How Fast is the Future?

Peter Diamandis & Steven Kotler cherry-pick evidence to paint a Panglossian picture of accelerating innovation, but fail to conceal the gloomy shadows of Secular Stagnation.

I have been reading Peter Diamandis & Steven Kotler’s new book, “The Future is Faster Than You Think”. If you are already familiar with the authors’ previous works, it will come as little surprise that the latest volume in their “Exponential Mindset Trilogy” doubles down on the same arguments made in “Abundance” (2012) & “BOLD” (2015). Basically, that a convergence of accelerating technologies such as Artificial Intelligence (AI) & Augmented Reality (AR) will lead us to a Singular Utopia — much sooner than you would expect.

The book is certainly worth reading end contains numerous enlightening insights about emergent exponential technologies. But, to get that out of the way first, the book’s biggest weakness is that it does not seriously engage with the counterarguments of the Secular Stagnationists; that the future is not so fast.

After all it has been close to a decade since the publication of “Abundance” & for most of the middle-class in most Western countries that has been a decade of stagnation in wages & wealth. For many the authors’ arguments that flying cars are right around the corner & that we will soon be enjoying an abundance of low-cost housing, healthcare & education may thus ring a bit hollow, to put it mildly. For example, what exactly does it mean that “exponential technology is dematerializing, demonetizing, & democratizing nearly every aspect of real estate”? Three funky words starting with “de-“; that will probably go down well in a TED speech. But try telling that to a Silicon Valley employee struggling to pay the rent on his ludicrously expensive San Francisco hut.

The authors do at least acknowledge Peter Thiel’s complaint that: “We wanted flying cars, instead we got 140 characters”. But they interpret Thiel too literally, without taking him seriously. Their retort that flying cars are here already is too clever by half. It is true that Uber plans to launch their aerial ridesharing scheme in 2023 – Uber Elevate – with a long-term goal of achieving lower marginal costs per passenger mile for flying cars than for traditional cars. However, the Covid-19 pandemic has thrown renewed doubt at the feasibility of both those targets.

The book would have been more credible if Diamandis & Kotler had offered a more comprehensive macro rebuttal of the Secular Stagnation story, instead of cherry-picking individual companies & sectors with outsize productivity gains. It weakens the book that the authors are hesitant to confront the decelerating productivity growth in the American (& other Western) economies that has been extensively documented by Robert Gordon & others – most notably in his masterly “The Rise & Fall of American Growth”. The only thing this evasiveness serves is to stoke suspicion that the crowd around Singularity University is turning into a techno-utopian cult.

The human factor is a bit missing from Diamandis & Kotler’s narrative. Throughout the book one often gets the impression that people, far from having any inherent value or free will of their own, are reduced to a pathetically mortal pieces of meat, whose intelligence has been transcended by their Alexa home entertainment system & who can be rocketed around in Hyperloops at 760mph serving some higher Kurzweilian purpose. & if the authors get their way society will soon reach Longevity Escape Velocity – i.e. that for every year you live you will add more than one year of life expectancy. If this is what the future holds, most sane people will probably wish to die before they will have the chance to spend eternity in AI purgatory.

Despite the claims of an “exponential mindset” the authors occasionally fall prey to rather linear, & dare one say shallow thinking. Such as their very linear view of the “time saved” by googling the answer to a question instead of having to look it up in a lexicon or applying your own cognitive faculties. Even Eric Schmidt has recognized the risk that Google – & the Internet more broadly – making us stupid, by producing a tendency toward shallow thinking, at the expense of (for schoolchildren especially) developing the ability to think (& read) deeply about things. Diamandis & Kotler on the other hand, have no such qualms.

There is no question that a lot of interesting stuff is happening in Silicon Valley. But will Avatars ever match the impact of the invention of electricity or Blockchain that of running water? Gordon has demonstrated very thoroughly how hard it is to match the truly life-altering significance of the General Purpose Technologies & subsidiary inventions of the Second Industrial Revolution: the internal combustion engine, automobiles, aircraft, the telephone, radio, “modern” sewage systems, the washing machine & assembly lines. These inventions powered a century of miraculous growth from ~1870 to ~1970. Their impact was so transformative that the feat can hardly be repeated. Diamandis & Kotler wax lyrical about the notion of a Kurzweilian “Law of Accelerating Returns”. But the law of diminishing returns will not easily be turned upside down. Yes it may be true that an iPhone would have cost 110 mUSD in 1980, but how much better is really the iPhone 11 than the iPhone 10? & will the 12 be leaps & bounds better than the 11?

