Regarding… Vol 2 no 28 – Crypto – Buy, Hold or Sell?

Dear Fellow Investors and Friends,

Welcome to this week’s edition of my investment letter, where I try to make sense of the world around me and share stuff that I find interesting.

I do appreciate you taking the time to read this.

Today is Thursday, July 25th, the 207th day of the year. There are 159 days left until the end of the year.

Monday was Pi day – you know, 22/7. Written as π. The most famous mathematical constant.

The formula for π is the circumference of a circle divided by its diameter. It works out to 3,14159265….to infinity; the sequence never repeats itself and has no ending. Some people find it a fascinating number because of these properties. Kate Bush is one of them and wrote a song about it, called…er, Pi. In the song, she sings the actual number to its 78th decimal place, then from its 101st to its 137th decimal place. Apparently, she gets the order wrong in one or two places – but I haven’t checked. You can listen to it here. See if you can determine whether she made a mistake. It’s better than watching the Nvidia share price.

On to more maths, in the form of all things crypto. As promised last week, today I am exploring the value of crypto, if any.

The market value of all cryptocurrencies is $2.5tn, with Bitcoin’s market value at $1.3tn, just over half the total. The whole crypto market is smaller than Nvidia and a fifth of the estimated $14tr value of gold above ground.

Dylan Grice of Calderwood Capital wrote a piece about Bitcoin in 2020, in which he developed the framework for thinking about the “value” of Bitcoin as follows:

“There are two arguments around the “value” of gold.

The first is that gold has no ‘valuation’. In other words, gold does not fit into the standard capital asset valuation techniques of cashflow forecasting and discounting. In real terms, its NPV is its current value. Its long-term expected real return is zero. Therefore, its long-term expected nominal return is the expected inflation rate. It’s simple enough, but not a satisfying conclusion for anyone trying to figure out if the metal is undervalued or overvalued.

The other argument is that it has no “use value” like other commodities, oil, copper, etc. However, something that has the property of being useless for everything else is exactly the property that makes that something useful as a currency. That’s because a currency is a numeraire in whose terms all other goods can be expressed. Its role is to facilitate the distributed computation we call economic activity by encoding information that signals each market’s scarcity conditions into a single price for participants to process. That function is performed best (i.e. the most accurate information is conveyed) when the numeraire is stable, and it will be at its most stable when it has no other uses.”

Most assets are liabilities on somebody else’s balance sheet. The cash you have on deposit at your bank is the bank’s liability. Government bonds are a liability of the government. Equities are even more junior than debt on the counterparty balance sheet. For an asset to be a useful long-term store of value, it shouldn’t be beholden to a debtor’s actions.

The more ‘useless’ an item, therefore, the better a candidate it is to act as a store of value. That’s why humanity has, at various times, used shells, beads, lead ingots and other useless items as a means of exchange.

Bitcoin, specifically, and Crypto in general, also ticks this box.

For example, gold, which has limited economic use, has held its value over the centuries.

Just like gold, it’s hard to put a specific value on crypto. Various attempts have been made, including this one:

Personally, I wouldn’t put much weight on such quasi-mathematical attempts at valuation. It didn’t work in 2021, and it won’t work now.

It’s more helpful to think about an incentive price for Bitcoin. In other words, what does it cost to mine it?

Just like any other kind of commodity, over time, this should act as a floor price. If the price of a physical commodity drops below the incentive – floor – price, mining becomes unprofitable, and mines are closed. Eventually, the supply of the commodity declines to a level where it can no longer satisfy demand. At this point, there will be upward pressure on the price to the point where it becomes profitable to mine again. Ultimately, new mines are incentivised to open if the price goes high enough. And so it goes.

In the case of crypto, if the price drops below the incentive price, “mining” becomes unprofitable, and computing power will be taken offline, causing the hash rate to decline. This reduces the cost to “mine,” eventually making “mining” profitable again. As it becomes more profitable, more computing power is added to the system, increasing the hash rate. And so it goes.

