Tech Giants Face Market Challenges: AI, Cloud, and Hardware Sectors Under Scrutiny

Super Micro Computer: Sell-Off Is Warranted - Finally Nearing The Buy Zone (NASDAQ:SMCI)

We shall discuss further why we have maintained our Buy rating, despite the stock market still reeling from the ongoing correction. We previously covered Super Micro Computer, Inc. (NASDAQ:SMCI) in May 2024, discussing why we had upgraded the stock to a Buy, thanks to the robust long-term generative AI trends, Nvidia's (NVDA) promising forward guidance, and Taiwan Semiconductor Manufacturing Company's (TSM) positive commentary. Combined with the improved margin of safety after the April 2024 pullback and the relatively healthy balance sheet, we had recommended a Buy once a floor had materialized. Since then, SMCI has further pulled back by -22% attributed numerous market forces, worsened by the bottom-line miss in the FQ4'24 earnings call, despite the excellent FQ1'25 guidance. We believe that this dip remains a Buy for growth oriented investors looking to capitalize on the robust long-term generative AI market trends, as signaled by numerous market players. We shall further discuss. The Sell-Off Is Warranted - SMCI Is Nearly Back To Earth SMCI YTD Stock Price SMCI has had a rather volatile 2024 indeed, as observed in the stock's immense rally by +330.5% since the start of the year, with much of those gains also moderated by the time of writing. The same trend has also been observed in varying degrees with other semiconductor stocks/ ETFs, including Nvidia (NVDA) and VanEck Semiconductor ETF (SMH), as the market rotates from high growth/ generative AI stocks. With most of these stocks already recording massive YTD rallies and the market expecting an overly rapid AI monetization cadence in the SaaS layer, it is unsurprising that these stocks, SMCI included, have been drastically corrected upon the slightest whisper of recession. This is despite SMCI already reporting an excellent FQ4'24 earnings call, with revenues of $5.3B (+37.6% QoQ/ +143.1% YoY) and adj EPS of $6.25 (-6% QoQ/ +78% YoY) and FY2024 numbers at $14.94B (+109.8% YoY) and $22.09 (+87% YoY), respectively. While some of these numbers may have missed the lofty consensus estimates, we believe that the market may have been "looking a gift horse in the mouth," since the company's ability to rapidly ramp up its capacity to meet the insatiable data center capex demand has been impressive indeed. This is especially since the FQ4'24 bottom-line miss is attributed to supply chain issues, which "delayed about $800 million of revenue shipments to July, which lowered our EPS for June and will be recognized in our September quarter." At the same time, while there has been near-term noise in SMCI's FQ4'24 gross margins of 11.2% (-4.3 points QoQ/ -5.8 YoY) and FY2024 margins of 14.2% (-3.9 points YoY), the management has already highlighted that it is partly attributed to the "higher initial costs in ramping production of new DLC AI GPU clusters" and competitive pricing to win new designs. With its Malaysia manufacturing facility set to ramp up in capacity by H2'24, we believe that we may see the company's gross margins eventually improve to its long-term target range of between 14% to 17%. This is especially since SMCI has continued to report a relatively healthy balance sheet at a reasonable net-debt-to-EBITDA ratio of 0.37x in FQ4'24, compared to the average ratio reported in the Semiconductor Equipment & Materials industry at 0.39x and Computer Hardware industry at 0.47x. The Consensus Forward Estimates Despite the mixed FQ4'24 performance, SMCI has in turn offered a rather optimistic FQ1'25 midpoint guidance as well, with revenues of $6.5B (+22.6% QoQ/ +206.6% YoY) and adj EPS of $7.48 (+19.6% QoQ/ +118% YoY). Combined with the extremely impressive FY2025 revenue guidance of $28B at the midpoint (+87.4% YoY), it is unsurprising that the consensus have raised their forward estimates to a top/ bottom-line growth at a CAGR of +57.7%/ +50.9% through FY2026, respectively. This is compared to the original estimates of +33.2%/ +30% and the historical growth at +18.2%/ +35.1% between FY2016 and FY2023, respectively. These numbers further demonstrate why the SMCI CEO's commentary of: We are well positioned to become the largest IT infrastructure company, driven by our technology leadership including rack-scale DLC liquid cooling and business values of our new Datacenter Building Block Solutions. The investments in Malaysia and Silicon Valley expansions will further strengthen our supply chain, security, and economies of scale necessary for the growing AI revolution, has not been overly ambitious indeed, significantly aided by the expanded capacity in Malaysia and the US - as the management looks to tap into hyperscale datacenter business, building upon its existing enterprise market. Most importantly, the consensus forward estimates and the CEO's commentary have been well-supported by TSM's higher FY2024 capex guidance, with AI-related demand still structurally robust no matter the pessimism observed in the stock market. The same has been observed with data center REITs, such as Digital Realty Trust (DLR) and American Tower (AMT) in their latest earning calls, as they guide intensified capacity build outs "to maximize sellable capacity on the back of ongoing record demand." Combined with Gartner's projections of the global spending on public cloud services growth from $561B in 2023 to $823B by 2025, expanding at a CAGR of +21.1%, with "AI-related workloads driving a significant portion of this growth," we believe that SMCI remains well positioned to generate multi-year profitable growth ahead. SMCI Valuations As a result, we believe that the ongoing normalization in SMCI's FWD P/E valuations has been unwarranted - one that we had observed since the peak March 2024 levels of 43.99x. Even now, the stock's FWD P/E continues to moderate to 19.20x, down from its 1Y mean of 22.84x and likely to near the 2022 mean of 9.29x. Despite so, it is undeniable that SMCI remains extremely cheap at current levels attributed to the consensus raised top/ bottom-line growth prospects, as discussed above. Even when compared to SMCI's direct peers, including Hewlett Packard Enterprise Company (HPE) at FWD P/E valuations of 9.06x with the projected adj EPS growth at a CAGR of +0.7% through FY2026, and Dell Technologies Inc. (DELL) at 12.67x/ +15.1%, respectively, we believe that the stock sell-off and discounted valuations have been overly done. Even so, with the stock market still reeling from the ongoing correction, we will also be adjusting our fair value estimate and long-term price target calculations (in the next segment) using its 1Y P/E mean of ~22x for an improved margin of safety. So, Is SMCI Stock A Buy, Sell, or Hold? SMCI YTD Stock Price For now, SMCI has already lost a staggering -49.8% or the equivalent -$30.32B of its market capitalization from the YTD highs of $1.22K, while trading below its 50/ 100/ 200 day moving averages. For context, we had offered a fair value estimate of $598.20 in our last article, based on the LTM adj EPS of $19.18 ending FQ3'24 and FWD P/E valuations of 31.19x. This is on top of the long-term price target of $1.18K, based on the consensus FY2026 adj EPS estimates of $37.84. Based on the FY2024 adj EPS of $22.09 and the lower P/E valuation of 22x (as discussed above), it appears that SMCI continues to trade at a premium to our updated fair value estimates of $485.90. Even so, based on the consensus raised FY2026 adj EPS estimates of $40.57 and the same P/E, there remains an excellent upside potential of +44.6% to our updated long-term price target of $892.50. As a result of the still attractive risk/ reward ratio, we are maintaining our Buy rating for the SMCI stock, though with a few caveats. Risk Warning With SMCI continuing to chart lower lows and lower highs during the recent sell-off, it appears that the stock has yet to meet bullish support - an expected development given that it continues to trade at a notable premium to our estimates. At the same time, with NVDA's Blackwell delivery likely to be delayed from Q4'24 to H1'25 instead, we may see SMCI's revenue recognition temporarily delayed, and stock performance impacted over the next few months. At the same time, FY2025 is about to bring forth a relatively tougher YoY comparison at double digits, after the company reports triple digit YoY growths in FY2024, with it potentially triggering a further moderation in its valuations. This is on top of the numerous macroeconomic events, including the unwinding Yen Carry Trade, the inverted SPY movement against the 125-day moving average, the elevated VIX at heightened levels, and the perceived delay in AI payoff in the SaaS layer. Combined with the uncertain Fed rate cut and the ongoing US election, it goes without saying that there may be more volatility in the near-term, prior to the start of the next recovery cycle upon the normalization in market sentiments and macroeconomic outlook. Therefore, while we remain optimistic about SMCI's long-term prospects, we recommend investors wait for further pullback before adding for an improved margin of safety. I am a full-time analyst interested in a wide range of stocks. With my unique insights and knowledge, I hope to provide other investors with a contrasting view of my portfolio, given my particular background.Prior to Seeking Alpha, I worked as a professionally trained architect in a private architecture practice, with a focus on public and healthcare projects. My qualifications include:- Qualified Person with the Board of Architects, Singapore.- Master's in Architecture from the National University of Singapore.- Bachelor in Arts from the National University of Singapore.If you have any questions, feel free to reach out to me via a direct message on Seeking Alpha or leave a comment on one of my articles. Analyst's Disclosure: I/we have no stock, option or similar derivative position in any of the companies mentioned, and no plans to initiate any such positions within the next 72 hours. I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article. The analysis is provided exclusively for informational purposes and should not be considered professional investment advice. Before investing, please conduct personal in-depth research and utmost due diligence, as there are many risks associated with the trade, including capital loss. Seeking Alpha's Disclosure: Past performance is no guarantee of future results. No recommendation or advice is being given as to whether any investment is suitable for a particular investor. Any views or opinions expressed above may not reflect those of Seeking Alpha as a whole. Seeking Alpha is not a licensed securities dealer, broker or US investment adviser or investment bank. Our analysts are third party authors that include both professional investors and individual investors who may not be licensed or certified by any institute or regulatory body.

