GrafTech International Ltd. (EAF) Q1 2019 Earnings Report, Transcript and Summary

Good morning. My name is Julie, and I will be your conference operator today. At this time, I would like to welcome everyone to the GrafTech International First Quarter 2019 Conference Call. All lines have been placed on mute to prevent any background noise. After the speakers’ remarks, there will be a question-and-answer session. [Operator Instructions] Thank you. I will now turn the call over to Meredith Bandy, Vice President of Investor Relations and Communications. Meredith, you may begin your conference.

Okay. Thank you, Julie. Good morning, and welcome to GrafTech’s first quarter 2019 conference call. On the call with me today is GrafTech’s Chief Executive Officer, David Rintoul; and Chief Financial Officer, Quinn Coburn. Turning to our first slide. As a reminder, some of the matters discussed on this call may include forward-looking statements regarding among other things results, performance and strategies. These statements are based on current expectations and are subject to risks and uncertainties. Factors that could cause actual results to differ materially from those indicated by forward-looking statements are shown here. We will also discuss certain non-GAAP financial measures, for which you will find reconciliations in these slides. The slides are posted on our website at www.graftech.com in the Investors section. For your reference, a replay of the call will also be available on our website. And now, I’m pleased to turn the call over to Dave.

Thank you, Meredith, and good morning, everyone. I will begin the call with a discussion of our safety performance, first quarter highlights and commercial strategy. Quinn Coburn will then cover the financial aspects of the business. I will then conclude our prepared remarks with a discussion of the steel market, after which we will open the call to questions. With that, I’ll begin, as we always do, with safety. We recognize the need to remain focused upon the safety of each and every team member. Our year-to-date total recordable injury rate stands at 1.32. I’m pleased to see that the trend is moving in the right direction, but we also know that our focus must remain on continuous improvement of our safety performance, driving towards our ultimate goal of zero injuries, that means every worker going home safely every day. I would like to take this opportunity to thank our team for their continued focus on our safety mission, while also delivering solid operational and financial results. During the first quarter of 2019, GrafTech delivered net income of $197 million, or $0.68 per share and adjusted EBITDA from continuing operations of $284 million. Our first quarter weighted average realized price for GrafTech manufactured electrodes was $9,954 per metric ton, roughly in line with the prior year quarter and previous quarter pricing. To date, GrafTech’s short-term and spot UHP pricing has averaged about $11,800 per metric ton for Q2 2019. Q1 production increased by 12% from the previous year quarter with the completion of our debottlenecking projects. Capacity utilization in the quarter was 94%, in line with our previous expectations. The finishing operations at our St. Marys plant are running to support overall flexibility within our manufacturing footprint. Market conditions don’t suggest the need for restart for additional capacity anytime soon. GrafTech is substantially vertically integrated into petroleum needle coke with our wholly-owned Seadrift facility. This competitive position differentiates us from our competitors and gives us a secure, low-cost, high-quality supply of petroleum needle coke. Our previously announced efficiency improvement at Seadrift is proceeding on time and on budget. As a reminder, Seadrift’s 2019 production is expected to be about 125,000 metric tons in 2019. That’s up from the 111,000 metric tons we produced last year since we do not have a planned outage in 2019 and we also expect to see some of the benefits of the efficiency improvement project later this year. On average, Seadrift produces more than two-thirds of our long-run needle coke requirements. This vertical integration aligns with our commercial strategy to sell electrodes on long-term contracts. Turning to Slide 5. As a reminder, GrafTech has over 70% of our 2019 production capacity sold under long-term take-or-pay contracts at weighted average prices of about $9,800 per metric ton. These contracts provide profitability and visibility of earnings. I’ll now turn it over to Quinn on Slide 6 for more details of our financial results.

