Baron Global Advantage Fund | Q1 2024

Baron Global Advantage Fund^® (the Fund) gained 3.6% (Institutional Shares) during the first quarter, compared to the 8.2% gain for the MSCI ACWI Index (the Index), and the 9.5% gain for the MSCI ACWI Growth Index, the Fund’s benchmarks.

Global equity indexes continued to move higher as the calendar turned to 2024. Similar to last year, the gains were not uniform or particularly broad based. While the U.S. (+10.3%) and developed markets in general (+8.9%) did well, the performance of their emerging markets counterparts was more muted (+2.3%) with Brazil down 7.4% and China down 2.2%; and more relevant to us, Argentina which represented, on average, 11.6% of the Fund’s net assets and is neither here nor there for the purposes of MSCI’s classifications, was down 3.8%. Giant-cap stocks continued to outperform, gaining 10.2% during the quarter. Giant-cap and large-cap stocks represented a whopping 83.0% of the Index, while stocks domiciled in the U.S. were 63.4% on average during the quarter. The Fund was 15.9% underweight in giant/large and 16.7% underweight the U.S., which did not bode well for relative returns.

Of course, the Index weights are market-cap based and the better the large U.S. multi-nationals perform, the larger they become in the Index. The U.S. accounts for approximately 25% of Global GDP, and though we would expect it to be over indexed, 63.4% does not make sense to us. For a mutual fund to call itself Global it must invest at least 40% of its net assets in companies outside of the U.S. and so, we are definitionally underweight this safe haven geography. Similarly, 83.0% giant/large cap works really well in a heightened anxiety, wide range of outcomes environment, but makes little sense to us for a Big Idea all-cap Fund.

It was a good thing that we owned a lot of NVIDIA with our thesis playing out much as we had laid out in this letter over the last six quarters. The bad thing is that we had to own other stocks too and while that in itself was not so bad, our investments in electric vehicles (EVs), IT services, and software could genuinely be described as ugly during this quarter. All in all, as much as we did not feel good about last year’s 25.6% gain, the first quarter’s +3.6% return felt like a win to us.

From a company-specific perspective, we had 19 contributors against 15 detractors, which was just too many against the Index that rose 8.2%. Many of our companies experienced extreme stock price volatility during the quarter, both up and down. In addition to NVIDIA, which gained 83%, Codere Online (+140%), Resident Home (+113%), Astera Labs (+106%), Zomato (+48%), Adyen (+31%), ASML (+29%), and CrowdStrike (+26%) all posted outsized gains. With the exception of Astera Labs, these investments contributed over 50bps each to absolute returns. Unfortunately, Endava (down 51%), Tesla (down 29%), and Rivian (down 53%) all cost us over 100bps each, while Snowflake and a final write down of Think & Learn cost us over 50bps each, which offset a lot of the above gains.

This quarter showcased some of The Good, the Bad, and the Ugly of what many of our investments go through over their full life cycles. In a way, it could be perceived as a microcosm of applying our investment philosophy and process.

The Good – NVIDIA, AI, disruptive change, and Big Ideas

At its core, our investment philosophy is centered on identifying and investing in Big Ideas, which we define as businesses that enable or benefit from disruptive change, have many characteristics of being or becoming platforms with network effects and attractive unit economics, and are likely to become materially larger in the future than they are today.

We identified NVIDIA as the company at the epicenter of one of the biggest technological paradigm shifts of the last 50 years as computing is shifting from sequential to accelerated and as we begin to see the early stages of the use cases of generative AI (GenAI) enter the mainstream. Is GenAI real? Is it going to be material, sustainable, and disruptive? Will NVIDIA (and other GenAI leaders and disruptors) benefit from this disruptive change? Our research suggests that the answer to all of these questions is an unequivocal – yes.

