Digital Tulips or: How I learned to stop worrying and embrace the coin.

In 1637, a tulip bulb in Amsterdam would sell for several multiples of the average annual wage. Bitcoin, which hit $5,000 yesterday, has drawn more than a few comparisons to the old Dutch irrational exuberance.

There are people who believe cryptocurrencies will upend the Fed, bankrupt Visa, or become the major mechanism for remittances. At the extreme, some people compare cryptocurrencies today to the Internet in 1993. Personally, I’m a skeptic who admires the math and technology of cryptocurrencies, but doesn’t think it makes a ton of sense practically as either a store of value or means of transaction.

Yet, despite my skepticism, I own cryptocurrencies.

At this point, I should mention this article isn’t an argument about whether cryptocurrencies are good or not. There are plenty of well-written pieces around the interwebs to confirm whatever you want to believe. This article also doesn’t touch on the more general applications of blockchain technology. Rather, this is meant to be a concrete look at handling “uncertainty of uncertainty” in investment risk using cryptocurrencies as an example.

Let me explain my ownership starting with Investing 101. A typical way to evaluate an investment is modeling its expected value. Econ nerds may prefer expected utility or even something fancy like the capital asset pricing model. Whatever your decision process, you’re relying on the existence of good estimates of underlying returns. Whether you’re looking at bonds, stocks, angel investments, or bets on a casino floor, methods exist to predict the range and distribution of outcomes and invest appropriately. They’re not always perfect but they do offer a decent baseline. (Pro-tip: table games at the casino are almost never a good investment. Unless you’re FedEx.)

Great. So with that knowledge we just need to come up with a probability distribution for cryptocurrency outcomes and invest appropriately. Here’s the problem: it’s impossible to come up with an outcome distribution for bitcoin. There are two causes of this uncertainty: lack of historical precedent and lack of it being an unalloyed replacement for something already in widespread use. Given either we could at least construct a starting point.

Millions of cryptocurrency users here?

If you’re confused about the impossibility of deriving an underlying distribution of cryptocurrency outcomes, consider a thought experiment. Take a set of hypothetical civilizations that evolved a currency system somewhat like ours (e.g. fiat or precious-metal-backed-paper). Let’s further say they’ve all developed an idea like cryptocurrency. What percent of those civilizations end up adopting the new technology as either a primary means of exchange or a store of value?

I honestly have no idea and nor does anyone. Maybe a cryptocurrency revolution is completely inevitable, comparable to the way civilizations seem to always transition from schlepping around goats for barter to using precious metals and eventually slips of paper for transactions. Maybe no civilization ever ends up moving away from fiat and crypto will forever be a side project for nerds. Instead of trying to quantify the results or speculate on unknowable probabilities, we should bask in our ignorance and proceed.

Normally when presented with an investment for which I cannot create some distribution of outcomes, I pass. No need to take unknown risk when more confident investments are available. This doesn’t mean we shouldn’t take risk — after all, angel investing is probably the riskiest investment on the planet. But I can somewhat quantify it: most companies fail, a few get acquired, and an even smaller number drive overall returns. But, if I literally have no idea how to evaluate a company, I like to keep the dry powder.

Cryptocurrencies are an entirely different beast. If my skepticism is misplaced, then my assumptions about economics, politics, and the dynamics driving the world are wrong. Since my current investments are predicated on those beliefs, I expect cryptocurrencies to be negatively correlated with the rest of my investments. Owning cryptocurrency isn’t merely another investment to be compared against a stock or bond, but rather a hedge against my model of the world, the uncertainty of uncertainty.

Of course the range of possible outcomes for cryptocurrencies isn’t binary. Perhaps they maintain value as a technophile niche or the less scrupulous parts of the Internet continue using bitcoin for illicit transactions. But these aren’t really interesting to me as they don’t call into question any fundamental assumptions about the world. However, if the dollar/yen/euro/pound collapse or cryptocurrencies become a big player in the world economy, I can confidently say “I was wrong in a really significant way.”

More generally, I think we should all constantly consider the question, “what if I’m wrong?” Not if we’re wrong about a single decision or outcome, but rather what if our way of analyzing the world is fundamentally flawed. Importantly this is different than merely being unlucky. For example, suppose I jump from an airplane and my parachute fails to open. During my rapid plunge I’d at least know I wasn’t wrong in any fundamental sense, but merely that I unfortunately drew the short straw (or rip-cord). On the other hand, if I construct an economic model that utterly fails to predict reality, it necessitates a reevaluation of the model entirely. It requires I rethink my assumptions about how the world functions. In this way, owning cryptocurrencies is best thought of as a hedge against a potentially significant defect in my schema of the world.

To the moon?

So, personally, I own a basket of cryptocurrencies in what I call the “Justin is wrong about the world portfolio.” In the scenario best for Bitcoin/Ethereum and the worst for my epistemic confidence, cryptocurrencies go to the moon!