Ferroglobe PLC (GSM) 2018 Q3 法說會逐字稿

Good day, ladies and gentlemen, and welcome to the Ferroglobe Third Quarter Conference Call.

(Operator Instructions) I would now like to turn the conference over to Phillip Murnane, Ferroglobe's CFO.

You may begin.

Good morning, and thank you for joining the Ferroglobe Third Quarter 2018 Conference Call.

I'm going to read a brief statement and then hand the call over to Pedro Larrea, Ferroglobe's Chief Executive Officer.

Statements made by management during this conference call that are forward-looking are based on current expectations.

Risk Factors that could cause actual results to differ materially from these forward-looking statements can be found in Ferroglobe's most recent SEC filings and the exhibits to those filings, which are available on our web page, www.ferroglobe.com.

In addition, this discussion includes references to EBITDA, adjusted EBITDA and adjusted diluted earnings per share, which are non-IFRS measures.

Reconciliations of these non-IFRS measures to IFRS measures may be found in our most recent SEC filings.

On the call today, we will review the Q3 results across our core products, followed by some consolidated financial highlights.

Lastly, we will provide an update on the company's operating environment and the near-term outlook.

At this time, I will turn the call over to Pedro Larrea, Ferroglobe's Chief Executive Officer.

Thank you, Phil, and good morning, everyone.

If we turn to Slide 4, please.

The third quarter results were disappointing as the business was challenged with the confluence of softer flat prices and lighter volumes.

There are specific drivers impacting each of these variables, and we will cover them on today's call as we review each of our product categories.

Some of the factors are ones we have touched on previously, which have carried over into the back half of the year, while others are newer developments impacting our results this quarter.

Volumes across our product portfolio were down quarter-over-quarter.

In addition to some seasonal impacts, there were one-off factors contributing to this result.

Despite a drop in volumes in the third quarter, I would highlight that volumes overall have increased by 17% during the first 9 months of 2018 compared to the same period last year.

The most significant driver of the Q3 results was reduced pricing, specifically average sales price for silicon metal declined 4.9% versus Q2 2018, while average sales price for silicon-based alloys and manganese-based alloys fell by 5.6% and 7.1%, respectively.

To provide some context, I would highlight that prices are up over 13% on average in the first 9 months of 2018 compared to the same period in 2017.

The net impact of weaker volumes and pricing during the quarter yields a 9.6% decline in our top line revenue.

Reported and adjusted EBITDA in the quarter was $45 million, with adjusted EBITDA down 48% from $86.3 million during the prior quarter.

The decline in revenues led to an EBITDA margin compression in the third quarter to 8.5%, a decline of 625 basis points from the prior quarter.

Ferroglobe posted a quarterly net loss of $1.2 million and an adjusted net profit of approximately $0.1 million in Q3 compared to $25.7 million in the previous quarter.

Given these trends, we are active in making changes to our commercial, operational and financial strategies, which we will touch upon during this presentation.

Next slide, please.

Year-to-date, Ferroglobe generated $220.9 million of adjusted EBITDA, reflecting a 69% increase over the same period last year.

The adjusted EBITDA generated in the first 9 months of 2018 amounts to 120% of the company's total adjusted EBITDA in 2017.

While the year-over-year trend is certainly positive, the quarterly results were clearly impacted by pricing and the continued lag in the recovery of the manganese alloys business.

Next slide, please.

Sales for the quarter were $527 million, a 9.6% decline in Q3 relative to the prior quarter.

Revenues across our 3 primary product categories were down quarter-on-quarter due to lower volumes and average sales prices.

When compared to Q3 of the prior year, however, we have significantly increased volumes across all products, and we have significantly improved our average realized selling prices across both silicon metal and silicon-based alloys.

I will go into more detail into specific drivers behind our various product categories later in the presentation.

Next slide, please.

Adjusted EBITDA declined $41.3 million over the previous quarter.

Volume declines across the portfolio negatively impacted this quarter's adjusted EBITDA by $1.6 million.

The biggest contributor to the decrease in quarter-over-quarter adjusted EBITDA was decline in average selling prices, which were lower across all products.

In the aggregate, the realized selling price evolution across all products resulted in a negative adjusted EBITDA impact of $28.5 million during the quarter.

Costs and foreign exchange movements, which have been a major contributor to adjusted EBITDA variances in prior quarters, actually had a much lower impact this quarter.

It is to be noted that the cost increase of $2.1 million compared to previous quarter was impacted by $5.8 million of inventory write-down in manganese alloys.

So net of this impact, cost levels are actually starting to marginally improve.

Some commercial changes around the transfer pricing for our coal contributed to a negative $7 million evolution in EBITDA from the Mining division.

And finally, our Energy business was down $3.3 million in the quarter.

As we highlighted in prior quarters, the first half of 2018 was exceptional for this business, and we did not expect this trend to continue.

We are now back in line with historical levels.

Next slide.

On the next few slides, we will discuss pricing and volume trends, earnings contributions and market observations for each of our key products.

Turning first to silicon metal on Slide 8, in line with pricing declines in the U.S. and European indices, Ferroglobe's realized average selling price for silicon metal declined by almost 5% to $2,636 per metric ton as compared to $2,773 per metric ton in the second quarter.

