A weekend in Omaha 2017

Machine Learning And A Learning Machine

Chetan Parikh | JUN 08 , 2017

Illustration: Kishore Das

In a digital era, Google, Apple and Amazon are examples of companies with the right trappings that will reinforce their moats

“It was a very good sign that Warren bought Apple. Either he’s gone crazy or is learning. I prefer to think he’s learning.”

— Charlie Munger, Berkshire Hathaway Annual Meeting 2017

A recent issue of The Economist (May 6, 2017) had a cover story titled, “The world’s most valuable resource...”– data – which the magazine called “the oil of the digital era”. It went on to state: “Machine learning will extract more value from data...This abundance of data changes the nature of competition. Technology giants have always benefited from network effects...With data there are extra network effects. By collecting more data, a firm has more scope to improve its products, which attracts more users, generating even more data, and so on...Vast pools of data can thus act as protective moats.”

Mr Buffett noted at the recent Berkshire Hathaway Annual Meeting: “The five largest US market cap companies are $2.5 trillion, close to 10% of the US market. One could run them with no equity capital at all. It’s a very different world than Carnegie’s steel mills and Rockefeller’s tank cars.”

So, how different are the moats and the economics in this world – the world of the Apples, Amazons and Googles – from those of the past century on which much of our understanding of competitive advantage and strategy is based?

Go back to an article in the April 2016 issue of the Harvard Business Review titled, “Pipelines, Platforms and the New Rules of Strategy”. The authors distinguish between conventional “pipeline” businesses that have been dominant for a long time and whose value creation can be understood by examining the value chain, versus technology-based “platform” businesses which facilitate interactions between participants and increase their value as the network grows. Modern “platform” businesses rely extensively on information technology that reduces the need to own physical infrastructure, makes scaling up relatively simpler and cheaper and enhances the ability to capture, analyse and exchange large amounts of data that increases the platform’s value.

Some businesses, as will be seen later when analysing Apple, have both “pipeline” and “platform” assets, which reinforce and strengthen their moats.

Building your moats
Recent history has not been kind to pure play “pipeline” businesses, owning high quality tangible and intangible assets, when “platforms” (whose assets are hard to copy like community and the resources its members contribute) enter the same marketplace – “platforms” usually have won. And when there have been “platform wars” for dominance in a given domain, the outcomes have had a “winner take all” stamp about them.Supplier and buyer power are menacing (“depletive”) in the “pipeline” world; they may actually add value and be “accretive” in the “platform” world. Governing the ecosystem and understanding which activities are “accretive” and “depletive” is a key challenge in the “platform” world.

Google and Apple offer contrasting studies of “platform” rules and architecture. Google kept Android “open”, betting on scaling quickly in a then highly fragmented mobile operating space. Apple based the iPhone on a “closed” iOS, preferring control. Both made sense in those days given their respective strategies and objectives. But today, in the era of machine learning, which is one of the main artificial intelligence techniques, open systems have the edge.

Turn to Apple which has both “pipeline” and “platform” parts that are so harmoniously inter-meshed. Apple is vertically integrated, owning its own chip manufacturing, controlling manufacturing, having extremely strict software standards and having a tight control over all parts of its value chain right up to its proprietary retail stores.

Apple’s retail stores give a sense of what is so distinctive about the “pipeline” part of the company. Apple’s retail innovations (“concierge” greeting customers; the Apple Genius Bar) were inspired by the luxury hotel chain Four Seasons and its founder Isadore Sharp, who like Steve Jobs was a perfectionist. Sharp’s guiding principle was: “So much of long-term success is based on intangibles. Beliefs and ideas. Invisible concepts.” And Steve Jobs did ask, “What would the Four Seasons of retail look like?”

Jobs was also Disney’s largest shareholder and he studied the consistently superior guest experience that was the hallmark of Disney. So, branding has always been incredibly important to Apple and its founder. Apple has never gone for early adopters or the budget-conscious, targeting instead customers who had the patience and were willing to pay for highly sophisticated product designs. It has always sought the premium end.

Jobs from his early days saw computers as consumer accessories, believing that “it is the intersection of technology and liberal arts that make our hearts sing”. According to Jobs, all activity at Apple started with the question, “How easy will it be for the user?”

Any consumer products company will agree with what Jobs once said: “Our brand is like a bank account. We have the opportunity to put deposits in the bank account by making a great product, something they really love, or we can do withdrawals, putting something out there that we know is not good enough but still putting our name on to it.”

Mr Buffett was spot on in stating at the AGM that Apple was much more of a consumer products company, in terms of analysing the moat around it. But there is also a “platform” part of Apple which complements the “pipeline” parts of products, retail and branding.

In 2007, there were five major mobile phone manufacturing companies – Nokia, Samsung, Motorola, Sony Ericsson and LG – which collectively made 90% of industry global profits. That feasting did not last with the introduction of the iPhone that year. By 2015, iPhone generated 92% of the industry’s profit and only Samsung on the Android operating system (from Google) made money. The incumbents had moats of brand, scale and intellectual property. However, at that time, they were all “pipeline” businesses. Apple brought a “platform” business (iTunes and the App Store) with its iPhone, connecting app developers on one side and app users on the other. Switching costs were non-existent between Apple products but severe for a user migrating out of the ecosystem.

