Gemfields Group — Still good value under varied COVID-19 scenarios

While in the context of current global uncertainties investors may be more focused on the outlook for 2020 and risks relating to nearer-term cash flows, we think it is important to look at the strength of the underlying business that is clearly evidenced in Gemfields’ solid 2019 full-year results, which were released on 6 April 2020.

Gemfields generated record revenue of US$216.2m (2018: US$206.1m) and EBITDA of US$80.9m (2018: US$58.9m), a 37% EBITDA margin (2018: 29%). Those strong results are despite the imposition of a 15% export tax on Zambian emeralds during 2019 (since removed), which reduced 2019 EBITDA by US$12.4m.

Gemfields ended 2019 with a strong balance sheet, with cash of US$78.2m and net cash of US$25.4m (up from net cash of US$9.8m at the end of 2018). Free cash flow from operations before working capital movements of US$31.1m was offset by a US$25.7m negative movement in working capital largely as a result of an increase in receivables from the large (US$71.5m) ruby auction, which only concluded on 14 December 2019. The bulk of the US$56.7m in auction receivables outstanding at 31 December 2019 were collected by the end of February 2020. The strong balance sheet position means that Gemfields begins 2020 well placed to weather the storm.

Given the uncertainties surrounding the ability to conduct emerald and ruby auctions, which require customers to travel internationally, the 2019 results contain a note in relation to the going concern basis of accounting (note 12). Specifically, investors should be aware that if Gemfields is not able to conduct any emerald or ruby auctions by October 2020, the company may need to renegotiate existing debt facilities (covenants) and potentially seek additional funding and/or take more drastic action to reduce costs.

On 30 March, Gemfields announced that all but critical operations at Kagem had been suspended for at least a month. The company also announced it was likely that the emerald (high quality, HQ) and ruby (mixed quality, MQ) auctions originally scheduled for May and June 2020, respectively, are highly likely to be rescheduled, with the possibility that the subsequent emerald (HQ) and ruby (MQ) auctions originally scheduled for November and December 2020 may be cancelled altogether. And on 22 April Gemfields further announced that operation at MRM would also be suspended in response to COVID-19. The company is also increasing security at MRM in response to increased local security risks. UK staff (including directors) will take a 20% temporary cut in pay in May.

Gemfields generates more than 90% of its revenues from emerald and ruby auctions usually comprising two HQ emerald auctions (May and November, usually in Singapore), two commercial-quality (CQ) emerald auctions (February and August, usually in Lusaka) and two MQ ruby auctions (June and December, usually in Singapore). The challenge from a practical perspective is that these auctions rely on customers travelling from multiple countries to carefully inspect the stones and determine the value of their bids. This process cannot take place remotely without the customers having physical access to the stones. For Gemfields to hold auctions thus requires not only the lifting of COVID-19 lockdowns in the respective locations of the planned auctions, but also lifting worldwide travel restrictions to the extent required to allow free movement of both stones and customers between countries.

In note 12 to the company’s final 2019 results (‘going concern’) Gemfields also notes that it believes it would be prudent, even once COVID-19 restrictions are lifted, to allow customers ample time to recover before the first auctions are held. As such, Gemfields expects to hold auctions only in the fourth quarter of 2020, with one HQ emerald, one further CQ emerald and one ruby auction held in Q4 (in addition to the CQ auction already held in February 2020, which generated sales proceeds of US$11.5m).

We have looked at three potential scenarios for Gemfields in terms of the potential impact of COVID-19 on 2020 and longer-term metrics. Each scenario is described below with the key assumptions and impact on forecast key metrics compared in Exhibits 2 and 3 below.

In terms of auction sales, our central case assumes that no mid-year auctions will be held and that one CQ emerald, one HQ emerald and one ruby auction are held in Q4. We assume the Q4 auctions will be of a similar volume to what would normally have been the case but now assume prices 10% lower than our previous forecasts. While the risk to demand and pricing is on the downside given the current challenging economic conditions, we believe that the lack of May/June auctions could potentially result in at least some pent up demand from customers, which should help to partially mitigate the impact.

Operationally, our central case assumes a two-month shutdown of operations at Kagem (ie extending the one-month shutdown announced on 30 March) and a similar length shutdown at MRM.

