Bloom Energy Corporation (BE) Q4 2019 Earnings Report, Transcript and Summary

Good afternoon, and welcome to the Bloom Energy Fourth Quarter 2019 Earnings Call. [Operator Instructions]. I would now like to turn the conference over to Mark Mesler, Vice President of Finance and Investor Relations at Bloom Energy. Please go ahead.

Thank you. Good afternoon all, and thank you for joining us on Bloom Energy's Fourth Quarter 2019 Earnings Conference Call. To supplement this conference call, we have filed our Q4 2019 shareholder letter and earnings release with the SEC and have posted it along with supplemental financial information that we will periodically reference throughout this call to our Investor Relations website. Today, we also filed Form 12b-25 with the SEC, indicating that we would file our Form 10-K no later than March 31, 2020. The matters we will be discussing today include forward-looking statements regarding future events and the future financial performance of the company. These statements are subject to risks and uncertainties that we discuss in detail in our documents filed with the SEC, specifically the most recent reports on forms 10-K and 10-Q, which identify important risk factors that could cause actual results to differ materially from those contained in the forward-looking statements. We assume no obligation to revise any forward-looking statements made on today's call. During this call and in our Q4 2019 shareholder letter, we refer to GAAP and non-GAAP financial measures. These non-GAAP financial measures are not prepared in accordance with U.S. generally accepted accounting principles and are in addition to and not a substitute for or superior to measures of financial performance prepared in accordance with GAAP. A reconciliation between the GAAP and non-GAAP financial measures is included in our Q4 2019 shareholder letter. Joining me on the call today are K.R. Sridhar, Principal Co-Founder and Chief Executive Officer; and Randy Furr, Chief Financial Officer. K.R. and Randy will review the operating and financial highlights of the quarter, and then we will take questions. I will now turn the call over to K.R.

Thank you, Mark. Good day, and thank you for joining the call. Let's start with the topic that's top of mind for all of us, COVID-19. Given what's happening, it's impossible to not feel vulnerable and somewhat helpless. This is a natural reaction to a human issue. At Bloom Energy, we are placing the health and safety of our extended workforce and that of our customers and community as our highest priority. Today, like with most other businesses, our normal operations are impacted by the coronavirus. In the short term, we are committed to doing everything possible to mitigate its impact on our people, business and customers without compromising safety. I also want to take this moment to address the restatement we announced a few weeks ago. What happened was unfortunate. Following the announcement, I was able to speak with many of you to make sure you had the facts. I appreciate the time that all of you took to look at and study the issue and understand what it is and what it is not. The restatement is solely driven by an accounting adjustment for the capital equipment and installed portion of our managed services agreements, less than 10% of the total revenue over the effective period. This is unrelated, unrelated to our service business, the life of our servers or our service contracts. These accounting adjustments have no impact on cash and no revenue is lost. What we are doing at Bloom is innovative, groundbreaking and critically important for the world. I want to acknowledge that the path hasn't always been a straight line, and there have been hurdles along the way. Tumbling as these hurdles are, thanks to an amazing team that is dedicated to the mission, we have overcome obstacles and made progress. We went from 0 to where we are today, a nearly $1 billion a year publicly traded company. Let me now discuss the momentum we saw coming out of 2019. We report our sales backlog once a year. During 2019, we expanded our system sales backlog to 1,983 systems, a 43% increase from 2018. We had reported a very sluggish first half 2019 order book. Clearly, we witnessed very strong business momentum in the second half of the year. So what changed? As I see it, there were 4 major reasons for this very positive turn. Number one, customers are realizing that the electric grid is not able to offer reliability and resiliency when faced with frequent natural disasters, and they are increasingly recognizing and valuing the reliability and resiliency of our Bloom Energy Server deployments. Number two, several utilities in our sales territories have requested significant multiyear rate increases, making Bloom's value proposition of cost and cost predictability more attractive to customers. Number three, customers are now focusing on their need to adapt to climate change and make their businesses resilient. They are asking, "What do I need to do to keep my lights on securely." Number four, while remote renewable projects may offer carbon credit benefits to a customer, they do not provide energy security and business continuity benefits to keep its business operating during a power outage. The dual need of sustainability and resiliency has gained traction with our customers, and we provide a unique solution in the marketplace that addresses them with equal force. The summary of those 4 points, the Bloom Energy solution and its value proposition are resonating strongly with our customers. Let me elaborate on the dynamics in our top 2 markets: the U.S. and South Korea. In the U.S. market, we deployed our always-on microgrid solutions at two Stop & Shop stores in New York. These pilot projects are meeting the customers' economic, reliability and resiliency expectations for several quarters now. Stop & Shop placed a repeat order for 40 of their grocery stores in Massachusetts and New York. These resilient deployments will allow Stop & Shop to better serve the local community at a time of need, even when severe weather events, such as winter storms and extreme heat, disrupt the power grid. We witnessed the increasing customer awareness for resiliency on both coasts now. Here are a couple of metrics. Traditionally, in the past years, the percentage of the total business that was microgrids was in the low teens. Our 2019 ending backlog had that number at over 25%. Also, the microgrid contribution mix in our sales pipeline had more than doubled. Whereas in 2018, the contribution mix was 21%, it grew to 45% ending 2019. We see microgrids as a significant growth opportunity for Bloom. Manufacturing industries have become very sensitive to the price of not having power. Manufacturers cannot afford power outages because they lose in process inventory, labor and overhead and capacity. Power outages also imperil safety and undermine the company's ability to fulfill customer commitments. Because of all of these factors, we are experiencing a healthy demand for our products from the manufacturing sector. Now let me address the Korean market. Bloom sales volume has increased significantly in Korea since we entered that marketplace. Customers recognize that our fuel cells are a preferred solution for reducing emissions and providing high-quality power to the grid. We ended 2018 with 2 major Korean generation companies as our customers. We have won contracts with 4 out of 6 major gencos exiting 2019. Here are a few other highlights for 2019. For clarity and ease of comparison with the estimates that we provided on our Q3 earnings call, when I reference revenue and non-GAAP gross margin metrics in my comments, I will be removing the impact of ASC 606 adoption and accounting adjustments due to the restatement. We provide detailed reconciliations of these adjusted financial metrics to the GAAP metrics in our shareholder letter. So the highlights. One, we had a year-over-year revenue growth of 25.2%. Number two, for the second half of 2019, we achieved a non-GAAP gross margin of 25.8%. Number three, we got a 17.7% reduction in our average product cost year-over-year, from $3,672 per kilowatt in 2018 to $3,021 in 2019. Number four, our prototype units of system 7.5 is performing to design goals, and the program is on track. Number five, we entered into a collaboration with CalBio to deploy our energy servers for the conversion of dairy waste into renewable electricity here in the U.S., as well as a joint effort with Energy Power in India to deploy Bloom's first commercial-scale on-site biogas solution from agricultural waste to municipal waste. Number six, we powered 55 microgrids to 679 grid outages, resulting in over 1,350 total hours of saved productivity for businesses. We as a team are very proud of this. And now I will turn over to Randy.