Okay. Thanks, Dave. As Dave mentioned, we’re pleased to report another strong quarter of financial performance. First quarter net sales of $475 million were up 5% from Q1 on higher sales volumes. During the quarter, we sold 45,000 metric tons of graphite electrodes, up from the prior year period due to benefits of our debottlenecking. GrafTech’s first quarter average realized price was $9,954 per metric ton, in line with the prior year quarter. Approximately 83% of our first quarter net sales were to customers with long-term agreements. As noted on the slide and in our earnings release, reported sales volumes and weighted average price now include only GrafTech manufactured electrodes. As our manufactured products contribute the vast majority of the company’s gross margin, this better reflects the way we manage the business and report internally. Net sales on the income statement also includes byproduct revenues and resales of some third-party manufactured electrodes. The margin contributions from these revenue streams is minimal. Now turning to Slide 7 for financial results. First quarter 2019 net income totaled $197 million, or $0.68 per diluted share. Q1 2019 adjusted EBITDA from continuing operations was $284 million. Compared to the prior year quarter, higher sales volumes were more than offset by higher raw materials costs, specifically related to third-party needle coke costs. These higher third-party needle coke costs continue to impact the cost of sales. We estimate that increased third-party needle coke costs will increase our Q2 2019 cost of sales by approximately $7 million compared to Q1 2019. First quarter 2019 free cash flow increased to $142 million from $127 million in Q1 2018. Free cash flow in Q1 is not as high as it was in the fourth quarter, primarily because the majority of our annual cash tax payments are made in the first quarter of the year. In Q1, these tax payments totaled $61 million. Now turning to Slide 8. I’ll quickly walk you through the first quarter adjusted EBITDA from continuing operations reconciliation. We began with $197 million of net income, add back $16 million depreciation and amortization, add back $33 million net interest expense and $33 million GAAP income taxes. Finally, add back other adjustments of just $5 million, the largest of which was about $3 million of non-cash fixed asset write-offs. Now turning to Slide 9. GrafTech has a strong track record of shareholder returns. Since our IPO, GrafTech has returned 71% of free cash flow to shareholders. Capital allocation, since our April 2018 IPO includes $181 million in debt repayment, this represents the required amortization of our term loan plus almost $100 million of prepayments in Q1 of 2019, $94 million in regular quarterly dividends, $225 million share repurchases and $203 million special dividend. Now turning to Page 10. We expect continued solid results enabling us to balance the need for capital reinvestment, debt repayment and returns to shareholders. 2019 capital expenditures are expected to be between $60 million and $70 million, similar to last year. Our focus is on operational improvements to maintain productivity, quality and cost position of our asset base. We repaid $125 million in debt in Q1 2019 and expect to prepay additional debt in the second-half of 2019, as we work towards the long-term leverage and capital structure consistent with the nature of our business. We’ve continued to analyze our capital allocation strategy and have refined it to reflect a target of approximately 50% to 60% of 2019 free cash flow to be returned to shareholders in the form of dividends and share buybacks, of course, subject to Board approval. The balance of the free cash flow is expected to be used for debt repayment. This results in a somewhat lower debt repayment trajectory than we discussed on our last quarterly earnings call. Our regular quarterly dividend is designed to be in line with market dividend yields and be sustainable through the cycle. Further shareholder returns will likely be in the form of share repurchases, which are accretive to all shareholders and the tax-efficient distribution. Direct share repurchases also help address the equity overhang and preserve liquidity in our shares. I’ll now turn it back to Dave on Slide 11.

Thanks, Quinn. Graphite electrodes are vital to our customers’ operation and EAF’s steelmaking trends remain positive. In the United States, steelmakers have announced approximately 12 million metric tons of electric arc furnace capacity additions, which represent a potential 15% to 20% increase in U.S. graphite electrode demand by 2022. According to the World Steel Association, electric arc furnace share of global steel production is at approximately 28%. We expect electric arc furnace share to increase by just over 30% in 2019. While steel production ex-China is expected to grow modestly, total global steel production is expected to remain at levels similar to 2018, also according to World Steel Association. In summary, GrafTech is a leading provider of high – highly-engineered graphite electrode services, solutions and products to the growing EAF steel market. We delivered another solid quarter results in Q1 2019 and we continue to execute our strategy to deliver long-term sustainable value. Ongoing operational improvements and vertical integration give GrafTech economies of scale and a competitive cost structure. This in turn enables us to offer secure long-term contracts to our customers and strong earnings visibility. That concludes our prepared remarks. We’ll now open the call up for questions.

[Operator Instructions] Your first question comes from Curt Woodworth from Credit Suisse. Please go ahead. Your line is open.

Thanks. Good morning.

Good morning, Curt.

First question is just on, I guess, ASP progression to the business. You talked about being 70% sold on LTA for the year, but then I assume that you priced a lot of your shorter duration contracts through first-half of the year, even into 2019. So, can you give us a view on how you think, I guess, your ASP would trend sequentially going forward from the 1Q number?

