About Axel-Springer’s CEO letter to Google – and upcoming UE antitrust decisions on Google

[switch to italian]

First of all, let me warn you that this post is quite long, but I hope you will find it substantial and interesting. 

Mathias Döpfner, is  Axel Springer‘s CEO. Axel Springer is one of the major European publishing groups and has embraced digital with significant success. 42% of their revenues (€2.8bn,£2.3Bn, $3.9Bn) derive from international activities; traditional brands account for only 54% of revenues . 65% of their EBITDA comes from digital, up from 27% two years ago.

He wrote this harsh letter (pdf) to Google’s chairman Eric Schmidt. I invite you to read it.

I’ll quote some parts, criticize some elements of Döpfner’s letter and then I’ll make a prediction

 

Two months ago, reagarding the 2010 EU antitrust case against Google, two months ago, EU commissioner Almunia released this statement. It looks like the proposed solution to ensure that Google’s competitors gain more visibility for their advertising and sites on the search engine is to add a new area for sponsored results. Something Döpfner comments as follows:

The Commission is seriously proposing that the infrastructure – dominating search engine Google be allowed to continue to discriminate against its competitors in the placement of search results critical to success. As “ compensation”, however, a new advertising window will be set up at the beginning of the search list, in which those companies who are discriminated against will be able to buy a place on the list. This is not a compromise. This is an officially EU – sanctioned introduction of the business model that in less honorable circles is referred to as protection money – i.e. if you don’t want me to kill you, you have to pay me

 

Then he talks about privacy 

There is a quote from you in this context that concerns me. In 2009 you said: “ If you have something that you don’t want anyone to know, maybe you shouldn’t be doing it in the first place.” The only sentence that is even more worrying comes from Mark Zuckerberg when he was on the podium of a conference with you and I in the audience. Someone asked what Facebook thinks of the storage of data and the protection of privacy. And Zuckerberg said: “I don’t understand your question. If you have nothing to hide you have nothing to fear.” Ever since then I have thought about this sentence again and again. I find it terrible. I know that it was certainly not meant that way. Behind this statement there is a state of mind and an image of humanity that is typically cultivated in totalitarian regimes – not in liberal societies. Such a statement could also have come from the head of East Germany’s Stasi or other secret police in service of a dictatorship. The essence of freedom is precisely the fact that I am not obliged to disclose everything that I am doing, that I have a right to confidentiality and, yes, even to secrets ; that I am able to determine for myself what I wish to disclose about myself.

I wrote in  “eu privacy officers against Google” (italian, 2012)

…this is not about being “good” or “evil”. It’s about a different hierarchy of values. On some topics, our positions are very different….

I understand the US point of view and, to some extent, I admire it. But they’ve lived through neither Gestapo nor  Stasi. We have different cultural roots.

In the Charter of Fundamental Rights of the European Union, the very first principle is “Article 1 – Human dignity: Human dignity is inviolable. It must be respected and protected.” The US Constitution’s first amendment deals with free exercise of religion, freedom of speech/press/assembly. [note: I’m not implying they are equivalent documents, just a proxy indication of attention].

 

Prof. Rodotà once said :

We must be wary of the argument of who says that an honest citizen has nothing to fear from the knowledge of the information concerning him. The “glass man” is a totalitarian metaphor, because it is the basis for the State’s claim to know everything, even the most intimate aspects of the lives of citizens, transforming into “suspect” anyone who asks for protection of private life.

I would not like to have persons who think like Schmidt or Zuckerberg. I could not confide my secrets to them.

Suppose you win a lottery and have a young child. maybe you wouldn’t like that everyone knows you won and where you livesuppose you are covertly collaborating with justice
Suppose you are hiding from a violent husband
Suppose you have some probability to have a disease that can discriminate you for a job
Suppose you are a famous person who would receive too many calls and wants to sleep
Suppose some practices are blamed by the community where you live and maybe punished with exclusion
Suppose your son is too smart and would be rejected by the group

Etc etc etc.

Statistical distributions in nature are not uniform; for each measured phenomenon there’s someone who falls outside the “normal” range and lack of privacy can turn into exclusion, discrimination or worse.

Schmidt’s argument is captious because it implies a “normality” while the protection is needed precisely for those – by choice or by chance – who fall outside this “normality”.

