Unpopular Opinions in Crypto

The truth of a proposition has nothing to do with its popularity. And vice versa. This is as true in investing as in anything else, and especially important to keep in mind in a nascent area like crypto where there are still far more open questions than answers. We wanted to share some of our reactions, thoughts, and perspectives on the relevance of some recently-expressed “unpopular opinions” (some of which are hopefully not so unpopular anymore) to crypto and investing more broadly.

Big hat tip to Nic Carter for the great series of tweets that inspired this post (selections linked throughout, but the whole thing is worth reading). There’s a lot to unpack, so let’s get into it.

New investment into BTC from larger institutional investors may also lead to decreased usage in the dark web and other unregulated areas. In particular, regulatory control, chain-data forensics, and KYC/AML at the on- and off-ramps between fiat and crypto (desirable for large institutions considering taking positions) will tend to make Bitcoin less desirable for other market segments, and potentially make some other assets inaccessible or unusable.

There’s already a substantial and growing trend toward regulation of exchanges. With the announcement of Binance soon halting services on its main platform and DEX for US customers in favor of launching a regulated exchange business, we’re losing the largest remaining unregulated exchange, and a primary gateway to trading down-market “altcoins”, operating in the US. Even decentralized exchanges are feeling pressure to geofence trading in the US—Bancor snuck in an announcement on Tuesday morning. Exchange and usage of crypto is becoming more regulated across the board. And that’s okay.

There are massive cognitive/behavioral biases we carry with us from the experiences we’ve lived through—and, just as importantly, those we haven’t lived through. It’s why one of the first questions I ask people I meet in crypto is when and how they first got interested in the space.

There’s a lot more to talk about here that I plan to expand on in a future post, but for now remember this: being early doesn’t give you automatic credibility, and being late doesn’t automatically disqualify you. Nor does your background. But not being aware of your biases and potential blind spots can get you into a lot of trouble. That applies equally whether you come from traditional software, finance, or some other background; whether you got into crypto to make a quick buck or because you’re passionate about the ideals of decentralization; whether you’ve been here since 2009 or just started learning yesterday.

This is one of my favorite points in the whole thread. It’s relevant for any hybrid public/private chain project that checkpoints external data from a private system onto a public chain, as well as for (some) projects integrating external assets on top of Ethereum and other smart contract platforms. Some early examples are Microsoft’s “ION” decentralized identity project and MakerDAO (as they begin adding support for lending against collateral other than ETH).

Security in public blockchains (whether proof-of-work or proof-of-stake) boils down to the cost/benefit analysis of attacking the chain for a meaningful period of time. In proof-of-work systems, the relevant cost is the cost of accumulating or temporarily renting sufficient hash power, which (while the exact relationship is still up for debate) is generally related in quasi-equilibrium to the price of the native asset earned by miners on the base chain. In proof-of-stake systems, the relevant cost is the cost of accumulating or borrowing sufficient stake—which is directly related to the price of the base chain’s native asset. The economic benefit of the attack is the potential revenue to be gained from subverting consensus in the chain. That could take the form of reversing or blocking certain transactions, modifying data stored on chain, etc. The potential reward is greater when economic value external to the blockchain relies on data stored on chain.

Projects that use public blockchains to create consensus and trust without having any meaningful interaction with the native asset (apart from paying transaction fees) are effectively freeloading—they use the blockchain to secure additional value that is external to the chain, without contributing directly to the security of that blockchain. Transaction fees are generally designed to price in delay externalities imposed upon other users, not as a core component of the security model—hence the “oblique threat” to the security model posed by freeloading on consensus. We (luckily) haven’t run up against this yet in any significantly painful and public ways in Bitcoin or Ethereum.

There will be some point (whether due to increased value of the native assets or advancements in consensus algorithms raising the threshold for an economical attack) past which base-layer consensus will be so secure that freeloading won’t matter. But there is also a limit to how far this piggybacking can go in the short term. This is a threat to be aware of and work to mitigate as we move forward.

