Stratechery Plus

Subscribe

Learn More
Member Forum

Menu

Stratechery Plus

Subscribe

Learn More
Member Forum

Latest Podcast

Stratechery Plus Update

  • The Uber Dilemma

    By far the most well-known “game” in game theory is the Prisoners’ Dilemma. Albert Tucker, who formalized the game and gave it its name in 1950, described it as such:

    Two members of a criminal gang are arrested and imprisoned. Each prisoner is in solitary confinement with no means of communicating with the other. The prosecutors lack sufficient evidence to convict the pair on the principal charge. They hope to get both sentenced to a year in prison on a lesser charge. Simultaneously, the prosecutors offer each prisoner a bargain. Each prisoner is given the opportunity either to: betray the other by testifying that the other committed the crime, or to cooperate with the other by remaining silent. The offer is:

    • If A and B each betray the other, each of them serves 2 years in prison
    • If A betrays B but B remains silent, A will be set free and B will serve 3 years in prison (and vice versa)
    • If A and B both remain silent, both of them will only serve 1 year in prison (on the lesser charge)

    The dilemma is normally presented in a payoff matrix like the following:

    What makes the Prisoners’ Dilemma so fascinating is that the result of both prisoners behaving rationally — that is betraying the other, which always leads to a better outcome for the individual — is a worse outcome overall: two years in prison instead of only one (had both prisoners behaved irrationally and stayed silent). To put it in more technical terms, mutual betrayal is the only Nash equilibrium: once both prisoners realize that betrayal is the optimal individual strategy, there is no gain to unilaterally changing it.

    TIT FOR TAT

    What, though, if you played the game multiple times in a row, with full memory of what had occurred previously (this is known as an iterated game)? To test what would happen, Robert Axelrod set up a tournament and invited fourteen game theorists to submit computer programs with the algorithm of their choice; Axelrod described the winner in The Evolution of Cooperation:

    TIT FOR TAT, submitted by Professor Anatol Rapoport of the University of Toronto, won the tournament. This was the simplest of all submitted programs and it turned out to be the best! TIT FOR TAT, of course, starts with a cooperative choice, and thereafter does what the other player did on the previous move…

    Analysis of the results showed that neither the discipline of the author, the brevity of the program—nor its length—accounts for a rule’s relative success…Surprisingly, there is a single property which distinguishes the relatively high-scoring entries from the relatively low-scoring entries. This is the property of being nice, which is to say never being the first to defect.

    This is the exact opposite outcome of a single-shot Prisoners’ Dilemma, where the rational strategy is to be mean; when you’re playing for the long run it is better to be nice — you’ll make up any short-term losses with long-term gains.

    Silicon Valley’s Iterated Game

    What happens in Silicon Valley is far more complex than what can be described in a simple game of Prisoners’ Dilemma: instead of two actors, there are millions, and “games” are witnessed by even more. That, though, accentuates the degree to which Silicon Valley as a whole is an iterated game writ large: sure, short-term outcomes matter, but long-term outcomes matter most of all.

    That, for example, is why few folks are willing to criticize their colleagues or former companies:1 today’s former co-worker or former manager is tomorrow’s angel investor or job reference, and memories are long and reputations longer.2 That holds particularly true for venture capitalists: as Marc Andreessen told Barry Ritholtz on a recent podcast, “We make our money on the [startups] that work and we make our reputation on the ones that don’t.”

    Note the use of plurals: a venture capitalist will invest in tens if not hundreds of companies over their career, while most founders will only ever start one company; that means that for the venture capitalist investing is an iterated game. Sure, there may be short-term gain in screwing over a founder or bailing on a floundering company, but it simply is not worth it in the long-run: word will spread, and a venture capitalists’ deal flow is only as good as their reputation.

    The most famous example of this is cemented in Valley lore. From The Facebook Effect:

    Facebook’s success was beginning to make waves. And in Silicon Valley, success attracts money. More and more investors were calling. Zuckerberg was uninterested. One of the supplicants was Sequoia Capital. Among the bluest of blue chip VCs, Sequoia had funded a string of giants—Apple, Cisco, Google, Oracle, PayPal, Yahoo, and YouTube, among many others. The firm is known in the Valley for a certain humorlessness and a willingness to play hardball. Sequoia eminence grise and consummate power player Michael Moritz had been on Plaxo’s board and was well acquainted with Sean Parker. It was not a mutual admiration society. Parker saw Moritz as having contributed to his downfall. [Parker was fired from the company he founded by the board, including Moritz] “There was no way we were ever going to take money from Sequoia, given what they’d done to me,” says Parker.

    Plaxo raised a total of $19.3 million in the rounds in which Sequoia participated; was whatever percentage of that $19.3 million Sequoia put in worth missing out on the chance to invest in one of the greatest grand slams in the history of venture investing?

    The entire point of venture investing is to hit grand slams, and that calls for more swings of the bat. After all, the most a venture capitalist might lose on a deal — beyond time and opportunity cost, of course — is however much they invested; the downside is capped. Potential returns, though, can be many multiples of that investment. That is why, particularly as capital has flooded the Valley over the last decade, preserving the chance to make grand slam investments has been paramount. No venture capitalist wants to repeat Sequoia’s mistake: better to be “nice”, or, as they say in the Valley, “founder friendly.”

    Benchmark Sues Kalanick

    This is why what happened last week was so shocking: the venture capital firm Benchmark Capital filed suit against former Uber CEO Travis Kalanick for fraud, break of contract, and breach of fiduciary duty. From Axios:

    The suit revolves around the June 2016 decision to expand the size of Uber’s board of voting directors from eight to 11, with Kalanick having the sole right to designate those seats. Kalanick would later name himself to one of those seats following his resignation, since his prior board seat was reserved for the company’s CEO. The other two seats remain unfilled. Benchmark argues that it never would have granted Kalanick those three extra seats had it known about his “gross mismanagement and other misconduct at Uber” — which Benchmark claims included “pervasive gender discrimination and sexual harassment,” and the existence of confidential findings (a.k.a. The Stroz Report) that recently-acquired self-driving startup Otto had “allegedly harbored trade secrets stolen from a competitor.” Benchmark argues that this alleged nondisclosure of material information invalidates Benchmark’s vote to enlarge the board.

    Moreover, Benchmark alleges that Kalanick pledged in writing — as part of his resignation agreement — that the two empty board seats would be independent and subject to approval by the entire board (something Benchmark says was the reason it didn’t sue for fraud at the time). But, according to the complaint, Kalanick has not been willing to codify those changes via an amended voting agreement.

    Giving three extra seats on the board to the CEO was certainly founder friendly; that the expansion happened at the same time Uber accepted a $3.5 billion investment from Saudi Arabia’s Public Investment Fund, which came with a board seat, suggests Benchmark viewed the board expansion as a way to protect its own interests and influence as well. After all, longtime Benchmark general partner and Uber board member Bill Gurley had been pursuing ride-sharing years before Uber came along, and the investor had penned multiple essays on his widely-read blog defending and extolling Kalanick and company.

    Then again, by June 2016, when the board was expanded and the Saudi investment was announced, Gurley’s posts had taken a much sterner tone: specifically, in February 2015 Gurley warned that late-stage financing was very different than an IPO, and that it had “perverse effects on a company’s operating discipline.” A year later, in April 2016, Gurley said that the “Unicorn financing market just became dangerous…for all involved”, and that included Benchmark:

    For the most part, early investors in Unicorns are in the same position as founders and employees. This is because these companies have raised so much capital that the early investor is no longer a substantial portion of the voting rights or the liquidation preference stack. As a result, most of their interests are aligned with the common, and key decisions about return and liquidity are the same as for the founder. This investor will also be wary of the dirty term sheet which has the ability to wrestle away control of the entire company. This investor will also have sufficient angst about the difference between paper return and real return, and the lack of overall liquidity in the market. Or at least they should.

    I suspect this, more than anything, explains this unprecedented lawsuit.

    The Uber Outlier

    Benchmark is one of those most successful venture firms ever. Founded in 1995 with a commitment to early stage funding, the firm has, going by this chart from CB Insights been an investor in 14 IPOs, 11 in the last five years (the chart shows 13 and 10; I added Snapchat, which IPO’d earlier this year).

    The company’s investments include Twitter, Dropbox, Instagram, Zendesk, Hortonworks, New Relic, WeWork, Grubhub, OpenTable, and many more; according to CB Insight, since 2007, the companies Benchmark has invested in have exited (via IPO or acquisition) for a combined $75.96 billion.3

    That, though, simply highlights what an outlier Uber is, at least on paper. Uber’s most recent valuation of $68.5 billion nearly matches the worth of every successful Benchmark-funded startup since 2007. Sure, it might make sense to treat company X and founder Y with deference; after all, there are other fish in the pond. Uber, though, is not another fish: it is the catch of a lifetime.

    That almost assuredly changed Benchmark’s internal calculus when it came to filing this lawsuit. Does it give the firm a bad reputation, potentially keeping it out of the next Facebook? Unquestionably. The sheer size of Uber though, and the potential return it represents, means that Benchmark is no longer playing an iterated game. The point now is not to get access to the next Facebook: it is to ensure the firm captures its share of the current one.

    This, I would note, is a lesson founders should learn: Kalanick was resolutely opposed to an IPO, claiming he would wait “as long as humanly possible”; his delay, though, completely flipped the incentives of Kalanick and his early investors. While in most companies the venture capitalists have to worry about their reputation along with their capital, in the case of Uber there is simply too much money at stake: transforming a $68 billion paper return to a real return (and guaranteeing a per partner return in the nine figures) is worth whatever reputational damage is incurred along the way.

    In other words, an iterated game is good for founders: it ensures venture capitalists are nice. Single move games, though, which Uber has become, often end badly for everyone, particularly founders.

    Diminished Uber

    Understanding that Benchmark is focused on achieving liquidity on its all-time greatest investment suggests two potential outcomes:

    • The most straightforward is that Benchmark hopes to push Uber to an IPO sooner-rather-than-later; clearly Kalanick was an obstacle as CEO, and according to reports, has sought to reestablish control of the company via his control of the board, driving away Meg Whitman, who was reportedly Benchmark’s choice for CEO.4 This also explains the urgency of this suit: Benchmark is trying to prevent Kalanick from naming two more members to the board, further complicating the CEO selection process.
    • The other potential outcome is that Benchmark is looking for an exit. Softbank, which is looking to dominate car-sharing globally, has reportedly had discussions with Benchmark and other investors about buying their shares; reports have been mixed as to who wants to make a deal — Kalanick or Benchmark — but if it is the latter a lawsuit is an excellent way of getting the former to agree to a sale.

    There is a third possibility: that Uber broadly and Kalanick specifically are in big trouble when it comes to Waymo’s lawsuit against the company, and that Benchmark is making clear that it is not culpable. A full six pages of Benchmark’s lawsuit were dedicated to describing Kalanick’s role in the Otto acquisition and Benchmark’s obliviousness to alleged wrongdoing; I noted when the lawsuit was filed that it, more than any of Uber’s scandals, had the potential to be Kalanick’s doom, and apparently Benchmark agrees (although, of course, one should question why Gurley, then an Uber board member, apparently declined to do more digging on a $680 million acquisition).

    What is without question, though, is that whatever outcome results from this mess will be a suboptimal one; most Uber critics still fail to appreciate that the ride-sharing market is demand driven, which meant Uber really did have a chance to be the transportation behemoth of much of the world. Now the company is retreating throughout Asia, is on the regulatory run in Europe, and stuck in a fight it should have never drawn out with Lyft in the United States. Perhaps Benchmark will get its all-time great return, reputations be damned. It seems unlikely its return will be what it once might have been.

    1. Above and beyond problematic arbitration agreements  ↩

    2. This isn’t always a good thing: one reason serious issues like sexual harassment by venture capitalists go underreported is that the harassed worry about the long-term effect on their reputation — will future investors simply see them as trouble? ↩

    3. According to CB Insight, IPO valuation is based on first day closing price; acquisition valuation is based on public pronouncements or whisper valuations ↩

    4. Whitman is most famous for her stewardship of eBay, Benchmark’s first big breakthrough investment ↩

  • Apple and the Oak Tree

    On October 5, 1999, Steve Jobs introduced the iMac DV and a new application called iMovie, declaring:

    We think this is going to be the next big thing. Desktop video…which we think is going to be as big as desktop publishing was.

    The problem is that “the next big thing” had already arrived: four months earlier Shawn Fanning and Sean Parker had released an app called Napster; I can personally attest that, by the time Jobs introduced the iMac, the music-sharing app had swept over university networks in particular.

    Jobs told Fortune that it took until the following year to realize his error:

    “I felt like a dope,” says Jobs, thinking back to summer 2000, when his fixation on perfecting video editing on the Mac distracted him from noticing that millions of kids were using computers and CD burners to make audio CDs and to download digital songs called MP3s from illegal online services like Napster. Yes, even Jobs, the technological visionary of his generation, occasionally gets caught looking in the wrong direction. “I thought we had missed it. We had to work hard to catch up.”

    What followed was one of the great pivots in tech history. Less than a year later, in January 2001, Jobs was again on stage, now declaring that the future of the PC was to be a digital hub that made digital devices 10x more valuable than they could ever be on their own, and that Apple’s focus was no longer video but audio. After all, “there is a music revolution happening.”

    One month later, Toshiba showed Jon Rubinstein their new 1.8″ hard drive that they didn’t know what to do with; eight months later Jobs was back on stage introducing the iPod. Two years after that Apple brought iTunes to Windows, unshackling the iPod from the presumption it existed to sell Macs, and completing the foundation for the greatest corporate run of success ever.

