Technology Investing Spurs Vertigo-inducing Response
Whoops! How can I write you a Gilder’s Guideposts, possibly steering your investments in a time of turmoil, when I am suffering from the sudden onset of Non-Benign Paroxysmal Positional Vertigo?
They say it’s benign (BNNV), but that’s only if you do not have to make your way through the DFW (the Dallas Fort-Worth Airport), or through the mazes of the markets without wobbling or crashing as your world spins dizzily around you. As remedies, my doctor yesterday prescribed a motion sickness pill named Bromine and physical contortions called Epley maneuvers.
By now, they are working well enough that I can finally sit in seat 10E on an American Airlines Airbus 311neo (“final assembly completed in the USA”), and I am able to compose this weekly message.
However, as I contemplate the gyrations of the market, the volatile wobbling of stock values, the deceptiveness of much official data, the long plummet of the dollar measuring stick against gold, the tidal flows of index funds, mutual funds, exchange-traded funds, and other putative trillions of dollars of “passive investment vehicles” that do not even pretend to fathom the real values of their so-called securities, I find myself again dizzied and vertiginous, and no head and neck contortions or Nexaprexine tablets offer any relief.
The current issue of Harper’s, which printed my first major articles some 55 years ago, emblazons its cover with a story on the problem entitled “401K Doomsday: Will Passive Investing Spell Catastrophe?” by a savvy writer named Andrew Lipstein. He shows that dominating current markets are some $15 trillion of passive investments, some $3 trillion alone from Vanguard, Black Rock, and State Street, that make no effort to research or evaluate their securities choices.
Lipstein quotes “passive investment” critic Michael Green, who gained credibility by recently orchestrating a $250 million profit in one trade for Peter Thiel. Green says these investors are, in fact, “not passive at all. What they are is active managers with the world’s simplest quantitative strategy. Did you give me cash? If so, then buy. Did you ask for cash? If so, then sell.”
This means they shun the usual notion of buying low and selling high, and instead in general follow the opposite rule of buying high (when all are buying) and selling near the bottom (when everyone wants cash). This process renders the associated markets mindlessly volatile, and investment analysts seeking “guideposts” increasingly clueless.
Nonetheless, at the end of the Harper’s piece, which flirts with doom theories for several salacious pages, Lipstein suddenly loses his apocalyptic bravado. He interviews several authorities, led by Danish economist Lasse Heje Pedersen, who is cited heavily by Green. Contending that “everyone can make money together,” Pedersen favored the view that “active investors will always be able to correct distortions caused by index funds.”
Ending a 10-page article on the stultifying and possibly catastrophic impact of passive investing, this reassuring view was anti-climactic: doom fended off, depreciated, deferred, even reversed by sage and visionary active traders.
What are the active traders doing, though? They are mostly relegating their trading to computers, or what Jaron Lanier — the shaggy sage who invented “virtual reality” — calls “Siren Servers.” Wrote Lanier: “Siren Servers are usually gigantic facilities, [data-centers] located in obscure places where they have their own power plants and some special hookup to nature, such as a remote river that allows them to cool a fantastic amount of waste heat.”
The most successful of all has been Nobel Laureate James Simons, co-author of the Chern-Simons algorithm in string theory and founder of the Renaissance Funds. Under a team led by former IBM speech recognition luminary Robert Mercer, Renaissance reached stratospheric heights, averaging roughly 40% yield every year, through up-and-down markets, for some 20 years (some $65 billion in 2018). Included was an 80% gain during the crash of 2008.
Simons and Mercer, in my view, were the real discoverers of the breakthrough in artificial intelligence in the now famous “Transformers paper” by eight Google researchers in 2017. The Google team showed that in achieving results beyond human capabilities, algorithmic logic, however complex and ingenious, is far less effective than pattern recognition by parallel processing across vast databases. Mercer proved the point at IBM by far outperforming everyone else in speech recognition by throwing out language models, grammars, and verbal theories. Nearly all the recent AI boom is based on this insight that underlies the two decades of Renaissance investment heroics.
Renaissance spurned any direct kibitzing from fundamental analysts or from anyone who knows anything about particular companies. The flaw of this AI approach is that functioning chiefly with price signals within markets, it is intrinsically self-referential and circular. As Simons explained in a speech in 1999: “Efficient market theory is correct in that there are no gross inefficiencies. But we look at anomalies that may be small in size and brief in time… We’re always in-and-out and out-and-in. So, we’re dependent on intense activity to make money.”
“Intense activity” meant huge data centers in Syosset, Long Island, parsing terabytes of data on market movements and associations in parallel, at gigahertz speeds, neutral between up and down-market movements, shorts and longs. Thus came the amazing performance during the two years of the financial crash.
This kind of computerized processing in AI systems is not investing at all. It makes near zero contribution to the discovery of new technologies and business plans and the channeling of capital to them.
The heart of the investment process is in depth research of companies, technologies, entrepreneurial skills, and paradigms. Governed by information theory, it seeks unexpected insights, measured by the element of surprise.
“Anomalies small in size and brief in time” are nearly always trivial, even if they can be harvested by siren servers in nanoseconds. What we seek are major new paradigms and scientific breakthroughs that yield transformative new technologies. At present, our focus is on the discovery of graphene, a single layer of carbon atoms discovered to be 200 times stronger than steel, more flexible than rubber, a thousand times more conductive than copper, and the best heat transmitter ever studied.
It will transform nearly every industry, including electronics. The most exciting breakthrough came early this year at Georgia Tech, where physics Professor Walter de Heer discovered that a perfect film of near superconductive graphene can be created on the surface of a silicon carbide wafer by subliming out the silicon one atom deep.
This promises cool, terahertz, wafer scale electronics within the next decade. We will be on top of it, exempt from the dizziness and vertigo of siren servers shuffling noisy data.