Wheaton Precious Metals — Putting WPM onto GMT

On 20 February, WPM announced that it had produced and sold 374,585oz and 327,336oz gold and 17,220koz and 14,326koz silver, respectively (among others), in FY23, implying a notably strong fourth quarter performance in all of its business areas with the single exception of palladium sales, which anyway accounted for only 1.2% of the total (Edison estimate). A full summary of WPM’s production and sales results for all metals for the year, relative to both prior guidance and Edison’s prior expectations, is as follows:

In addition to adjusting our Q423 and FY23 forecasts to reflect actual compared to expected production and sales levels, we have also adjusted them to reflect actual metals prices prevailing during the fourth quarter compared with those expected at the time of our last note, published on 20 November 2023. These are similarly summarised below:

Given its copper production (reported to the market by Vale on 29 January 2024) of 55.2kt in Q423, we suspect that almost all (if not all) of WPM’s gold production outperformance will have been attributable to Salobo, where we believe that gold production for WPM’s account will have been close to record levels (see Exhibits 3 and 4, below).

Also shown in Exhibit 3 is our forecast quarterly production of gold attributable to WPM from Salobo in FY24 in the light of the company’s guidance. In the meantime, we suspect that most of the outperformance in silver sales in Q4 will have been attributable to Peñasquito, where Newmont reported production and sales of 4Moz and 5Moz, respectively, during Q4 (of which 25% is attributable to WPM), compared with our prior forecasts of 1.3Moz in production and zero ounces in sales attributable to WPM during the quarter.

Notwithstanding the unanticipated sales at Peñasquito, we calculate that aggregate sales of silver were still 1.3Moz, or 29.7%, below production for the fourth quarter, which is close to the bottom end of the range for the period Q112–Q423e (see Exhibit 5). By contrast, we calculate that gold production and sales were closely in line (probably mostly attributable to Salobo, where we suspect there will have been a modest over-sale of material in Q4 after a material shortfall in Q3 when sales lagged the ramp-up of Salobo III):

As a result, we calculate that gold ounces produced but not yet delivered to WPM will have decreased from 99,293oz at the end of Q3 to c 95,941oz at the end of Q4 (or 3.07 months of production), while silver ounces produced but not yet delivered will have increased from a historically low level of 1.1Moz at the end of Q3 to a more normal level of c 2.4Moz (1.68 months of production) at the end of Q4. Note that these compare with WPM’s target levels of two to three months for gold and palladium production and two months for silver production:

At the time of its Q422 results, WPM provided guidance for non-stock general and administrative (G&A) expenses of US$47–50m, or US$11.75–12.50m per quarter, for FY23 (cf US$47–49m in FY22, US$42–44m in FY21 and US$40–43m in FY20), including all employee-related expenses, charitable contributions, etc, but excluding performance share units (PSU) and equity settled stock-based compensation.

As at the end of Q323, the analysis of stock-based G&A expenses relative to the change in WPM’s share price (also in US dollars) over the previous 16 quarters continued to exhibit a close Pearson product moment (correlation) coefficient between the two of 0.78, which can be said to be statistically significant at the 5% level for a directional hypothesis (ie there is less than a 5% probability that this relationship occurred by random chance). Given the strong performance of WPM’s shares during Q4, we therefore anticipate that stock-based G&A expenses during the quarter will have been in the order of US$8.7m (±US$2.5m), as shown by the point on the best-fit line in Exhibit 7, below:

Consequently, we forecast that the total G&A charge for WPM in Q423 will have been US$21.2m, of which 59% will have been attributable to non-stock-based G&A expenses and 41% to stock-based expenses.

As a result of WPM’s disclosures, therefore, we have increased our earnings forecasts for Q423 by 25.2% and for the full year by 6.5%:

Moreover, our forecasts for WPM’s adjusted EPS for both Q423 and FY23 are towards the top end of the market range:

At the same time as updating the market for its actual sales and production in FY23, WPM also took the opportunity to provide detailed production guidance for FY24 and beyond. Whereas in the past, this had entailed providing numbers for the current year and then five- and 10-year averages looking forward, in this case the company provided detailed guidance for FY24, a production target for FY28 and then an average for the five years from FY29 to FY33. A comparison between these two formats and our current forecasts over the equivalent periods is as follows:

In 2024, WPM forecasts GEO production to be consistent with levels achieved in 2023, as expected stronger attributable production from Peñasquito and Voisey’s Bay is forecast to be offset by lower production from Salobo, the suspension of operations at Minto and the temporary halting of production at Aljustrel. Attributable production is forecast to increase at Peñasquito as a result of uninterrupted operations and at Voisey’s Bay owing to the ongoing transition from the Ovoid pit to the underground mines. Attributable production is forecast to decrease slightly at Salobo owing to lower grades, which are expected to be partially offset by increasing throughput as the Salobo III expansion project continues towards completion. However, we note that – if we are correct in our assumptions about Salobo in Q423 (see Exhibit 4) – then quarterly production at Salobo will have to fall sharply into Q124 and Q224 (see Exhibit 3) for WPM to not exceed its gold production guidance at least. In addition, WPM anticipates that production from the Blackwater and Platreef projects will commence in Q424. As such, we consider the ‘risks’ to WPM’s FY24 guidance to be very much skewed to the upside.