I’d like to know mr. Schimdt’s home address and phone number. but, maybe, based on his own argument, he doesn’t use a telephone and has not a fixed dwelling.

No doubt Mr. Schmidt knows my address and phone number. 

 

And this is a key point: we’re not all, everyone of us,  on the same level playing field.

And it’s not only Schmidt, but Page as well:

He dreams of a place without data – protection laws and without democratic accountability. „There’s many, many exciting and important things you could do that you just can’t do because they’re illegal“, Page said back in 2013, continuing „ …we should have some safe places where we can try out some new things and figure out what is the effect on society, what’s the effect on people, without having to deploy kind of into the normal world.“

We would never try a substance without an experimentation protocol and strict controls, which lead to trial on humans only at the very last stage, after innumerable verifications (and of course this limits “innovation” for the  pharma industries and increases their costs.)

But, on Page’s opinion, we should try innovative things on parts of society.

Many centuries ago the importance of communication and its possible damages were well understood. Think of Old Testament’s Book of Sirach, chap. 28, v. 18:  “Many have fallen by the edge of the sword: but not so many as have fallen by the tongue.”

I wonder if he’d try these exciting things on his children (even the name of one of them is not known to public)

But we’re not all on the same level playing field

 

 

Döpfner says Google could demonstrate some good will…

By not saving IP addresses, automatically deleting cookies after each session , and only saving customer behavior when specifically requested to do so by customers. And by explaining and demonstrating what it intends to do with its floating group headquarters and development labs.

The imprint of Google on our privacy has grown over time

Their first privacy policies (june 1999) stated

Upon your first visit to Google, Google sends a “cookie” to your computer. A cookie is a file that identifies you as a unique user. It can also store personal preferences and user data. A cookie can tell us, “This is the same individual who visited Google two days ago” but it cannot tell us, “This person is Joe Smith” or even, “This person lives in the United States.”
Google uses cookies to track user trends and patterns, in order to better understand our user base and to improve the quality of our service. Google may also choose to use cookies to store user preferences.

That was it, up to 2001 at least, when they changed to include retention of personal data provided by users

In 2004 they started an account-based strategy, but they were divided service by service

Then in 2012 they made their big leap and unified all accounts. Back then I wrote (italian)

  1. three little points:It doesn’t say for how long they keep it (and on telephone/SMS traffic [not yet applicable in italy], I’m not able to judge whether they comply to regulations on data retention or not)
  2. The consents’ limitations to different companies within the Google group, provided for different uses, specific to each service, are now ignored because all data is now aggregated (but someone who gave consent to Picasa for pictures did not provide consent to Orkut (I consented for A, now you take all the data and aggregate with B))
  3. They don’t allow for deletion. If you “delete” the informations, they may keep them : “unless we have to keep that information for legitimate business” (and they don’t even get deleted from archived versions)

Regarding the third point, I wrote recently about NSA’s data collection

Now, we all love Obama very much and we trust him completely when he guarantees us

But what if, for example following some unforseeble crisis, in 15 years a dictator comes to power ?

And what if in 30 years (who knows why? because of a war?) something happens that determines the conditions that my daughters will be discriminated because today I’m writing that in my opinion this great country I love, the US, is making a mistake ?

Because no Hebrew, before the start of WWII, could have imagined what would have happened just few years later.

And no Ukrainian could have imagined one year ago what is happening today.

 

 

Point number 2., in my opinon, is thorny

Let’s imagine google buys ACME, a medical testing company

Would I have given ACME consent to retain my data, if I knew it would end up in Google’s hands ? Or would I have rather preferred a company who promised me to anonymize my data before selling their business unit ?

Would I give Waze my data about my movements had I known that later Google would have integrated it with the rest of my data ?

IMHO, privacy authorities should express themselves on non-integrability of data deriving from acquisitions, mandating that the data should be given back to users and then deleted by companies. (I’d even do it for acquisitions already made; don’t think there’s a term…)

 

As for a solution to the competition problem Döpfner says that…

But Google could – for its own long – term benefit – set a good example. The company could create transparency, not only by providing search results according to clear quantitative criteria, but also by disclosing all the changes to algorithms.