Thanks, K.R. Throughout my prepared comments, I'll be referring to the slides in the earnings call presentation that Mark referenced earlier. Before I dive into the highlights for the quarter, given our announcement from February 12, at a high level, I would like to walk you through the accounting changes you're seeing at Bloom. Recall from our press release that day, the accounting adjustments related to our managed service contracts impacted slightly less than 10% of our revenues over the effective periods, which were from Q1 2016 to Q3 2019. Furthermore, less than 7% of the systems in our year-end December '19 backlog are managed services related. So with that said, I will now summarize the accounting changes. First, as communicated on February 12, the accounting treatment for managed service agreements are moving from a sale subject to an operating lease with upfront revenue and profit recognition to a financing lease. Under this financing lease treatment, revenue will be recognized over the 10-year term of the customer agreements as power is generated from the Bloom Energy systems and cost against that revenue will flow through the P&L at both the cost of goods sold line as well as the interest expense line. Even though we do not hold title to the assets, the Bloom Energy servers will be put on our balance sheet and depreciated over the estimated life of the servers. So our cost of goods sold for these transactions will be in the form of depreciation. The upfront cash proceeds that we receive from our financing partners, the banks, will be offset with a liability in the form of a loan on our balance sheet. This loan will be reduced or offset over time as we record electricity revenue as power is generated from the systems. The amount of that offset or the electricity revenue will be the portion of the monthly payments made by our end customers that represent the lease payments to our finance partner. Given that the upfront cash received is booked as a loan, we will impute an interest charge, and that will flow through our P&L as interest expense. While making these adjustments, we also identified and made some additional minor adjustments for stock-based compensation and derivatives expenses. This falls within the impacted range that we communicated on February 12 and reflected in the adjustments to our consolidated financial statements. So in summary, the revised accounting treatment for our managed service agreements is, revenue recognized over the 10-year contract term, cost of goods sold against that revenue will be equal to period depreciation, and we will see a noncash interest charge as a result of the balance sheet loan. The second major area of accounting change will be the adoption of the new revenue standard ASC 606. Like many other companies, we are required to adopt the new standard with our new fiscal year ending December 31, 2019. As a reminder, ASC 606 is a standard that intends to improve the comparability of revenue recognition practices across entities, industries, jurisdictions and capital markets. Going forward, under the new standard, we will see minimal impact on both revenue and profit. The overall impact on 2020 P&L is relatively small, and the $17 million-ish negative range for revenue, less than 2% and in the $4 million-ish negative range for net income. This, of course, is relative to the current consensus estimates for 2020. Again, this revenue and related profit is not lost, just pushed into future periods and is noncash. The ASC 606 impact to our P&L was larger in 2019 than we expect to be on a go-forward basis. This is due to high ASPs experienced in 2019. In order to provide greater clarity on both the impact of the accounting changes as well as the adoption of ASC 606, we have provided you with accounting change reconciliation schedules to take our GAAP reported numbers and reconcile those for the impact of the ASC 606 adoption as well as the impact of the accounting change restatement, and we adjust for the noncash stock-based compensation to get you to the adjusted financial metrics. We did this as it reconciles our results with the estimates we provided you at the end of our third quarter. And we will use these adjusted financial metrics for our discussion this afternoon. Now on to the financial highlights for the quarter, and I will include some full year data since we entered our fiscal year. Note that all revenue and profit numbers that I reference will tie back to the adjusted financial metrics presented in the slides. For your reference, detailed reconciliations of these adjusted financial metrics with the accounting changes are found on slides 10 through 14 in the presentation deck. So on to Slide 3. In summary, this was a very respectable quarter. Year-end systems backlog rose by over 43% to 1,983 systems. Acceptances were 386, up 50.2% from Q4 2018's 257 systems. Revenue was $261.2 million, up approximately 22% year-over-year. Non-GAAP gross margin came in at 25.8%, up 7.7 percentage points from Q4 2018 and essentially flat with Q3. Non-GAAP operating income was $21.3 million with adjusted EBITDA coming in at $32.5 million. Adjusted EPS was $0.04. And finally, we ended the quarter with $377.4 million in consolidated cash and short-term investments, which includes $19.5 million of PPA cash. Excluding the PPA cash, we have $357.9 million of cash and short-term investments. For the 2019 year, acceptances were 1,194 systems, a record and up 47.6% from 2018's 809 systems. Revenue totaled $929.1 million for the year, up 25.2% from 2018. Non-GAAP gross margin come in at 22.6% for the year, non-GAAP operating income was $28.8 million and adjusted EBITDA