The 3 Laws that Rule Silicon Valley

I've uncovered something about Silicon Valley tech companies that nobody is talking about. They all capitalize on the same three basic principles. Did you know that Apple generates enough profit every single year to buy the entire airline industry in the United States? United, Delta, Southwest and American Airlines are truly massive companies operating some of the world's most impressive machinery. But they haven't been able to successfully capture the value of what they create in the same way that tech companies have. Why is this? It's impossible to understand tech company valuations by just looking at one thing. But there are three laws that combine important properties, that make them unique, and which together explain why Silicon Valley companies have been able to achieve incredible levels of success and profitability. To understand the first law, we have to go back in time to Bell Labs and what became the most notorious and dramatic betrayal in the history of Silicon Valley. In the middle of the 20th century, Bell. Bell Labs was performing the world's most cutting edge scientific research. The list of technologies invented by researchers working at Bell Labs is radio astronomy, the laser information theory, the Unix operating system, programming languages like C. But the most important was the transistor. Jointly created in 1947 by William Shockley, John Bardeen and Walter Bratton, the transistor made modern computing possible. Shockley was a brilliant scientist whose status approached the divine. But by all reports, he wasn't easy to get along with. By 1953, he had fallen out with his colleagues at Bell Labs and left to return to Caltech. But he couldn't stop thinking about the transistor and how it could be improved. Backed by his friend Arnold Beckman, who was a great inventor in his own right, Shockley opened the Shockley Semiconductor Laboratory in a small lot in Mountain View in 1956. By September, the lab had 32 employees. The standards were unbelievably high. Shockley hired all of the smartest scientists he could find, many of whom compared getting a phone call from him to talking with God. Later that year, Shockley was awarded the Nobel Prize for Physics, shared with his former colleagues, Bardeen and Bratton, ensuring their scientific immortality. But before long, things at his lab started unraveling. Remember how I said that Shockley wasn't the easiest guy to get along with? Yeah, he was a tyrant. He staged public firings. He laughed at one scientist for agreeing to work for too little money. At one point, he even tried to force his entire lab staff to take a lie detector test. Shockley's young researchers were, well, shocked. They soon realized they needed to get out of there before his Behavior killed their passion for science. Huddling over lunch after one especially bruising confrontation, eight of them, Julius Blanc, Victor Grinich, John Erne, Eugene Kleiner, Jay Last, Gordon Moore, Robert Noyce and Sheldon Roberts decided to seek out Arnold Beckman, who technically managed Shockley with an either Shockley goes or we go. After hearing what these rebels had to say, Beckman initially agreed to shift Shockley out of his management responsibilities. But within a few days, he lost his nerve. Shockley may have been a tyrant, he reasoned, but he was a brilliant tyrant. And besides, Beckman didn't have any outside investors holding him accountable. He could dodge unpleasant decisions. He informed the rebels that they'd have to deal with the situation themselves. Unsatisfied, the eight young scientists went back to the drawing board. Eventually, they hit on a possible solution. Eugene Kleiner had a connection via his father to a New York investment firm named Hayden Stone. He proposed to write the firm explaining the situation and request requesting assistance. Surely a well to do investment firm could find new roles for these brilliant young men. The broker who'd worked with Eugene Kleiner's father was getting ready to retire. So he passed the letter to a young associate, Arthur Rock. As it happened, Rock was the perfect person to receive the Rebel's proposal. Two years earlier, he'd worked on a funding package for General Transistor, the first independent manufacturer of germanium semiconductors. So he knew about the promise of this emerging technology. And he also knew about William Shockley's reputation. To Rock, the fact that these eight guys were willing to quit working with God was a sign of their strong character. He was intrigued. He wrote back to Kleiner and before long was meeting him and his colleagues for dinner in San Francisco. Rock understood that the eight rebels wanted to continue working together. So he suggested something which, even though it seems so obvious today, had never occurred to them. The way you do this is you start your own company. That's how the Rebels could work independently. What's more, as founders, they would capture the true value of their technical skills. The eight scientists were stunned. The thought of starting their own company had never occurred to them. It simply wasn't done in the conformist 1950s. Raising the money to do so wasn't straightforward. The idea of raising money for these young scientists, who, although they were brilliant, had no real profile of their own, were was a hard sell. At the time, after approaching 35 potential investors, Rock had failed to raise a single cent. Then a colleague of his suggested he approach Sherman Fairchild, a self described putterer who had inherited a fortune and was interested in science. Fairchild loved what the rebels were up to, so he put in the money they needed to start their own company and they called it Fairchild Semiconductor. Shockley was furious. He called the departure of the young scientists a betrayal and predicted their new venture would be a failure. There were many reasons he was wrong, but let's just focus on one. Let's Fast forward to 1965 and zoom in on a single member of the team, Gordon Moore. Moore, who by then was making a name for himself as the director of R and D at Fairchild, was asked to contribute to the 35th anniversary of electronics magazine by making a prediction about the future of the semiconductor industry. His contribution was a short article called Cramming More Components onto Integrated Circuits. In the paper, Moore observed that the number