Recently, I've spotted several article headlines warning Nvidia (NVDA -0.30%) investors to be cautious because its stock's recent big run-up shares key similarities with Cisco Systems (CSCO -0.22%) stock's huge run-up in the 1990s before it crashed in 2000 when the dot-com bubble burst. In just one year, Cisco stock plummeted 77% after hitting its all-time high on March 27, 2000.
Let's dig into the qualitative reasons why the comparison of Nvidia now to Cisco in the late 1990s isn't a good one. A second article on this topic will focus on the quantitative reasons, including differences in profit margin trends and stock valuations.
Nvidia 2024 = Cisco Systems late 1990s theory has more holes than an AI supercomputer has GPUs
The above subheading is a reference to Nvidia's business -- and yes, it's an exaggeration because artificial intelligence (AI) supercomputers can have thousands of graphics processing units (GPUs).
Nvidia is the dominant maker of GPUs for accelerating the processing of AI and high-performance computing (HPC) workloads in data centers. It has an estimated 90% share of the data center AI GPU chip market, and 80% share of the overall data center AI chip market. It also targets other markets, but its data center business is its largest, accounting for 83% of its revenue last quarter.
To be clear -- I'm not saying Nvidia stock is immune to a big decline or even a crash. No doubt, it would plunge if the company's GPUs lost their status as the favored chips for speeding up the training of AI models and the running of AI applications in data centers, and if it failed to develop whatever tech had displaced GPUs for these fast-growing uses. My premise is simply that the comparison of Nvidia now to Cisco in the late 1990s is not a good one. The similarities are only surface deep.
Key qualitative differences between Nvidia now and Cisco in the late 199Os
Cisco's CEO in the below chart refers to John Chambers, who led the company from 1995 to 2015. Cisco was founded in 1984 by two Stanford University computer scientists and went public in 1990.
Nvidia has had only one CEO since its founding in 1993, co-founder Jensen Huang. The company went public in 1999.
| Qualitative Metric | Cisco Systems Late 1990s | Nvidia Now |
|---|---|---|
| Run by a founder-CEO? | No | Yes |
| CEO educational background* | Business (including an MBA) and legal | Electrical engineering (including a master's degree) |
| CEO work background* |
- Technology sales at IBM - VP of Operations at Wang Laboratories (a now-defunct computer company) |
Microprocessor design for Advanced Micro Devices (AMD) |
| Company's primary growth driver | Internet adoption and growth | AI adoption and growth, particularly generative AI.*** (A notable secondary driver is the growth of computer gaming.) |
| Main products | Internet networking products (hardware and software) and services. Main hardware products in the 1990s included switches (LAN and WAN)**, access servers, and routers. | Primarily GPU chips and related hardware, software, and services. It also offers other chips, such as central processing units (CPUs) and a "superchip," which is basically a combined CPU-GPU. |
| Competitive moat | Moderate-to-high overall | High overall -- Its GPUs and associated hardware and software have a high degree of complexity. |
| Growth strategy | Considerable focus on acquisitions | Heavily organic growth. Acquisitions have mainly been small, except for 2020's $7 billion acquisition of Mellanox, which makes high-performance networking products. |
*According to public records. **LAN = local area network; WAN = wide area network. *** Generative AI is the tech behind OpenAI's extremely popular ChatGPT chatbot, released in late 2022.
Studies support the theory that stocks of public companies run by founder-CEOs tend to perform better over the long term than those run by non-founders. Huang is a co-founder of Nvidia; Chambers was not involved in founding Cisco.
Huang's educational and work background before starting Nvidia was in electrical engineering and semiconductor design, respectively. These factors give him a major advantage over CEOs of technology companies whose backgrounds are not as technical, in my view.
Cisco's moat to keep competition at bay in the late 1990s was not as high as Nvidia's current moat. Nvidia's mighty moat isn't just due to the complexity of its products, but also stems from the fact that its ecosystem is huge and growing, thanks to its early mover status in the data center AI chip space. This ecosystem includes "over 4.7 million developers worldwide using CUDA and our other software tools to help deploy our technology in our target markets," the company said in its annual report filed in February.
The moat differences will be more evident in my second article on this topic focusing on quantitative factors. Suffice it to say here that Cisco's profit margins were steadily declining for a few years before its stock plunged. One main reason was competitive pressures on pricing. Nvidia's profit margins are currently at record highs.
Let's move to the last category in the chart -- growth strategy. Cisco in the 1990s was a corporate equivalent of the iconic video game Pac-Man -- it devoured a lot of companies. From 1993 through July 2000, the company "acquired or announced our intent to acquire 65 companies," it said in its fiscal year 2000 annual report.
Nvidia, on the other hand, is mostly growing organically. Granted, it relatively recently made one huge acquisition, Mellanox, but it had already been growing at a robust pace before this deal. In other words, Nvidia didn't need to "buy growth." And prior to this acquisition, Mellanox was a Nvidia partner, not a competitor, so it didn't buy the company in order to get rid of a competitor.
Growth strategies that heavily focus on acquisitions are very challenging to pull off successfully. They take the time and attention of top execs away from focusing on internal innovation. They eat up money that could be spent on research and development. In addition, it's difficult to meld different corporate cultures.
Not surprisingly, Apple, one of the most successful and innovative companies of the last few decades, is known to prioritize organic growth.
