Network Effects and Other Moats

Morningstar has five economic moats many investors use to categorize companies. They include Network Effects, Intangible Assets, Cost Advantage, Switching Costs, Efficient Scale. There are various iterations of these five.

Ah, if in case you are not familiar with the concept of moats please type in "Warren Buffett Moats" in Google to get up to speed. It'll improve your life going forward.

Helmer’s 7 Powers includes Counter Positioning, Cornered Resources with some similarities between Process Power and Economies of Scale. With a growing and shrinking list of various ‘competitive advantages I’ve heard it things get narrowed down to the four essentials of Intangible Asset, Switching Costs, Network Effects, and Cost Advantage.

The fact is that when companies have one type of advantage, they tend to have others. AMZN has scale but it also has a brand and it also has cost advantages and there are switching costs involved for some products (i.e. AWS, where I buy toilet paper).

Naturally, I was thinking about whether there was a foundational moat. Does one come before the other? I started with the biased view that network effects encompassed most, if not all, the various moats people talk about.

By definition, a network effect exists if each incremental user/customer adds additional value to the entire ecosystem/product. The greater the users on the platform the exponentially greater its value to the next one.

As Facebook gained more users it continued to become exponentially valuable. I might’ve joined because 30 of my friends are on it but as more people go on it becomes even more valuable to me. Same for the advertisers who are Facebook’s customers.

Same for languages. English gets more valuable as more and more people learn it, use it, and publish important material with it. It is highly unlikely Korean will become a universal language let alone German with their larger population. Network effects follow this power law where it can continue compounding at greater rates until it eventually becomes a monopoly.

It’s funny to read some books that talk about how Google was a monopoly with a 60% market share in search because it’s at ~97% in 2020. By default, network effects may become monopolies.

So, if a company has network effects…will it not eventually gain scale advantages? If it truly is something where each additional user adds value to incumbents and makes it of greater value for future users…..it should follow an exponential growth curve.

Most businesses that achieve this also have zero marginal costs. Usually, it’s a set product/software that may get updates but it’s the same product for all users. You build it once and people will come. So, you end up front-ending the development cost with venture dollars to pay the developers.

By that nature, these companies should have a cost advantage given its products have zero marginal costs. Then, as the network grows and its value compounds…it becomes harder for people to switch away from it to move to something else.

I had moved away from Google for about 6-10 months. I used Firefox as a browser and DuckDuckGo as a search engine but I’m making the move back to Google because the products are so exponentially superior. I like Firefox and DuckDuckGo as companies but as a user, I need utility with such tools and they aren’t there for me yet.

Same with merchants on AMZN or homeowners on Airbnb. There is a cost to switching out of these services.

The network also creates the intangible asset of "brand". One can also argue that it’ll have way more data too. Data scientists love working at big tech companies because of the sheer number of data points they can play with to arrive at some statistically meaningful result that is not possible at a startup.

On the brand side, there is a trust that brews up over time through review ratings, or just the sheer number of people who use it signal a social trust towards it. This is probably why people feel betrayed when they buy something off AMZN and it turns out it was a fake from a fraudulent merchant (this was a common case for USB drives last time I checked).

This makes me believe a company with network effects will have (or will be on the path to establishing) every other form of competitive advantage with time. This won’t necessarily be the case for others.

A business with economies of scale may not be able to have network effects. This would be the case for companies that become viral. Virality is rapid growth but this could’ve been achieved with lots of marketing dollars. It doesn’t necessarily mean the product has network effects or even stand-alone value for that matter.

Some companies with the brand as a moat have that value precisely because it does not have network effects. Ferrari’s have value because of exclusivity. If someone else has a Ferrari, the value of my Ferrari declines.

ERPs like Oracle or SAP have high switching costs for its customers but it doesn’t matter to me as a big bank if millions of others use Oracle or SAP. It might contribute to the trust factor to its brand but it doesn’t necessarily make the product better.

In some ways, the test for whether a network effect exists for a product could be on how it’s distributed. If it grows from organic word of mouth (i.e. Slack, Atlassian, FB) where one person adopts it and it leads to others adopting it…it probably has a network effect. But, if the company has to continuously spend lots of marketing dollars to acquire users…..it might eventually get it but it doesn’t seem as clear cut. It might be a battle between a handful of players for a commoditized product (i.e. food services).

This doesn’t mean using ad dollars is bad. Some can argue Google succeeded because it spent up to become the default search engine for browsers when it was first starting. They still do it…I think ~$20b of Apple’s revenue is Google paying them to be the default search engine for Safari.

But in the pursuit of making things simpler for myself….analyzing whether network effects exist or not might be the first (if not only) step when analyzing the business model of an investment. By default, having multiple competitive advantages compounds a company’s strength and if network effects lead to more…then that’s what I want in my investments. Especially because the game is to find the few companies that will become monopolies which will be responsible for most of the returns in the market.