🔍Deep Dive: A Model for Bitcoin's Epoch V

1. Introduction

Welcome to the inaugural issue of Bitcoin News Research, an exclusive collaboration between Bitcoin News and 21st.Capital aimed at providing you with high-signal Bitcoin-focused research and analysis.

What to Expect:

Our goal is to provide you with unique, data-driven insights with a focus on the three core themes:

• Bitcoin fundamentals (on-chain data, valuation models risk metrics).

• Macroeconomic and financial trends.

• Investor behavioral (spot, futures, and option markets).

Estimated Read Time: 0.8 Blocks (~8 Mins)

About Me:

My name is Sina (@Sina_21stCapital), and I’m both a professor of business and current co-founder of 21st.Capital, a specialized Bitcoin custody and consulting service. I am predominantly focused on data-driven business research.

Before my current role, I completed a Ph.D. and a post doctorate in business and worked as an analyst in Silicon Valley. I am dedicating all my attention to Bitcoin as I believe it is a worthy cause that will drive enormous positive change in our world.

Newsletter Overview:

We have divided this newsletter into three sections, namely:

• The Supply Story: In the first section, we discuss Bitcoin as the hardest money known to human-kind and top it off with what Mises, Rothbard, Hayek, and Keynes thought about money and inflation.

• The Demand Story: The second section offers an update on the status of Bitcoin spot ETFs, which are among the most important demand-side factors this year.

• Bitcoin Model: The final section offers a statistical analysis of Bitcoin’s potential price trajectory in the new cycle taking into account diminishing returns and cyclical movements.

2. The Supply Story

2.1. Apolitical Monetary Policy

Halvings are pivotal events in Bitcoin's lifecycle where its monetary policy adjusts transparently and automatically. These events underscore the notion that monetary policy need not be political. In contrast to traditional fiat systems, Bitcoin operates independent of any central authority and does not require technocratic adjustments to its supply schedule. Bitcoin is a permissionless bank in cyberspace with automated, transparent, and apolitical monetary policy.

Since its inception, Bitcoin has undergone four halvings, creating five distinct cycles. With each halving, the block reward halves, significantly impacting bitcoin miners' earnings. Below is a summary of the block rewards paid to miners for every Bitcoin cycle. As seen below, the most recent halving occurred on April 19, 2024, reducing the block reward to 3.125 BTC.

2.2. Harder Than Gold

This latest halving was particularly special since it made Bitcoin the ‘hardest’ asset in history. Bitcoin's inflation rate dropped by half to 0.8%, below Gold's inflation rate of 1.4%.

Historically, a high initial inflation was seen as the necessary evil tolerated to bootstrap the network and subsidize miners while the network was young. The network is increasingly moving away from relying on inflationary miner subsidies, transitioning instead to a transaction fee-based model. In the process, Bitcoin becomes a harder form of money, aligning with core principles of sound money. 

By lowering its inflation rate, Bitcoin also becomes more attractive to investors seeking to divest from other inflationary assets, potentially triggering the next price appreciation cycle. If centuries of monetary history are any guide, markets prefer the hardest asset for wealth preservation and allocate capital to these asset classes.

2.3. Why is ‘Hardness’ Important?

For centuries, economists have debated which monetary properties are most ideal and how best to conduct optimal monetary policy. Famed economist John Maynard Keynes believed that money should be managed by the state to ensure stable prices are maintained. He thus formulated the theory of state-managed money that aimed at keeping prices stable (neither inflation nor deflation). Contrary to popular belief, Keynes himself understood and discussed the dangers of inflation.

However, he was prepared to accept temporary inflation as a lesser evil compared to deflationary episodes. In practice, when governments gained full control over money and abolished the gold standard, they did away with the “price stability” part of Keynesianism and began an era of indefinite inflation, as outlined by Austrian-American economist Ludwig von Mises.

Other Austrian economists such as Hayek believed in the fixed supply of money and favored the gold standard. Hayek believed that Keynes’s ideal of artificial price stability is counter to the nature of human action and hurtful. He favored apolitical money.

Foundational to Austrian economists’ criticisms of the fiat standard was an understanding that managed money is political money. As such, political money inevitably becomes corrupt money which then functions as a tool for enriching insiders and impoverishing outsiders – the Cantillon effect. Murray Rothbard, one of the most ardent advocates of the gold standard writes: 

As attractive as gold is as a store of value, its supply is still subject to manipulation. As the price of gold rises, so too does the incentive to mine it. The chart below shows how strongly the production of gold follows its price. This increased supply prevents gold from reaching its full potential for price appreciation. 

Bitcoin does not have this limitation. Its supply is mathematically capped, and it is the only asset known to humankind with a truly fixed supply. Compare this with the virtually unlimited demand for other ‘store of value’ assets, and one realizes that Bitcoin is well positioned to become the best-performing asset in the coming decades. Bitcoin represents the manifestation of unlimited demand mapped against a mathematically-certain fixed supply. Bitcoin: unlimited demand meets fixed supply!

2.4. Not Enough Bitcoin!

If you heed Rothbard’s warning and want to avoid relying on central bankers to manage the supply of your money, you now have access to the world’s hardest asset. Bitcoin’s limited supply of only 21 million coins means that most people won’t be able to own a full Bitcoin (there aren’t even sufficient for every millionaire!). To put the math into perspective, there are currently ~8.1 billion people on the planet; 62 million of them millionaires – 1 BTC for every 400 people; 1 BTC for every 3 millionaires. 

Therefore, if everyone wanted Bitcoin, they’d be able to own 0.26% of a bitcoin or 259K SATs on average. If every millionaire wanted Bitcoin, they’d be able to own ⅓ of a Bitcoin each, on average.

