Advanced Energy Industries, Inc. (AEIS) Earnings Call Transcript & Summary

Ladies and gentlemen, thank you for standing by, and welcome to the Advanced Energy 2020 Investor Briefing Event. [Operator Instructions] And I would now like to hand the conference over to your speaker today, Mr. Edwin Mok. Please go ahead.

Thank you, operator. Good afternoon, everyone. Welcome to Advanced Energy 2020 Investor Briefing. I'm Edwin Mok, Vice President of Strategic Marketing and Investor Relations. On behalf of the AE management team, I'd like to thank you all for joining us today, either on our live webcast or over the phone. Let me start with our safe harbor statement. Today's presentation contains forward-looking statements, which are subject to risks and uncertainties that could cause actual results to differ materially and are not guarantees of future performance. Information concerning these risks and uncertainties can be found in our filings with the SEC. Aspirational goals and long-term vision targets presented today should not be interpreted as guidance. Financial figures in today's presentation will represent on a non-GAAP basis. Explanations and reconciliations of these non-GAAP financial measures can be found in today's presentation and in our filings with the SEC. Lastly, let me remind you that we are in our quiet period. Therefore, any comments we make today for the year 2020 or for the fourth quarter based on guidance we provided during our Q3 2020 earnings conference call on November 5, and they are not a reiteration. We are not providing any near-term updates and our comments will be focused on the long-term goals for Advanced Energy. Let's move to the agenda. With me today are Yuval Wasserman, our President and CEO; Paul Oldham, our Executive Vice President and CFO; and Dana Huth, our Chief Revenue Officer. Yuval will start off with an overview of discussing our secular drivers, key growth areas and key messages. Dana will follow with a deep dive into both the data center hyperscale and the industrial medical markets. Paul will then present the progress we have made and a detailed review of our financials. We will then take your questions. With that, let me pass the presentation to our President and CEO, Yuval Wasserman. Yuval?