    Rip. Mix. Burn.

    One of my favorite artifacts from the brief period between the introduction of iTunes and the release of the iPod was Apple’s “Rip. Mix. Burn.” advertising campaign.

    What is particularly amazing (that is, beyond the cringe-inducing television ad) is that Apple was arguably encouraging illegal behavior: it was likely legal to rip and probably legal to burn, presuming the CD that you made was for your own personal use. It certainly was not legal to share.

    And yet, as much as “Rip. Mix. Burn.” may have walked the line of legality, the reality of iTunes — and the iPod that followed — was well on the other side of that line. Apple knew better than anyone that the iPod’s tagline — 1,000 songs in your pocket — was predicated on users having 1,000 digital songs, not via the laborious procedure of ripping legally purchased CDs, but rather via Napster and its progeny.1 By the spring of 2003 Apple had introduced the iTunes Music Store, a seamless and legal way to download DRM-protected digital music,2 but particularly in those early days the value of the iTunes Music Store to Apple was not so much that it was a selling point to consumers, but rather a means by which Apple could play dumb about how it was that its burgeoning number of iPod customers came to fill up their music libraries.

    To be clear, I’m not very bent out of shape about this; the reality is that piracy was happening before Apple woke up to the music revolution, and would have continued whether the iPod came along or not.3 In fact, by offering a legal alternative that not only matched but exceeded the convenience of piracy, Apple pointed the way to a surprisingly bright future for the music labels.

    What is worth noting, though, is that Apple’s breakthrough product — the one that started Apple down the road to the iPhone, iPad, App Store, everything that contributed to yesterday’s financial results — was not simply a product of Steve Jobs vision, or Rubinstein or Tony Fadell or Jony Ive or any of the other folks at Apple. The iPod was very much a product of the company that created it and the world in which it came to be: somewhat lawless, with nothing to lose.

    The End of the iPod

    Given the context of the iPod’s introduction, there has been, if you squint, a certain symmetry in the circumstances surrounding its death. The iPods Shuffle and Nano, the last two iTunes-dependent (i.e. non-iOS) MP3 players Apple sold, were quietly discontinued last Tuesday. The revelation two days later that Apple was, at the behest of the Chinese government, removing VPN apps from the App Store in China, drew considerably more interest.

    These two stories are related by more than their timing. The iPod was obsoleted by the iPhone, reduced to just an app on a general purpose device, while the very concept of individual songs synced over a wire is a relic in a world where over 30 million songs can be streamed at any time (“Mix” is the one word that has endured from Apple’s old slogan).

    Apple’s preference, of course, is that you stream via Apple Music, one of the key parts of Apple’s Services businesses; the “Services” line on Apple’s income statement is now the second-largest (after the iPhone), and has loomed largest in Apple’s quarterly presentation to analysts for the last year-and-a-half. The pitch is a valid one: while iPhone customers mostly stick to the platform, and Apple attracts switchers, the upgrade cycles are elongating and the low-hanging fruits of country and carrier expansion are in the past. What is and will continue to be the case are that phones are the most important devices in people’s lives, which means the monetization of that importance — via the App Store, Apple Music, and iCloud Storage — is a business that is pure upside.

    Pure upside, that is, from a dollars-and-cents perspective. The broader implications are a little more complex.

    Apple’s Elegant Business Model

    Apple’s traditional model — selling hardware differentiated by software at a premium — had an elegant simplicity that went beyond the impact on the company’s income statement. Just look at the iPod: as much as the music industry may have whined about Apple earning profits they insisted were theirs, there was nothing record labels could actually do about it. iPods were transactional products that could be filled with legal music, pirated music, podcasts, music of one’s own creation — it didn’t matter to anyone but the iPod’s owner. Apple sold a product that benefited from openness into a relatively lawless market and reaped the rewards.

    China has long been another example of the advantage of Apple’s model. A consistent point I made in the early years of Stratechery was that the company, relative to its Western peers, was uniquely equipped to compete in China. For example, from 2014:

    The truth is that China is not and will not be a meaningful market for nearly every Western consumer-focused tech company. The Great Firewall makes all service-based companies unviable, including Facebook, Twitter and of course Google, while widespread piracy makes pure software plays (i.e. Windows and Office) widely available even as they drive negligible revenue.

    The exception is Apple: because the company monetizes through hardware and differentiates through exclusive software, its products physically exist inside the Great Firewall even as they avoid the piracy trap. It’s a big advantage relative to other U.S. based companies.

    Later that year Apple would release the iPhone 6 and reap the rewards of that advantage: Greater China quickly became Apple’s second-largest market, buying an incredible $59 billion worth of Apple products in the company’s 2015 fiscal year.4 Naturally, despite the fact Apple’s China sales have faltered with the iPhone’s increasingly stale design, services revenue has only grown; according to App Annie, App Store revenue in China surpassed App Store revenue in the United States last fall, making China the most important market for Apple’s fastest growing segment.

    Small wonder, then, that Apple has deemed it prudent to stay in the government’s good graces. Tim Cook argued on Apple’s earnings call — correctly and fairly, to be sure — that in the case of removing VPN apps the company is simply following the law; of course there is no law that says Apple, contrary to the company’s behavior in other countries or markets, ought to invest $1 billion in a Chinese company (then) competing with a Western challenger, or open R&D facilities worth $500 million when the company has been reticent for years to let technology-focused employees work in San Francisco, much less across the Pacific (although the law certainly figured in opening a new data center in China).

    The reality is that Apple’s elegant transaction-based business model, centered on selling software-differentiated hardware, died along with the iPod. One could certainly argue that Apple’s services don’t differentiate their hardware, at least not in a positive direction, but it is impossible to deny that said services play an ever more important role in monetizing said hardware, above-and-beyond increasingly infrequent (on an individual basis) up-front purchases. And while that is great for Apple’s continued growth, it is a limit on Apple’s freedom to maneuver — and now, for the company’s Chinese customers, a limit on their freedom to circumvent China’s Great Firewall.

    Apple’s Services Shift

    None of this is a criticism of Apple; if anything the company deserves praise for developing a revenue stream that is growing even as iPhone sales growth has slowed (or in last year’s case, declined). What it is, though, is an example of how success carries its own curse: Apple today has far more to lose than it did two decades ago, and that means less of a focus on doing what is best for non-Apple stakeholders, not more.

    There are others:

    • Apple’s services revenue is largely built on the App Store, particularly in-app purchases in free-to-play games. That means the company has no incentive to lower its 30% take, or offer side-loading (which would, as John Gruber astutely noted, make it far more difficult for the Chinese government to enforce its VPN app ban).
    • Apple is bringing the HomePod, its Siri-based home speaker product, to market a full three years after Amazon’s Echo; it seems likely the company was blind to the home opportunity because it had such a strong position in smartphones.5
    • As for the HomePod, Cook highlighted on the earnings call that it is “designed to work with your Apple Music subscription”; if you have a Spotify subscription and want voice control, you will have to get an Echo instead.

    Indeed, Apple’s attempt at services lock-in is steadily increasing: HomePod supports only Apple Music and Siri, CarPlay supports only Siri and Apple Maps, iOS still doesn’t let one change default applications. None of these decisions are based on delivering a superior experience, the key to Apple’s differentiation with a hardware-based business model; all are based on securing an ongoing relationship with the company that can be monetized over time.

    Again, this all makes sense, particularly for the bottom line: every bit of lock-in makes Apple’s business stronger. Stronger, that is like an oak tree.

    A Giant Oak stood near a brook in which grew some slender Reeds. When the wind blew, the great Oak stood proudly upright with its hundred arms uplifted to the sky. But the Reeds bowed low in the wind and sang a sad and mournful song.

    “You have reason to complain,” said the Oak. “The slightest breeze that ruffles the surface of the water makes you bow your heads, while I, the mighty Oak, stand upright and firm before the howling tempest.”

    “Do not worry about us,” replied the Reeds. “The winds do not harm us. We bow before them and so we do not break. You, in all your pride and strength, have so far resisted their blows. But the end is coming.”

    As the Reeds spoke a great hurricane rushed out of the north. The Oak stood proudly and fought against the storm, while the yielding Reeds bowed low. The wind redoubled in fury, and all at once the great tree fell, torn up by the roots, and lay among the pitying Reeds.

    — Aesop’s Fables

    The Burden of Success

    Let me be as explicit as I can be: Apple is not doomed. Indeed, the company’s future is bright, particularly in the short term; I expect the next iPhone, particularly the rumored high-end model, to be a big hit in China in particular (close readers will note that that was one of the counterintuitive conclusions in my piece originally pointing out Apple’s China Problem).

    Indeed, what has always made the “Apple is doomed” argument so dumb is that it has always implied that Apple was some sort of special snowflake, incapable of leveraging its massive user base or demonstrated ability to iterate on its industry-leading products. As if the company would somehow forget how to make a phone,6 or that developers would give up on a user base in the hundreds of millions, or that users would suddenly not care about nice things. All nonsense.

    That, though, is my point: Apple has had a special run, thanks to its special ability to start with the user experience and build from there. It is why the company is dominant and will continue to be so for many years. Apple as an entity, though, is not special when it comes to the burden of success: there will be no going back to “Rip. Mix. Burn.” or its modern equivalent.

    In short, Apple is no longer the little reed they were when Jobs could completely change the company’s strategy in mere months; the push towards ever more lock-in, ever more centralization, ever more ongoing monetization of its users — even at the cost of the user experience, if need be — will be impossible to stop, for Tim Cook or anyone else. After all, such moves make Apple strong, until of course they don’t.

    Such is life, and time, inexorably flowing past oak trees standing and fallen alike.

    1. Or, if you’re an old fogey like me, dicey web directories and FTP clients ↩

    2. DRM would eventually be removed in 2009 ↩

    3. This is no different than how YouTube grew, for example ↩

    4. Which ran from October 2014-September 2015 ↩

    5. Editor: Lots of pushback this isn’t late. Fair points! ↩

    6. Not that that matters  ↩

  • Microsoft’s Monopoly Hangover

    Microsoft announced something very impressive last week: revenue for the company’s 2017 fiscal year (which ended June 30) increased 5% year-over-year. That may not seem particularly meaningful until you realize 2016 was only the second year in the company’s history that revenue declined; the first included the worst economic slowdown since the Great Depression:

    Moreover, all indications are that growth will continue, defeating the presumption that tech companies that start to decline do so inexorably. The most famous example that said inexorable decline need not be inevitable is IBM, which, in the early 90s, found itself in far more dire straits than Microsoft, only to recover under the leadership of Lou Gerstner:

    Microsoft’s earnings report isn’t the only thing that has made me think of IBM lately; two weeks ago, at Mirosoft’s annual partner conference, CEO Satya Nadella introduced a new offering called Microsoft 365. Nadella said:

    Microsoft 365 is a fundamental departure in how we think about product creation. This is the coming together of the best of Office 365, Windows 10, Enterprise Mobility and Security…

    We have decided that the time has come for us as a company and us as an ecosystem to talk about this in the terms that customers can get the most value. We want to bring these products together as an integrated solution. A complete solution that has got AI infused in it with intelligence, whether it is intelligence that is helping end users be more productive and creative and teamwork, or intelligence in security. It’s that complete solution for intelligent teamwork and security that we want to bring about with Microsoft 365.

    A cynical take is that this is typical Microsoft, cribbing a successful naming scheme (‘365’) to rebrand a SKU that Microsoft actually announced a year ago. That’s true! A slightly more generous take is that Microsoft 365 is the latest implementation of the company’s decades-old bundling strategy, and, well, that’s true too!

    The way Nadella framed the announcement though — associating customer value with integration — that is straight from Gerstner’s IBM playbook.

    The IBM and Microsoft Monopolies

    When Gerstner signed on as IBM CEO in the spring of 1993, the company had just recorded the biggest annual loss in American corporate history: -$4.97 billion. In his memoir about the turnaround he led at IBM, Who Says Elephants Can’t Dance?, Gerstner noted that 1993 was going just as badly:

    At the end of May I saw April’s [numbers] and they were sobering. Profit had declined another $400 million, for a total decline of $800 million for the first four months. Mainframe sales had dropped 43 percent during the same four months. Other large IBM businesses—software, maintenance, and financing—were all dependent, for the most part, on mainframe sales and, thus, were declining as well.

    Gerstner expanded on this point in various sections of his book:

    Despite the fact that IBM, then and now, was regarded as a complex company with thousands of products…IBM was a one-product company—a mainframe company—with an array of multibillion-dollar businesses attached to that single franchise…It didn’t take a Harvard MBA or a McKinsey consultant to understand that the fate of the mainframe was the fate of IBM, and, at the time, both were sinking like stones.

    IBM’s mainframe business was being hammered on two fronts: Unix-based alternatives offered modular lower-cost alternatives for back-end operations, while PCs were taking over many of the jobs mainframes used to do — and, in the long run, threatening to take over the data center itself. IBM was not only stuck with a product that was too expensive for a market that was simultaneously shrinking in size, but also an entire organization predicated on that product’s dominance.

    This was Microsoft a few decades later: the company loved to brag about its stable of billion dollar businesses, but in truth they were all components of one business — Windows. Everything Microsoft built from servers to productivity applications was premised on the assumption that the vast majority of computing devices were running Windows, leaving the company completely out of sorts when the iPhone and Android created and captured the smartphone market.

    The truth is that both companies were victims of their own monopolistic success: Windows, like the System/360 before it, was a platform that enabled Microsoft to make money in all directions. Both companies made money on the device itself and by selling many of the most important apps (and in the case of Microsoft, back-room services) that ran on it. There was no need to distinguish between a vertical strategy, in which apps and services served to differentiate the device, or a horizontal one, in which the device served to provide access to apps and services. When you are a monopoly, the answer to strategic choices can always be “Yes.”

