- The Leicester City fairytale - May 15, 2016
- Off the Charts interview: former board member Niels van Iwaarden - April 9, 2016
- How will the merger between Vodafone and Ziggo affect our telecom costs? - March 4, 2016
- Big data will change the product market - January 30, 2016
- 50 Porsches and Mark Rutte: the Economics Dinner in a nutshell - December 10, 2015
It has been a tough week and you have therefore decided to reward yourself with a present. You visit your favorite website and order that fresh pair of new sneakers or brand new smartphone that you have been looking at for months now. You quickly walk through the ordering process and happily click the ‘confirm’ button as the final step to complete the transaction. Within minutes later the doorbell rings, and yes indeed if you open the door the just ordered package has arrived. Sounds fairly impossible right? It is not according to the American e-commerce company Amazon, who was granted a patent on a ‘method and system for anticipatory package shipping’ in the end of 2013.
Amazon claims that they can predict that you will buy that new pair of sneakers in the future, and therewith already ship them before you have even actually bought these sneakers at their website. This claim is a consequence of the constantly developing Big Data Era that we are in at the moment, in which privacy is a utopia and all your personal data has become freely accessible to several big companies. Since the rise of the digital revolution there are, as the Wall Street Journal calls it, ‘oceans of information’ available. A striking example of this is Acxiom, a company you probably have never heard of but claims to have a database that contains information about more than 500 million consumers (approximately 7% of the total world population!).
The patent Amazon has gained heavily trusts on their own database, a database that has kept growing ever since their foundation in 1994. By analyzing several business variables such as historical buying patterns, preferences expressed explicitly via surveys/questionnaires, demographic data, browsing habits, and wish-lists Amazon could already route a package to a certain geographical zone before it is actually being ordered, according to TechCrunch. The patent also entails an even more speculative example, in which ‘a package may be more or less continually in transit to a given geographical area or delivery address after being tendered to a common carrier for speculative shipment’. So your new pair of sneakers might have been in transit and possession of a courier for several weeks before it is actually delivered at your home address. This fascinating example is one of the possibilities of the Big Data Era and was the start of a recent guest lecture by Sebastian Dengler at Tilburg University, who held a talk about his research on the economics of privacy. Privacy issues are an increasing concern with respect to the use of big data. As I was eager to hear a little more about this, I asked him to shed some more light on this topic.
Within what time do you expect Amazon to be able to use this patent in their daily business?
Dengler: “The patent has been registered in August 2012 and published in December 2013. So the system itself might already be in use as the blog Mashable explains: ‘Anticipatory shipping may not be as sexy as delivery drones, but it’s a lot more practical. Plus, for all we know, may already be part of Amazon’s day-to-day business practices.”
The increasing existence of Big Data is the key element that makes this patent possible. But the awareness for privacy issues has increased among consumers over the years. Would it be possible that these privacy issues make the patent unusable in the near future?
Dengler: “This particular patent is unlikely to be affected as it is more likely to be used for some kind of ‘speculative shipping’. This might be informed by somewhat aggregate data analysis and results in filling in the precise address during the shipping process. Hence, the patent itself does not necessarily give rise to privacy issues in a narrow sense. However, if people would radically reduce their data footprint, this patent might become less useful (but so would pretty much the entire data-related business sector). Although shifts in privacy concerns can be observed, such a drastic shift is rather unlikely to occur as far as I can tell.“
” It illustrates the almost unbelievable precision that data-driven companies (are about to) have when making predictions. “
– Sebastian Dengler
The shortened delivery time may not be the only consequence of Big Data Era. If Amazon or Acxiom discovers that a consumer really needs that pair of sneakers today, they could easily raise the price of their product. Because of the increased time preference of the consumer he may be tempted to buy it, even at this increased price. It may empower producers to come closer to perfect price discrimination and therewith significantly decrease consumer surplus. Do you actually think that this will be ever feasible in the future?
Dengler: “This patent itself does not necessarily mean that Amazon can perfectly price discriminate, but rather shows how good predictions already are. It illustrates the almost unbelievable precision that data-driven companies (are about to) have when making predictions. Amazon, however, has already tried personalized pricing in the past but experienced negative publicity and hence tried to come up with less visible ways. You can find some more detailed information at this article.
This attempt was not strictly speaking “perfect price discrimination” and we are probably not exactly there yet either, but I guess the Andrew Odlyzko’s conclusion from 2003 still holds that “in the Internet environment, the incentives towards price discrimination and the ability to price discriminate will be growing.”
Hence, we might see further attempts to get closer and closer to perfect price discrimination. But I cannot confidently say that we will arrive there, especially not, if people become more aware of this possible future“.
Follow-up innovation is one of the effects of the current existing patent system. Existing patents are often at the base of further research. Have you already heard about any patents that follow up on this patent by Amazon? And if not do you expect to hear about any of these follow-up patents in the coming years?
Dengler: “To be honest, I have not thoroughly checked on further quotes of the patent and follow-up patents. However, combined with the recent advertisement of Amazon’s efforts to get drone delivery started up, it is easy to imagine that this is not the end of the line of innovation – and the aspect of price discrimination that I try to point out would be one of them. In my research I am currently analyzing a model where perfect price discrimination is already possible to see what we should expect when consumers try to anonymize themselves again. It turns out that the effects can crucially depend not only on the cost of anonymity, but also on cognitive constraints of consumers vis-à-vis a big data algorithm that outsmarts them. It is very exciting to work on such a topic that has the potential to (silently) affect our everyday life.”