Follow The Data

Does antitrust law need changing for the data age? The discussion has started, but it is clear that new business models are not fully understood.


What does antitrust mean in the age of data? Data is in many cases gathered for free and has huge value to those who can turn that data into information and make decisions based on it.

Amazon has become an example in retail because they see more sales data than any other retailer – not just what they sell, but what others choose to sell on their platform. Amazon gets to see all the data and the resellers only see their own private data. With this data, Amazon can potentially influence who wins and loses based on how they rank products and where they appear on the page.

Amazon can also choose to stock and resell items that are successful, thus taking the profit for themselves, rather than just a commission. They can choose to develop in-house brands for those that have the highest profit margin and add a little more to their total profit.

In a Planet Money podcast, it was argued that this was wrong because it could limit competition in the long run because nobody will now choose to break a new product into the market. The company that attempt to sell a new product has to pay to produce the product and then attempt to sell it. If it flops, the company is left holding product and they may end up taking a loss. Amazon bears no risk in this scenario. If they are successful, Amazon will reap a commission on the sale. Amazon can choose to start selling the product directly themselves, often purchasing from the same manufacturer. Ultimately, Amazon is the winner and thus the balance of risk has been transferred from the company that introduced the product initially to Amazon, which bears no risk.

I see this as a classic example of not having a competitive protection moat around a product. How many times do we see a product listed on Amazon from a dozen or more companies, the only difference being the printing on them, or different packaging? The profit margins here must have been squeezed to the limit to start with.

The same type of situation exists for many tech products. Some companies do not try to be the leader, but a fast follower. Allow someone else to break a new product into the market and, if successful, produce something that is slightly better or cheaper. Patent laws can provide a moat, but even then the time and effort to obtain the patent and defend it is often longer than the time to generate the next generation of a product, or for the company to close up shop and run with whatever profit they have managed to milk out of it. Patents are also more difficult in a pure commerce situation, and there is risk associated with the development of the competing product.

I am not a lawyer, and definitely not an antitrust expert, but it seems to me as if this is just business having to come to grips with new risk models based on information. Does anyone fully understand the implications today? I don’t think so. This is why the whole IoT world is getting so much attention. Companies have already demonstrated how they can obtain small amounts of data from a sensor array and turn that into tangible ways for a company to become more efficient, to lower costs and improve productivity.

IoT companies understand that it is the information that has value and they are fighting to be the owner of that information, sometimes even denying access to the owner of the installed system. They want to be the ones to be able to see and sell more opportunities to the same company later.

Personal data is a little different. The person receives benefit when a service is offered to them for free in exchange for data, but in most cases the person does not understand the value of that data or the ways in which it could be used. They have no control over how it is used once it has been given up. It amazes me how many people take quizzes or play games on Facebook where it is evidently clear that the only purpose of them is to skim data from the person, sometimes obtaining data that can be used to compromise accounts and passwords.

So, the question is: should anything be done about it? Should governments get involved and regulate it today? It would be crazy to do so before we fully understand it, but that may be too late. Damage may have already been done at that point and it may be twice as difficult to undo it. Or is this just the fear of the unknown?

I would be interested in hearing from people who have an opinion on this subject.

Leave a Reply

(Note: This name will be displayed publicly)