Is there hype around GenAI? Sure. There is always a hype cycle around major new technologies. Is GenAI a bubble similar to what we saw during the Internet bubble of late 1990s/ early 2000s? We don’t think so. First of all, it is important to recognize that while there were many stocks trading at silly valuations on newly invented metrics (peak multiples on peak eyeballs), the internet itself proved to be a paradigm changing disruption, giving birth to a plethora of Big Ideas. But even more importantly, while the rise in NVIDIA’s stock price has been nothing short of unprecedented for a company of its size, it was fueled almost entirely by rapid growth in revenues, earnings, and cash flows – not multiples. NVIDIA’s stock price exited 2023 with a P/E ratio of 24.7 and ended the first quarter with a P/E ratio of 35. We can debate whether it is cheap or expensive, but it cannot be compared to the triple-digit multiples that were assigned to the perceived market leaders of the early Internet era that were clearly not in as strong of a competitive position then, as NVIDIA is today.

It is not lost on us that semiconductors is a notoriously cyclical industry. Historically, the hyperscalers (AWS, Azure, GCP, etc.), who are among NVIDIA’s largest customers, have not invested/spent/consumed CapEx in a straight line. It will be more than a mild surprise then if there was no pullback in demand leading to a significant growth deceleration and a potentially meaningful correction in the price of the stock, sometime in the near future. So, it is incumbent upon us to manage the size of this investment appropriately, while continuing to imagine what the future will likely look like without losing sight of what reality on the ground is today.

Then again…NVIDIA is not just a semiconductor company. Many investors have missed the boat thinking that Apple is just a smartphone company, Amazon is just a retailer, and Tesla is just a car company. We have long argued that just like the other three, NVIDIA is a platform. We are more certain of this now than ever before.

In March, we spent the better part of a week in San José, attending NVIDIA’s annual developer conference GTC 2024 and got to experience firsthand what Forbes magazine called the Nerd Woodstock. After several years of being held online due to COVID-19, over 17,000 nerds (us included) attended the four-day event in person. With over 900 featured sessions, 1,700 presenters, over 300 exhibits, and more than 20 technical workshops, all centered around AI, there was a lot to choose from. From our perspective, there was nothing better than watching start-up founders debate the merits of Large Language Models (LLMs) compared to domain-specific Small Language Models, and how to get LLMs to have long-term memory, or what are the key challenges that need to be solved to enable reasoning, planning, and multi-agent LLMs (AI models that rely on and work together with other models). When it was all said and done, we came away with several observations:

• AI is developing rapidly across industries – near term, there is a lot of excitement around AI for areas such as consumer chatbots, AI-based customer service, AI-based assistants for a variety of business tasks such as coding, marketing, back-office, and more. Longer-term avenues of development are broad and include drug discovery, in which the opportunity for AI is significant due to the long timelines for drugs to reach approval and the high probability of failure (90% of drugs fail); planning and running factories and supply chains using digital twins (with help from NVIDIA Omniverse – NVIDIA’s real-time collaborative simulation); and using AI to build robots across a variety of use cases (from autonomous machines to humanoids). Multi-domain, multi-industry disruption.
• We are early on the S-curve – most companies are still in the proof of concept stage while very few are ready for production today. Hurdles in implementing AI include data prep, model adaptation and fine-tuning, and embedding of AI into existing workflows. There is a lot of innovation taking place to reduce these hurdles – from tools and infrastructure that help companies build and run AI models more easily, to third-party AI models exposed via Application Programming Interfaces (APIs) that enable companies to use them without building their own models from scratch. NVIDIA’s ecosystem across developers, system integrators, cloud providers, and independent software vendors, and internal software innovation are lowering these hurdles as well. For example, one of the most interesting announcements at the GTC Conference were NVIDIA Inference Microservices – or NIMs, which are APIs to easily access open-source models (NVIDIA already has dozens of models available) without the need to worry about model optimizations, security, patching, or sending data to third parties. NIMs could ease AI adoption for enterprises while also driving incremental monetization for NVIDIA, priced at $4,500/GPU or at $1/GPU hour if used on the cloud, and increase the stickiness of NVIDIA’s platform.
• We are rapidly coming down the demand elasticity curve – while in the Moore’s Law era, performance improvements were driven by cramming more transistors into a piece of silicon, AI is a March 31, 2024 Baron Global Advantage Fund data center scale problem with performance improvements driven by every layer in the stack:
  • GPUs – The GPUs themselves are getting faster and better – NVIDIA’s latest Blackwell GPUs are composed of 2 dies with 208 billion transistors, compared to 80 billion for the prior generation (Hopper), showing a performance boost of 2.5 times to 5 times depending on the use case.
  • The system layer – NVIDIA is innovating across the system from the accelerator (the GPU) to the CPU (Grace), server design and networking (between GPUs, between GPUs and CPUs, within a rack and between racks). For example, NVIDIA announced the latest fifth generation NVLink, a networking solution that connects multiple GPUs together, enabling order of magnitude higher memory bandwidth (1.8TB/sec) as compared to the standard PCIe (NVLink has 14 times more bandwidth than the fifth generation PCIe), which alleviates a significant bottleneck for many AI models. The latest fifth generation NVSwitch also enables connecting up to 576 GPUs together, which creates significantly higher overall bandwidth for a much larger unit of compute, which is especially important for very large LLMs (previously only 8 GPUs were connected with NVLink, while connecting a higher number required using the slower bandwidth, PCIe). Overall, the Blackwell GPU can see performance improvements of up to 30 times for inferencing compared to the prior Hopper generation¹:

