Productive Fragmentation
Today: Recent readings and conversations inspired another round of reflections on fragmentation in Europe.
The Agenda 👇
Competition across borders
The concept of optimal fragmentation
Jared Diamond on optimal fragmentation
Competition at two levels
Let’s have a look at China
The importance of pluralism
An interesting case: the US
The global distribution of power
The right playbook for Europe
Further readings
1/ Competition across borders
I just wrote a column in Sifted about football, specifically about Spotify’s Daniel Ek push to buy Arsenal. There’s a uniquely European dimension in the project that I like a lot (and which reflects how much football matters on this continent, as opposed to in the US):
The nature of the consortium formed by Ek reveals the pan-European dimension of the effort. Ek is Swedish, and to my knowledge not especially connected to England — where Arsenal is based. As for his business partners, they all used to play for Arsenal, but two are French (Thierry Henry and Patrick Vieira) and the third, Dennis Bergkamp, is Dutch. What’s more European than a Swede partnering with personalities from France and the Netherlands to purchase a football club in North Islington, London?
And these days I’m watching Drive to Survive, the (excellent) Netflix series about Formula One, which is now in its third season. Like with football, I’m not that interested in Formula One in itself, but I like to learn about the economics and sociology of the sport. In this case, the thing that strikes me most is how diverse the sport is from a cultural perspective. It was invented a long time ago somewhere between the UK, Italy, and France (hence the continued use of the term Grand Prix). People literally come from all over the world, including emerging countries like Brazil. And they all speak English fluently, albeit with very strong accents that betray their national origins.
Here’s a compilation about Guenther Steiner, who’s the Team Principal at Haas (a tier-2 team), where you’ll see him interact with drivers Romain Grosjean (a French-speaking Swiss) and Kevin Magnussen (from Denmark). Steiner himself, a figure I like a lot, is from South Tyrol in Italy, which means he has two native languages: Italian and German (but not English, hence the thick accent). Have a watch:
What interests me here is the paradox. Europe is very fragmented, including by language, and most Europeans don’t cross borders much nor do they speak multiple languages. But there are some worlds, including football and Formula One, where it’s perfectly normal for people from different countries to work together, usually in English. This coming-together doesn’t in any way negate the plurality of countries: people on the outside, starting with fans, still check to see where the personalities (and constructors, in Formula One) are from before deciding what team they’ll support.
So you can have competition and come together to pool resources at the same time!
2/ The concept of optimal fragmentation
It reminded me of my conversation with Noah Smith on the Building Bridges podcast a few weeks ago. Noah mentioned the idea of “optimal fragmentation”, from a time when European countries were large enough to aspire to be world powers, yet small enough that they still had to compete against each other, pulling various levers in the process: colonial empires, industrial power, the military. The end of the 19th century and the first decades of the 20th century were a period of extraordinary progress in Europe thanks to the competition fostered by this particular context.
Here’s the relevant extract from the Noah Smith podcast:
There's this idea that Europe's period of outperformance and extreme success in the 19th and early 20th centuries came from optimal fragmentation, the fact that you had the UK and France and Germany and the smaller countries competing with each other (and Russia)—competing with each other in imperial conquest, in wars, but also just for nationalistic bragging rights.
It was like soccer before soccer. Because you had these smallish countries, each country had to really think about its global position, really get everyone on board for rapid modernization, really build up these strong bureaucracies and leverage local cultural advantages or local technological or resource advantages like Germany's coal, etc. And this competition was effective.
But then once you got better information technology and better transportation technology such that you were able to ride herd on a larger populace and bind together a larger country and create infrastructure across a larger country like the United States, like Russia, like China, at that point, when the technology changed, fragmentation may have gone from a benefit to a liability.
3/ Jared Diamond on optimal fragmentation
Needless to say I was very interested in this concept, so I googled the term “optimal fragmentation” and didn’t find much, except in this Jared Diamond interview with Strategy+Business, where they discussed both countries/continents and companies:
S+B: How does Microsoft — or any company that’s promoting internal competition among small groups — manage the potential chaos so that it doesn’t spin out of control?
DIAMOND: In Microsoft, each group has some critical mass. At the opposite extreme, if you’ve got so many groups that none has critical mass, chaos can result. That’s part of what has dragged down India. India has been hyper-fragmented.