was $97.3 million. Now on to some color for the quarter. I'd like to start with backlog, Slide 4. A as you all know, we only report our backlog once per year. We ended the year with 1,983 systems in product and install backlog, a 43.3% increase over our December 31, 2018 number. This is 66.1% higher than the 1,194 systems recognized as revenue in 2019. Recall that our goal is to maintain 9 to 12 months of product and install backlog at any point in time. We are clearly at the higher end of that window as we enter the 2020 fiscal year. The upfront system contract value of the 1,983 system backlog is approximately $1.1 billion. And this backlog excludes any service agreements as we recognize those billings as occurred on an annual basis. If you were to include the estimated value of the service revenue, assuming all service contracts for these 1,983 systems are renewed over the full contract term, then another approximate $1.1 billion would be added to that backlog value. Also excluded from both the $1.1 billion of upfront system contract value and the $1.1 billion of service billing value just discussed is the value of the future service billings for our existing installed base. That estimated value of future service billings for systems in our installed base, again assuming that the service contracts are renewed over the remaining contract term, is approximately $2.1 billion. And finally, also excluded from backlog is future electricity revenue from our existing installed base for our previous Bloom Electrons program. Total electricity revenue in 2019 was approximately $72 million. So summing up, all the current contract backlog and future contracted revenue yields approximately $4.3 billion of revenue that will be recognized in the future. The majority of our current backlog is domestic and includes U.S. commercial and industrial customers, utility-scale projects and international customers. On to Slide 5. The 386 acceptances and $261.2 million of revenue were both Q4 records for Bloom. Quarterly acceptances were up 50.2% year-over-year and up 27.8% sequentially. Quarterly revenue was up 22.3% year-over-year and up 11.9% quarter-over-quarter. The majority of our revenue growth was driven by the mix of acceptances where we have low or virtually no install revenue associated with them. I will add more color on this in my ASP discussion. On a year-over-year basis, we achieved 1,194 acceptances and $929.1 million in revenue. Again, both records for Bloom. On an annual basis, acceptances were up 47.6% in 2019, and revenue was up 25.2% relative to 2018. Most of Q4's mix of acceptances were from existing customers and represent a broad range of verticals to include health care, pharmaceutical, universities, utility scale projects, food and beverage retail and even a sports venue. In total, the 386 systems were spread over 12 different end customers with the majority of the installations in the United States. On to Slide 6. In our Q3 shareholder letter, we provided you with a range of Q4 sales price estimates as well as a range of total installed system cost estimates. For Q4 '19, our average selling price, or ASP, come in at $5,906 per kilowatt, a number just below the lower end of our estimated range. Total installed system cost, or TISC, come in at $4,289 per kilowatt, a number within, and on the positive side, or put another way, below the midpoint of our estimated range. While I'm on the topic of cost, I'd like to take this opportunity to say 2019 was another excellent year in our product cost reduction efforts. Average product cost dropped 18% year-over-year, a number in line with the average cost reductions over the past 5 years. And as I mentioned in the past, both ASP and TISC are impacted by a number of factors to include: site location and applicable utility tariffs for that location; whether the site includes grid outage protection and/or is mission-critical; the size of the site being installed, generally, the larger the installation, the lower the cost on a per kilowatt basis; and whether or not, the scope of our work includes installation. Typically, our international business does not include installation. So once again, I continue to stress that the important element is not the trend of the ASP or the TISC but the trend and the delta between the 2. This delta represents our unit level profit of the acceptances during the quarter, which directly correlates to our overall gross profit and gross margin. The midpoint of the estimated ASP and TISC yielded a delta or margin estimate of $1,670 or $1,670 per kilowatt. As you can see on Slide 6, our actual margin delta was $1,617 per kilowatt. The mix of customer sites that yielded from our pool of acceptances had ASPs that drove us towards the lower end of the range on our ASP and thus drove a margin slightly below the midpoint of our estimates. However, as mentioned, acceptances or the volume metric slightly exceeded the top end of our estimated range. I note that for 2020 and beyond, we're going to try to simplify our forward-looking estimates for you, and I will share more specifics later on in our outlook for Q1 2020. Turning to Slide 7. Gross profit on an adjusted financial metrics basis was up almost 74.4% from $38.7 million in Q4 '18 to $67.5 million in Q4 '19. On a sequential basis, gross profit increased 11.9%. Gross margin for Q4 came in at 25.8%, a significant increase from last year's 18.1% and flat with Q3 19's 25.8%. Our operating income in Q4 was $21.3 million, up significantly both on a year-over-year and sequential basis. Our reported adjusted EBITDA was $32.5 million for the quarter. Nonoperating expenses, worker plan and EPS came