Microsoft: Its Overvaluation Finally Bites Back After The Pullback

With MSFT still reporting rich Free Cash Flow generation and a healthy balance sheet, we believe the market overreacted. We previously covered Microsoft Corporation (NASDAQ:MSFT) (NEOE:MSFT:CA) in May 2024, discussing its robust FQ3'24 earnings call results, with the growing partnership with OpenAI and expansion of in-house AI capabilities delivering excellent Azure cloud results. Combined with the tailwinds from the new PC refresh cycle, we had continued to rate the stock as a Buy for those looking to dollar cost average at every dip. Since then, MSFT has rallied by another +11.4% to new heights of $460s, before drastically pulling back as the market rotated from high-growth stocks and the labor market was impacted, triggering intensifying recession fears. This is worsened by the management's softer forward guidance and the potential implication from the previous CrowdStrike outage. Even so, we shall discuss why we are maintaining our Buy rating, with the recent selloff triggering an improved risk/ reward ratio and an attractive entry point, with the article focusing on MSFT's Cloud business opportunities. The Mag 7 Pullback Is Likely Here MSFT YTD Stock Performance We believe it is no secret that a correction is here, with it moderating much of MSFT's YTD gains along with the other Mag 7 stocks, worsened by the perceived slower AI payoff (an erroneous notion, in our honest opinion). The former's optics have also been further worsened by the CrowdStrike (CRWD) outage, with 8.5M Windows machines affected by the software update being just a small "subset of devices actually affected." Given that the update operates at the kernel level - effectively taking the whole computer system down, we believe that the impact may be worse than the reported sum of 8.5M, potentially in the tens of millions - impacting global operations across hospitals, banks, airlines, TV broadcasters, supermarkets, and many others. While the management is currently looking at "reducing the need for kernel drivers to access important security data," it remains to be seen how their future remediation may be implemented indeed. Otherwise, MSFT has reported double beat FQ4'24 earnings call with revenues of $64.7B (+4.5% QoQ/ +15.1% YoY) and adj EPS of $2.95 (in line QoQ/ +9.6% YoY), with most of the tailwinds naturally attributed to Intelligent Cloud revenues at $28.5B (+6.7% QoQ/ +18.7% YoY) and Azure at +30% YoY growth. With the management already reporting over 60K Azure AI customer (+60% YoY), along with higher cross-selling within its vertically integrated data, developer, Copilot for Microsoft 365, Customer Relationship Management [CRM], and Enterprise Resource Planning [ERP] platforms, among others, we believe that its AI adoption and monetization has been robust indeed. This development plays into the market's expectations as well, naturally justifying MSFT's hefty FY2024 capex of $44.47B (+58.2% YoY) and the management's guidance of "FY2025 capital expenditures expected to be higher than FY2024." MSFT remains more than well capitalized to continue investing in its growth driver as well, based on the still healthy balance sheet at a net cash position $23.93B despite the expensive Activision acquisition and the rich Free Cash Flow generation of $74B in FY2024 (+24.5% YoY). Unfortunately, while these numbers appear to be excellent, it is undeniable that its cloud delivery may fall somewhat short of the market's lofty expectations in the near-term. For one, MSFT has offered a lower FQ1'25 Azure growth guidance of +28.5% YoY, with it naturally disappointing analysts attributed to their expectations at over +30% YoY. This is despite the management's commentary that the lower guidance is attributed to "capacity constraints and non-AI growth trends," with things to accelerate by H2 of FY2025, "as our capital investments create an increase in available AI capacity to serve more of the growing demand." Unfortunately, with Blackwell delivery delayed due to production issues and MSFT's AI capacity expansion likely to be temporarily impacted, we believe that the stock couldn't have avoided the recent correction indeed. This is especially attributed to the tougher YoY comparison to Azure's growth at +27% YoY in FQ4'23 and +27% in FY2023. Secondly, the same underwhelming cloud development is also observed, when comparing MSFT Intelligent Cloud's lower unearned revenues of $23.11B (+7.1% YoY), compared to the AWS backlog at $156.6B (+18.5% YoY) and Google Cloud at $78.8B (+30% YoY). Combined with Intelligent Cloud's lower market share at 23% (-2 points QoQ/ +1 YoY), compared to AWS at 32% (+1 points QoQ/ in line YoY) and Google Cloud at 13% (+2 points QoQ/ YoY), likely attributed to the same capacity headwinds, it is unsurprising that the stock was sold off after the recent earnings call. MSFT Valuations Lastly, these teething issues may have also contributed to MSFT's relatively slower projected bottom-line expansion at a CAGR of +15% through FY2027 (CY2026), compared to AMZN at 36.8% and GOOG at 19.7%. While MSFT has gotten increasingly larger and well-diversified in capabilities over the past few years, especially due to the recently completed Activision acquisition, it is apparent that the stock has been overly expensive at the recent peak of 36.9x pre-correction. This is compared to the previous article at 34.8x, the 5Y mean of 31.3x, 10Y mean of 25.7x, and post-correction levels of 29.9x. Anyone concerned about MSFT's recent dip must also note that the same moderation has been observed in its hyperscaler peers, Amazon (AMZN) at 31.1x and Google (GOOG) at 19.7x by the time of writing, compared to their 5Y means of 79.6x/ 25.8x and 10Y means of 93x/ 23.72x, respectively. Nonetheless, when combined with MSFT's underwhelming forward guidance and impacted cloud market share, it is undeniable that this is a classic case of where elevated P/E valuations come with great expectations, with any earning misses and/ or underwhelming forward guidance likely to bring forth potential corrections, despite the highly strategic partnership with OpenAI. As a result of the near-term uncertain market conditions, we will be revising our fair value estimate and long-term price target (in the next section) using MSFT's more moderate 5Y P/E mean of 31.3x for an improved margin of safety. So, Is MSFT Stock A Buy, Sell, or Hold? MSFT 4Y Stock Price For now, MSFT has already pulled back drastically by -14.6% from the recent peak of $460s, while appearing to be well-supported at the previous support levels of $390s and nearing our previous recommended buy ranges of between the $385s and $400s. For context, we had offered a fair value estimate of $402.90 in our last article, based on the LTM adj EPS of $11.55 ending FQ3'24 and the FWD P/E valuations of 34.8x. This is on top of the long-term price target of $545.60, based on the consensus FY2026 adj EPS estimates of $15.64. Based on MSFT's LTM adj EPS of $11.80 ending FQ4'24 and the more moderate 5Y P/E mean of 31.3x, it is apparent that the stock is still trading at a notable premium to our updated fair value estimates of $369.30. Even so, despite the lowered consensus FY2026 adj EPS estimates of $15.33 (-1.9%) and the same P/E, there remains an excellent upside potential of +20% to our updated long-term price target of $479.80, thanks to the pullback. As a result of the still attractive risk/ reward ratio, we are reiterating our Buy rating for the MSFT stock, with a few caveats. Risk Warning Market Momentum Turning Pessimistic It is no secret that we have temporarily entered pessimistic market conditions over the past few weeks, with the CBOE Volatility Index hitting heights unseen before and the SPY breaching the prior 125 trading days rolling average, similar to October 2023 pessimism levels. We believe that the next few weeks may bring forth more volatility until the Fed pivots as projected by the September 2024 FOMC meeting. At the same time, after four quarters of robust top/ bottom-line growth, we believe that FY2025 may bring forth a tougher YoY comparison as observed in the market's over reaction to MSFT's softer FQ1'25 Azure guidance, with sentiments likely to remain impacted in the intermediate term. As a result of the potential volatility, we are lowering our recommended buy range to its next support levels in the $365s, with those levels also nearer to our fair value estimate while offering interested investors with an improved margin of safety. This dip is unlikely over yet. It pays to be prudent for now. I am a full-time analyst interested in a wide range of stocks. With my unique insights and knowledge, I hope to provide other investors with a contrasting view of my portfolio, given my particular background.Prior to Seeking Alpha, I worked as a professionally trained architect in a private architecture practice, with a focus on public and healthcare projects. My qualifications include:- Qualified Person with the Board of Architects, Singapore.- Master's in Architecture from the National University of Singapore.- Bachelor in Arts from the National University of Singapore.If you have any questions, feel free to reach out to me via a direct message on Seeking Alpha or leave a comment on one of my articles. Analyst's Disclosure: I/we have no stock, option or similar derivative position in any of the companies mentioned, and no plans to initiate any such positions within the next 72 hours. I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article. The analysis is provided exclusively for informational purposes and should not be considered professional investment advice. Before investing, please conduct personal in-depth research and utmost due diligence, as there are many risks associated with the trade, including capital loss. Seeking Alpha's Disclosure: Past performance is no guarantee of future results. No recommendation or advice is being given as to whether any investment is suitable for a particular investor. Any views or opinions expressed above may not reflect those of Seeking Alpha as a whole. Seeking Alpha is not a licensed securities dealer, broker or US investment adviser or investment bank. Our analysts are third party authors that include both professional investors and individual investors who may not be licensed or certified by any institute or regulatory body.

Amazon Stock: AWS Reloaded, Retail Charging (NASDAQ:AMZN)