Even if one remains sceptical about Diamandis & Kotler’s grand thesis of accelerating change, their analyses of several industries are spot on. In the retail industry for instance, the changes are obvious to see & have been accelerated by the coronavirus pandemic. The transition from analog to digital has been sped up. Amazon is the big winner. Shopping malls will die. It will be the end of cashiers as automated checkout becomes the new norm. The transition from physical to online retail has profound consequences for the advertising industry as well, as evidenced by the fact that Google & Facebook have gobbled up a combined quarter of ALL global ad revenue – leaving Old Media as the big loser.

The Covid-19 Pandemic has been a Luther moment for the “Higher Education” racket. As lectures have moved from campus to cyberspace it seems to have finally dawned on students that they are indebting themselves for life only to pay exorbitant tuition fees for knowledge that is essentially available for free on the Internet. The authors are right that this relic from the past is no longer fit for purpose (if it ever was – note the disparaging judgments of Oxbridge from the likes of Adam Smith or Edward Gibbon as long as 250 years ago). Peter Thiel has rightly compared modern university diplomas to the indulgence letters of the medieval Catholic Church. But students’ obedient faith in orthodoxy has held up surprisingly long. With Covid – judging for example from reports suggesting that foreign students will not bother with elite educations at prestigious US or UK universities if classes will anyway be held online – the moment when students (& employers) see that the Professor is not wearing clothes may at long last have arrived.

AR, VR & Avatars are all exciting fields, but will they truly transform the lives of common people like the dishwasher did? Or perhaps more likely & more sinisterly; will these technologies provide avenues for resigned people to retreat from reality? A second chance to create in virtual worlds the life they could not have in the real world? At the very least this should bode well for the gaming industry.

Blockchain & Cryptocurrencies are perhaps the emerging technologies with the biggest potential to radically change the way the economy is ordered – by challenging the incumbency-biased hegemony of central banks & fiat currencies. In theory at least. Not that I believe it will happen in practice. The “Fintech” revolution held early promise but has by now been quite effectively co-opted by the legacy banking system, rendering the chances for radical change minimal.

Legacy banks have today become probably almost as big a break on entrepreneurial activity as government regulation. If blockchain, cryptocurrencies, Digital Autonomous Organisations (DAOs) & seasteading perhaps could be combined in a way that would enable people to conduct (legitimate) business across borders without being reliant on legacy financial institutions then it could unlock a new dimension. But for the moment the whole cryptosphere, although it has practical utility, for all intents & purposes remains a casino.

Speaking of money, Diamandis & Kotler claims that entrepreneurs today enjoy easier access to capital than ever before. While that may be true for some tech startups, it is not the everyday reality for most small businesses. Legacy banks have largely abandoned their traditional role of providing finance to new business creation, in favour of loans to big corporates & mortgages to the salaried classes. With the exception of a few much-publicized success stories, the concept of crowdfunding was dead on arrival. At any rate, cheap capital is no magic potion for entrepreneurship. If that had been the case, the Kamikaze monetary experiments of the Federal Reserve & other central banks should have unleashed a Wirtschaftswunder like the world had never seen before. It has not happened. That is because the stagnation we see in Western economies today is due to deeper problems on the supply side, which cannot be solved by funny monetary policies. Or to take a case in point, look at SoftBank’s 100bUSD Vision Fund, which was created with the aim of throwing billions of dollars at Unicorns to accelerate the future, & the best they could find was a Ponzi office sub-letting scheme? Building the future is hard & requires more vision than an abundance of Petrodollars from hallucinating desert princes.

The authors do pay lip service to concerns for rising technological unemployment – concerns that suddenly have become much more acute in the wake of the coronavirus pandemic, as many of the job losses triggered by Covid-19 are unlikely to be reversed. For example, the cashiers that have been rendered obsolete as shopping has migrated online. What are they going to do next? Diamandis & Kotler cite a McKinsey report (congratulations) claiming that the Internet has created 2,6 new jobs for every job it has destroyed. That claim may have been supported by the low pre-pandemic unemployment levels. But, as David Graeber has documented, there has been an awful lot of bullshit jobs out there. What the pandemic has brutally revealed is that society manage to cope pretty well without many of those jobs, (a consequence that can be both encouraging & disheartening, depending on your perspective).

