We have argued in a previous article that the five titans of technology – Amazon (NASDAQ:AMZN), Google (GOOG, GOOGL), Facebook (NASDAQ:FB), Apple (NASDAQ:AAPL) and Microsoft (NASDAQ:MSFT) – are basically unstoppable.
In particular, Amazon, Google and Facebook are driven by multiple increasing return feedback loops, which makes their position virtually unassailable, we concluded. We also pointed out some downsides (here and here). These positions are getting on the nerves of some antitrust regulators, more specifically in the European Union.
Antitrust cases used to be simple. Companies had a dominant position, which led to a loss of consumer welfare and/or unfair competitive practices. Both of these were fairly easy to identify. A dominant position can be identified with the help of market share. A loss of consumer welfare or unfair competitive practices are a little harder to demonstrate with iron-clad proof, but usually these involve some price hikes or acti ons that limit competition or ties in buyers.
In the digital economy, this is all quite a bit harder to do. Take, for instance, the EU case against Google. The EU argues that Google has a dominant position in search and in operating systems of mobile phones.
So far, so good – these are not controversial by any means and extremely hard to dispute (Google’s market share in search is above 90%, for instance). Where it becomes much more tricky is when the EU argues that Google abuses both dominant positions.
With respect to search, the EU argues that bundling it with its own ad business (AdSense), it has abused its dominant position in search. But Google can (and has) argue that its search engine has made e-commerce more efficient for consumers despite the bundling with AdSense.
Also, it’s difficult to point out the consumer harm as a result of the dominant position for Android. Google simply argues it has made it cheaper and faster for device manufacturers to bring devices to market, thereby lowering prices for smartphones and increasing the distribution opportunities for apps.
The company could also point out that both its search and Android are free products. So where is the consumer harm?
Why EU and not the US?
Why has the EU taken the lead in the cases against the digital behemoths? Is this just an example of US bashing? Not necessarily, the issue is more likely to be the result of a different philosophy that informs antitrust policies in the EU compared to that of the US.
The US antitrust policy is informed by increasing consumer welfare that could be described as “efficiencies are the goal; competition is the process.” Seen in that light, it is difficult to start a competition case against Google, as there doesn ‘t seem to be much, if any, consumer harm.
However, while similar concerns also inform EU competition policy, the latter is more concerned with dynamic efficiency, i.e., how things are likely to play out in the longer term.
Here is where it gets interesting. Google allows mobile device makers to alter Android, that is, to install their own version of Android (a so-called Android fork), but not if they want to install Google search and Google Play Store as default options. So, according to Harvard Business Review:
Consequently, device manufacturers that adopt the Google stack must bundle both Google Search and Google Play, which essentially locks out the other search engine providers and store fronts. In other words, while Google supports divided technical leadership for part of the stack, it tightly controls the applications layer.
Here where the EU balks, as it argues that this reduces future competition in search and reduces the incentives for d evice makers to make their own version of Android (limiting competition here as well).
There are good reasons to rewrite at least some of the rules of antitrust policies. After all, these were written for an era when the industrial economy produced the first abuses of power, and that is a fundamentally different economy than that of today. Not only is industry much less important, but a new sector emerged, armed with new ways to build dominant positions that could have detrimental effects for consumers.
One fundamental issue that needs to be addressed is the changed form of competition that the digital economy produces, which can perhaps best be described as the interlocking of ecosystems, or platforms.
Platforms are often subject to strong increasing return mechanisms, making it difficult for newcomers to unseat established power. What’s more, a dominant position on one platform can often be leveraged to enter others. This is something Microsoft has shown first, when it tried to leverage its dominance in PC operating systems to leverage its b rowser and hobble the competition (Netscape).
We are inclined to argue that competition authorities should be very careful before dismissing any harm to consumers. Before we know it, we would have allowed the creation of companies which are truly dominant in one form or another without authorities having the means to counter this. Here is MKM Partners’ Rob Sanderson pointing this out (from Barron’s):
Internet ecosystems are like black holes where mass attracts mass, competitive advantage widens and eventually strangles sub-scale competitors. This may be even more so in the future as the usefulness of A.I. will be determined by the amount of data and ability to invest. Mobile is making Internet delivery applicable to many more sectors of the consumer economy than the desktop Internet could address. Cloud computing enables rapid innovation cycles and allows applications to reach global scale almost immediately. This continues to strengthen the value proposition o f juggernauts. Internet-based systems are far more efficient than those of incumbents. These businesses drive very high margin at lower price, but this shift also means fewer jobs.
And sometimes a picture says more than a thousand words. Mind you, it’s not all due to digital dominance, but then you probably have to add a few Chinese players:
Philosopher Jeremy Bentham designed a prison system in which one guard, placed at the center, could watch over all prisoners in a circular glass design. Light would come from the outside, making every prison cell completely transparent for the observer in the middle.