In light of the continued strong profit growth prospects, the valuation of shares seems quite conservative at current levels. The past several quarters at Amazon (NASDAQ:AMZN) have been characterized by a swift rebound in retail margins. The previous status quo, where profits from cloud services financed investments in the retail segment came to an end. However, the most recent earnings print of the company showed that the majority of margin improvements in the retail segment could have been accomplished, which could turn back the focus to AWS. Luckily, demand for cloud services began to pick up again in recent quarters as companies continued their migrations to the cloud after a prolonged period of cautiousness. This, coupled with increasing AI-related revenues, could continue to fuel Amazon's profit engine in 2024 and beyond, supporting the share price after the recent pullback. I find the valuation of shares appealing at current levels, making me stick to my Strong Buy recommendation. Retail profit growth shifting into lower gear The complete overhaul of Amazon's fulfillment network has brought significant margin improvement to the North America segment since the beginning of 2023. The company's regionalization strategy, which took shape in the form of creating 8 self-sufficient regional networks, enabled it to handle increased transportation volumes more efficiently. Thanks to these efforts, Amazon managed to improve its North America segment operating margin to pre-pandemic levels of ~6%: In the light of the fact that Amazon has increased its offerings (including those available for same-day delivery) significantly over the past 5 years, this is a significant achievement in my opinion. This resulted in operating profits for the segment of ~$5-5 billion over the previous two quarters, not to mention the International segment, which began to finally generate operating profit as well. Based on management comments, the pace of margin improvement in these segments could shift into a lower a gear as low-hanging fruits have been picked, while further optimizations could take more time to bear fruit: "We have a number of opportunities to further reduce costs, including expanding our use of automation and robotics, further building out our same-day facility network, and regionalizing our inbound network. These cost improvements won't happen in one quarter or one fell swoop. They take technology and process innovation with a lot of outstanding execution, but we see a path to continuing to lower our cost to serve, which as we've discussed in the past, has very meaningful value for customers in our business." - Andy Jassy, CEO on Q2 earnings call Besides continued cost optimizations in fulfillment, the increasing proportion of advertising revenues within the North America segment should contribute to further margin improvement as well. In Q2 advertising revenues reached $12.8 million, growing 20% yoy significantly faster than online store's revenue growth rate of ~5%. Looking at competitors, Q2 growth rates have been somewhat softer in advertising than those for Q1, but I expect Amazon to turn the tide soon: The reason I believe that Amazon's advertising growth could diverge positively from competitors in the future is that there are still many untapped opportunities among its assets on the top of sponsored ads. The most important one of these could be Prime Video ads, which Amazon began to monetize really this year. Now that NBA games will join the streaming platform soon, combined with Thursday Night Football, the value of the platform to advertisers could increase further. Another important event in the advertising space has been the turning off of third-party cookies by Google this year, thanks to increasing security regulations. These are those cookies that are attached to users by websites other than those visited, making them an important tool in ad targeting. Now, advertisers are looking for efficient third-party cookie alternatives, so Amazon launched its new AI-assisted, Ad Relevance service in June. This could attract additional interest for its demand-side platform, where advertisers can display their expressions on Amazon's retail and non-retail properties. Another important development worth monitoring in the North America segment is the impact of Amazon's new fee structure on third-party sellers. Amazon introduced a new inbound placement service fee, which increased placement fees for sellers if they don't split up their shipments between different locations. This makes Amazon easier to deal with spreading third-party inventory, but it increases the charges for sellers, resulting in potentially lower seller activity. On the top of that, a new low-inventory fee came also into effect recently, making many sellers disappointed. The result of these changes could have already been evidenced in Q2 numbers, where revenue from third-party seller services grew only 13% yoy after growing 16% in Q1 and 19% in Q4 2023. As third-party sellers represent the core pillar of the company's advertising business, it's worth monitoring these trends closely. Based on the Q2 earnings call, management has also been somewhat surprised by these developments: Third, our seller fees are a little lower than expected given the behavior changes we've seen from our latest fee changes." - Andy Jassy, CEO, on Q2 earnings call Finally, everybody began to talk about a possible U.S. recession after the recently published non-farm payroll report, which sparked a significant sell-off in stocks. Although currently we are far from a recession, the shopping trends at Amazon seem to signal some caution among customers: ...we're seeing lower average selling prices, or ASPs, right now because customers continue to trade down on price when they can. More discretionary higher ticket items, like computers or electronics or TVs, are growing faster for us than what we see elsewhere in the industry, but more slowly than we see in a more robust economy. - Andy Jassy, CEO on Q2 earnings call These observations are in line with recently softening consumer confidence, which is represented by the survey of University of Michigan: After peaking around 80 in the first quarter of the year, there has been a decline in recent months with a reading of 68.2 in June. These developments are reflected in Amazon's online store business as well, where revenues grew 6% yoy to $55.4 billion after growing 7% in Q1 and 8% in Q4 2023. Although an unemployment rate ticking up to 4.3% doesn't give reason for concern in my opinion, in case these trends amplify, it could put pressure on Amazon's online and physical store businesses. The increasing share of everyday essentials in sales provides a natural hedge against a possible economic backdrop, but even with that in mind, it could still act as a significant drag on profit growth. Based on these observations, I believe the Retail segment of Amazon consisting of North America and International segments should continue to increase operating profits meaningfully over the upcoming years. A key driver of this will be continued growth in ad revenues, where a substantial portion of sales make it to the bottom line. Meanwhile, further investments into the fulfillment network should positively impact margins going forward, although to a lesser extent than over the previous 2 years. These investments also result in an increased product lineup that could positively impact sales, but this is currently counterbalanced by cautious spending patterns. Finally, the growing number of Prime memberships and the company's ability to increase subscription prices is a key element of maintaining profit growth in the Retail segment over the medium term as well. AWS stealing the spotlight again After the exponential growth in retail operating profits seemed to come to an end, investors' attention slowly turned back to Amazon's cloud and AI businesses. The AWS segment made an important comeback in recent quarters as its annual revenue growth rate started accelerating again. Besides customers completing their cost optimization efforts and reaccelerating their migrations to the cloud, the AI business of AWS has also been an important contributor to this. Looking at competitors, we can see a same tendency at Google Cloud, while Microsoft's cloud business growth experienced some moderation in recent quarters: I believe this could be an important cornerstone for AWS as it looked like at the beginning that Microsoft was better at monetizing AI, however based on these numbers, AWS is catching up nicely. I believe an important driver behind this is Amazon's chip business, especially its Graviton lineup, which entered its 4th generation recently and made up close to 10% of AWS revenues already a few years back. Lengthening server replacement cycles and cloud cost optimization efforts could've pressured Graviton revenue growth over previous quarters, but based on management comments, this is slowly coming to an end. The launch of Graviton4 could accelerate this rebound in the second half of 2024, providing a strong pillar for continued growth in the AWS business. AWS also launched its 2nd generation of Trainium chips recently, which provide a cost-efficient solution for training large language models. For inferencing workloads, there is Inferentia2 a chip designed in similar perception. Although the chips themselves seem to provide excellent price performance, their software support is lagging Nvidia's CUDA by a mile, a fact that even AWS employees admit. Currently, it doesn't seem there could be a breakthrough soon, so investors must wait until these chips achieve similar success than its CPU equivalents. The reason that the company's AI business reached a multibillion-dollar revenue run rate could be the early success of its AI services like Amazon Bedrock for building gen AI applications based on pre-trained models, or Amazon SageMaker that allows data scientists and developers to build their own ML models. Besides accelerating revenues in the segment, margins are holding up strong as well despite increasing investments in GPU capacity: Non-GAAP operating margin jumped significantly above 30% in recent quarters, showcasing real strength in AWS's operating model. However, this could change soon. Capex could top $60 billion in 2024, and most of it could support the growth of AWS business. This could be at least 25% higher than the $48 billion in 2024, where a large part went into the reorganization of the Amazon's fulfillment network. Based on this, Capex in the AWS segment experiences a significant increase during 2024, which will lead to increased depreciation expense going forward resulting in lower operating margin. If we assume that the company invests an additional $30 billion in AWS infrastructure this year compared to 2023, which will be amortized over 6 years (similar to useful life of servers) this means an additional $5 billion annual burden on the bottom line. Calculating with ~100 billion AWS revenue for 2024 this means a drag of 5%-points on operating margin, which will be fully felt in 2025, but already impact the company's bottom line this year. However, even with this negative impact we're talking about an operating margin of 30%, which is still very impressive among well-established companies. So, investors should prepare for decreasing margins going forward, which could be counterbalanced by a better price mix resulting from increasing share of AI-related revenues to some extent. Finally, another important risk factor to point out is the muted growth of remaining performance obligations over recent quarters, which are tied mostly to the AWS business: After a significant bump at the end of 2023, we could have witnessed stagnation over the previous two quarters. The yoy growth rate fell back to 19%, which is in line with the current revenue growth rate for the business, so this isn't concerning yet. However, if this tendency continues, it would mean a bad omen for continued topline growth reacceleration. I can imagine that more and more customers began to renew their commitments at the end of the calendar year, which could provide an explanation for this phenomenon, but anyway, it's worth monitoring this metric closely over the upcoming quarters. In a nutshell, Amazon's cloud and AI businesses showed encouraging signs in recent quarters, which could continue into the year. However, increasing investments could pressure margins going forward, which should be measured against improving topline growth going forward. Let's see how this could impact the valuation of share over the upcoming years. Valuation Based on the developments described above, I updated my relative valuation framework for Amazon. Resulting from somewhat softer consumer spending, I decreased the growth trajectory for the Retail segment. On the top of that, I also decreased my operating margin estimates for the segment, as further cost optimizations will take time to exert their effect. These effects result in lower operating income for the upcoming years than in my previous estimate. However, this is counterbalanced by better-than-expected growth rates and operating margin for AWS, which could remain the key profit driver through 2024 and 2025. Looking at the numbers, my valuation framework looks as follows: Calculating with an operating income of $62 billion for 2024, Amazon shares trade at a forward EV/EBIT ratio of 27.2. This is only 24% higher than the U.S. market average, which premium would decrease to 8% in 2025 and turn into a discount for the years that follow. As Amazon has significant growth levers to pull for the upcoming years, I believe its operating profit growth rate will be significantly higher than the market average, so its shares should continue to trade at a premium in my opinion. Based on this, I think its shares should outperform the market in general. To keep its current valuation premium, shares should outperform the market by almost 40% in 2 years, which is a realistic scenario in my opinion. In the case of positive fundamental surprises - like higher growth rates or margins in the AWS segment - this could be even higher. However, in the case of a more severe economic downturn leading to a collapse in consumer spending, this outperformance could be postponed for later years. Hi there! I'm a former Equity Analyst and a CFA charterholder writing mostly about holdings in my investment portfolio. These are overwhelmingly transformative technology names with perceived significant long-term upside potential. Once I've covered a company on SA, I try to update my views on it regularly.I like to do dig deep into company fundamentals and cover topics previous articles didn't elaborate on. Rigorous valuation is a must in my world. Feel free to share your opinion in the comment section, constructive critique is more than welcome! Analyst's Disclosure: I/we have a beneficial long position in the shares of AMZN either through stock ownership, options, or other derivatives. I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article. Seeking Alpha's Disclosure: Past performance is no guarantee of future results. No recommendation or advice is being given as to whether any investment is suitable for a particular investor. Any views or opinions expressed above may not reflect those of Seeking Alpha as a whole. Seeking Alpha is not a licensed securities dealer, broker or US investment adviser or investment bank. Our analysts are third party authors that include both professional investors and individual investors who may not be licensed or certified by any institute or regulatory body.

AMD: Finishing In Second Place During The AI & X86 Race (NASDAQ:AMD)