Thank you, Edwin. Good afternoon, everyone, and thank you for joining us at today's virtual investor briefing. One year ago, at our 2019 analyst event, we discussed our strategy and plans to transform Advanced Energy into an industrial growth company. Today, we would like to provide you an update on our progress, discuss our strategy for future growth and show how Advanced Energy is becoming a best-in-class industrial technology growth company. Advanced Energy is uniquely positioned to deliver growth and shareholder value based on our power-centric strategy, market growth drivers, track record and focus on growing earnings and creating value. The core of AE's strategy is our pure-play power leadership. AE excels in the highly specialized area of precision power delivery, leveraging our power electronics core competency throughout our highly engineered product portfolio to provide industry-leading solutions. Being a pure-play power leader, our strategic focus and deep experience enables sustainable competitive advantage and allows us the ability to scale the company as we grow. Second, the fourth industrial revolution is a megatrend driving growth across all our markets. We have a broad base of leading products, applications and solutions that directly power the key applications behind the data economy and the digital transformation of many industries, making us a leader in each of our markets. Third, we have a solid track record of outperforming the markets we serve. As the undisputed power conversion innovator, we have grown faster than the markets we serve by gaining share and increasing content. We also have a history of broadening our portfolio, entering new application verticals and expanding our served available market. Finally, as we have demonstrated over the past year, we are accelerating earnings growth. Our goal is to grow earnings at more than 2x the rate of revenue growth and have demonstrated that we can meet this target while delivering top-tier return on invested capital. Over the past year, our trailing 4-quarter ROIC rose from about 13% to over 20%. We believe this represents true value creation and return for our shareholders. Advanced Energy is a clear leader in precision electric power conversion and control solutions, with decades of experience in enabling our customers to power their most critical applications. In a nutshell, we make advanced power supplies and related products that take in electrical power from the grid and convert it to an optimal usable form for a variety of applications, from semiconductor capital equipment and industrial tools to medical and life science equipment, data center servers and 5G radio towers. Our power solutions are embedded in our customers' equipment, converting raw power to usable, controllable, stable and clean power necessary to operate these different types of equipment with precision and reproducible results. We are experts in power conversion, measurement and control technologies. Our focus on pure-play power and our leadership in the markets we serve has enabled us to deliver solid financial performance as we have grown and diversified the company over the last several years. Our served available market of $9 billion consists of 4 market verticals, which are driven by the data economy. Our diverse exposure across these different markets provides less volatility while allowing us to capture meaningful growth opportunities. We are #1 or #2 player in each of our vertical markets, with the ability to continue and grow our share position and content driven by our innovation and technology leadership. Our leadership, synergistic operation model relying on our functional organization structure, and the resulting execution have delivered solid financial performance. Since completing the Artesyn acquisition a year ago, we have more than doubled the revenue, operating profit, earnings and cash flow generation. These results reflect our strong execution and operational excellence despite the COVID-related economic downturn. Our solid performance in the third quarter of 2020 validated our business model, giving us confidence in meeting last year's aspirational goals. Therefore, I want to kick off today's presentation with our updated 3-year aspirational goals. We are targeting to reach revenue of over $1.65 billion; non-GAAP EPS of greater than $7.50; and return on invested capital of greater than 23%. The revenue and EPS goals represent solid increase over the 3-year aspirational goals we provided a year ago. Let me turn now to the key secular drivers behind our markets, starting with 5G. 5G might be the most impactful new technology driving our 4 markets. 5G handsets are already driving higher demand in both logic and memory chips, supporting the need for increased semiconductor manufacturing capacity and leading-edge technology. In data center computing, 5G will accelerate the adoption of cloud and mobile edge computing leading to dramatic increases in demand for our power solutions. On the infrastructure side, 5G is driving new investment by telecom providers in cell towers, base stations and backhaul networks. 5G will change how enterprises and factories operate, how industries manage themselves and how medical care is delivered and administered. Artificial intelligence and machine learning are another secular driver. It is driving new thinking, new designs and new architectures across our markets. AI will require new innovation in chips, high-performance computing and high-speed data connectivity. It will enable new industrial and medical applications. Higher power consumption will be needed in the data center to enable AI, making power conversion efficiency and power density even more important to address the need to reduce total cost of ownership. AE is a global leader in both power conversion efficiency and power density where our competitive position has been proven in our recent share gains, particularly in hyperscale. The Industrial Internet of Things or IIoT, is the third macro driver that will accelerate growth and innovation across our markets to enable smart manufacturing operations, faster processing, more computing and high-speed connectivity are required, both within and outside the factory. Our power solutions are enabling IIoT by embedding data analytics, connectivity and system integration capabilities. Our customers can leverage these advanced features to accelerate their IIoT innovations. 5G, AI and IIoT are the key technologies behind the Fourth Industrial Revolution. Together, They are accelerating investment in technology, capability and infrastructure for years to come. The macro drivers behind the Fourth Industrial Revolution will directly benefit each of the market verticals we serve, increasing demand for our advanced power solutions. Not only will the power content in these markets grow, but the advantage to the power innovators will become greater as our customers look to partner with the most innovative suppliers to stay ahead in the race. Advanced Energy is a technology leader in all these markets, and we are partnering with customers to help drive their technology road maps. Let me move to our growth strategies. The 3 core pillars of our growth strategy are market share gains, innovation and capitalizing on new opportunities both organically and through acquisitions. We apply these strategies to grow our business across all our market verticals and product categories. More specifically, I'd like to focus on 4 key areas for growth. In semiconductor, we are extending our leadership position with new products, higher capability and added content. In hyperscale, we are quickly moving from a fast follower to a market leader. In industrial and medical, we developed a wide range of Internet-enabled smart applications. And finally, through value-creating inorganic growth, we will leverage our scale as a pure-play power leader to increase our addressable markets and grow scope. Let me provide you with some details behind each one. In semiconductors, we have been the undisputed leader in process power for decades. This is reflected in our share gains over time and our history of growing faster than the markets and our competitors. In RF, we have been the #1 market leader for a long time and have 2x more market share than our next competitor. Year-to-date, our revenues from the semi market have grown 61% over 2019. Based on the midpoint of our Q4 guidance, we are on track to grow our semi revenue at a 13% CAGR over the last 10 years, growing at over 2x faster than the wafer fab equipment market or WFE. With our leadership and solid pipeline of potential design wins, we remain confident in our ability to continue and outgrow the market over the long term. Two quick examples illustrate our technology and leadership. RF matching is a fast-growing category. As chip designs have become more complex, demand for advanced RF matching is accelerating due to the need for more sophisticated RF power tuning and methodology. We are the industry leader in this area as is demonstrated in our accelerated growth. Our new solid state match and our integrated generator and match solutions will further extend our leadership position. Also, with last week's announcement of our new remote plasma source product, the MAXstream, we are changing the game in this $150 million market. MAXstream solves some of our customers' most pressing challenges that were unmet by competing and incumbent solutions. With 60% higher power accuracy and dependability plus a smaller form factor, the MAXstream is the most capable RPS solution on the market. In hyperscale, we entered the market as a fast follower but are quickly becoming a market leader as the industry focuses on power efficiency, density and system-level solutions. We have seen strong market share gains in the last year and are targeting to outgrow the hyperscale market by more than 4x, enabling our total data center computing annual revenue to more than double from 2017 to 2023. Dana will provide you with a deeper review of hyperscale in this presentation. The third key area for growth is in industrial and medical. We are enabling smart applications by adding advanced functionality to our power solutions. We believe this is an important step in enabling our customers' adoption of Industry 4.0. As an example, recently, we announced 3 new products, each with advanced digital capabilities such as connectivity, big data analytics and self-diagnostics that deliver on the promise of Industry 4.0. Dana will discuss more details behind our leading product capability, configurability and market reach as part of the strategy to grow in this market. In addition to organic growth, we believe there are significant opportunities to continue to expand our position and leverage our scale through smart acquisitions in this large and fragmented market. Over the last several years, we have made 8 successful acquisitions, deploying $534 million to acquire over $750 million of pro forma revenue. Our track record with these acquisitions has been strong as we added complementary high-voltage, low-voltage embedded power and measurement and control capabilities. We have integrated each of these businesses and improved their profitability. We continue to focus on power or power adjacent businesses, which can expand our reach in our target markets and to our portfolio of products and technologies and leverage our scale. We have a strong funnel of opportunities and are using a disciplined and analytical approach to ensure value creation and drive both top and bottom line growth. In addition to being a power leader and enabler for our customers, we believe that how we operate is equally important to our stakeholders. Our products reduce energy waste and lower carbon emissions, helping to create a more sustainable future. In addition, we strive for the best environmental practices in how we operate. For example, our new manufacturing facility in Malaysia runs on 100% renewable energy and we just recently received the top award for manufacturers in the Philippines for the third consecutive year. We are partnering with industry leaders to reduce waste and lower carbon footprint through us recycling a new transportation strategies. We are committed to our employees and communities and support a number of volunteer and education opportunities, including a recently announced scholarship and internship program designed to increase gender and racial diversity in STEM education and careers. And finally, we have recently expanded our Board, adding 2 new independent directors with complementary capability, industry experience and perspective. These actions reflect our core values and support our overall business success. Looking forward, Advanced Energy has a significant opportunity to deliver growth and earnings beyond the 3-year aspirational goals I discussed earlier, Enabled by the ongoing drivers of the Fourth Industrial Revolution, our leadership in precision power solutions, execution of our strategies and the potential contributions from inorganic growth, we see a well-defined path to achieve our long-term vision of $2.5 billion in revenue and over $12 per share in non-GAAP earnings. We believe this long-term vision gives a compelling view into how AE continue to increase shareholder value over time. Paul will go into more details on this long-term vision in his remarks. In conclusion, as a global diversified power leader, I'm excited about our target markets, our technology leadership and our growth prospects. We have demonstrated solid performance over the last few years. This track record gives us the confidence that we are well on our way to becoming a top-tier, mid-cap industrial growth company. With that, I will turn the floor over to Dana Huth, who will go into more detail in data center, hyperscale and the industrial and medical markets.