My term for appropriate critical mass is the “optimal fragmentation principle.” The idea behind optimal fragmentation is that if you’re trying to organize some large group, there are perils at both extremes. There are perils of excessive unity — China — and perils of hyper-fragmentation, which would be the Indian subcontinent. There is an optimal intermediate fragmentation, which would be represented by Europe from the Middle Ages on.
But the idea of optimal fragmentation is only a good starting point for thinking about how best to organize groups. Then come the complexities. Is intermediate fragmentation necessarily good? Is competition necessarily good?
Competition in general tends to be good. But does this mean that any country ought to require its businesses to compete with each other and with the outside world? Think of the development of South Korea. After the Second World War, there was a stage where its companies simply could not be exposed to competition from the outside world because they would have gone under, and therefore protectionism made sense. Now, in South Korea, the disadvantages of protectionism outweigh the advantages. Think of Europe in 1914, when competition became destructive and led to the catastrophe of the First World War. That’s why I say that these management issues involve some simple principles, but applying them is complicated. Competition can be stimulatory or it can be utterly disastrous.
So within this simple idea of intermediate fragmentation, one has to recognize that there are different forms of competition, which can be stimulatory or non-stimulatory. Critical is the communication between the different entities, and critical also are the relations of the entities to the leader. So I would say that the idea of intermediate fragmentation is a good first chapter for a whole book.
I’d note that Jared Diamond is the very same author I quoted in my Will Fragmentation Doom Europe to Another Lost Decade? (January 2020):
In Guns, Germs & Steel, Jared Diamond explains why fragmentation once helped Europe rise as the main Eurasian power at the expense of China, which was much more monolithic than fragmented Europe and thus depended on the (occasionally stupid) decisions of a single sovereign:
Christopher Columbus, an Italian by birth, switched his allegiance to the duke of Anjou in France, then to the king of Portugal. When the latter refused his request for ships in which to explore westward, Columbus turned to the duke of Medina-Sedonia, who also refused, then to the count of Medina-Celi, who did likewise, and finally to the king and queen of Spain, who denied Columbus’s first request but eventually granted his renewed appeal. Had Europe been united under any one of the first three rulers, its colonization of the Americas might have been stillborn.
4/ Competition at two levels
There’s something else: the idea that nation states and capitalists need and support each other as they compete in their respective dimensions—nation states on the world stage, capitalists in business.
As you remember from my discussion of Fernand Braudel’s history of capitalism, capitalists insert capital into their value chain so as to generate increasing returns to scale and escape competition. (Peter Thiel coined the formula that “competition is for losers”, but it’s really all Braudel). Then their scaling up pits them against capitalists from other countries, at which point enjoying the support of your home state becomes critical if you want to survive.
I wrote as much in Capitalism and the Future of Nation States (December 2019):
Every state has an interest in supporting local capitalists with access to a supportive domestic market and various forms of industrial policy. The goal is to have capitalists create and capture even more value that will then be realized locally. In exchange for this support and “protection”, capitalists have to contribute to local development through consumption, investment, and redistribution. The state contributes to capitalists succeeding; capitalists, in turn, contribute to lifting up the local “material life” and “market economy”, thus strengthening the state.
It might very well be that Europe’s fragmentation is not optimal anymore now that we’ve shifted from the Fordist Age of the 20th century to the Entrepreneurial Age of the 21st. Let me explain.
In the past age of the automobile and mass production, the ultimate capitalist game was played in manufacturing, most specifically in the car industry. And for a long time, in part due to trade barriers but also because of diminishing returns beyond a certain scale, it was difficult for any car manufacturer to expand beyond its domestic market. Therefore many countries had local car manufacturers, each with a stronghold on their domestic market and an ability to grow up to a certain scale.
Fast forward to today, the age of computing and networks, and there are many tech businesses that don’t encounter diminishing returns before they reach a much larger scale. Again, it’s partly due to the rules in international trade (the economy is more globalized than it was after World War II). But it’s also due to the nature of today’s tech companies, which can sustain increasing returns much longer thanks to the unique characteristics of the technology that is computing and networks.
That means global markets are much more concentrated, which means that not every country that had a successful car manufacturing company can have their national champion anymore, therefore the level of fragmentation might not be optimal anymore: we can’t grow a Google or a Facebook in each European country like we once grew car manufacturers in Germany, France, Italy, Britain, Sweden, and Spain.
Only one or two European countries can still compete, and therefore it’s not all European countries competing against each other—rather it’s a few of them potentially competing against the US, China, and any other countries that have grown today’s equivalent of a car manufacturer.