Howard Marks (co-founder of Oaktree Capital) spoke at our 2019 Collaborative Fund Summit back in February, sharing his experience and insights in a conversation with Morgan before taking some audience questions. Our LPs and guests had heard from cofounders Marc and Jen of Tagomi the previous hour, in addition to meeting Steve and learning about Collab Crypto for the first time, so we probably shouldn’t have been surprised when one of the first questions asked pertained to Marks’ views on Bitcoin.

His response was exactly in line with Nic’s tweet. He said that Bitcoin, as an asset that does not produce cash flows, is not something he (or anyone else) knows how to value analytically—period. He hasn’t put his or his clients’ money in as a result.

Marks expressed a similar opinion of gold in one of his wonderfully written (and publicly available!) investor memos from 2010: “…because it can’t be assessed quantitatively, no one can say definitively that the current price for gold doesn’t already recognize and reflect all of the dollar’s problems (and all of gold’s merits)”. Replace “gold” with “Bitcoin” in the above and, boom.

While quantitative approaches to the valuation of commodity cryptoassets do exist (many based around market supply/demand analysis, transaction momentum, or the equation of exchange), it’s generally not possible to directly use the discounted cash flow (DCF) analysis most familiar to traditional investors. It’s possible that we will never have a reliable DCF-like quantitative method for intrinsically valuing Bitcoin in particular. And that’s okay.

Like gold, as long as Bitcoin continues to prove itself as an independent commodity that can serve as a store-of-value (with the added benefit of being digital, portable, and censorship resistant), it will have achieved a measure of success. More of Howard Marks on gold, from the same memo: “I also concluded that since gold has ‘worked’ for hundreds of years, it probably will keep on doing so. It might not do so forever, but what’s the probability this will be the year that it stops? So I wouldn’t bet against it, and I might recommend a position ‘just in case’”.

This idea of “minimum viable success” for Bitcoin also highlights the immense blue-sky design space for other cryptographically secured assets and smart contract platforms to fill in other use cases taking advantage of the trust and coordination properties of public blockchains: stablecoins, DAOs, tokenized securities, privacy protocols, Web3 applications, digital commodities. The Ethereum community has made the most progress to date in these areas, but we are still in the early days of learning what is possible.

Two thoughts here. One, the importance of exchanges (as points of access) to the crypto ecosystem cannot be overstated. Those who control the ability to convert fiat currency into digital assets wield tremendous power, and are both valuable and influential players in the space as well as major concentrations of risk (corruption, capture by special interests, hacking and theft, etc.)

Two, on the threat of partition—the idea of “forking” a blockchain is a pretty foreign concept at first. How can you split an asset, creating two identical copies, that go on to have independent histories proceeding from the date of the fork? And what does this mean practically for the various forks of Bitcoin, Ethereum, and other assets we are living with today?

It’s useful to think about this in terms of game theory, as Stephanie Hurder of Prysm Group laid out in an excellent talk at the Stanford Blockchain Conference earlier this year. Hard forks that die off, with users eventually recombining, are not Nash equilibria. Some forks that persist are both Nash equilibria and Pareto optimal, meaning that if you were to recombine the two groups, at least one of them would be worse off. And some end up in what Hurder called “the Brexit section”—where a hard fork is in Nash equilibrium but is not Pareto optimal. One or both of the groups is worse off, and they wish it hadn’t happened, but without some external mechanism to bring everyone together, people are stuck in the suboptimal situation.

It doesn’t take a lot to end up in “the Brexit section”. We need better, more mature governance and coordination mechanisms to help us avoid this.

I’m proud to be working with the team at Collaborative Fund who have been investing in innovative economic models for the past 10+ years.

Tech remains an important component of what we do. But fundamentally the innovation in crypto is an economic one: using well-designed incentives to change the structure of trust and enable new ways of interacting with economic value in a digitally native way, to the benefit of all. What the internet did to media, crypto will do to finance. Tech is the great enabler, the means to an end, but not the end in itself.

Cover image: Richard Bartz (CC)