But the point is that search neutrality does not exist. I wrote (italian) some years ago:

to me, “search neutrality” is an oximorom

a search engine task is selecting and presenting data in an order defined in an algorithm

Would the algorithm be public, online content would be all mathematically engineered to maximize the chance of ending up on the first page, something which is clearly not possible. The system would annihilate itself.

However, given their absolute dominance in search market, if Google’s algorithm willfully favors one their own services over the competitors, we would clearly have an antitrust problem.

But, how can we know it ? How can we know if one of Google’s own services is artificially listed on the first page because of an unfair exploitation of an advantage, if the algorithm cannot be published ?

That  manager of Dailymotion accusing Google of favoring youtube videos in search results is surely right!

However are those search results “right” ? Are they there because they’re based on the “right” criteria to give the user the best possible service ? Or is it an abuse by Google to their own advantage ?

 

In my opinion this is the important point: the question is undecidable with the present market structure.

Therefore, to get a level playing field, we cannot rely on google promise of “don’t be evil” alone. We need to change market structure.

3 years ago I wrote:

the one who is the gatekeeper, can also make the content?

IMHO,we’d be better off if he didn’t. I wrote a post about it a long time ago.

Even the financial market is regulated in a similar way (not very successfully, but imagine if it was not!) with Chinese walls: those who make studies (and give opinions on investment, on where to put your money) should not communicate with those who benefit from selling these investments.

Those representing interests of demand-side and those representing interests of offer-side should be separated

I don’t see any other solution to a separation between the search business and the rest.


 

And once again (2009):

… None of the investigations take aim at Google’s core advertising business. And unlike other technology giants in years past, Google has not been accused of anticompetitive tactics.

No one has aimed to the heart, so far.

It’s not true that AT&T was broken up because of anticompetitive behaviour

In 1984 AT&T broke up as a consequence of an antitrust case not because of a market abuse, but because it was too big, based on a principle of Sherman Act:

“If we will not endure a king as a political power we should not endure a king over the production, transportation, and sale of any of the necessaries of life,”

in 1982 it was considered that the telephone was one of the necessities of life.

But the investigations and carping from competitors and critics have Google fighting to dispel the notion that it has a lock on its market… “They describe where they are in a market under a kind of a fairy-tale spun gloss that doesn’t reflect their dominance of key sectors,” said Jeff Chester, executive director of the Center for Digital Democracy.
“Google search is an absolute must-have for every marketer in the world.”

Google tries to dispel the idea that it has a grip on its market. “they describe themselves as if they were in a fairy tale,” says the Director of the Center for digital democracy. “Google is mandatory for every marketer in the world”.

 … some experts say that the steady stream of headlines about antitrust investigations could tarnish Google’s image with consumers, who by and large still view the company, and its growing list of free and innovative online services, positively.

Why free? In order to colonize.

This one is from 2009 as well:

one thing we can be sure of: Google thinks a lot about the future of online monetization

And as long as there’s no online tool for interoperable micropayments or an antitrust intervention, they can sleep tight

BTW, as long as there is no way to make interoperable online micropayments, advertising is the only way to massive online monetization and certainly it’s not in Google’s interest for this situation to change. In mobile phones, as soon as the Apple app store became a clear monetization platform, an alternative to advertising, Google launched android and conquered the micropayments mechanism used by most people in the world (even here there is an antitrust issue, but we’ll talk about it in the future.)

Google has a public image of open and free or – at least -flat. As a matter of fact, they have very powerful user-lockin elements to ensure they keep their audience, and a very sophisticated billing system, absolutely non-flat, to secure their revenues in a strongly dominant position, with high margins (way above normal capital remuneration in competitive markets.)

Being a monopoly is not a crime. According to some doctrines, unless it doesn’t abuse its market power, it can have a very positive outcome.

We cannot consider an abuse the fact that they exploit their monopolistic revenues to control key segments of innovation in markets that are adjacent to the Search market. They offer users free chocolate, often of very good quality.

When asked about marketing Wave during the launch Q&A, the Google reps said “We really haven’t thought about that too much.” What about advertising? “We haven’t thought about that yet.” What about competition? “It’s not something we really thought a lot about.” So what have the Googlers thought about?

They’re right. there’s no need to think about it. They shouldn’t. Everyone else would budget on the dollar, but not Google, because of the reason above. Furthermore, filling a space helps avoid that something unpredictable gets created, something which could potentially harm Google interests in the long terms. Something that could happen with Facebook.