in at $0.04 for the quarter. For full fiscal year, gross profit was $209.9 million, up 32.3% from $158.6 million for FY '18. Gross margin came in at 22.6%, an increase of 1.2 percentage points over FY '18. Operating income, as reported in the adjusted financial metrics, increased to $28.8 million, up 9.1% from FY '18 and adjusted EBITDA come in at $97.3 million, an increase of 45.4% over FY 2018. Let me now switch to the balance sheet on Slide 8. We ended the quarter with $377.4 million of consolidated cash and short-term investments. This includes a total of $19.5 million of PPA cash. So excluding PPA cash, we ended with $357.9 million of total cash and short-term investments. This is an increase of $19.5 million from Q3. However, included in the $357.9 million of Bloom cash is $157.2 million of restricted cash. This is up about $44.6 million from Q3. The driver of this increase in restricted cash is a cash reserve committed to our financing partner in exchange for their commitment to increase their financing limit. This will reduce over time, starting in July of 2021, with the full amount of this increase expected to roll off by the end of 2025. Keeping on the balance sheet, returning to debt, I wanted to provide an update on the $330 million of debt that I discussed on last quarter's call that we are in the process of refinancing, with Jefferies leading the process. The focus of our efforts has been on refinancing the approximately $289 million of the 6% convertible notes that are due in December of this year. Since that call, we have considered the range of options that I outlined last time. We have narrowed that range of options with the goal of pursuing a debt combination of term debt alongside another convertible note. We will provide full details upon completion of the offering. Given the strength of our business and the positive momentum we have, we remain confident that we will work through this in the first half of the year and that our refinancing approach will be in the best interest of our equity and debt holders as well as our partners in the business. Referencing Slide 9. Days of sales was down 1 day from Q3 to 12 days driven by the higher volume in Q4. Our days of inventory outstanding was up by 6 days from Q3 to 81 days driven by reduction in service and electricity cost of goods sold. And our payable base was down from Q3 by 1 day to 42 days. I would now like to change the conversation to our outlook. But first, I wanted to add that in Q4, we did a pretty thorough investor perception study. One of the items coming out of that study is our investors' preference to simplify our outlook and provide more transparency into our full P&L. So here is our attempt to do just that. We are simplifying our estimates to 2 metrics as part of our outlook, one for the top line and 1 for the bottom line, or total revenue and adjusted EBITDA. Providing these 2 metrics afford additional insight into other aspects of the P&L to include our service and electricity revenue streams and stock-based compensation, which is part of our adjusted EBITDA calculation. This was a challenge under the prior methodology. Even though we are only providing estimates for revenue and adjusted EBITDA, we will continue to report acceptances, ASP and TISC. Periodically, we may also provide visibility into other metrics but they have a meaningful impact on the quarter. For Q1 2020, we expect total consolidated revenue to be between $140 million and $160 million. We expect operating expenses to be between $48 million and $51 million, and this includes $3 million to $4 million of onetime expenses related to the restatement and restructuring. Finally, to complete our outlook, we expected adjusted EBITDA to be between a $15 million to $25 million loss. Q1 is challenged with respect to mix. Mix will be more favorable throughout the balance of the year. Also, we expect the cadence of both revenue and profit to be similar to prior years, with each successive quarter seeing higher revenue and corresponding profits. As we enter the new year, I'd like to add a little color to our outlook, especially as it relates to the 2020 full year. We clearly had a record second half for Bloom with respect to bookings. Given this and the fact that for our U.S. C&I business, it generally takes 9 to 12 months from order booking to order acceptance. We do expect a far better second half of 2020 than the first half as there's just not sufficient time to transition in the strong second half 2019 bookings in the first half 2020 revenue. In fact, many of the Q4 orders will represent 2021 revenue. So clearly, a stronger second half. Ideally, this would have allowed for an improved outlook. You might recall, on our Q2 earnings call, we discussed certain headwinds we were seeing in the markets. And we provided a high-level outlook for 2020 at that time. Obviously, with respect to the headwinds, that pendulum swung completely the other way. And by Q4, we were seeing some strong tailwinds. So again, why not an improved outlook at this time. It all has to do with what we're seeing in the world today. But the recent events primarily driven around the coronavirus, we felt it not prudent to be increasing estimates at this time as the world is simply a difficult place to judge today. With that said, and as already mentioned, we believe we are in excellent position today. We have a strong backlog and the demand for resilient power is real. Today, our customers no longer look to Bloom for simple grid power arbitrage or for our sustainability attributes. They look to Bloom to keep their business operating. I will now turn back to K.R. before we take questions.