Fundamentally, this incentive price is the fulcrum around which the value of cryptocurrencies oscillates. This price is determined by:

1. Hash rate: The aggregate processing power in the system and equipment quality. The higher the hash rate and the faster the chips, the harder it is to mine, and the cost of mining goes up.
2. Network effect: If your equipment gets better, the cost of mining goes down, as you can process the algorithms faster. Also, older, slower machines can’t compete and get switched off. The system’s processing power declines, and it becomes easier to mine, reducing costs.
3. Energy cost: Running the algorithms that validate blocks is energy intensive. As the cost of energy changes, mining costs increase or decrease.
4. Halving: The rewards for mining Bitcoin are halved every four years, guaranteeing an 18% p.a. increase in the cost to mine or incentive price.

There are currently around 18.5 million bitcoins and a maximum possible supply of 21 million. The number is encoded and can’t be changed. Supply inflation is around 1% p.a., but this will decline to zero by 2040.

Those who dismiss Crypto as nothing more than “number go up” might be identifying a kernel of truth.

Here is a chart showing how closely the price of Bitcoin tracks its production cost over time:

Source: https://en.macromicro.me/charts/29435/bitcoin-production-total-cost

The yellow line denotes the cost of “mining” Bitcoin, and the blue line shows the price of Bitcoin. They track each other quite closely. Historically, the cost to “mine” Bitcoin, and, therefore, its price, grows at c.40% p.a. (note the logarithmic scale on the chart above).

The bottom green line shows the deviation from the “fair” value or the incentive price. According to this, Bitcoin is currently priced a little bit rich. However, the overarching point is that the value of Bitcoin is not some arbitrary number but is fundamentally determined by the cost of “mining” it.

Next week, I’ll examine some of the standard arguments against Crypto.

UPS is the world’s premier package delivery company and a leading global supply chain management solutions provider.

UPS could be regarded as a bellwether for the global economy. If so, things are not looking good:

Like many other businesses globally, the company faces lower volumes of goods transported and higher costs due to inflation. This is compounded by excess market capacity because of rapid expansion in response to the post-COVID e-commerce boom.

If you want a reliable leading indicator of trouble in any sector, look for where expansion capex is being (over) spent. Supply chains – of which UPS is a crucial component – went awry during COVID-19 and haven’t yet recovered to normal. It will take a while for normal levels of trade to fill the expanded capacity in the system.

My take: How will we know when this is the case? The share price will tell us. In July 2020, amid the COVID lockdown idiocy, it told us that demand would explode. Right now, it’s saying there is too much capacity for the current level of demand.

Chow Tai Fook is a Hong Kong-based retailer of luxury jewellery. Like the rest of the luxury goods sector, it is facing headwinds.

In a recent result announcement, they announced that retail sales were down by 20% year over year. That’s a big decline! They are closing stores in mainland China (95 out of 7504, to be precise). Both gold and gem-set jewellery declined by double digits.

It’s not looking good out there in the land of Chinese luxury goods:

As discussed here, the downturn in China has spread from property to drinks and now to luxury goods. Any business that relies too heavily on China will be tested. Also, when the communists are forced to choose between “shareholder value” and “common prosperity” – you know what their choice will be.

My take: Buying opportunities will emerge, but it’s early days.

In South Africa, the market is broadening out, and more stocks are making new highs than new lows. But in the USA, this is not the case; the market had become narrow with just a few big stocks making all the running. Now, even the leaders are struggling. So, I think it’s noteworthy when a stock makes a decisive new high in this environment.

Lockheed Martin describes itself as a global security and aerospace company. However, the majority of its business is with the U.S. Department of Defense and federal government agencies. It is one of the biggest defense companies in the USA.