• The software layer – NVIDIA’s rapid software innovation improves performance on the same hardware while lowering the hurdles for adoption. From NIMs described above to NeMo microservices (which make it easier for organizations to fine-tune existing models for their own data, needs, and requirements) to their innovation with optimizing existing models to run better on NVIDIA’s existing hardware. For example, its recent TensorRT LLM software, optimizes LLMs for inference, showing a 2.9 times improvement in performance thanks to incorporating innovations such as in-flight batching.²
• AI algorithms – The AI algorithms themselves are rapidly improving – some recent innovations include models from the Mixture-of-Experts (MoE) family. MoE models split the total number of parameters they have between a number of “experts.” These experts are each trained to be particularly good at inferring a particular part of the data (essentially becoming experts on some of the data). A router layer then routes each token to the appropriate expert. Each of the experts, which could be MoEs by themselves, and the router layer are also trained. This specialization enables reaching high-quality results with fewer parameters activated at each run. The benefit of MoE models is that they are faster (in both training and inferencing) because not all parameters are activated on all data points. For example, Databricks’ new DBRX model³, a MoE-type model, beat existing models on various quality benchmarks including a score of 70 in a programming benchmark (even better than the much larger GPT-4’s 67%). Databricks compared its current model with the MPT-7B model it released in May of 2023, and the new model required 3.7 times less compute (as measured by FLOPs) to reach the same quality – thanks to various Baron Global Advantage Fund algorithmic improvements including in the MoE architecture, various optimizations and better pre-training data. Additionally, “DBRX inference throughput is 2-3x higher than a 132B non-MoE model”. Another example for the pace of innovation is Claude 3 Opus (from Anthropic), which reached a 84.9% score on the same programming benchmark, 86.8% on multiple choice questions (vs. 86.4% GPT-4), a 95.4% on Reasoning (vs. 95.3%), 95% on grade school math (vs. 92%) and 60.1% in math problems benchmark (vs. 52.9%).⁴
• Another way to look at the pace of innovation is to compare the price OpenAI charges for its latest model, GPT-4 Turbo, compared to its previous model, GPT-3.5 Turbo. For the earlier model, the price is $10 for 1 million input tokens and $30 for 1 million output tokens. For GPT-4 Turbo, the price is 95% lower at $0.5 for 1 million input tokens and $1.50 for 1 million output tokens.