Thank you, Randy. I would like to thank Randy and the entire finance and accounting teams for their professionalism and integrity during this process. The extraordinary work required to get our financial results completed in a timely fashion was significant. Thank you, Randy, also, for your dedication and leading the team to get us to this point. And in terms of Randy's successor, we are nearing the end of the process with our final candidates. We thank Randy for agreeing to stay with us through the selection and on-boarding of the new CFO. We wish you well in your retirement. Looking ahead to 2020, we will continue on the cost reduction of our 5.0 platform. The development of our 7.5 platform is on track, and we will see the rollout of first systems to our customers. We will continue to develop and demonstrate low end 0 carbon solutions with our technology platform. I am confident that we are building a transformational energy technology business that is world-changing and highly relevant for our current times. Before the COVID-19-related disruptions, we had increased business momentum and healthy backlog and were well positioned for the year. It is our hope that our nation and the world will emerge out of this crisis soon. After this crisis is over, we know that our strategy and offerings will be very relevant and needed. We, as a company, are using this time to be prepared to serve our customers after this crisis is over. COVID-19 is a reminder that it is absolutely important for us to build resiliency and adaptability in our daily lives to protect ourselves against macro disruptions. And just as this is true with our health and wellbeing, this is very true with electricity as well. Our microgrid resiliency solutions are designed to build resiliency and adaptability for business. As I see it, going forward, business leaders and boards will prioritize the implementation of their business continuity plan. As we have gone through the few weeks in 2020, we have taken stock of who we are, what our values and beliefs are and how we run the company. First, we are about doing the right thing all the time. Second, we work hard to deliver on promises we make and are dedicated to constant and continuous improvement. Third, we are fully committed to our mission of delivering cost-effective electricity solutions that are cleaner and more reliable to address the causes and the impact of climate change and power disruptions around the world. Thank you all, and we will now take your questions.