Another trend that the authors are very keen on that is also being challenged by Covid is urbanisation. Are ever bigger cities with ever-taller towers with ever-more people crowded together in tiny closets still the answer to the future? It does not look very appealing to me.

As a matter of formality Diamandis & Kotler go through the list of threats to their thesis – from water woes to climate change in addition to the risk of technological unemployment. But they remain optimistic on all fronts. Politics do not seem to play much of a role in their future – except that: “Governments around the world, encouraged by Estonia’s example, are going digital”. That sounds great, but where is the evidence that digitalization has increased public sector productivity – other than in governments’ tax-collecting arms & citizen surveillance capabilities? When it comes to immigration Diamandis & Kotler still cling to, the rosy-eyed view that: “Migration is an innovation accelerant”, which to a European reader sounds dangerously naive. As if their example of Jews emigrating from Hitler’s Germany holds any relevance whatsoever for the migration flows of today.

All in all, the book is absolutely worth reading, as it is representative of how Silicon Valley’s most Panglossian minds envision the best of all possible futures. But while Diamandis & Kotler have their AR goggles on & are ready to hyperloop into the future, the rest of humankind will be joining them not so fast.

Will there be life after Google?

The Google system of the world is an evil empire that stifles entrepreneurial activity and is doomed to fail, before a decentralised Internet will rise from the ashes, argues George Gilder in his latest book.

life after google

On the face of it the main argument George Gilder is making in his latest book is preposterously bold: “The Google era is coming to an end because Google tries to cheat the constraints of economic scarcity and security by making its goods and services free. Google’s Free World is a way of brazenly defying the centrality of time in economics and reaching beyond the wallets of its customers directly to seize their time.”

Considering that Google generated revenues of 136,8 billion dollars and profits of 30,7 bUSD in 2018 and has a market capitalisation not far shy of a trillion (870 bUSD as of April 22), its decline and fall does not seem imminent, but a rather far-fetched prospect.

Google’s system of the world

Google is not only one of the most valuable (on some days the most valuable) companies in the world, but more than any other of the tech giants Google represents a system of the world, (borrowed from Neal Stephenson’s Baroque Cycle trilogy). A system of the world necessarily combines science and commerce, religion and philosophy, economics and epistemology.

Google’s theory of knowledge, its religion, is Big Data. Whereas good Christians may go to heaven “good googlers” are on a determinist path to a place called Singularity, where technology has transcended biology and artificial intelligence surpassed human intelligence.

Reaching Singularity rests on two conditions:

All data in the world can be compiled in a single “mind”.
Algorithms sufficiently comprehensive to analyse the data can be written.

As Gilder points out the Google theory of knowledge and mind are not mere abstract exercises. They are central to Google’s business model, which has progressed from “Search” to “Satisfy”. Google can already show considerable evidence that with enough data and processors the search engine can know better what will satisfy your longings than yourself. Notwithstanding that it in many cases makes us stupider: Just last weekend I was in Vienna and Google could tell me a restaurant I fancied was closed for lunch. We walked by to double-check and lo and behold open the restaurant was.

The fatal conceit of Google-Marxism

Gilder – an unreconstructed Reaganite free-marketeer – sees Google’s idea of a universal omniscient logic machine as deterministic and ultimately dictatorial. He accuses Google of repeating Karl Marx’s erroneous belief that we have reached the final stage of history. An immanentized escathon, in the words of William Buckley. Ironically both the technology Utopians and Dystopians share the belief/delusion of the coming all -powerful AI. Gilder thinks they are exaggerating the attainments of their own era, and sees the gospel espoused by the Google guys, Ray Kurzweil (since 2014 employed by Google) and the likes of Life 3.0 author Max Tegmark and Yuval Noah Hariri as hubristic Google-Marxism. Gilder accuses the AI champions of being blind to the realities of consciousness and of having forgotten Gödel’s lesson that all logical systems, such as AI, is incomplete and in need of an “oracle” (such as human consciousness outside the machine). He does not share Kurzweil’s position that when a machine is fully intelligent it will be recognised as conscious.

