Marketing by Elon Musk

While the Uber story shows that a poorly regulated industry may be a good place to start a new company, Elon Musk suggests another opportunity borne by government:

But government is inherently inefficient. So it makes sense to minimize the role of government such that government does only what it has to do, and no more.

After this quote, some people cut their Social Security cards into pieces and run to a libertarian sea platform, away from government. This is, however, not what Musk means. Here’s some background.

It’s not a secret that, since 1958, NASA received $1 trillion dollars from federal budget to create the stack of technologies that SpaceX currently uses in its own commercial projects. SpaceX’s initial capital of $100 million makes 0.01% of this investment in space odysseys. The other 99.99% came from the government, which is presumably the necessary minimum mentioned by Musk.

And as Musk rightly reminds in the same interview:

But funded by the government just means funded by the people. Government, by the way, has no money. It only takes money from the people. [Laughter.]

So SpaceX took away dozens of engineers trained by publicly funded NASA and secured at least $500 million in government contracts.

Tesla Motors, another company founded by Musk, sells cars eligible for a $7,500-worth federal subsidy and numerous of state subsidies of a comparable amount. It’s about 20% off each car to help Tesla compete with fossil fuel vehicles.

His third company, Solar City, also advertises solar tax credits and rebates as its competitive advantage over traditional utilities. It promises that “some [state governments] are generous enough to cover up to 30% of your solar power system cost.”

The subsidies are, of course, not the point here. They are the second way toward clean and renewable energy, after complete pricing of fossil fuels (which is broadly supported by economists, see Pigou Club). In practice this transition will happen very much like what Tesla and Solar City do now.

But for anyone practically or intellectually interested in how this business works, executives happen to be a pretty misleading source. Even when these executives write long books about their companies or hire well-known economists without giving them complete data. Instead of the story how the company really works, the reader gets ideological cliches about business, management, and government. With teachers like them, it’s not a surprise that 9 out of 10 startups end up nowhere.

This happens mostly in hi-tech, with all this sudden success and media exposure. But the most competent executives manage to keep a low profile even here, because they know that they are best at running companies, not at teaching people how things work.

Markets and Behavioral Economics

In recent years, the mentions of “behavioral economics” in the books reached 1/3 of the mentions of The Simpsons, which is a big success of science. As usual, this research revolves around cognitive biases. But how far do these biases go? I’d say, biases end where markets start.

While most economic theories try to be general (often unsuccessfully so), behavioral economics suffers from the opposite. Experiments with college students—which the field was producing over the years—describe how humans think when they are in their early 20s filling questionnaire right before the lecture. These results don’t describe an experienced stock broker making million-worth transactions each day for the last 10 years.

Behavioral economics loses its explanatory power as bets go up. Nudging works in ecommerce, in everyday services. But as money and competition appear, irrationality vanishes. And it happens way before multimillion deals.

One landmark finding by Kahneman was the weighting function in his prospect theory:

Kahneman and Tversky - 1979 - Prospect Theory
Kahneman and Tversky – 1979 – Prospect Theory

It says that people overestimate the probability of unlikely events. Kahneman and Tversky derived this from the experiments with their undergrads, and you see how the solid line deviates from the “correct” dotted line by a few percentage points.

These few points get the magnitude in big decisions. A high school grad who wants to be a movie star plays against the odds. Movie appearances are skewed toward few superstars, and the probability of being one of them has many zeros after the decimal point. How does the grad percept his chances? He misses a few zeros, thus taking success as more probable. The person makes choices he wouldn’t make if he knew the true probabilities.

But in money markets, humans learn probabilities faster. The market does start with misestimated probabilities. For example, profitable sports betting strategies make money out of the people who bet on underdogs:

Hausch and Ziemba - 2008 - Handbook of sports and lottery markets
Hausch and Ziemba – 2008 – Handbook of sports and lottery markets

The bets don’t break even because the winners pay the house, too. This strategy yields positive returns in other popular sports, like soccer. But professional bettors leave these markets as big bets behave rationally and kill successful strategies. The bettors choose unknown sports, as jai alai. Or they bet on the goal difference, or particular players. All in attempts to find small markets, where irrationality resides.

Financial markets dwarf sports betting. How’s likely any human bias then? Not at all. Money managers with biases leave the table quickly. The markets have inefficiencies, but for other reasons and with other implications. A famous example of statistical arbitrage by LTCM:

Long-Term bought the cheaper off-the-run bond, while simultaneously selling the more expensive on-the-run bond. This allowed it to lock in the spread between the two bonds while immunizing it from interest rate movements. LTCM didn’t want to bet on the future of interest rates, instead, it wanted to make a very specific bet on liquidity. By creating a long position in one bond and a short position in another similar bond, LTCM knew that any losses from interest rate movements in one bond would be wiped out by equivalent gains in the other bond.