In terms of costs, our central case sees a 15% decrease in 2020 cash mining and production costs in absolute terms relative to our previous forecasts as a result of:

We also forecast a 25% decline in selling, general and admin costs relative to our previous forecasts, partly as a result of direct costs not being incurred in relation to May and June auctions and taking into account the announced 20% reduction in UK (including board) salaries for May.

From 2021 onwards, our central case sees a more limited longer-term impact on MRM and Kagem. We assume::

Some 80% of Faberge’s wholesale customers are currently closed, and the impact on demand is likely to be significant. We now forecast revenue of just US$9m for Faberge in 2020 (versus US$17m previously), rising to US$20m in 2022 (from US$27m previously).

Our faster recovery scenario assumes there is a nearer-term change in countries’ responses to the COVID-19 pandemic in terms of re-opening their economics and lifting travel restrictions. In this scenario we assume some pent-up demand from customers sees slightly larger auctions than in our central case held either late in Q3 or early in Q4. In this scenario, 2021 and 2022 benefit from some sales of excess emerald and ruby inventory produced in 2020. While this scenario may seem unlikely to most observers, we note that the current situation whereby multiple countries prevent their residents from freely leaving their own homes would have seemed even more unlikely from the standpoint of December 2019 – a lot can change in a small space of time.

Operationally, the faster recovery largely mirrors the central case but with resumption of operations two weeks earlier and thus slightly higher costs in absolute terms.

Our slower recovery scenario assumes certain COVID-19 restrictions are still in place in Q4 and thus Gemfields is unable to hold standard auctions even late in the year. We assume that some sales would still be possible through more flexible mechanisms, for example holding a CQ and small HQ emerald auction in Jaipur, and selling selected parcels of rubies direct to customers in key locations. In this scenario in addition to lower volumes, we assume a 15% negative impact on 2020 prices relative to our previous forecasts.

In this scenario the company would need to renegotiate and extend debt facilities and covenants, although we think it is unlikely it would need to turn to equity markets. In this scenario we also assume the company closes both Kagem and MRM for three months and takes more drastic measures to reduce costs across the business. As a result, in this scenario we see 2020 cash mining and production costs a further 5% lower than in the central case and SG&A a further 9% lower in absolute terms.

Our slower recovery scenario also assumes the more severe global economic recession affects longer-term coloured gemstone demand and thus pricing. Despite the company’s ability to offer higher-quality stones for sale from inventory built in 2020, we nevertheless reduce forecast 2021 and longer-term prices by 10% from our previous forecasts and assume 10% lower 2021 auctions sales volumes.

Exhibit 2 below provides an overview of the three COVID-19 scenarios and the changes relative to our previous forecast assumptions.

Exhibit 3 on the following page sets out the key metrics for Gemfields in each of these three scenarios (and for reference also includes our previous forecasts).

We would expect Gemfields to generate US$102m in 2020 revenue in the central COVID-19 scenario (a fall of 56% of our previous 2020 revenue forecast). In the faster recovery case, we would expect 2020 revenue of US$128m, but we see revenue of just US$53m in the slower recovery case. Gemfields remains EBITDA positive in 2020 in both the faster recovery and central cases generating 2020 EBITDA of US$26m and US$11m, respectively (our previous 2020 EBITDA forecast was US$75m). That falls to a 2020 EBITDA loss of US$14m in the slower recovery scenario. In 2021 EBITDA grows strongly to US$102m in the faster recovery case, US$87m in the central case and to US$79m even in the slower recovery case. In 2022 we forecast EBITDA of US$118m, US$104m and US$81m in the faster, central and slower recovery cases, respectively.

Gemfields is fortunate to be starting 2020 in a strong balance sheet position with cash of US$78.2m and net cash of US$25.4m. The company should also benefit from a sizeable positive move in working capital in 2020 as it ended 2019 with US$56.7m in auction receivables largely relating to the December 2019 ruby auction (and with the majority of this collected by the end of February 2020). In the faster recovery scenario, Gemfields would still end 2020 with net cash of US$15m, but that reduces to US$4m net cash in the central case and to net debt of US$21m in the slower recovery case.

As previously, we value Gemfields on a discounted cash flow sum of the parts. However, given the uncertainties surrounding the pandemic, we now present three valuation scenarios in line with our three COVID-19 scenarios.

Rand per share valuations (of interest mostly to South African investors) benefit from the sharply weaker rand (ZAR18.95/US$ vs ZAR14.87/US$ previously) used to translate our predominantly US dollar driven valuation of Gemfields to rand per share.