Which areas could Apple potentially disrupt with its pipeline/platform complementary assets? It has already introduced the watch in the “wearable technology” category. Other areas could be in gaming (where the combination of high process speeds and video refresh rates in the iPad along with streaming technologies in the TV market could prove a threat to high-end gaming consoles); home entertainment (Jobs talked to Walter Isaacson in his biography of a uniquely easy-to-use television); payments (the Apple Pay product on the iPhone 6 could leverage on the installed base of 800 million iTunes customers who have credit cards on file); retail services (building on its in-store tracking technology iBeacon, Apple could assist brick-and-mortar retailers in providing personalised shopping recommendations) and mobile wellness and health care solutions (health and fitness monitoring applications to be incorporated on all its mobile computing devices) among others.

Retail juggernaut
At the recent Annual Meeting, Mr Buffett also spoke on Jeff Bezos and Amazon. He said: “We didn’t think that Bezos would succeed like he has in retail. We underestimated the brilliance of the execution. You can read Bezos’ annual report in 1997, and he lays it all out. And he has done it and done it in spades. It just always looked expensive.”

Amazon is a “platform” business and accounts for 51 cents of every dollar spent on online retail owing to its broad product offering and Prime membership benefits (Prime is Amazon’s premium home delivery service and attracts a growing following of highly loyal and profitable customers). Its founder and CEO Bezos is in ‘Day 1’ mode because according to him in his 2016 letter: “Day 2 is stasis. Followed by irrelevance. Followed by excruciating, painful decline. Followed by death. And that is why it is always Day 1.”

Amazon is the 800 pound gorilla of online retail but is facing attacks from the world’s biggest retailer, Wal-Mart, on its online turf. So, what are Amazon’s moats and what does it need to execute correctly to keep on widening and deepening them?

Wal-Mart has been making acquisitions to build its e-commerce presence, having spent almost $4 billion in the past eight months. It is not betting on books, electronics or toys but on product areas that are becoming popular online such as apparel, fresh food and drugstore items. Amazon prefers to grow organically, largely owing to the lessons learnt from its acquisition spree in the late 1990s when it failed in the numerous start-ups that it bought. And of the few acquisitions that it has made in the past two decades, it has preferred marketplaces and not brands owing to their larger addressable size and scalability (inventory is carried on the books of merchants who deal on its platform; the inventory though is stored in its numerous fulfillment centres). Wal-Mart has been buying brands and multi-brand retailers so that it can deal with ‘long-tail’ categories (endless assortment of products that online retailers can offer). As a result, Wal-Mart’s online inventory will increase. A final difference in strategy is that as Wal-Mart continues to buy brands, Amazon is bringing out private labels (with higher profit margins) based on the huge data set that it has accumulated over the past two decades.

What is Amazon doing which gives it an edge and which Wal-Mart will not be easily able to replicate? More fulfillment centres and quick deliveries? Wal-Mart has probably more distribution centres and certainly more storage space. Rejigging that network for e-commerce fulfillment would not pose a challenge. Last-mile delivery? Amazon has drones, delivery robots and an on-demand human delivery network. These can be outsourced by Wal-Mart. They are not Amazon’s moats.

Amazon built its own tech infrastructure and later converted it into an external product – Amazon Web Services. Beyond the obvious financial benefits, opening that operational piece to external customers helped that part remain in a “Day 1” mode.

Many parts of Amazon’s operations could be converted into externally consumable services, just like the fulfillment services for even non-Amazon orders. There would be financial benefits but the primary objective would be that customer-centricity remains and “Day 2” does not arrive. A culture of forcing many parts of its internal operations to compete for external clients would more than any other factor ensure lower prices, lower costs of shipping, faster deliveries and operational excellence.

Could Wal-Mart not replicate this? Maybe, but there would be a long learning curve and high costs associated with the tuitions.
The ‘Day 1’ culture is Amazon’s biggest moat.

Master of the web
“Google has a huge new moat, in fact; I’ve probably never seen such a wide moat.”
— Charlie Munger in 2009

Alphabet’s Google relies on its search-based advertising platform for the bulk of its current revenues. Google’s key assets are the quality of its algorithms that run on its data and the “smart creatives” and the talent that the firm has hired to develop them. And the recipes (algorithms) have kept improving with the amounts of ingredients (data). The accumulated knowledge that resides in Google’s data refineries makes it difficult to replicate asset. Artificial intelligence-powered services by Google will probably remake everything the company does. Google’s artificial intelligence platform could, potentially, impact the ecosystem of every single industry.

The threats to its moat go beyond challenges such as downward pressure on its mobile ad monetisation. Can Google maintain its massive lead in data capture? Social networks are becoming popular enough to limit the usefulness of internet searches. Niche search engines could gain more traction against Google’s generalised system, although it has attempted to counter this. The early lead that Amazon has going in its Alexa voice-operated service could translate into dominance in voice search which could dent Google. Finally, regulatory risks concerning data protection and antitrust cannot be ignored, but this is a risk faced by all tech- based “platform” companies.

What really matters
There are three questions that every investor should ask when looking at “moats”: 1) How wide and deep are the moats? 2) Are the moats growing? and 3) How sustainable and durable are the moats? The last question is probably the most difficult and the most pertinent when looking at tech-based “platform” businesses. But in the digital era, where “platforms” can quickly change shape and have been and can be a disruptive threat to every “pipeline” business, the last question on moat durability is probably the most important for every investor while analysing any business.

Henry Ford once said, “Anyone who stops learning is old, whether at twenty or eighty. Anyone who keeps learning stays young. The greatest thing in life is to keep your mind young.” Little wonder then that Mr Buffett keeps getting younger every year. Investors should too.

The writer is a co-promoter of Jeetay Investments Pvt Ltd and Jasmine India Fund

Here's your chance to read the latest issue of Outlook Business for free! Download the Outlook ‚ÄčMagazines app now. Available on Play Store and App Store
More from A weekend in Omaha 2017