Curated Insights 2019.09.27

Why these two innovations in artificial intelligence are so important: Eye on A.I.

The first item was news that a Hong Kong-based biotechnology startup, InSilico Medicine, working with researchers from the University of Toronto, had used machine learning to create a potential new drug to prevent tissue scarring. What’s eye-popping here is the timescale: just 46 days from molecular design to animal testing in mice. Considering that, on average, it takes more than a decade and costs $350 million to $2.7 billion to bring a new drug to market, depending on which study one believes, the potential impact on the pharmaceutical industry is huge.

What’s also interesting here is that InSilico used reinforcement learning, an A.I. technique that hasn’t yet impacted business much. Reinforcement learning is notable because it doesn’t require the vast pools of structured, historical data that other A.I. methods do. Here researchers used reinforcement learning to rapidly design 30,000 new molecules and then narrow them down to six, which were synthesized and further tested in the lab. Look for more A.I. breakthroughs like this to start upending the balance of power between biotech startups and Big Pharma.

The second piece of underappreciated news is that researchers at DeepMind, the London A.I. shop owned by Google parent Alphabet, and Imperial College London, successfully used a deep neural network to find more precise answers to quantum mechanical problems. That’s basically the physics that underpins all of chemistry.

To date, the only element for which we can completely solve the underlying quantum equations is the simplest, hydrogen, which has just one proton and one electron. For every other element, we rely on approximations. Get better approximations, and you potentially get new chemistry – and that means new materials. Think room temperature superconductors or new kinds of batteries that will vastly extend the range of electric vehicles. DeepMind’s A.I.-powered approximations were in some cases almost an order of magnitude better than previous methods. If you’re Dow or DuPont, or Formosa Plastics or LG Chem, that sort of advantage could be worth billions.

Why prescription drugs cost so much more in America

All over the world, drugmakers are granted time-limited monopolies — in the form of patents — to encourage innovation. But America is one of the only countries that does not combine this carrot with the stick of price controls. The US government’s refusal to negotiate prices has contributed to spiralling healthcare costs which, said billionaire investor Warren Buffett last year, act “as a hungry tapeworm on the American economy”. Medical bills are the primary reason why Americans go bankrupt. Employers foot much of the bill for the majority of health-insurance plans for working-age adults, creating a huge cost for business.

Other drugs are more innovative — and their development undeniably expensive. According to Tufts University, the average is $2.6bn per drug, up 145 per cent in the past 10 years. Most drug candidates fail; those that do make it to later stages must go through expensive clinical trials. In support of the drug companies’ argument, one 2015 study found that for every extra $2.5bn a company made in sales, it produced one extra drug.

Robot pilot that can grab the flight controls gets its plane licence

A recent conversion of US military F-16 fighter jets into drones cost more than a million dollars each. ROBOpilot can be inserted into any aircraft and just as easily removed afterwards to return it to human-controlled operation.

“It looks like an impressive achievement in terms of robotics,” says Louise Dennis at the University of Liverpool. “Unlike an autopilot which has direct access to the controls and sensors, the robot is in the place of a human pilot and has to physically work the controls and reads the dials.” The makers suggest that ROBOpilot will be useful for tasks including transporting cargo, “entry into hazardous environments”, and intelligence, surveillance and reconnaissance missions.

Entrepreneurs hope microbes hold the key to a food revolution
A taxonomy of moats

Curated Insights 2019.09.20

The financial Turing test

Imagine we could simulate the universe where each time you are born to different set of parents with a different genetic makeup. Sometimes you are born a man. Sometimes you are born a woman. Sometimes black. Sometimes white. Sometimes smart. Sometimes not. Etcetera etcetera. What would you do to have the highest probability of becoming financially secure regardless of your background?

If you wanted to re-state this question more simply, it is: How do you get rich without getting lucky?


Product-user fit comes before product-market fit

The jump from product-user fit to product-market fit is no trivial leap. Skipping what to focus on during the product-user fit stage and prematurely racing to spark the market adoption can actually decelerate your path to product-market fit. Forcing growth on a product that isn’t yet ready for broader adoption will not ultimately convert to a market of highly retained, happy users. And if you don’t listen to the early power users closely enough, you may never discover the insights that get you to a world-class product.