Roughly 18 months after the ChatGPT moment, GenAI is already showing rapid real-world adoption with revenues of GenAI companies exceeding $3 billion, excluding the revenues that the large cloud providers (like Google and Meta) are generating from AI due to better engagement and better ad targeting. To put this in perspective, it took the cloud Software-as-a-Service (SaaS) companies 10 years to reach $3 billion in revenue.⁵

We believe we are in the early stages of a multi-decade disruption. Jensen Huang, NVIDIA’s co-founder, president, and CEO suggested at the conference that similar to how in the industrial revolution, raw materials came into the plant and final products came out, in the GenAI era, companies would become AI factories with data as a raw material and tokens as the output. Tokens can represent words, images, videos, or controls of a robot. Over time, as models continue to improve, and the cost of running them declines, an increasing number of human tasks could be augmented or replaced entirely by AI. We expect decision making to become much more data-driven, enabled by AI, as consumers, corporations, and governments alike, take advantage of the vast amounts of unstructured data we generate, which is estimated to represent 90% of all data generated.⁶ With increasingly challenging demographics across many economies (especially in developed markets), a greater proportion of global growth must come from productivity enhancements. AI, in our view, is likely to be a key driver behind these productivity gains, and potentially, the basis for technological breakthroughs that help humanity solve a host of the most difficult challenges from climate change to finding cures for diseases that have remained unsolved. We believe this disruptive change will be truly profound.

The Bad and (for brevity’s sake) the Ugly

A long-term ownership mindset is foundational to our philosophy and process. As owners of businesses, we assign less weight and importance to short-term trends and macro/geo-political events and evaluate them almost exclusively in the context of how they impact our long-term investment thesis. We rarely trade around positions let alone initiate or exit an investment based on these shorter-term trends. This approach works exceedingly well during some parts of the market cycle, but it works poorly during others. To be clear, we do not ignore those trends or a challenging macro environment, but we are honest in acknowledging that we have absolutely no edge in forecasting near-term trends better than the market and we can’t forecast investors’ expectations of companies’ reported results better than the market.

Whenever our businesses undergo cyclical headwinds, we tend to do nothing as long as we believe long-term prospects remain intact. This structurally results in increased volatility for our portfolio, especially during times of heightened uncertainty and stress. This quarter was a good example of our investments in IT services, EVs, and software getting hit particularly hard on what we believe are largely cyclical and short-term issues. 

We held positions in digital transformation consultants Endava, Globant, and EPAM for many years. Until 2022, these were some of the most successful investments in the 10-year history of the Fund. The fundamentals of their businesses began to weaken in early 2022, which coincided with the Fed’s tightening cycle (and specifically for EPAM, Russia’s invasion of Ukraine). While we eventually exited EPAM, we held on to Endava and Globant through ’22 and ’23 expecting their businesses to return to historical growth rates once the cycle turned. Though both companies’ fundamentals stabilized and started improving, their stock prices were hit hard when growth failed to materialize in the first quarter. While we may have underestimated the magnitude of the cyclical slowdown and these business’ sensitivity to it, we believe their intrinsic values have not changed nearly as dramatically as the contraction in their multiples would suggest. It was just as bad for software holdings Snowflake and Zscaler, who reported far more impressive growth numbers but offered muted guidance due to reduced visibility and continued macroeconomic softness (though Snowflake did have an unexpected change in senior management). And then it was downright ugly for our EV manufacturers Tesla and Rivian where classic product cycle transitions created a bit of a vacuum in demand, which caused the short-term focused market participants to question the long-term attractiveness and viability of the overall addressable market for EVs.

While we kind of saw most of these things coming (and certainly saw all of them as they were unfolding) we spend most of our time focusing on researching the business’ long-term prospects, the sustainability of their competitive advantages, and the durability of their growth drivers. Our investment theses are built around what companies will likely look like 5 or 10 years down the road and while cyclical or economic downturns often push out the timing, they rarely imperil the theses themselves and hence, more often than not, we will choose to ride it out.

