The link above is an excellent analysis of Bloomberg’s moats that I found on a tech start-up blog, which highlights an interesting point:
The best way for an investor to examine a company’s economic moat and competitive positions is to take the point of view of a potential entrant into the industry, and figure out how much it would cost to replicate/disrupt the incumbent’s business model.
I’m sure there are many sharp minds out there who already employ such a framework in business analysis. What really differentiates the quality of the analysis though, in my opinion, is the approach people take in understanding the relationship between cost of replication and economic moats.
First – Level Thinking
I teach a little value investing night class and read a lot of analysis every year from the students. One thing people often do is they look at a business and make an educated guess about the cost of replicating its business model. When that cost is very high, they conclude that the business has a high barrier to entry, which entails the existence of some forms of a “moat”.
But that’s not a complete analysis – most capital intensive businesses have a high nominal cost when competitors seek to replicate their resources, yet very few of them have moats. At the end of the day, the size of the cost of entrance should not be a prohibitive factor as long as the said entrant has ready access to capital markets (which is especially ample for large public companies merely making a push into a different vertical).
For example, a skyscraper office tower can cost hundreds of millions of dollars, if not billions to build – a significant sum, yet 80-story office buildings get built all the time, often right next to four or five other 80-story office buildings! The nominal cost of entry in this case is obviously not a significant enough deterrent.
Second- Level Thinking: An Improved Framework
This is because it’s not the nominal sum of cost of entry that creates moats. What really matters is the return on capital that an entrant can generate and the “cost of capital” associated with entrance. While giant office towers cost a ton of money to build, office tower operators do not have much of a moat – new entrants trade a low cost of capital (interest rate on development debt) for a reasonable return on capital (margin on sale, or net rentals yield should he choose to rent it out). As long as they can obtain proper financing, the economics of the business does not discourage them from entering into the business.
A real moat exists when the 1) cost of capital (for a well-financed and rational new entrant) is greater than the 2) long-term steady-state return on capital after entrance. This in turn forms a barrier to entry that discourages new players from emerging and allows the incumbent to earn above-average economic returns. Potential entrants are kept at bay not because they are discouraged by the initial capital requirement, but because they figure that there are better ways to make money. A complete analysis of moats, therefore, would require the analyst to make an educated guess about each of the two variables.
A Case Study: Part of the rationale behind Berkshire’s purchase of Coca Cola in 1987 was that Buffett estimated it would cost over $100 billion for a new entrant in the soft drink business to replicate Coke’s brand and distribution scale. Coke made a little over $900 million in 1987 and could reasonably make $1 billion the next year. That means if I were to start a viable competing brand, I would have to spend $100 billion and many years of energy to get $1 billion of annual earnings (in reality it would be much lower than $1 billion since returns for all industry players would shrink once a new entrant turned a duopoly into an oligopoly). That’s a sub-1% return on capital. Even if you factor in the long-term growth in earnings, the steady-state return would be nothing spectacular.
On the other side of the equation, natural risk aversion, exacerbated by the sheer size of the capital requirement (risks in gathering $100 billion of financing) and the potential execution risks of frontally challenging a well-entranced incumbent, means the “cost of capital” to a rational investor would be extremely high. How much return would investors/lenders demand if you asked them for $100 billion to start a new global cola brand in 1987? There’s no way to know that precisely but it would be a very ugly number, certainly ugly enough to discourage someone from chasing a sub-1% return.
In many cases, even the framework above carries significant limitations. This is because one aspect of competitive analysis that is frequently under-appreciated is the role played by irrational financiers.
Suppose you own a cable system in Los Angeles that makes 14% on capital a year. You have a decent business on the video side that only competes with two satellite companies and a very good business on the broadband internet side where there’s not much competition. You have a nice little semi-regulated natural monopoly whose moat derives from the fact that your infrastructure efficiently serves a market that is limited in size – there’s simply not enough room to provide adequate returns to two local players. Plus, if a telco operator wanted to invade your territory with a fiber rollout, they would also be sub-scale in content acquisition, face franchising issues with municipal governments, and cannibalize their existing DSL service.
Well, it all sounds good but even with the odds heavily stacked against them, Verizon decides to overbuild in your city anyway, with no compelling economic justification. It’s unlikely that the return on the fiber overbuild can justify the cost, but they can afford it, and the “growth” their shareholders would like to see has to come from somewhere. The net effect is that despite the presence of a strong moat, the uneconomic entry permanently impairs the value of your cable system, and there’s nothing you can do except to share your misfortune with your competitor’s shareholders.
The irrational growth in the example above is financed internally, likely a result of some forms of institutional imperatives – escalation of commitment, empire building, you name it, but it’s not the only way moats can be threatened by irrationality. The existence of a dynamic, but perpetually inefficient capital market means there will always be some irrational behaviors backed by equally irrational financing, and the adverse economic effects on incumbents can be especially enduring when the scale of irrational competition snowballs over a positive feedback loop, or as George Soros calls it, “reflexivity” (his 1970 essay on investing in REITs illustrates this effect well and is a must-read).
This is because even if the entrant’s business plan is uneconomic, as long as it is perceived favorably, investors will finance the venture at a low cost of capital (happens more often in the equities market). The entrant takes the cash and reinvests in profitless growth, which reinforces investor recognition in its growth prospects. The achievement of growth causes capital markets to continue to reward the venture with ever-higher multiples over whatever irrelevant financial metric (just not free cash flow) they can stretch to justify, further lowering the cost of capital, and allowing the firm to finance more “growth”. The end product of a flawed investor perception is a self-fulfilling cycle that allows many uneconomic entrants to stay in business at a significant scale for prolonged periods, wreaking havoc to the incumbents and causing permanent damage to their value. There are quite a few good examples of such a phenomenon today in companies that operate under a hot idea but never seem to make money.
I think there are four major reasons for the existence of long-term investor misperception:
1. Accounting shenanigans that make the economics of the entrant appear artificially appealing, such as under-depreciating assets, or paying salaries in options/using non-GAAP standards to show a much higher EPS.
2. Concepts that appeal to hope/imagination/emotions. Air travel used to be such a hot idea (investing in it turned out to be not so hot). Incumbents in such industries often do not have any moats to speak of in the first place, which means excess capacity financed through investor misperception often leads to across-the-board value destruction on a massive scale. Other examples include renewable energy, gold mining, etc.
3. Investors extrapolating abnormal returns during cyclical upturns into perpetuity.
4. Most important of all, the presence of a promotional management that is very capable of consistently selling its idea to the investing public.
The reason I mention this is that it adds a new dimension to competitive analysis – an investor must not only understand the rational economic behavior that drives investment return outcomes, but also the effects of market sentiment on the fundamentals of real businesses.
In my opinion, the safest way to invest in quality (moat) and make sure the value of the moat does not get diluted by irrational competitive behavior is to invest in companies with moats so strong that no irrational entrant can sustain competition, even backed with significant financing.
The network effect often constitutes such qualities. Companies that benefit from the network effect are usually “cooperatives” that provide some forms of a service platform. They tend to be monopolies (since there is no reason for anyone to use a smaller network that brings less value to the party) and operate with a very large fixed cost structure. Successful entrants must overcome years of huge losses while attempting to present to their customers a difficult value proposition that a smaller network can be superior to a larger one – a virtually impossible task given the nature of the effect itself. Consequently, new entrants are killed off very quickly and cannot generate the momentum necessary to form a positive feedback loop in capital markets.
Not sure how I wandered off from Bloomberg to promotional managements, but modern competition has multi-dimensional dynamics and is pretty confusing to think about.