Power users are the biggest sign of product-user fit. Making the leap from product-user fit to product-market fit is about listening to these users to evolve your product to attract more users. When exploring products that have only been in market for a short amount of time, the behavior of power users is often more interesting and important than any aggregate metrics. If the goal is to “make something people want,” then continuously talking to and observing early power users is the only way to really understand what drives both user retention and non-user activation.

5 reasons to consider buying Berkshire Hathaway

First, we think Berkshire’s broad diversification provides the company with additional opportunities and helps to minimize losses during market and/or economic downturns. Berkshire remains a broadly diversified conglomerate run on a completely decentralized basis, with a collection of moaty businesses operating in industries ranging from property-casualty insurance to railroad transportation, utilities and pipelines, and manufacturing, service, and retailing. The economic moats of these operating subsidiaries are built primarily on cost advantage, efficient scale, and intangible assets, with some of these businesses being uniquely advantaged as well by their ability to essentially operate as private companies under the Berkshire umbrella. The operating subsidiaries also benefit from being part of the parent company’s strong balance sheet, diverse income statement, and larger consolidated tax return.

Berkshire’s unique business model has historically allowed the company to–without incurring taxes or much in the way of other costs–move large amounts of capital from businesses that have limited incremental investment opportunities into other subsidiaries that potentially have more advantageous investment options (or put the capital to work in publicly traded securities). The managers of Berkshire’s operating subsidiaries are encouraged to make decisions based on the long-term health and success of the business, rather than adhering to the short-termism that tends to prevail among many publicly traded companies. Another big advantage that comes from operating under the Berkshire umbrella is the benefit that comes with diversification not only within the company’s insurance operations, but also within the organization as a whole. In most periods, it is not unusual to see weakness in one aspect of Berkshire’s operations being offset by the results from another or from the rest of the organization.


We can be weird, or it can be public

WeWork seems to be facing the traditional tradeoff: Stay private, keep control, but lose access to billions of dollars of funding, or go public, raise unlimited money, and have to act normal. If it does either of those things, that will mark a sort of end of an era. At the height of the unicorn boom, big tech companies could stay private without giving up the benefits of being public, or they could go public without taking on the burdens of being public. Now they might have to make hard choices again.

Shopify is now a major player in e-commerce. Here’s how it happened, according to the COO

Over the years, we’ve also realized as we grow bigger, we have incredible economy of scale. If you were to aggregate all our U.S. stores [customers’ sales volume] we would be the third-largest online retailer in the U.S. Amazon is first, eBay second, and Shopify is a very close third. What that means is when we go to the payment companies, when we go to the shipping companies or go to anyone, we negotiate on behalf of more than 800,000 merchants. Instead of keeping the economies of scale for ourselves, we distribute [the benefits] to the small businesses. I think that’s why we have been really successful.


The foodoo economics of meal delivery

The newbies, born more recently, have turned a once-tidy business into a food fight. They include listed firms such as Meituan of China and Delivery Hero of Germany, Uber Eats (part of Uber), Ele.me (owned by China’s Alibaba), and privately held DoorDash, based in San Francisco, and Deliveroo, from London. For most of them, delivery is their core business, so they share their cut of the bill with riders as well as restaurants. This substantially broadens the market to restaurants offering everything from steak to Hawaiian poké bowls. But margins suffer. Funded largely by venture capital, they have thrown subsidies at customers, forcing their veteran rivals onto the defensive. To catch up, the veterans are investing in advertising and delivery networks—at a big cost. This week Grubhub and Just Eat reported slumping earnings, and Takeaway mounting losses, as they spent heavily to fend off the upstarts.

The only mouthwatering aspect of the delivery business is its potential size. According to Bernstein, a brokerage, almost a third of the global restaurant industry is made up of home delivery, takeaway and drive-throughs, which could be worth $1trn by 2023. In 2018 delivery amounted to $161bn, leaving plenty of room for online firms to expand; the seven largest increased revenues by an average of 58%. Their businesses support the trend of 20- and 30-somethings to live alone or in shared accommodation, with less time and inclination to cook. In China, by far the biggest market for food delivery, one-third of people told a survey that they would be prepared to rent a flat without a kitchen because of the convenience of delivery. Delivery also fits neatly with the gig-economy zeitgeist, alongside ride-hailing firms such as Uber, Lyft and China’s Didi.

