Existing crypto ( COIN ) investing literature tends to focus on either finding the “intrinsic” value of Bitcoin ( BTC-USD ) and other cryptocurrencies using network-based models (Metcalfe’s Law, NVT ratio etc) or technical analysis to profit off short-term fluctuations. I’ve covered the former quite extensively in my past work on Bitcoin ( here , here , here , here and here ).
This note is intended to offer a new perspective on Bitcoin investing by identifying event-driven strategies to take advantage of crypto-specific short-medium term dislocations in the price discovery process.
Key events I believe crypto event-driven investors should look out for include – 1) Flow events, 2) Forks, 3) Bitcoin-altcoin cycles, and 4) Credit decoupling events.
Event #1a – Flow Events – ICOs
While the crypto universe is far less prone to indexing and institutional position sizing, forced liquidation events occur far more frequently and can lead to prolonged price dislocations. The most infamous liquidation event in crypto is the ICO-led “fiat drag” where treasury liquidations create artificial selling pressure on funding currencies such as BTC and ETH ( ETH-USD ), leading to a reflexively exaggerated downward move.
The chart below illustrates a typical step-wise ICO process – 1) ICO advertises offering online and conducts raise, 2) Exchanges provide BTC, ETH for fiat currency, 3) Investors buy tokens with BTC or ETH, and 4) ICOs use the BTC, ETH to fund growth.
(Source: Goldman Sachs)
The nature of the ICO process creates a timing difference between when an ICO raises funds (creating buying pressure on BTC/ ETH) and when it utilizes these funds (creating selling pressure on BTC/ ETH). Without further ICO inflows to support the inevitable outflows, treasury liquidations can artificially depress the price in the short run, creating opportunity for investors.
Because most projects are built on Ethereum and fundraise in ETH, ICO flows tend to be most pronounced in Ether. The data reflects this – fundraising has shown a strong correlation to Ethereum’s market cap as ~2% of Ether (~0.5% of crypto) is reinvested into ICOs.
(Source: Autonomous NEXT)
The shift toward ETH as the de facto ICO funder has led to elevated ETH volatility – on most days, ETH vol overshoots BTC’s in both directions. This tends to result in more frequent ICO-driven dislocations in ETH.
(Source: Standard Kepler)
Mapping ETH moved from ICO treasury accounts against price action would seem to confirm this view – the steep post-April ETH drawdown coincided with the highest level of treasury liquidations on a rolling 30-day basis in its history.
(Note: Annotations by Author)
The primary contributor to the spike in ETH moved (note the rolling 30 day average peak) was likely EOS ( EOS-USD ) – a check of their known wallet balances (see below) shows exactly 0 ETH left from the total 654 ETH collected.
The pace and magnitude of the sale (the 654 ETH was liquidated promptly following ICO close) likely sparked reflexive selling and contributed to the massive post-April ETH drawdown.
(Source: Santiment)
The implication is this – the nature of flows and timing gaps in the ICO cycle (illustrated below) creates a time arbitrage opportunity for event-driven crypto investors to short ICO-driven rallies (typically led by mega-ICOs) and subsequently buy into ICO-driven dips caused by the inevitable wave of forced liquidations. For long-term HODLers, ICO liquidation events offer great entry points.
Event #1b – Flow Events – Taxes
The second big flow event – tax outflows – typically revolves closely around key tax dates.
In 2017, it was April 17th in the US and March 15 th in Japan (Japan and US were the two largest crypto players globally). The 2017 tax event was significant because of the sheer size of of taxable gains created by the 2017 bull-run (~$71bn per Burniske) created from the late-’17 crypto bull run (see sensitivity table below).
(Source: Burniske via Medium )
In Burniske’s piece on flow and reflexivity , he then estimates the resulting fiat drag using the following assumptions – 1) 75% of this liability is dutifully paid, 2) 75% of it is paid by selling cryptoassets. The resulting outflow amounts to an estimated ~$14bn.