    That, though, is why it is so interesting to think about what happens — and the problems that arise — when the monopoly ends.

    Post-Monopoly Problem One: Nature

    The great thing about a monopoly is that a company can do anything, because there is no competition; the bad thing is that when the monopoly is finished the company is still capable of doing anything at a mediocre level, but nothing at a high one because it has become fat and lazy. To put it another way, for a former monopoly “big” is the only truly differentiated asset.

    This was Gerstner’s key insight when it came to mapping out IBM’s future:

    I am not sure that in 1993 I or anyone else would have started out to create an IBM. But, given IBM’s scale and broad-based capabilities, and the trajectories of the information technology industry, it would have been insane to destroy its unique competitive advantage and turn IBM into a group of individual component suppliers more minnows in an ocean.

    In the big April customer meeting at Chantilly and in my other customer meetings, CIOs made it very clear that the last thing in the world they needed was one more disk drive company, one more operating system company, one more PC company. They also made it clear that our ability to execute against an integrator strategy was nearly bankrupt and that much had to be done before IBM could provide a kind of value that we were not providing at the time—but which they believed only IBM had a shot at delivering: genuine problem solving, the ability to apply complex technologies to solve business challenges, and integration.

    So keeping IBM together was the first strategic decision, and, I believe, the most important decision I ever made—not just at IBM, but in my entire business career. I didn’t know then exactly how we were going to deliver on the potential of that unified enterprise, but I knew that if IBM could serve as the foremost integrator of technologies, we’d be delivering extraordinary value.

    In Gerstner’s vision, only IBM had the breadth to deliver solutions instead of products; the next challenge would be changing the business model.

    Post-Monopoly Problem Two: Business Model

    The natural inclination for former monopolies, at least if Microsoft and IBM are any indication, is to stick with the monopoly-era business model. That meant doubling down on the device (or OS, as it were).

    The problem with this approach is twofold:

    • First, as I just noted, the nature of the company is set: being big — which in this case means offering services to everyone — is much easier to accomplish than being better, a critical factor in selling a differentiated device in a competitive market.
    • Second, as long as the business-model is device-centric, there is a risk in destroying the services component of the business. In Microsoft’s case, that meant holding Office for iPad back to prop up Windows, for example, or building Azure (née Windows Azure) around Windows Server. IBM, in far more dire straights, was, as Gerstner noted, close to splitting up the company so that individual divisions could sell their respective devices on their own without corporate overhead.

    The reality is that while changing business models is hard, for both Microsoft and IBM it was necessary to preserve what strengths they still had. This is why defenders of former-CEO Steve Ballmer miss the point when pointing out that Microsoft Azure and Office 365, the keys to Microsoft’s renewed growth, both got started under his watch. Look again at Gerstner’s account of IBM:

    If you were to take a snapshot of IBM’s array of businesses in 1993 and another in 2002, you would at first see very few changes. Ten years ago we were in servers, software, services, PCs, storage, semiconductors, printers, and financing. We are still in those businesses today…

    My point is that all of the assets that the company needed to succeed were in place. But in every case—hardware, technology, software, even services—all of these capabilities were part of a business model that had fallen wildly out of step with marketplace realities.

    This is why I don’t give Ballmer too much credit for Office 365 and Azure: the products of Microsoft’s future were there, but the Windows-centric business model was constricting every part of the company to an ever-shrinking share of the overall market; Nadella’s greatest success has been taking off that straitjacket.1

    Post-Monopoly Problem Three: Culture

    Four years ago, while announcing a company-wide reorganization (that I thought was a bad idea), Ballmer wrote a memo called One Microsoft. This was the key paragraph:

    We will reshape how we interact with our customers, developers and key innovation partners, delivering a more coherent message and family of product offerings. The evangelism and business development team will drive partners across our integrated strategy and its execution. Our marketing, advertising and all our customer interaction will be designed to reflect one company with integrated approaches to our consumer and business marketplaces.

    I wrote in Services, Not Devices:

    The crux of the problem is in that paragraph: no one is asking Microsoft to design its “customer interaction” to “reflect one company.” Customers are asking Microsoft to help them solve their problems and get their jobs done, not to make them Microsoft-only customers. The solipsism is remarkable.

    The solipsism, at least if IBM was any indication, was also inevitable. Gerstner writes:

    When there’s little competitive threat, when high profit margins and a commanding market position are assumed, then the economic and market forces that other companies have to live or die by simply don’t apply. In that environment, what would you expect to happen? The company and its people lose touch with external realities, because what’s happening in the marketplace is essentially irrelevant to the success of the company…

    This hermetically sealed quality—an institutional viewpoint that anything important started inside the company—was, I believe, the root cause of many of our problems. To appreciate how widespread the dysfunction was, I need to describe briefly some of its manifestations. They included a general disinterest in customer needs, accompanied by a preoccupation with internal politics. There was general permission to stop projects dead in their tracks, a bureaucratic infrastructure that defended turf instead of promoting collaboration, and a management class that presided rather than acted. IBM even had a language all its own.

    Sounds familiar!

    Comic from Bonkers World

    Gerstner’s response was to restructure IBM, change the company’s promotion and compensation policies, and most importantly, push IBM to better understand customers and then leverage its size to offer services they actually needed:

    Our bet was this: Over the next decade, customers would increasingly value companies that could provide solutions— solutions that integrated technology from various suppliers and, more important, integrated technology into the processes of an enterprise. We bet that the historical preoccupations with chip speeds, software versions, proprietary systems, and the like would wane, and that over time the information technology industry would be services-led, not technology-led.

    This is why the Microsoft 365 announcement and Nadella’s talk of integration is so interesting, and IBM plays a role in this story as well.

    IBM’s Cloud Miss

    I’ve previously written about how IBM, specifically Sam Palmisano, who succeeded Gerstner as CEO, missed the cloud. Remarking on Palmisan’s declaration that “You can’t do what we’re doing in a cloud” I wrote:

    Something that is interesting about most cloud solutions is that few are really doing anything new. Rather cloud service providers are simply taking operations that were formerly done on premise and moving them to a cloud that is available for any enterprise to use. And, as Palmisano realized, the inherent lack of customization in such a model means that most cloud services are on a feature-by-feature basis inferior to on-premise software.

    The reality, though, is that the businesses IBM served — and the entire reason IBM had a market — didn’t buy customized technological solutions to make themselves feel good about themselves; they bought them because they helped them accomplish their business objectives. Gerstner’s key insight was that many companies had a problem that only IBM could solve, not that customized solutions were the end-all be-all. And so, as universally provided cloud services slowly but surely became good-enough, IBM no longer had a monopoly on problem solving.

    To put it bluntly, enterprises don’t need a systems integrator for their data center if they no longer have a data center. Once again IBM is stuck competing for a shrinking market, which is why the company’s revenue has now declined for 21 straight quarters.2

    Microsoft’s Cloud Opportunity

    Still, the fact that enterprises no longer have data centers doesn’t mean integration is no longer valuable; rather, the locus of needed integration has shifted to the cloud as well. The average enterprise customer uses 20~30 apps, data is often scattered on and off premise, or stuck in email or personal accounts, and while IT departments may be happy to no longer upgrade servers, managing identity and security across all of these services and on a whole host of new devices far more likely to be used outside a company’s intranet calls for the same sort of integrator Gerstner wanted IBM to be.

    This seems to be the long-term goal of Microsoft 365. Microsoft said in a blog post:

    [Microsoft 365] represents a fundamental shift in how we will design, build and go to market to address our customers’ needs for a modern workplace. The workplace is transforming—from changing employee expectations, to more diverse and globally distributed teams, to an increasingly complex threat landscape. From these trends, we are seeing a new culture of work emerging. Our customers are telling us they are looking to empower their people with innovative technology to embrace this modern culture of work.

    With more than 100 million commercial monthly active users of Office 365, and more than 500 million Windows 10 devices in use, Microsoft is in a unique position to help companies empower their employees, unlocking business growth and innovation…

    Microsoft 365 Enterprise:

    • Unlocks creativity by enabling people to work naturally with ink, voice and touch, all backed by tools that utilize AI and machine learning.
    • Provides the broadest and deepest set of apps and services with a universal toolkit for teamwork, giving people flexibility and choice in how they connect, share and communicate.
    • Simplifies IT by unifying management across users, devices, apps and services.
    • Helps safeguard customer data, company data and intellectual property with built-in, intelligent security.

    Wait, inking?

    Here’s the big concern I have about the Microsoft 365 rollout, and Microsoft generally: Nadella and team deserve plaudits for working through the first two post-monopoly problems. Microsoft has embraced its bigness and focused on services, and has the business model to match (although, it should be noted that it was Ballmer who was responsible for shifting most of Microsoft’s enterprise business to a subscription model years ago). That’s great!

    I’m troubled, though, that I just articulated what I think is the Microsoft 365 strategy — or what it should be — far more clearly than either Nadella or Kirk Koenigsbauer, the corporate vice-president for the Office team that wrote this blog post. Indeed, Gerstner articulated the strategy best of all, and he wasn’t even talking about Microsoft or the cloud!

    Then again, I’m not entirely sure a focus on cloud integration is Microsoft’s strategy after all: maybe the cynical take — that Microsoft is just stealing a successful name for yet another enterprise licensing bundle — is closer to the truth. It is striking that the primary reason Microsoft gives for Microsoft 365 is that it already has a lot of users.

    Stepping back even further, Nadella loves to say “Our customers tell us” or some derivative thereof, but an actual articulation of customer use cases is consistently missing from his presentations. This keynote was not dissimilar to Nadella’s Build keynote, which featured a full 30 minutes of theory about the future of computing, that, while fascinating, seemed much more like a justification for Microsoft’s continued relevance as opposed to an articulation of demonstrated customer needs.

    Can Culture Change?

    The most bittersweet paragraph in Who Says Elephants Can’t Dance? is the final one:

    I was always an outsider. But that was my job. I know Sam Palmisano has an opportunity to make the connections to the past as I could never do. His challenge will be to make them without going backward; to know that the centrifugal forces that drove IBM to be inward-looking and self-absorbed still lie powerful in the company. Continuing to drive change while building on the best (and only the best) of the past is the ultimate description of the job of Chief Executive Officer, International Business Machines Corporation.

    Palmisano completely failed the challenge: what was the aforementioned reliance on IBM’s seemingly impregnable position as a systems integrator and dismissal of the cloud anything but the result of being “inward-looking and self-absorbed”? The same point applies to Palmisano’s obsession with profit-per-share: customers, Gerstner’s obsession, were totally forgotten.

    That is why Gerstner’s IBM should be a inspiration to Microsoft, but Palmisano’s (and current CEO Ginni Rometty, who has hewed far more closely to Palmisano’s example than Gerstner’s) IBM a warning: culture is a curse, and for better or worse, a company can recover but never be fully cured.

    1. Update: A few folks have written in to note that I’m being a bit harsh on Ballmer. After all, not only did he start Azure, he fired the well-respected and successful head of the Server and Tools Business, Bob Muglia, because he wasn’t moving fast enough in the cloud. Muglia’s replacement? Satya Nadella ↩

    2. By the way, Gerstner predicted the public cloud in the first appendix of his book, which was published in 2003, four years before AWS was launched:

      Put all of this together—the emergence of large-scale computing grids, the development of autonomic technologies that will allow these systems to be more self-managing, and the proliferation of computing devices into the very fabric of life and business—and it suggests one more major development in the history of the IT industry. This one will change the way IT companies take their products to market. It will change who they sell to and who the customer considers its “supplier.” This development is what some have called “utility” computing.

      The essential idea is that very soon enterprises will get their information technology in much the same way they get water or electric power. They don’t now own a waterworks or power plant, and soon they’ll no longer have to buy, house, and maintain any aspect of a traditional computing environment: The processing, the storage, the applications, the systems management, and the security will all be provided over the Net as a service—on demand.

      The value proposition to customers is compelling: fewer assets; converting fixed costs to variable costs; access to unlimited computing resources on an as-needed basis; and the chance to shed the headaches of technology cycles, upgrades, maintenance, integration, and management.

      Also, in a post-September 11, 2001, world in which there’s much greater urgency about the security of information and systems, on-demand computing would provide access to an ultra-secure infrastructure and the ability to draw on systems that are dispersed— creating a new level of immunity from a natural disaster or an event that could wipe out a traditional, centralized data center.

      IBM misses him. ↩

  • Publishers and the Pursuit of the Past

    Editor’s Note: This article was originally published as a Daily Update for Stratechery members. To receive the Daily Update please subscribe here.

    David Chavern, the president and CEO of the News Media Alliance, a trade association representing 2,000 North American newspapers, has an op-ed in the Wall Street Journal that I’m going to take a longer-than-usual excerpt from:

    The rapid growth of digital connectivity has pushed demand for information to unprecedented heights. Never in history have so many people consumed so much news. This is a boon for democracy. Although reporting is often an irritant to those in power, high-quality news and analysis is essential to any political system that depends on giving citizens the facts so they can draw their own conclusions.

    The problem is that today’s internet distribution systems distort the flow of economic value derived from good reporting. Google and Facebook dominate web traffic and online ad income. Together, they account for more than 70% of the $73 billion spent each year on digital advertising, and they eat up most of the growth. Nearly 80% of all online referral traffic comes from Google and Facebook. This is an immensely profitable business. The net income of Google’s parent company, Alphabet, was $19 billion last year. Facebook’s was $10 billion.