 

And it’s not only that.

Why did Google, over the years, move from “recognizing that it’s the same user” to “having a unique user profile across all services” ? (even those with a very low contribution margin, even those that are a pure cost).

I think the main reason is that a “vanilla search engine” has a problem, competition (other search engines) are just a click away.

To solve this problem, you need to introduce reasons fort the user to come back frequently to your property, to type the address of your domain, to integrate everything. And here’s the mail and the shared documents. To foster the network effect, you create an aquarium in which the user can swim, but where you control the source of oxygen.

“Google’s system” masters the exploitation of the only market where it is an absolute worldwide monopolist to colonize adjacent markets.

I wonder why Zoho has never started an antitrust case in Europe. (*)
 

 

Regulators (not only European) have historically put a great deal of attention on telecom operators and largely ignored Over The Top (OTT) operators (and the idea that OTT should pay telcos  is a nonsense, as I wrote elsewhere). Perhaps it is due to the speed of the internet market and the poor comprehension of the business, particularly in Italy, where it is considered with great superficiality by nearly all Italian leaders who don’t realize the extension and the relevance for all the economy.

 

(this one is from 2011)

let’s think about it…

  • One of the objectives of telco regulators is to accelerate number portability up to enabling it on the same day: same day portability. the objective is to enable the maximum possible level of “migrability” from one provider to another
  • Core elements of internet companies’ strategies are user lock-ins and network effects.

it seems obvious that there’s a regulatory asymmetry

 

Maybe today, with this letter from Döpfner, we are at a tipping point.

It happens at a moment when we have to change our EU antritrust Commissioner and a long awaited decision about Google is pending.

I don’t believe Commissioner Almunia will decide before the new Commission, but rather think that as soon as the new Commissioner gets appointed, he will face this topic which is very thorny but is fundamental for our future, with all the EU industry that, in the meanwhile, will likely sustain the pro-competitive side.

We’ll see if she will have the same determination Mr. Monti had back in 2007 re. Microsoft’s antitrust case.

 

 

P.S. find below some updates re. some emails I received. 


(*) Update: one of the effects of the information economy rules is that building a system costs a lot, then the incremental cost per user is virtually nil. If, thanks to your financial power at the time of the investment, you can consider the investment itself as a sunk cost, you can afford to price advertising completely by the customer with an auction mechanism without base price! (as variable costs incurred are nil)

This implies that the price-setting power for advertising moves from publishers (who have to pay variable costs and hence had a trading basis) to the advertiser. Moves from being tied to production costs to be tied exclusively to the advertiser’s margin. Something that has a massive (negative) lever effect on systems that include atoms and not only bits, dragging down revenues throughout the advertising industry that is based on (even for a small part) variable costs.

 

(**) Update 2 : to clarify about shared documents : the architecture of shared documents in Google requires all users to be registered users of Google and working simultaneously connected to the cloud server of Google.

It ‘s like the “doc” format , but on steroids : if tomorrow I have to co- write up a document, on which platform will I do it ?

On Google, which has extended its dominance from search to mail to documents, and where everyone is registered, or on another platform where very few people are already registered? And once the document is created, I’m sure all of its contributors will return there, day and again.

 

It is not the only possible architecture!

(2010)

I love the Internet, the one of protocols, of rough consensus and running code, not the one of centralized or proprietary services .

Suppose… the “email” gets invented now : no one uses email. mailbook.com starts to offer email . How do they do it? Open and interoperable? No; they’d (legitimately) say “everybody comes to my server, ’cause I offer the e-mail.”

The Internet could not have been born outside universities & research: the private industries seek to maximize their own (short term) interests, to autonomously achieve the critical mass because once you’ve got it , you’ve conquered the world.

On the opposite end, universities built an email ecosystem that was open and interoperable and all together helped the system reach a critical mass.

Of course, in the telco industry it is not so because the rules require interoperability, because for decades we have understood that  interoperability and plurality of the offer is of value to society.

Is it possible to conceive a “doc” system that would be completely interoperable and not resident in the servers of just one organization? The answer is yes.

And Google was building it. It was called Google Wave: a system based on a protocol and a federation of interoperable servers, each with its users who could collaborate across all networks to cooperativley write documents.

Such a system would not have generated lock-in. As a matter of fact, Google stopped the project.