The BTC chart seems to support this narrative – the March sell-off coincided with Tax Day in both Japan and the US, with a subsequent rally immediately after, allowing Bitcoin to recoup a large portion of its losses.
(Source: Coindesk, Annotations by Author)
Notably, prices seemed to bottom out before Tax Day. Thus, from the perspective of the event-driven crypto investor, there are two key things to watch out for to leverage tax events – 1) key tax deadlines, 2) the likely size of the capital gains for the year, and 3) the likely size of the outflow.
Now I’ve never been an advocate of trying to time one-off events, but I think the key takeaway here is that tax events artificially push crypto prices down before the event, creating an event-driven dip-buying opportunity.
Event #1c – Flow Events – Flow Gaps
The conventional view among crypto investors when thinking about flow events goes something like “if $x amount enters the crypto market, prices should easily go above $y.” Inverting the narrative makes a lot more sense in my view – “if prices reach $y, $x amount of flow would be required to sustain it”.
Adjusting for three key factors – capital gains taxes, ICO treasury liquidations and miner treasury liquidations, the implied flow required to sustain a $6,000 price rose almost six-fold to ~$62bn in 2018, from ~$11bn in 2017.
(Source: Author)
Breaking down the required flow by source yields some interesting insights.
(Source: Author, Note: Assumes tax outflows occur the following year, ICO and miner outflows the same year)
Firstly, the “natural” level of flow required (implied miner-driven outflows) to maintain $6000 has reduced to ~$4.2bn in 2017 and ~$4.1bn in 2018.
Due to the nature of Bitcoin’s supply schedule, the “natural” run rate should skew lower over time per the coinbase reward schedule.
(Source: Bitcoin Wiki)
As non-linear scaling comes into play, transaction fees are also likely to decrease. The underlying “natural” flow run rate required to sustain Bitcoin prices would thus decrease over time, all else equal. This creates a nice long-term flow tailwind for proof of work (“POW”) cryptos such as Bitcoin.
The second key point is the outsized contribution by capital gains taxes in 2018 at ~$45bn post the 2017 bull run.
Note that I have used a conservatively broad ~20% capital gains tax assumption here to approximate the “natural” (post-regulation/ steady state) tax-related outflow – the actual outflow in 2017 is likely to be much lower. As 2018 has seen a massive drawdown in cryptos, it seems unlikely that there will be much, if any, tax outflow in 2019.
Intuitively, this implies that major crypto bull runs create an unnaturally high bar for flows the following year and vice versa. This creates an opportunity for event-driven investors to sell the year after a major bull run and buy the year after a major bear rout.
The third factor – ICO flows – has been covered in more detail above (“Event #1a – Flow Events – ICOs”). Because ICO inflows tend to occur before treasury liquidations (costs need to be paid in fiat) with varying time differentials, this creates a time arbitrage opportunity for event-driven investors to profit from artificially depressed (or inflated) prices following a major sale (or raise).
On balance, tax events tend to be the most dominant flow event (depending on the magnitude of the prior year’s bull run), followed by ICO and mining flows. While mining flows should be more of a long-term tailwind, the timing shift around tax and ICO events create compelling opportunities for event-driven investors.
Event #1d – Flow Events – ETFs and Multipliers
With crypto ETFs a-coming, consensus has been that an influx of ETF flows would introduce a flood of institutional money into the market and bring Bitcoin optimism levels back to the dizzying 2017-highs. I think consensus is right on with this theory but may be misjudging the magnitude.
To measure the impact of a potential Bitcoin ETF, I mapped out net flows at the exchange level against subsequent changes in market cap on a yearly basis. The result is interesting – the net multiplier effect of a $1 net inflow into Bitcoin yields a ~$2,600 change in market cap, implying a multiplier effect of ~2600x, up significantly since 2015.
(Source: Author)
Now, prior attempts at calculating the multiplier has focused solely on estimated outflows as opposed to net flows. In my view, this simultaneously underestimates the multiplier effect and ignores the impact of aggregate inflows/ outflows on the crypto market.