NVIDIA Corporation sells semiconductors, systems, and software for accelerated computing, gaming, and GenAI. NVIDIA’s stock rose 83.2% in the first quarter, driven by continued strong demand for its GPUs that stand at the epicenter of the GenAI revolution. NVIDIA closed 2023 with unprecedented revenue growth at massive scale, with a fourth quarter revenue run-rate just shy of $90 billion, growing over 3.5 times year-over-year with operating margins of 67%. NVIDIA’s fourth quarter was even more impressive than numbers suggest considering both the fact that sales to China declined significantly due to U.S. regulations, and as we are nearing the end of the current generation of the Hopper architecture, with the next generation, Blackwell coming out in the third quarter. NVIDIA continues to improve the performance of its chips and systems significantly from generation to generation, thanks to its full-stack approach, optimizing every layer from the chips, through systems, networking, and software. NVIDIA is taking a page from Apple’s vertical integration book, adding to its competitive advantage. For example, the recently introduced GB200 NVL72, a liquid-cooled rack with 72 GPUs, interconnected with the latest generation NVLink technology, offers a significant boost to performance with one-fourth the GPUs required to train a 1.8 trillion parameter model, using one-fourth the power. NVIDIA is also removing hurdles for AI adoption through software innovation, such as the recently announced NIMs, which make it easier for companies to adopt GenAI at scale. Jensen Huang, NVIDIA’s CEO spoke about NVIDIA’s full stack innovation in the company’s last Investor Day:

“Blackwell is both a chip at the heart of the system but it’s really a platform. It’s basically a computer system. What NVIDIA does for a living is not build the chip, we build an entire supercomputer, from the chip to the system to the interconnects, the NVLinks, the networking, but very importantly, the software. Could you imagine the mountain of electronics that are brought into your house, how are you going to program it? Without all of the libraries that we’ve created over the years in order to make it effective, you’ve got a couple of billion dollars worth of assets you just brought into your company. And any time it’s not utilized, it’s costing you money…

“And so our ability to help companies, not just buy the chips, but to bring up the systems and put it to use and then working with them all the time to make it – put it to better and better and better use, that is really important, okay? That’s what NVIDIA does for a living. The platform we call Blackwell has all of these components associated with that. This is the part that’s incredibly hard about what we do.”

CrowdStrike Holdings, Inc. is a cloud-architected SaaS cybersecurity vendor offering end – point security, threat intelligence, and cyberattack response services. Shares increased 25.6% in the first quarter, following one of the strongest reported quarterly results in the company’s history. With market share gains in end-point detection and response (EDR) accelerating, emerging products (Cloud, Identity, and SIEM) scaling to about $850 million in annual recurring revenue (ARR), and new partnership channels like Dell and Pax8 already making meaningful contributions, the outlook suggests sustained revenue growth of 30% or more over the next two years. Fiscal year 2025 guidance looks conservative, as it projects 8% to 12% net new ARR growth, a modest increase from the 6% growth it reported in fiscal year 2024. This guidance would suggest net new contributions from emerging products would significantly decelerate, landing in the range of 30% to 35% (on a 25% larger base), or that core EDR contributions would contract by roughly 15%, which are unlikely scenarios, in our view. With its leading competitive positioning in cybersecurity, the growing threat landscape, which is also driven by the advancements in AI and is making hackers more dangerous, its unique single-agent architecture, and its platform approach, we retain conviction in CrowdStrike, which is emerging as the security platform to beat in terms of scale, profitability, and free-cash-flow conversion.

Codere Online Luxembourg, S.A. is an operator of online gaming and sports betting assets in Latin America and Spain. Shares increased 140.4% during the quarter as the company reported net gaming revenue and EBITDA growth that beat consensus estimates, with net gaming revenue growth of 33% year-over-year, while guiding for positive adjusted EBITDA for 2024. Profitability is inflecting as Codere is exiting a period of accelerated investments to build its brand and gain share in Latin America, a key growth market. This is also helped by the growing mix of existing cohorts of users (whose overall spending continues to grow), as marketing spending is focused almost exclusively on new users (note that net gaming revenues are already accounted for net of promotional spending on existing users), driving a structural profitability improvement with the growing mix of existing cohorts. This should accelerate earnings growth and drive a re-rating in the stock, which currently trades at a steep discount relative to its global peers – Codere is trading at 1.3 times EV/Revenues (on consensus 2024 estimates, though we believe the company will grow revenues much faster, so the real multiple is likely closer to 1.1-1.2 times), whereas competitors such as DraftKings or Flutter trade at 3 to 4.5 times. We remain shareholders and believe the company remains early in addressing the €4 billion TAM in its core markets in Mexico, Colombia, Panama, and Argentina.