So, as you know, we don’t provide guidance going forward into the future. What we can share with you is that, if you look at our smart pricing numbers as of today looking into our second quarter, we’re at that $11,800 per metric ton value and feel pretty good about that as a spot number for what we expect – like to unfold in the second quarter. Beyond that, Curt, I don’t think we would provide guidance on that beyond the second quarter.

Okay. So would it be fair to assume, I guess, for your spot book, which is roughly 30%, number close to that $11,800 number and then the 70% would be priced at the $9,800 number. So that would imply a decent sequential improvement in ASP, or are there other factors to think about in the 2Q versus 1Q?

I think, generally, you’re pretty close with that. I mean, there is a couple of orders that I’m aware of that we’re, if you will, cleaning up from some previous commitments that have an impact are not major in the scheme of life, but they might have a tendency to lower that by a very minimal amount.

Okay. And then in terms of capital allocation going forward, the 50% to 60%, would you, for example, in times like this where your stock price, I would think, given the value would be significantly more accretive to either buyback stock or pay special dividends out relative to deleveraging. Is that a hard and fast rule on the 50% to 60%, would you potentially look to go above that threshold for capital return in an environment such as this?

Hey, Curt, this is Quinn.

Hey, Quinn.

How are you doing today?

Good.

Good. Obviously, look, we will always reassess and discuss this type of thing with the Board and readjust as necessary going forward. But our target, and I emphasize our target is definitely 50% to 60% of free cash flow return to shareholders in 2019. So that’s – we wanted to set a clear target, we wanted to clarify that and be a little bit more clear about our capital allocation strategy for the year.

Great. And then just one final question on needle coke. Can you comment on pricing trends there? Is your sense that pricing or your needle coke cost would continue to be a headwind sequentially moving into 3Q? And have you – are you starting to see any stabilization in needle coke pricing?

Sure, Curt. So as we mentioned last quarter, we stopped giving guidance on exactly what our needle coke prices were, but we did last quarter indicate that we expected a sequential increase. We did have that sequential increase. Last year – last quarter, we indicated we thought it would be $13 million, it turned out to be just a little bit less around $11 million. And then in our remarks today, we indicated that we expected the sequential increase over Q1 to be about $7 million in Q2. Should flatten out a little bit going forward, and I’ll speak to that the specifics of Q3 and Q4 in future quarters.

Great. I really appreciate it. Thank you.

Your next question comes from David Gagliano from BMO Capital Markets. Please go ahead. Your line is open.

Okay, thanks for taking my questions. I just had a quick follow-up on the near-term. First of all, we had – looks like we had production of, I think, it was 48,000 tons roughly, shipments were 45,000 tons. So for the second quarter, should we expect that inventory build to reverse?

Sorry, David, could you repeat that question?

The question is, basically what I’m asking is, how should we model volumes for the second quarter – shipment volumes, because it looks like we had an inventory build based on the production versus the shipments in the first quarter?

Yes, that’s right. So we did have a production of 48,000 in the first quarter and shipments of 45,000, did have a slight inventory build, and you see that a little bit in our free cash flow, slight inventory build and overall working capital build. As Dave mentioned earlier, we haven’t given guidance for the future and haven’t given volume guidance, so we’ll refrain from any guidance. But you’re correct, if we were to sell at levels of our production, this could be a little bit higher in the following quarter.

Okay. And then just bigger picture, thanks for the update on the steel industry on Slide 11. Obviously, helpful to frame the demand side, quite a bit of consternation out there on the UHP graphite electrode market now. You gave us come commentary about spot pricing, I think, you said basically holding steady. But I’m wondering if you can provide a similar update on the UHP graphite electrode industry trends similar to what you provide on Slide 11 for the steel industry, what you’re seeing in your UHP graphite electrode industry right now?