Indeed, there are eerie similarities between the Google theory of progress and the Soviet economic theories of central planning. The Soviet economist Leonid Kantorovich, who won the Sveriges Riksbank’s economics prize in 1975, the year between Friedrich Hayek and Milton Friedman, never shed the belief that advances in information technology would give Gosplan (close to perfect) knowledge and make central planning more efficient than market choices.

One foundational feature of the Google system is the Zero Price. Informed by Internet pioneer Stewart Brand’s slogan that “Information wants to be free”, Google has made (almost) all of its content and information free, in a digital version of the medieval Commons – i.e. public resources available to all. Private data are the mortal enemy of the Google system of the world.

In Gilder’s view the Zero price is the fatal flaw that dooms Google, just as Hayek predicted that the lack of price signals would doom the Soviet economy. Jeremy Rifkin heralds a Zero Marginal Cost Society, where prices for near every product and service will reach (close to) zero as every device and entity become connected in an Internet of Things and exponential network effects will unleash a Utopia of leisure and abundance – (a vision for the future that needless to say is much in vogue at the Googleplex).

Gilder does not buy into that prophesy. He sees “free” not only as a lie – as Apple CEO Tim Cook acerbically pointed out: “If the service is ‘free’ you are not the customer but the product” – but as a return to the barter system, a system so inefficient that it was left behind in the Stone Age: “Above all, you pay in time. Time is what money measures and represents – what remains scarce when all else becomes abundant in the “Zero Marginal Cost” economy. Money signals the real scarcities of the world concealed in the false infinities of free.” Furthermore, “free” entails a slippery avoidance of liabilities to real customers and no concern for security.

Disconcertingly for Google there are many signs that internet users have had enough of attention-grabbing ads and lack of data privacy.

Adoption of ad-blockers are skyrocketing to the degree that even Google itself has felt forced to launch their own “ad-blocker” – and it is the most attractive demographics for advertisers that are leading the trend. In the US 25 percent of internet users were blocking ads in 2016.
Internet users are simply developing “banner blindness”, becoming more or less immune to the bombardment of ads.
In the product-search category Google is rapidly losing ground to Amazon. Internet users still prefer Google for informational searches, such as “What is the capital of Kazakhstan?”, “Was Jesus a Muslim?” or “When will the world end?”. But by 2017 Amazon had captured 52 percent of the product-search market; intentional searches for products to buy, vs 26 percent for Google and other search engines.

Nobody really wants the value-subtractive ads that are the underpinning of the Google system. Gilder, who is 78, thinks this will doom Google within his lifetime, although he qualifies his prediction by saying that search will likely remain a valuable business.

Crashing into Hölzle’s Wall

If you meet any technology investor today his first question will likely be if your business is “scalable”? Meaning if you are able to grow revenues with less than proportional, or ideally no increases in marginal cost – the key to ever-fatter profit margins. That is what the likes of Google and Facebook have achieved, and what investors hope the likes of Amazon, Uber, Spotify, AirBnB et al. will achieve.

The question is if the margin story can continue for Google? An inherent challenge in Google’s business model is that “free” inevitably leads to over-consumption. “Free” has enabled Google to capture (effectively) the entire market, but will the company eventually collapse under the weight of the unlimited demand for its free services? As Gilder writes: “The nearly infinite demand implicit in ‘free’ runs into the finitude of bandwidth, optical innovation, and finance – a finitude that reflects the inexorable scarcity of time. This finitude produces not zero marginal costs but spikes of nearly infinite marginal costs – Hölzle’s wall. The latter a reference to Google employee number 8, Urs Hölzle’s lament that Google’s cloud infrastructure was “rapidly approaching a wall”.

The exponential growth in demand for Google’s services (see chart below) – which roughly doubles every year – has forced Google to build out the largest cloud infrastructure on the planet over the past 15 years, with three underwater cables across the Pacific and more on the way. The 12.899km third cable from California to Hong Kong (that was scheduled to be completed last year but is delayed) will carry data at a rate of 144 terabits per second – up 29-fold from Google’s first 5tbps cable from California to Japan in 2010.