Thank you, Yuval. Good afternoon, everyone, and thank you for joining us for today's event. For new investors or analysts who I have not met yet, I'm Dana Huth, Advanced Energy's Chief Revenue Officer. I have responsibility for sales and marketing at AE and today, I will share with you how we are realizing AE's growth potential. In today's presentation, I will focus on 2 important verticals: the data center hyperscale and industrial and medical markets. First, I will walk through the market dynamics in hyperscale and explain how AE is able to gain share in this market. I will then focus on the industrial and medical market, discussing our strategies and success in this vertical. Let's start with data center hyperscale. We are seeing an accelerated adoption of cloud computing and continuous growth of the Internet as we experience the Fourth Industrial Revolution. This growth is expected to continue for many years to come. To meet the growing demand, hyperscale companies are investing in their own infrastructure which is driving increased demand for our products in the data center computing market. The power needs of hyperscale customers are different from those of traditional data center server manufacturers. Traditional server and storage systems provided by OEMs and ODMs have power supplies embedded in each server. These power supplies are typically standardized for economies of scale with multiple providers competing based on a combination of efficiency and cost. As system architectures evolved, larger companies began customizing their own servers for their very specific and unique applications. AE works closely with these customers, design organizations to develop custom power supplies that are optimized for their system architecture. Relative to standard power supplies, there are only a limited number of suppliers who can offer our level of customization. As the hyperscale landscape evolved, our customers realized that a system-level centralized power system, delivering energy to a whole rack of servers would maximize flexibility and energy savings. Known as a power shelf, these power systems have many built-in capabilities that go beyond power conversion. We believe very few suppliers have the technical expertise to deliver and support this type of complex system-level solution. Hyperscale companies design, own and operate their own IT infrastructure and data centers. And as a result, are highly focused on the total cost of ownership or TCO. So why is TCO so important? As our customers take a holistic view on their infrastructure spending, they have realized that a better power system design can optimize rack level costs and high-powered solutions that enable increased compute density are more important than just the cost of the power supply. At the same time, since electricity is the biggest cost component in operating a data center, maximizing energy efficiency is also critical. A customer told me recently that increasing the efficiency of the server power system would allow them to save millions of dollars per year. To optimize cost, they also want to have the visibility into where energy is used. With industry-leading power efficiency, density and system-level controls, our power shelf solutions help customers to optimize CapEx and reduce the OpEx of running the data center. The server migration to 48-volt is another important trend in the data center hyperscale market. Currently, Only one of the top 10 hyperscalers has deployed 48-volt servers in its core infrastructure, but we expect half of them will move to 48-volt by 2023. The main benefit of 48-volt is improved power efficiency, which lowers TCO. With compelling cost savings, we believe 48-volt will accelerate the adoption of power shelf architectures for both Tier 1 and Tier 2 hyperscale customers. AE is well positioned to capture the growth driven by this 48-volt migration. In fact, our 48-volt power shelf product launched earlier this year and is already in prototype stage with our customers. We are working closely with multiple customers who are developing their own custom 48-volt power solutions. The transition to 48-volt servers also creates new opportunities for AE's board-mounted DC-to-DC business and new opportunities for upgrading existing assets to 48-volt. In addition, power content is growing in the data center rack, driven by advanced processes, rack density and AI. Every generation of Intel processor comes with increased power ratings, enabling higher performance and driving increased rack density into the data centers. According to the leading data center and IT infrastructure association, AFCOM data center rack density has increased by more than 26% over the last 2 years. This growth in compute density has been accelerated by the increased number of artificial intelligence and machine learning applications, which are both compute and power intensive. As a result, AI server racks have 2x more power content than a traditional server rack. Our technology leadership and competitive advantages position us well to capitalize on these trends in hyperscale. Efficiency is one of the most critical performance factors in power supplies and 96% or above efficiency has become the standard requirement for hyperscale customers. We are shipping products today with efficiencies above 96% and have already demonstrated 98%. While many companies claim to be experts in power conversion, only a few have reached this milestone. As I mentioned, higher compute density allows our customers to deliver more in less space at a lower cost. Power density is an important enabler for high rack compute density. Our power supplies lead the industry in small footprint and watts per inch. In addition to our leadership in efficiency and density, our power shelf is highly differentiated. With communication, monitoring and control capabilities, our power shelves are customized to integrate with our customers' data center management systems and allow data center owners direct visibility into the power consumption and energy loss of the servers and battery backup units. This added functionality and increased product content of our power shelf expands our revenue opportunities to broader system-level solutions. By applying these technologies and working intimately with our hyperscale customers, we have been able to increase our market position from just 1 hyperscale customer at the beginning of this journey to now shipping in volume to 3 hyperscale customers. Over the last 3 years, we have added over $100 million of annual incremental revenue and we are just getting started. As we look forward to the next 3 years through 2023, we expect to add another $100 million annually or more. We are in the early stage of the ramp of our initial design wins with more new wins to come. We're confident that our system-level solutions, best-in-class quality, operations and customer relationships will enable us to gain meaningful market share, both at Tier 1 and eventually Tier 2 hyperscale customers. With the growing demand of the hyperscale market and our increased product content, we believe our SAM growth rate is 2x faster than the server power supply market. And as we gain share, we expect hyperscale revenue to grow at a compound annual growth rate of 40% to 50% from 2017 to 2023. Over the next 3 years, this target will add over $100 million of incremental annual revenue to our data center computing business. Moving on to industrial and medical. The Fourth Industrial Revolution or Industry 4.0 is driving many new smart applications. Enabled by advanced communications, IoT and the cloud, we're seeing a dramatic increase in the amount of data generated and consumed in today's factories and medical applications. We have the technology, configurability and channels to deliver custom power solutions to support these leading-edge applications. The industrial and medical market represents a large number of many applications. Over the last 12 months, medical represented just over 20% of revenue over the last 4 quarters and has been driven recently by critical care applications. We also provide solutions to many other applications in this highly regulated market. The characteristics of the medical market are very attractive to AE because it is highly regulated, requires high technical capabilities to serve with solutions that are mostly custom and comes with sticky relationships. It is also expected to see long-term growth driven by new innovative medical equipment. Our industrial revenue can be split between production and general. Industrial production consists of applications that are typically capital investments in manufacturing and automation, heavy industries and advanced material processing. Lastly, we sell into a wide range of industrial technology end markets with configurable solutions that typically track macro growth. Let me give you a few examples of where our power supplies enable smart products. In manufacturing, we believe power is at the center of the smart everything trend. And our advanced power solutions are enabling our customers to deliver on the promise of Industry 4.0. Power supplies, we sell into manufacturing applications, have built-in control, monitoring, connectivity and digital functionalities. In horticulture, we recently announced a new patented power and control system for lighting that is transforming the indoor farming industry. Our solution enables indoor vertical and greenhouse farms to reduce power consumption and lower cost while increasing crop yield. In materials processing, we have the most advanced process power products. Our solution is enabling our customers to develop equipment with the highest performance at the lowest cost. We have started to incorporate machine learning-based controls in these solutions, and we lead the way in industrial IoT integration with our customers' system. To address the diverse industrial and medical market, we have developed configurable power solutions. Our highly flexible platform allows customers to use existing off-the-shelf modules to create custom solutions. Our products also feature advanced monitoring and communications capability, so customers can more easily integrate these products into their systems. We have the broadest portfolio of medium to high-power configurable power systems and lead this $140 million market with over 35% share. This morning, we announced our latest addition to our configurable system family, the CoolX 3000. This 24-output platform is designed for a wide range of medical and industrial applications giving designers leading power density, unique flexibility, monitoring and control functionality. Beyond technology, our world-class customer engagement model enables both customer intimacy and the ability to serve thousands of customers across our markets through our global channel partners. Typically, we put the most energy and resources into the customers who require the highest level of customization. And in return, we earn the most value. These are key accounts where we have dedicated teams to engage deeply at both a strategic and commercial level. We often co-develop solutions and technology with these top customers and our field sales application engineers and design engineers often work at the customer site. In addition, we have what we call named accounts. These are customers who know our products well and with whom we've built long-term relationships. To best serve them, we use a mix of internal and partner resources which allows us to maximize serving our current business and increases our design win rate. Finally, our regional accounts consist of thousands of customers who use our standard products and configurable solutions for unique applications. For this group, we leverage our top-tier channel partners who have thousands of sales, technical and product delivery resources to support them. In addition to providing reach, our global distribution channel has become an engine to drive our design win funnel and growth. Over the last few years, as we acquired new businesses and diversified into new markets, our sales through the channel have grown substantially. With the Artesyn acquisition, we took a holistic approach in optimizing our global sales channel, deepening relationships, driving scale, improving customer outreach and accelerating cross-selling. The end results are a growing design win funnel and share gains with our customers. Before I finish, I want to share with you some of our initial success stories in cross-selling. We recently won a critical auxiliary power design for flat panel display equipment, leveraging our configurable power supply into a legacy AE customer. In addition, we are at the final stage of qualification for embedded power in a heater design for an electrostatic chuck. This design highlights new opportunities we uncovered from cross selling our embedded power products into semi. As a result, we have increased our WFE auxiliary power SAM from an estimate of $170 million we provided at last year's event to now closer to $300 million. In summary, we are strategically focused on hyperscale and the Smart Everything trend, which are the fastest-growing parts of the data center computing and the industrial and medical markets. We are leading these markets with advanced technologies and differentiated solutions. Our performance, market share gains and design wins have already yielded results. And we are still in the early innings of growth. We are realizing the opportunities presented by the Fourth Industrial Revolution, and we expect this megatrend will drive continued growth in our business. With that, let me pass the presentation to our CFO, Paul Oldham.