This is a nice model with which to describe what these big tech platforms are doing. Google has a point when it argues that it doesn’t produce consumer harm, as discussed above. But what if it is simply offering these services for free (despite a dominant position) in order to amass a wealth of data about users, which it can then monetize in other ways?
One way is offering these users better data, simply because the company has a better understanding of their needs. Of course, these are personalized ads in the first place, and you can even argue that the more personalized they become, the less intrusive they might seem.
But we have highlighted previously that this info can also be used for other purposes, like hitting impressionable voters with just the right message to swing his/her vote. This isn’t necessarily done by Google or Facebook or another platform but by parties who know how to use part of the information available via these platforms or have ways of synthesizing it.
The end result is what another Harvard Business Review article calls a digital replica:
A digital replica is a digital representation of an individual, object, or asset. Such a representation is constructed based on an individual, object, or assets interactions with its environment.
For physical assets, like jet engines, having a digital replica is very useful, as it allows stuff like predictive maintenance (sensors and algorithms spot problems before they happen) and outcome-based pricing. People can tune into some of this with wearables, which are about to revolutionize healthcare. But at least you’re doing it yourself, with a clear purpose and benefits.
Companies like Google and Facebook have multiple ways to gather highly personal data of its users and – this is new – consolidate these across numerous fields. From HBR:
Historically, many firms have had deep knowledge about only specific facets of an individual’s life. For example, financial institutions knew the financial lives of their customers; retailers accumulated knowledge about their customers’ buying habits; and even libraries had information on users’ reading habits. What they lacked, however, was a composite knowledge about individuals that came through the pooling of such information.
For instance, APIs are very powerful tools for data integration, and one of the main reasons why you see these companies moving into other services is simply to open up another front on data collection.
An example is Facebook’s social login, which simply looks like a handy service for users, as it allows them speed and ease of use, using Facebook login details for a host of other sites. But social login is a big integrator of data. Cookies don’t work very well on mobile and are not cross-platform compatible, so it’s difficult to track people’s activities on different platforms. Social login provides a neat way to pull this off. And, of course, Facebook completely dominates this social login space with 69% of the market (and guess who’s number 2? Google – with 29%).
So, one can conclude that the ability to collect, connect and integrate data from various sources will become the new competitive arena. This would not be the problem it is if ma rkets for data would work like other markets. But they don’t. Markets for data are subject to considerable increasing returns. That is, those who already have a lot of personalized data are likely to get more, simply because they can offer more personalized services.
So, it allows the companies who sit on top to do things competitors cannot match. This is already blatantly obvious in the ad market. From Business Insider (our emphasis):
The 10 leading ad-selling companies accounted for 73% of total revenues in Q4 2016, according to the report. So who are these 10 companies that grab the largest share of these revenues? The report didn’t say. But analysts for the Pivotal Research Group, cited by Reuters, reported the only two names that really matter: Facebook and Google. In terms of the industry growth, so in terms of the 22% or $12.9 billion year-over-year increase in total internet advertising revenue, Facebook and Google together grabbed 99% of the growth! They’re sitting at the sweet spot. Everyone else is fighting for crumbs.
AI, the next data frontier
Dominating advertising is one thing, dominating the future is quite another. The latter is a bit of an exaggeration, of course, but artificial intelligence (“AI “) is rapidly growing into one of the key – perhaps the key – innovative battlefields in many sectors of the economy, from healthcare to autonomous cars to finance to fraud detention and a whole host more.
This is a revolution that has only just begun. AI is based on feeding machine learning with Big Data. Control the data allows you to build better algorithms, which allows you to offer superior solutions, which then allows you to gather even more data.
Libertarians greatly object to Big Brother government, but Big Brother private for-profit organizations might be much the bigger threat. It allows them to amass unique and difficult-to-replicate knowledge and expertise in all kinds of fields, which can then be monetized, often without having real competition.
Here is the Guardian (our emphasis):
Google’s AI-powered health tech subsidiary, DeepMind Health, is planning to use a new technology loosely based on bitcoin to let hospitals, the NHS and eve ntually even patients track what happens to personal data in real-time. Dubbed Verifiable Data Audit, the plan is to create a special digital ledger that automatically records every interaction with patient data in a cryptographically verifiable manner. This means any changes to, or access of, the data would be visible. DeepMind has been working in partnership with Londons Royal Free Hospital to develop kidney monitoring software called Streams and has faced criticism from patient groups for what they claim are overly broad data sharing agreements. Critics fear that the data sharing has the potential to give DeepMind, and thus Google, too much power over the NHS.
Stuff like this is very much a double-edged sword. In and by itself, this is a very welcome development. AI allows to deliver all kinds of medical insights that have significant potential, and blockchain could make electronic patient record secure.