Combined with the abrupt departure of Victor Peng from the acquisition of Xilinx, we believe that market sentiments surrounding AMD's prospects are likely to remain depressed ahead. We previously covered Advanced Micro Devices, Inc. (NASDAQ:NASDAQ:AMD) in May 2024, discussing its underperformance in both stock and financial performance, as Nvidia (NVDA) took the lead in the generative AI race and Intel (INTC) in the x86 CPU race. With the gaming segment still reporting elevated inventory channels and slower IoT/ automotive recovery in the embedded segment, we had been uncertain about its intermediate term prospects, resulting in our reiterated Hold rating. Since then, AMD has fallen by another -11%, well underperforming the wider market at +3.4%. Part of the headwinds may be attributed to the ongoing market correction observed in high growth/ generative AI stocks, with the balance likely attributed to the lackluster FQ2'24 earning results and underwhelming FQ3'24 guidance. Combined with the abrupt departure of Victor Peng from the acquisition of Xilinx, it appears that AMD is likely to continue being relegated as the second best, resulting in our reiterated Hold rating. AMD's Data Center Investment Thesis Remains Underwhelming AMD YTD Stock Price At a time of generative AI boom and a supposedly insatiable AI chip demand from enterprises/ hyperscalers, two semiconductor companies have stood out as laggards, namely AMD and INTC. For context, NVDA has guided excellent FQ2'25 revenues of $28B (+7.5% QoQ/ +107.4% YoY), well exceeding the consensus estimates of $26.84B, with most of the tailwinds likely attributed to Grace Hopper while building upon the FQ1'25 data center sales of $22.56B (+23% QoQ/ +427% YoY). With that in mind, AMD only reported FQ2'24 Data Center revenues of $2.83B (+21.4% QoQ/ +114.3% YoY), with the absolute number being rather underwhelming despite the robust QoQ/ YoY growths. With INTC also reporting and guiding poor Data Center/ AI related revenues in FQ2'24 and FQ3'24, it is no secret that both companies have been unable to capitalize on the ongoing AI boom, underscoring why NVDA has been estimated to command the leading share of up to 90% in the AI GPU chip market/ 80% of the entire data center AI chip market. This is despite Microsoft's (MSFT) CEO having praised AMD's Instinct MI300X AI accelerators as the best "price-to-performance" AI chip on the market, as hyperscalers increasingly place orders for NVDA's next-gen chips, as observed with GOOG (GOOG) at $10B, Meta (META) at $10B, and Microsoft (MSFT) at $1.62B, with Amazon (AMZN) also opting to swap Grace Hopper orders for Blackwell. Based on these developments, we believe that NVDA's CUDA platform has been very sticky indeed, with it naturally being a barrier of entry for latecomers, such as AMD. At the same time, this development also demonstrates the hyperscalers willingness to spend and wait for top-tier performance, despite the supposedly hefty price tags and long-lead time. Therefore, while Blackwell's launch may have delayed due to technical issues, we believe that the potential boost on AMD's top-line may not be material, with hyperscalers likely to continue waiting for NVDA's next-gen offerings. The only silver lining to AMD's investment thesis is the sustained gains in its x86 CPU market share to 33.6% by Q3'24 (+0.2 points QoQ/ -1.5 YoY/ +10.5 from F2'19 levels of 23.1%), naturally eroding INTC's share to 63.5% (-0.4 points QoQ/ +0.9 YoY/ -13.4 from FQ2'19 levels of 76.9%). At the same time, AMD has reported expanding Client Segment operating margins of 6% (+13 points YoY), though still far from the 30.2% reported in FY2021 and 12.2% in FY2019. These developments imply that the ongoing PC refresh cycle has been a boon indeed, significantly aided by the new launch of AI-ready PCs with MSFT already projecting up to 50M units of AI PCs sold in 2024. The latter includes AMD's Ryzen AI 300 chipsets, along with other AI PCs featuring INTC's x86 Lunar Lake chipsets and Qualcomm's (QCOM) ARM-based Snapdragon X Elite CPU chips. As a result, we believe that AMD's Client Segment may continue to record robust top/ bottom-line numbers over the next few quarters. Even so, with the other segments, such as Gaming and Embedded still drastically impacted, it is unsurprising that sentiments surrounding its intermediate-term prospects have turned sour. Combined with the abrupt departure of Victor Peng, with it backpedaling AMD's unified AI strategy and vertically integrated ecosystem based on the MI300 Accelerated Processing Unit across the open-source ROCm and Xilinx's proprietary Vitis platform, we can understand the drastic correction observed in its stock prices and valuations. The Consensus Forward Estimates On the one hand, the consensus forward estimates for AMD's top/ bottom-line growth through FY2026 remain somewhat optimistic, with it signaling the eventual recovery in financial performance. On the other hand, at the risk of repeating ourselves, it is apparent that AMD is likely to be left behind by the ongoing data center/ AI capex boom, with the management only offering a FQ3'24 revenue guidance of $6.7B (+14.9% QoQ/ +15.5% YoY), compared to NVDA's annualized revenues at $112B. With the sales gap to NVDA painfully widening, AMD likely has to contend with INTC in both the data center processor and AI accelerator markets, with the latter's Xeon already powering data centers (thanks to the partnership with NVDA's H100 GPUs) and Gaudi 3 set to launch in Q3'24, with it supposedly offering "two-thirds the cost of competitive offerings." Only time may tell how AMD may perform. AMD Valuations As a result of the potential underperformance, the market has temporarily discounted AMD's FWD P/E valuations to 29.22x, down from its 1Y mean of 39.84x, 5Y mean of 37.57x, and pre-pandemic mean of 39.56x. While the stock may appear to be cheap at current levels, we believe that the market has merely re-priced AMD's valuations according to its (temporal) inability to directly compete with NVDA, currently at FWD P/E of 33.48x, while nearing INTC at 18.78x. Even when comparing AMD's projected top/ bottom line growth at a CAGR of +13.4%/ +20.2% through FY2026, to NVDA at +62.6%/ +89.3% and INTC at -0.6%/ +1.4%, it is painfully apparent that the former has a lot to catch up on - warranting the moderated P/E valuations. So, Is AMD Stock A Buy, Sell, or Hold? AMD 4Y Stock Price For now, AMD has already lost -43.3%, or the equivalent $133.23B of its market capitalization since the March 2024 peak, while trading below its 50/ 100/ 200 day moving averages. For context, we had offered a long-term price target of $244.50 in our last article, based on the consensus FY2026 adj EPS estimates of $7.41 and the 1Y P/E mean of ~33x. While those numbers remain largely in-line, with the moderate downgrade to $7.30 (-1.4% YoY) resulting in a still promising updated long-term price target of $240.90, we maintain our belief that the market sentiments surrounding AMD are likely to remain depressed, nearing that of INTC's. Prior to a healthier competition between AMD's and NVDA's Data Center prospects and the eventual reversal in the former's prospects, we are uncertain if it is wise to recommend a rating upgrade at current levels. This is especially since AMD has continued to chart lower highs and lower lows over the past few weeks as the market-wide correction occurs, with bullish support yet to be found. With the AMD stock likely to underperform its peers and the wider market, we believe that it may be more prudent to reiterate our Hold (Neutral) rating here. This may be a value trap. I am a full-time analyst interested in a wide range of stocks. With my unique insights and knowledge, I hope to provide other investors with a contrasting view of my portfolio, given my particular background.Prior to Seeking Alpha, I worked as a professionally trained architect in a private architecture practice, with a focus on public and healthcare projects. My qualifications include:- Qualified Person with the Board of Architects, Singapore.- Master's in Architecture from the National University of Singapore.- Bachelor in Arts from the National University of Singapore.If you have any questions, feel free to reach out to me via a direct message on Seeking Alpha or leave a comment on one of my articles. Analyst's Disclosure: I/we have no stock, option or similar derivative position in any of the companies mentioned, and no plans to initiate any such positions within the next 72 hours. I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article. The analysis is provided exclusively for informational purposes and should not be considered professional investment advice. Before investing, please conduct personal in-depth research and utmost due diligence, as there are many risks associated with the trade, including capital loss. Seeking Alpha's Disclosure: Past performance is no guarantee of future results. No recommendation or advice is being given as to whether any investment is suitable for a particular investor. Any views or opinions expressed above may not reflect those of Seeking Alpha as a whole. Seeking Alpha is not a licensed securities dealer, broker or US investment adviser or investment bank. Our analysts are third party authors that include both professional investors and individual investors who may not be licensed or certified by any institute or regulatory body.

Arista Networks: Pureplay Ethernet Switch Company At Its Peak (NYSE:ANET)