Thank you, Dana, and good afternoon, everyone. As a quick reminder, all financial measures presented today are on a non-GAAP basis. So far, you've heard Yuval discuss our strategy as a diversified technology growth company, and Dana talked about the exciting opportunities in 2 of our newer markets, data center hyperscale, and industrial and medical, all driven by investments in the Fourth Industrial Revolution and the data economy. Last December, we discussed our financial journey as a company, which was built on our strong foundation as the leader in power for the semiconductor market. We built a best-in-class financial model and strengthened the company's financial position while maintaining profitable earnings and positive cash flow every quarter through the last semiconductor downturn. We also laid out our strategy to diversify beyond semi, building a platform for earnings growth and delivering top-tier financial performance as a diversified industrial technology company. With our 8 acquisitions, we've added complementary technical capabilities, new products, and expanded our addressable markets to include a large exposure to smart industrial and medical applications, data center hyperscale and telecom verticals. One year after closing the Artesyn Embedded Power acquisition, we have significantly diversified our revenue base, delivered record financial results and more than doubled our earnings from a year ago. Our integration of Artesyn has gone well with synergies and accretion achieved beyond our Phase 1 target in about half the expected time. And we strengthened the balance sheet with strong cash flow, reducing our operating leverage from over 2x a year ago to just 1.3x today. As a result, we are increasing our aspirational goals and long-term financial vision, and believe we are delivering on our goal to be a top-tier diversified industrial technology company. Since we completed the Artesyn acquisition in late Q3 of 2019, we have steadily improved our financial performance, driven by market recovery in semi, increased share in our target markets and executed on our product and synergy plans. Despite the COVID-19 pandemic, we have delivered record levels of revenue, gross profit, non-GAAP operating income, earnings and cash flow. In particular, our third quarter gross profit dollars were 35% higher than our previous peak, and both operating income and earnings increased more than 3x from the third quarter a year ago. This strong financial performance has come from more diversification across our end markets. In semi, we are delivering very strong performance with growth well ahead of the market and our peers. Looking forward, we remain confident in our ability to grow faster than the market. Our smart industrial and medical market has been the most impacted by the global pandemic over the last year, partially offset by our success in medical applications. Looking forward, a global economic recovery should be a tailwind for this market. We have seen significant growth in our data center and computing market, primarily driven by share gains in hyperscale. While the industry is facing near-term digestion, our new trough level should be higher than the previous peaks at over $60 million per quarter. Finally, telecom remains somewhat limited by geopolitical factors and COVID, but our mix of 5G-related product is rising. We have executed well on our integration of Artesyn and are ahead of plan. As Yuval has said before, a crisis can drive organizations apart or bring them together. In our case, responding to customer and operational challenges brought the organizations together, allowing us to seamlessly integrate into a single functional company with strong leadership for each of the organizations. As a result, we have already achieved greater than $30 million in annualized synergies and over $1 per share in accretion, well ahead of our Phase 1 targets, and as I said, in approximately half the time. While we have additional synergies to deliver, our Q3 results were already approaching our long-term fully integrated model of over 40% gross margin and 20% operating margins, giving us confidence in our ability to exceed our aspirational goals. As a result of our strategy to expand our addressable market, we now have a much more diversified revenue portfolio than we did just 18 months ago. We are more than just a semi subsystems provider with broader market exposure to multiple secular investments and macro factors. And while each one of our markets can still be lumpy, the cycles do not usually move in tandem, reducing total revenue cyclicality while keeping the benefits of the upside. You can see from historical pro forma data that our peak to trough improves by over 1/3 and our standard deviation for the mean is reduced by more than 50%. Based on these characteristics, we have moved beyond the relatively small and captive semi process power market to become a broad-based industrial technology leader. As a result, our peer group is also expanding beyond our traditional names to include global power solution providers and diversified industrial technology companies. Over time, with our accelerated earnings growth and diversification, we believe AE can earn the higher and more consistent PE multiples associated with this peer group. We believe the higher multiple will come as we continue to execute and deliver top quartile financial performance relative to this group. On every measure from gross margin to operating margin, working capital efficiency or return on invested capital, we are either already in the top tier or expected to be there as we execute our strategy. Now let me switch gears to discuss our markets and financial model. Overall, our $9 billion SAM is growing at a low single-digit rate. However, by focusing on the faster-growing parts of our SAM, we will grow faster than our addressable markets. And as we execute our strategies, we can grow faster still. In semi, we believe the process power market will continue to grow faster than WFE as a result of increased power intensity in etch and deposition applications. In addition, our strategy to extend our leadership by delivering compelling products and growing our content should allow us to grow at over 1.2x WFE growth, excluding the impact of EUV. In data center, our focus on hyperscale and ability to continue to gain share should add over $100 million in annual revenue growth over the next 3 years. Although this market will be lumpy, we believe we will continue to have higher highs and higher lows. Industrial and medical is a very broad market and generally grows at a little better than long-term GDP growth rate of 2% to 3%. However, our focus on smart applications, leadership and configurable power supplies and expanded channels should allow us to grow more than 2x GDP. Finally, in telecom and networking, we see stable growth and expect to grow faster than the market as we focus on 5G infrastructure and next-generation networks. Overall, our market growth plus our targeted strategies would lead to underlying long-term compound annual growth rates in the high single digits. At the same time, we have been clear about our strategy to optimize our product portfolio. which will be a headwind to revenue but improve our returns on R&D dollars invested. We have already begun this process and have seen the initial benefit in our financial results. In all of our markets, our products had a distribution of gross margins with some below the corporate average and some above. Our strategy is to shift that distribution to the right, phasing out lower-margin products and focusing our R&D and operating resources on higher value-added opportunities. This is a gradual process, an evolution, not a revolution. As a result, revenue may be lower than it would otherwise be, but we believe we can improve gross profit and the profitability. In broad terms, we expect revenue growth to be impacted by 200 to 300 basis points reducing our net growth from high single digits to 5% to 6% CAGR. However, earnings leverage will be higher, and we will have a stronger portfolio of precision power products over time. As a result of these strategies, we expect earnings to grow over 50% in the next 3 years. This will be a function of executing on our remaining synergies, market growth, our strategy is to grow faster than the market and the impact of portfolio optimization. Overall, this should result in incremental margins of 35% to 45% and earnings growing at over 2x revenue growth rate. This supports our updated 3-year aspirational goals. Relative to our recent results, we are targeting to grow revenues by approximately 20%, earnings by well over 50% from $4.62 to over $7.50 per share while continuing to invest in critical programs to drive growth. Underpinning this model is a framework of net revenue growth in the mid-single digits, leveraging our cost structure to deliver 35% to 45% incremental operating margins while investing in critical programs, as I mentioned, and delivering earnings that grow greater than 2x revenues. We also expect to maintain our top-tier performance on return on invested capital. Although revenues may vary year-to-year, depending on the economic and market cycles, we believe this framework is very reasonable and sustainable given the growth drivers in our markets, our strategies and our track record of execution. Looking beyond our original financial framework, growth can be enhanced through smart M&A. We serve a large and still fragmented market for precision power solutions and believe we can strategically add products and scope while leveraging our growing scale as a company. Our assumption would be to add around $500 million in revenue at a reasonable purchase price and initial profitability that we can then leverage through synergies over time. This is basically what we've done over the last 6 years. Assuming we can add $500 million in new inorganic revenue, and extend the same organic financial framework I just discussed, we believe it is very reasonable to be able to deliver $2.5 billion in revenue; operating margins greater than 21%; and earnings of $12 per share in the next 6 to 8 years. We believe this vision of our financial performance represents a guiding path for creating shareholder value over the long term. Our long-term vision is supported by our strong balance sheet and capital allocation strategy in which we allocate the majority of our cash flow to growing the company. Over the last 6 years, we have spent over 70% of free cash flow in acquiring 8 companies. As we look forward, we are adjusting our capital allocation to 75% of free cash flow for growth, primarily through M&A. We have strong financial liquidity with $105 million in net cash, gross debt leverage around 1.3x and an additional $400 million available in our current debt facility, allowing us to use debt for larger transactions if we choose to do so. Our capital structure also allows us to return a portion of cash to shareholders. We have historically done this through an opportunistic share repurchase plan. Over the last 6 years, we repurchased over $200 million of stock at an average price of just under $39 per share, taking advantage of market variability. Our overall goal is to offset dilution, but the opportunistic nature of the program has allowed us to realize a net reduction in shares over time. Looking forward, today, we are announcing that we will also be initiating a regular quarterly cash dividend program as part of our shareholder return allocation. This decision is supported by our proven financial strength, increased scale of the company and our track record of execution and being profitable and generating operating cash in both up and down cycles. The initial dividend will be $0.10 per share per quarter. This reflects approximately 10% of free cash flow and yields approximately 0.4% based on the current share price, with room to increase over time. Overall, we believe this more balanced capital allocation approach will not impede our ability to grow and reflects our confidence in our strategy and commitment to consistently return capital to shareholders. In summary, our focus as a pure-play power leader of precision power conversion solutions enables our market leadership, sustainable advantages and the ability to drive scale across our business. Our markets have good long-term growth drivers, underpinned by the Fourth Industrial Revolution and the data economy, including investments in 5G, artificial intelligence and IIoT in a smart interconnected ecosystem, which needs higher power and more complex power solutions. Our strategies of extending our leadership in semi, growing share in hyperscale, leveraging our portfolio and sales channel for smart industrial and medical applications and focusing on 5G and telecom will enable us to grow faster than our markets. Finally, our continued efforts to drive synergy and leverage in our model will enable us to grow earnings more than 2x faster than revenues. And when combined with inorganic growth, enable us to achieve our long-term financial vision of $2.5 billion in revenue and $12 in earnings per share. With that, I will open it up to your questions.

