Man maketh markets

From Kingsman: The Secret Service (2014)

William Horman’s one-liner, “Manners maketh man”, could perhaps credit its re-entry into the vernacular to Colin Firth (and subsequently Taron Egerton) in the iconic pub brawl scenes in the Kingsman films, wherein these words preceded some fantastic fight moves, particularly involving an umbrella.

Just as “Manners maketh man”, markets are made by the multitude of participants that engage in bids and offers on an ongoing basis.

At their very core, markets perform a coordination function, seeking to discover what the prevailing price of any asset is. Mechanically, it’s simply a game of trial and error in real time: parties interested in buying bid, parties interested in selling offer, and where they cross up a trade is done.

We put this note together with the backdrop of recent volatility in the cryptocurrency space, but these principles are applicable when looking at any market, anywhere in the world. The arguments about how the volatility in the space compromises its investability are a false dichotomy: volatility is the outcome that we observe, rather than an intrinsic feature of any asset class.

Of course, certain asset classes have different inherent scope for volatility: for example, fixed income instruments tend to be less volatile than equities, simply because expectations of whether the issuer is a going concern don’t tend to vary frequently between “viable” and “going bankrupt”. Yet we don’t point to the volatility of equity instruments vis-à-vis fixed income and declare equities uninvestible on that count, not least because that would belie a profound misunderstanding of how markets work.

Likewise, we take the view that the volatility in what is surely a fledgling asset class of cryptocurrencies needs to be looked at in a broader context. There are many angles we can take on this, but top priority for us is market structure.

What makes a price?

Defining market price is simple: it’s where the highest bid meets the lowest ask and a trade gets done. Working out what those bids and offers are based on is the real challenge.

For decades, the greatest names in finance and economics have tried their hand at securities pricing: from the aptly named “Capital Asset Pricing Model”, to the expanded versions built by Fama and French, taking in inputs like size and valuation metrics, to the ill-fated (but still widely used) Black-Scholes option pricing model, to the plethora of magic formulae that purport to be able to tell what “fair value” is for any given instrument.

Yet the market refuses to yield to formulae and regressions. And why would we want it to? If the market did yield and price everything perfectly (efficient market hypothesis, anyone?), price charts would comprise horizontal lines, jumping in binary fashion from one level to another as new information is priced in perfectly and efficiently. The prospect of a market that behaves in this way is absolutely terrifying – not least because of the volatility that would cause.

As a result, it doesn’t happen, because in reality “the market” isn’t this exogenous thing that sits out in the cosmos and feeds us prices and day-to-day excitement by swinging prices up and down. “The market” is everyone that participates. The idea that “fair value” is an absolute number is a notion that can only be entertained in academia (or a math exam).

Where the game theorists got it right is in their use of “fair value” as an input into the process of modelling auction outcomes, where each individual participant has a “prior” belief of what the fair value of the object being auctioned is.

They go on to bid for it, and depending on the format of the auction, an outcome is observed where the item is sold. That price is a function of the bids, which are function of each participant’s “fair value” prior assumption, which are themselves a function of the nature of each participant.

Of course, markets are not a single-event auction – they are an infinite, continuous game. But our point has hopefully been clear so far: the prices are made by the participants, and “fair value” not only doesn’t need to be the same for all participants, but it is also at best indirectly responsible for what the final price ends up being.

Rate of change

Extending that logic, one could argue that given a sufficiently large number of participants, the volatility of the price of an instrument is a proxy for the rate of change of the opinions around “fair value”.

The condition for a large number of participants is important: consider a situation where only two participants are making the price in a market, and they have drastically different views on what “fair value” of an instrument is. The price would be highly volatile at best, assuming any trade could happen at all, but no “change” in beliefs around “fair value” would be needed.

On this basis then, an increasingly clear picture starts to form: the more “institutionalised” an asset class is, the lower the likely volatility in their prices.

This isn’t just hypothetical: work by Faias and Ferreira at the ECGI from 2016 asked the question “Does Institutional Ownership Matter for International Stock Return Comovement?”, looking at more than 18,000 instruments over a period of 10 years from 2001-2010. Based on their work, the answer seems to be a resounding “yes”. In particular, they show that stocks with high institutional ownership benefit less from diversification, thanks to increased comovement.