Prices across North America, Europe and China have for the first time returned to levels in line with 2017, which are also in line with historical averages.

The net impact of silicon production at high rates, the impact of customers stocking up in anticipation of the trade case and the availability of aluminum scrap are weighing on the pricing environment in North America.

In Europe, the decline in index pricing is largely attributable to seasonally lower activity, coupled with the impact of freight flow movements, mainly from Brazil and China into Europe.

The bar chart on the top right of Slide 8 shows a decline in volumes over the prior quarter.

Volumes during the quarter were negatively impacted by a seasonal slowdown in July and August, trade flow movements, customer inventory builds and some unexpected customer outages during the quarter, mainly in the United States.

Additionally, silicon unit sales in the U.S. were impacted by the availability of aluminum scrap, which is now burdened by a 25% tariff on inputs from the U.S. into China.

The cost increases we have been facing in silicon metal production have started to stabilize and technical performance as well as active portfolio management are contributing to a better cost structure.

Next slide, please.

Turning to silicon-based alloys, overall, EBITDA contribution from this product category declined primarily as a result of lower pricing.

During the quarter, the average selling price decreased by 5.6% to $1,802 per metric ton, down from $1,908 in the second quarter.

There has been some pricing pressure in Europe, although from record levels, as a result of increased inputs from Malaysia.

At the same time, pricing in the U.S. has remained firm on the back of solid demand from the steel sector.

Sales volume was steady at 75,964 metric tons in Q3, with ferrosilicon continuing to deliver stable demand.

Globally, the steel market has been strong despite trade actions.

In particular, Ferroglobe is in a prime position to continue to serve 2 protected markets: the United States, 2, 3 to import duties and Europe enacted safeguard measures.

Additionally, the ongoing environmental crackdown in China has resulted in ferrosilicon production capacity curtailments there.

The current pricing environment has resulted in attracting new capacity in our products, and we are monitoring those developments, mainly out of Malaysia and Eastern Europe.

Similar to silicon metal, the silicon-based alloys business benefited from decreased costs in the third quarter.

Foundry products, which represent approximately 1/3 of our overall silicon-based alloys segment continued to perform well.

These are tailor-made, noncommodity products with much greater stability in prices and with double-digit volume growth year-over-year.

Next slide, please.

Turning now to our manganese-based alloys.

This product category has been a laggard in our portfolio for most of 2018.

The unforeseen dynamic, which has prevailed this year, de-linking manganese alloys prices from oil prices has lasted longer than anticipated and weigh on our results in the near term.

Our average realized price for manganese-based alloys decreased 7.1% to $1,211 per ton, down from $1,304 last quarter.

The manganese business has been under pressure as the prices of these alloys in Europe have dropped again in the quarter, while manganese ore prices remain high, all having an adverse impact on our spread.

While orders were strong during the quarter, we were faced with some logistical challenges along the supply chain, which resulted in some orders not being shipped during the quarter.

In addition to the pressure on the spreads, product margins were affected by high power costs in Spain, which have remained near multi-year highs.

We also note that costs and results in Q3 have been negatively affected by inventory write-downs for a value of $5.8 million.

Without these write-downs, costs would actually have started to show some recovery, thanks to efficiency measures being implemented.

At this point in time, we are starting to see a few trends on the ore supply side and on alloy prices, which reinforce our belief that this business will turn around.

However, it will take some time to realize the benefit of a price drop in ore, given our current inventory and the lag in shipments.

Next slide, please.

As you saw in the others -- in the adjusted EBITDA bridge on a prior slide, the biggest factor contributing to the decline this quarter has been pricing.

For the year, we have also been challenged with increasing input cost of gross products.

A year ago, there was a lot of focus on the graphite situation, and we were expecting significant further increases in electrode prices.

This year, on the heels of a strong global economy, we have seen other input cost increase such as manganese ore, coal, met coke and power.

Silicon metal costs have increased by 11% since Q1 2017, while silicon-based alloy costs and manganese-based alloy costs have increased by 20% and 15%, respectively, over the same period.

Next slide, please.

As an additional disclosure this quarter, we have included a detailed reconciliation of EBITDA, including both the core products that we have historically reported and other contributors.

One first item I would like to highlight is that as you can see from this historical EBITDA summary, the manganese business has contributed as much as $25 million of EBITDA in the quarter even before the acquisition of the new assets this year.

After the acquisition, this is an approximately $600 million revenue business.

With historical EBITDA margins of 10% to 15%, the manganese alloys business should deliver EBITDA results of $60 million to $90 million per year.

Silicon metal and silicon-based alloys show a very different cyclical behavior when compared to manganese-based alloys, underlining once more the benefit of our diversified product portfolio.

Other metals, which includes silicon fume, cored wire and other products such as fines and flat contributed adjusted EBITDA of $7 million in Q3.

Mining, which represents the accumulated results of the various Ferroglobe-owned mines, has historically contributed significantly more than the $4.2 million achieved during Q3.

The drop is attributable to lower volumes and a lower transfer price of coal from our captive mines to our plants, including the joint ventures.

Energy had contributed above-average returns during the first half of 2018.