Screenshot 2019-04-22 14.10.56 — Source: Urs Hölzle keynote OFC Plenary, 2017

This is putting huge Capex pressure on Google, with the need to continue building massive data centres and undersea cables. As Hölzle pointed out in his 2017 keynote apart from the change from copper to fibre, cable building has not changed much over the past 150 years and the system’s reliance on a small amount of cables leaves it vulnerable to tail risks. Still, Google is to a significant degree essentially free-riding on the backs of carriers such as AT&T, Verizon and T-Mobile, whose infrastructure investments Google is totally reliant on. The cost is ultimately passed on to the end user: On average (US) smartphone users pay 23 dollars per month for ads, trackers, scripts and other spam that strikes them with malware, slows load times and depletes battery life. Strikingly, on popular publishers’ sites as much as 79 percent of the mobile data are adds.

What will replace the Google World Order?

The latter parts of Gilder’s book are somewhat unstructured and anecdotal but contain insightful observations. Obviously, it would be a futile exercise to foretell exactly how the Google system will meet its end. But Gilder argues interestingly that the overarching threat to Google will come from the Cryptosphere. That the un-secure and centralised Google Empire will be replaced by a decentralised and secure Crypto order. That is admittedly far away – 70 percent of all Internet links are handled through Google or Facebook. But alternatives are emerging, such as Blockstack, whose mission is to “foster an open and decentralised Internet that establishes and protects privacy, security and freedom for all internet users.” Tim Berners-Lee, who fears the Internet is being broken by the centralisation of user data in the giant silos of Google, Facebook, Amazon and Microsoft – is one Blockstack admirer.

Gilder also goes through the emergence of Bitcoin and Ethereum and discusses their potential and limitations as currency systems. He sees the upper limit of 21 million total bitcoin units as one of the main challenges that would likely be highly deflationary over time. The high price volatility of Bitcoin could likely be addressed by more issuance and circulation of Bitcoin-redeemable liabilities – something which SEC’s clampdown on ICOs and crypto in general may put up obstacles for. Nevertheless, despite all the challenges it is clear that it is in the Cryptosphere the most interesting economic and political ideas are to be found today.

In the “Googlesphere” on the contrary, true entrepreneurial activity is being stifled by the narrow focus on developing largely parasitic apps and profitless Unicorns inside the Google/Android, Facebook or Apple iOS ecosystems — (a charge I would stand guilty of myself). Gilder also laments investors excessive belief in Marc Andreessen’s mantra that “Software will eat the world”, which has the unfortunate effect of crowding out innovative hardware investment. Though there are pockets of light, such as Luminar Technologies, which manufactures advanced Lidar censors for autonomous vehicles. What makes Luminar founder Austin Russell stand out from the crowd is that he does not believe in the consensus that autonomy is chiefly a software problem, but a hardware one, and that no amount of big data, artificial intelligence and software can make up for bad data from hopelessly outdated hardware. In contrast for example, two of the three finalists in the startup pitch at the Web Summit in Lisbon before Christmas were self-driving startups focused on getting more out of existing hardware with better software, one of them even claiming that a $12 camera will suffice for self-driving – Luminar would beg to differ.

Most business books today are boring. Life after Google is definitely not boring. And even if Gilder will be proved wrong in his hypotheses about the fall of the Google system of the world and the rise of the Blockchain economy, his analyses are insightful as well as idiosyncratic. That makes the book worth reading for everybody seeking to understand the new and uncharted economic landscape we find ourselves in.

Reid Hoffman’s not fully convincing book on “Blitzscaling”

Businesses can achieve global scale faster than ever in the Networked Age, argues LinkedIn-founder Reid Hoffman in his new book, but must grow fast or die slow in a brutal market where the winner takes all.

Blitzscaling — The lightning-fast path to building massively valuable companies. Reid Hoffman, Chris Yeh, 336 pages, Penguin Random House, 2018

Reid Hoffman may be the closest Silicon Valley has to a philosopher king. Stanford- and Oxford-educated Settlers from Catan enthusiast. Original Communist, who has dedicated his formidable brain capacity to “playing monopoly” on the Internet. Member of the “PayPal Mafia”, the eccentric group of nerds — including Elon Musk and Peter Thiel — who founded the pioneering payment service before moving on to building some of the most powerful businesses in the Internet age. In Hoffman’s case, the professional network LinkedIn. Since Linkedin was acquired by Microsoft for $26,2 billion in 2016, Hoffman has served as a board member of the software giant along with his numerous other investing and philanthropic endeavours.