WPM’s updated guidance for FY24 and beyond is based on standardised pricing assumptions of US$2,000/oz gold (cf US$1,850/oz in FY23), US$23.00/oz silver (cf US$24.00/oz), US$1,000/oz palladium (cf US$1,800/oz), US$1,000/oz platinum (cf US$1,100/oz) and US$13.00/lb cobalt (cf US$18.75/lb). Of note is the implied gold/silver ratio of 87.0x, which compares with an equivalent figure of 83.0x in FY23, the current ratio of 90.0x and a long-term average of 60.1x (since gold was demonetised in August 1971). At the updated standardised prices indicated, our production forecast of 613.9koz gold equivalent (GEO or AuE) for FY24 is towards the top end of WPM’s guidance range of 550–620koz AuE, although our gold equivalent sales forecast of 558.2koz is towards the bottom end of the same range (see Exhibit 12). This compares with our prior GEO forecast for production in FY24 of 770.9koz (at the prior standardised prices) or 714.8koz (at the updated standardised prices). Stated alternatively, were it not for the change in standardised prices, our FY24 forecast for GEOs produced would have been 40.5koz (or 6.6%) higher at 654.4koz.

Otherwise, readers will note that our longer-term production forecasts are within 2.2% of WPM’s guidance for the period FY29–33. However, at the moment, these exclude any contribution from either Kutcho, Fenix or Toroparu or WPM’s royalty interests. Note that both our and WPM’s estimates necessarily exclude potential future stream acquisitions (of which we expect there to be a number in the time horizon considered).

In August 2023, the Canadian Department of Finance released draft legislation for a Global Minimum Tax Act, which will introduce a GMT in Canada. In summary, this will entail:

The tax will apply to members of multinational groups that have annual consolidated revenues in excess of €750m, with a business presence in Canada and at least one foreign jurisdiction. As proposed, the Canadian GMT is expected to apply to the fiscal years of those multinationals from 31 December 2023 and, we expect, will be applied to WPM’s non-Canadian – or more specifically, its Cayman Islands – earnings (which we estimate at 95% of the total) in FY24.

Currently, the Cayman Islands does not levy corporate tax. As a result, prior to FY24, WPM only paid tax on its domestic Canadian earnings – for these purposes, on its precious metals purchase agreements (PMPAs) at Sudbury and Voisey’s Bay. However, owing to the PMPA capital values, – including carried forward losses – we do not expect either of these Canadian assets to be in a taxable position for some time. In due course, these streams will also be joined by those from Blackwater, Marathon, Goose, Kutcho and Kudz Ze Kayah (and potentially other Canadian mines). In similar fashion, however, these will also not be in a cash taxable position until their capital values have reached zero. As a consequence, we estimate that WPM will not pay tax on its Canadian assets until FY30 and that it will not reflect tax expense at or near its full (effective) rate of 27% until FY34.

By contrast, in FY24, we expect that WPM will become liable for GMT at close to the full rate of 15% on its Cayman Islands earnings (minus the proportion of G&A costs that are recharged from its Vancouver HQ to its Cayman Islands subsidiary). Under IFRS, there is no requirement to account for deferred taxes related to GMT. At the current time, we think that it is highly unlikely that the required Canadian legislation will be in place before the end of March – in which case, WPM’s income statement in Q124 will not show any GMT liability. However, we think that it is quite possible that it will be in place by the end of June, in which case the Q224 GMT liability will effectively be a ‘double dose’ also including the retrospective liability for Q124 back to 1 January 2024. In the first year of operation, the tax will be payable 18 months after the year in which it is incurred (ie mid-2026 for FY24) and, thereafter, every 15 months after the year in which it is incurred – although we recognise that this might move to an instalments-based payments profile in due course. As a result, whereas we had previously forecast a negligible effective rate of tax for WPM into the future, we are now forecasting a rate that will range in scope from 12.9% to 16.3% for the period FY24–35 (all other things being equal).