Shares of IT services provider Endava plc fell 51.1% after management cut guidance for the fiscal year ending June 30, 2024, by 7% to 8%. Growth has slowed over the last year as business customers pulled back on discretionary IT spending due to macroeconomic uncertainty. Last fall, management was seeing early signs of a recovery, but new projects have been taking longer to materialize as customers delay spending decisions. Higher expenses due to increased staffing to meet anticipated demand weighed on margins as well. Management acknowledged that it misread the market and is taking steps to right-size the cost structure to improve margins. We remain invested because we expect these near-term headwinds to abate over time, leading to better growth as the demand for digitization remains strong, with AI likely to serve as a tailwind to digitization over the long term.

Tesla, Inc. designs, manufactures, and sells EVs, related software and components, and solar and energy storage products. Shares fell 29.2% in the first quarter as the core automotive segment remained under pressure due to a complex macroeconomic environment, factory shutdowns, growing competitive risks in China, and Tesla’s price reductions throughout 2023. During the first quarter of 2024, production was negatively impacted by the Red Sea maritime supply-chain interferences, sabotage in Tesla’s factory’s power supply in Berlin, and factory closure for the launch of the refreshed Model 3. We remain shareholders. Tesla commenced delivery of its highly anticipated Cybertruck pickup, which features new technologies within the car and its manufacturing lines. Tesla also launched version 12 of its Full Self Driving product, which shows significant progress from prior versions and increases the probability that Tesla’s unique data collection, and verticalized software and hardware approach will position Tesla as a leader in the future of autonomous driving and shared mobility. We also expect energy storage sales to continue to grow over the coming years as the adoption of renewable energy continues. Lastly, we believe Tesla’s core automotive segment will recover with the company remaining a leader in the EV market, which continues to expand with EVs still accounting for only around 10% of vehicle sales globally.

Shares of Rivian Automotive, Inc., a U.S.-based EV manufacturer, declined 53.3% in the first quarter. Despite substantial improvements in production and delivery volumes in 2023 as well as improved unit economics, Rivian’s business remains constrained by its limited scale, negative gross margins, and elevated cash outflows. Additionally, Rivian expects to temporarily shut down its production facilities for upgrades, impeding anticipated production growth in 2024. Compounding these challenges is the potential for demand constraints, which may not keep pace with production due to the continued complex macro environment, and the relatively small automotive segments that Rivian’s initial products target. Nevertheless, the recent unveiling of Rivian’s mass-market products, the R2 and R3, garnered enthusiastic responses, evidenced by over 68,000 pre-orders within the first 20 hours post-launch. In a strategic move, management opted to produce the R2 in Rivian’s existing facility, deferring the construction of a new factory. This decision should help reduce mid-term capital expenditure obligations while ensuring higher utilization of current facilities as the R2 ramps production in 2025. We remain shareholders.

The portfolio is constructed on a bottom-up basis with the quality of ideas and conviction level having the most significant roles in determining the size of each individual investment. Sector and country weights are an outcome of the stock selection process and are not meant to indicate a positive or a negative “view.”

As of March 31, 2024, the top 10 positions represented 59.4% of the Fund’s net assets, and the top 20 represented 87.6%. We ended the first quarter with 35 investments compared to 34 at the end of 2023. Note that our top 25 investments represented over 95% of the Fund.

Our investments in the Information Technology, Consumer Discretionary, Industrials, Financials, and Health Care sectors, as classified by GICS, represented 99.6% of the Fund’s net assets. Our investments in non-U.S. companies represented 52.8% of net assets, and our investments in emerging markets and other non-developed countries (Argentina) totaled 26.1% of net assets.

During the first quarter, we initiated three new investments: a biotechnology company developing a drug for obesity, Viking Therapeutics; a networking-focused fabless semiconductor provider, Astera Labs; and an automotive-focused fabless semiconductor provider indie Semiconductor.

We reduced 21 existing positions and sold our investment in the life sciences simulation software provider, Schrodinger. We also had a positive realization event with one of our private investments – the direct-to-consumer retail company, Resident Home, which was acquired by Ashley Home. Following this acquisition, we now hold four private investments, representing just under 5.5% of net assets, with SpaceX accounting for 5.0% of that total.