of components in a dense integrated circuit had doubled approximately every year and speculated that it would continue to do so for at least the next 10 years. Essentially, more components on an integrated circuit meant more computing power. By 1975, exactly 10 years later, Moore had seen enough progress in the field to revise his forecast to say that the number of components would double every two years. His predictions weren't based on empirical evidence, but Moore even called his original forecast a wild extrapolation. But his forecast proved so accurate that before long it was dubbed Moore's Law. Calling it a law, though, didn't make it one. Moore didn't actually create a technology which allowed the number of transistors on a circuit to double every two years. Instead, the semiconductor industry adopted this as its goal. Companies used Moore's insight to guide their long term planning and and set targets for research and development. Almost every advancement in modern computing and electronics is tied to Moore's Law. Gordon Moore brought the future of computing and electronics into clear view. He was about to bring his own future into view as well. In 1968, he and his fellow Traitorous Eight member Robert Noyce decided to start a new company to harness the innovations in miniaturization and processing power which the transistor had made possible. Their impeccable credentials convinced Arthur Rock to arrange $2.5 million to help get the new company off the ground. Noyce and Moore called their company NM Electronics after their last initials. But before long, they changed its name to the one it still bears today, Intel. Intel is a truly pioneering company. In 1971, the year it went public, it created the world's first commercial microprocessor chip. Its big investments in new designs like the Pentium microprocessor in the 1990s and accelerated the rise of PCs as a viable industry. It developed the x86 family of microprocessors, which still forms the basis of most PCs, laptops and game consoles. In the run up to the 40th anniversary of Moore's article in Electronics, intel put out a call for an original issue of the magazine. They didn't have a copy, and neither did Gordon Moore. They eventually found one, thanks to an engineer from the UK named David Clark. Clark started collecting copies of Of Electronics and other trade magazines in the 1970s when his local library started throwing them out. Despite his wife's pleas, he held onto them for decades. He ended up finding the issue of Electronics with Moore's famous article under his floorboards. When he was asked what he would do with the $10,000 reward, Clark said that he and his wife intended to use it to pay for their daughter's wedding. Even when he didn't mean to, Gordon Moore was creating value the second law that excludes explains Silicon Valley was also born from ambition and ego. Let's talk about a guy named Robert Metcalfe. In 1973, while working at Xerox PARC, a division of Xerox tasked with creating computer related products and hardware systems, Metcalfe and his colleague David Boggs co founded Ethernet. They were inspired by Alohanet, a pioneering computer networking system developed at the University of Hawaii. Metcalfe grew frustrated with Xerox, who, despite filing a patent application for ethernet in 1975, showed no interest in building a business from the technology. So In June of 1979, he quit and founded 3Com in his apartment in Palo Alto based on his belief that Ethernet would emerge as the standard protocol for linking PCs. By September of 1980, Metcalfe had set out to raise money. The Mayfield fund valued Threecom at $2 million, but Metcalfe thought he could get more. He declared that 3Com was worth $6 million and that he could fetch $20 a share. Metcalfe wasn't exactly short on self confidence. He was known to refer to himself as a Viking American and admitted that he always resented MBAs because they got paid more than him and he was smarter than them. Besides, he knew computer networking was a hot business that was about to get even hotter. He noticed that VCs would always bring up three reasons why startups like his excessive founder ego, insufficient focus on the best product, and insufficient capital. So Metcalfe decided to flip the script. He would demonstrate his own lack of ego by bringing in an outside executive before proceeding with fundraising. Eventually, a junior partner at Mayfield recommended a guy named Bill Krause. They were polar opposites. Where Metcalf was a bold visionary, Kraus was a methodical technocrat who loved writing memos. But they also had a lot in common. Over highly competitive games of tennis, the two would discuss their shared belief that the utility of computing would be greatly enhanced when connected with other computers in a network. That central insight came to be formalized and eventually known as Metcalfe's Law. Metcalfe's Law states that the value of a network rises with the square of the number of devices connected to it. It's a close complement to Moore's Law. The two exist hand in hand, but there are also important differences between them. Remember that Moore's Law predicts that the number of components on an integrated circuit, computing power basically doubles every two years. This is called an exponential relationship. After two years, the number of components will have doubled. After four years, it will have grown fourfold. After six years, it will have grown Eightfold, then sixteen fold, and so on. But if Metcalfe's Law was true, then the value of a network would increase with the square of the number of units. So a network with three users is nine times more valuable than one with just a single user, while a network with 10 users will be 100 times more valuable. In this case, progress would be quadratic, not exponential. This was possible because networks are about the relationships created between all users. As the number of users increases, the number of possible links between them increases by the square of that number. Dynamic and fast growing networks connected by new technologies would create totally unprecedented collaboration opportunities. Another major difference between Moore's Law and Metcalfe's Law is that the latter is not constrained by the passage of time. It's purely a function of the number of users. So if you're confident that you can quickly and massively scale up