In the current quarter, adjusted EBITDA from the energy assets yielded $2.4 million.

And finally, overheads continues to decrease steadily in 2018.

In Q3, the overhead cost of $18.5 million was 4.1% lower than the prior quarter.

Next slide, please.

Despite the company's weaker third quarter results, as I highlighted on previous slides, performance through the first 9 months of 2018 has significantly improved compared to the same period in 2017.

Demand across our product categories is up year-on-year, with volumes increases in silicon metal up 7%, silicon-based alloys, up 8% and manganese-based alloys up 37%.

Average sales price have improved across both silicon metal and silicon-based alloys, driven by stronger demand across our end markets.

For most of this year, we have been faced with a disjointed manganese industry, whereby, prices have remained under pressure, despite the manganese ore prices staying at very high levels.

For the 9 months of 2018, Ferroglobe posted a net profit of $102.9 million and an adjusted net profit of approximately $59.1 million compared to $10.5 million in the same period in 2017.

Next slide, please.

Adjusted EBITDA through the first 9 months of the year is up 69%, totaling $220.9 million year-to-date 2018 compared to $130.9 million over the same period in 2017.

The biggest driver behind this improvement is pricing despite some pullback in the recent quarter.

Higher prices in 2018 have contributed approximately $196 million to our adjusted EBITDA.

The cost side of the equation has had an adverse impact of $133.1 million, which does not include the impact of higher manganese ore prices.

For instance, on a year-on-year basis, electrode prices, coal, met coke and power prices in Spain have had a negative impact on our costs of approximately $70 million.

Manganese ore prices have had a $27.4 million negative impact on our adjusted EBITDA year-to-year and this has not been coupled with an increase in manganese alloys prices, which history suggests we should expect.

As a matter of fact, manganese alloy prices have had a negative impact of $1.1 million.

As we have reset some pricing relating to our internal coal supply and scaled back production in Laudun, there has been an impact on the mining division, hence the rather flat contribution this year.

The energy division has contributed $12.7 million and both other businesses and reduced overheads have benefited results by $14.7 million in the first 9 months.

With that, I would like -- I would now like to turn the call back to Phil, who will review the financials in some more detail.

Thank you, Pedro.

Focusing on Slide 16, please.

As you've already heard, sales volumes are down to approximately 256,000 tons in the quarter and revenues are approximately $527 million.

The quarter saw revenues impacted by weaker prices across all our major products and smaller volume declines.

Adjusted EBITDA of $45 million is, therefore, well down from prior quarter, with an 8.5% EBITDA margin.

Although results are down from prior quarter, as Pedro highlighted, year-to-date revenues, profits, EBITDA and margins are still well above the same period in 2017.

Moving to Slide 17, overall working capital increased to $443 million.

This quarterly increase is primarily driven by an unexpected build in finished inventory.

The growth in working capital in both the quarter and in the first 9 months has had a corresponding impact on both net debt and free cash flow.

During the quarter 2 results, we announced that we'd implemented various initiatives to return this working capital during the second half.

With the unexpected capital -- working capital build during the third quarter, achievement of all of these initiatives has become challenging, which I will discuss in more detail later in the presentation.

Although net debt has increased in the period, our balance sheet metrics remain strong.

We will continue to focus on deleveraging and, although challenged by the net debt build in the quarter, are committed to the targets we announced in Q1.

Next slide, please.

The working capital increase in the third quarter includes a $31 million increase in working capital of the existing assets and a further $5 million increase in the recently acquired manganese plants, bringing total working capital to $443 million.

The working capital build in the existing plants has been driven primarily by finished inventory builds across the majority of our plants and particularly silicon metal and manganese alloys finished inventories as Pedro highlighted.

We experienced some seasonal slowdown in July and August, which is typical given the summer holidays.

But the turnaround in sales in September did not materialize as we expected, and we experienced the impact of oversupply in the market as producers continued to run operations at high utilization rates.

This combined with the impact of the availability of aluminum scrap on silicon unit sales in the U.S. drove the finished inventory increases.

As previously highlighted, the manganese alloys although -- in manganese alloys, although orders were strong during the quarter, we were faced with some logistical challenges along the supply chain, which contributed to the inventory build.

Working capital at the recently acquired manganese plants increased to $95 million in the quarter, reflecting the increase in manganese alloys finished inventory.

Next slide, please.

The working capital build across the plants, which totals $36 million in the quarter continued the previous quarter's trends of working capital increases.

This drove an increase in our overall gross and net debt levels of $12 million and $36 million, respectively, in the quarter, ending the quarter at net debt of $511 million.

Next slide, please.

Looking in more detail at our cash generation.

Profit for the 9-month period was $99 million.

Our profit adjusted for noncash items was approximately $224 million.

This cash was then invested in operating assets $188 million, with significant working capital investment in the 9 months, including as I said, the working capital build in the recently acquired manganese plants and the finished inventory build.

Interest payments of $39 million, income tax payments of $29 million and payments due to property, plant and equipment of $78 million.

This resulted in a negative free cash flow for the first 9 months of $111 million.

Highlighting a few lines.

The working capital increase in the period, as I said, includes those recently acquired plants of $95 million and the build in the existing assets of $68 million.