Moreover, potential growth may be overstated. Subsidies make true demand hard to gauge. When delivery charges and service fees eventually rise, which they will have to if profits are to materialise, some customers may flee. In the meantime, cheap money lets firms undercut rivals but distorts incentives. The war of attrition could get even worse if giants like Amazon muscle in, as it has tried to do by buying a stake in Deliveroo (the deal is stalled at present because of antitrust concerns). Alibaba, Amazon’s Chinese counterpart, uses Ele.me as a loss leader helping drive traffic to its profitable e-commerce sites.

Untangling the threads: Stitch Fix is a bargain

There have been numerous ecommerce 2.0 flameouts over the past decade (e.g. Gilt Groupe, Fab.com, Birchbox, Shoedazzle, Beachmint, One Kings Lane). Venture capitalists flocked to these businesses due to large addressable markets and strong top-line growth. To be fair, there have been some big winners (e.g. Wayfair) which can justify the VC game. But as Bill Gurley points out, innovations around pricing or distribution — think flash sales and subscriptions in a box — don’t represent core differentiation or sustainable competitive advantages. Additionally, these startups had access to hundreds of millions of VC funding and therefore weren’t forced to prove out the unit economics before scaling rapidly.

Are Airbnb investors destroying Europe’s cultural capitals?

The definitive story of how a controversial Florida businessman blew up MoviePass and burned hundreds of millions

Farnsworth’s pitch to MoviePass: $25 million for 51% of the company, two seats on the five-member board, and a promise to drop the monthly subscription price, temporarily, from $50 to $9.95, with the goal of hitting 100,000 subscribers. If all went well, the next step would be taking MoviePass public. But Farnsworth’s plan worried Spikes; to him, $10 a month was too low. At that price MoviePass would start losing money when a subscriber used the service more than once a month.

In the US, the average price for a movie ticket is about $9; if a customer ordered a ticket every day for a month (the maximum the MoviePass plan allowed), it would cost MoviePass about $270, of which the subscriber’s fee would cover just $10. But in July 2017, the MoviePass board agreed to the deal. And on August 15, the price drop went into effect. Thanks to word-of-mouth buzz and press attention, within two days subscriptions jumped from about 20,000 to 100,000. MoviePass had transformed from a scrappy startup trying to keep the lights on to a disrupter in the making.

But Spikes saw a looming disaster. The company was overwhelmed by its overnight success and couldn’t keep up with demand. A quarter-million new subscribers were signing up every month, and MoviePass customer-service lines were flooded with complaints from people who had been waiting weeks for their cards. MoviePass had lowballed the number of cards it would need after the price drop. It got to a point where the vendor making the MoviePass cards didn’t have enough plastic and had to call on its competitors to fulfill all the card orders. “We all knew we were selling something we couldn’t deliver on,” one former staffer said.

Pat Dorsey: Never put any moat on a pedestal

The same way you evaluate any other business, which is trying to think about the present value of future cash flows. This is an area where the world has changed pretty significantly over the past couple of decades because, 30 years ago, most investments were done via the balance sheet. They were investments in buildings, in factories, in railroads, in locomotives and all those came out of the balance sheet. Today, a lot of investment happens out of the income statement. If you are a software company, and you are acquiring new customers, who might have a nine to 10-year lifespan with the business, that comes out of sales and marketing, and so that depresses your current margins.

But it seems insensible to me to argue that I should not invest in a customer who could be with me for 10 years and who will pay me 3% more every year as I raise prices. Why is that not just as valuable an investment as a machine that will wear out in 10 years? One is an appreciating asset and the other is a depreciating asset. The former — the customer — comes by way of investing through the income statement and depresses current margins. As for buying the machine, it is just a capital expenditure. If you have a business that is re-investing heavily today, a software company or an Amazon for example, you have to think about the incremental unit economics. How much does it cost to acquire each customer and how much value do they deliver over some span of time, and then try to think about what does this business look like at steady state, say in a five or 10-year timeframe. You know what margins it will have once the investment slows down and then you discount those cash flows back to the present.