So a couple of comments to address, I think, what you’re asking. We recognize that we added capacity in the latter part of 2018 going into the beginning of this year, couple of our competitors did the same. And of course, when electrodes were extremely tight in 2018, there were some folks that dabbled with some Chinese electrodes, because they couldn’t get them anywhere else. And I think all of that has brought us to a place in 2019, whereas we came up – came through 2018 with a pretty significant undersupply to the market, if you will, combined with, I think, people building some electrode inventory in their plants, because they got caught short. And we transition, as I’ve referenced in the previous chorus call into 2019, where the market is clearly in a balanced – in a more balanced position. So that’s why you’re seeing us in a spot – place where we’re talking about spot prices at $11,800, whereas two quarters ago, we were talking about numbers that were well in excess of $15,000 per metric ton. So absolutely the market is more balanced than it was and we would expect that to continue.

Okay. And if I can just ask one related follow-up, I guess. What are your internal and external sources telling you about the potential growth in China, UHP specifically, UHP graphite electrode volumes over the next one to two years?

Right. So we recognize there has been a lot of activity and attention to that subject matter over the last week to 10 days. And look, here is what we would say about that. We recognize that China clearly is adding capacity and they’re going to continue to grow, because their electric arc furnace industry is growing. It’s sitting at about just almost 10% today, not quite 100 million tons per year, with a trajectory to grow to 20% by 2020, 2021, that’s another 75 million to 100 million tons in round numbers. And so we’re talking about an industry in China that’s set to in the next couple of years basically double in its size and add the equivalent of the entire United States electric arc furnace steel industry in that short period of time. So no question they’re adding capacity to service that need. Now also recognize that when you speak about the Chinese competition, and I think everyone knows this, it clearly is an opaque information situation. And there is a number of the graphite electrode designations that are called UHP in China that we would consider more of a ladle-type product. There are, to be fair, there are a couple of Chinese producers that have been working for years to produce UHP melter type electrodes. And we have seen customers selectively purchase some of these electrodes on trials, and particularly last year when there were shortages. But I think it’s important to recognize, it’s taken the best of these couple of well-established Chinese producers years to get to a point where they have a UHP electrode that has a consumption rate and a breakage rate that people would trial. And for the most part, they’re not being trialed or used on high-intensity applications in melt shops that are demanding and high-production melt shops. The majority of the capacity additions that we’ve seen in China that are claiming they are bringing online UHP electrodes have no experience producing them. So it’s a bit unclear, clearly to take that announced capacity and translate how much of it actually reports to the market and how long it will take for them to get to a place where if they do actually build these plants and get them up and running, how long it will be before they can practically be considered viable competition, we think it’s sometime. So add on top of that the – there’s a wide disparity of information from various resources and sources coming out of China and the lack of clarity on what’s really UHP versus HP versus ladle. Lack of clarity on the needle coke situation, I think, everybody knows that the electric vehicle world, the big action is in China. They’re leading that activity and thus leading the consumption of needle coke type products for that – supporting that battery market. So having said all of that, I think, there is a recognition on our part that look, yes, they are building things. We don’t think all that’s announced reports to the market quickly, but we’re happy with our place with the long-term contracts that we have, the take-or-pay contract. We have a very diverse customer base. We have over 100 customers that are procuring products from us, and we know that we make high-quality products and services that provide a competitive advantage for our customers. And lastly, we believe our cost structure is best-in-class, and we’re capable of competing globally anytime, anywhere.

That’s great. Thank you. It’s helpful.

Yes.

Your next question comes from Arun Viswanathan from RBC Capital Markets. Please go ahead. Your line is open.

Great. Thanks. Good morning. I guess, just similar line of questioning. Do you have any reaction, I guess, to some of the recent reports on the consultant side that are kind of discussing increases in Chinese UHP capacity and needle coke? I mean, what is the capability that you’re aware of for them to produce UHPs? I mean, I know you just detailed some of it and your belief is that most of it’s ladle. But it seems that there is a little bit more UHP capacity there and that has impacted spot pricing. And if it continues to go that way, I mean, maybe you can just reiterate your confidence in the long-term contracting process, and what you’re hearing from your customers as far as willingness to pay higher prices in spot? Thank you.