    But the two digital giants don’t employ reporters: They don’t dig through public records to uncover corruption, send correspondents into war zones, or attend last night’s game to get the highlights. They expect an economically squeezed news industry to do that costly work for them.

    The only way publishers can address this inexorable threat is by banding together. If they open a unified front to negotiate with Google and Facebook—pushing for stronger intellectual-property protections, better support for subscription models and a fair share of revenue and data—they could build a more sustainable future for the news business.

    But antitrust laws make such coordination perilous. These laws, intended to prevent monopolies, are having the unintended effect of preserving and protecting Google and Facebook’s dominant position. The digital giants benefit from legal precedent against collective action that has a chilling effect on publishers. Yet each newspaper or magazine on its own has only limited negotiating power.

    Chavern’s solution is twofold: one, that regulators should do a better job of enforcing antitrust laws against Google and Facebook, and two, that Congress should grant publishers a safe harbor to negotiate collectively with Google and Facebook.

    That a safe harbor is necessary to negotiate collectively was made apparent in the Apple ebooks case; while reasonable people can disagree as to whether or not Apple was rightly found to be guilty, no one disputes that the publishers colluded to raise prices, a per se violation of the Sherman Antitrust Act.1

    Aggregation Theory Redux

    That publishers feel the need to pursue the same strategy is driven by the reality of Aggregation Theory. To quickly recap, in a world of abundance, as long as the aggregator controls demand, suppliers have no choice but to commoditize themselves according to the strictures of the aggregator, which is another way of saying that suppliers have no choice but to engage in perfect competition. This is extremely ruinous for publishers in particular:

    • In a perfect market, the price of an undifferentiated commodity will be its marginal cost
    • The marginal cost is the cost to produce one more item (as opposed to the total cost, which includes the fixed costs necessary to create the item; those fixed costs, though, are already spent whether or not an additional item is created)2
    • The marginal cost of a digital item is zero, which means in a perfect market the inevitable price of a digital item is zero

    In short, aggregators are market makers, and the markets they make — thanks to the aforementioned lack of distribution costs and transactions costs — are pretty darn close to perfect, particularly in the case of purely digital goods. The problem is that publishers have huge fixed costs, even if you ignore old world infrastructure like printing presses and delivery trucks; specifically, publishers have to pay salaries, but those costs, by virtue of being fixed, not marginal, have zero impact on a publication’s pricing power in a perfect market.3

    The end result is dire for newspapers in particular: in commodity markets the winning companies have superior cost structures, which means they can sustainably sell at the market-clearing price; newspapers, though, by virtue of being built for a world of print, will never have the cost structure or mentality to succeed in the long run.

    Thus this solution: Chavern and the big publishers want permission from Congress to escape the perfect competition fostered by Aggregation Theory via collusion. The theory seems to be that, were the 2,000 newspapers party to this proposal able to present a unified front, they could force concessions from Google and Facebook that would make their businesses viable.

    News Versus Advertising

    There’s just one problem with this analysis — and that problem extends to Chavern’s proposal as a whole: it’s based on a myth. Specifically, newspapers have never succeeded by selling news, a point I made explicitly in The Local News Business Model:

    By owning printing presses and delivery trucks (and thanks to the low marginal cost of printing extra pages), newspapers were the primary outlet for advertising that didn’t work on (or couldn’t afford) TV or radio — and there was a lot of it. Maximizing advertising, though, meant maximizing the potential audience, which meant offering all kinds of different types of content in volume: thus the mashup of wildly disparate content listed above, all focused on quantity over quality. And then, having achieved the most readership and the ability to expand to fit it all, the biggest newspaper could squeeze out its competitors.

    Too many newspaper advocates utterly and completely fail to understand this; the truth is that newspapers made money in the past not by providing societal value, but by having quasi-monopolistic control of print advertising in their geographic area; the societal value was a bonus. Thus, when Chavern complains that “today’s internet distribution systems distort the flow of economic value derived from good reporting”, he is in fact conflating societal value with economic value; the latter does not exist and has never existed.

    This failure to understand the past leads to a misdiagnosis of the present: Google and Facebook are not profitable because they took newspapers’ reporting, they are profitable because they took their advertising. Moreover, the utility of both platforms is so great that even if all newspaper content were magically removed — which has been tried in Europe — the only thing that would change is that said newspapers would lose even more revenue as they lost traffic.

    This is why this solution is so misplaced: newspapers no longer have a monopoly on advertising, can never compete with the Internet when it comes to bundling content, and news remains both valuable to society and, for the same reasons, worthless economically (reaching lots of people is inversely correlated to extracting value, and facts — both real and fake ones — spread for free).

    A Better Solution for Publishers

    I just said that “good reporting” has never had economic value; this wasn’t quite right. Rather, the articles that result from “good reporting” don’t have economic value: once published on the Internet they have zero marginal cost in a world of perfect competition. This is why publishers have to shift their mindset about their product and their market.

    Specifically, publishers should be selling “good reporting”, that is, the commitment to the regular production of content that the buyer would like to see. This is a slight distinction but a critical one: successful subscription products do not sell content but rather the production of the content, and that production, unlike the articles themselves, can be differentiated and sold as a scarce product. That, though, means knowing who the buyer is: it’s not advertisers, but rather readers.

    Moreover, this approach addresses the Facebook and Google problem: the issue with Aggregation Theory from a supplier perspective is that the aggregator owns the consumer relationship; however, because Facebook and Google are advertising companies, they are not even competing in the subscription market. They, like advertisers, only care about content that has already been produced, not how it came to be.

    In fact, this is the single most ridiculous part of this proposal: one of the issues Chavern wishes to collectively bargain with Facebook and Google about is “better support for subscription models”. In other words, Chavern wishes to bring in Facebook and Google as an aggregator in the one market — subscriptions — where newspapers actually have a viable business model.

    It’s easy to envision how this could play out: Google and Facebook set up subscription offerings for publishers, eventually create the bundle of the future, and, by virtue of owning the consumer, skim off most of the profits, leaving publishers desperately pursuing page views to get their minuscule share of revenue. Sound familiar?

    The fundamental issue is this: there is a business model that works for publishers, but it requires a dramatic shift in mindset and the long hard slog of building a business. That means understanding what customers want, building a product that appeals to them, reaching them, moving them down a sales funnel, and retaining them. What it does not mean is the suffocating sense of entitlement and delusion that underlies not just this proposal but the majority of commentary from newspapers themselves that expects someone — anyone! — to give journalists money simply because they are important.

    That’s the thing: journalism is important. It is so important that the sooner publishers let go of a long-gone world where publications earned advertising simply by existing, and actually build publications that can not just survive but thrive on the Internet, the better off society will be. And, in that fight, this move by the News Media Alliance is actively hurting the cause.

    1. Apple was ruled to have been a co-conspirator which meant they were per se guilty; the company argued their relationship with the publishers was a vertical one, which would have led to a rule of reason analysis that I suspect would have exonerated the company. I wrote more about the case here  ↩

    2. Technically all costs, at least in the long-run, are marginal costs; however, in the short-run, the determination of whether or not to produce one more item is based purely on the cost of that item alone; a useful overview of the difference is here  ↩

    3. Salaries are fixed costs in terms of producing one more article; they are much closer to variable costs, though, than something like a printing press ↩

  • Ends, Means, and Antitrust

    The European Commission levied a record €2.42 billion ($2.73 billion) fine on Google yesterday for having “abused its market dominance as a search engine by giving an illegal advantage to another Google product, its comparison shopping service.” Commissioner Margrethe Vestager said in a press release announcing the decision:

    “Google has come up with many innovative products and services that have made a difference to our lives. That’s a good thing. But Google’s strategy for its comparison shopping service wasn’t just about attracting customers by making its product better than those of its rivals. Instead, Google abused its market dominance as a search engine by promoting its own comparison shopping service in its search results, and demoting those of competitors.

    “What Google has done is illegal under EU antitrust rules. It denied other companies the chance to compete on the merits and to innovate. And most importantly, it denied European consumers a genuine choice of services and the full benefits of innovation.”

    It is tempting when these decisions come down to start with the ends: specifically, does the outcome in question agree with one’s pre-existing views on such matters as regulation generally, antitrust specifically, and even nationalism (or continentalism, as it were)? The means matter, though, especially in this decision: there are three meaningful questions in this case that cut to the heart of antitrust regulation of digital companies:

    • What is a digital monopoly?
    • What is the standard for determining illegal behavior?
    • What constitutes a competitive product?

    The European Commission’s decision was impressive on some of these questions, and very problematic on others; all sides of the antitrust debate should be wary of standing in resolute opposition or support.

    What is a Digital Monopoly?

    This is perhaps the most consequential aspect of this case, and I think the European Commission got it exactly right. Last year in Antitrust and Aggregation I explained why the unique dynamics of the Internet push towards dominant players that look very different from the monopolies of the past:

    Aggregation Theory is about how business works in a world with zero distribution costs and zero transaction costs; consumers are attracted to an aggregator through the delivery of a superior experience, which attracts modular suppliers, which improves the experience and thus attracts more consumers, and thus more suppliers in the aforementioned virtuous cycle. It is a phenomenon seen across industries including search (Google and web pages), feeds (Facebook and content), shopping (Amazon and retail goods), video (Netflix/YouTube and content creators), transportation (Uber/Didi and drivers), and lodging (Airbnb and rooms, Booking/Expedia and hotels).

    The first key antitrust implication of Aggregation Theory is that, thanks to these virtuous cycles, the big get bigger; indeed, all things being equal the equilibrium state in a market covered by Aggregation Theory is monopoly: one aggregator that has captured all of the consumers and all of the suppliers. This monopoly, though, is a lot different than the monopolies of yesteryear: aggregators aren’t limiting consumer choice by controlling supply (like oil) or distribution (like railroads) or infrastructure (like telephone wires); rather, consumers are self-selecting onto the Aggregator’s platform because it’s a better experience.

    That bit about self-selection is the most obvious reason to critique this decision: how can Google have a monopoly when users — over 90% of the population in most European countries, according to the European Commission’s Factsheet — could choose to use another search engine simply by typing a URL, or opening another app? The Factsheet has the answer:

    There are also high barriers to entry in these markets, in part because of network effects: the more consumers use a search engine, the more attractive it becomes to advertisers. The profits generated can then be used to attract even more consumers. Similarly, the data a search engine gathers about consumers can in turn be used to improve results.

    This is exactly right, and in my view a real breakthrough in antitrust regulation. In the physical world, limited by scarcity, economic power comes from controlling supply; in the digital world, overwhelmed by abundance, economic power comes from controlling demand, and that control stems from a virtuous cycle that, for the reasons explained in the excerpt above, accrues to the dominant player in a space. In other words, to note that end users could go elsewhere is to ignore the reality that users are not dummies, and that network effects are the foundation of digital monopolies.

    What is the Standard for Determining Illegal Behavior?

    The United States and European Union have, at least since the Reagan Administration, differed on this point: the U.S. is primarily concerned with consumer welfare, and the primary proxy is price. In other words, as long as prices do not increase — or even better, decrease — there is, by definition, no illegal behavior.

    The European Commission, on the other hand, is explicitly focused on competition: monopolistic behavior is presumed to be illegal if it restricts competitors which, in the theoretical long run, hurts consumers by restricting innovation. From the Factsheet:

    Market dominance is, as such, not illegal under EU antitrust rules. However, dominant companies have a special responsibility not to abuse their powerful market position by restricting competition, either in the market where they are dominant or in separate markets. Otherwise, there would be a risk that a company once dominant in one market (even if this resulted from competition on the merits) would be able to use this market power to cement/further expand its dominance, or leverage it into separate markets…

    As a result of Google’s illegal practices and the distortions to competition, Google’s comparison shopping service has made significant market share gains at the expense of rivals. This has deprived European consumers of the benefits of competition on the merits, namely genuine choice and innovation.

    The European Commission approach, relative to the U.S. when it comes to determining illegality, has both pluses and minuses: the good thing is that it is an approach that is actually applicable to digital markets, particularly those monetized by advertising. Given the fact that Google is free for consumers, it is basically all but impossible for the company to be found guilty of antitrust behavior by the U.S., as the FTC determined a few years ago. The truth is that proxies are always problematic, including price, and there is no better example than the absurdity of the U.S. Justice Department successfully suing Apple for building a competitor to Amazon, the actual e-book monopolist.

    On the other hand, isn’t consumer welfare the entire point? Sure, a narrow focus on price is perhaps a bad proxy, but if dominant services are winning by being better — which is my argument in Aggregation Theory — why should regulators busy themselves with demanding worse alternatives be given the right to succeed?

    This is a point where many of those focused on the antitrust ends go wrong: in their crusade against big companies, they fail to grapple with the reality that on the Internet being big comes from being the best, leaving their arguments vulnerable to the critique that they are, in fact, anti-consumer, or at a minimum, anti-competence. To put it another way, in contrast to the previous point, regulators are treating people like dummies, assuming they can’t figure out how to find a competitive service, when in fact the truth is they don’t want to.

    What Constitutes a Competitive Product?

    This is by far the most concerning part of the European Commission’s decision, for two reasons.

    First, if I search for a specific product, why would I not want to be shown that specific product? It frankly seems bizarre to argue that I would prefer to see links to shopping comparison sites; if that is what I wanted I would search for “Shopping Comparison Sites”, a request that Google is more than happy to fulfill:

    Screen Shot 2017-06-28 at 6.40.22 PM

    The European Commission is effectively arguing that Google is wrong by virtue of fulfilling my search request explicitly; apparently they should read my mind and serve up an answer (a shopping comparison site) that is in fact different from what I am requesting (a product)?