Arista's profitability, margins, and FCF growth indicate a strong business model, but high customer concentration poses a risk to its financial outlook. In this analysis, we covered Arista Networks, Inc. (NYSE:ANET), a communication equipment company that has experienced an impressive share price surge of 80.48% over the past 12 months and a staggering 448.79% over the past 5 years. Firstly, we determined whether the company's increase in share price is justified and measured if it will continue to surge by assessing the company's revenue generation model, its competitive landscape within the Ethernet switch market, key drivers of growth, and projected revenue growth based on average 5-year P/S ratio among the top competitors within the market. Secondly, we examined the company's earnings and margins growth trajectory to understand the company's business model and determine projected the earnings growth based on the average 5-year P/E ratio among the top companies within the Ethernet switch market. Lastly, we examined Arista's FCFs to determine the company's business model and whether it justified the surge in its stock price and compiled the P/FCF ratio across the top companies within the market to project the company's future FCF growth. Pure Play Advantage Driving Further Growth We first determine whether the company's surge in stock price is justified and examine whether it will continue to increase by first examining its revenue generation model, assessing the competitive landscape, and identifying the key drivers of growth within the market in which Arista is in. We updated the Ethernet Switch market share since our previous coverage of Cisco and compared the company's P/S ratio to the competitors from the Ethernet Switch market with the historical ratios and average. In the competitive landscape of our Ethernet market share chart, we further categorized the companies in terms of pure-play Ethernet switch vendors, communication equipment vendors, and non-communication equipment vendors. Based on the chart, Cisco (CSCO) continues to maintain its market leadership in the Ethernet switch market with a 43.7% share by Q4 2023. Despite being the lead, Cisco has been experiencing a steady decline in market share, dropping from 64% since Q4 2011. Besides, Arista's share has seen a consistent rise, climbing from 1.5% when it was first reported by IDC in Q3 2013, to 11.1% by Q4 2023. HPE (HPE) has also shown modest growth, with its market share inching up from 8.5% in Q4 2011 to 9.4% in Q4 2023. Another key player, Huawei, is also in a gradual uptrend, growing from a 2% market share in Q4 2011 to 9.4% in Q4 2023. In contrast, H3C has experienced an overall decline, with its market share decreasing from 3.6% when it was first reported in Q2 2016, to 2% in Q4 2023. The collective market share of smaller competitors, which are categorized as Others has also slightly increased, moving from 20.1% in Q4 2011 to 24.4% in Q4 2023. Overall, as of Q4 2023, the Ethernet Switch Market was still dominated by Cisco, followed by Arista, with Huawei and HPE tied closely behind, and H3C trailing behind. Category Definition Listed Companies Pureplay Ethernet Switch Vendors Companies that derive a majority of their revenues from manufacturing and selling Ethernet switches. Arista Network Communication Equipment Vendors Large companies that produce and sell diverse networking and communications equipment. Cisco, Huawei, HPE, Juniper Networks (JNPR), H3C, Alcatel-Lucent, ROM Devices Non-communication Equipment Vendors Companies that do not belong to the communication equipment industry, but have strategic partnerships and acquisitions that aim to expand reach and product offerings within the Ethernet switch market. Nvidia (NVDA), Broadcom (AVGO), Dell (DELL) Click to enlarge Source: Company Data, Khaveen Investments The above shows the categorization of companies into the 3 main groups. Pure play Ethernet switch vendor includes Arista, which has a strong focus on deriving revenues from manufacturing and selling Ethernet switches. Communication equipment vendors are defined as companies with relatively high market shares that produce and sell diverse networking and communication equipment such as LANs, WANs, and routers, this includes companies like Cisco, Huawei, HPE, Juniper Networks, H3C, Alcatel-Lucent, and ROM Devices. Additionally, we categorized companies like Nvidia, Broadcom, and Dell under Non-communication equipment vendors, which are companies that do not belong to the communication equipment industry but have strategic partnerships and acquisitions that aim to expand reach and product offerings within the Ethernet switch market. For instance, Broadcom, a semiconductor company, acquired VMware "to offer more comprehensive product lines, including cloud networking solutions". Similarly, Nvidia has strategically partnered with Cisco to develop "Ethernet networking-based solutions will be sold through Cisco's vast global channel", which expands Nvidia's reach towards the Ethernet switch market. Ethernet switches are devices that connect multiple computers or other network devices, allowing them to communicate within the same network by directing data to the correct destination. These switches are essential for building networks in homes and especially in environments like data centers and campus networks. In data centers, Ethernet switches play a role in managing the amount of data traffic between servers and storage systems. Besides, in the AI aspect, Ethernet switches also play important roles in handling AI workloads, which require efficient and fast network infrastructure. Campus networking, which involves the networking of large coverage areas, relies on Ethernet switches to connect different buildings and departments and ensure consistent network access. According to Global Market Insights, rising cloud demand and IoT devices are growth drivers. Cloud services require robust network infrastructures to handle larger volumes of data traffic while data centers support growing data storage and processing needs, particularly for processing AI workloads, which drives demand for high-performance switching solutions. Also, more IoT devices spur the demand for more network ports and Ethernet switches. P/S Comparison 2019 2020 2021 2022 2023 Current Average Arista Networks 6.82x 9.96x 15.57x 8.76x 12.75x 18.80x 10.77x Average 2.38x 2.76x 4.62x 3.15x 3.88x 5.06x 3.36x Click to enlarge Source: Company Data, Khaveen Investments We compiled the P/S ratios across the companies within the Ethernet switch market. In terms of Arista, the company's average P/S ratio has increased by 47.5% since 2023 and is observed to have the highest P/S across all its competitors over every period during the past 5 years. From our research, we believe Arista is a competitive company in the Ethernet switch space with an average revenue growth of 23.6% over the past 5 years. Additionally, the company is expected to grow at an average of 17.6% over the next 5 years based on consensus. We believe the growth drivers such as the increasing demand for cloud services, the expansion of data centers, and the surge in internet-connected devices, will benefit all three categories of companies but should benefit pure-play Ethernet switch provider Arista the most. This is reflected by Arista's strong revenue growth and highest market share gain (9.6% since 2013). While broad Communication equipment vendors such as Cisco, Huawei, and Juniper Networks may leverage cross-selling of their products to derive higher sales, we believe these companies lack the focus to develop more advanced and competitive Ethernet switches. This is observed with Cisco's Ethernet switch market share dropping significantly over the past decade due to its "greater exposure to the enterprise segment where demand is in a secular decline." In contrast, Arista's "focus on the hybrid cloud" has likely contributed to its market share growth as global demand shifts towards the cloud segment. Additionally, we believe the Non-Equipment communication vendors such as Nvidia and Broadcom also lack the strategic focus to innovate better Ethernet switches than pure play companies like Arista, and diverse Communication vendors such as Cisco and ROM Devices which both have relatively high market share in Data Center High Speed Ethernet ports. Arista's CEO also mentioned that the company "don't see NVIDIA as a direct competitor yet on the Ethernet side" and instead sees Nvidia as a partner for Arista to be "able to be the scale-out network for NVIDIA's GPUs and NICs based on Ethernet". We believe Arista will remain highly competitive moving forward given AI solutions already developed such as the Etherlink AI Networking Platforms designed to enhance the performance of networks that handle AI applications, as well as collaboration with Nvidia for new solutions such as the Arista AI Agent that mainly enhances the managing of AI workloads in GPUs and network interface cards (NICs). Valuation (P/S) 2024F 2025F 2026F 2027F 2028F Revenue ($ mln) 6,894 8,109 9,539 11,221 13,200 Growth % 17.6% 17.6% 17.6% 17.6% 17.6% P/S Ratio 12.49 9.00 6.48 4.67 3.36 Equity Value 86,107 72,947 61,798 52,354 44,352 Shares Outstanding 314.15 314.15 314.15 314.15 314.15 Price Target ($) 274.09 232.20 196.71 166.65 141.18 Current Price ($) 323.54 323.54 323.54 323.54 323.54 Upside % -15.3% -28.2% -39.2% -48.5% -56.4% Click to enlarge Source: GuruFocus, Khaveen Investments All in all, based on the P/S method, we valued the company by projecting the company's revenue in 2028 with an average of 17.6% over the next 5 years based on consensus. This is fairly in line with the H1 2024 actual revenue growth (16.1% YoY) as well as the Q3 2024 management guidance (14.9% YoY) according to its latest earnings transcript. Based on the current price of $323.54 and the 5-year average P/S ratio we compiled across the top companies within the Ethernet switch market, we derived a price target of $141.18 in 2028. Furthermore, we prorated this price target to derive our price target of $274.09 in 2024 which is a downside of -15.3%. Growing Profitability Next, we further examine the company's earnings growth and whether it justifies the surge in stock price by focusing on the company's business model and examining its margin outlook. We analyzed the earnings and margins growth of Arista and compiled the current and historical P/E ratio for the top companies within the Ethernet switch market and assessed the current position of Arista by comparing it with the market average. The company's margins overall show a positive trend in revenue growth with a 10-year average of 33.43%. EBIT margin has overall increased from 21.49% to 38.54% over the past 10 years. Net margins from the past 10 years have increased from 14.88% in 2014 to 35.62% in 2023. However, the company's gross margin has been in a slight downward trend from 67.13% to 61.95% over the past 10 years. Furthermore, the chart indicates economies of scale, which describe the cost advantages that a company achieves due to an increase in the scale of output, where the cost per unit decreases as production becomes larger. This is shown from 2015 to 2017 as well as from 2021 to 2023, where revenue uptrends together with EBIT and Net margin. This is also reflected from 2019 to 2020 where revenue declined along with a downward trend in the two margins. We identified the reasons for the increase in margins to be primarily attributed to a decrease in the R&D and SG&A expense in % of total revenue, from 25.5% in 2014 to 8.84% in 2023. To understand the factors behind the cost efficiencies, we compiled the past 14 quarters of earnings briefing transcripts since Q1 2021 and examined the list of reasons regarding the decrease in R&D and SG&A expenses. The reasonings from management include reduction in "new product introduction costs" and "other costs associated with the development of next-generation products". In terms of SG&A expenses, past declines are due to the reduction in "product demo costs" and variable expenses. During the latest earnings transcript, Arista guided operating margins to rise to 44% up from 38.5% in FY 2023. We believe this is achievable considering the economies of scale that Arista benefits particularly from SG&A and R&D cost efficiencies. Arista mentions specifically "supply chain productivity gains led by the efforts of John McCool, Mike Kappus, and his operational team". John McCool is the Chief Platform Officer which we assume leads the rollout of the Arista Extensible Operating System, a software platform that is "used on the network switch, to configure, monitor, and debug network problems on the server". P/E Comparison 2019 2020 2021 2022 2023 Current Average Arista Network 19.12x 36.32x 54.66x 28.42x 35.79x 50.01x 34.86x Market Average 15.31x 24.92x 23.98x 19.13x 23.72x 34.88x 21.41x Click to enlarge Source: GuruFocus, Khaveen Investments Similarly, we compiled the P/E ratios across the top companies within the Ethernet switch market. We observed that the company currently has a P/E ratio of 50.01x, which is the second highest among the listed rivals, trailing behind Juniper's P/E ratio of 53.23x. Arista's current P/E ratio is also moderately higher than the industry average (34.88x). Overall, we believe Arista has demonstrated notable improvements in its earnings and profit margins over the past decade. We identified that this is primarily due to a reduction in R&D expenses, which declined from 25.5% in 2014 to 8.84% in 2023, as well as a decrease in SG&A expenses from 22.2% to 8.8% over the same period, which illustrates an effective cost management over the past decade. Besides, the company further guided an EBIT margin expansion to 44% in FY2024, showing an increase of 5.5% compared to FY2023, which we believe is achievable as the company could continue benefiting from economies of scale. Valuation (P/E) 2024F 2025F 2026F 2027F 2028F Revenue ($ mln) 6,894 8,109 9,539 11,221 13,200 Net Margin 37.1% 39.0% 40.9% 42.8% 44.7% Earnings ($ mln) 2,557 3,162 3,900 4,800 5,896 P/E Ratio 41.51 35.06 29.68 25.18 21.41 Equity Value 106,142 110,842 115,750 120,876 126,229 Shares Outstanding 314.15 314.15 314.15 314.15 314.15 Price Target ($) 337.87 352.83 368.45 384.77 401.81 Current Price ($) 323.54 323.54 323.54 323.54 323.54 Upside % 4.4% 9.1% 13.9% 18.9% 24.2% Click to enlarge Source: Company Data, Khaveen Investments We further valued the company using P/E valuation by projecting the company's earnings until 2028. In 2024, we derived a net margin of 37.1% as we factored in the company's expected increase in operating margins based on management guidance (5.5%) as well as an addition of 1.9% per year for 2025 and beyond which we derived from the past 9-year EBIT average increase. Based on the current share price of $323.54 and the 5-year average P/E ratio which we compiled across the top companies within the Ethernet switch market, we derived a price target of $401.81 in 2028, which we then prorated to derive our price target of $337.87 in 2024, at an upside of 4.4%. Strategic Cash Position for Future Growth Lastly, we examine the company's FCF growth, as well as its acquisition expenditures and capex requirements and whether it justifies the surge in its stock price by analyzing its business model and deriving our outlook. We analyzed the FCF growth of Arista and the categories under cash flow from investing activities (cash acquisition, investment in marketable & equity security) and compiled the current and historical P/FCF ratio for the company and its top competitors within the Ethernet switch market. From the chart, the company's 10-year average FCF margin (capex Only) is 26.11%, and the 10-year average for investing FCF margin is at 9%. The reason for the difference comes from non-capex cash outflows. We observed large cash outflows from investments in marketable & equity securities throughout the past decade at an average of 24% of total revenue (excluding years 2015 and 2022 which the company sold these investments for positive cash flows). Additionally, in 2018, 2020, and 2022, Arista's acquisitions also increased its investing cash outflows, with an average of 5.87% of total revenue over these three years. In terms of capex, the company's capex was minimal at an average of 1.4% of total revenue in the past 10 years. In terms of acquisitions, the company acquired Mojo Networks (wireless LAN provider) and Metamako (low-latency networking provider) in 2018, which helped expand its reach in the wireless networking market and enhanced its switches performance. In 2020, the company acquired Big Switch Networks (SDN provider) and Awake Security (NDR provider) to further expand its reach in the SDN market and its capabilities to deliver "the next generation of operationally efficient network security and visibility". Further in 2022, the company acquired Pluribus Network (unified cloud networking provider) and Untangle Holdings (ETM provider), which helped the company to "automate the management of cloud networks". Additionally, we believe that the company's continuous investments in marketable securities are mainly aimed at maintaining its cash-to-debt ratio to prepare for future acquisitions. We can observe that this strategy is on track as its cash-to-debt ratio has remained stable at a 10-year historical average of 2.1x. The company's low capex across the past 10 years could be attributed to its business model, where Arista delivers products with software solutions and it also outsources most of its "manufacturing and supply chain management operations to third-party contract manufacturers", which we believe overall avoids the need for large investment in physical infrastructure. P/FCF Comparison 2019 2020 2021 2022 2023 Current Average (5-yr) Arista Network 17.37x 32.09x 48.25x 85.70x 37.35x 53.40x 44.15x Average 15.35x 15.89x 19.20x 41.97x 17.00x 21.67x 21.88x Click to enlarge Source: GuruFocus, Khaveen Investments Additionally, we compiled the past 5-year P/FCF ratios of Arista Network and its competitors. We observed that Arista's current P/FCF has risen by 43.0% to 53.4x despite a negative 56.4% growth in 2023 and it is well above the market average of 21.67x, which is a positive indicator along with the surge in stock price. Arista's high P/FCF ratio, which is also higher than the market average, is believed to be a result of low capital expenditure requirements and strategic investments in acquisitions and marketable securities. The company's software business model and reliance on third-party manufacturing overall contribute to its minimal capex needs. The company's consistent investment in marketable securities also helps to maintain a healthy cash-to-debt ratio for future acquisitions. Thus, we believe Arista could continue its trajectory of growth and further strengthen its market position given the benefits from minimal capex needs and a cash-to-debt ratio maintained for potential future acquisitions. Valuation (P/FCF) 2024F 2025F 2026F 2027F 2028F Revenue ($ mln) 6,894 8,109 9,539 11,221 13,200 FCF Margin (Capex Only) 33.8% 35.7% 37.6% 39.5% 41.4% Free Cash Flows (Capex Only) ($ mln) 2,330 2,895 3,586 4,431 5,462 P/FCF Ratio 45.05 37.46 31.24 26.11 21.88 Equity Value 104,986 108,440 112,009 115,695 119,502 Shares Outstanding 314.15 314.15 314.15 314.15 314.15 Price Target ($) 334.19 345.18 356.54 368.27 380.39 Current Price ($) 323.54 323.54 323.54 323.54 323.54 Upside % 3.3% 6.7% 10.2% 13.8% 17.6% Click to enlarge Source: Company Data, Khaveen Investments Similarly, we valued the company using a P/FCF valuation by projecting the company's FCF until 2028. In 2024, we projected an FCF margin of 33.8% factoring in an increase in operating margin of 5.5% as well as an increase of 1.9% per year from 2025 which we derived from the past 9-year EBIT average growth. Based on the current share price of $323.54 and the 5-year average P/E ratio which we compiled across the top companies within the Ethernet switch market, we derived a price target of $380.39 in 2028, which we then prorated to derive our price target of $334.19 in 2024, at an upside of 3.3%. Risk: High Customer Concentration Arista's financial outlook could be influenced by its heavy reliance on a small number of large customers, which is highlighted in the company's annual report, that "Microsoft (MSFT) and Meta Platforms (META) in fiscal 2023 and 2022 collectively represented 39% and 42%" of revenue. We believe its dependence on the two customers which account for approximately 80% of Arista's revenue indicates a risk that could largely impact Arista's financial performance if they switch to competitors. Verdict To conclude, Arista Network has gained a significant 9.6% market share since 2013 in the Ethernet switch market. This contrasts with broader communication equipment vendors like Cisco which have seen declines due to a lack of specialization to keep up with market trends as competition intensifies. We expect Arista's potential could be further supported by the strategic AI partnerships with Nvidia, to develop offerings in AI-driven networking solutions. Despite its strong revenue growth outlook, our P/S valuation indicates limited upside. Additionally, we believe the company to continue benefiting from economies of scale as well as from minimal capex needs given its business model to produce software solutions and the reliance on third-party manufacturing. However, our P/E and P/FCF valuations also indicate limited upside in 2024. Overall, we chose to base our price target of the company on our valuation using P/FCF, at $334.19 in 2024 as the company demonstrated robust cashflows throughout the past 10 years at an average of 26.1%. However, this translates to an upside of only 3.3%, thus we rate it as a Hold given the strong rise in the company's stock price in the past 1 year has increased by a whopping 80.48%. Khaveen Investments is a Macroquantamental Hedge Fund managing a portfolio of globally diversified investments. With a vested interest in hundreds of investments spanning diverse asset classes, countries, sectors, and industries, we wield a multifaceted investment approach that combines top-down and bottom-up methodologies, integrating three core investment strategies: global macro, fundamental, and quantitative strategies. We serve accredited investors throughout the globe, which include HNW Individuals, Corporates, Associations, and Institutions. At the heart of our investment prowess lies specialized expertise in cutting-edge technologies that are reshaping the fabric of numerous industries. Our strategic focus revolves around the transformative fields of Artificial Intelligence, Cloud Computing, 5G, Autonomous & Electric Vehicles, FinTech, Augmented & Virtual Reality, and the Internet of Things.www.khaveen.com Analyst's Disclosure: I/we have a beneficial long position in the shares of ANET either through stock ownership, options, or other derivatives. I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article. No information in this publication is intended as investment, tax, accounting, or legal advice, or as an offer/solicitation to sell or buy. Material provided in this publication is for educational purposes only, and was prepared from sources and data believed to be reliable, but we do not guarantee its accuracy or completeness. Seeking Alpha's Disclosure: Past performance is no guarantee of future results. No recommendation or advice is being given as to whether any investment is suitable for a particular investor. Any views or opinions expressed above may not reflect those of Seeking Alpha as a whole. Seeking Alpha is not a licensed securities dealer, broker or US investment adviser or investment bank. Our analysts are third party authors that include both professional investors and individual investors who may not be licensed or certified by any institute or regulatory body.