During the first quarter, we initiated a new position in Viking Therapeutics, Inc., a developer of metabolic disease medicines with a focus on diabetes, obesity, and metabolic steatohepatitis (MASH) (i.e., fatty liver). Viking’s lead asset is an injectable and oral version of a GLP-1/GIP combo medication that is a direct competitor to the well-known Mounjaro/Zepbound medicines from Eli Lilly. Viking’s second asset is a competitor to Madrigal’s just approved MASH asset. Both of Viking’s drugs appear to be more efficacious than their competitors, which is the primary basis of the investment thesis bolstered by what are potentially the largest revenue end-markets for the industry ever. The pharmaceutical industry has returned to primary care medicines, meaning large volumes as opposed to high price, after about a 20-year period working in the rare disease space, including oncology. This macro trend is led by the obesity space. For a reference point, 50 million people (or one-third of the current U.S. market) treated at only $200/year, which is significantly lower than current pricing, would result in a $100 billion addressable opportunity just in the U.S.

Another new addition was indie Semiconductor, Inc., a fabless designer, developer, and marketer of automotive semiconductors for applications including advanced driver assistance systems (ADAS), car connectivity, user experience, and electrification. The automotive semiconductor vertical is attractive for long-term investors as semiconductor content is increasing over time from several hundreds of dollars of content in legacy internal combustion engine vehicles to as much as several thousands of dollars in high-end, feature-rich EVs driven by improving safety features (including autonomous driving), electrification and digitization, and premiumization of interior and exterior features for all types of vehicles driven by both consumer demand and regulation. Indie is a small but rapidly growing share gainer leveraging mixed signal expertise, and 100% automotive focus to design more highly integrated, power efficient, and lower cost solutions than larger competitors that are focused on a variety of different end-markets. Specific to ADAS, despite indie’s size, it is the only company to offer all key ADAS modalities (vision, radar, ultrasonic, LiDAR) under a single roof. Indie’s size and focus also enable it to be more agile in supporting OEM’s needs whereas competition, which has significantly consolidated in the past 10 years, is encumbered by inertia of large organizations and has been slower to innovate. These advantages have led to indie securing more than $6.3 billion in lifetime design awards, of which $4.6 billion is in ADAS applications, compared to approximately $220 million in 2023 revenue. 2023 was the third year in a row that indie doubled its revenues.

Driven by a growing backlog of design wins and a strong opportunity funnel, we believe indie can grow to over $500 million in revenues in the medium term and more than $1 billion in revenue towards the end of the decade as its large program wins in radar and driver/occupant monitoring systems start ramping. On top of this, the company continues to expand its gross margins benefiting from the growing mix of new products, which have higher-than-average gross margins, with the company targeting a 60%-plus gross margin (from the lows 50s% today) and 30%-plus operating margins (with breakeven expected by the end of 2024). Despite the company’s long runway for growth, making it a potentially Big Idea, valuation is still significantly below both other high-growth semiconductor companies and slower growing mature peers, as the stock price was negatively impacted by its de-SPAC history, the stock still being small and obscure, and the cycle challenges impacting the broader automotive industry, which also creates uncertainty with indie’s near-term numbers. We see significant upside for indie’s stock in both the medium and the long term and have therefore decided to initiate a position.

We also initiated a small position in Astera Labs, Inc., a fabless semiconductor company focused on addressing connectivity challenges in data-centric systems. It offers multiple product lines of mixed-signal connectivity products with embedded COSMOS software and have been first to market with their PCIe retimers, Ethernet smart cable modules, and CXL memory controller products, securing over 300 design wins in the short time since its founding in October 2017. These products are critical in solving data and networking bandwidth bottlenecks, signal integrity challenges, and memory capacity limitations that hyperscale customers are facing across their entire data center networks but especially in AI-related applications. Today its Aries PCIe retimers are shipping alongside every NVIDIA GPU with virtually 100% market share, driving significant growth, and the company sells to every hyperscale customer and AI platform provider. In the coming years, its Taurus Active Electrical Cables modules and Leo Compute Express Link controllers could potentially further expand its market, ramping into AI and general-purpose server applications, respectively.