5/ Let’s have a look at China
Yet I still think that there should be a playbook to make the most of Europe’s current level of fragmentation, even in an Entrepreneurial Age that seems to reward companies operating on a very large market. One obvious case is that of China, which I discussed in the Building Bridges podcast with Lillian Li, published yesterday.
Let me quote Lillian about the fact that China is far from being as centralized as you’d think:
Not that I am an expert on Chinese industrial policies, so I’ll just talk about what I've seen. What I think is quite interesting is the fact that, as I mentioned before, China is incredibly decentralized. And what you have is cities competing against each other. A city like Shanghai doesn’t see itself as doing things on behalf of China. Rather Shanghai is thinking it needs to compete against Beijing and Shenzhen and Guangzhou for the best talents.
Also, the Shanghai government has a lot of autonomy to be able to carry these things out. They are able to think more holistically about what it means in terms of providing attractive offers for talent. So now there are policies designed to help local companies attract and retain Chinese workers with special advanced degrees and specialization in certain subjects. The local government provides these companies with benefits, subsidies, the ability to buy housing to make sure that they're attracting the best people to a given location.
Companies are also able to have tax breaks and subsidies, especially in key areas that are deemed to be crucial to the nation's science and technology progress. And that varies and changes, with quite a lot of flexibility. For instance, livestream e-commerce has been very big in China in the past few years. This particular business is not in line with the national directives, but you still see cities supporting that line of business by providing subsidies and a kind of ease of doing business for people who're working as livestreamers.
Interestingly, Noah Smith also said as much in a recent podcast discussion with economist Brad Delong:
There’s this short online book by Barry Naughton called Growing Out of the Plan. Basically, it says that China didn’t have a national industrial policy until very recently, it had local industrial policies. It allowed local governments a wide latitude to set these industrial policies. Only after 2010 have they really shifted to having a national industrial policy.
I find it interesting to observe that, yes, China is a very large market and that scale explains its success at growing very successful tech companies. On the other hand, it should be pointed out that part of this success is due to the ability of China to foster fragmentation and competition on the inside.
6/ The importance of pluralism
An article by historian Walter Scheidel, published in Aeon, made the rounds earlier this week. It’s about the fall of the Roman Empire and how it gave birth to a fragmented, prosperous Europe:
Historians, economists and political scientists have long argued about the causes of these transformative developments... Almost without fail, all [the] different arguments have one thing in common. They’re deeply rooted in the fact that, after Rome fell, Europe was intensely fragmented, both between and within different countries. Pluralism is the common denominator.
If you side with those scholars who believe that political and economic institutions were the basis for modernising development, western Europe is the place for you. In an environment where bargaining trumped despotism and exit options were plentiful, rulers had more to gain from protecting entrepreneurs and capitalists than from fleecing them. Size also mattered: only in moderately sized countries could commercial interests hope to hold their own against aristocratic landlords. Smaller polities enjoyed greater capacity for inclusion, not least by means of parliamentary deliberations. The better medieval legacies of pluralism survived, the more such states developed in close engagement with organised representatives of civil society. International competition rewarded cohesion, mobilisation and innovation. The more governments expected from their citizens, the more they had to offer in return. State power, civic rights and economic progress advanced together.
It reveals another dimension of the system whereby fragmentation creates prosperity. It’s not only that you need several geographical entities competing against each other (like European countries in the 19th century or Chinese cities these days). It’s also that within these entities you need pluralistic institutions—in other words, you need to turn pluralism on the inside into a weapon for better competing on the outside. This leads to a collection of random/interesting observations:
China has been doing well in the Entrepreneurial Age because, while the Chinese Communist Party doesn’t tolerate challenges to its authority, it provides local governments free rein to implement bold industrial policies at the local level and finds it perfectly acceptable for entrepreneurs to wage brutal wars against one another on a given market. (Lillian mentioned the hundreds of startups that once were competing on the food delivery market, a war from which Meituan emerged as the victor.)
It remains to be seen if the tightening implemented under Xi Jinping over the recent years will lead to China losing ground in the global race for success and prosperity in the Entrepreneurial Age. Noah Smith is pessimistic, as recently explained in China's government is starting to screw up.