Meta Platforms: I Will Stay To The Sidelines (NASDAQ:META)

Nevertheless, we should be very cautious right now as the next few years might be challenging, and I'm staying on the sidelines. Last week, on July 31, 2024, Meta Platforms, Inc. (NASDAQ:META) reported second quarter results and not only did the business beat estimates for revenue as well as earnings, but the company also reported solid second quarter results and increased its top and bottom line with a high pace. Following earnings, the stock also jumped as one would expect on this news, but in the last few days since earnings were published it seems like the "bigger picture" took over and indices all around the world are suddenly in free-fall - especially in Japan the situation seems dire. In my last article about Meta Platforms published in February 2024, I was still bullish about Meta Platforms but also argued that the bullish case was getting weaker. My conclusion included the following statement: Over the long run, Meta Platforms is still a great investment, and I am confident Meta Platforms will trade for higher stock prices in five or ten years from now. And I also assume that the stock can run up to about $600 in the coming months (or quarters). Nevertheless, I would be a little more cautious at this point as a larger correction in the coming months gets more and more likely. A technical correction would also be the major risk I am seeing for Meta Platforms in the next few quarters and whoever is buying now might be a little late to the party or should at least not expect similar returns as in the last 12 months for the next 12 months. And it seems like I might have been right with my statement, although Meta Platforms did not run up to $600 (it reached $542 at least). It probably was a good call to turn rather cautious, but looking back I probably should have been even more cautious and rated the stock as a "Hold" already. In the following article, I will get more cautious and definitely rate Meta Platforms only as a "Hold". But I will explain once again why I think Meta Platforms is a great long-term investment with investors needing nerves of steel in the next few years (not only with Meta Platforms investors, I think investors will need nerves of steel in general, almost every stock investor will). Quarterly Results Let's start with the last quarterly results - and as mentioned above they were great once again. But it is difficult to draw conclusions from past results as the fundamental business is often lagging investor sentiment, and it also takes several months before results reflect troubles. Revenue for Meta Platforms increased from $31,999 million in Q2/23 to $39,071 million in Q2/24 - resulting in 22.1% year-over-year top line growth. Income from operations increased 58.1% year-over-year from $9,392 million in the same quarter last year to $14,847 million this quarter and diluted earnings per share increased even 73.2% year-over-year from $2.98 in Q2/23 to $5.16 in Q2/24. Only free cash flow declined slightly from $10,955 million in the same quarter last year to $10,898 million this quarter - resulting in a 0.5% year-over-year decline. And not only the income statement is great. When looking at other metrics, Meta Platforms is also performing great. The daily active people within the Family of Apps increased from 3.07 billion in Q2/23 to 3.27 billion in Q2/24 - resulting in 6.5% year-over-year growth. Additionally, ad impressions delivered also increased 10% year-over-year (with growth being especially high in the Asia-Pacific region). And average price per ad also increased 10% year-over-year (with growth being especially high in Europe and Rest of the World). Overall, the average revenue per person for the Family of Apps increased 14.1% year-over-year from $10.42 in the same quarter last year to $11.89 this quarter. Zuckerberg Going All-In Again? It seems like investors are getting scared once again that CEO Mark Zuckerberg might start spending huge amounts of money on projects that are not contributing to revenue or take several years or a decade before they might pay off (a situation that seems familiar). It probably was statements like the following take by Zuckerberg: At the end of the day, we are in a fortunate position where the strong results that we're seeing in our core products and business give us the opportunity to make deep investments for the future. And I plan to fully seize that opportunity to build some amazing things that will pay off for our community and our investors for decades to come. Or the outlook provided by CFO Susan Li had a similar vibe: For Reality Labs, we continue to expect 2024 operating losses to increase meaningfully year-over-year due to our ongoing product development efforts and investments to scale our -- to further scale our ecosystem. While we do not intend to provide any quantitative guidance for 2025 until the fourth quarter call, we expect infrastructure costs will be a significant driver of expense growth next year. As we recognize depreciation and operating costs associated with our expanded infrastructure footprint. And we can understand the skepticism of investors a little bit. When looking at the segment results, it is still the Family of Apps, which is responsible for the entire operating income and almost all revenue. Reality Labs generated $353 million in revenue. Compared to the same quarter last year ($276 million in revenue), the segment increased revenue by 27.9% year-over-year, but it is still only responsible for less than 1% of total revenue and still had to report a huge operating loss of $4,488 million. Compared to the same quarter last year, the operating loss increased as it was "only" $3,739 million in the same quarter last year. While revenue growth was driven by Quest headset sales, operating loss was driven by headcount-related expenses and inventory costs. I, personally, believe in the long-term vision of Meta Platforms and Mark Zuckerberg, and I think the investments will pay off at some point in time. But we should not forget that Meta Platforms are not only about the metaverse and a long-term vision but also have a profitable business with a wide economic moat. Long-term Vision Meta Platforms is not only investing in long-term growth (the metaverse) but also investing in its core business and future growth projects that will generate revenue not only in a decade from now. Threads For starters, Threads is progressing well and during the earnings call, Zuckerberg reported that the app is about to hit 200 million monthly active users. And although Zuckerberg is reporting deeper engagement it probably will take years before the app is generating enough revenue or contributes to profits. WhatsApp Business Messaging Another promising segment is business messaging on WhatsApp, which is the biggest contributor to the "Other revenue" among the Family of Apps. This segment generated $389 million in revenue in the last quarter and year-over-year reported 72.9% year-over-year growth. During the last earnings call, Mark Zuckerberg also pointed out that Business AI, which is currently in alpha testing, could be a major contributor to revenue: Over time, I think that just like every business has a website, a social media presence, and an email address, in the future I think that every business is also going to have an AI agent that their customers can interact with. And our goal is to make it easy for every small business, and eventually, every business, to pull all of their content and catalog into an AI agent that drives sales and saves them money. When this is working at scale, I think that this is going to dramatically accelerate our business messaging revenue. There are a lot of other new opportunities here that I'm excited about too, but I'll save those for another day when we're ready to roll them out. Core Business And while Threads and Business Messaging can drive growth in the years to come, the major part of revenue and operations is still stemming from the core business - the advertising business. And this business will be driven by two different factors (as management pointed out during the last earnings call). Revenue performance will be driven by the ability to deliver an engaging experience as well as the effectiveness of monetizing that engagement over time. The first factor is Meta Platforms' ability to deliver engaging experiences. And Meta Platforms is especially focused on including video and in-feed recommendations. More than half of recommendations in the United States are coming from original posts and especially in India retention and engagement on WhatsApp is particularly promising. Aside from driving engagement, Meta Platforms also must monetize this engagement. And in improving monetization efficiency, artificial intelligence is playing a central role. Meta Platforms is improving ad delivery by adopting more sophisticated modeling techniques made possible by AI advancement (like the Meta Lattice ad ranking architecture). And with the company's Advantage + suite of solutions, a 22% higher return on ad spend for U.S. advertisers can be reported. During the last earnings call, Zuckerberg also mentioned that Facebook is obviously becoming more popular again with young users, which is a good sign: I'm particularly pleased with the progress that we're making with young adults on Facebook. The numbers we are seeing, especially in the US, really go against the public narrative around who's using the app. A couple of years ago, we started focusing our apps more on 18 to 29-year olds and it's good to see that those efforts are driving good results. Intrinsic Value Calculation Now it gets trickier. As always, we are also calculating an intrinsic value by using a discount cash flow calculation, or we look at simple valuation metrics to determine if a stock is a good investment or not. And in the case of Meta Platforms, we have a business constantly reporting high growth rates (still double-digit growth, although growth rates slowed down) and analysts are also expecting high growth rates for the years to come. And as argued above, we have plenty of reasons to expect solid growth for Meta Platforms over the long run. In the last three years, Meta Platforms grew revenue with a CAGR of 16.21%, operating income with a CAGR of 12.69%, and earnings per share with a CAGR of 13.80%. When looking at longer timeframes, the annual growth rates are even higher. And between fiscal 2024 and fiscal 2029, analysts are expecting earnings per share to grow with a CAGR of 11.64%. On the other hand, we have a stock trading for reasonable valuation multiples. At the point of writing (and in a volatile environment), the stock is trading for around 25 times earnings as well as 25 times free cash flow and for a business which seems to be able to grow in the double digits with a high level of consistency such a valuation multiple certainly seems reasonable. We can also use a discount cash flow calculation to determine an intrinsic value for the stock. As a basis for our calculation, we can take the free cash flow of the last four quarters, which was $48,573 million. Additionally, we are calculating with a discount rate of 10% and with 2,610 million outstanding shares. When calculating with these assumptions, Meta Platforms must grow 6% annually from now till perpetuity in order to be fairly valued and when looking at previous growth rates as well as expected growth rates, we should argue that Meta Platforms is at least fairly valued and if the stock will continue to drop it might quickly become a bargain. But we must keep in mind that the free cash flow in the last four quarters was the highest number of Meta Platforms reported on a trailing twelve-month basis. However, when comparing free cash flow to revenue, Meta Platforms is generating about 33 cents of free cash flow from every dollar of revenue - a metric in line with past numbers and far away from the peak a few years ago. And although current revenue is not unrealistic high, we must keep in mind that in case of a global recession advertising revenue will probably stagnate or even decline. And with a lower free cash flow and maybe declining top and bottom line, investors suddenly might not attribute a 25 times valuation multiple to the stock anymore. And we must also keep in mind that 6% growth till perpetuity is a high growth rate. A business growing at such a rate till perpetuity would constantly grow (at least) at twice the pace of the overall economy in the United States. History has shown us several examples of companies outpacing GDP growth for several decades and I think Meta Platforms could also be one of these companies. But we can also make the argument that we should not calculate with more than 3% growth till perpetuity. In that case, Meta Platforms has to grow between 12% and 13% in the next ten years to be fairly valued - a growth rate that the company might not achieve in the next few years (considering a looming recession). A Warning In the last few days, the global stock markets suddenly shocked investors with high levels of volatility, and especially the stock market in Japan declined steeply. And I certainly will not say I told you so at this point as I certainly did not know the stock market will decline now. But I have mentioned again and again that we should be cautious as the stock market is extremely overvalued, and I warned that the inverted yield curve is sending a strong warning sign for an upcoming recession and bear market in the United States. The market will probably not plummet in a straight line and, of course, there is always the possibility of the stock going higher again. But the yield curve (especially the 10-year minus 2-year treasury yield) is giving us another warning sign. After the yield curve inverted several quarters ago, the yield curve is now getting close to re-inverting again - another step that almost always happens before a recession. And Meta Platforms is neither cheap nor expensive at this point and the stock is not trading at unreasonably high valuation multiples like some other AI-related stocks or other hyped stocks. But I still see the risk of Meta Platforms declining rather steeply in case of a global recession and steep bear market. Conclusion Meta Platforms is a great business and a great long-term investment. I, personally, will hold on to my shares, and I am pretty confident that 10 years from now the stock will trade at a higher price than right now. Nevertheless, I don't think now is the time to buy and over the coming quarters, I rather see the stock at $400 or $300 than at $500 or $600. And although the stock might seem undervalued at $400 or lower, we should be extremely cautious. The stock will plummet first before the business will follow and although Meta Platforms will grow over the long run, it might certainly take a hit in the next few years. We should stay on the sidelines at this point and Meta Platforms - like many other high-quality businesses - is only a "Hold". My analysis is focused on high-quality companies, that can outperform the market over the long-run due to a competitive advantage (economic moat) and high levels of defensibility. Focused on European and North American companies, but without constraints regarding market capitalization (from large cap to small cap companies).My academic background is in sociology and I hold a Master's Degree in Sociology (with main emphasis on organizational and economic sociology) and a Bachelor's Degree in Sociology and History. Analyst's Disclosure: I/we have a beneficial long position in the shares of META either through stock ownership, options, or other derivatives. I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article. Seeking Alpha's Disclosure: Past performance is no guarantee of future results. No recommendation or advice is being given as to whether any investment is suitable for a particular investor. Any views or opinions expressed above may not reflect those of Seeking Alpha as a whole. Seeking Alpha is not a licensed securities dealer, broker or US investment adviser or investment bank. Our analysts are third party authors that include both professional investors and individual investors who may not be licensed or certified by any institute or regulatory body.