Okay. Alex, [sic] [Arun] good morning. Thanks for your question. Look, I think that in the same vein or the same theme is what I answered in the earlier question, there, yes, are a number of, if you look in the ICC reports, you see a listing of people that supposedly are building what they claim to be plants and when they might come online. And we know that from previous experience, not just this particular announcement of additions, but over the last number of years that there are some degree of those that happen, some that don’t. And we would acknowledge that there is a couple and a definition of a couple is two, maybe three producers that have been at it long enough that they have produce a product that is getting trialed in certain areas. Sometimes the trials go okay, sometimes the trials are a dismal disaster. So we know that they’re not – even those that have been at it a while and think that they have a good product are not up to the standard of the kind of qualities that we or some of our ex-China competitors produce. Now having said all of that, it’s incumbent upon us to be prepared and ready for this competition and we’re doing that. And as I said, we believe we have a cost structure that is quite competitive and happy to compete on a level playing field at any day, anytime, any place in the world. I think there is a large number of those announcements that are from people that are entering the field in China, and they will have the same learning curve that a couple that I referenced that have been at it for years have been producing a UHP electrode. There is a heck of a difference between producing a ladle electrode with a very mild service duty and an electrode that’s good for a melt application, and particularly high production shops that require a grade A-type product. So we’re being prudent and managing our business accordingly. But I think, there is a number of those announcements that will take sometime to work their way out and come to fruition, and we’ll continue to advance our business accordingly.

Great, thanks. And just as a follow-up on the customer behavior. Appreciating the detail you provided in the slide deck on your next five years or so of contracting. Are you still seeing strong demand for the LTAs? And as a percentage of your volume, is there kind of a targeted range where you’d like to move all your volumes into contracting versus spot, or what kind of percent would you expect to keep in that level going forward?

So I think, we’ve been pretty consistent in our description of – we get two-thirds to 70% of our needle coke covered by Seadrift and we like how that matches up with our LTA business. And I think it’s a good arrangement not only for us, but we think it’s a good arrangement for our customers. So let’s bear – let’s not forget that when we signed these things in late 2017, you think through what happened in 2018, spot pricing was $20,000, $25,000 and even crazy $30,000 numbers that were floating around and people couldn’t get them. But our customers were sleeping well at night knowing that they had signed their LTAs with us at a fixed price and knew they were – they had absolute assurance that they were going to get their product and at a price that we had agreed to. And so I think those people look pretty smart right now. They saved a nice chunk of change for their company and weren’t running around last year with their hair on fire wondering whether or not they were going to be able to get electrodes. So I think those procurement managers look pretty smart right now.

Great, thanks. And just lastly, I appreciate the target of free cash flow that you plan to return to shareholders in 2019. Assuming that the long-term contracts hold for the next several years, is that a level that you would continue to target going forward 50% to 60% of free cash flow? And maybe you can just also discuss your uses of cash between returning capital to shareholders, deleveraging and if at all potential M&A or capital investment? Thanks.

Sure. So this is something that we would reevaluate on a year-by-year basis, certainly, but I want to emphasize our commitment to returning cash to shareholders, that remains a key priority. As far as allocation goes, we will manage the debt prudently, but shareholder returns remains the key priority.

Thanks a lot.

Your next question comes from Alex Hacking from Citi. Please go ahead. Your line is open.

Hey, good morning. Thanks for taking my questions. I have two questions. The first one, I guess, just coming back to the last question, asking it in a slightly different way, like this year you’re going to allocate a certain amount to reducing debt. How should we think about, what – I guess, what level of net debt are you comfortable with on the long term?

Sure.

Obviously, given the contract book, your current level of net debt is very comfortable, right? Wouldn’t necessarily need to decrease it. So I guess, on the long term, how do you think about it? Thanks.

Yes. That’s absolutely right, Alex. The current level of debt is very comfortable. The long-term target will really depend on how business evolves over time. How much of the LTAs we renew or how far we renew them in advance. So, so much of a long-term target really depends upon what happens three, four, five years from now. For now, we consider it prudent and in the best interest of shareholders to apply some portion of our cash flow to debt, but we specifically haven’t defined an exact long-term target for that very reason.

Okay, makes sense. And then, I guess, more of an accounting question. The $7 million of additional cost in needle coke, just to clarify, does that assume like the same amount tons of needle coke purchased or tons of electrodes produced, or is there some assumption that baked in a sort of higher volume as well?

You can think of that as roughly the same. It’s not materially different, given the range of volumes that we’re talking about. In the reasonable range, it will vary $1 million or $2 million at most, so you can consider roughly the same volume.

Okay, thanks makes sense. Thanks for the questions.

Thanks, Alex.

Your next question comes from Michael Gambardella from JPMorgan. Please go ahead. Your line is open.