[Operator Instructions] Your first question will come from the line of Scott Graham of Rosenblatt.

Yes. I really have one question and one follow-up. Could you talk a little more about your 48-volt ambitions here? Do you have revenues from 48-volt? Or is it more -- is it just a prototype right now? And how many design wins do you have?

We'll have Dana answer the question.

Sure. I'd be happy to, Yuval. 48-volt right now is a product that is being evaluated and ready for designing a next-generation server. And really the timing of the design wins and when we will get revenue from the stream will really be up to the road maps for each one of the customers that we have. We did launch our 48-volt power shelf product earlier this year, and it's already entered prototype testing with some of our customers.

Got it. This one is more for you, Paul. And on the aspirations, certainly, like the strategy of moving toward more of a margin-centric model with giving up a little bit of top line. How do you see that playing out? Are you essentially saying that you're thinking 8% to 9% growth over the next year? And then once that gets going, you kind of morph down 2 to 3 years from now to 5% to 6% while margins rise? Can you just maybe give us a little bit of a time frames from 8% to 9%, to 5% to 6%, if you would?

Yes. It's a good question, Scott. So it's probably the inverse of what you said. We've already begun this program to optimize our portfolio and look at where we focus our energies. And we're seeing some of those results already in our improved margins that we've delivered over the last few quarters. At the same time, we've also probably seen some impact of lower revenues as a result of that. And so that is more of a real-time impact. And we talked about this being a 2- to 3-year program, kind of an evolution, not a revolution. We're not exiting any markets. We're not canceling any customers. It's more a matter of where we focus our energy in the portfolio. So the net impact of that, if you will, the 8% to 9% underlying growth potential is facing a current headwind of that 2% to 3% that we talked about.