So far, Uber and Lyft have competed very heavily on price. That was evident in both of their IPO filings, they have been trying to undercut each other on price, which is not the sign of a healthy competitive dynamic that’s going to result in great return for shareholders. Maybe that will change, I don’t know. But, when I see two big companies trying to basically undercut each other on price and, it’s not really clear who is going to win, I’d rather just stay on the sidelines and watch. One of the most important things for an investor to do is to maximise return on time. By analysing Uber and Lyft, we probably aren’t going to get a lot of advantage, because everybody and their mother is trying to have an opinion on these things, and it’s just not clear how the competitive dynamics will pan out long term. So we’ve spent literally zero time on them!

A lot of it comes down to the unit economics of the business. Boeing and Airbus need to absorb a lot of fixed costs. Building an aircraft factory, investing and designing a new aircraft, requires a lot of very high fixed costs, and so they need to absorb that. And so, each incremental plane sold is very important to both companies. So they need to take market share from each other. Whereas for Visa and Mastercard, their fixed cost for the payment networks, those costs were sunk decades ago. Their network is there. It exists. So there’s no incentive to compete on price, because they don’t have the same economics of cost absorption.

When to sell and when a moat is weakening are really two different questions. But I would say, the biggest signal that a moat is weakening is the lack of pricing power. If a business historically had been able to raise prices and is no longer able to raise prices, that generally indicates that its competitive advantage is weakening or disappearing.

Howard Stern is getting ripped off

Take a look at Joe Rogan, who currently has the most popular talk show podcast with over 200 million downloads per month. This number comes from Joe himself¹, but let’s assume he was exaggerating and it’s only 100 million downloads per month.

Assuming he sells ads at a low $18 CPM (cost per thousand listeners) and sells out his ad spots, he’s making approximately $64mm in annual revenue. If he’s on the higher end, at $50 CPM, he could be making as much as $240mm per year². The only factor that would change this is how many free ads Joe gives to companies that he has a personal equity stake in (like Onnit, the supplement brand he co-owns).

That means that Joe makes somewhere between $64-$240 million per year in revenue from his podcast advertising alone—and that’s handicapping his audience by half what he claims to have. That number also doesn’t include any additional revenue generated from his wildly popular YouTube channel, which has over 6 million subscribers.

Based on existing advertising revenues alone, Joe Rogan could easily be worth over a billion dollars, even if he doesn’t realize it. If estimates are correct, he owns a business that produces somewhere in the neighborhood of $60-$235 million/year in profit and is likely growing at 30–50% annually (assuming his audience is growing alongside the podcast ecosystem)³. If it were publicly traded, his podcasting business could easily fetch a valuation in the billions.

Even the small stresses of daily life can hurt your health, but attitude can make a difference

When people talk about harmful stress — the kind that can affect health — they usually point to big, life-changing events, such as the death of a loved one. A growing body of research suggests that minor, everyday stress — caused by flight delays, traffic jams, cellphones that run out of battery during an important call, etc. — can harm health, too, and even shorten life spans.

Curated Insights 2019.09.06

Where the opportunities are in regulated marketplaces: Managed marketplaces

The licensing of workers was more critical in a “pre-internet” world, since licenses established consumer trust by signaling the skills or knowledge required to perform a job. But today, digital platforms can mitigate the need for (some) licensing by establishing trust and ensuring quality through other means — such as user reviews, platform requirements, and other mechanisms like pre-vetting and guarantees.

“Managed marketplaces” models in particular can be helpful in establishing user trust, because they intermediate parts of the service delivery, adding value by taking on functions like identifying high-quality providers, standardizing prices, and automating matching between demand and supply. As scrutiny around safety for marketplaces continues to rise, the importance of trusted labor becomes even more significant. In childcare, for instance, people don’t want to just see a list of all possible caregivers — they want to know with certainty that the providers they’re hiring are trustworthy and qualified, and a managed marketplace can capitalize on this user need by thoroughly vetting all supply.