[Operator Instructions]. And your first question comes from Stephen Byrd with Morgan Stanley.

I wanted to just first touch on care, the very first topic you mentioned, the coronavirus and impacts on the business and just explore that a little bit further, whether it be on supply chain or on customer sales outlooks or on installation capability. I know things are quite fluid and it's challenging to predict exactly the extent of impacts from the coronavirus. But would you mind just elaborating a little bit further on how you think about potential impacts both today, but also potentially in the future?

Sure. That's extremely important. And as you mentioned, the key thing here is it's dynamic and fluid. And I want to emphasize where I started. For us, it's about doing the right thing all the time, which means safety first. Safety and well-being of our employees, of our community and of our customers. So in the immediate term, what we see is normal business is going to be affected, whether it's people being able to come to work, whether supply chain is getting disrupted. And there are a lot of our customers who may not want any installation work being done at this point in time or us not feeling that it's the right thing to do. So that's your immediate issue. But if you go past that to the midterm issue, if you look at it, the good news at Bloom is, we have 15 months' worth of backlog. We have products and these are blue chip customers, such as a Home Depot, Walmart, a Stop & Shop, wanting to keep their operations running. If you look AT&T or customers like that, that need to keep the communication links running, if you look at somebody like Kaiser Permanente wanting the hospitals running. The important thing to remember in the midterm is while other things may pop-in or take a pause, mother nature and climate is not going to take a pause and resiliency is as important going forward. If you look at the products and the service we offer, it's the base load for the power, not even the peak load. And that base load for these kind of blue chip customer business that we need is going to be there. Longer term, we really believe in a very connected centralized world. While that has to continue and be strong, it is equally an important aspect of localization in terms of protection, safety, business continuity. This is going to become front and center, and I think the COVID puts a very good exclamation point on saying, "Are we thinking that way about every aspect of our life?" We think that's going to play a very important role going forward. Stephen, that's the immediate midterm, long term. But I'm more than happy to take this further if you want to.

Understood. I mean it sounds like there could be potentially just some delays in the very near term, given customers thinking through exactly how they plan their business in this remarkable time. I mean many employers have requested employees to remain home. And so I think I understand that, that message, if I got that right?

Yes. Yes. That's correct. And that would affect our suppliers, that would affect our own employees, depending on how things turn out in the next few days. So everything is fluid in that situation in the immediate short term. But the business was extremely strong before this event happened. It is a crisis, but this will end. And how are we going to look when we emerge out of this crisis. And so we are trying to focus with safety in mind first, but we're trying to focus with our energies how are we going to look and how strong we're going to be when we emerge out of this crisis. And I'm extremely confident we're going to emerge out very strong coming out of this crisis.

Understood. And then just a separate question just on thinking through the potential for carbon capture. Would you mind, just given your latest thoughts on, I guess, both the technical status of being capable of capturing CO2, but also maybe a little bit more about, just at a high level, the business prospects, what could that mean? And I'm thinking there about really the cost of capturing the practical uses of the CO2, and just more broadly about how you think about that opportunity?

Sure. So if you look at carbon capture, I think, while -- there are 3 ways that we think about this. The first thing -- above and beyond what we do, the first point I want to make is when we use natural gas, traditionally in our system today, we have the least carbon footprint way of generating power without the smog emissions with our current technology. But you are asking the next steps forward. Three different ways to think about this. Number one is we are trying to get as much green molecules as we can, which is either renewable hydrogen, biogas to come into our systems and in the pipeline. That takes carbon out in the first place even before you get started. When you use biogas and get carbon capture, you get to negative carbon. Now let's talk about natural gas to 0 carbon using carbon capture. That is a highly scalable time and -- a time-sensitive -- something we can do in the next few years in large-scale across the board. How can we do this? In a 1 acre footprint, we can do 100-megawatt power plant. In that 100-megawatt power plant, if you're able to get carbon capture at a sub-penny level and then be able to use Bloom's aggressive cost reductions as we have seen, we believe that we can get to a single-digit cents per kilowatt hour price, starting with natural gas and going to zero carbon sequestrant. That's what we are focused on for the long term when we look at carbon capture. And we think it's one of the most viable methods of providing reliable, resilient and zero carbon electricity, with a fuel that's abundantly available today to supplement the renewable electricity we generate.

Your next question comes from Michael Weinstein with Crédit Suisse.