Our preliminary quarterly financial forecasts for WPM for FY24 are based on its production guidance for the year (albeit we believe that the ‘risks’ relating to its guidance are skewed to the upside), recently prevailing metals prices and the (retrospective) imposition of GMT from Q224. While this, in aggregate, has resulted in a 22.0% reduction in our earnings estimates for the year, it is worth noting that 69% of the decline is attributable solely to the new GMT regime and only 31% to other factors:

While lower than our previous forecast, our updated adjusted basic EPS forecast of US$1.13 per share for FY24 is nevertheless near to consensus, as shown below:

WPM is a multi-asset company that has shown a willingness and desire to buy streams in the past to maintain production and maximise shareholder returns. As a result, rather than our customary method of discounting maximum potential dividends over the life of operations back to FY24, in the case of WPM (as with Newmont and Endeavour), we discount forecast cash flows back over three years to the start of FY24 and then apply an ex-growth terminal multiple to forecast cash flows in that year (ie FY26) based on the appropriate discount rate.

Relative to our previous note published on 20 November 2023, the principal change to our financial model has been the imposition of GMT from Q224. As a consequence of this change, our estimate of WPM’s ‘terminal’ cash flow in FY26 has decreased by 14.2%, from US$2.96 per share to US$2.54 per share – with the majority of the decline being attributable to GMT.

Assuming 4% growth in nominal cash flows beyond FY26 (ie 0% growth in real cash flows) and applying a discount rate of 9% (being the expected long-term required nominal equity return), our ‘terminal’ valuation of the company at end-FY26 is US$52.72 per share (cf US$61.41/share previously), or C$71.19 per share. However, it should be noted that this valuation is inherently conservative in that it assumes zero growth in (real) cash flows beyond FY26. This is inconsistent with the gold price, which has risen at a compound average annual growth rate of 7.4% per year from 1967 to 2023, a simple average annual growth rate of 9.3% per year (cf a compound average inflation rate over the same period of 4.0%) and a compound average real annual growth rate of 3.0% per year.

It is also inconsistent with WPM’s longer-term historical performance, wherein operational cash flows have increased at a compound average annual growth rate of 20.8% pa for the 17 years between FY05 and FY22, while its operational cash flows per share have increased at a compound average annual growth rate of 13.9% pa.

Excluding FY04 (part-year), WPM’s shares have historically traded on an average P/E multiple of 30.4x current year basic underlying EPS, excluding impairments (cf 38.2x Edison and 39.2x Refinitiv consensus FY23e currently – see Exhibit 17).

Applying this 30.4x multiple to our updated EPS forecast of US$1.60 in FY26 (cf US$2.01/share previously – again with the majority of the decline attributable to the imposition of GMT) implies a potential value per share for WPM of US$48.77 (cf US$61.19 previously) or C$65.86 in that year.

From a relative perspective, WPM’s valuation compares as follows to its most prominent peers:

Of note are the relatively slow pace of earnings and cash flow increases implied by the market’s consensus multiples between years two and three (ie FY24 and FY25) when we anticipate that the company will have begun an aggressive upward trajectory in both GEOs produced and EPS and cash flows per share. This could be explained either by the market’s anticipating a slower ramp-up in production or that the ramp-up will be partially offset by lower metals prices. Also of note is the market’s lower forecast yields for WPM in years two and three compared with year one, which seems to imply that it believes that WPM will cut its dividend – a contingency that we would regard as highly unlikely, except in extenuating circumstances (eg materially lower precious metals prices).

As at 30 September, WPM had US$833.9m in cash on its balance sheet and no debt outstanding under its US$2bn revolving credit facility. As such (including a modest US$6.2m in lease liabilities), it had US$827.7m in net cash overall after generating US$171.1m in operating cash flow, investing US$98.9m and paying out US$67.0m in dividends during the quarter. In Q4, we estimate that WPM will have generated a near-record c US$225.5m from operating activities (before working capital changes), but that its net cash position will have declined to US$592.3m – principally on account of the payment of a US$370m instalment to Vale in respect of Salobo III’s successfully concluded completion test during the quarter.

In addition, WPM has long-term investments, in the form of equity share- and warrant-holdings, in listed companies in the sum of US$200.9m as at 30 September, or the equivalent of US$0.56/share.

For FY23 as a whole, we estimate that WPM will have generated US$754.4m from operating activities (after working capital), before consuming US$584.5m in investing activities (excluding dividends received) and paying out up to US$271.7m in forecast dividends. However, readers should note that the timing of PMPA payments is uncertain and, inasmuch as investments are advanced or delayed, it is possible that WPM could register either a larger or smaller net cash position on its balance sheet by the year end. Either way, all other things being equal, in the absence of any major new asset acquisitions, we do not expect that WPM will require recourse to its debt facilities at any point in the foreseeable future.