Managed marketplaces can greatly mitigate the need for licensing because users trust the marketplace itself, particularly on the highly managed side of the spectrum. Such platforms can enable high-quality, but unlicensed, suppliers to offer services alongside licensed providers — and in doing so, promote entrepreneurship and alleviate supply constraints.


The Big Short’s Michael Burry explains why index funds are like subprime CDOs

The dirty secret of passive index funds — whether open-end, closed-end, or ETF — is the distribution of daily dollar value traded among the securities within the indexes they mimic. In the Russell 2000 Index, for instance, the vast majority of stocks are lower volume, lower value-traded stocks. Today I counted 1,049 stocks that traded less than $5 million in value during the day. That is over half, and almost half of those — 456 stocks — traded less than $1 million during the day. Yet through indexation and passive investing, hundreds of billions are linked to stocks like this. The S&P 500 is no different — the index contains the world’s largest stocks, but still, 266 stocks — over half — traded under $150 million today. That sounds like a lot, but trillions of dollars in assets globally are indexed to these stocks. The theater keeps getting more crowded, but the exit door is the same as it always was. All this gets worse as you get into even less liquid equity and bond markets globally.

This structured asset play is the same story again and again — so easy to sell, such a self-fulfilling prophecy as the technical machinery kicks in. All those money managers market lower fees for indexed, passive products, but they are not fools — they make up for it in scale. Potentially making it worse will be the impossibility of unwinding the derivatives and naked buy/sell strategies used to help so many of these funds pseudo-match flows and prices each and every day. This fundamental concept is the same one that resulted in the market meltdowns in 2008. However, I just don’t know what the timeline will be. Like most bubbles, the longer it goes on, the worse the crash will be.


Debunking the silly “passive is a bubble” myth

So index funds hold less than 15% of shares in public companies. And according to former Vanguard CEO Bill McNabb, indexing in stocks and bonds globally represents less than 5% of global assets.

When you buy an index fund of the total stock market, you are literally buying the stock market in proportion to the shares held by all active investors. If you sum up the collective holdings of active managers, what you basically get is a market-cap-weighted index. Index fund investors are simply buying what the active investors have laid out for them.

Charley Ellis wrote in his book, The Index Revolution, that indexing accounts for less than 5% of trading, with the remaining 95% or so done by active investors. This will always be the case, no matter the amount of money flowing into index funds.

When an index fund investor sells, they’re technically selling their holdings in direct proportion to their weighting in the index. So there is little market impact involved. Again, index fund investors are simply owning stocks in the proportion that all active investors own stocks. Plus, index funds never lever up your holdings. They never receive a margin call. They don’t put 30% of your holdings in Valeant Pharmaceuticals. And no index fund has ever closed up shop to spend more time with their family.

Why the Periodic Table of Elements Is More Important Than Ever

How Amazon’s shipping empire is challenging UPS and FedEx

Amazon now delivers nearly half of its orders, compared with less than 15% in 2017, according to estimates from research firm Rakuten Intelligence. It is now handling an estimated 4.8 million packages every day in the U.S. … The U.S. Postal Service, once the primary carrier of Amazon parcels, delivers about half the share of packages than it did two years ago.

Asahi’s voracious thirst sees it take crown as king of M&A in Asia

At a news conference in Tokyo this month, Mr Koji said the company would focus on strengthening its three core markets in Japan, Europe and Australia: “That will be our priority and we’ll subsequently consider whether we will do further merger and acquisition deals.”

Suntory, known for its Yamazaki whisky, bought US spirits maker Beam for $16bn in 2014, creating the world’s third-largest spirits maker. Before that, Kirin had made a disastrous foray into Brazil with a $3.9bn acquisition of family-owned Schincariol in 2011.

Asahi went on a buying spree of its own with a $1.3bn acquisition of New Zealand’s Independent Liquor in 2011 and other smaller deals in Australia, China and Malaysia. In the decade before 2016, it spent $3.9bn on 24 outbound deals, according to Dealogic, but none had any serious impact on its balance sheet. Their geographical reach was limited, with its overseas business making up less than 15 per cent of revenues.