Could you talk a little bit more about the gross -- the upfront margin, which dropped, I guess, 44% quarter-over-quarter to $1,617? It looks to me like the main thing here is total installed system costs up while ASPs are down. I understand the ASP being dependent on your regional mix, especially if it doesn't include install. Is the -- what -- how do you see the total installed system cost shaping up in 2020 and beyond? Where do you see gross margins going? I know you briefly mentioned that you think it will improve, but if you can give more color?

Yes. So Michael, good question. So obviously, the gross margin there is impacted primarily by two things. One is our average selling price, and that varies significantly from order to order within the mix of what falls into the quarter. And the other is our product and install cost, the combination of those, which we refer to as total -- TISC, total installed product cost. And that, as you can see on the long term, that trend has been trending down. Now from quarter to quarter, you could see some fluctuations on whether -- and that fluctuation is all tied back to the installed cost. But generally, that tracks down. From quarter to quarter, we're going to have a richer mix, meaning just the mix of business in there. It's going to be -- have a higher ASP and also higher profit and other quarters, like we just guided to, it's going to -- the mix is going to be, as I pointed out in my prepared remarks, it's going to be more challenged. In the short term, if you go back to -- in the short term, I meaning, say, over the next 6 months here, if you go back to comments that we made at the end of the second quarter, we certainly pointed out that we were being challenged and to fill that order book up through the first half of last year, we were a little more aggressive in booking orders with -- that translated to some more ASPs than we normally like to see and normally, we think we would see in the future, especially given the strength that we saw in Q4 and the tailwinds that we're facing today. And you are seeing that flow through in Q1 and a bit more in Q2. I do want to -- without giving too much forward guidance here, I do want to get back to the prepared comments, I thought were key, when I said, we expect 2020 to follow the cadence of 2019, and if you go back to that, what you'll see is it was a loss in Q1, give or take, breakeven in Q2 and then pretty decent profits in Q3 and 4. I -- based on what we see today, we would expect that cadence to continue here going forward. I'll pause there and see if I answered your question or you want to go down on that some more.

Okay. It sounds like the second half, which is what you said before, the second half should be a solid second half, especially as the backlog starts to flow into -- starts to flow into second half.

Michael, and then I think you asked a question on the long term, and here is the way to think about it, right? I spoke to the contribution coming from the microgrids. If you think about the shift in conversation happening among our customers for all the reasons that you can imagine of the price of not having power versus the cost of power, that is a very different discussion, and that's where we see the dynamic going. You combine that with our constant cost reduction that you saw, we feel that going forward, the margin is going to be healthy.

And the increase in utility.

Yes. And the third point, like Randy currently points out, is the utility rates are going up. So that combination of those 3 things will or should lead to a higher margin as we go forward.

Got you. On the term loan and convert, are you confident -- I guess are you confident after talking to Jefferies that you're not going to have a liquidity problem here in the first half, if you're going to be able to get convert refinance in the first half. Is that the reason why there's a term loan being considered that, that might be a little easier to do at this point?

Yes. Look, we picked this path because we want the right capital structure for the company long term. And we certainly felt like some amount of, call it, permanent or long-term debt makes sense for the company. And we felt like the fact to get some equity -- more equity into the company and the best vehicle to do that, given our share price today being low was a convert made sense for the company to give us the right long-term capital structure. We're -- I don't know what inning to put it, but we're kind of in the middle of the game of getting that whole transaction done. And until it's done, it's not done, but we felt good about the path we're going. And as I pointed out, the goal is to get that done here in the first half.

Yes. But I guess my comment is more aimed at the current market and whether you're seeing any changes to the process getting that refinance completed as a result of that, and whether the term loan specifically helps with that -- with liquidity, making sure that you get something done, something refinanced even if the market conditions and the equity margins are terrible.

You're exactly right. And look, the term -- you hit the nail on the head, it's a lot easier to get to where we needed to be with the term loan. And we want to do it. The markets are challenged out there today, getting folks to focus on this today is a little bit more of a challenge. But we still feel confident of where we sit today. And our goal is to make an announcement on this sooner than later, but we want to keep our commitment that it will occur in the first half.

[Operator Instructions]. Your next question comes from Paul Coster with JPMorgan.

I'm just trying to understand the lag between the bookings that are going into backlog and the subsequent acceptances. It can't be production. I mean it seems to me that you could easily run rate your production forward and grow from that. But -- so it has to be something to do with the lead time on installations. Can you just talk about that a little bit? And why it's why there is that gap between -- and timing between the two?