Over time, as AI continues to advance, solving networking bottlenecks becomes more important, which should serve as a tailwind to Astera’s average content per server opportunity (the faster the PCIe standard becomes, the worse signal loss becomes, requiring either the use of more expensive PCB materials or the use of more retimers – which is more cost efficient). This dynamic together with the strong AI server unit growth should drive solid revenue expansion from $200 million to around $1 billion towards the end of the decade. We also expect the company to benefit from operating leverage with operating margins expanding towards 40%. Lastly, we believe that Astera benefits from solid competitive advantages thanks to its first mover advantage, having solved interoperability challenges between their retimers and dozens of different third-party devices that its chips can connect to.

After significant runs we took some profits in NVIDIA Corporation, MercadoLibre, Inc., and Bajaj Finance Limited. As mentioned earlier in the letter, Resident Home Inc. was acquired for cash and we exited our small four-year investment in Schrodinger, Inc. with a small profit.

The market is starting to come to grips with higher for longer. One cut? Two cuts? No cuts this year? We have no idea. The 10-Year U.S. Treasury yield is now back to yielding 4.5% after declining to 3.8% towards the end of last year. We continue to believe that the Fed’s tightening cycle is over, and that the next move in interest rates is going to start an easing cycle, whenever that will be.

We believe that the disruptive change ushered in by the commercialization and use of AI is real, material, and that it will very likely have a long-term deflationary effect (despite the high cost of GPUs). Major technological innovations have generally been a deflationary force – from hardware to software, to the internet. Moore’s Law for example, demonstrated that we get a doubling of performance for the same cost every two years or so, for decades. AI chips have been improving at a rate of 1,000 times every eight years, or a double every nine to ten months.⁷ Software enabled a significant productivity boost for information workers as the amount of manual work was substantially reduced and the quality of work improved. The internet proved to be a massive deflationary force as it reduced distribution costs effectively to 0 and gave us cloud computing, which reduced the cost of software materially, while significantly reducing the cost and the risk for startups that no longer had to buy hardware for millions of dollars enabling entrepreneurs to align costs with their progress. AI, in our view, will further accelerate these deflationary forces. If we are right about that, then longer-term interest rates are heading lower and then it won’t matter whether the Fed cuts in July or September, and once or thrice (although it is unlikely to matter even if we are wrong…).

As we do every quarter, we analyzed the change in the weighted average multiple of the Fund and the weighted average change in consensus expectations for 2024 (for revenues, operating income, and operating margins). The weighted average multiple for the Fund as a whole increased by 3.5% during the first quarter, though it declined by 1.6% excluding NVIDIA). Revenue expectations for 2024 increased by 2.1% and it declined by 0.2% excluding NVIDIA. Operating income expectations increased by 0.9% and declined by 2.2% excluding NVIDIA. Operating margin expectations declined by 36bps or 64bps if we exclude NVIDIA. Looking under the hood, the headwinds were mostly due to our IT consulting and EV holdings, where the stocks suffered from significant multiple contraction and a cyclical reduction in short-term expectations. The other sub-industries were much more balanced, and the trend we have seen in the second half of 2023 of stable fundamentals continues to hold true.

Every day we live and invest in an uncertain world. Well-known conditions and widely anticipated events, such as Federal Reserve rate changes, ongoing trade disputes, government shutdowns, and the unpredictable behavior of important politicians the world over, are shrugged off by the financial markets one day and seem to drive them up or down the next. We often find it difficult to know why market participants do what they do over the short term. The constant challenges we face are real and serious, with clearly uncertain outcomes. History would suggest that most will prove passing or manageable. The business of capital allocation (or investing) is the business of taking risk, managing the uncertainty, and taking advantage of the long-term opportunities that those risks and uncertainties create.

We are optimistic about the long-term prospects of the companies in which we are invested and continue to search for new ideas and investment opportunities while remaining patient and investing only when we believe the target companies are trading at attractive prices relative to their intrinsic values.

Sincerely,