Indeed, some countries are trying hard to compete, but they’re still failing or lagging behind for lack of pluralism. I find this especially true in my home country of France, as I explained in France, As Revealed by its Elite (July 2020):
All paths lead to Paris. If you manage to arrive there (‘monter à Paris’), you mingle with the rest of the elite and enjoy a lifetime of rents that come from being part of that small world. There is an internal paradox as well: the French elite is diverse from a background perspective, but the elite life is concentrated in some specific neighborhoods in Paris. Once you arrive there, there’s no real incentive to move elsewhere, either functionally (by, say, leaving that world to found a startup) or geographically (moving back to the province or emigrating, as I did).
There are countries that are less than pluralist but which have decided building startups and tech companies is their new challenge. Saudi Arabia comes to mind, and they seem to have realized you can’t become competitive as a nation in the Entrepreneurial Age if you don’t provide for some pluralism on the inside. Here’s a recent chronology of reforms there, including ending a prohibition on cinemas, lifting a ban on women driving cars, and allowing music to be played in restaurants—in Reuters: Factbox: Saudi Arabia's main economic and social reforms, investor concerns.
And then there’s the case of India which until last year seemed to be riding high as a tech powerhouse in the making, but that seems to be going backward when it comes to encouraging pluralism on the inside. Have a look at my What's Happening in India? (June 2020), as well as articles such as the BBC’s Soutik Biswas’s 'Electoral autocracy': The downgrading of India's democracy.
7/ An interesting case: the US
Here’s another question raised by Noah Smith in the above-mentioned discussion with Brad Delong: how come the US was so successful despite being so large? Historically, very large countries have proven so difficult to govern that they’ve either been ruled in a centralized and rather tyrannical way, like Imperial China, or they have simply collapsed, like the Roman Empire.
I think there are two things to keep in mind to understand this American paradox. One is that fragmentation and pluralism are deeply entrenched in the country’s institutions as well as in its history and culture. Various instances of this are the federal system and Louis Brandeis’s idea of states acting as “laboratories of democracy”; the distribution of power between the various branches of government; the importance of the First Amendment to the Constitution and the freedom of speech it provides to everyone; the legacy of immigration and the effective coexistence of very different cultures; Bruno Maçães’s “escapism” and the new frontiers it constantly provides for American dreamers and wanderers.
On the other hand, being so pluralistic while being so large is a luxury only the US can afford (remember Scheidel’s “Smaller polities enjoyed greater capacity for inclusion”) because it’s been so blessed from the perspective of geography and natural resources. Here’s Peter Zeihan:
The United States isn’t very good at national governance. When geography takes care of all the big issues, there is little need for a large, overarching, competent, national government. And it shows. The U.S. isn’t Germany or Korea, countries that live in geographic pressure cookers and so governance has to be top notch to ensure survival. This isn’t Russia which is paranoid for good reason and so must excel and intelligence operations. This isn’t Brazil where the terrain and climate are hostile to development and so excellence at infrastructure policy is essential. America’s lack of federal competence means that when there is a crisis it all comes down to the personality, skill and contacts of the person at the top. America’s initial reaction to the coronavirus isn’t its first failure of presidential leadership. But America’s sublime geography means the country will survive this failure to have others down the road.
Of course this extreme version of pluralism at a large scale creates problems (Tyler Cowen discussed them in our conversation a few weeks ago), but in the end it’s a unique formula for success that explains why America has been so consistently good at racing ahead in each great surge of development since the age of steel and heavy engineering at the end of the 19th century.
8/ The consequences on the distribution of power
Here’s the thing with tech companies: sooner or later, power down the stream translates into power up the stream. The first stage is that tech giants emerge on consumer markets thanks to their unique ability to harness the power of their multitude of users. In turn, the existence of downstream giants gives birth to economic powerhouses up the stream. There are two manners in which this story plays out:
The downstream giant diversifies up the value chain and grows another line of upstream business that complements its consumer business, before eventually becoming a standalone, profit-making business. That is the story of how Amazon (very much a consumer company) built Amazon Web Services (the global leader in cloud computing). Or,
Relationships are built within a given ecosystem (often with the blessing of the government) between the new dominant players down the stream and emerging players up the stream, with the two groups complementing each other and together cementing their common ecosystem’s success and ability to scale. This is what happened with Apple playing a role in the rise of Qualcomm, for instance (Qualcomm beating the incumbent Intel due to their superior ability to build chips for smartphones as opposed to chips for traditional computers).