These are not normal working capital builds and would not be expected to reoccur in 2019.

Regarding interest paid of $39 million, refinancing the company's 9.375% 2022 senior notes was explored during the quarter and will be reconsidered when the market conditions improve and as we deliver on our daily regime.

And finally, in respect to the payments due to property, plant and equipment, normally -- normalized recurrent capital expenditure should in a full year be similar to the 9 months spend in 2018.

Given the outflows required during H1 2018, we announced various cash-generating initiatives at the Q2 results call.

I would like to give a status update on those various actions.

The first action was a planned increase in the A/R securitization program, targeting additional liquidity of $35 million.

It has been completed and $20 million of that liquidity has been drawn, with no further plans to draw additional liquidity.

The second action was to reduce manganese ore inventory levels by $20 million.

We achieved significant tonnage reductions by November, and we are on track to complete this action by the end of Q4.

The third action was to increase the rotation of finished product inventories in key products and adapt production capacity to commercial commitments, which we have targeted to deliver $20 million.

Although significant volumes are expected to be shipped during November and December, given the unexpected inventory build during the quarter, this action is now delayed.

And finally, the fourth action was to complete noncore asset divestitures during the remainder of the year of about $20 million.

This is on track, with final negotiations underway.

Given our reported Q3 results, including the working capital build, our free cash flow targets for the second half of 2018 have become a stretch goal.

Next slide, please.

We continue to be committed to a conservative capital structure, and will focus on deleveraging the company's balance sheet.

Given our focus on cash generation, I would like to highlight a few updates relating to the solar project.

Design changes have been approved to increase capacity of the solar project, which have a budget impact of EUR 8 million, increasing the budget to $58.5 million, but given the focus on cash generation, we have decided to adapt the pace of construction and the expected ramp up, which will reduce cash outflows by EUR 20 million across 2018 and 2019.

EUR 12 million.

EUR 12 million.

And finally, in light of the financial performance in Q3 2018, the near-term market outlook and the company's continued focus on cash generation and deleveraging its balance sheet, no interim dividend is being declared or is being payable in respect of Q3 2018.

Now I would like to hand back to Pedro.

Thank you, Phil.

Next slide, please.

Thus far, in 2018, we have witnessed an interesting evolution in our business and the global markets for our products.

There were several prominent factors impacting our results, both positively and negatively, which we will need to monitor in the coming quarters.

On the positive side, we have seen increasing volumes across our end markets.

Steel, aluminum and chemicals in the near term, we expect this stability to continue, notwithstanding the uncertainties of the global macroeconomic picture going into 2019.

While the pricing environment improved in 2018, it was really a return to historical levels.

As noted, we are monitoring the supply-demand balance, and we keep a watchful eye on new capacity and conversions, which can impact this balance and thus, impact pricing.

On the other hand, we have also been challenged by a number of factors, some of which should play less of a role in the future, while others will continue to impact our business for some time.

On the cost side, we have witnessed cost inflation for inputs beyond graphite electrodes, which were the main topic a year ago.

With the growth in the global economy, many of the inputs we use have been in high demand and as a result, the pricing for these inputs has gone up.

This includes manganese ore, coal, met coke and power.

In 2018, we have experienced $77 million of cost increase as a result of price increases in electrode, power and coal alone.

While we do not have a crystal ball indicating when exactly these costs will improve again, we do feel inputs such as manganese ore and power in Spain are poised for a price decrease next year.

Other costs, such as electrodes, particularly graphite electrodes, will remain high into 2019.

Lastly, we are still in the same position in the manganese alloys business.

Previously, we had communicated our view that the trend would start to change in the back half of 2018, but unfortunately, manganese ore prices have held up longer than expected.

As new mining capacity hits the market, we expect to start to see pressure on the supply side, which will be beneficial to our business.

Next slide, please.

The projected growth in our end markets across aluminum, chemicals and solar industry should be positive for our business.

As the charts on the bottom of the slide illustrate, these industries have demonstrated growth in 2018, a trend which independent analysts expect to continue with the exception of the photovoltaic industry.

Aluminum demand globally is expected to increase in 2019 and is supported by the increased use of aluminum due to trends such as recyclability, light-weighting of orders, investment in construction and greater consumption of everyday products.

Some headwinds in this sector could come from the availability of cheap scrap in North America and from the pressure on vehicle emissions in Europe.

We continue to monitor the developments in the others industry.

The chemical sector has shown GDP plus growth this year.

And the market is expected to remain in supply deficit in 2019.

As global chemical producers create advanced solution for their customers, the silicon metal industry should benefit from pull-through demand in the near to medium term.

Lastly, regarding the polysilicon market, this portion of the silicon metal end market has created some uncertainty this year, following the unexpected cutting of solar-related subsidies in China.

It is important to remember that global photovoltaic installations almost doubled from 2015 to 2017.

So a temporary setback in China seems unlikely to affect what is an exciting story in this segment of the market.

As uncertainty looms around the trade war and as photovoltaic prices remain depressed, there will be some impact on silicon sales into this market as customers have curtailed production.

However, we agree with the view that the photovoltaic market will continue to grow in the long term.

Next slide, please.