So maybe to be more crisp about it, we have already begun the process of optimizing our pricing strategy, optimizing the applications we serve and optimizing the products that we would like to extend the life of and products would like to end the life of. However, we have really long, deep relationship with customers that have multiple product lines that they buy from us. And for that reason, it's a long-term process where we work closely with our customers with them to decide together which applications we exit off and which applications we pursue. And the whole idea is to increase our margins. And obviously, because this is a migration of product life cycle management and the investment in new products for new applications, it's something we expect to see over time, over time. The good thing, we have already saw the initial results quarter-by-quarter as we have demonstrated in 2020 that we are increasing our margins every quarter.

Yes, I can see that's very helpful. Paul, if I can just ask just sort of an adjacent question to this with the accretion on Slide 41. Are you essentially saying that you're sort of at a run rate of over $1 non-GAAP accretion right now. And that the 1 -- over -- the over $1.50 is that within your 3-year aspiration as of today?

Yes, that's right. We actually said in our third quarter earnings release that in the last 4 quarters, we had seen over $1 accretion. So we experienced that today and would expect to see that sort of annualized accretion continue to increase over time.

Your next question will come from the line of Quinn Bolton of Needham & Company.

Thank you guys for the presentation materials this evening. Maybe the first question for Yuval, as you look at 2021 and your sort of expectations to outgrow semiconductor WFE, it seems to be dependent on the growth of etch and depth. When we look at Applied Materials, in their recent comments, they sort of said looking into 2021 that DRAM and foundry/logic would grow faster than NAND, NAND maybe flattish. Given the etch and depth intensity in NAND, if NAND is flat, does that change your thinking about how 2021 shapes up?

Well, I don't think we're trying to be market forecasters, Quinn. However, we are in a fantastic position to outgrow the market. And I think on the last earnings call, we shared our view relative to the first half of 2021 and what we see coming. We still have a really strong position, accelerated by recent growth we saw in areas like RF matching, our growth in RPS and additional products we introduced in the market that are going to support not only the capacity add, but also the migration to new tools and new technologies that will accelerate our growth. So without commenting on the general semi forecast for 2021, which, in general, we believe, is going to be a growth year, we expect to outgrow the market.

Appreciate it. And a second question for Dana. You sort of outlined the efficiency benefits of moving to 48-volt in the data center. Why do you think it's taken so long for the hyperscalers to move en masse to 48-volt? You mentioned only 1 of 10 hyperscalers is there today with wide scale adoption and I think even some of the compute intensive racks today are still 12-volt, what's taking so long to enable that conversion?

I think it's just the evolution of where they've come from a product set. They started out buying standard-based products from the OEMs out in the market, began to then slowly migrate to a platform where they built their own products. Then they started developing racks, then they started working with ODMs and CMs to build broader solutions that they could deploy themselves. And in general, they've been following along, trying to drive efficiency and power density levels to the point where they could get to the next level. So 48-volt, I think, is a natural step. And as you know, there's only one that's publicly doing it and announced doing it today, we really see the traction for them gaining in the coming year or so with what we've been prototyping with them. And we really think it's a matter of the timing of when they're willing to roll out new solutions based on new data center builds.

And I guess that forecast 2023 data for 5 of the 10 to adopt 48-volt, is that just in their compute intensive racks that require 30 to 60 kilovolts or kilowatts per rack? Or is that across the entire data center from your perspective?

It's more or less in the compute section at this point. I think it will expand though pretty quickly.

Great. And then last question for me for Paul, you didn't address the manufacturing consolidation in this presentation. I think in the Analyst Day last year, you talked about 6 facilities ultimately getting down to 3. Can you give us an update sort of where you are on that facility consolidation program? Because I know that's a key part to some of the margin benefits over the next 3 years.

Yes. Good question, Quinn. As we look forward, we've said that the Phase 2 of our integration is where some of the heavy lifting in manufacturing occurs, where we would see the benefits of closing factories and consolidating some of the activities. And that's still the case. So we've talked that we'd expect our Shenzhen factory to complete the closure during next fiscal year. We have another factory that's being consolidated kind of as we speak. And so what we've said is that we'd expect those things to happen over the course of the next several quarters.

Your next question will come from the line of Krish Sankar of Cowen and Company.

I had three of them. First, and also, by the way, congrats on the dividend initiation. So the first question I had is for Paul. In the long-term model, like the $1.65 billion revenue, what does the revenue composition looks like between semi and Artesyn? Is semi less than 40% of the mix? Because I would have thought the op margins would have been higher if you had more semi in the mix on the 3 years, $1.65 billion revenue target. And then I have a few -- 2 more follow-ups.

Yes, it's a good question. We haven't broken out what that mix looks like when we go out to 3 years. But if you look at the growth rates we talked about, so you would expect to see fast growth in hyperscale because of our share gains. You'd expect to see some improvement in industrial and medical in part because there should be some tailwinds in those markets because of the data economy -- I'm sorry, because of the macro economy and how much that's been impacted this year. So then it comes down to your assumption on semi market growth, which we hadn't talked about only that we would grow faster than the market. So on balance, I don't anticipate that mix plays a large portion of the change in gross margins. It's more a function of the synergies that we would drive over time.

Got it. Got it. And then a question for Yuval on the remote plasma source. My understanding is that one of the -- is the value proposition on this that it has a very high lifetime and has enough plasma that can generate to clean the whole chamber, and that's where the cost of ownership advantage comes to the customer. Or is there other value proposition? And if so, is there a way you can quantify how much better yours is versus the incumbents like MKS?

So good question. So if you look at where RPS is being used, the large application space goes to chamber clean. There are some other remote plasma source applications that go into processes like ALD. The fact is that we have been a leader in plasma enhanced ALD using our remote plasma source for many years, with our older remote plasma source that was practically the best in the market. The chamber clean area was an area that, although very large, we did not have access to, due to, as you know, in IP limitations that we had until 2019. Now in 2019, the IP limitation was removed, and we basically worked on launching our next-generation remote plasma source that is aimed at the largest application space, which is a chamber clean. The problem is that because of lack of investment in the industry and lack of innovation, the market was hungry for remote plasma source products that will be reliable, that will ignite every time, that will be much more precise in terms of power control and accuracy and have much higher lifetime of the applicator of the chamber, where the plasma is. And our product, the MAXstream that we launched in the market have all these attributes. And in fact, we're being pulled across the board by a massive amount of companies that are asking for a new supplier to enter this market. And it's a sizable -- it's not insignificant. It's about $150 million revenue per year and our market share in this specific application space was very small because of the IP limitations that now is practically removed because we have the freedom to operate. So we look at that area as an area for focus and growth. We have invested in the talent and the R&D teams and the experts that basically allowed us to develop this new product that we just launched last week and it's pretty exciting. It's really exciting.

Got it. Super helpful, Yuval. And then final question for Dana. You spoke about the hyperscale adoption down the road of the 48-volt product. I was under the impression that the telecom for customers already use 48-volt and ABB is the main player there. So I'm kind of curious when hyperscale starts looking at 48-volt, is there a risk that you run into the bigger players like ABB down the road?

