Wheaton Precious Metals — ROICs alive

ROIC is a calculation used increasingly by investors to assess a company's efficiency at allocating the capital under its control to profitable investments. The formula for calculating ROIC (expressed as a percentage) may be rendered as follows:
ROIC = net operating profit after tax × 100 ÷ invested capital

Net operating profit after tax (NOPAT) may be calculated as operating profit (or EBIT) rendered on a post-tax basis and invested capital is typically calculated as debt + equity. Note that debt may be net debt or gross debt depending on the extent (or not) to which any cash on the balance sheet is immediately available to pay down structural debt.

In general, a comparison between a company's ROIC with its WACC may be said to reveal whether invested capital is being used effectively by the company and, as a result, many companies are being tempted to quote specific targets for returns. Like many financial ratios, ROIC may be used by investors in either an absolute or a relative fashion. Used as a relative measure of a company’s performance against either its own historical performance or against its peers with comparable capital structures and investment strategies, we do not believe that it poses many risks to investors. However, as with all financial ratios, it requires thoughtful application and, in this case, we believe that ROIC has certain deficiencies when applied to streaming companies as an absolute measure of capital efficiency, particularly when calculated for any specific year rather than over the life of any investment as a whole. We could argue this point in a number of ways. However, for the purposes of this note, we believe that ROIC’s shortcomings as a measure of capital efficiency are best highlighted by considering one theoretical stream and the accounting treatments that surround it.

Consider a stream of income for which an investor pays 100 units of currency (perhaps best conceived of as US$100m) in return for an annual income of 23.9 (henceforth US$23.9m) per annum for the next 10 years. The internal rate of return (IRR) of this investment – over the life of the investment – is 20.0%.

The 20.0% IRR is somewhat analogous to the yield to maturity of a bond. The so-called running yield of the same bond would be said to be 23.9%.

For the purposes of this example, the return of US$23.9m per annum may be conceived of as the difference between revenue of US$31.9m and direct costs of US$8.0m. As far as a streaming company such as WPM is concerned, this stream of income could, in turn, be conceived of as being derived from a stream of 17,722oz gold per annum (sold at a price of US$1,800/oz) for which it will pay an ongoing cost of US$454/oz and for which it paid an upfront cost of US$100m (ie roughly approximating a typical WPM stream).

However, the application of western (eg IFRS, GAAP) accounting standards to a company owning such a stream of income would render its accounts (albeit simplified) over the life to the investment as follows:

A number of conditions and assumptions have been applied to the company, which may be summarised as:

Readers should note the fact that the calculated IRR of the cash flows relating to the company’s debt providers of 5.0% reconciles with our stipulated 5% cost of debt. Among other things this acts as check on the overall rectitude of the model, in general, and therefore also on the returns available to equity holders, in particular (see ‘Potential remedies’ on pages 5–6).

One of the main consequences of applying generally accepted accounting standards to the company can be seen to be the introduction of a (non-cash) US$10m annual charge for depletion (analogous to depreciation) to reflect the initial US$100m investment into the stream over its 10-year life. This has two major effects:

The overall consequence of these two effects is to reduce ROIC (see line marked ‘Company ROIC (%)’ in Exhibit 3) in the early years of the life of the stream and to inflate it in the later years. Relative to the stream’s 20.0% IRR calculated earlier, the variation in ROIC over the life of the investment may be depicted graphically as follows:

To some extent, the 20.0% lifetime IRR of the stream may be considered to be a time-weighted average of the variable ROICs calculated year by year. Of note, within this context however, is the fact that the calculated ROIC for the company does not exceed the 20.0% IRR of the stream until the fourth year of the stream’s life and that it does not exceed the 23.9% running yield of the stream (see above) until the fifth year of the stream’s life. From this, we conclude that looking at a ROIC for any one particular year is not a good indicator of a streaming company’s capital efficiency unless it is considered critically within the context of that year’s chronological position within the life of the steam.

Two further points are pertinent to this finding:

In its consideration of its shortcomings, Edison has looked at three potential alternatives to ROIC in Exhibit 2, including cash flow return on investments (pre-tax, pre-interest cash flows divided by invested capital), return on equity (net income divided by balance sheet equity) and cash flow return on equity (pre-tax, pre-interest cash flows divided by balance sheet equity). These are detailed in the exhibit. However, as can be seen from the outcome of the analyses, none of these alternatives overcome the central distorting effect of applying the depletion charge to earnings and thereby to equity as a component of ‘investments’ resulting in an upwards trend in all four of these measures over the life of the stream.

In recognition of this, Edison proposes two further alternatives:

The problem of analysing capital efficiency at Wheaton is precisely that it has already aggregated some 32 streams and early deposit agreements, all of which are at different stages of development. However, we are able to observe that the years FY12, FY13, FY15, FY16 and FY18 were years of relatively heavy investment for Wheaton – eg from US$0.6bn up to as much as US$2.1bn in each of those years – often in streams with lives anticipated to be greater than 10 years. As such, we may speculate that Wheaton’s ROICs since FY12 should approximate the ROIC line on the left-hand side of the graph in Exhibit 3 (ie when it is below the equivalent IRR line), that is to say, ROIC is likely to understate the level of returns being earned by the company relative to their actual, or ultimate, level over the full life of the stream.

ROICs can be calculated in slightly different ways depending on the institutions conducting the analysis. For example, for the purposes of the analysis below, Edison calculates invested capital as total assets minus current liabilities. In addition, we calculate the ROIC based on a year-end number, rather than an average over the course of the year. With that caveat, a comparison of Edison’s historical, calculated ROICs for Wheaton compared to those calculated by Bloomberg is as follows:

Two particular features of the analysis in Exhibit 4 are notable:

Notwithstanding the shortcomings and caveats surrounding ROIC as a measure of capital efficiency in absolute terms, it may have some relevance as a relative valuation tool, albeit with the proviso that companies being compared should be at approximately the same stage of development relative to their expected future cash flow streams. For these purposes, we have construed this to mean that peers of Wheaton must be large, multi-asset companies with long future asset lives. Within this context a comparison of ROIC for Wheaton and eight of its peers (as calculated by Bloomberg) over the 13 years from FY08 until FY20 is as follows:

One notable feature of the calculation of ROIC that is relevant to the above analysis is that a higher proportion of Wheaton’s operating costs (used in the calculation of net operating profit after tax) is typically accounted for by depletion than for operating companies. In FY20, for example, depletion accounted for 43.8% of Wheaton’s cost of sales (including depletion), whereas for a sample of mining majors depreciation averaged 33.4%, ie the distorting effect of depletion on ROIC for streamers in general and Wheaton, in particular, appears to be larger than the equivalent distorting effect of depreciation for operating companies. Nevertheless, immediately apparent from the analysis is the fact that: