Executives
Joe Racanelli - Director, IR & Communications David Pathe - CEO Andrew Snowden - CFO Stephen Wood - COO
Analysts
Don DeMarco - National Bank Financial Greg Barnes - TD Securities, Inc.
Operator
Good morning, ladies and gentlemen. Thank you for standing by.
Welcome to the Sherritt International Corporation Fourth Quarter 2017 Results Conference Call and Webcast. [Operator Instructions] I would like to remind everyone that this conference call is being recorded today, Monday, February 12, 2018 at 10:30 AM Eastern Time.
I will now like to turn the conference over to Mr. Joe Racanelli, Director of Investor Relations and Communications.
Please go ahead.
Joe Racanelli
Thank you, good morning, everyone. And before we begin, I'd like to remind everyone that we will be following the presentation that is available from our website, sherritt.com.
We will also be making a number of forward-looking statements, and the risks associated with these statements are detailed in our presentation. With me is David Pathe, CEO of Sherritt; Andrew Snowden, Chief Financial Officer; and Steve Wood, Chief Operating Officer of the company.
In the interest of time, we're going to restrict our remarks and we'll focus primarily on our Q4 results and performance. Please go ahead, David.
David Pathe
Okay, hello, thank you, Joe, and good morning, everyone. Thanks once again for joining us.
Quite a lot going on at the moment, so a number of things to cover from Andrew, Steve and I but we'll try and get through those and then take your questions. For the quarter, obviously the closing up of the restructuring joint venture was the biggest highlight from our perspective of the quarter and with that the elimination of the $1.4 billion in debt really is the key highlight.
That restructuring really is the culmination of over two years of effort that began back with the financial completion on the project financing down there in October 2015 which enabled us to stop funding the project with 40% work, led subsequently to the renegotiation of the deferral and the amortization schedule of the project financing down there and ultimately the renegotiation of the rest of the capital structure with our partners to get to an ownership level that more closely aligns with our economic interest. So it was a big accomplishment for us to -- a big part of our priority is for 2017 and it was nice to get that in the books.
Other highlights for the quarter include our net direct cash cost at Moa which was the lowest since Q3 of 2004, at $1.80 a pound. As of third consecutive quarter where Moa is in the lowest quartile relative to our peers.
Higher commodity prices are helping us out obviously on that but so is the benefit of that asset plant that's still delivering about $0.50 of pound and cost savings is really what's making the difference at our net direct cash cost at this point in time. Higher commodity prices are obviously helping us in Moa and from the cash flow perspective to that end we received just under $20 million in cash flow from the Moa JV in the quarter.
That's on the back of the commodity prices I mentioned; nickel volatile and we'll look at it a bit more closely in a moment but it was up on an average for the year, up about 7% to 5% to in a quarter. Cobalt is a big part of the story in the last 12 or months, cobalt was up 134% quarter-over-quarter, Q4 over Q4, it's at $31.60 a pound.
We're seeing some strength on the oil side as well with the GCF6 which is the reference price we used in our oil business, it was up 28% over Q4 in the previous year to almost $53 a barrel. We here continue to be focused on trying to strengthen our balance sheet and then the improving marketing fundamentals that can help us do that.
We'll get into more -- into the Q4 results in a moment and Andrew will take you through to some of that as well but I want to spend a couple of minutes just looking at the market environment; that's certainly been part of the story for us in the last couple of quarters and seems to be continuing into 2018. Slide 6 in the presentation shows you what nickel and cobalt prices did over the course of the year.
Nickel, as no surprise continues to be volatile; we saw strong start to the year but then decline through Chinese New Year and really took into the summer before we start to see it recover. Nickel bottomed out at about $4.25 a pound in July and then with some volatility you can see in the second half of the year it did definitely have a trend upwards and actually touched $6 a pound briefly earlier this year.
The graph on the right illustrates the trajectory that we've seen in cobalt, that 134% is reflected there from $15 or so where we started the year to well over $30 and that trend has continued. Just to put that in their perspective, going forward here every dollar in the nickel price and I think this is ignoring [indiscernible] for the moment but just cash flow out of Moa, every dollar in the nickel price is worth almost $40 million a year in cash flow out of Moa and every dollar on the cobalt price is worth almost $4 million out of Moa.
Over on Page 7, you see what's driving the nickel price and cobalt prices; there are looming concerns about particularly for Class 1 nickel in the years ahead, and certainly concerns about security of cobalt supply and adequacy of cobalt supply, particularly that the majority of the world's cobalt supply comes from the DRC. Driving those concerns and the supply concerns really is this electric vehicle theme and the pace of investment in electric vehicle seems to be accelerating.
Ford, for example, at the moment -- at recent to try to honor a show [ph], announced an additional $11 billion investment in electric and hybrid vehicles and committed to having 40 new hybrids and electric vehicle models by 2022. Other volume makers like Volvo and Volkswagen have made similar investments as well.
I think this is all very positive for the industry, it's certainly driving demand for cobalt and we think in time will continue to drive demand for Class 1 nickel and we do continue to expect to see that differentiation between Class 1 pure nickel products that are more amendable for battery manufacturing and nickel contained in iron type products that differentiation and pricing should continue to emerge in the years ahead here. On Page 8, I want to just touch briefly on battery composition and what that means for cobalt and nickel supply.
Certainly a lot of effort is being expended on the cobalt content in batteries, driven in part by higher cobalt prices and these security and supply concerns. Here you see an illustration of the composition of the cathode and the most common type of battery used for electric vehicles; the nickel or manganese, cobalt or so-called NMC battery.
Today the majority of those batteries are done with using cobalt, manganese and nickel in roughly equal proportions; these so-called NMC111 battery and research towards that is trending to try and up the nickel content to diminish the importance of cobalt. So while we do believe that there is substitution risk for cobalt in electric vehicle batteries, the most likely uptake of that engineering out of cobalt will be nickel because it's nickel that gives the batteries the ability to maintain that energy, stability and density; certainly the pace of that research and we've kind of tracked it out there today to 2025 but the pace at which that evolution from the 111 to the 811 composition for cathodes and electric vehicle batteries will play a big part in nickel and cobalt prices going ahead.
On Page 9, you see the cost curves that we typically share for you; this shows the industry cost curves cumulative for year-to-date for 2017. Our net direct cash cost of Moa for year-to-date was $2.35 a pound which actually came down just outside the lower -- the bottom 25 or bottom quartile on the cost curve.
That $1.80 though was well within the cost curve and was our third consecutive quarter inside the bottom quartile and that is read and out just the benefits of higher cobalt pricing but also the benefits of that asset plant. At Ambatovy, notwithstanding the production challenges there that we experienced through asset reliability and the cyclone; Ambatovy still put up a net direct cost near the 50th percentile, and certainly a lot of opportunities see that cost come down as we get to more continuous -- more constant production out of Moa.
I'm now going to let Steve touch on a few operational highlights before we have Andrew then take us through some financial highlights and I'll come back and talk about a couple of things that we're dealing with as we head into 2018 here. Steve?
Stephen Wood
Thanks, Dave, and good morning everyone on the line. As always, I will start with a few points on safety.
We continue to work on fulfilling our commitment to all employees going home safely every day without injury by implementing some fatality prevention standards, visible felt leadership program, and process safety across the site. As a result, we continue to see significant improvement in safety performance in the operations.
Unfortunately in Q4, we reported a fatal accident in Cuba involving an interim [ph] gas employee. Following an investigation by the authorities there the cause of the accident was deemed to be non-industrial and therefore it's not classified as a fatality.
That said, we've done our own investigation and did find some areas for improvement and we continue to work on those diligently to fulfill our commitment. Also as previously announced, Cyclone Ava landed in Ambatovy in January.
I would like to make a point now of thanking the people at Ambatovy for their efforts before, during and after the storm which resulted in no serious injuries and damages being kept to a minimum. I'd like now to turn to our production results and I'll start with the Moa joint venture.
On a 50% basis, Moa produced 4,134 tons of nickel and 465 tons of cobalt in the quarter and that's up 9% and 22% respectively from the same quarter in 2016. The higher number is due to the fact that in Q4 of 2016 production was impacted by Hurricane Matthew.
I would like to point out that production totals in Q4 2017 also allowed us to reach our guidance that we had issue for 2017. Having said that, Q4 2017 was impacted by abnormally heavy rainfalls that limited mining and required a dry down of inventory stockpile.
And now over to costs; NDCC was $1.80 per pound and that's the lowest we've experienced since Q3 of 2004. The low NDCC was driven by cost savings from the third asset plant as Dave mentioned earlier and a cobalt credit of about $3.54 per pound.
I turn over to the Ambatovy slide now; I point out that nickel production in the quarter was 3,111 tons while cobalt production was 245 tons. As you can see from the slide -- that being Slide 12, these totals were down from our results for Q4 of 2016.
Let me put this into perspective; first, I would like to point out that our production in the quarter -- in quarter four of 2016 represented the highest quarterly production ever at Ambatovy. And second, the production in Q4 of 2017 was impacted by reliability issues at the asset plant that are being addressed.
I should also point out that our production totals at Ambatovy are presented on a 40% basis through December 10 and 12% thereafter. This reflects the impact of the restructuring of the Ambatovy JV on our ownership and resulted in a modest impact on our production totals for the quarter.
Despite the reliability issues experienced at Ambatovy, NDCC was only moderately higher in Q4 compared to the same quarter in 2016, higher maintenance and operating costs in the quarter were largely offset by the cobalt byproduct credit of about $2.66 a pound. Now turning over to our oil and gas operations on Slide 13; we produced just over 6,100 barrel of oil equivalent per day on a net working interest or approximately about 25% lower than the same quarter in the previous year.
The decrease was due mainly to two factors; the natural decline of the mature fields and the exploration of the Varadero West production sharing contract in November. As one would expect, this decrease in the number of barrels produced had a negative impact on cost and our unit cost for the quarter were $12.24 per barrel and up from the $10.95 per barrel for the same quarter in 2016.
Now over to the power highlights on Slide 14; we've produced 201 gigawatts of electricity in Q4 and that's down marginally from Q4 of 2016. The lower number is due to reduced supply of the natural gas that is used to produce the electricity.
Despite the decline in production, we were able to lower our unit cost per megawatt of electricity from $24.73 to $23.43, partly due to the deferral of some maintenance activities. Now before I hand over to Andrew to talk about the financial highlights, I'd like to review a couple of developments subsequent to year end.
At Moa, the heavy rains of November and December have continued into Q1 of this year and that's limiting our ability to mine and resulting in a dry down from our ore stockpile. This impacts the mixed sulfide production for the quarter, however, I'd like to point out that we did take this into account when we put together our guidance for 2018.
Now turning over to Ambatovy, as previously disclosed, this first quarter has been impacted by Cyclone Ava and that's a category two hurricane that occurred in January. Ava caused damage to equipment and resulted in a temporary halt in production.
We've since completed the number of repairs and have restarted the asset plant or one of the asset plants and production is now at about 50% capacity and we expect this to ramp up through the end of the second quarter. As we did with Moa, the forecast for Ambatovy for 2018 reflects the impact of this cyclone.
That concludes my remarks on our operational results and I will now hand it over to Andrew who will review the financial highlights. Andrew?
Andrew Snowden
Thank you, Steve, and good morning, everyone. I'd like to begin my remarks by discussing the impacts of the Ambatovy joint venture restructuring on our financial results.
Turning to Slide 16, I'll just make a few comments on the impact of this restructuring on our balance sheet and this slide clearly highlights the changes to our balance sheet from September 30 to year end December 31. The most significant impact which is highlighted on this slide is the elimination of approximately $1.4 billion of debt, the balance is $1.3 billion at September and that climbed to $1.4 billion by the time restructuring ended or was finalized in December.
And this elimination of debt was a key driver in our decision to restructuring the Ambatovy joint venture and reduced our earnings shift from the 40% now, 12% interest. This slide also highlights how significantly we are strengthening our balance sheet while at the same time by maintaining exposures to long-life [ph] nickel and cobalt asset through our 12% ownership interest in Ambatovy.
The total cost of the restructuring to show was approximately $70 million and this included a catch up of 12% cash cost and we've declined some mix in financial completion in September 2015. There were also transaction costs and an additional $12 million that we're holding in escrow for future cash costs and that $12 million is currently disclosed as restructured cash as you can see that on the balance sheet on this slide also.
In addition to the elimination of debt, the restructuring has also impacted a number of other items on our balance sheet including reducing our investment in Ambatovy by about $480 million and reducing our Ambatovy subordinated loans by around $560 million and both of those items reflect a change in ownership interest from 40% to 12%. And in addition, that was around $270 million of unrelaxed foreign exchange which is being reclassified out of accumulated other comprehensive income on our balance sheet.
Turning now to Slide 17; I just wanted to touch on the impact of this restructuring house on our income statement. So the results of all of those changes in our balance sheet has led to a gain of around $629 million which impacted our Q4 and year end 2017 results.
And this gain primarily arises because of the liabilities that were eliminated from our balance sheet, a larger than the assets that were reduced from our balance sheet for the larger liabilities and there have been eliminated results from this gain. Concluding this gain, Sherritt's reported net earnings for the fourth quarter was $553 million or $1.85 a share and the Ambatovy restructuring gain was a one-time non-cash item this slide also presents our adjusted earnings which eliminates that one-time gain, I mean our adjusted earnings were around $50 million or negative $0.17 for the quarter.
So turning to Slide 18 now; I presented here an update on our Cuban energy receivables. As you can see from this slide, during the quarter we received around US$7.5 million for the quarter and this was a lower amount than we drew during the quarter and some of the results of that our overdrew balance increased from US$100 million at Q3 to US$133 million at year end.
Q4 is typically a liked quarter for collections but Q4 was also impacted by Hurricane Irma and the impact that had on the Cuban economy. By reducing the amount of overdue receivables is a strategic priority for us but is important to remember this amount has always fluctuated over the 25 years we've operated in Cuba and it's been at levels much higher than it is today.
During this time we've never had any disputes on the amount said and always recovered these amounts. So there is few comments I'll make on our consolidated cash position now on Slide 19; this is our typical waterfall slide which shows the movement in our cash balance from December 2016 through to December 2017 and highlights our year end cash equivalents and short-term investment balance of $203 million at year end.
The main use of cash during the year is $70 million Ambatovy restructuring payments I referred to earlier on. In addition, the typical interest that we pay on our outstanding debentures of $57 million is highlighted here as well.
These uses of cash are partly offset by positive operating cash flow of about $50 million from our oil, gas and power operations. And also during the year we received $32 million of advances from Moa and $20 million of that was in the fourth quarter as they refer to earlier.
The net cash received from Moa reflects the ability of the Moa joint venture to generate surplus cash flows in the current commodity price environment and as long as that continues we all expect ongoing cash coming out of Moa through 2018. Just heading for Slide 20; I just wanted to remind everyone on the call of our production unit cost and CapEx guidance for 2018 which we released last month in middle of January.
The production numbers here are presented at the midpoint of our -- of the ranges and also based on our ownership interest which is 50% from Moa and now 12% from Ambatovy. And as Steve mentioned earlier, these guidance numbers do include and have factored in some of the production issues that we're seeing and expecting through the first quarter of 2018.
In addition on the CapEx side for the outlook, the [indiscernible] is highlighted as our oil and gas CapEx is dependent on and tied to our rate of collections on overdue receivables, so we will be looking at the amount that we'll spend on oil and gas CapEx as being dependent on the receivables which we hope to get through the course of 2018. The final put on my commented slide is just in relation to our net direct cash cost or unit costs at Moa and Ambatovy.
Again, these numbers here represents the midpoints of the range and they are all heavily impacted by market pricing of byproducts and other commodity prices which I've tried to explain in a bit more detail on the next slide, Slide 21. Here you will see the midpoints of our guidance range for Moa's NDCC of $2.75 a pound, and that number was forecasted based on assumed cobalt price in 2018 of $30 a pound and also sulphur price on a landed basis of $200 a ton.
You will see there is some sensitivity but clearly the cobalt price is a key sensitivity to our NDCC and for every dollar change in the cobalt price, that will impact our NDCC by approximately $0.10 a pound. If we were to look at today's spot pricing of both cobalt and sulphur, so I think cobalt's in the region of US$36 a pound and sulphur pricing on a landed basis is approximately $170 a ton.
And if we were to utilize those assumptions and assume they hold through the course of 2018 then that would reduce our NDCC expectation at Moa from $2.75 to about $2 a pound, so you can see that some of the percentage have achieved but might occur through the course of 2018. That concludes my remarks and review of our financials.
So now I'll turn the call back over to Dave for his closing remarks.
David Pathe
Hi, thanks, Andrew. Couple of things for me, just has been in 2018, it's been a busy start to the New Year for us here as well and so there are a couple of things I want to touch on before we take your questions.
Couple of things first in oil and gas; you have seen that we've -- just recently we were able to execute an extension to our production sharing contract at PUMRE [ph] in the oil business in Cuba; that contract was originally scheduled to mature and the production revert back to the Cubans in March of this year, it has now been extended out to 2021 and that production was reflected in our 2018 guidance. I also want to just touch on where we stand in our Block 10 drilling.
You've seen our announcement just prior to the holidays that we're taking a break from drilling there to reassess how we were going to go about getting down to the target reservoir. The issue there is the same issue for those that you've been following the Block 10 drilling over the second half of last year is around these own we were losing circulation.
And our first attempt to get through those loss circulation zones resulted on us having to suspend when we started to lose stability in -- the geotechnical stability in the well bore. Just to backup a little bit; loss circulation is not an uncommon phenomenon for us in the drilling that we've done in Cuba over the years but historically most of the time we've been able to overcome those loss circulations just my pumping sea water down the well hole and restoring enough pressure to resume circulation on the drilling margin then get our casing it [ph].
Block 10, the first to go around, obviously that sea water wasn't able to do it and we took a break and then started the last side track once we brought into the country a number of different compounds, including things as diverse as rubber platelets and walnut shells which are widely used in the industry for over comfort healing fractures that lead to loss circulation. The results of that however though and over the course of December were that we weren't able to sufficiently heal those loss circulation zones as sufficiently to be able to restore circulation and get on with the drilling and get our casing in.
And after a few attempts of that we decided to take a break and step back and look at what other options are available to us. We have since then been exploring several different options and we're probably a couple of weeks away from making a filed decision here.
There are some other technologies that have been used in other parts of the world quite extensively that historically we have not had access to impart been the embargo -- what we've identified some technologies that are being used out of the UK and have been used in the North Sea and elsewhere; I mean we hope that we'll be making a decision on that in the next couple of weeks that will most likely involve a new side-track and what we did last time, so backing upto about the 3,300-meter level where we had the wider diameter casing set and breaking out the casing and growing out this again with some new technology in the form of casing that we think can actually help us overcome these zones. But ultimately we'll have the decision on that in the next few weeks here and their expectation is it will take some time to get those supplies on the island and then we should be drilling again, probably in April or sometime in a couple of months with the drilling before we can start sharing with you the results.
At the moment over the last few weeks and ongoing is the technical work to assess the viability of the solution with the different vendors and to pricing out so we know what we we're talking about. We'll have more definitive cost numbers for in future when a final decision is taking here and that work is ongoing but the capital guidance that you saw for the year for us in oil and gas is certainly designed to contemplate finishing this hole and the second hole on Block 10 later this year.
Well, actually I want to touch on our most recently -- our corporate capital structure activity. We recently completed our first equity raise in more than 10 years, raising gross proceeds of $132 million.
The offering consisted of shares and warrants, the warrants are linked to the price of cobalt, that was a very innovative structure that made possible as a successful offering and we are really quite pleased with the overwhelming interest and support that we received from investors around the world. The offering was executed to help us continue to pursue our strategy of deleveraging and strengthening our balance sheet, over the last few years we've reduced our total indebtedness by most of $2 billion and this activity is designed to continue in that direction.
Sentiment around nickel and cobalt continues to improve, the biggest impediment to new investor interest in Sherritt continues to be the extent of our leverage and we're determined to manage that issue. Actions like this demonstrate that determination and make Sherritt an even more attractive opportunity for new investors.
So consistent with that deleveraging objective, concurrent with the offering we also launched the modified Dutch auction tender process to repurchase some of that outstanding debenture debt. The bidding process for the Dutch auction culminates tomorrow night, so we'll have a better idea where we stand by then but thus far we've been quite encouraged by the level of interest expense expressed by bond holders and by tomorrow evening we'll have a sense where we stand and you'll hear more from us later this week as the Dutch auction process comes to a close.
So that is what we wanted to cover for you this morning. And with that, operator we'll take any questions anybody may have.
Operator
[Operator Instructions] And we'll move to our first question from the line of Don DeMarco of National Bank Financial. Please go ahead.
Don DeMarco
A question on direct cash costs; so I see that if you were to update your guidance at Moa to use prices for cobalt and sulphur the cost come down quite a bit to $2. What would the cost at Ambatovy be rather than the guidance cost if you were to similarly update for current spot prices of cobalt and sulphur?
Stephen Wood
I don't have a sense of the sensitivities at the top of my head. I'm looking to see whether Andrew does or not; I mean obviously, NDCC at Ambatovy is a function of production as well but Andrew, do you have a feel for the sensitivity?
Andrew Snowden
Yes, it would be around $0.50 per pound lower than the guidance we disclosed.
Don DeMarco
So here the midpoint is about $3.25, so we'd see that coming down to about $2.75?
Andrew Snowden
Correct, yes.
Don DeMarco
So another question on Ambatovy cash costs ; so I see that in Q4 you had $10.5 million cash costs, what do you expect for 2018? Is there any kind of cyclone impact or anything?
What can we put down for this?
David Pathe
Our expectation is that -- and obviously we're getting some help from the commodity prices on this at this point in time but our expectation is still notwithstanding the cyclone is that any further cash costs that are required for Ambatovy will be able to be funded out of the money that we've set aside for the closing until -- for our 12%. I mean it does, a 12% rate rather than a 40% rate grows upto a decent amount of cash for Ambatovy and with the cash levels -- with the net direct cash costs where they are, there is positive margins to be made at Ambatovy and the improved significance if we can catch up with some production on the second half of the year.
So that is currently -- our expectation is that we're not anticipating any further cash going into Ambatovy off our balance sheet beyond what's in the escrow account.
Don DeMarco
And I think there is about $12 million in the escrow account?
David Pathe
Yes, that's right.
Don DeMarco
So you're not anticipating any more than $12 million in cash costs in 2018? And any event that there were cash calls, would you have to top-up the escrow account?
Do you have to maintain a certain level in the escrow account?
David Pathe
No, there is no obligation for us to fund anything further into the escrow account, that was just -- that figure was part of the deal in the restructuring and there is nothing that obligates us to continue on further cash costs.
Operator
[Operator Instructions] And we'll move to our next question from the line of Greg Barnes of TD Securities. Please go ahead.
Greg Barnes
All else being equal, what does the production profile looks like at Moa over the next five years or so, both in terms of nickel and cobalt?
David Pathe
All things being equal, we expect to see production being pretty consistent over the next five years and there is -- the hole distances in that to be managed in terms of what that will do to net direct cash costs but we're not expecting to see any and I'm looking over to Steve to see whether he disagrees with me on this but I'm not -- I don't think we're expecting to see any material variations in the production over the next few years.
Stephen Wood
No, we don't expect any material variation. On the JV, the increased hole distances will be offset by the fact that we'll be getting some new equipment in there that will help us out.
Greg Barnes
The caution should remain more or less the same as well is what we're saying. Thank you.
Operator
[Operator Instructions] We'll move to our next question from the line of [indiscernible]. Please go ahead.
Unidentified Analyst
Can you speak to the level of interest that you're getting from auto makers or other parties that are potentially trying to enter into long-term contracts for cobalt sales? I'm just interested in the current level of interest you're seeing.
Thanks.
David Pathe
There has been a lot made of that in the media and companies like Volkswagen have tried to run processes to secure amounts of cobalt, BMW is in the process of doing that as well. We have participated in some of those conversations, mainly because so far the Cuban origin of some of our nickel out of Cuba, some of that is -- in some cases less attractive for higher profile uses of cobalt.
But thus far we haven't seen any advantage in doing that, we get market price for all of our cobalt products as we have done for many years and we sell a lot of it to -- under long-term relationships with battery manufacturers, primarily in Asia. So we haven't as of yet seen any compelling reason to commit any significant volumes, most of our volumes are committed on an annual basis but they are committed on an annual basis to long standing customers and it's all priced off at market rates and we haven't seen anything that would suggest that there is any better way for us to maximize our net backs [ph] in doing that.
Operator
Thank you. It appears that there are no further questions at this time.
I'd like to turn it back to management for any additional or closing remarks.
David Pathe
All right. Well, thank you very much once again everybody for joining us.
You will hear more from us later this week as our Dutch auction process for our debt repurchases comes to a finish and we will speak to you all again when it's time to discuss Q1 results in a few weeks.
Operator
Thank you. This concludes today's call.
Thank you for your participation. You may now disconnect.