    The second reason is even more problematic: “Google Shopping” is not actually a search product; it is an ad placement:

    Screen Shot 2017-06-28 at 6.56.13 PM

    You can certainly argue that the tiny “Sponsored” label is bordering on dishonesty, but the fact remains that Google is being explicit about the fact that Google Shopping is a glorified ad unit. Does the European Commission honestly have a problem with that? The entire point of search advertising is to have the opportunity to put a result that might not otherwise rank in front of a user who has demonstrated intent.

    The implications of saying this is monopolistic behavior goes to the very heart of Google’s business model: should Google not be allowed to sell advertising against search results for fear that it is ruining competition? Take travel sites: why shouldn’t Priceline sue Google for featuring ads for hotel booking sites above its own results? Why should Google be able to make any money at all?

    This is the aspect of the European Commission’s decision that I have the biggest problem with. I agree that Google has a monopoly in search, but as the Commission itself notes that is not a crime; the reality of this ruling, though, is that making any money off that monopoly apparently is. And, by extension, those that blindly support this decision are agreeing that products that succeed by being better for users ought not be able to make money.

    Long-time readers of Stratechery know that I have shifted my position on antitrust over time. At the end of the day, I tend to agree with the European Union that competition is an end worth pursuing in and of itself, and that antitrust regulation is fundamentally different from the sort of red tape that limits entrepreneurship. Indeed, it is directly opposed: red tape regulation entrenches incumbents and limits new entrants, while antitrust regulation limits incumbents and enables new entrants.

    Moreover, I believe that Google has been a bad actor: the company’s scraping of content from Yelp, TripAdvisor, and Amazon was clearly anticompetitive, and a much better example of illegally favoring Google’s results over superior competition. And, more broadly, I started writing about the Google-Facebook duopoly in online advertising earlier than most; I have been building the case that that is where the problems of monopoly might most clearly be seen.

    In short, I agree with the ends as far as the European Commission’s ruling is concerned: Google is a monopoly, and they act badly. In this case, though, I simply can not tolerate the means: I can not see any compelling case in which consumer welfare is better served by offering answers they didn’t actually ask for, and attacking an ad unit feels a lot more like an attempt to hurt a company as opposed to helping competition.

    More broadly, antitrust advocates have to appreciate that, when it comes to digital monopolies, there is a very fine line to walk between opposing products that are better for consumers and promoting competition: I do think competition is ultimately pro-consumer, but simply presuming that “big” is bad when “big” comes from a superior customer experience is little more than a shortcut to political irrelevance.1

    1. In other words, wait for the Android case! More on the long-term repurcussions for Google tomorrow in the Daily Update  ↩

  • Amazon’s New Customer

    Back in 2006, when the iPhone was a mere rumor, Palm CEO Ed Colligan was asked if he was worried:

    “We’ve learned and struggled for a few years here figuring out how to make a decent phone,” he said. “PC guys are not going to just figure this out. They’re not going to just walk in.” What if Steve Jobs’ company did bring an iPod phone to market? Well, it would probably use WiFi technology and could be distributed through the Apple stores and not the carriers like Verizon or Cingular, Colligan theorized.

    I was reminded of this quote after Amazon announced an agreement to buy Whole Foods for $13.7 billion; after all, it was only two years ago that Whole Foods founder and CEO John Mackey predicted that groceries would be Amazon’s Waterloo. And while Colligan’s prediction was far worse — Apple simply left Palm in the dust, unable to compete — it is Mackey who has to call Amazon founder and CEO Jeff Bezos, the Napoleon of this little morality play, boss.

    The similarities go deeper, though: both Colligan and Mackey made the same analytical mistakes: they mis-understood their opponent’s goals, strategies, and tactics. This is particularly easy to grok in the case of Colligan and the iPhone: Apple’s goal was not to build a phone but to build an even more personal computer; their strategy was not to add on functionality to a phone but to reduce the phone to an app; and their tactics were not to duplicate the carriers but to leverage their connection with customers to gain concessions from them.

    Mackey’s misunderstanding was more subtle, and more profound: while the iPhone may be the most successful product of all time, Amazon and Jeff Bezos have their sights set on being the most dominant company of all time. Start there, and this purchase makes all kinds of sense.

    Amazon’s Goal

    If you don’t understand a company’s goals, how can you know what its strategies and tactics will be? Unfortunately, many companies, particularly the most ambitious, aren’t as explicit as you might like. In the case of Amazon, the company stated in its 1997 S-1:

    Amazon.com’s objective is to be the leading online retailer of information-based products and services, with an initial focus on books.

    Even if you picked up on the fact that books were only step one (which most people at the time did not), it was hard to imagine just how all-encompassing Amazon.com would soon become; within a few years Amazon’s updated mission statement reflected the reality of the company’s e-commerce ambitions:

    Our vision is to be earth’s most customer centric company; to build a place where people can come to find and discover anything they might want to buy online.

    “Anything they might want to buy online” was pretty broad; the advent of Amazon Web Services a few years later showed it wasn’t broad enough, and a few years ago Amazon reduced its stated goal to just that first clause: We seek to be Earth’s most customer-centric company. There are no more bounds, and I don’t think that is an accident. As I put it on a podcast a few months ago, Amazon’s goal is to take a cut of all economic activity.

    This, then, is the mistake Mackey made: while he rightly understood that Amazon was going to do everything possible to win in groceries — the category accounts for about 20% of consumer spending — he presumed that the effort would be limited to e-commerce. E-commerce, though, is a tactic; indeed, when it comes to Amazon’s current approach, it doesn’t even rise to strategy.

    Amazon’s Strategy

    As you might expect, given a goal as audacious as “taking a cut of all economic activity”, Amazon has several different strategies. The key to the enterprise is AWS: if it is better to build an Internet-enabled business on the public cloud, and if all businesses will soon be Internet-enabled businesses, it follows that AWS is well-placed to take a cut of all business activity.

    On the consumer side the key is Prime. While Amazon has long pursued a dominant strategy in retail — superior cost and superior selection — it is difficult to build sustainable differentiation on these factors alone. After all, another retailer is only a click away.

    This, though, is the brilliance of Prime: thanks to its reliability and convenience (two days shipping, sometimes faster!), plus human fallibility when it comes to considering sunk costs (you’ve already paid $99!), why even bother looking anywhere else? With Prime Amazon has created a powerful moat around consumer goods that does not depend on simply having the lowest price, because Prime customers don’t even bother to check.

    This, though, is why groceries is a strategic hole: not only is it the largest retail category, it is the most persistent opportunity for other retailers to gain access to Prime members and remind them there are alternatives. That is why Amazon has been so determined in the space: AmazonFresh launched a decade ago, and unlike other Amazon experiments, has continued to receive funding along with other rumored initiatives like convenience store and grocery pick-ups. Amazon simply hasn’t been able to figure out the right tactics.

    Amazon’s Tactics

    To understand why groceries are such a challenge look at how they differ from books, Amazon’s first product:

    • There are far more books than can ever fit in a physical store, which means an e-commerce site can win on selection; in comparison, there simply aren’t that many grocery items (a typical grocery store will have between 30,000 and 50,000 SKUs)
    • When you order a book, you know exactly what you are getting: a book from Amazon is the same as a book from a local bookstore; groceries, on the other hand, can vary in quality not just store-to-store but, particularly in the case of perishable goods, day-to-day and item-to-item
    • Books can be stored in a centralized warehouse indefinitely; perishable groceries can only be stored for a limited amount of time and degrade in quality during transit

    As Mackey surely understood, this meant that AmazonFresh was at a cost disadvantage to physical grocers as well: in order to be competitive AmazonFresh needed to stock a lot of perishable items; however, as long as AmazonFresh was not operating at meaningful scale a huge number of those perishable items would spoil. And, given the inherent local nature of groceries, scale needed to be achieved not on a national basis but a city one.

    Groceries is a fundamentally different problem that needs a fundamentally different solution; what is so brilliant about this deal, though, is that it solves the problem in a fundamentally Amazonian way.

    The First-And-Best Customer

    Last year in The Amazon Tax I explained how the different parts of the company — like AWS and Prime — were on a conceptual level more similar than you might think, and that said concepts were rooted in the very structure of Amazon itself. The best example is AWS, which offered server functionality as “primitives”, giving maximum flexibility for developers to build on top of:1

    The “primitives” model modularized Amazon’s infrastructure, effectively transforming raw data center components into storage, computing, databases, etc. which could be used on an ad-hoc basis not only by Amazon’s internal teams but also outside developers:

    stratechery Year One - 274

    This AWS layer in the middle has several key characteristics:

    • AWS has massive fixed costs but benefits tremendously from economies of scale
    • The cost to build AWS was justified because the first and best customer is Amazon’s e-commerce business
    • AWS’s focus on “primitives” meant it could be sold as-is to developers beyond Amazon, increasing the returns to scale and, by extension, deepening AWS’ moat

    This last point was a win-win: developers would have access to enterprise-level computing resources with zero up-front investment; Amazon, meanwhile, would get that much more scale for a set of products for which they would be the first and best customer.

    As I detailed in that article, this exact same framework applies to Amazon.com:

    Prime is a super experience with superior prices and superior selection, and it too feeds into a scale play. The result is a business that looks like this:

    stratechery Year One - 275

    That is, of course, the same structure as AWS — and it shares similar characteristics:

    • E-commerce distribution has massive fixed costs but benefits tremendously from economies of scale
    • The cost to build-out Amazon’s fulfillment centers was justified because the first and best customer is Amazon’s e-commerce business
    • That last bullet point may seem odd, but in fact 40% of Amazon’s sales (on a unit basis) are sold by 3rd-party merchants; most of these merchants leverage Fulfilled-by-Amazon, which means their goods are stored in Amazon’s fulfillment centers and covered by Prime. This increases the return to scale for Amazon’s fulfillment centers, increases the value of Prime, and deepens Amazon’s moat

    As I noted in that piece, you can see the outline of similar efforts in logistics: Amazon is building out a delivery network with itself as the first-and-best customer; in the long run it seems obvious said logistics services will be exposed as a platform.

    This, though, is what was missing from Amazon’s grocery efforts: there was no first-and-best customer. Absent that, and given all the limitations of groceries, AmazonFresh was doomed to be eternally sub-scale.

    Whole Foods: Customer, not Retailer

    This is the key to understanding the purchase of Whole Foods: from the outside it may seem that Amazon is buying a retailer. The truth, though, is that Amazon is buying a customer — the first-and-best customer that will instantly bring its grocery efforts to scale.

    Today, all of the logistics that go into a Whole Foods store are for the purpose of stocking physical shelves: the entire operation is integrated. What I expect Amazon to do over the next few years is transform the Whole Foods supply chain into a service architecture based on primitives: meat, fruit, vegetables, baked goods, non-perishables (Whole Foods’ outsized reliance on store brands is something that I’m sure was very attractive to Amazon). What will make this massive investment worth it, though, is that there will be a guaranteed customer: Whole Foods Markets.

    stratechery Year One - 270

    In the long run, physical grocery stores will be only one of Amazon Grocery Services’ customers: obviously a home delivery service will be another, and it will be far more efficient than a company like Instacart trying to layer on top of Whole Foods’ current integrated model.

    I suspect Amazon’s ambitions stretch further, though: Amazon Grocery Services will be well-placed to start supplying restaurants too, gaining Amazon access to another big cut of economic activity. It is the AWS model, which is to say it is the Amazon model, but like AWS, the key to profitability is having a first-and-best customer able to utilize the massive investment necessary to build the service out in the first place.

    I said at the beginning that Mackey misunderstood Amazon’s goals, strategies, and tactics, and while that is true, the bigger error was in misunderstanding Amazon itself: unlike Whole Foods Amazon has no particular desire to be a grocer, and contrary to conventional wisdom the company is not even a retailer. At its core Amazon is a services provider enabled — and protected — by scale.

    Indeed, to the extent Waterloo is a valid analogy, Amazon is much more akin to the British Empire, and there is now one less obstacle to sitting astride all aspects of the economy.

    1. To be clear, AWS was not about selling extra capacity; it was new capability, and Amazon itself has slowly transitioned over time (as I understand it Amazon.com is still a hybrid)  ↩

  • Podcasts, Analytics, and Centralization

    Tucked into the last day of WWDC was a session on podcasting, and it contained some big news for the burgeoning industry. Before getting into the specific announcements, though, the session itself is worth a bit of analysis, particularly the opening from Apple Podcasts Business Manager James Boggs:

    First we want to talk for a moment about how we think about modern podcasts. Long-form and audio. We get excited about episodic content that entertains, informs, and inspires. We get excited and many of our users have gotten excited too.

    I went on to transcribe the next 500 or so words of Boggs’s presentation, which included various statistics on downloads, catalog size, and reach; a listing of Apple “partners” organized by media and broadcast organizations, public media, and independents; and even started in on Boggs’s review/promotion of individual podcasts like “Up and Vanished” and “Masters of Scale” before I realized Boggs was never going to actually say “how [Apple] think[s] about modern podcasts.” I won’t make you read the transcript — take my word when I say that there was nothing there.

    Still, that itself was telling; Boggs’s presentation perfectly reflects the state of podcasting today: Apple is an essential piece, even as they really don’t have anything to do with what is going on (but naturally, are happy to take credit).