We discussed some of the near-term silicon metal overhang in supply earlier in the presentation, but when we look beyond the next few quarters, a few data points stand out.

The recent trends and reform actions in China will be a net positive for our business.

While Chinese production continues to increase, as shown on the top left of this slide, the pace of increase is stabilizing as producers are forced to adhere to new environmental regulation.

This, coupled with growing demand within China means China will likely be exporting lower volumes in the future, as shown in the chart on the bottom left-hand side of the slide.

When we take a step back and look at the supply risks outside of China, the picture is positive for the industry.

With the recovery in silicon prices in the beginning of 2017, major producers that had previously idled capacity were incentivized to restart, most noticeably, the Brazilians.

Additionally, marginal producers in Thailand and Laos, which were within the scope of the trade case, have also restarted their facilities.

So the market is now running at fairly high utilization from a global perspective.

We expect that some incremental capacity could hit the market in 2019, but the net amount of incremental merchant volume is less than the anticipated growth in demand, as shown on the graph on the bottom right of this slide.

Next slide, please.

Earlier in the presentation, I summarized the factors which are creating some near-term downward pressure.

The graph on the left-hand side of the slide shows the quarterly supply and demand progression according to independent analysts.

Over the past 2 quarters, there has been a supply build, which has been an active topic at recent industry conferences.

When we account for the global supply-demand picture, including the point just made around the limited new production which could potentially hit the market, we see support for the case that the basic supply-demand fundamentals remain favorable.

For this reason, we believe that the short-term supply-demand imbalance of approximately 50,000 tons should not be mistaken for a significant market oversupply that will linger.

Next slide, please.

As we noted during the second quarter call, we have been prepared to take action based on our market pulse and developments on the supply and demand sides.

As we mentioned, the surplus aluminum scrap in the U.S., which is rich in silicon content, weighed negatively on demand for silicon units during the quarter.

This coupled with the fact that inputs into the U.S. increased following the trade case led to some oversupply buildup.

While this was enough reason to take action, we were further propelled to curtail production in the U.S. because of some favorable trends relating to foreign exchange, particularly in South Africa, which is now our lowest-cost production base.

We have talked about our ability to shift production and convert capacity on numerous occasions, and our recent announcement is a good example of the type of action we continually consider taking to capitalize on various trends.

The graph on this slide summarizes the actions we have taken globally to curtail silicon production, given the current apparent overhang of supply and increasing cost trends.

In Europe, France specifically, we have reduced production capacity by 40,000 tons.

This is tied due to silicon metal furnaces at Château-Feuillet and one furnace at Laudun.

Additionally, we curtailed 36,000 tons of silicon metal capacity in the U.S., 24,000 tons by idling 2 furnaces at Selma and the balance by idling 1 furnace at Beverly.

However, the depreciation of the South African rand and the renegotiated power supply contract support our production out of Polokwane in South Africa, hence, we are not modifying any production plans in that region.

In the quest to continue to optimize our production portfolio, the mentioned silicon metal curtailments are being partially compensated with increases in the production of other products.

For instance, the 2 furnaces at Château-Feuillet will be taking over production of calcium, silicon and foundry products from our factory in Mendoza, Argentina, while Beverly has increased its ferrosilicon production.

As a global producer, we are fortunate to be able to navigate changes in the competitive environment by leveraging those of our facilities that provide the most upside.

This provides cost savings, improved margins and enhanced flexibility.

Next slide, please.

Similarly, in the manganese alloys part of the business, we have taken action during Q4 to adapt to market circumstances.

Our Q4 production capacity has been reduced by 28,000 tons, which on an annualized basis represent 112,000 tons or around 17% of our capacity.

This decision is mainly driven by a need to draw down inventories and generate cash.

However, demand is shaping up to be strong during Q4 and there are some initial signs of improving prices.

We will remain attentive to these trends before we decide whether to continue these capacity reductions into 2019.

Next slide, please.

2018 is on pace to be a record year for the global steel industry in terms of demand.

According to the World Steel Association, steel demand is forecast to set a new demand record in 2019, implying further growth from peak levels.

As both developed countries and emerging markets continue to build, our ferrosilicon and manganese alloy segments are positioned to benefit from such demand.

Today, there has not been much impact from recent trade war rhetoric on global steel demand.

Our global reach and production footprint enables us to benefit from this trend in manganese alloys, ferrosilicon and foundry sales.

Whenever you have several strong years, there will be new capacity attracted to the market.

We have certainly seen that play out in respect of Malaysian producers of manganese alloys and ferrosilicon as well as some from Eastern European producers.

Overall, we anticipate seeing stable volume demand for these products in the near term.

Next slide, please.

Manganese alloys demand has remained strong this year, on the heels of the record demand in steel.

However, there has been no contribution to EBITDA from this business year-to-date as input costs such as manganese ore and power have been at higher-than-expected levels for an extended period of time.

When we compare the historical trends in manganese alloys pricing against the ore pricing, there is a strong correlation exceeding 85% according to independent analysis.

This implies that manganese ore price should eventually decline along with the alloys prices earlier this year, helping maintain a stable spread.

The manganese alloys business is one we have been in for over 25 years, and we have seen anomalies like this in the past.

We remain confident that weathering the storm will yield favorable results.