I think the key insight here is less the quantum and more the delta . From 2015 to 2018TD, the net multiplier has consistently increased, indicating higher through-cycle sensitivity to flows/ reflexivity.
The recent coining of the term “Coinbase effect” is a great indicator of the rising multiplier effect. The addition of Bitcoin Cash ( BCH ) to Coinbase for instance, added tens of billions to BCH’s market cap, enabling BCH to outperform the late-2017 bull run.
(Source: Yahoo Finance)
The tricky part about buying into exchange/ ETF-driven events is that the multiplier effect from the additional flows is already well understood and thus, gets priced in relatively quickly. Barring a speed advantage, I believe event-driven crypto investors should instead turn to the less-understood flow/ liquidation events described in 1a, 1b and 1c instead of chasing rallies like these.
I’d view cancellations of mass flow events however, as buying opportunities – the multiplier effect works both ways and tends to lead to reflexive selling, which creates dip-buying opportunities for opportunistic investors.
Event #2a – Forks – The “Unloved” Splinter
Forks are to event-driven crypto investors what spinoffs are to event-driven equity investors –one-off events to buy an unloved, financially-engineered splinter.
There are two types of forks in crypto – the soft and hard fork.
While soft forks are typically just backward-compatible software upgrades, hard forks introduce a new rule into the system which isn’t backward-compatible, forcing a split.
Event-driven investors should keep a close eye on forks – this event is in practice, a crypto-specific form of financial engineering which splinters the chain and creates new financial assets. These assets are subsequently gifted out to recipients a la dividend in specie.
Hard forks can arise out of philosophical differences, to bootstrap a community or to create liquidity among others.
While the number of coins in circulation doubles post-fork, the law of parity does not hold. In theory, the total package of assets before and after distribution should be worth the same, but in practice, the total package (forked + main chain) will be worth (much) more than before distribution.
The Bitcoin Cash fork for instance, created “money out of thin air” (~$5bn worth) in the immediate aftermath, worrying some analysts who found it worrisome that a “relatively large, liquid asset can seem to allow for the creation of ~$5bn of value out of thin air” through the issuance of fission products.
Prevailing theories behind the value creation post-hard fork include reduced uncertainty, illiquidity and the availability of alternatives (increasing the pool of investable opportunities). The first two reasons may ring true in the short run, but in the long run, it fails to explain the sustainability of value creation post-fork.
Case in point – the largest forks continue to hold philosophical and financial value – as a % of their parent’s market cap, both ETC and BCH have been able to retain significant value (~3-5% for ETC and ~5-20% for BCH).
(Source: Autonomous NEXT)
The third reason (increasing the investable opportunity set) makes intuitive sense. Powerful forks are akin to corporate take-overs for an original use case. In BCH’s case, its existence provides a hedge if BTC breaks, as BCH would then “take-over” the community. Similarly, for ETC, it exists as a hedge for potential ETH governance failures.
The tradeable fork event can be divided into two stages in my view – 1) the splinter event rally and 2) the “unloved” splinter sell-off.
The ideal way to take advantage of stage 1 would be to buy the main chain’s crypto pre-fork and sell immediately following the splinter event. But barring a speed advantage, this is not the approach an event-driven crypto investor should take. Instead, the best way to approach fork events is akin to spinoffs in my view – buy the “unloved” forked chain following sell-off (typically begins once exchanges are willing to trade it).
In the case of BCH, the rally post-fork was immediately followed by a sell-off as many BTC holders likely disposed of their BCH for the free gain. The subsequent sell-off artificially depressed BCH prices, creating an attractive buying opportunity for investors – BCH fell from ~15+% of parent market cap to its ~5% floor.
(Source: CoinMarketCap)
Event #2b – Forks – Hash Power Arbitrage
A lesser-known way to profit from fork-triggered inefficiencies is to arbitrage hash power differentials.