3. The Demand Story:

In addition to the reduction in supply due to the halving, this year also saw a significant boost in potential demand. On January 11^th , 2024, Bitcoin spot ETFs were approved by the United States Securities and Exchange Commission. This paved the way for Wall Street and traditional finance firms to integrate Bitcoin into their portfolios. So far, the ETFs have attracted a cumulative flow of ~$12 Billion.

The big winners among the newly launched ETFs were Blackrock’s IBIT and Fidelity’s FBTC which brought in $15.5B and $8.1B, respectively. All ETFs combined, have attracted a staggering $29B, excluding GBTC.

On the other hand, Grayscale’s GBTC relentlessly sold a cumulative $17.5B worth of Bitcoin and has lost 53% of its assets under management, down from 619K coins at the beginning of the year to 292K as of May 2024. 

Since its launch, IBIT has been scooping up most of the coins that GBTC has sold.

Interestingly, Grayscale’s nominal management fee (denominated in dollars) has only seen a minor impact due to Bitcoin’s strong year to date price appreciation. Up until mid-March, GBTC’s management fees had remained unchanged compared to early January. This may explain their reluctance to swap over to more competitive fee schedule.

However moving into May, their monthly collected fees (blue chart below) strted declining more rapidly; we estimate that Grayscale was collecting ~$36M/month before the ETF launch, a figure that is now down by 33% to ~$23M/month.

We also note that the ETF flows have slowed down. After several weeks of explosive growth, marking the most successful ETF launch in history, the excitement seems to have now subsided. Ever since the launch, we have been fluctuating between positive and negative weekly inflows. 

This is to be expected though following the initial burst of pent-up demand that was unleashed from institutions and investors. Now that this demand is satisfied, a slow and steady stream of investors will likely continue allocating toward these ETFs as they gradually complete their due diligence.

Notably, we saw a major buy-the-dip move on both Monday (April 29th) and Friday (May 3rd) where all ETFs including GBTC received positive inflows.

4. What to Expect this Cycle?

So the question that remains - what is next for Bitcoin’s price trajectory? To answer this question, we modeled Bitcoin’s growth from day one and projected it up until the end of the current (i.e., fifth) cycle or “Epoch”. Bitcoin’s price has grown from $0.05 to more than $60,000 over the last 14 years. 

If the same pattern more or less continues, what prices will we see in this upcoming fifth cycle? 

4.1. Linear Growth Model

A simple way to project Bitcoin’s growth path is to use a regression model to analyze what happened after prior halvings. Below, you can see our projected growth trajectories modeled after the percentage returns of cycles 2-4. For example, if Bitcoin follows the same trajectory as cycle 4, we expect the price to go from $62K to approximately $500K at the cycle peak.

The regression model used for the above predictions is given Price_t=α+ βCycleDays_t+ ε_t , where CycleDays is a vector of dummy variables representing different days since the cycle began.

4.2. Average Price - Diminishing Returns

However, it is not enough to apply the previous percentage returns to the next cycle. The average price follows a log-log curve (i.e., a Power Law¹) showing diminishing returns. Modeling Bitcoin in the log-log space produces a much better fit, and in my assessment, it fits amazingly well with strong out-of-bag predictions.

If we regress the logarithm of price on the logarithm of time, we get the following prediction as a baseline to model the average price level. You can see how well this curve fits Bitcoin.

Note that diminishing returns do not mean “weak returns” by any means. If the historical pattern continues (which it has been following very closely so far), a 200x or even 50x would be unlikely in a 4-year cycle.

All this is natural. As an asset matures, it gets less volatile, and explosive price moves give way to measured, more predictable, and lower-risk moves. This may even result in an improvement in risk-adjusted returns over time, helping the asset attract additional institutional capital. Even with diminishing returns, Bitcoin can still reach +$1M this decade.

1. Santostasi, G. 2024, April 15. The Bitcoin Power Law Theory - Medium

4.3. Adding Cyclical Movements

After modeling the average price, we will add the cyclical movement patterns using the term CycleDays which is a vector of dummies for each week after the cycle begins. So, we are modeling not only the average long-term price trajectory but also the cyclical ups and downs.

Important note: this is only an example illustrating what will happen if Bitcon follows a pattern similar to the previous cycles. Whether the cyclical patterns will actually be similar to the past is another question. We have only had four prior cycles, and that is not enough sample size for reliable predictive purposes, but this analysis provides a good illustration of what happens if the general pattern in prior cycles repeats.

4.4. Final Model

Finally, the variance has been dampening too, i.e. the max spike in each cycle gets smaller over each consecutive cycle. We can allow the cyclical deviations to change over time by interacting CycleDays and Time, helping us build a model that more accurately projects the future price trajectory. 

This model now produces more reliable predictions, and we can then train this model on cycles 1-4, helping us estimate the price for cycle 5. Below, you can see the logarithmic chart of these estimations. 

Let us zoom in on cycle 5 and switch to the original non-logged scale, as shown below. The model predicts that the price will reach $400K at the top, with the average prediction reaching $200K by the end of the cycle. Again note that this is if the future cyclical movements follow the general pattern seen in the past.

The fundamental supply and demand-side stories that we discussed earlier support Bitcoin’s continued growth implied by this empirical model. This next cycle will be full of exciting news for Bitcoin (which will also bring in more hostility from the losing actors).

Thank you for reading the first edition of Bitcoin News Research. As Bitcoiners, we firmly believe in the value of proof of work. We’re offering this unique content completely for free. All we ask is that you share it on Twitter with one friend to help spread the word and show your support.