But this is stuff done by a private company with a for-profit motive in a market that will be difficult for others to replicate. And in case you think we’re imagining things, here is The Verge:
A deal between UK hospitals and Googles AI subsidiary DeepMind “failed to comply with data protection law,” according to the UK’s data watchdog. The Information Commissioner’s Office (ICO) made its ruling today after a year-long investigation into the agreement, which saw DeepMind process 1.6 million patient records belonging to UK citizens for the Royal Free Trust – a group of three London hospitals.
We could become terribly dependent on companies like Google. It is likely they will develop expertise on broad swa ths of modern life, which governments simply cannot match. For instance (from HBR):
Tomorrow’s AI aggregators will be able to detect and counter “fake news” by scanning for inconsistencies and routing people to alternative perspectives.
We basically have to trust them doing a good job, but even if they intent to do just that, the problem with algorithms is that we don’t really know how they are reaching their conclusions, as they don’t really produce any reproducible logic. That is, they could contain all kinds of hidden biases. From MIT Technology Review:
But these AI systems also make decisions for reasons we may never understand. That’s why researchers, consumer rights lawyers, and policy makers have begun to voice concern that unintentional or intentional bias in machine-颅learning systems could give rise to patterns of algorithmic discrimination with causes that may be difficult to identify. This isn’t theoretical: studies have already found evidenc e of bias in online advertising, recruiting, and pricing, all driven by presumably neutral algorithms.
We’re only at the very start of AI, which, for decades, was an interesting idea but hobbled by a lack of Big Data, cloud storage and processors that enable to use massive amounts of data for machine learning. But now that these have arrived, AI is making very rapid progress. Computers are closing in on understanding the world like we do. From Quartz:
If we ever want future robots to do our bidding, they’ll have to understand the world around them in a complete way – if a robot hears a barking noise, what’s making it? What does a dog look like, and what do dogs need? AI research has typically treated the ability to recognize images, identify noises, and understand text as three different problems, and built algorithms suited to each individual task. Imagine if you could only use one sense at a time, and couldn’t match anything you heard to anything you saw. That’s AI today, and part of the reason why we’re so far from creating an algorithm that can learn like a huma n. But two new papers from MIT and Google explain first steps for making AI see, hear, and read in a holistic way – an approach that could upend how we teach our machines about the world.
And here is Scientific American (our emphasis):
The amount of data we produce doubles every year. In other words: in 2016 we produced as much data as in the entire history of humankind through 2015. Every minute we produce hundreds of thousands of Google searches and Facebook posts. These contain information that reveals how we think and feel. Soon, the things around us, possibly even our clothing, also will be connected with the Internet. It is estimated that in 10 years time there will be 150 billion networked measuring sensors, 20 times more than people on Earth. Then, the amount of data will double every 12 hours. Many companies are already trying to turn this Big Data into Big Money. Everything will become intelligent; soon we will not only have smart phones, but also smart homes, smart factories and smart cities.
And it continues by asking a highly relevant question:
Should we also expect these developments to result in smart nations and a smarter planet?
And an ominous prediction:
in the coming 10 to 20 years around half of today’s jobs will be threatened by algorithms. 40% of today’s top 500 companies will have vanished in a decade.
It all depends on how we use these powers and to whom we allow control.
The EU’s antitrust case against Google, while it might have merit on its own, is basically a case against the Google of 2010. This isn’t surprising, as the case was 7 years in the making.
However, the company has evolved, and so has technology. We think antitrust policy should be thoroughly overhauled for the digital age. We’re now fighting 21st century market power with essentially a 19th century approach.
More especially, developments in AI are accelerating at a dizzying pace. The take-off happened only very recently, when ways to manage and store the massive amounts of da ta required and the processing powers to enable machine learning became available.
The implications are going to reshape society in a fundamental way, and at present, those few organizations that have access to the massive amounts of data required to drive this forward are shaping the agenda, potentially leading to enormous market powers and fostering a dependency of the rest of society.
At the minimum, this should give pause for thought. It could be that antitrust (or perhaps other policy areas, like data policies) will slow the likes of Google down or attempt to build in some safeguards that these massive powers won’t be abused. But for the moment, the likes of Google don’t seem to have much to fear from policy, despite the EU antitrust case which is almost an anachronism.
Disclosure: I/we have no positions in any stocks mentioned, but may initiate a long position in GOOG over the next 72 hours.
I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.
About this article:ExpandAuthor payment: $35 + $0.01/page view. Authors of PRO articles receive a minimum guaranteed payment of $150-500.Tagged: Investing Ideas, Long Ideas, Technology, Internet Information ProvidersWant to share your opinion on this article? Add a comment.Disagree with this article? Submit your own.To report a factual error in this article, click hereFollow Shareholders Unite and get email alerts