One problem with this trade, however, was that the spread between the two types of treasuries tended to be very small. For example, in August 1993, before Long-Term entered the market, 30-year bonds yielded 7.24%, while 29½ year bonds yielded 7.36%. This 12 basis point spread would not allow it to earn the type of returns that its investors expected, so the traders at LTCM needed to leverage their trade in order to magnify this return.

Hoping that the yields converge, LTCM bought positions in bonds with maturity 30 and 29½ years. Then the 1997 Asian crisis and 1998 Russia’s default occurred, and investors fled to quality. Investors were buying 30-year bonds, and not the 29½. As 30-year bonds grew in price, their yield declined and the gap between the two types of bonds increased. LTCM suffered immediate losses that wiped out the profits from the previous four years.

LTCM was an unusual hedge fund. Its founders were careful academic researchers with solid models. So, they found a good balance between profitability and bold assumptions about the markets. The result? The 12 basis point spread is a sort of inefficiency that big markets offer, and hedge funds can’t take it free of risks. The spread had nothing to do with human rationality. Buying 29½ bonds just happened to be a different market.

In between college questionnaires and Treasury bonds, where do cognitive biases cease to be a good approximation of reality? Behavioral economics can’t say. Social sciences don’t come up with universal models; they only show how to do certain thing better. Here, psychologists showed how to do elegant experiments that predict the future within a specific domain. So far, these experiments have been successfully adopted by UX designers and advertising. Whatever one thinks about their ethics, ads now waste less than in the Mad Men period—exactly because pros became as disciplined as earlier scientists. Meanwhile, governments—which could learn a lot from behavioral economics—adopt these things sluggishly, as recent results by the UK nudge unit show. But that’s also connected with markets and competition.

One to n: Market Size, Not Innovations

In his popular Zero to One, Peter Thiel singles out original product development as the most important step for entrepreneurs to make. After that, “it’s easier to copy a model than to make something new. Doing what we already know how to do takes the world from 1 to n, adding more of something familiar.”

Of course, building a prototype is important. But it’s not the most important problem in the hi-tech industry. More often, startups passes the zero-to-one step trivially. They fail in what comes next: in going from one to n.

Right from the preface, Peter Thiel supports his thesis with the cases of Microsoft, Google, and Facebook. But these companies never went from zero to one. Their core products were invented and marketed by their predecessors. Unix was there ten years before Microsoft DOS release. AltaVista and Yahoo! preceded Google. LiveJournal had pioneered social networks five years before Mark Zuckerberg founded Facebook. Do a small research on any big company mentioned in the book’s index, and you’ll find someone else who did zero to one before the big and famous.

Now, there’s an obvious merit in what Microsoft, Google, and Facebook did. Reaching billions of customers is more difficult than being a pioneer. However, it principially changes the startup problem. Going from zero to one doesn’t make a great company. Going from one to n does.

And startups pay little attention to their one-to-n problem. Take the minimum: the product’s target market, the n itself. In their stylized business plans, founders routinely misestimate their ns by a few digits. For one example, developers of a healthy-lifestyle app equated this app’s market to all obesity-related spendings, including things like liposuction. Naturally, the number was large, but it wasn’t their n.

Many founders sacrifice several years of their lives to ideas with overestimated ns. Back to Thiel’s examples, Microsoft, Google, and Facebook knew their huge ns before their grew big. Moreover, they purposefully increased their ns by simplifying their products on the way. In the end, each human being with Internet access happened to be their potential (and often actual) customer.

What do other founders do, instead? They see a monster like Microsoft and run away from competition into marginal niches. A marginal niche leaves them with a small n, while requiring about the same several years of development. In fact, it’s cheaper to fail early with such a niche product because if a modest project survives, it distracts its founders from bigger markets. The project functions like a family restaurant: good people, nice place, but, alas, no growth.

How to escape competition right? For example, by building a path to a big market right from the start, as Y Combinator suggests when it welcomes a possible competitor to Google.

Here, Zero to One again may mislead if taken literally. The book’s emphasis on innovation and technology sidelines simple facts about successful companies. Successful companies are lazy innovators. In their early years, Microsoft, Google, and Facebook were too small to invest in serious innovations. They’ve been built on simple technologies. Google run on low-cost consumer hardware and Facebook was a simple content management system written on PHP in a few weeks. Common-sense creativity, not fancy innovations, supported these companies. While their simple initial products remain critical to business performance, their graveyard of failed zero-to-one innovations grows (look at Google’s).

The path to a big market is perpendicular to innovations. In the innovation scenario, founders become scientists who dig a single topic until the zero-to-one moment. Such as very advanced DeepMind, which was virtually unknown before Google’s acquisition. In the big market scenario, founders devote their attention to marketing, namely, how to earn new users and retain their loyalty. Often, this task is easier to complete with handwritten postcards to early adopters than spending years teaching a computer to recognize cat videos. And it’s clearly not a single zero to one step, but many steps back and forth, with the foreseeable n in mind.