the size of your network, then you'll be able to capture huge amounts of value. Moore's Law and Metcalfe's Law are highly complementary. The explosion of network size was fueled both by its growth in utility and by falling hardware costs. In case this sounds abstract, let's focus on some real life examples. Social media companies are the most famous examples of Metcalfe's Law in action. I think it's fair to say that even at its peak, the value of Facebook was never really about the quality of the product. We were on it because all of our friends were there too. Remember Google? It was meant to be a Facebook killer, but it failed because it could never attract enough regular users. So it was never able to replicate the social feeling as effectively as Facebook. Ebay is another cool example. Its incredible early popularity can be totally explained by Metcalfe's Law. The more people listed stuff for sale, the more that buyers were drawn to the site, which in turn drew more sellers who knew that they could get a fair price in a liquid market. It was the perfect positive feedback loop. What's more, furthermore, because ebay owned the network, it captured all the value created by its growth. We can see Metcalfe's Law in places other than the Internet. Tesla's self driving technology also benefits from network effects. You might think that having 100 self driving Teslas on the road is just 100 times as valuable as having one, but that's not quite right. Tesla is betting on self driving technology becoming dominant. So it's important for them to think ahead to a future where self driving technology is dominant by taking the time today to learn how their cars react when interacting with other autonomous vehicles. It's Metcalfe's law in action. Metcalfe's Law isn't always precisely correct, but it is a great way to understand why every type of tech company endlessly chases new users and subscribers, and why the companies that succeed capture huge markets and make massive profits. The third and final law that explains Silicon Valley wasn't actually discovered in the Valley, but that's where it's had its most powerful impact. This law is often referred to as the Pareto Principle, after the Italian economist Wilfrid Pareto. You might know it as the 8020 rule, meaning that roughly 80% of an outcome results from 20% of the inputs. But it's best known as the power law, and it's the final step to understanding modern tech company valuations. Let's go back to a man I mentioned when discussing Moore's Law, Arthur Rock. I said that Arthur Rock helped Gordon Moore and Robert Robert Noyce found Intel. I kind of buried the lead though, because in any honest history of the venture capital industry, Arthur Rock is one of the main characters. His history of picking winners is ridiculous. In 1957, when the traitorous 8 left Shockley semiconductor Laboratory, Rock helped them find a place to go. Davis and Rock, the firm he co founded with his colleague Thomas Davis Jr. In 1961, was one of the biggest and most successful and early VC firms. That same year, he put up $280,000 to help start Scientific Data Systems. Eight years later, that company sold for $950 million. He was so critical in the early days of Intel. He's often given equal billing with Moore and Noyce as a co founder. In 1978, Rock invested $57,400 in a fledgling personal computer company with a whimsical yet evocative Apple. The power law underpinned Rock's approach to investing for decades. In a speech in San Francisco in 1962, he explained that venture portfolios need grand slams to average out the goofs. It's actually a great description of how the VC industry has operated since its inception. Consider this. In early 2021, Y Combinator's list of top companies showed that about 5% of the more than 2,500 companies they had invested in were responsible for generating a huge majority of total returns. The top generated 3/4 of the returns. Just three companies, Airbnb, DoorDash and Stripe, represented more than half of YC's top companies portfolio value alone. Bill Gurley, who's a general partner at Benchmark, one of the top VC firms today, said it well, VC is not even a home run business. It's a grand slam business. But why do such a small number of companies consistently generate such a huge share of portfolio returns? You can see the power law in action in other industries, but it's more prevalent in tech because tech companies are themselves based upon technologies that progress at exponential rates. Arthur Rock understood this completely. When co founding Davis and Rock, he insisted that they would only invest in ambitious high growth companies whose value might jump at least tenfold in five to seven years. Most startups will fail, they reasoned. Sometimes it's because the technology they rely on isn't mature yet. Other times it's because they fail to establish the use case for their product. Sometimes it's just plain bad luck. Therefore, the winners would need to win big enough to make a success of the entire portfolio. As Davis put it, trying to play it safe in small companies is, to my mind, self defeating. But by realizing that paradoxically, the best way to manage investment risk was to embrace it, they recognized the fundamental insight of the power law. If just a handful of outliers hit grand slams and dominate entire markets, or even create new ones from nothing, then you've gotta swing for the fences. VCs don't always succeed. Critics love to focus on what Arthur Rock called the goofs. But then most companies don't succeed either. In a random group of 100 tech companies, 90 will probably fail. Eight or nine of them will deliver decent returns. But because of the power law in combination with Moore's Law and Metcalfe's law, one or two will probably generate insane returns. That's what explains why Google makes more profit than the entire global airline industry. And it's what VCs are looking for when they back speculative innovations like flying cars, artificial intelligence, or Elon's plan to fly a rocket to Mars. Their job is to look beyond the horizon to a future that has yet to be built. Understanding how Moore's Law, Metcalfe's Law, and the Power Law interact is key to understanding modern tech companies. And now that you understand the three laws that explain modern tech companies, watch my video about the amazing history of Silicon Valley next. Thanks for watching.