Precisely because of his somewhat paradoxical background, Hoffman brings illuminating perspectives on the 21st century Internet economy, in his new book — Blitzscaling: The lightning-fast path to building massively valuable companies — and explains why he believes the ability to act fast is the key to competitive advantage in the modern economy.

The title, Blitzscaling, derives from Blitzkrieg. The German concept of attacking warfare perfected by Heinz Guderian during the invasion of France in 1940. French forces were completely outmanoeuvred and overwhelmed by the surprise element of the Germans’ lightning-fast and densely concentrated combined tank and air attacks. Hoffman is fully aware of the nomenclature’s burdened history but has chosen analytical precision over political correctness — also considering that the German word (which in fact was popularised by the British tabloids and never officially used by the Wehrmacht. Guderian himself referred to the doctrine as Bewegungskrieg in his book Achtung — Panzer!) or its shortened form Blitz has become a common metaphor in sports and elsewhere. As Hoffman makes clear, Blitzscaling accurately captures the strategic essence of how technology companies such as Amazon, Google, Facebook, Apple, Uber and AirBnB have rapidly subjected huge markets to their rule.

Dawn of the Networked Age

At the end of 1996, the world’s five most valuable companies were General Electric, Royal Dutch Shell, The Coca-Cola Company, Nippon Telegraph & Telephone and ExxonMobil. The common denominator is that they all were traditional industrial or consumer goods companies. Fast forward to the end of 2017, and the the Top 5 list looks very different: Apple, Google, Microsoft, Amazon and Facebook. All technology companies, of which Google and Facebook were not even born in 1996, while Amazon was a two-year-old toddler.

What happened? The Networked Age happened — Hoffman’s preferred term for the era he dates from the stock market listing of Netscape, the pioneering web browser provider, in 1995. Today, over two billion people are connected to the internet through smartphones that have become an extended limb of the human body. At the same time, more and more industries and businesses are being run on software and delivered as online services, as Netscape founder Marc Andreessen wrote in a well-known essay in 2011, “Software is eating the world”. Binge watching on Netflix has replaced the drive to the video rental shop. Amazon has delivered mass death for physical bookstores. IBM produced its last PC in 2005, and today is essentially a software and services provider. Even industries that produce physical products (the economy of atoms), like cars or refrigerators, are becoming ever more integrated with software (the economy of bits). Such as when a Tesla’s acceleration is upgraded by an automatic software update while the car sits untouched in the garage overnight.

Goodbye Galapagos

Hoffman’s central thesis is that in this hyperconnected world it is possible to build global monopolies faster than ever in history. This has fundamentally changed the market economy. While the world economy previously consisted of geographically fragmented “Galapagos Islands”, where many businesses such as newspapers and bookstores were relatively sheltered from external competition, the networked age has tied these islands together in a hyper-competitive global market.

The Networked Ages opens up enormous opportunities for entrepreneurs with wet dreams of monopoly. The flip side is a merciless market where “thewinner takes all”. Hoffman illustrates with an analogy from the movie Glengarry Glen Ross, where Alec Baldwin’s character, Blake, tells a sales team: “As you all know, first prize is a Cadillac Eldorado. Anyone wanna see second prize? Second prize is a set of steak knives. Third prize is you’re fired. Get thepicture?” Or in the social network category: First prize to Facebook, second to MySpace, and third prize to Friendster, who hardly anyone remember.

Why did Facebook win? Mark Zuckerberg was not the market’s first mover — Both Friendster and MySpace got off the ground before Facebook. But Facebook was the first scaler. Being the first to achieve critical mass is decisive in a market with network effects. If all your friends choose Facebook, it is pointless to sign up with another social network that no one uses. This creates a positive feedback loop, eventually ending in an equilibrium where Facebook more or less monopolises the market while all the other players go the way of Friendster. De facto the consumer does not really have much of a free choice whether he wants to use Facebook or a competing social network. Like a political prisoner in a remote Siberian labour camp a Facebook user in theory is quite free to go anywhere, but in practice has nowhere else to go. The degree of customer lock -in Mark Zuckerberg has cunningly engineered would be admired by the designers of the Gulag.

