,649/oz). The company expects gold segment costs to bottom out in Q4. Similarly to silver, the gold segment cost performance should benefit from the inclusion of the low-cost Yamana operations. In particular, Jacobina is guided to have a cash cost and AISC of only US$750–810/oz and US$1,020–1,110/oz, while El Penon’s costs are forecast to be US$600–770/oz and US$758–985/oz supported by the significant silver by-product credits.

The company expects to spend US$305–320m in sustaining capital on a consolidated basis and another US$95–105m in project capital in FY23. At the project level, US$26–29m will be spent at Jacobina to stabilise the plant recovery rate at about 96% as part of the production expansion from c 200koz to c 230koz of gold, as was previously announced by Yamana. At the existing operations, La Colorada capital spend is estimated at US$36–38m as the company continues to advance the Skarn project through the PEA (due for completion in H223) and to upgrade the ventilation system at the existing mine.

The overall G&A expense is guided at US$75–80m. In FY22, Yamana reported US$120m in G&A cost, which gives a combined expense of almost US$150m. The FY23 guidance implies a reduction of c US$70m and is therefore in line with the company’s earlier guidance of c US$50m in annualised overhead cost savings. Another US$98–109m is expected to be spent on care and maintenance, with the MARA project contributing US$60–65m (on a 100% basis) in additional expenditure.

We have revised our earnings estimates to reflect the acquisition of Yamana as well as production and cost guidance for FY23. Based on our updated gold and silver price assumptions (see Exhibit 11), we forecast consolidated FY23 revenues and EBITDA of US$2.4bn and US$734m, implying a margin of 30%. We see further improvements in FY24 as the company benefits from the 12-month contribution from the Yamana assets. While costs in FY23 are expected to be mainly driven down by the addition of the lower-cost Yamana operations, in FY24 we expect cash costs to start gradually normalising on the back of easing global inflationary pressures. On the flipside, we forecast a slight reduction in the average gold price in FY24. Our silver price assumptions reflect the improving global economic outlook and growing renewable energy spend. As such, we see a gradual normalisation of the gold/silver ratio. All in all, we forecast FY24 revenues and EBITDA of US$2.9bn and US$1.1bn, at an improved margin of 38%.

Apart from Yamana, the biggest change in our modelling approach is for Escobal. While we previously assumed the project to be restarted in FY24, we now conservatively model its recommissioning in FY25. Further, we now assume the project to process higher-grade ore reserves in earlier years of operation, in a similar vein to the historical production numbers. Our previous production assumptions were largely based on the feasibility study averages. We have also increased our cost assumptions for the project to partially capture the recent industry-wide cost pressures.

While our overall approach to the valuation of PAAS remains unchanged in that we continue to employ a discounted cash flow (DCF) valuation methodology, we have introduced a number of changes to capture the value of the Yamana assets. Previously, our fundamental valuation of PAAS was based on a DCF to equity holders, whereby we assumed that the company pays out all net cash flow generated during the year as a theoretical dividend to equity holders. Our DCFE was based on a five-year explicit forecast period and a terminal value. We have now adopted a more traditional approach to valuing mining stocks and use the conventional DCF methodology over the mine life of the respective projects. For the existing PAAS operations and Escobal, we combine these cash flows into a single consolidated cash flow, which we discount by a CAPM-derived WACC. For the Yamana assets, our DCF-derived valuation is based on separate projects, to which we add the value of MARA and subtract the present value of the corporate G&A post anticipated cost savings.

Our valuation of MARA is based on the transaction value derived from the sale of Newmont’s 18.75% stake in the project to Glencore in September 2022. To this we apply a 25% premium for control, which implies an overall value for PAAS’s 56.3% stake of US$581m (US$1.1bn on a 100% basis). We believe that MARA does not fit into the company’s project portfolio and is a likely candidate for divestment.

Our approach to the valuation of PAAS’s exploration and other advanced projects remains unchanged. We use a peer group-based EV/Resource multiple method for the Skarn project and book values for Morococha, Navidad and La Arena.

Our approach to deriving the discount rate remains broadly unchanged. Our current nominal WACC is 7.5% (7.3% previously). It is based on the risk-free rate of 3.1% (Canadian long-term bond), an equity risk premium of 5.9% (using data by Damodaran) and a beta of 1x. Our current real discount rate assumption is 4%, which we apply to real cash flows post FY28. Our model for PAAS is a mixture of ‘real’ and ‘nominal’ approaches because, although we assume some cost normalisation post COVID-19 and the recent energy price spike, our long-term costs and commodity prices are expressed in real terms.

All in all, our combined equity valuation of the company is US$8.2bn, which translates into a per-share value of US$22.5 and implies 43% upside.

Peer group valuation of PAAS is complicated by the company’s increased scale and the lack of comparable medium-sized primary silver producers with a significant gold revenue stream. We have therefore subjectively put together separate peer groups for silver and gold companies. In addition to size, the analysis is distorted by different regional exposures of the broadly comparable companies.

On our revised estimates, PAAS trades on an FY24e EV/EBITDA multiple of 5.8x (consensus 5.4x) compared to the silver universe average multiple of 7.3x and the gold average of 6.9x. Applying an average silver/gold EV/EBITDA of 7.1x to PAAS’s FY24e EBITDA of US$1,099m, adjusting for net debt and adding our valuation of MARA, implies a per-share value of US$21.4.