With the Chinese New Year still a few months away, we think any meaningful decline in ore prices will take some time to realize.

Furthermore, the additional ore capacity expected to hit the market in 2019 further supports our view that prices for ore shall decline.

Next slide, please.

As we continue to negotiate contracts in the silicon metal business for next year, all the trends and factors discussed today come into play.

The uncertainty we're facing around costs and perceived oversupply seems apparent in the way the negotiating season has progressed thus far as deals are taking longer to conclude.

We will head into 2019 with an order book, which is less committed overall than prior years in silicon metal as we feel it would be foolish to sign deals when customers are bottom fishing despite strong fundamentals.

Some customers may be finding some marginal producers will take a more aggressive stance and are even ready to sell at a loss, but given that costs are rising and end markets are performing well, we will not do business at those type of levels.

Shifting to the silicon-based alloys portion of our business, we are experiencing a pace of contracting, which is consistent with prior years with some very significant orders already booked.

Our focus is also to optimize our specialty sales where we have good discussions going on at the moment.

With manganese alloys, we will continue to weather this storm as we have and have focused on securing higher-margin business in the refined products.

We have curtailed some capacity, 28,000 tons in Q4, but demand seems to remain strong and there are some early signs of margins recovery through better alloys prices.

Next slide, please.

In closing, whilst the third quarter confirms an overall solid 2018 performance, the quarter itself has delivered very disappointing results.

Unexpected weakness in the pricing of our products, continuous headwinds in some of our cost components and the exceptionally long underperformance of our manganese alloys business are all factors that are negatively impacting our financials.

We continue to believe that the overall supply-demand fundamentals of our industry remain essentially solid and there could start to be some signs of recovery in the manganese business, but we cannot deny that we will be facing some headwinds in some areas of our business and that there is some inertia in the negative pricing trends that will affect our financials in the near term.

In this near-term context, our top priority is to remain focused on generating cash flow and deleveraging the company.

For that purpose, we count on the advantage of a very solid balance sheet and a comfortable liquidity position.

Ferroglobe is the leading company in its industry with an unrivaled asset base, a unique market position and a diversified portfolio of products, markets and production platforms.

We are proving that we are ready to actively deploy those competitive advantages, and we are confident that our efforts in optimizing the production platform and remaining disciplined in the commercial strategy will yield results.

At this time, I would like to open the line to any questions.

(Operator Instructions) Our first question comes from Ian Zaffino of Oppenheimer.

Couple of questions here.

Maybe if you could give us an idea of when you think you might hit free cash flow positive?

Also the $250 million of total liquidity you mentioned, what EBITDA is that based on?

Is there any type of covenant issues or anything related to that?

And then also, if we do go into a protracted downturn here, how are you thinking about your maturities?

Even though they're not till 2022, manganese business hasn't been doing well.

So I'm wondering maybe that could be a source of liquidity for you guys or how you think about that.

And then I have a follow-up.

Well, thank you, Ian.

And your second question, I'll hand over to Phil.

In terms of free cash flow positive, Q4 will be free cash flow positive.

We are pretty certain about that, and it's just that we have said that the overall targets we had established for second half of the year are now considered a stretch goal.

In terms of our $250 million liquidity and possible covenants in them, Phil?

In terms of liquidity, clearly, as we move forward, I mean, Pedro's talked about a lot of uncertainty in terms of the results, but we certainly see that we will continue to be committed to delivering on those actions we announced in Q2.

Liquidity will improve today and will continue to improve by that as we deliver on those cash-generating initiatives.

Looking forward to the 2022 bonds, there's no major covenant restrictions in terms of the impact that lower EBITDA would have on the bond.

The restrictions in the covenant are really around additional indebtedness or additional restrictions, rather than a limitation upon us or increased covenants as leverage would rise.

Certainly, as we look forward, and when I think about refinancing, the focus is clearly going to be the cash generation -- cash-generating initiatives that we announced in Q2 and continuing to focus on cash as we roll forward.

As such, our first priority is going to be continuing to deleverage the balance sheet.

We want to ensure irrespective of those of how the market outturns.

And on your third question, yes, it is clear that we have now a stable, solid balance sheet structure and that gives us, I would say, the confidence to do what is necessary to turn the business around.

And we certainly do trust that our manganese alloys business will turn around and will contribute positively to our financials.

Okay.

Just to be clear that $250 million is free and clear irregardless of movements in EBITDA?

As I said, yes, it's a limitation on additional indebtedness or the payments, it's not going to be that it would be an event of default if leverage rose.

Okay.

I'm just looking at EBITDA declining meaningfully and as you look back 12 months, you have a very strong quarter rolling off and then, as we progress throughout the next 12 months that EBITDA based on a trailing 12-months basis will be less.

So I just want to make sure that I'm understanding that properly.

You are.

There's no impact there.

We have a strong liquidity position.

And as I said, I would expect that liquidity position to strengthen as we deliver on the cash-generating initiatives.

Okay.

And then I guess, can you now touch upon what the price talk now is?

I think we've been hearing kind of 105, 110 on the silicon metal side.

Is that kind of what you're seeing?

And I know you'd referenced some people willing to sell at a loss.