The Magic of Network Effects

Hoffman prefers the layman’s definition of network effects: A product or service is subject to positive network effects when increased usage by one user increases the value of the product or service for other users. Network effects occur in several forms: Increasing numbers of passengers or guests attracts more drivers and landlords to marketplaces such as Uber and AirBnB, and vice versa, (two-sided network effects); The dominance of operating systems such as Windows, iOS or Android, encourage third-party application developers to adapt their apps to these platforms, (indirect network effects); Microsoft Word’s dominance meant that its document file format became the industry standard and swept all non-compatible competitors off the pitch, (network effects driven by compatibility and industry standards). Similarly, that Internet Explorer came pre-installed on all Windows PCs was the “pincer movement” that broke Netscape.

A feature in markets with network effects is increasing returns to scale, which often results in a monopolistic/oligopolistic equilibrium where one single or a few dominant players eat all of an industry’s profits — as demonstrated today by for instance Facebook. Facebook enjoys a gross margin of a whopping 82 percent. Yet, as Hoffman reminds, many failed to see the potential value of Facebook in the company’s early days, when Zuckerberg turned down repeated takeover offers. Perhaps the doubters were not without reason; as late as 2012 Facebook struggled with the transition from desktop to mobile.

Move Fast and Break Things

Facebook’s problem with finding the right product/market fit was nevertheless a luxury problem. Thanks to the record-breaking user growth, Facebook accumulated enormous amounts of data on billions of people — a digital gold mine. It was just a question of time to figure out the profit-maximising business model. MySpace or Friendster didn’t have this privilege. Without users, no business model. With users, one can find a business model later.

In this market landscape, speed is as vital as it was for Guderian’s Panzerkorpsin World War II. The only difference is that while Guderian had to capture land, business have to capture data. To achieve scale quickly, businesses must be willing to sacrifice efficiency for speed. Companies must grow fast or die slow. Summarised in Mark Zuckerberg’s previous motto: Move fast and breakthings. The classic approach to business strategy, carefully gathering information and making decisions with a fairly high confidence level, is dead. Blitzscaling forces businesses to make much faster decisions in the face of much higher levels of uncertainty, and they must commit to making big investments being half-blinded in the dark.

In a market where the winner takes all (or most), it becomes more important to cement market share before focusing on profitability. For Jeff Bezos in Amazon, profit has been secondary to growth for a quarter of a century — expressed by his bon mot that: “your margin is my opportunity”. Uber has subsidised both drivers and passengers when they have launched in new markets, in order to take the pole position before their competitors. Naturally, it would not have been possible without investors who have been willing to pump in nine billion dollars to fund Uber’s growth — an investment they can be far from sure to recoup. The growth imperative in the networked age also explains the popularity of Freemium business models, which companies from Spotify to Dropbox employ, luring new customers with a free version in the hope that they will upgrade to a paying premium subscription later on.

Can Software Digest the World?

It is a small paradox that technological developments are taking place so quickly, at the same time as productivity growth in the economy overall is stagnating. This must probably be seen in light of the fact that only a very small number of companies achieve lightning-fast growth and monopoly profit, while many others are lagging behind. Information technology also appear to be a contributing factor to the rising market concentration seen in many industries, raising the productivity gap between market leading and average firms.

As Peter Thiel, among others, have highlighted there is a dichotomy between the economy of bits and the economy of atoms, with rapid technological development in the former and far slower in the latter.

Thiel’s bifurcation can be illustrated by Tesla, a company at the intersection of new technology and old industry. Due to physical bottlenecks in the value chain, Tesla has not managed to increase car production in line with rising demand — hampering the company’s growth and exposing its fragile finances. One big reason why fast-growing online businesses like Dropbox have been able to navigate around such infrastructure limitations is Amazon’s cloud services offering, Amazon Web Services (AWS) — which almost by accident thus has become Amazon’s primary cash cow. However, so far nobody, not even Elon Musk, has found a way to manufacture Teslas in the cloud. Unfortunately, it must still be done in the physical world. This is the Achilles heel of Hoffman’s thesis: Companies can grow exponentially in the internet age, as long as they confine their business to the realm of bits. Confronted with the sticky realities of the world of atoms, things tend to slow down. Perhaps software is eating the world, but there are still some chunky pieces of the physical economy left that may prove tricky to chew and digest.

Previously published on Medium