We don't see that happening today. And really, the adoption in the hyperscalers is also driven around Intel and what Intel has been driving from the development of the next-level chips. So it's something we're keeping an eye on with them, but we also are working closely with the telecom providers. Right now, we just don't see that as a competitive threat. I mean the designs are so different. They're completely different use of the technology.

Your next question will come from the line of Tom Diffely of D.A. Davidson.

Paul, I'm curious, when you went through and you decided to focus on the higher-margin business, how did you balance going for margin percentage versus maybe giving up some margin dollars from the other programs?

Yes. It's a great question, Tom. And this is partly why it's an evolution. As we looked at the effort we were spending in various applications that we try to do a more longer-term financial analysis so that we could migrate our efforts towards areas that we believe we had we could add more value with the customers. I mean that should be a win-win between us and the customers, and I think the customers see it that way. And so it was more of a financial analysis that looks at a multiyear view about what's the value that we can add and what's the cost that we have to apply to continue to support a set of products that have -- that may have lower margins. And again, it's not a phase out immediate of things. It's more of an evolution. And we believe we can do that and continue to grow revenue at the same time and improve our gross profit dollars.

Okay. And most of those programs are on the Artesyn side of the business? Or did you rationalize the semi side as well?

Well, I think when we acquired Artesyn, we were pretty open that, that was one of our strategies because the prior ownership of Artesyn kind of view any revenue as good revenue and any margin is incremental maybe without taking into account all of the costs or opportunity costs involved. But some of the tools that we've implemented to drive this, things like 80/20 analysis and things like that, we actually -- have actually applied across our entire business, and there's some areas into the legacy business where we can make some improvements as well. But predominantly, it's on the embedded power side.

Yes. Tom, we implement that similar methodology across other acquisitions as well, including, as Paul said, some of our legacy products. It's just being disciplined about which products need to end life and which products need to basically be renewed or invested in. And thus far, we managed to really move the needle on ASP, ASP and gross margins.

Great. Okay. And then final question, when you look at the manufacturing side of Artesyn, were you able to take anything from the Artesyn side to benefit the semiconductor side of your manufacturing?

I'll take it.

Go ahead, Paul.

I'll take the cut at and then ask Yuval to add some color. The answer to that is absolutely. If you look at Artesyn, this is a well-run organization with an extremely capable manufacturing footprint. And we've been able to take some aspects of supply chain management, volume purchasing. We've looked at some business practices. We've been able to apply more broadly. And even in some cases, we've begun the process where certain subsystems or components that we had outsourced previously. We can effectively in-source to ourselves into our vertically integrated factories. So there are several things that I think that we've been able to leverage from the Artesyn capability even into our more traditional markets. And the spirit of that is try to leverage the best of both companies in terms of practices and capabilities. So Yuval, you want to add anything?

That's a good answer. Without going into detail which products, that's a good answer.

Okay. I assume that just sets you up for became a very nice kind of multi-product, multi-industry manufacturer going forward?

Right. So Tom, what you're going to see is overall, higher factory utilization of the combined company, overall, you're going to see [ better version ] of the combined company because we're shrinking our operation -- manufacturing footprint. The consolidation of the supply chain gave us tremendous buying power and synergy in terms of approved vendors list, et cetera, and more strategic relationship with suppliers and lastly, to the point that Paul made selectively insourcing some components to ourselves allows us to reduce our cost of material for products we would like [ insert ], and ultimately, our flexibility to be able to manufacture close to our [indiscernible] because of our multiple factories, global footprint, and in some cases, purposeful redundancy, allows our customers to decide where to buy from [indiscernible] for us to decide where to manufacture, overall reducing cost, reducing complexity, reducing tariff, reducing issues related to trade wars, et cetera. So that's -- overall, it's a multifaceted strategy that address business continuity, cost and flexibility.

Your next question come from the line of Paretosh Misra of Berenberg.

So first question on your portfolio optimization, just curious like what kind of products you might be looking to deemphasize and what end markets do they go to?

It's across all verticals. No one specific area. Especially since -- due to the acquisition, we have inherited a very broad portfolio of products. And if you look at the slide that Paul presented, every portfolio will have a distribution of margins. Some of them, older products, been there, have a low margin. Some of them are products that have unique applications, maybe customization that have higher margins. So it's really -- I don't like to use the word cherry picking, but it's like laser shooting at targets, right? So it's across the board. It's not only one area, it's broad, and it presents for us a target-rich environment, but at the same time, as we said earlier, we have really long-term relationship with really important customers, and we worked really closely with them to make decisions together how do -- how we migrate from a certain portfolio to another. And that's why we call it evolution, not revolution.

Understood. Next, probably for Paul. How should we think of capital spending in the years ahead to achieve your organic growth targets, particularly for the -- some of the things you described for the data center business. Would that require any large investments?

Yes. We believe that CapEx, broadly speaking, should run about 3% per year. It's a little less than that this year because we've scaled back on some things given the uncertainty of COVID. But I think if you look at that range, 3% plus or minus from a capital expenditure, that allows us to replenish and maintain capacity in the factories, upgrade some of our IT systems, things like that and make the other facility and engineering investments that we need to make.

Got it. And last one for me, any thoughts on the electric vehicle market? Any opportunities there that might make sense for you?

At this stage, we're not looking at the EV market as an area that we would like to pursue now. Highly commoditized, many players and in fact, AE historically was late to the market. However, what we see as an interesting outcome of this market is technology adoption, a new power technology that emerges and is being adopted in other areas adjacent to EV in the market. Now we do have some presence. We sell power supplies to charging stations. But it's not a giant revenue stream yet. Strategically, there are other target applications that we believe we can bring sustainable, competitive advantage and value to our customers. One example is the medical equipment.

Your next question will come from the line of Pavel Molchanov of Raymond James.

First one for Paul, great to see the dividend being initiated. Are you ready to commit at this point to a progressive dividend policy on annual increases? Or should we assume some other approach?

Yes. I think what we said, Pavel, is that at the level we introduced the dividend, there's room to grow. Obviously, that's something that we will evaluate each year.

Okay. In Slide 45, you talk about 8% to 9% growth -- top line growth for the company on average versus SAM growth of 4% to 5%. Just to clarify, does that ascribe credit for uplift from M&A, the fact that you're growing at twice the level of SAM?

Yes. So just to make sure you have all the pieces, so we're growing twice the level of SAM with the headwind of the portfolio optimization, which we talked about earlier, which gets us more into the mid-single digits or mid-single digits plus. That is all organic. So any M&A would be in addition to that.

Got it. And as it relates to M&A, historically, before Artesyn, you talked about industrial and medical being kind of the principal focus area. Is that remaining the case? Or are you going to cast the net more widely, potentially into some adjacent areas?