This method focuses solely on the equilibrium enforced into a fork by one key party in the network – miners (recall that crypto networks are four-sided with developers, nodes, users and miners playing key roles).
This arbitrage opportunity follows one key philosophy – the price ratio between the forked and main (POW) chains must be approximately equal to their difficulty ratio. Thus, in long-run equilibrium, both chains will inevitably converge towards 100% – the point where both chains are equally profitable.
To understand why this opportunity exists, we must view the fork event from the perspective of the miner.
Forks present each miner with a continuous decision point – mine the main chain, mine the forked chain or a mixture of both (by allocating partial capacity). Due to the difficulty adjustment mechanism embedded within each chain, the key decision criteria for resource allocation is the relative difficulty and the relative reward per block i.
e. the coinbase reward and transaction fees.
Now, pre-fork, the circular miner feedback loop would go something like this – 1) More miners, 2) Increases hash power, 3) Which raises difficulty, 4) Which triggers both miner exits and higher buying activity, 5) Which raises prices, 6) Which increases the reward and 7) Induces more mining. Illustration below.
(Source: Author)
Now, the loop doesn’t quite work seamlessly in practice – the embedded difficulty adjustment mechanism is inefficient and thus, introduces some peculiarities. For instance, the introduction of new hash power leads to changes in profitability for all participants for two reasons – 1) the difficulty adjustments take place slowly (once every 2 weeks for Bitcoin), and 2) difficulty adjustments are based on an unweighted average over that time.
As a result, it can take up to two adjustment periods to fully correct for a spike (new hash power hastens adjustment).
On the flipside, the sudden removal of hash power leads to a destabilizing cycle – 1) it takes longer to mine as the difficulty has not fully adjusted, and 2) the adjustment period takes longer. With an unforked POW chain, this should have little bearing on the miner’s optimal decision point (no interoperability).
But post-fork, things change.
Because we now have two separate, but interoperable chains, miners can reallocate hash power from their ASICs on two (or more) different chains. This creates an opportunity for miners to arbitrage hash power between blockchains, effectively breaking the feedback loop.
(Source: Author)
Because miners now have the added optionality of reallocating hash power (vs mining/ buying or exiting), a miner-enforced equilibrium is reached where the relative price to difficulty ratio on both chains converge at a certain point (parity for BTC/ BCH). The following relationship must hold:
(Source: Author; Note: alpha denotes a constant to account for appropriate adjustment factors e.
g. governance-driven difficulty adjustments etc)
Now, in the case of BTC/ BCH, the difficulty element remains constant in the short run for BTC (adjusts every 2016 blocks) while BCH does not (adjusts every block). The design differences of the respective difficulty adjustment mechanisms can create prolonged arbitrage opportunities in the short run. In the long run however, the miner-enforced relationship essentially forces convergence toward the point where both chains are equally profitable.
The relationship holds very well in practice – following the initial turbulence (BCH began with a wildly fluctuating difficulty adjustment algorithm), the price/ difficulty ratio of both chains eventually settled at parity.
(Source: Author)
A closer look at the relationship in action shows a fair bit of prolonged divergences in the short run, followed by subsequent difficulty adjustments and eventual convergence toward price/ difficulty parity.
(Source: Author)
In the short-medium term, differences in the respective difficulty adjustment algorithms and relative prices can lead to temporary dislocations from equilibrium. This creates an arbitrage opportunity for investors to buy the main/ forked chain trading at a relative discount to the other chain while subsequently shorting the chain trading at a premium.
Alternatively, longer-term investors can use temporary mispricings as attractive entry points.
A simplified backtest shows the fork arbitrage strategy would have yielded ~241% YTD, outperforming a HODL strategy for BTC and BCH by 289%pts and 319%pts respectively.