Deep dive into the General Electric-Markopolos case – Here: The Baker Hughes accounting

Admittedly, GE has never been at the forefront of conservative accounting application. Looking into the history of the company we can find a couple of examples of quite aggressive representations of its economic situation. But with regard to the Baker Hughes accounting we cannot find anything wrong (but of course, it could be that we missed something). Moreover, the Markopolos report does not come even close of what is necessary to assess the accounting treatment here. We do not want to judge to harshly on the report (with regard to the Baker Hughes accounting) because at least the economics are correct – but also disclosed by GE – but all in all the Markopolos report really seems to be a bit light in accounting from our subjective point of view.

The Amazon is not Earth’s lungs

The Amazon produces about 6 percent of the oxygen currently being made by photosynthetic organisms alive on the planet today. But surprisingly, this is not where most of our oxygen comes from. In fact, from a broader Earth-system perspective, in which the biosphere not only creates but also consumes free oxygen, the Amazon’s contribution to our planet’s unusual abundance of the stuff is more or less zero. This is not a pedantic detail. Geology provides a strange picture of how the world works that helps illuminate just how bizarre and unprecedented the ongoing human experiment on the planet really is. Contrary to almost every popular account, Earth maintains an unusual surfeit of free oxygen—an incredibly reactive gas that does not want to be in the atmosphere—largely due not to living, breathing trees, but to the existence, underground, of fossil fuels.

After this unthinkable planetary immolation, the concentration of oxygen in the atmosphere dropped from 20.9 percent to 20.4 percent. CO2 rose from 400 parts per million to 900—less, even, than it does in the worst-case scenarios for fossil-fuel emissions by 2100. By burning every living thing on Earth. “Virtually no change,” he said. “Generations of humans would live out their lives, breathing the air around them, probably struggling to find food, but not worried about their next breath.”

Why Indonesia is shifting its capital From Jakarta

As well as bursting at its seams, the city is sinking. Two-fifths of Jakarta lies below sea level and parts are dropping at a rate of 20 centimeters (8 inches) a year. That’s mostly down to the constant drawing up of well water from its swampy foundations. Stultifying traffic congestion and polluted air are a daily reality for Jakarta’s 10 million inhabitants. The gridlock costs an estimated 100 trillion rupiah ($7 billion) a year in lost productivity for the greater Jakarta area, known as Jabodetabek, encompassing 30 million people.

Jakarta will keep growing. The population is on course to reach 35.6 million by 2030, helping it topple Tokyo as the world’s most populous city. Since the greater metropolitan area generates almost a fifth of Indonesia’s GDP, Jakarta will continue to be the country’s main commercial hub. There’s a $43 billion plan to sort out the traffic, including a Mass Rapid Transit rail line that opened in 2019. As for Jakarta’s submergence problem, the president is planning a giant wall to keep big waves out.

Better is fragile — different is king

Most of us grew up believing that, to compete, we need to be better than the competition. We need better skills, better players, better résumés. But what happens when your best is no longer good enough? What happens when that amazing software application you just spent beaucoup bucks developing is blindsided by an even better program? One that’s less expensive, to boot?

Better is fragile. It can be trampled in a nanosecond. Attempting to be better puts companies on a hampster wheel, running faster and faster—and in the same direction as everyone else—to keep up. Better is weak.

Different is king. When you can differentiate yourself in the market, you step off the hamster wheel, never to return. You only look back to witness the frenzy your brand is causing in the hamster cage you left behind.

Here’s your choice: Spend a lot of time and money in pursuit of better. Or find what makes you different, and then do it on purpose.


A silent interlude

Warren Buffett hasn’t been reading five newspapers every day for seven decades for no reason. The trick is to find the right balance between exposure to the news while honing the ability to distinguish between news and noise.


Running your trading as a business

Imagine that you are pitching your trading business to a venture capitalist. How will you convince the VC that this is a business worth investing in?

Reasonable investing philosophies

Personal finance > investing, at all income levels, because a good saver who doesn’t invest will be fine but a great investor mired in debt and overspending can be wiped out.