    A Brief History of Podcasts

    Probably the first modern podcast was created by Dave Winer in 2003, although it wasn’t called a “podcast”: that was coined by Ben Hammersley in 2004, and the inspiration was Apple’s iPod. Still, while the media had a name, the “industry”, such that it was, was very much the wild west: a scattering of podcast creators, podcatchers (software for downloading the podcasts), and podcast listeners, finding each other by word-of-mouth.

    stratechery Year One - 267

    A year later Apple made the move that cemented their current position as the accidental gorilla of the industry: iTunes 4.9 included support for podcasts and, crucially, the iTunes Music Store created a directory (Apple did not — and still does not — host the podcast files themselves). The landscape of podcasting was completely transformed:

    stratechery Year One - 268

    Centralization occurs in industry after industry for a reason: everyone benefits, at least in the short term. Start with the users: before iTunes 4.9 subscribing and listening to a podcast was a multi-step process, and most of those steps were so obscure as to be effective barriers for all but the most committed of listeners.

    • Find a podcast
    • Get a podcatcher
    • Copy the URL of the podcast feed into the podcatcher
    • Copy over the audio file from the podcatcher into iTunes
    • Sync the audio file to an iPod
    • Listen to the podcast
    • Delete the podcast from the iPod the next time you sync’d

    iTunes 4.9 made this far simpler:

    • Find a podcast in the iTunes Store and click ‘Subscribe’
    • Sync your iPod
    • Listen

    Recounting this simplification may seem pedantic, but there is a point: this was the most important improvement for podcast creators as well. Yes, the iTunes Music Store offered an important new discovery mechanism, but it was the dramatic improvement to the user experience that, for the vast majority of would-be listeners, made podcasts even worth discovering in the first place. Centralized platforms win because they make things easier for the user; producers willingly follow.

    Interestingly, though, beyond that initial release, which was clearly geared towards selling more iPods, Apple largely left the market alone, with one important exception: in 2012 the company released a standalone Podcasts app for iOS in the App Store, and in 2014 the app was built-in to iOS 8. At that point the power of defaults did its job: according to the IAB Podcast Ad Metrics Guidelines released last fall, the Apple Podcast App accounts for around 50% of all podcast players across all operating systems (iTunes is a further ~10%).1

    The Business of Podcasting

    It’s not clear when the first podcast advertisement was recorded; a decent guess is Episode 67 of This Week in Tech, recorded on September 3, 2006 (Topic: “Does the Google CEO’s place on Apple’s board presage a Sun merger?”). The sponsor was surprisingly familiar — Visa (“Safer, better money. Life takes Visa.”), and Dell joined a week later.

    Over the ensuing years, though, the typical podcast sponsor was a bit less of a name brand — unless, of course, you were a regular podcast listener, in which case you quickly knew the brands by heart: Squarespace, Audible, Casper Mattress, Blue Apron, and recent favorite MeUndies (because who doesn’t want to hear a host-read endorsement for underwear!). Companies like Visa or Dell were few and far between: a study by FiveThirtyEight suggested brand advertisers were less than five percent of ad reads.

    The reason is quite straightforward: for podcasts there is neither data nor scale. The data part is obvious: while podcasters can (self-)report download numbers, no one knows whether or not a podcast is played, or if the ads are skipped. The scale bit is more subtle: podcasts are both too small and too big. They are too small in that it is difficult to buy ads at scale (and there is virtually no quality control, even with centralized ad sellers like Midroll); they are too large in that the audience, which may be located anywhere in the world listening at any time, is impossible to survey in order to measure ad effectiveness.

    That is why the vast majority of podcast advertisers are actually quite similar: nearly all are transaction-initiated subscription-based services. The “transaction-initiated” bit means that there is a discrete point at which the customer can indicate where they heard about the product, usually through a special URL, while the “subscription-based” part means these products are evaluating their marketing spend relative to expected lifetime value. In other words, the only products that find podcast advertising worthwhile are those that expect to convert a listener in a measurable way and make a significant amount of money off of them, justifying the hassle.2

    The result is an industry that, from a monetization perspective, looks a lot like podcasting before iTunes 4.9; there are small businesses to be built, but the industry as a whole is stunted.

    Apple Podcast Analytics

    This is the context for what Apple actually announced. Jason Snell had a good summary at Six Colors:

    New extensions to Apple’s podcast feed specification will allow podcasts to define individual seasons and explain whether an episode is a teaser, a full episode, or bonus content. These extensions will be read by the Podcast app and used to present a podcast in a richer way than the current, more linear, approach…

    The other big news out of today’s session is for podcasters (and presumably for podcast advertisers): Apple is opening up in-episode analytics of podcasts. For the most part, podcasters only really know when an episode’s MP3 file is downloaded. Beyond that, we can’t really tell if anyone listens to an episode, or how long they listen—only the apps know for sure. Apple said today that it will be using (anonymized) data from the app to show podcasters how many people are listening and where in the app people are stopping or skipping. This has the potential to dramatically change our perception of how many people really listen to a show, and how many people skip ads, as well as how long a podcast can run before people just give up.

    The new extensions are a nice addition, and a way in which Apple can enhance the user experience to the benefit of everyone. As you might expect, though, I’m particularly interested in the news about analytics. Problem solved, right? Or is it problem caused? What happens when advertisers realize that everyone is skipping their ads?

    Advertisers: Not Idiots

    In fact, I expect these analytics to have minimal impact, at least in the short run. For one, every indication is that analytics will only be available to the podcast publishers, although certainly advertisers will push to have them shared.3 More pertinently, though, all of the current podcast advertisers know exactly what they are getting: X amount of podcast ads results in Y number of conversions that result in Z amount of lifetime value.

    Indeed, contrary to what many folks seem to believe, advertisers, whether they leverage podcasts, Facebook, Google, or old school formats like radio or TV, are not idiots blindly throwing money over a wall in the vague hopes that it will drive revenue, ever susceptible to being shocked, shocked! that their ads are being ignored. Particularly in the case of digital formats advertisers are quite sophisticated, basing advertising decisions off of well-known ROI calculations. That is certainly the case with podcasts: knowing to a higher degree of precision how many ads are skipped doesn’t change the calculation for the current crop of podcast advertisers in the slightest.

    What more data does do is open the door to more varied types of advertisers beyond the subscription services that dominate the space. Brand advertisers, in particular, are more worried about reaching a guaranteed number of potential customers than they are tracking directly to conversion, and Apple’s analytics will help podcasters tell a more convincing story in that regard.

    In truth, though, Apple’s proposed analytics aren’t nearly enough: advertisers still won’t know who they are reaching or where they are located, and while brand advertisers may not have the expectation of tracking-to-purchase no one wants to throw money to the wind either. The problem of surveying effectively to measure things like brand lift is as acute as ever, and it simply isn’t worth the trouble to do a bunch of relatively small media buys with zero quality control.

    Apple’s Opportunity

    This, though, is why Apple’s centralized role is so intriguing. Remember, the web was thought to be a wasteland for advertising until Google provided a centralized point that aggregated users and could be sold to advertisers. Similarly, mobile was thought to monetize even worse than the (desktop) web until Facebook provided a centralized point that aggregated users and could be sold to advertisers. I expect a similar dynamic in podcasts: the industry will remain the province of web hosting and underwear absent centralization and aggregation, and the only entity that can accomplish that is Apple.

    One can envision the broad outlines of what the business for a centralized aggregator for podcasts might look like:

    • The centralized aggregator would likely offer hosting to podcast creators, not only to secure the user experience and get better analytics (including on downloads through other apps) but also to dynamically insert advertisements. Those advertisements would also be available to smaller podcasts that are currently not worth the effort to advertisers.
    • Advertisers would get their own dashboard for those analytics and, more importantly, the opportunity to buy ads at far greater scale across a large enough audience to make it worth their while. Ideally, at least from their perspective, they would actually be able to target their advertising buys as well.
    • Users would, at least in theory, benefit from a far broader array of content made possible by the growth in revenue for the industry broadly.

    There are already companies trying to do just this: I wrote about E.W. Scripps’ Midroll and their acquisition of podcast player Stitcher last year. The problem is that Stitcher only has around 5% of listeners, and it is the ownership of users/listeners, not producers/podcast from which true market power derives. Apple has that ownership, and thus that power; the question is will they use it?

    Surely the safe bet is “no”. iAd, Apple’s previous effort at building an advertising business, failed spectacularly, and Apple’s anti-advertising rhetoric has only deepened since then. That’s a problem not only in terms of image but culture: Apple seems highly unlikely to be willing to put in the effort necessary to build a real advertising business, and given how small such a business might be even in the best-case scenario relative to the rest of the company, that’s understandable.4

    To be sure, should Apple decline to seize this opportunity it will be celebrated by many, particularly those doing well in the current ecosystem. Podcasting is definitely more open than not, with no real gatekeepers in terms of either distribution or monetization. That, though, is why the money is so small: gatekeepers are moneymakers, and while podcasts may continue to grow, it is by no means inevitable that, absent a more active Apple, the money will follow.

    Disclosure: Exponent, the podcast I host with James Allworth, does have a (single) sponsor; the revenue from this sponsorship makes up a very small percentage of Stratechery’s overall revenue and does not impact the views in this article

    1. For what it’s worth, Exponent has a much different profile: Apple Podcasts has about 13% share, while Overcast leads the way with 26% share, followed by (surprisingly!) Mobile Safari with 23% ↩

    2. This shows why Casper mattresses are the exception that proves the rule: mattresses are not a subscription service, but they are much more expensive than most products bought online, which achieves the same effect as far as lifetime value is concerned ↩

    3. I’m less worried about the fact other podcast players may not offer similar analytics: the Apple Podcast app will be used as a proxy, although this may hurt podcasts that have a smaller share of downloads via the Apple Podcast app (as total listeners may be undercounted absent similar analytics from other apps)  ↩

    4. It’s Google’s challenge in building a real hardware business in reverse ↩

  • Apple’s Strengths and Weaknesses

    The San Jose location of WWDC, Apple’s annual developer conference, felt a bit odd, but Apple sought to strike a familiar tone: the artwork on and around the San Jose McEnery Convention Center featured a top-down view of humans, and a familiar message:

    FullSizeRender

    The idea of Apple existing at the intersection of technology and liberal arts was central to the late Steve Jobs’ conception of Apple and, without question, a critical factor when it came to Apple’s success: at a time when technology was becoming accessible to consumers and their daily lives Apple created products — one product, really, the iPhone — that appealed to consumers not only because of what it did but how it did it.

    That said, it was telling that this artwork and the sentiment it signified were not referenced in the keynote itself; after a humorous skit about a world without apps, Tim Cook delivered platitudes about how Apple and its developers were on a “collective mission to change the world”, and immediately launched into what he said were six important announcements. It was not dissimilar to Sundar Pichai’s opening at Google I/O: when the announcements that matter are grounded on the realities of a company’s core competencies and position in the market, vision can feel extraneous.

    Apple’s Announcements

    Cook’s first four announcements spoke to those core capabilities and the position they afford Apple (or don’t, as the case may be) in the markets in which it competes:

    tvOS: It was generous of Cook to give tvOS top billing: the only announcement of note was the upcoming availability of Amazon Prime Video on Apple TV. That itself is a reminder of Apple’s diminished position in the space: winning in TV is not about hardware or software, much less the integration of the two, but rather content. The brevity of this announcement — there wasn’t even the traditional executive hand-off — spoke to Apple’s status as an also-ran.

    watchOS: This garnered more time, and the headline feature was the Siri watch face. The watch face, which implied a broadening of Siri’s brand from voice to context-based general assistant, seeks to anticipate and deliver the information you need when you need it. The model is Google Now; the difference is that Siri is now housed in an attractive and increasingly popular watch that works natively with an iPhone, while the equivalent Google service requires not simply a different watch but a different phone entirely. It is a testament to Apple’s biggest advantage: thanks to the iPhone the company already owns the “best” customers,1 frequently rendering moot Google’s superiority in managing information.

    macOS: This actually encompassed two of Cook’s six promised announcements;2 the separation of MacOS and Mac computers was, I suspect, born of Apple’s desire to convince developers and other pro users that the company was not abandoning their favorite platform. Moreover, the addition of hardware announcements, after several years in which WWDC was software only, resulted in a very different feel to this keynote: after all, hardware is exciting, even if, in the long run, it is software that actually matters. That feeling, though, goes to the very core of what Apple sells: superior hardware differentiated — and thus sold at a handsome margin — by exclusive software.

    As is always the case with the modern incarnation of Apple, though, the announcements that truly mattered centered around iOS.

    iOS 11

    The iOS-related announcements, despite being only one of Cook’s “Big Six”, could have been their own keynote; given the importance of mobile generally and iOS specifically that would have been more than justified. Taken as a whole the iOS segment in particular highlighted what Apple does best, what it struggles with, and what reasons there are to be both optimistic and pessimistic about the company’s fortunes in the long run.

    Strength: Defaults

    Controlling one of the two dominant mobile operating systems grants Apple the power of defaults. That means iMessage is both an iPhone lock-in and a channel to introduce new services like person-to-person Apple Pay. Siri can be accessed both via voice and the home button, and, just similar to the WatchOS update, is increasingly integrated throughout the operating system. Photos and Maps are used by the majority of iPhone customers, even if alternatives offer superior functionality.

    Weakness: Limited Reach

    At the same time, iMessage will never reach the dominance of a service like WeChat because it is limited to Apple’s own platforms — as it should be! iMessage is the canonical example of how strengths and weaknesses are two sides of the same coin: it is iMessage’s exclusivity that allows it to be a lock-in, and it is that same exclusivity that limits the standalone value.

    Strength: Hardware Integration

    Peppered throughout Apple’s presentation were seemingly small features like new compression algorithms that depend on Apple controlling everything from Messages to the camera to the processor that makes it all work. The most impressive example was ARKit: in one fell swoop Apple leaped ahead of the rest of the industry in the race to realize the promise of augmented reality. The contrast to Facebook was striking: while the social network is seeking to leverage its control of content distribution to lure developers to build on Facebook’s “camera”, Apple is not only offering the same opportunity (the results of which can, of course, be shared on Facebook or Instagram), but also delivering a superior set of APIs that, by virtue of being part of that vertical stack, are both more powerful and accessible than anything a 3rd-party application can deliver.