Is that the kind of prices that you're talking about when you're saying people are agreeing to these types of numbers and they're going to be producing at a loss or maybe any type of color there?

Yes.

And again, let's remind the composition of our business.

You're asking specifically about silicon metal, which today represents 35% to 40% of our total business and within silicon metal, North America is around 1/3 of our business.

So let's say, 13% of our total revenues is silicon metal North America.

Silicon metal in North America, we are certainly not selling at those kinds of levels today, so not even close, and we are remaining very disciplined in not getting there.

We have certainly not been requested to supply at those kind of price levels at all.

So that is what we are seeing right now.

It is true that total -- and now, I'm talking both Europe and North America -- our order book today for silicon metal is around 45% of what would be our expected sales for 2019, which is slightly below where it generally is at this time of the year.

But again, in terms of prices, we are not seeing prices that are significantly below where the indexes are today.

Okay.

So basically, I could then take just this past quarter or, I guess, your guidance of the fourth quarter and annualize that and that's sort of a run rate you would be at in '19, given where the current price talk is, is that the right way to think about it?

I would say that would be a right thing or right way of thinking for the beginning of 2019.

We continue to believe that medium term, as we see it today, and of course, with the number of uncertainties around that factor, but medium term, the supply-demand dynamics in silicon metal, still talking about just silicon metal, remain positive.

So going forward, I wouldn't be surprised that there could be some price recovery in the medium term.

Our next question comes from Martin Englert of Jefferies.

Well, some of the costs seem to have improved sequentially across the different product lines, their prices are deteriorating faster than that.

So which of these do you believe are structural versus transitory and would you expect to roll off?

You called out maybe some of the power costs in Spain, maybe if you could just go down a list of some of your major costs and what you expect to be rolling off maybe in the near to medium term here and what you can do to better contain costs and lower costs still on a go-forward basis?

So in terms of costs, the main items that have been affecting our costs year-over-year, we have talked extensively about electrodes.

Our view is that electrode costs into 2019 will remain stable.

Maybe even with a slight increase and that is more due to the time lag.

So market prices in electrode are certainly going down, but given our existing orders and our existing stock, the impact on our costs is more or less stable into 2019.

Coal, international coal prices are clearly going down.

So that today, we see as a positive going into 2019.

And all that refers to energy-related costs worldwide: oil, gas, coal, et cetera, are going down so that should have a positive impact on our power prices in Spain, for instance.

The same applies to other factors like met coke, which has had a very significant, again, cost impact on our manganese alloys.

And as we sit today, I would say, that the trend looks positive, but again, with all the uncertainties that are inherent to these kind of markets.

Okay.

And are you able to give some cost numbers as far as your year-to-date headwinds for some of these different categories?

Can you just remind us what those were again breaking down your coal and your electricity, met coke, electrodes?

Yes.

In Slide 14 of the presentation, we're saying that the total cost deterioration year-on-year is $133 million, power in Spain is almost $24 million, electrode is almost $21 million, met coke is almost $17 million, coal is $9 million.

So all those factors are in total around $70 million.

I would say, again, that electrodes could remain at levels that are similar with the others.

Today, we are seeing signs of costs going down.

So those are -- those should be positives going forward.

Okay.

And on the idled capacity here, when exactly was the silicon metal capacity curtailed there?

And also, do you believe that the capacity curtailments are aggressive enough given the quarterly results today?

The curtailment is taking place as we speak sequentially, depending on various operational restrictions such as existing raw materials that we want to consume or labor regulations.

So they are happening sequentially.

We believe that the kind of capacity we will be running in 2019 is the one we need for the sales we are expecting, given what we see in terms of demand.

So -- and the only thing we are doing is, ensuring that, that production is run in the most efficient way, assigning production to the least cost -- to the lowest-cost production units.

And that is what we're doing and yes, we think the curtailments are enough, with our expectations of demand as we stand today.

Of course, we always say, we're ready to do more if needed, but we don't see that need as of today.

Okay.

One last question if I could.

Was there some recent news maybe today or overnight on refinancing for the parent company, Grupo Villar Mir?

I have seen that there is announcement out there.

I don't know more details than what the announcement is saying.

So I really -- I don't have further information.

Our next question comes from Vincent Anderson of Stifel.

So when you look at the potential switch silicon metal to ferroalloys and the margin's holding up pretty well there.

Is there any discussion internally about maybe writing that business more to maximize cash flow in the near term, rather than hitting a return requirement, particularly just given that you have this curtailed capacity sitting on hand?

We are certainly looking into Q4.

We are certainly very focused on generating cash flow.

And when we have higher-than-required inventories, well, the cost of producing those inventories is already incurred.

So we're looking at following demand in the different product categories.

We're seeing strong demand on manganese alloys, and we think that when you look at Q3 plus Q4, you will see that volumes are at expected levels.

And the issue, of course, right now is margins, but we will certainly be looking at generating cash through depleting inventory levels.

On the silicon alloys, in general, it's not only for silicon but also the other silicon alloys, particularly, foundry products.

We are running at full capacity, and we are selling what we produce.

So in that regard, both in terms of cash flow generation and in terms of our financial results, we think we are -- we're doing what we need to do.