I spotted this idea in a very interesting discussion between China expert Jordan Schneider and analyst Emily de la Bruyere, which I chronicled in my Industrial Policy: China Gets It, We Don't:
In China’s case, there’s a high return on invested capital in their efforts in semiconductor research because a direction is provided by the Chinese tech giants down the stream. Investing in research with that clear direction (in this case, serving the strategic interests of Huawei, Tencent, Baidu, and Alibaba in fields such as 5G networks, cloud computing, and artificial intelligence) generates a much higher return on investment than funding research merely for the beauty of it.
I can’t help but think that the current global showdown that’s happening in the semiconductor industry is a result of a previous round of competition in which only some countries were able to grow large and successful consumer tech companies.
Clearly the US and China, both very large, integrated consumer markets, had an advantage on that one, and it’s telling that Europe, despite its excellence in basic research and engineering, is missing out on semiconductors just as it has missed out on consumer markets.
Have a look at these articles:
China Revs Up Grand Chip Ambitions to Counter U.S. Blacklistings (Bloomberg)
EU Seeks to Double Share of World Chip Market by 2030 in ‘Digital Sovereignty’ Drive (Daniel Michaels and Stu Woo, The Wall Street Journal)
Intel's Plan to Win Big in Chips Is Full of Big Risks (Tae Kim, Bloomberg)
Europe Is Trying to Reclaim Its Lost Chipmaking Glory (Natalia Drozdiak, Bloomberg)
(By the way, Emily de La Bruyere will be one of my next guests on the Building Bridges podcast.)
9/ The right playbook for Europe
This is a recurring theme of this newsletter, but I’ll never get enough of refining my idea because 1/ it matters, and 2/ it is still unclear what exactly Europe should do to catch up in the Entrepreneurial Age.
I believe one favorable trend is the fact that as software eats more of the world, the more tech founders enter industries that are more localized by nature, either because they rely on tangible assets or labor (as in food delivery) or because they are more regulated (as in financial services) (or both). These “default local” industries, to quote Andreessen Horowitz’s Anish Acharya, are ones in which startups start playing at a local level before they reach escape velocity and start expanding.
In my view, it makes it rather easy to determine the right playbook is in this context:
In any industry, fragmentation and pluralism are critical to foster the emergence of successful startups. Silicon Valley is a pure product of American pluralism (Balaji S. Srinivasan has been talking for years of “Silicon Valley’s ultimate exit”), and Europe being fragmented with generally pluralistic institutions makes it probable that similar ecosystems will emerge on the continent. Obviously, you can tell that not all cities are created equal: London, for instance, is doing well thanks to the English language and the local liberal tradition, while Paris is lagging behind in no small part due to the language and its position in elitist, backward-looking France where pluralism is absolutely not a given.
But it’s not enough to have a few thriving ecosystems able to give birth to successful companies. Once they reach escape velocity, you also need to provide these companies with the ability to expand at as large a scale as possible, which usually means crossing borders and finding that you’re set up to fall victim to European fragmentation. Again, I already made this argument in Will Fragmentation Doom Europe to Another Lost Decade? but it’s worth articulating it again in light of more recent findings. If fragmentation is an advantage when it comes to starting businesses but a disadvantage when it comes to scaling them up, then it’s critical that Europe as a whole (national governments, the EU, entrepreneurs themselves) to work very hard on hedging against the adverse consequences of scaling up.
But can we have our cake and eat it, too—as in, can we enjoy the advantages of the fragmentation without suffering its disadvantages? Can we make fragmentation more productive, like in football or Formula One?
I think so, because it’s been done in financial services, as I once wrote in What European startups can learn from the success of the fintechs (in Sifted—pre-Brexit):
Having London as their base would not be enough for fintech entrepreneurs to succeed. They also need a large market to scale up their business and, on that front, they’ve been able to count on another ally: the European Union. Indeed, financial services is one of the few sectors, like energy and railways, that truly benefits from the advantages of the single market. EU-enforced rules (in this case, the payment services directives) ensure that there’s a level-playing field, making it easier for startups to enter the market. In practical terms, this means that once they’ve taken off on the British market, it’s easier for all those UK-based fintech startups to expand on the continent.
I guess we’ll figure out soon if Europe can emulate this approach in every industry that’s not yet dominated by a US tech giant. In the meantime it’s incredibly important to get the conversation going.
10/ Further readings
That’s all, folks! No need to add a reading list since you can find it here: All About the Great Fragmentation
Obviously I’ll iterate on what I see as the very promising concept of productive fragmentation!
If you enjoyed this edition of European Straits, you should subscribe so as not to miss the next ones.
From Munich, Germany 🇩🇪
Nicolas