Super Micro Computer: Just The Report I Was Waiting For (NASDAQ:SMCI)

Valuation assumptions suggest a 30% upside to the company's fair value, with potential risks including market fluctuations, margin deterioration, and competition. Super Micro Computer (NASDAQ:SMCI) recently reported FY24 numbers that were met quite negatively by the investing community, which drove its share price 20% on the announcement. I believe the report wasn't as bad as everyone thought. The community got hung up on the margin deterioration, which was not unsurprising, therefore, this report allowed me to start a position for the long run. The company's share price tumbled over 46% since the last time I covered it and gave it a hold rating due to uncertain margins and the overall valuation being stretched. With such a plummet in price in such a short time, I believe the company is attractive once again, therefore, I am upgrading it to a buy. Starting from the top, the company's top line came in at $14.9B vs $7.1B a year before. A whopping 109% y/y increase. It is even better than what I estimated in my previous article (104%). Diluted EPS came in at $20.09, compared to $11.43 a year ago, an increase of 75%, not as impressive as the top-line growth but that is mainly due to the company's efficiency deteriorating slightly more than anyone expected. Gross margins came in at 11.2% vs 15.5% for the third quarter, and 17% the same quarter a year ago. Many analysts were expecting GM to come in at around 14%, therefore, the company missed EPS estimates quite a bit, $6.25 vs $8.14 expected. The depressed margin numbers did not help the company's share price from a free fall on the day, even after announcing better-than-expected sales figures for the next year. The company expects to make $26B to $30B vs consensus estimates of $24.6B. That is a 90% growth from the FY24 figure (at the midpoint). The demand for its products is alive and well. So, everyone was expecting further margin deterioration this and a couple of more quarters probably. That was one of my biggest concerns in the previous cover of the company. That and the company's stretched valuation. As I mentioned in my previous article, I don't think it is too much to worry about, and I believe the market is announcing the death of the company just because the margins deteriorated slightly more than expected. Also, the fantastic growth in the top line more than offsets the loss in margins, as the absolute dollar amount of profit did rise, as evidenced by a 75% increase in EPS. In my opinion, the negative sentiment and some misses on key reports provide us with a decent opportunity in the long run. In terms of revenues, as I mentioned earlier, the demand for the company's products is there. This view is supported by the ever-increasing demand for AI-enabled data centers. Major cloud service providers, or CSPs, like Alphabet (GOOG), Amazon (AMZN), Microsoft (MSFT), and Meta (META), are investing billions of dollars in data center infrastructure annually. Just look at the below graph to see how much it has grown in just 8 years, and it is not showing any slowdown, which will bode well for SMCI, which is at the forefront of the revolution because it had the sense to aggressively expand its offerings when the management saw where data centers are heading. During their latest quarter reports, the management of the mentioned big tech CSPs said that most of the increases in CapEx will be prioritized on improving infrastructure regarding data centers and AI. As Gen AI becomes more and more sophisticated, power consumption will increase tremendously. Cooling of the data centers may not be a major issue as of now, however, the more power-hungry data centers become, traditional air cooling may not be enough in the next couple of years. There is a limit to how well it can cool down. The company's investment in direct liquid cooling racks is a testament to the management's strategic foresight. As the management correctly anticipated the AI boom and invested heavily in infrastructure to support the explosion in demand, now by recognizing the limitations of traditional air cooling and investing in DLC, will solidify the company once again as the market leader. The management expects to grow its DLC rack capacity to 3,000 per month by this year-end, essentially doubling the capacity. The company wouldn't do that if there was no demand for it, so I am confident the company's revenue growth will remain robust for quite a while. This supports the management's view of continued strength in this AI boom, citing the Q2 earnings call, "Overall, I feel very confident that this AI boom will continue for another many quarters if not many years. And together with the related inferencing and other computing ecosystem requirements, demand can last for even many decades to come, we may call this an AI revolution", as Charles Liang put it. The demand is not an issue. The big concern that drove down the company's share price a whopping 20% on the day of the announcement was the margin contraction that was larger than expected. So, what is the company going to do about that? Are they going to be able to ease these concerns? I think these are just momentary inefficiencies directly attached to the company's efforts to be the go-to solutions provider. This means aggressively investing in DLC, which clearly did a number on the company's margins. Once the company builds out its infrastructure over the next couple of years, like the new plant in Malaysia and scaling up in Taiwan, should help ease margin pressures and improve the overall cost structure, according to the management. Furthermore, the company's Datacenter Building Block Solutions 4.0, aims to reduce data center build time considerably for new data centers, from three years to two years, while for the existing data centers with older infrastructure, the BBS aims to reduce build times to a year, or even to half a year, depending on the center. This should translate into higher margins going forward. The management is confident that the company will achieve its long-term gross margin target of 14% to 17%, and with the initiatives the company is taking on, I could see this happen. Maybe not by the end of FY25, but in the long run, I can see the company reaching these numbers. Let's now look at some updated valuation assumptions. For revenues, we already have a sense of where the top line is going to stand by the end of FY25. I went with around $28.7B, which is 92% y/y growth, indicating a strong demand. After that, I decided to lower the growth quite considerably to keep it on the conservative end of things. For the next couple of years, the company will see 20% growth, which will taper off to around 5% by FY34, giving me a 21% CAGR. I think that is very achievable given how the demand for AI is still in the ramp-up stage, and if the company can keep up with the demand, I could see higher growth for the next few years at least. Nevertheless, I'll keep it somewhat conservative. I can always go back and change if some new information surfaces that changes my thesis. For margins, I went with the theme that over time, the company will manage to become much more profitable and regain its margins. I am modeling that the company's gross margins will remain subdued for the first couple of years, after that, they will start to climb up and by FY34, will reach 17%, which is where the company expects to. To be honest, I believe gross margins will get there quicker, but to keep it safe, I am taking my time to improve these. For the DCF model, I decided to use the company's WACC of around 9% and a 2.5% terminal growth rate because I would like the company, in the long run, to at least match the US long-term inflation goal. Furthermore, to give myself more room for error in the estimates above, I am adding a 20% discount to the final intrinsic value calculation. It's always better to be safe than sorry. With that said, SMCI's (conservative) intrinsic value is $672.5 a share, meaning there is at least a 30% upside to its fair value. In the short run, I don't doubt that we will see the company's share price fluctuating quite violently due to the overall jitters of the market. Many people are screaming that recession is on the way, so any further concerns regarding this will certainly affect a low-float company like SMCI. For the good or the bad. Further deterioration in margins will continue to impact the company's share price negatively, especially if we are going to get more surprises to the downside. An unexpectedly larger decrease in margins will wreak havoc on the short-term valuation of the company. Of course, we cannot forget the competition. Companies like Dell (DELL), and HP Enterprise (HPE). The industry is becoming commoditized, so it will be tough to maintain margins, especially when other companies are just as competent, and have deep pockets to undercut to gain market share. With the rapid technological advancements, companies with such deep pockets will come out on top. Even with what I believe to be rather conservative estimates, the company is a bargain at these prices. The short-term pain of lower margins will continue to weigh on the company's share price and may bring it down further, therefore, I only opened a small position right now, in anticipation of further price decreases. If it doesn't happen, at least I am having some exposure here. I am going to be following how the company's margins progress over the next year or two. This will give me time to decide whether I would like to increase my position or cut losses if the margins I modeled become unreachable. On the other hand, fantastic top-line growth may offset any loss in margins, especially if the growth is just as robust as it has been recently in terms of absolute dollar profits. The biggest concern I had with the company just 5 months ago was its stretched valuation, now that it has come considerably down, I believe the risk/reward is amazing at these levels.