    Weakness: Services

    While Apple bragged about Siri’s natural language capabilities and alluded to a limited number of new “intents” that can be leveraged by apps, it is not an accident that there were no slides about accuracy, speed, or developer support: Siri is well behind the competition in all three. More fundamentally, all of Apple’s services are intrinsically limited by the fact that they exist to sell Apple hardware: those services, and the teams that work on them, will never be the most important people in the company, and their development will be constrained by the culture of Apple itself.

    Strength: Privacy

    Apple not only touted its privacy credentials, it also showed off new features to actively limit things like autoplaying videos and advertising networks that follow you across sites. As a user both are very welcome; strategically, both features follow from the fact that Apple makes money on its hardware, while companies like Google, Facebook, and other online businesses rely on advertising and the collection of data.

    Weakness: Data

    Collecting data is useful for more than advertising, though. Here Google is the obvious counter: certainly the search company wants to better target advertisements, but the benefits gained from data go far beyond overt monetization. It is data that drives Google’s superior machine learning capabilities and the customer-friendly features that follow in apps like Google Photos. Interestingly, Apple made moves in this direction, syncing things like facial recognition data and Messages across devices, favoring convenience over a very slight increase in the risk to privacy. To be clear, the data will still be encrypted, both in transit and at rest, but that is my point: encryption means that Apple cannot leverage the data it will now store to make its services better.

    Strength: The App Store

    The strategic role of 3rd-party apps has shifted over time: once a differentiator for iOS, Android has largely reached parity, and apps are now table stakes. They are also a big moneymaker: Apple has been pushing the narrative on Wall Street that it is a services company, fueled by the $30 billion the company has collected from app sales and especially in-app purchases in free-to-play games; 30% of that total has come in the last year alone. Make no mistake, this is a compelling narrative: iPhone growth may be slowing in the face of saturation and elongated update cycles, but that only means there is that large of a base from which to earn App Store revenue.

    Weakness: Developer Economics

    The success of free-to-play games and the associated in-app purchases has come at a cost, specifically, management blindness to the fact that the rest of the developer ecosystem isn’t nearly as healthy, and that the App Store is no longer a differentiator from Android. The fundamental problem remains that for productivity apps in particular it is necessary to monetize your best customers over time; Apple has improved the situation, particularly with the addition of subscription pricing and de facto trials (basically, starting a subscription at $0), but hasn’t made any moves to support trials or upgrade pricing for paid apps, despite the fact that is the proven successful model for productivity applications on the Mac. I have long argued that bad developer economics is the fundamental reason that the iPad hasn’t fulfilled its potential; yesterday’s iPad software enhancements were welcome and will help, but I suspect letting developers set their own business models would be even more transformative.

    Strength and Weakness: Business Model

    This point is part and parcel with all of the above: Apple’s strengths derive from the fact it sells software-differentiated hardware for a significant margin, which allows for exclusive apps and services set as defaults, deep integration from chipset to API, a focus on privacy, and total control of the developer ecosystem. And, on the flipside, Apple only reaches a segment of the market, is less incentivized and capable of delivering superior services, has less data, and can afford to take developers for granted.

    HomePod

    Apple’s final announcement encapsulated all of these tensions. The long-rumored competitor to Amazon Echo and Google Home was, fascinatingly, framed as anything but. Cook began the unveiling by referencing Apple’s longtime focus on music, and indeed, the first several minutes of the HomePod introduction were entirely about its quality as a speaker. It was, in my estimation, an incredibly smart approach: if you are losing the game, as Siri is to Alexa and Google, best to change the rules, and having heard the HomePod, its sound quality is significantly better than the Amazon Echo (and, one can safely assume, Google Home). Moreover, the ability to link multiple HomePods together is bad news for Sonos in particular (the HomePod sounded significantly better than the Sonos Play 3 as well).

    Of course, superior sound quality is what you would expect from a significantly more expensive speaker: the HomePod costs $350, while the Sonos Play 3 is $300, and the Amazon Echo is $150. From Apple’s perspective, though, a high price is a feature, not a bug: remember, the company has a hardware-based business model, which means there needs to be room for a meaningful margin. The Echo is the opposite: because it is a hardware means to the service ends that is Amazon, it can be priced with much lower margins and, as has already happened, be augmented with even cheaper devices like Echo Dots (or, in the case of the Echo Show, offer more functionality for a price that is still more than $100 cheaper than the HomePod).

    The result is a product that, beyond being massively late to market (in part because of iPhone-induced myopia), is inferior to the competition on two of three possible vectors: the HomePod is significantly more expensive than an Echo or Google Home, it has an inferior voice assistant, but it has a better speaker. That is not as bad as it sounds: after all, the iPhone is significantly more expensive than most other smartphones, it has inferior built-in services, but it has a superior user experience otherwise. The difference — and this is why the iPhone is so much more dominant than any other Apple product — is that everyone already needs a phone; the only question is which one. It remains to be seen how many people need a truly impressive speaker.

    This, broadly speaking, is the challenge for Apple moving forward: in what other categories does its business model (and everything that is tied up into that, including the company’s product development process, culture, etc.) create an advantage instead of a disadvantage? What existing needs can be met with a superior user experience, or what new needs — like the previously unknown need for wireless headphones that are always charged — can be created? To be clear, the iPhone is and will continue to be a juggernaut for a long time to come; indeed, it is so dominant that Apple could not change the underlying business model and resultant strengths and weaknesses even if they tried.

    1. Speaking strictly of which customers generate the most monetary value either through their purchases or advertising targeting ↩

    2. iPad hardware and optimizations all fell under the iOS umbrella ↩

  • Faceless Publishers

    When I first worked for a (student) newspaper, the job of a publisher seemed odd to me; as far as I and my editorial colleagues were concerned, the publisher was the person the editor-in-chief, who we viewed as the boss, occasionally griped about after a few too many drinks, usually with the assertion that he (in that case) was a bit of a nuisance.

    That attitude, of course, was the luxury of print: whatever happened on the other side of the office didn’t have any impact on the (in our eyes) heroic efforts to produce fresh content every day. We were the ones staying in the office until the wee hours of the night, writing, editing, and laying out the newspaper that would magically appear on newsstands the next morning, all while the publisher and his team were at home in bed.

    The moral of this story is obvious: the publisher represented the business side of the newspaper, and the effect of the Internet was to make the job and impact of editorial easier and that of a publisher immeasurably harder, in large part because many of a publisher’s jobs became obsolete; it is the editorial side, though, that has paid the price.

    The Jobs a Publisher Did

    In the days of print, publishers provided multiple interlocking functions that made newspapers into fabulous businesses:

    • Brand: A publisher had a brand, specifically, the name of the publication; this was the primary touchpoint for readers, whether they were interested in national news, local news, sports, or the funny pages.
    • Revenue Generation: Most publishers drove revenue in two ways: some money was made through subscriptions, the selling, administration, and support of which was handled by dedicated staff; most money was made from advertising, which had its own dedicated team.
    • Human Resources: Editorial staff were free to write and complain about their publishers because everything else in their work life was taken care of, from payroll to travel expenses to office supplies.

    What tied these functions together was distribution: a publisher owned printing presses and delivery trucks which, combined with their established readership and advertising relationships, gave most newspapers an effective monopoly (or oligopoly) in their geographic area on readers and advertisers and writers:

    stratechery Year One - 264

    Each of these functions supported the other: the brand drove revenue generation which paid for editorial that delivered on the brand promise, all underpinned by owning distribution.

    Publishing’s Downward Spiral

    It is hardly new news, particularly on this blog, to note that this model has fallen apart. The most obvious culprit is that on the Internet, distribution, particular text and images, is effectively free, which meant that advertisers had new channels: first ad networks that operated at scale across publishers, and increasingly Facebook and Google who offer the power to reach the individual directly.

    modeldisintegration

    I wrote about this progression in Popping the Publishing Bubble, and the intertwined functionality of publishers explains the downward spiral that followed: with less revenue there was less money for quality journalism (and a greater impetus to chase clicks), which meant a devaluing of the brand, which meant fewer readers, which led to even less money.

    What made this downward spiral particularly devastating is that, as demonstrated by the advertising shift, newspapers did not exist in a vacuum. Readers could read any newspaper, or digital-only publisher, or even individual bloggers. And, just as social media made it possible for advertisers to target individuals, it also made everyone a content creator pushing their own media into the same feed as everyone else: the brand didn’t matter at all, only the content, or, in a few exceptional cases, the individual authors, many of whom amassed massive followings of their own; one prominent example is Bill Simmons, the American sportswriter.

    Vox Media + The Ringer

    I wrote about Simmons two years ago in Grantland and the (Surprising) Future of Publishing, and noted that media entities needed to think about monetization holistically:

    Too much of the debate about monetization and the future of publishing in particular has artificially restricted itself to monetizing text. That constraint made sense in a physical world: a business that invested heavily in printing presses and delivery trucks didn’t really have a choice but to stick the product and the business model together, but now that everything — text, video, audio files, you name it — is 1’s and 0’s, what is the point in limiting one’s thinking to a particular configuration of those 1’s and 0’s?

    In fact, it’s more than possible that in the long-run the current state of publishing — massive scale driven by advertising on one hand, and one-person shops with low revenue numbers and even lower costs on the other — will end up being an aberration. Focused, quality-obsessed publications will take advantage of bundle economics to collect “stars” and monetize them through some combination of subscriptions (less likely) or alternate media forms. Said media forms, like podcasts, are tough to grow on their own, but again, that is what makes them such a great match for writing, which is perfect for growth but terrible for monetization.

    My back-of-the-envelope calculations estimated that Simmons’ Ringer podcast network was likely generating millions of dollars, and in an interview with Recode earlier this year, Simmons confirmed that is the case, claiming that podcast revenue was more than covering the cost of creating not just podcasts but the website that, at least in theory, created podcast listeners.

    Still, given Simmons’ ambitions, it would certainly be better were the site more than a cost center, which makes the company’s most recent announcement particularly interesting. From the New York Times:

    The Ringer, a sports and culture website created by Bill Simmons, will soon be hosted on Vox Media’s platform but maintain editorial independence under a partnership announced on Tuesday. Mr. Simmons, a former ESPN personality, will keep ownership of The Ringer, but Vox will sell advertising for the site and share in the revenue. The Ringer will leave its current home on Medium, where it has been hosted since it began in June 2016.

    Jim Bankoff, Vox’s chief executive, said in a phone interview that the partnership was the first of its type for the company and would allow it to expand its offerings to advertisers. Mr. Simmons said in a statement: “This partnership allows us to remain independent while leveraging two of the things that Vox Media is great at: sales and technology. We want to devote the next couple of years to creating quality content, innovating as much as we can, building our brand and growing The Ringer as a multimedia business.”

    Simmons is exactly right about the benefits he gets from the deal: instead of building duplicative technology and ad sales infrastructure, The Ringer can simply use Vox Media’s. This is less important with regards to the technology (Vox’s insistence that Chorus is a meaningful differentiator notwithstanding) but hugely important when it comes to advertising. It’s not simply the expense of building an infrastructure for ad sales; the top line is even more critical: it is all but impossible to compete with Google and Facebook for advertising dollars without massive scale.

    Make no mistake, Simmons is the sort of writer that many advertisers would be happy to advertise next to (his podcast has had an impressive slate of brand names, in addition to the usual mainstays like Squarespace and Casper mattresses); the problem is that when it comes to the return-on-investment of buying ads, the “investment” — particularly time — is just as important as the “return”: a brand looking to advertise directly on premium media is far more likely to deal with Vox Media and its huge stable of sites than it is to do a relatively small deal with a site like The Ringer.

    Indeed, the bifurcation in the Internet’s impact on editorial and advertising — the former is becoming atomized, the latter consolidated — explains why the implications for Vox Media are, in my estimation, the more important takeaway from this deal.

    Vox Media’s Upside

    To date Vox Media has been a relatively traditional publisher, albeit one that has executed better than most: the company has built strong brands that attract audiences which can be monetized through advertising, and that revenue, along with venture capital, has been fed into an impressive editorial product that builds up the company’s brands.

    The Ringer, though, is not a Vox Media brand: it is Simmons’ brand, a point he emphasized in his statement, and that’s great news for Vox. The problem with editorial is that while the audience scales, production doesn’t: content still has to be created on an ongoing basis, and that means high variable costs.

    Infrastructure, though, does scale: Vox Media uses the same underlying technology for all of its sites, which is exactly what you would expect given that software can be replicated endlessly. Crucially, the same principle applies to advertising: one sales team can sell ads across any number of sites, and the more impressions the better. Presuming The Ringer ends up being not an outlier but rather the first of many similar deals,1 then that means that Vox Media has far more growth potential than it did as long as it was focused only on monetizing its owned-and-operated content.

    Publishers of the Future

    The new model portended by this deal looks something like this:

    stratechery Year One - 265

    In this model the most effective and scalable publisher is faceless: atomized content creators, fueled by social media, build their own brands and develop their own audiences; the publisher, meanwhile, builds scale on the backside, across infrastructure, monetization, and even human-resource type functions.2 This last point makes a faceless publisher more than an ad network, and crucially, I suspect the greatest impact will not be (just) about ads.

    Earlier this month I wrote about the future of local news, which I argued would entail relatively small subscription-based publications. Said publications would be more viable were there a faceless publisher in place to provide technology, including subscription and customer support capabilities, and all of the other repeatable minutiae that comes with running a business. Publishers still matter, but much of what matters can be scaled and offered as a service without being tied to a brand and a specific set of content.