They do not, however, propose a reason why; we’d venture a guess though – institutional views on “fair value” tend to be largely convergent around a “consensus”, and for various reasons (from indexation to benchmarking to “career risk”), deviation from that “consensus” is less than ideal for a manager to make.

Furthermore, this piece by in the Journal of Financial Economics (Nagel 2004) points out an additional dynamic: the availability of short sell borrow supply in institutionally well-owned instruments actually attenuates volatility in prices. Particularly, Nagel points out that “prices of stocks with low institutional ownership also underreact to bad cash-flow news and overreact to good cash-flow news, consistent with the idea that short-sale constraints hold negative opinions off the market for these stocks.”

Added to the mechanics of passive flows which are increasingly dominant in traditional markets today and it’s hard to imagine any situation where opinions around what an instrument is worth fluctuate violently.

Institutional capabilities with no institutional opinions

Taking this framework over to the land of cryptocurrencies, where institutional investors are barely present and participants are largely individual (or small funds of at best US$200-300m in AUM, no larger than a “small” family office), one can hardly avoid conceding that using volatility as a metric for comparison is spurious at best.

The market for cryptocurrencies is large, but highly heterogenous in terms of the opinions each individual participant holds. Additionally, while able to access huge amounts of financial leverage, most accounts and the trades they make are tiny. Consider this data put together by Kaiko.com on what is arguably the “most popular” traded pair in crypto, BTC/USD:

LMAX Digital, an exchange skewed towards institutional traders holds the #1 position at $17,648/trade, while the rest of the other exchanges see average trade sizes of US$5-6k. Binance has been excluded from this analysis, but extrapolating these trends from prior research by derivatives exchange Bitmex suggests that Binance’s figures aren’t too far off that US$5-6k mark.

While individual traders’ accounts are relatively small compared to the size of the market, the amount of leverage these traders are given access to further contributes to the occasional spurts of violent unwinding.

As of the time of writing, we’ve just seen a violent unwind of futures-driven leverage, where a total of almost US$10bn of open interest was liquidated in a single day across the major exchanges, according to data compiled by data provider bybt.com. This was the largest 24hr liquidation ever, which was itself followed by another US$5bn of liquidation (so far) today.

Arguably, leverage of that scale in the “traditional” world would be given out with much more reluctance, as intermediary banks would be reluctant to finance small counterparties. For better or for worse, these limits don’t exist in the crypto world, although one would hope that after having had positions liquidated a couple of times, risk management amongst smaller traders would improve.

The differences go beyond the mechanics of trading, however. The diffusion of information on various crypto projects also behaves very differently from traditional assets: there isn’t a Bloomberg, Reuters or PRNewsWire-like service for crypto. It’s called Twitter, and sometimes Discord.

Founders and their communities interact in largely separate groups, and with many crypto projects (including the relatively “matured” ones) still in the early years of their formation, the aim is often “community building”. Rightly or wrongly, these communities rally around founders that are often charismatic in their own ways, building for themselves identities and memes which are now characteristic of this Twitter-driven space.

As a result, information transmission is far from perfect, and while each individual community may be highly enthusiastic about a certain piece of news, it is as much a mystery to them as to any other external observer if the initial positive reaction will be sustained by the rest of the market who may or may not be as enthusiastic about their pet project as they are.

Varying levels of conviction, involvement and skin in the game, built on a still-nascent ecosystem of participants who don’t necessarily overlap set the stage for the “overarching” opinion on any individual project to swing violently.

Early days

Ultimately, while the cheerleaders can scream “HODL” at the top of their twitter voices, the reality is that the bulk of participants in the crypto markets are yet to make up their mind as to its viability.

Every observer of the crypto markets needs to accept that these are only the first innings of crypto’s emergence as an asset class. It has to prove itself, and moments of “fear, uncertainty and doubt” do take place – add that to highly levered positions held by relatively small traders/investors and volatility is almost a given.

Our message is simple: a thorough understanding of the market structure of crypto markets is critical to make sense of the realised volatility in prices.

With that understanding, as well as an appreciation of where crypto is in its life cycle as an asset class, the volatility we see is probably much more a growing pain (think grumpy, teething babies or rebellious teenagers) rather than an intrinsic defect.

That’s our view, at least. Differences in views are, after all, what make a market.