Yes. Paul, but we're not only a company that builds -- we're a technology company that builds a product from the energy industry here. And you're right, we have a great both supply chain and internal manufacturing operations. And I can honestly say, in my five years here at the company, we've not missed 1 shipment related to anything to do with our supply chain. Our internal manufacturing folks are just terrific. Great quality. And you're right, we can increase that and the folks costs that we do. But also bear in mind that for all of our U.S. commercial and industrial, we're also, for lack of a better term, a construction company, and we go out and we install these systems. And that process itself -- I like to think of it as kind of three phases. The first phase, which takes in the neighborhood of about 3 months, we send the field application engineer out to the site, survey the site, find out where all the utilities are, get that back to a CAD. Person running a CAD machine, they do all of the detailed construction drawings, everything we needed for all the approvals improvements and that whole process takes in the neighborhood of 3 months. Then we have to get all of those approvals, and that includes customer approval, often sites leased, the landlord approval, the local building permit approval from the local city or the jurisdiction having the authority to approve those building permits, local gas, electric company. All of those approvals usually take in the neighborhood of three months as well. So we are somewhere around 6 to 7 months into the process. When we get those approvals, there's a person who's actually on my team, he signals to start the construction process and to start the actual building of the systems. From the time we start a fuel cell to the time we deliver that system is about 6 weeks. So it's a fairly short time from start to finish. And we build those systems. We construct them in the field. And that whole process takes from the time we break ground to the time it fully gets up and running. Again, another 2 to 3 months. So you add all of that up, on the average, we're in the 9 months, sometimes 10-month time frame. And then obviously, on some of our larger customers that gives you a large amount of sites at 1 time, even besides your books kind of start in that 9-, 10-month time frame and then go through 2 or 3 more quarters before we get all of that done. So the point being is that the orders that we booked in Q4, most of them in November-December time frame, would be orders that would be at best delivered in Q4 of this year. And a fair amount of that would spill over into Q1, Q2 of 2021.

And your last question comes from Pavel Molchanov with Raymond James.

I know that in 2018, 12% of the revenue came from Korea, and I imagine a significant percentage this past year as well. Given the virus impact in Korea specifically, can you talk about how your relationships with SK and the Korean customers are being affected currently?

Yes. Sure. That's a good question, Pavel. Here is what I would say is the Korean orders, unlike the U.S. C&I orders come from the utility companies. These are for the utility base loads, and they're usually in large industrial sites, and there's a cadence to it with a request for proposal and a process. So far, even during the 1.5 months to 2 months when Korea went through a peak and now it's civilized out there, we have seen that process in its normal cadence as opposed to anything very different. How will it translate in the future? It's very difficult for any one of us to say. But at this point in time, we have not seen any change that we can see from our end to that entire process. And they have a mandate to be able to do this. These are large RFPs for multimegawatts at a time. And they are to the 6 gencos. And there have been no signals from any of them that any of that process is going to be stalled in the near future. Does that answer your question?

Yes. I appreciate the color. And can we get a quick update on the marine shipping partnership with Samsung? I guess we're in 6 months of that now. Any progress?

Yes. We are making progress with them in terms of a work breakdown schedule and who is going to do what. And because there is a lot that has to be done on the ship onboard and lot that we will be doing in our own systems, and simultaneously, there are many authorities and regulatory agencies through whom this has to go through in terms of its design and approval process, very similar to a UL process that you have for ground-based applications. And that process is also going forward. And that is simultaneously working with the ship builder that is discovery and understanding of customer needs and customer requirements, and that's also happening in parallel. So those are the activities that are moving forward on that particular project.

That's all the time we have for questions. I'll turn the call back to K.R. Sridhar for closing remarks.

Well, apologies for keeping you 5 minutes longer than what we originally planned for on what is an unusual day for all of us in more ways than one. Look, what we want to emphasize out here is our strategy, our product, what we were doing is resonating with our customers. Customers are beginning to understand what we do and value it for what it can deliver for them. The COVID crisis is clearly a stumbling block along the way that closed the entire nation and the world on, but it is a crisis that it's going to be over. I am very confident that we will be fully ready to serve our customers when the crisis is over. We, as a company, have resilience and I'm extremely confident of our team out here being resilient and being creative and being nimble to be: a, safe; b, take care of our customers; and c, be able to serve our customers on once the crisis is over. So thank you all for your time, and we really appreciate your support and your faith in the company and our mission.

Thank you.

This concludes today's conference call. Thank you for joining. You may now disconnect.