    I suspect this is part of the endgame for publishing on the Internet: free distribution blew up the link between editorial and publishing and drove them in opposite directions — atomization on one side and massively greater scale on the other. And now, that same reality makes possible a new model: a huge number of small publications backed by entities more concerned with building viable businesses than having memorable names.

    Disclosure: I have previously spoken at the Vox Media-owned Code Media conference and was previously a guest on The Bill Simmons Podcast; I received no monetary compensation for either appearance

    1. There is already a parallel to The Ringer within Vox Media: the company’s vast network of team-specific sites that sit under the SBNation umbrella ↩

    2. This is where Medium went wrong: the company made motions towards this model — which is why The Ringer is hosted there — but has decided to pursue a Medium subscription model instead ↩

  • Tulips, Myths, and Cryptocurrencies

    Everyone knows about the Tulip Bubble, first documented by Charles Mackay in 1841 in his book Extraordinary Popular Delusions and the Madness of Crowds:

    In 1634, the rage among the Dutch to possess [tulips] was so great that the ordinary industry of the country was neglected, and the population, even to its lowest dregs, embarked in the tulip trade. As the mania increased, prices augmented, until, in the year 1635, many persons were known to invest a fortune of 100,000 florins in the purchase of forty roots. It then became necessary to sell them by their weight in perits, a small weight less than a grain. A tulip of the species called Admiral Liefken, weighing 400 perits, was worth 4400 florins; an Admiral Van der Eyck, weighing 446 perits, was worth 1260 florins; a Childer of 106 perits was worth 1615 florins; a Viceroy of 400 perits, 3000 florins, and, most precious of all, a Semper Augustus, weighing 200 perits, was thought to be very cheap at 5500 florins. The latter was much sought after, and even an inferior bulb might command a price of 2000 florins. It is related that, at one time, early in 1636, there were only two roots of this description to be had in all Holland, and those not of the best. One was in the possession of a dealer in Amsterdam, and the other in Harlaem [sic]. So anxious were the speculators to obtain them, that one person offered the fee-simple of twelve acres of building-ground for the Harlaem tulip. That of Amsterdam was bought for 4600 florins, a new carriage, two grey horses, and a complete suit of harness.

    Mackay goes on to recount other tall tales; I’m partial to the sailor who thought a Semper Augustus bulb was an onion, and stole it for his breakfast; “Little did he dream that he had been eating a breakfast whose cost might have regaled a whole ship’s crew for a twelvemonth.”

    Anyhow, we all know how it ended:

    At first, as in all these gambling mania, confidence was at its height, and everybody gained. The tulip-jobbers speculated in the rise and fall of the tulip stocks, and made large profits by buying when prices fell, and selling out when they rose. Many individuals grew suddenly rich. A golden bait hung temptingly out before the people, and one after the other, they rushed to the tulip-marts, like flies around a honey-pot. Every one imagined that the passion for tulips would last for ever…

    At last, however, the more prudent began to see that this folly could not last for ever. Rich people no longer bought the flowers to keep them in their gardens, but to sell them again at cent per cent profit. It was seen that somebody must lose fearfully in the end. As this conviction spread, prices fell, and never rose again. Confidence was destroyed, and a universal panic seized upon the dealers…The cry of distress resounded every where, and each man accused his neighbour. The few who had contrived to enrich themselves hid their wealth from the knowledge of their fellow-citizens, and invested it in the English or other funds. Many who, for a brief season, had emerged from the humbler walks of life, were cast back into their original obscurity. Substantial merchants were reduced almost to beggary, and many a representative of a noble line saw the fortunes of his house ruined beyond redemption.

    Thanks to Mackay’s vivid account, tulips are a well-known cautionary tale, applied to asset bubbles of all types; here’s the problem, though: there’s a decent chance Mackay’s account is completely wrong.

    The Truth About Tulips

    In 2006, UCLA economist Earl Thompson wrote a paper entitled The Tulipmania: Fact or Artifact?1 that includes this chart that looks like Mackay’s bubble:

    Screen Shot 2017-05-23 at 7.16.24 PM

    However, as Thompson wrote in the paper, “appearances are sometimes quite deceiving.” A much more accurate chart looks like this:

    Screen Shot 2017-05-23 at 7.17.16 PM

    Mackay was right that there were insanely high prices: those prices, though, were for options; if the actual price of tulips were lower on the strike date for the options, then the owner of the option only needed to pay a small percentage of the contract price (ultimately 3.5%). Meanwhile, though, actual spot prices and futures (that locked in a price) stayed flat.

    The broader context comes from this chart:

    Screen Shot 2017-05-23 at 7.17.32 PM

    As Thompson explains, tulips in fact were becoming more popular, particularly in Germany, and, as the first phase of the 30 Years War wound down, it looked like Germany would be victorious, which would mean a better market for tulips. In early October, 1636, though, Germany suffered an unexpected defeat, and the tulip price crashed, not because it was irrationally high, but because of an external shock.

    As Thompson recounts, that October crash was in fact a financial disaster for many, including some public officials who had bought tulip futures on a speculative basis; to get themselves out of trouble, said officials retroactively decreed that futures were in fact options. These deliberations were well-publicized throughout the winter of 1636 and early 1637, but not made official until February 24th; the dramatic rise in options, then, is explained as a longshot bet that the conversion would not actually take place: when it did, the price of the options naturally dropped to the spot price.2

    By Thompson’s reckoning, Mackay’s entire account was a myth.

    Myths and Humans

    Early on in Sapiens: A Brief History of Humankind, Yuval Noah Harari explains the importance of myth:

    Once the threshold of 150 individuals is crossed, things can no longer work [on the basis of intimate relations]…How did Homo sapiens manage to cross this critical threshold, eventually founding cities comprising tens of thousands of inhabitants and empires ruling hundreds of millions? The secret was probably the appearance of fiction. Large numbers of strangers can cooperate successfully by believing in common myths.

    Any large-scale human cooperation — whether a modern state, a medieval church, an ancient city or an archaic tribe — is rooted in common myths that exist only in people’s collective imagination. Churches are rooted in common religious myths. Two Catholics who have never met can nevertheless go together on crusade or pool funds to build a hospital because they both believe that God was incarnated in human flesh and allowed Himself to be crucified to redeem our sins. States are rooted in common national myths. Two Serbs who have never met might risk their lives to save one another because both believe in the existence of the Serbian nation, the Serbian homeland and the Serbian flag. Judicial systems are rooted in common legal myths. Two lawyers who have never met can nevertheless combine efforts to defend a complete stranger because they both believe in the existence of laws, justice, human rights – and the money paid out in fees. Yet none of these things exists outside the stories that people invent and tell one another. There are no gods in the universe, no nations, no money, no human rights, no laws, and no justice outside the common imagination of human beings.

    The implication of Harari’s argument3 is pretty hard to wrap one’s head around.4 Take the term “tulip bubble”: everyone knows it is in reference to a speculative mania that will end in a crash, even those like me — and now you — that have learned about what actually happened in the Netherlands in the winter of 1636. Like I said, it’s a myth — and myths matter.

    The Rise in Cryptocurrencies

    The reason I mention the tulip bubble at all is probably obvious:

    Screen Shot 2017-05-23 at 5.07.24 PM

    This is the total market capitalization of all cryptocurrencies. To date that has mostly meant Bitcoin, but over the last two months Bitcoin’s share of cryptocurrency capitalization has actually plummeted to less than 50%, thanks to the sharp rise of Ethereum and Ripple in particular:

    currencyshare

    As you might expect, the tulip is having a renaissance, or to be more precise, our shared myth of the tulip bubble. This tweet summed up the skeptics’ sentiment well:

    To be perfectly clear, this random twitterer may very well be correct about an impending crash. And, in the grand scheme of things, it is mostly true today that cryptocurrencies don’t have meaningful “industrial [or] consumer use except as a medium of exchange.” What he is the most right about, though, is that cryptocurrencies have no intrinsic value.

    Compare cryptocurrencies to, say, the U.S. dollar. The U.S. dollar is worth, well, a dollar because…well, because the United States government says it is.5 And because currency traders have established it as such, relative to other currencies. And the worth of those currencies is based on…well, like the dollar, they are based on a mutual agreement of everyone that they are worth whatever they are worth. The dollar is a myth.

    Of course this isn’t a new view: there are still those that believe it was a mistake to move the dollar off of the gold standard: that was a much more concrete definition. After all, you could always exchange one dollar for a fixed amount of gold, and gold, of course, has intrinsic value because…well, because us humans think it looks pretty, I guess. In fact, it turns out gold — at least the idea that it is of intrinsically more worth than another mineral — is another myth.

    I would argue that cryptocurrency broadly, and Bitcoin especially, are no different. Bitcoin has been around for eight years now, it has captured the imagination, ingenuity, and investment of a massive number of very smart people, and it is increasingly trivial to convert it to the currency of your choice. Can you use Bitcoin to buy something from the shop down the street? Well, no, but you can’t very well use a piece of gold either, and no one argues that the latter isn’t worth whatever price the gold market is willing to bear. Gold can be converted to dollars which can be converted to goods, and Bitcoin is no different. To put it another way, enough people believe that gold is worth something, and that is enough to make it so, and I suspect we are well past that point with Bitcoin.

    The Utility of Blockchains

    To be fair, there is an argument that gold is valuable because it does have utility beyond ornamentation (I, of course, would argue that that is a perfectly valuable thing in its own right): for example, gold is used in electronics and dentistry. An argument based on utility, though, applies even moreso to cryptocurrencies. I wrote back in 2014:

    The defining characteristic of anything digital is its zero marginal cost…Bitcoin and the breakthrough it represents, broadly speaking, changes all that. For the first time something can be both digital and unique, without any real world representation. The particulars of Bitcoin and its hotly-debated value as a currency I think cloud this fact for many observers; the breakthrough I’m talking about in fact has nothing to do with currency, and could in theory be applied to all kinds of objects that can’t be duplicated, from stock certificates to property deeds to wills and more.

    One of the big recent risers, Ethereum, is exactly that: Ethereum is based on a blockchain,6 like Bitcoin, which means it has an attached currency (Ether) that incentivizes miners to verify transactions. However, the protocol includes smart contract functionality, which means that two untrusted parties can engage in a contract without a 3rd-party enforcement entity.7

    One of the biggest applications of this functionality is, unsurprisingly, other cryptocurrencies. The last year in particular has seen an explosion in Initial Coin Offerings (ICOs), usually on Ethereum. In an ICO a new blockchain-based entity is created, with the initial “tokens” — i.e. currency — being sold (for Ether or Bitcoin). These initial offerings are, at least in theory, valuable because the currency will, if the application built on the blockchain is successful, increase in value over time.

    This has the potential to be particularly exciting for the creation of decentralized networks. Fred Ehrsam explained on the Coinbase blog:

    Historically it has been difficult to incentivize the creation of new protocols as Albert Wenger points out. This has been because 1) there had been no direct way to monetize the creation and maintenance of these protocols and 2) it had been difficult to get a new protocol off the ground because of the chicken and the egg problem. For example, with SMTP, our email protocol, there was no direct monetary incentive to create the protocol — it was only later that businesses like Outlook, Hotmail, and Gmail started using it and made a real business on top of it. As a result we see very successful protocols and they tend to be quite old. (Editor: and created when the Internet was government-supported)

    Now someone can create a protocol, create a tokens that is native to that protocol, and retain some of that token for themselves and for future development. This is a great way to incentivize creators: if the protocol is successful, the token will go up in value…In addition, tokens help solve the classic chicken and the egg problem that many networks have…the value of a network goes up a lot when more people join it. So how do you get people to join a brand new network? You give people partial ownership of the network…

    0*cSLRTnSGE6vw2VGi.

    These two incentives are amazing offsets for each other. When the network is less populated and useful you now have a stronger incentive to join it.

    This is a huge deal, and probably the most viable way out from the antitrust trap created by Aggregation Theory.

    Party Like It’s 1999

    The problem, of course, is that while blockchain applications make sense in theory, the road to them becoming a reality is still a long one. That is why I suspect the better analogy for blockchain-based applications and their associated cryptocurrencies is not tulips but rather the Internet itself, specifically the 1990s. Marc Andreessen is fond of observing, most recently on this excellent podcast with Barry Ritholtz, that all of the dot-com failures turned out to be viable businesses: they were just 15 years too early (the most recent example: Chewy.com, the spiritual heir of famed dot-com bust Pets.com, acquired earlier this year for $3.35 billion).

    As the aphorism goes, being early (or late) is no different than being wrong, and that’s true in a financial sense. As I noted above, I would not be surprised if the ongoing run-up in cryptocurrency prices proves to be, well, a bubble. However, bubbles of irrationality and bubbles of timing are fundamentally different: one is based on something real (the latter), and one is not. That is to say, one is a myth, and one is merely a fable — and myths can lift an entire species.

    1. This link requires payment; there is an uploaded version of the paper here  ↩

    2. Thompson’s take is not without its critics: see Brad DeLong’s takedown here  ↩

    3. If you’re religious, please apply the point about “gods” to other religions — the point still stands! ↩

    4. I first encountered this sort of thinking in an Introduction to Constitutional Law course in university, when my professor contended that the U.S. Constitution was simply a shared myth, dependent on the mutual agreement of Americans and its leaders that it mattered. It’s a lesson that has served me well ↩

    5. And, as @nosunkcosts notes, said claim, via taxes, is backed by military might ↩

    6. A useful overview of how cryptocurrencies work is here  ↩

    7. What happened with The Dao will not be covered here! ↩