Yes, good morning. I just wanted to drill down a little…

Good morning. Michael.

…good morning. I just wanted to drill down a little bit more on this whole threat from China. You’ve talked about demand increases on electrodes are pretty good as we’ve talked about from not just the U.S., but primarily from China, as you said, they’re looking to double where they are today over the next few years, which is significant increase. There is questions about just what type of electrode capacity is being announced in China, as you alluded to, a lot of people include ladle just as they would think it’s ultrahigh powered electrodes that you produce. So there’s some confusion there. But if you drill down even further, could you talk a little bit more about the petroleum needle coke that is needed in China and in the world? And how difficult that is to procure, because the raw material for petroleum needle coke is a byproduct in the refinery business. So it’s not as easy as just building a low-sulfur decant oil facility to feed raw material to your petroleum needle coke facility, which then feeds new ultrahigh powered graphite electrode capacity. What can you tell us since you’re in the business of petroleum needle coke as well as graphite electrodes, just some feel for how difficult that is going to be in China, the real petroleum needle coke that’s needed to compete with your product?

Right. Look, that’s a great question. I think, if you’re going to be in Tier 1 melter electrode business servicing shops in the manner that – excuse me, we do, you have to have a pin or a joining – a joint and the area around that joint in your electrode that can withstand the type of service duty that happens in an electric arc furnace melt shop. Those pins are to be of that nature generally are made with petroleum needle coke. And our understanding is that, the Japanese even today import some of that petroleum needle coke, which we believe those that – the handful that are – or the couple that I referenced earlier in my comments that know what they’re doing, that’s why they have to import some petroleum needle coke, so they can make their pins to be able to have half a chance to do anything. And your point is exactly right, making petroleum-based needle coke is not just as easy as though I’m going to build a plant and get some oil and away we go. We have taken years to develop not just our process at our plant, but the relationships that we have with specific decant oil suppliers, because like anything else in the world, decant oil is not decant oil is not decant oil. They all have various constituents that come with them, because you have to remember when an oil refinery is producing a product, their main goal in life is to produce gasoline and diesel fuel. They are not going through that process with a barrel of crude oil to make decant oil. They are doing it to make oil and gas – excuse me, diesel fuel and gasoline. And so the decant oil is a secondary product basically what’s leftover, no pun intended, at the bottom of the barrel when the day is done. And each of those refineries have different operations, and so the decant oil that comes out at the end of the day varies widely from refinery to refinery. So it is a complicated process. You have to put all these pieces together and know what the attributes are of the refinery that you’re doing business with and then how your process has to react to those attributes, so that you can produce a petroleum needle coke product that will do what you need it to do in the electrode and subsequently in the melt shop.

Right. And then also in China in terms of new petroleum needle coke capacity, could you just talk about the EV market, electric vehicle market demands. I was – I had heard that petroleum needle coke for the EV market, they’ve discovered have extended the range of batteries in the EV market and that’s what a couple of years ago started to lead to increased demand for petroleum needle coke from the EV market. Is that still ongoing? Have they found any substitute material for petroleum needle coke? Can you discuss that at all?

So what we do know and I want to be clear that we don’t proclaim to be battery experts by any shape or form, but we obviously monitor that space closely to try and stay as well informed as we can. And I would share with you and the people on the call that what is needed in the anode of the battery is a needle coke-like structure. And to the best of our understanding, they need in order to make the battery as efficient as possible and have the longest life, which you’re asking your question around, as possible, the higher the quality of the product, the better off that battery is. So the petroleum needle coke has certainly, as we understand, a bit of an advantage in that respect. I think, they do a lot of mixing of different products between petroleum needle coke and pitch needle coke and natural graphite to accomplish their goal. And it’s all about the ability for the product to transfer a charge in the most effective and efficient way possible. And I have now told you what, you know what we know.

Okay, all right. Thank you.

Thank you. We have now reached the time limit available for questions. I will now turn the call back over to Mr. Rintoul for closing comments.

Thank you, Julie. In conclusion, GrafTech is well positioned to leverage our unique competitive advantage. We will continue to maximize the value of our vertical integration and low-cost production base to provide industry-leading services, solutions and products to our customers. Again, thank you for your interest in GrafTech, and we look forward to speaking with you next quarter.

This does conclude today’s conference call. Thank you for your participation, and you may now disconnect.