Fork Arb
BTC HODL
BCH HODL
YTD Returns (to end-Aug)
241%
-49%
-78%
(Source: Author; Note: Backtest assumes accumulated long/ short trades initiated on the day following dislocation)
Event #3 – Bitcoin-Altcoin Cycles
Cryptos, not unlike equities, go through boom-bust cycles. But while the equity cycle is well-understood, the crypto cycle is not, creating compelling opportunities for event-driven investors. As discussed in #1a, the rise of ICOs has led to funding currencies (BTC and ETH) becoming the bellwether of ICO inflows and outflows. Intuitively, the rotation between the crypto funders and altcoins should result in some degree of cyclicality.
The data shows this to be true – cryptos move in well-defined cycles which have become increasingly exaggerated as ICOs gain popularity. As the chart below shows, a typical Bitcoin-altcoin cycle is led by relative gain in Bitcoin, followed by a sharp rally in altcoins.
From an investment flow perspective, a typical Bitcoin-altcoin cycle moves in the following stages – 1) BTC rallies as altcoins remain flat or lose value in relative terms (risk-off), 2) After the BTC run, newly created BTC wealth moves into altcoins (risk-on).
For ICOs, the cycle progresses as follows – 1) BTC/ ETH rallies as investors convert fiat to ICO funders (BTC/ ETH) to participate, 2) tokens are later issued to investors and begin to trade, while ICO treasuries liquidate BTC/ ETH to fund projects, leading to a relative altcoin rally.
The advent of the ICO as well as the rise in crypto funds have resulted in increasingly severe Bitcoin-altcoin cycles. While ICOs may slow down on regulatory concerns, crypto funds should continue to grow, which should sustain the current state of Bitcoin-altcoin cycles.
Opportunistic investors can leverage the cycle by either 1) taking a contrarian approach by accumulating altcoins in a Bitcoin rally and vice versa, or 2) adopt a trend following approach by accumulating more Bitcoin on a Bitcoin-altcoin rally and vice-versa.
Event #4 – Credit Decoupling Events
Existing Bitcoin pricing models such as Metcalfe’s Law depend entirely on network-specific metrics such as number of users and transaction activity. As I elaborated here , prevailing pricing models fail to account for the flow effects from Bitcoin’s existence as a parallel financial channel. As this key missing variable is not well-understood by the market, significant decoupling events create opportunities for investors to profit from temporary dislocations.
Mapping out Bitcoin prices with a BBB spread overlay (using ICE BofAML US Corporate BBB Option-Adjusted Spread as proxy) show a striking inverse correlation between both variables. Major decoupling occurred only in the ’14 – ’15 period when a crypto-specific external shock (the Mt Gox hack) occurred, and post ’18 when credit began tightening.
(Source: CoinMarketCap, FRED)
Accounting for flow effects using BBB credit spreads as a proxy not only improves robustness of a typical Metcalfe’s Law-based pricing model, it also negates the Granger Causality failure triggered by network models. Unlike network variables which are actually led by Bitcoin prices in practice, credit spreads have been found to lead Bitcoin prices.
The takeaway is this – widening/ tightening HY spreads should lead to lower/ higher BTC prices in the medium-long run.
Deviations from this trend in the short run creates an opportunity for investors to position accordingly to take advantage of a decoupling event.
Final Word
To recap, I’ve highlighted four key plays event-driven crypto investors should look out for including – 1) Flow events, 2) Forks, 3) Bitcoin-altcoin cycles, and 4) Credit decoupling events.
While there is no shortage of commentary on crypto investing from a long-term HODL and short-term technical perspective (most investors can play this theme through the Bitcoin trust ( OTCQX:GBTC ) and ETN ( OTCPK:CXBTF )), there is a clear dearth in literature on investing in special situations and event-driven plays. The inefficiencies and intricacies unique to the crypto universe is not well-understood and holds a wealth of opportunities for a different type of investor to capitalize. This note is intended to plug the existing gap and spark debate around such opportunities.
Disclosure: I am/we are long BTC-USD, ETH-USD.
I wrote this article myself, and it expresses my own opinions.
I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article..