And on silicon metal, as I just explained, we are adapting our production capacity, of course, to where we see our demand and our sales.

Great.

And I know you can't comment on Grupo Villar Mir's refinancing, but it is public knowledge that their balance sheet is certainly in better shape.

Given where shares of GSM are trading now, I mean, pennies on the replacement value of your assets, what is the motivation for Grupo Villar Mir keeping Ferroglobe a public company right now?

That is something I cannot answer.

We are a public company.

We're a traded company, that is what we are today.

And of course, we are devoted to all our investors as a company.

All right.

Fair enough.

And just turning back to manganese alloys briefly.

I'm just trying to work out the math on the changes in the total capacity.

You're running at somewhere just south of 400,000 tons per year.

The slide deck shows 552,000 tons of capacity.

I know that number changes depending on the product mix, but is that really an optimized utilization rate for those assets if we're going to stay at around 400,000 tons on 550,000 tons of capacity?

Or are you basically just anticipating enough of a recovery in margin that you don't want to curtail too aggressively at this point?

A couple of answers to that.

One is that as of today, what we're looking is that curtailing capacity only in Q4, and we will see how things develop into Q1 and further.

There are some signs of recovery in margins there.

And as I was saying, in terms of demand and volumes, we see strength.

So that's the first.

And the second is a technicality that I have explained sometimes, which is that the nominal capacity.

So if we look at total nominal capacity of 664,000 in Slide 28, there is an actual real capacity, which is significantly lower than that due to power price arrangements mainly in Spain.

So we have always said that our natural maximum production capacity is somewhere between 500,000 and 550,000 tons in total.

That is the maximum, even with no curtailments and again because of technicalities in terms of how our power contracts are arranged in Spain.

Our next question comes from Sarkis Sherbetchyan of B. Riley FBR.

Looking at the EBITDA reconciliation table, it seems like the bulk of the deterioration was from the manganese alloys business.

And it sounds like, from your comments, you expect that business to eventually normalize.

Can you just give us the puts and takes on what normalization means relative to the ore of capacity that's going to come online and what you expect prices should be doing in light of your actions to cut manganese production?

Well, the actions to cut manganese production, of course, is more for us to adapt to actual sales and to the fleet inventory levels.

So I cannot comment on that from any other point of view.

But I mentioned during the presentation that this is a business that -- and you can look at the numbers that we are disclosing.

This is a business that should be running at around $600 million revenue per year and this is a business that over the 25 years we have been in it, has been giving margins that are similar to our other products.

And one could say a normal EBITDA margin in this business is around 15% and just being cautious, we say, let's say, between 10% to 15%.

So the maths are relatively simple in terms of what is a normalized level of EBITDA from this product, which is again, between $60 million and $90 million, is something that should be normal.

Apart from the spreads and the manganese ore prices, what has been hitting significantly this business during 2018 is both met coke and power prices that we have also -- power prices in Spain -- that we have also mentioned and disclosed in that same page.

So all of those are factors that need to return to normal levels.

Understood.

And with respect to the silicon metal production curtailment, if we kind of look at the first 9 months of the year, maybe there's some aberration on the EBITDA level generated, but given where you expect to take capacity for silicon metal specifically, what would you maybe call the new EBITDA generation of that business with that production cut?

I mean, we don't give specific guidance, of course, in terms of what is EBITDA level.

If you look at Q3, silicon metal business margin, EBITDA margin is around 15%, almost.

That is something, again, that is not exceptional in this business.

Now in terms of the curtailments we have announced, will that have a very significant impact on our sales volumes?

I don't think there will.

If we think about depleting stocks and our flexibility to ramp up production if we need, we could be selling as much volume as we are selling in 2018, if the demand is there.

So I would say, we're just taking a cautious stance and making sure that we are not overproducing in 2019, but the flexibility is there to produce further if needed.

Got it.

And kind of bringing everything home with regards to the curtailments and the fact that you guys have built up quite a bit of working capital, you did talk about the cash flow levers that you'd like to pull, how much cash can you harvest from working capital, say over the next year?

I think most -- and I'll turn to Phil.

Most of the working capital we hope to harvest, as you say, in Q4 and then, beginning of 2019, where it may be just converge to normalize levels, but Phil, additional color on that?

I mean, I would say, if you look at the level of capital that was built above what we would expect in the recently acquired manganese plants, at the moment that sits $35 million above normalized levels.

We built $68 million in finished inventory in the remaining plants.

So there's sort of -- there's a $100 million there that is, of course, where we would expect.

And as you turn up or turn down different plants, clearly, there's a level of inventory that sits in each facility.

So if you chose to turn down a plant, then clearly, some of that working capital associated with that plant would also come back.

So there is significant levers there for us to pull in terms of cash flow.

Thank you.

And ladies and gentlemen, this does conclude our question-and-answer session.

I would now like to turn the call back to over to Pedro Larrea for any closing remarks.

Thank you, and this concludes our Q3 earnings call.

Thanks, again, for your participation.

We look forward to hearing you -- to hearing from you on the next call and on meetings we may be having with many of you.

Have a great day.

Ladies and gentlemen, thank you for participating in today's conference.

This concludes today's program.

You may all disconnect.