In an era defined by rapid advancements in artificial intelligence (AI) and data center technology, leading tech companies are finding it increasingly necessary to adapt their business models to these burgeoning sectors. Among the companies making significant strides in this direction is Nvidia Corporation (NVDA), a renowned leader in the technology industry. Recently, Nvidia announced its Q1 2024 earnings, highlighting its remarkable performance and strategic shift towards data centers and AI-driven growth.
Picture: Nvidia. Source: Unsplash
Q1 2024: A Quarter of Outstanding Performance
Nvidia's Q1 2024 financial results were particularly noteworthy, with the company exceeding expectations on multiple fronts. This strong performance led to an as much as 33% increase in extended trading, a testament to the market's positive reception of Nvidia's current trajectory. This is mainly thanks to the out of the world raise for Q2 revenue forecast to 11B plus/minus 2%, which the most bullish analyst projected only at 7.9B.
One of the standout figures in Nvidia's Q1 report was the earnings per share (EPS) of $1.09, adjusted, surpassing the anticipated 92 cents. Furthermore, the company reported impressive revenue of $7.19 billion, considerably higher than the expected $6.52 billion. In addition, they are projecting a higher gross margin for next quarter, making EPS for quarter 2 in the range of $1.67 to $2.0, which also beats the most bullish analyst at $1.39. This means a 40% growth in revenue translated to 30% growth in EPS, which is amazing for a giant firm.
Driving these impressive figures was the company's data center group, which reported sales of $4.28 billion, marking a 14% annual increase. This growth was primarily propelled by demand for Nvidia's GPU chips from cloud vendors and large consumer internet companies, which use these chips to train and deploy sophisticated AI applications like OpenAI's ChatGPT. Nvidia CEO Jensen also suggested in the earning call that the market current 1 trillion data centers need replacement or upgrade GPU due to AI training. For context, most data centers right now are still running with only CPUs.
A Strategic Pivot Towards Data Centers and AI
While Nvidia's strong Q1 performance is indeed notable, what's even more striking is the clear strategic shift the company is making towards data centers and AI technology. In the earnings call, Nvidia's CFO indicated that the company expects sequential growth to be primarily driven by data centers, a trend reflecting a sharp increase in demand related to generative AI and large language models.
The significance of this shift lies in the nature of AI training, which is fundamentally iterative and endless. AI models, particularly large language models, constantly require training and retraining to improve their performance and adapt to new data. On a broader view, AI integrated products, Google's for instance, further intensify the ongoing AI training. This process necessitates powerful GPUs and data centers, creating a virtually inexhaustible demand for Nvidia's products and services.
Picture: Jensen Huang. Source: Wikimedia Commons.
In fact, there was a real example given in the earning call. In that conference, Nvidia disclosed partnership with Microsoft's Azure to provide enterprise-ready generative AI cloud environment. The partnership is not an exclusive one, which opens the door for cooperation with others large cloud providers, which Amazon and Google accounts for a total of 42% of total market share. This success further cement Nvidia's CEO as one of the best visionary CEOs ever. It is twice now that he produced miracle vision into reality, with the first one being the creation of the world first GPU for personal computers in late 1990s. The reason for this shift to data center was his call for the Moore's Law being broken by GPU in 2018 was ridiculed not just by people in the industry but also mainstream media.
The Iterative Cycle of AI: A Catalyst for A Near Monopoly Business Transformation
As mentioned above, the endless cycle of AI training and retraining is central to understanding the transformation underway in Nvidia's business model. The company has been shifting from a heavy reliance on traditional hardware sales model, where revenue generation is typically a one-time event, to a continuous demand model rooted in the cyclical nature of AI training since 2018. As a result, the company is in a near monopoly environment for GPU for AI training right now, which the second runner up is AMD not in a position to catch up in at least the next 2 years according to this PhD. Using the Lambda Lab benchmark test on GPUs with PyTorch, which is a popular Machine Learning library by Facebook, there is no competition right now for Nvidia.
Picture: GPU benchmark for AI training. Source: Lambda
This transformation indicates a significant evolution in Nvidia's strategy, positioning the company as a frontrunner in the AI revolution and the expanding market it generates. Despite its clear focus on data centers and AI, Nvidia is not abandoning its other ventures. The company has a diversified business model that spans various sectors, including gaming and automotive. Although the gaming division reported a 38% drop in revenue, Nvidia's diversified portfolio cushions the impact of such industry fluctuations, allowing the company to weather sector-specific downturns while continuously advancing its strategic objectives. In fact, based on the latest earning report, the gaming sector seems to be on track of recover.