So Pavel, I think the important thing is we remain focused on pure-play power. And we have a lot of opportunities coming our way from all kind of industry to expand beyond power, and this is not what we're going to do. We continue to look at pure-play power. Within the area of pure-play power, we're looking at the niche where there is a need for highly engineered precision power solution for critical applications in regulated market. If you look at that, then medical has become a very interesting area, and it is indeed an area of interest for us. We have entered the medical equipment market and through acquisitions. And our intent is to continue at growing it, both organically and inorganically. However, if there are areas where we believe we bring unique value and become an enabler for difficult applications, we will definitely look at that. The last comment about M&A, we have always looked at that strategically. The Artesyn acquisition took us for the -- all the verticals for the Fourth Industrial Revolution in a data economy. As we look at other target applications, both small and large, we have opportunities to expand our technology, broaden our product portfolio and grow our SAM. And that's the way we look at that. And indeed, as we stated earlier, the medical equipment, life science equipment, diagnostic equipment, all of these areas are areas that are of high interest to us because we know we bring unique capabilities to the table, and these are highly regulated, very sticky long-term relationship between suppliers and customers, which we operate in really well in this environment, as we demonstrated in the semi world, and that's why indeed medical equipment is a target area.

Your next question will come from the line of Weston Twigg of KeyBanc.

I have two, the first one is around the telecom segment. And you may have explained it in the past, but I'm just wondering why you're not seeing more of a 5G tailwind right now, given that the ramp is well underway. Is it just that you're boxed out of certain geographies because you mentioned geopolitical risk? And then I'm also wondering, given that this market segment is the smallest TAM of the ones you outlined -- or SAM, I guess, I should say, the ones you outlined. Does it really still make strategic sense to keep this market going? Or is there some cross-selling or technology investment here that you can transfer to the other segments?

Dana, do you want to answer that or should I?

You want to go ahead and I can add on.

Yes. So two things, Wes. We have power content in 4G and 5G, right? What we see right now is happening in the world is the migration as 4G is displaced by 5G over time. We're still at the early stage, especially outside of China and early stage of the dynamic of the market investment. Two things, number one, due to geopolitical pressures, the Huawei equation, the split of the world between those who buy from Huawei and those that will never buy from Huawei, we are more aligned with the western world suppliers and have been for many, many years, strategic suppliers to them, and we expect to continue to grow with them as they migrate to 5G and also add additional content, the things like remote radio power supplies, et cetera. So we expect to continue to serve the market and grow within the market, and we expect to grow more than the market. However, if you look at 5G, it's a displacement strategy, right? As the infrastructure migrate to 5G, you're going to see a decline in 4G. Dana, do you want to add anything to that?

No, Yuval, I think you hit the key points there. I would just reiterate that our focus has been strategic alignment from an engineering-to-engineering standpoint with innovative products with the western suppliers for a long time. And what we're very focused on with them is taking advantage of next-generation efficiency and densities that allow them to do more with their next solutions and in more ruggedized applications, et cetera. So we feel well positioned with the partners that we're choosing to, and we're keeping that as a very focused, smaller set of customers.

And also to add to that, Wes, the rate of migration of the 5G will depend on the investment cycle and rate in North America and Europe. And that will determine the capacity requirements, capacity need. We do have the products, we do have the technology, we do have a partnership, really close and strategic relationships, right now, basically, are influenced by the investment cycle.

I see. That's very helpful. And then the other question is just really related to the strong semiconductor outperformance this year. And I get this question a lot. But I'm just wondering, I know there's some parts restocking, given the strong demand at some of your equipment customers. How much of a tailwind was that this year? And how much of that tailwind fades next year? And I guess I'm also wondering, do you have enough new products? You mentioned a couple today or other technology trends that would just offset any of the potential kind of headwind related to the restocking maybe being close to done?

So look at, as we explained in the last earnings call, we had a uniquely strong Q3. It was a combination of push ins and pull in and kind of a normalized -- the capacity constraint we initially had with COVID-19, now we believe that inventory is normalized. We believe inventories normalize and we agree with some of you guys when you talk about growth expected at least in the first half of 2021, driven by the investment cycles that, that we see in memory, et cetera. The other thing that's going on, and we're excited about that, 2 trends right now drive additional content in semi. One, as you know, Wes, the throughput of the new device, the process throughput is unacceptable. And there is an increased investment right now, not only in advanced technology, complex 3D dimensions and new materials, but there is an investment around the new design of platforms and tools that will allow the end user, the fab to get more throughput. I mean some of these process steps are so long. They can't get enough wafers at their fab. So throughput becomes an issue and that drives highly dense platforms with more chambers for platform. The complexity of the 3D devices will drive more power supplies and more advanced power supplies. And now we see an interesting trend. Pumping more power to the chamber is not a good enough solution. Historically, every generation -- remember, we used to talk about that as we go from one generation to another generation, we need more power and more complex controls. As we presented in the last Analyst Day, the industry right now is trying to avoid pumping more RF power to the chambers. So now it's not how much power you can get into the chamber, is can you develop the right power beyond RF to be able to enable the process and these very complex process steps and allow the customers to have the throughput and to have the uniformity, repeatability, reproducibility and shape control without using tremendous amount of RF power that you pump between the chamber? The problem is that not only create plasma damage, today, the chambers get so hot that now the fabs are getting problem with getting rid of heat, right? So the industry is moving to the next-generation type of power supplies, which is exciting for us because we're leading in this area.

Okay. That's helpful. But I guess in short, it sounds like you think you should be at your target goal of growing 20% plus of the industry next year and that the restocking is essentially over in the tech tailwinds and the power tailwinds kick in.

Wes, I can't provide any forecast. As Edwin said, we can't provide forecast. What we can tell you, our goal, as we have demonstrated, our ability to grow faster than the market.

And that's all the time we have for questions today. I will now turn the call back over to Mr. Yuval Wasserman for closing remarks. Please go ahead.

Okay. Thank you. Thank you, everyone, to -- for joining us today. We appreciate your time. And these are exciting time for AE. 2020, as you saw the first 3 quarters, were really amazing for us. As we look into the future, we have a clear strategy and a clear road map and a path to a higher level of performance. And for that reason, not only we have increased and updated our 3 years aspirational goals, we have shared with you our, long-term goal to be greater than $2.5 billion in revenue as we shared with you. Exciting times. The integration of Artesyn is going really well. And we're looking forward to talking to many of you going forward. And as you saw, over the last few weeks, we have a drumbeat of new product launches. Over the last 3 weeks, we continue to launch new products across multiple applications and multiple verticals, all bring unique value to the industry, to the world and to our customers. Thank you very much for joining us.

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