Fortinet: Smooth Execution Of The Business (NASDAQ:FTNT)

The 25% rise in one day has left FTNT at a fair, but not demanding, valuation. It will continue to perform well in the long term. It's been 8 months since the last time I wrote about Fortinet, Inc. (NASDAQ:FTNT). Since then, the company has had a very volatile 2024. It was up +25% in March but dropped to -5% a week ago. After the Q2 earnings release, which we will discuss in detail, Fortinet's shares have surged to more reasonable prices given the quality of the business. In January, I rated the stock as a buy, and thanks to the recent rise, we have a 19.87% return compared to the S&P 500's 10.69% and the 5.36% of the security ETF I follow the most, the iShares Digital Security UCITS ETF USD Acc. As I explained in the first article, The thesis for Fortinet is its great reputation and expertise in specific areas of cybersecurity, accompanied by a highly scalable product portfolio, successful penetration into new market niches like SASE, excellent capital management, and founders with over 20 years of experience in the company and the sector. We are all aware that cybersecurity is a very complicated sector for the generalist investor to understand, which is why in this type of company, the management teams play a critical role in the thesis. Fortunately for Fortinet, this might be its strongest point. In the article, we also talk about the cycles that its hardware segment usually experiences, which we will discuss later. Additionally, we cover its ecosystem, which is the second key to the thesis. In the words of my past article: In my opinion, the key strength of this company lies in its unified platform, FortiOS, which integrates various products and services rather than creating standalone solutions that are difficult to integrate. This not only generates a network effect but also presents high switching costs for customers, potentially leading to pricing power. The ecosystem created by the company makes it indispensable for the customer." The overall figures for the quarter were quite good, especially standing out in the margins. The increase in margins has been due to several factors. They have managed to expand margins at the bottom of the cycle. It can be argued that as software weighs more in the sales mix, margins grow, but you also have some operational deleveraging due to the drop in product sales. Therefore, I believe margins will continue to expand in future cycles. They also had lower OPEX expenses (especially in marketing expenses), increasing margins, but they will raise these again once demand picks up more, which is another way of reinvesting in the business. Management once again confirms one of the key points of the thesis: the cross-selling and upselling of products, creating a much stickier ecosystem for its customers. But on the other side, we also see a lot of new opportunity, whether in the OT area in the Unified SASE and also upsell, cross-sell which are all helping driving, I would say, probably like a 90% customer initially moved by a FortiGate getting the firewall and network security market first, which we have a huge advantage over competitors. But after that one, they're keeping expanding beyond the network security, go to the other area. So that's what's happening for the Unified SASE for the Secure Op. And now the product, especially on the FortiGate firewall side we're starting to see kind of go back to normal or starting growing with the market now." - Ken Xie, Fortinet's CEO and founder. My view of Fortinet's product segment is like a Mr. Potato Head with software pieces added on top, an example of upselling. It also reminds me quite a bit of Nintendo with its Switch console; the years with the highest hardware sales are accompanied by many software sales. In a way, this is also confirmed by the CEO. So once the product starts growing, because product has a lower gross margin, that probably will impact the margin, but the product is also the leading indicator of future service. So that's where we kind of also were happy to see the product also starting growing now, which I think going forward with the product has a higher percentage, that probably also will impact the margin." - Ken Xie, Fortinet's CEO and founder. While all of this is positive, I don't think these are the main reasons why the market drove Fortinet's shares up 25% in one session. One of the main reasons is that they expect billings to return to a positive growth trend in the next quarter. Billings are a leading indicator of sales, and it is starting to be discounted that we have reached the bottom of the cycle in hardware. 90% of these billings come from existing customers, with over 80% being large enterprises, which are generally more resilient to economic cycles and tend to cut back on essential services like cybersecurity. And speaking of the bottom of the cycle: Sales for the quarter grew by 11%, but were driven by services by 20%, hence the record margins. Product sales decreased by 4%. The gross margin rises to 81%, with a product gross margin of 66%, mainly due to an increased software mix and lower indirect costs. They are also benefiting from generative AI, especially in their threat analysis and detection components. It seems that the refresh cycle for the hardware segment is being pushed back to 2025 instead of 2024, yet the market has largely ignored this, likely due to the increase in billings. The company also provided its guidance for the full year: And again, for the full year, inclusive of the numbers we gave a moment ago, we expect billings in the range of $6.400 billion to $6.600 billion; revenue in the range of $5.800 billion to $5.900 billion, which at the midpoint represents growth of 10%. Service revenue remained of $3.975 billion to $4.025 billion, which at the midpoint represents growth of 18%. Non-GAAP gross margin of 79% to 80%, Non-GAAP operating margin of 30% to 31.5%. Non-GAAP earnings per share of $2.13 to $2.19, which assumes a share count of between 767 million and 777 million. Capital expenditures of $320 million to $360 million. Non-GAAP tax rate of 17% and cash taxes of between $525 million and $575 million." - Keith Jensen, Fortinet's CFO. They have also made two acquisitions during the quarter. They seemed quite optimistic about them and have clearly found value in these acquisitions, as they have not repurchased any shares despite recent price declines. This isn't something that worries me too much, as they still have $1 billion authorized for share buybacks, and Fortinet is very opportunistic in executing these, as shown in the chart. The acquisitions will impact margins in the short term. Having a strong management team focused on the long term allows you to invest in upcoming macrotrends regardless of market opinions. Additionally, this creates opportunities for long-term investors who have conducted prior analysis and know what they have in their portfolio. In this case, the following is true: That's also you can see the Gartner research we pointed out, the convergence of network in the network security also starting or accelerating. So originally, I think last year, they say by 2030, the secure networking will be larger than traditional networking. Now they say, 2026, 4 years ahead, the secure networking will be larger than the traditional networking. So that's where we really invest long term on this trend. And with all this FortiOS, FortiASIC and making the best both appliance and infrastructure, the ASIC technology and at the same time, also try to invest more in the sales and marketing area to really catch the trend and also keep gaining market share. So that's the strategy to be ahead." - Ken Xie, Fortinet's CEO and founder. Fortinet continues to invest and gain market share in areas where they are already the best. Secure networking customers are increasingly recognized our FortiOS and FortiASIC technology offering 5 to 10x better performance than our competitors while improving security effectiveness and providing a low total cost of ownership." - Ken Xie, Fortinet's CEO and founder. Their internally manufactured ASICs, specifically designed for their products, provide cost savings and increased power, which are passed on to their customers in a form of shared economies of scale. In fact, they haven't raised prices since early 2023 and have been securing better deals for their customers thanks to these cost savings. Additionally, the increase in sales of their firewalls in recent years serves as a Trojan horse for subsequent upselling and ecosystem creation. In this section, I will be brief because I will continue using the annual data from 2023. I could attempt to create a Free Cash Flow estimate for 2024 and update it to a factor (4/12, the remaining months of the year), but it wouldn't add much value, and I prefer to be more conservative using current data. In any case, if we use a reverse DCF with a Terminal Growth Rate of 3% and a discount rate of 10% (the minimum I require for my investments), at current prices, it would suggest that the market is discounting a FCF growth of 13.5% for the next 10 years. I believe this growth is quite achievable, perhaps not for this year, but once the cycle recovers, margins expand, and they continue with good capital allocation, I think this growth will be in the middle-lower range of the range I consider. Nevertheless, after such a steep increase, I doubt it will rise much more in the short term, and there might be an opportunity to buy on a pullback, gaining some margin of safety. For all these reasons, I rate the stock as a hold. These companies are not without risks, just ask CrowdStrike, but I believe this is a common risk across all companies in the sector and depends on internal execution. The management team and corporate culture at Fortinet reassure me in this regard. I believe the greatest short-term risk to the company's stock price (as I don't think it poses a high long-term risk) would be a deterioration in the hardware cycle or further delays. They have stated that they expect to start recovering in 2025; if this doesn't happen, the stock will suffer. Nevertheless, this is not a long-term risk and could actually create investment opportunities if the stock corrects as a result. The other risks are the same as in the previous article: distributor concentration, production in Taiwan, the risk of hardware obsolescence in inventories, and competition, as the cybersecurity sector is particularly challenging. These are further developed in the article. Fortinet's performance over the last 10 years has been completely exceptional. A total return of +1326% for its shareholders, which equates to a 30% CAGR. Drawdowns of 35% have been somewhat common for the company, but it rarely exceeds that percentage of declines. The Price/FCF multiple shown by Koyfin doesn't seem the most appropriate for Fortinet, as two adjustments need to be made. With the annual data from 2023, Fortinet had $1.4 billion in net cash and is also in an expansive CAPEX cycle, so this figure needs to be normalized. The adjusted multiple would be 28X, which, while high, I don't think is overvalued. Overall, I believe Fortinet's quarter has been very good. They continue to invest long-term in effective niches. Management is still executing well, and capital allocation remains strong. The bottom of the cycle is just around the corner, and the subsequent hardware refresh will drive an increase in software sales and fortify the ecosystem in many companies, ultimately leading to higher recurring margins and returns for the company. I'm pleased to be a shareholder in this great company.