Investing in Nvidia: A Commitment to Transformation
Nvidia's commitment to its business transformation is further evidenced by its financial strategies. The company is planning to increase investments in the business while maintaining operational leverage, highlighting its commitment to this new strategic direction. Additionally, Nvidia has a $7B share repurchase authorization through December 2023, demonstrating its readiness to deploy various financial tools to facilitate its business growth. This earning also flips the expectation for Taiwan Semiconductor Manufacturing Company, which is the provider for Nvidia.
However, in order to invest in Nvidia at the price of $380, there is a need to deeply understand the whole picture. Firstly, to gain a better understanding of NVIDIA's value proposition, we can consider its Price/Earnings to Growth (PEG) ratio. Using NVIDIA's guidance of an expected EPS of average $1.8 for the next quarter, we can annualize this to $7.2 for the full year. However, it should be noted that this is not recommended for other stocks since it is still not confirmed if this sudden growth is a one-time thing or a consistent trend. The only reason I am doing this is due to the historical quarterly performance by Nvidia, which suggested an average annualized calculation is suitable. With a current share price of $400, the P/E ratio thus stands at 55.5 ($400/$7.2).
#Full disclosure: I am a shareholder of Nvidia shares at the time of this writing. Please do your own analyst before making any decisions.
Picture: Investing. Source: Unsplash
Base Case
For the base case, given that the AI revolution is just at the beginning, we can assume an average EPS growth rate of 25% per year for the next three years (an average of the 20%-30% range) thanks to data centers and the recovery of gaming sector after a 2 year decline. From here, we can calculate the PEG ratio. The PEG ratio is the P/E ratio divided by the annual EPS growth rate, giving us a PEG ratio of 2.22 (55.5/25).
In Nvidia's case, a PEG ratio of 2.22 suggests that the stock's price is at a slight premium given the EPS and expected growth rate. This indicates the gold rush toward AI is over-hyping the price of AI-related stocks.
Bullish Case
For the bullish case, with a more rosy view of the AI revolution being in initial phase, we can assume an average EPS growth rate of 35% per year for the next three years (an average of the 30%-40% range) thanks to data centers and the recovery of gaming sector after a 2 year decline. From here, we can calculate the PEG ratio. The PEG ratio is the P/E ratio divided by the annual EPS growth rate, giving us a PEG ratio of 1.58 (55.5/35).
In Nvidia's case, a PEG ratio of 1.58 suggests that the stock's price is at a fair price given the EPS and expected growth rate. This is only fair thanks to Nvidia near monopoly in GPU market for AI training.
Bear Case
For the bear case, it means the AI revolution is overblown. We can assume an average EPS growth rate of 18% per year for the next three years (an average of the 15%-21% range), which is on track with analyst expectation before earning. From here, we can calculate the PEG ratio. The PEG ratio is the P/E ratio divided by the annual EPS growth rate, giving us a PEG ratio of 3.08 (55.5/18).
In Nvidia's case, a PEG ratio of 3.08 suggests that the stock's price is extremely expensive right now and it is prone to a pull back.
The Road Ahead
Looking towards the future, Nvidia's transformation reflects its strategic response to the AI revolution and evolving market dynamics. However, it is difficult to quantify any percentage number for each case to realize exactly because we are still at the initial boom of AI. Hence, I would not be providing the number without further information.
As the demand for AI technology continues to surge, Nvidia's business model transformation presents a compelling case study of strategic adaptation in an ever-evolving industry. The company's CEO, Jensen Huang, projected confidence in Nvidia's strategic direction, hinting at another potential increase growth in the second half of FY24.
In conclusion, Nvidia's transformation signifies a significant evolution in its business model. By embracing the iterative demand model spurred by AI training and retraining, Nvidia positions itself as a key player in the AI revolution. This shift, accompanied by continued commitment to its diversified portfolio and strategic financial decisions, fortifies Nvidia's standing in the tech industry. Nvidia's strategic shift is not only reshaping its own business model but is also setting a benchmark for the tech industry.
Comentarios