Ranked by market cap, the sector looks like this:

• RTX (aka Raytheon) has a market cap of about $135bn
• Lockheed Martin is $115bn
• Boeing is $110bn, but peaked around $250bn
• General Dynamics is $80bn
• Northrop Grumman is $65bn

Elon Musk’s business, SpaceX, founded in 2002, recently raised money at a $210bn valuation, giving it a bigger market cap than all the major defense companies. It’s probably unsurprising that Musk is cosying up to his biggest prospective customer, Donald Trump.

The bottom line is that it’s glory times for the military-industrial complex, according to the Lockheed Martin share price:

My take: I read a piece by Doomberg this week, which included the following quote: “For people who make bombs, peace is a working capital issue.” I think the people who make bombs will be fine. It’s us we should worry about.

I find it amazing that trips to the moon have taken off after almost a generation of nothing happening.

Source: Liberty’s Highlights Substack (I’m a subscriber)

Also from Liberty:

Why is all this happening now? Why is no one talking about this? As a 7-year-old, I remember staying up all night to watch Apollo 11 land on the moon. It was all anyone talked about. Now, it’s not even worth mentioning.

My take: In my first newsletter, I wrote about India’s (third!) mission to the moon, which did not make any headlines. Now Japan is up to 6, China is on 9, and the USA has gone from 40 to 49. All over the past 3 years. What’s going on?

These charts are all courtesy of YWR:

While this particular capex bubble is still going on, the market is all in on the stocks involved.

No one is short:

And large-cap funds (the kind that own the companies spending all the capex on Nvidia’s chips) have seen 4 years of massive inflows:

Also, Amazon, Meta, Microsoft, and Alphabet’s cumulative net income over the past three years was $579bn. Of that, they spent $418bn on property and equipment purchases. Capital light, much? What are shareholders paying for and what are they getting?

So far, only Amazon has said the quiet part out loud. They recently announced that, despite spending tens of billions on AI applications, the commercial case for such investments had yet to emerge.

Here’s an interesting fact: the Wiltshire 5000 index, which was created to include all the stocks in the U.S. market, currently contains only 3370 companies. No one cares about small caps. They are being taken off the market by knowledgeable insiders, both in the USA and here in South Africa.

Where should you put your money? With the commission-driven sellers of large-cap ETFs or with the knowledgeable buyers of their own (small) companies?

My take: I wrote about this capex situation in February of this year, using Microsoft as a specific example. You can read my comments here. Since then, the Microsoft share price has increased, and so has the capex spend.

I said it then and I’ll say it again: this will not end well. And if it does, it will be the first time ever. The odds are not on your side here.

Neal is the author of Snow Crash (and the person who originally coined the term Metaverse), and Tim Sweeney is the Founder and CEO of Epic Games, which makes the Unreal Engine and Fortnite. Matthew Ball is probably the pre-eminent writer of the tech scene.

The conversation is fascinating. The future is happening more quickly than we can fathom.

The money quote came from Sweeney: “But in the future, the ideas from various blockchains such as zero-knowledge proofs, the idea of cryptographic consensus protocols and so on, should be a key component of many systems.”

Read it here.

The Diff is a newsletter by Byrne Hobart (I am a subscriber), which usually has some very interesting takes on the world. In this letter, he talks about how software is changing its nature due to AI.

Software businesses are classically high fixed cost, low or no marginal cost – so as you scale, profitability goes through the roof.

Hobart argues that AI is reducing the cost of creating new software while increasing the cost of running and maintaining it, making it a less attractive business model.

You can read the whole article and other interesting riffs here.

Rick Rubin, a legendary music producer, hosts the podcast Tetragrammaton. He is also a co-founder of Def Jam Recordings and American Recordings and a former co-president of Columbia Records. You could call him the father of hip-hop music, as he worked with LL Cool J and Run-DMC and produced luminaries like the Beastie Boys and Public Enemy.

This is a picture of him:

Yes, he’s a middle-aged white guy.

Wikipedia says the Tetragrammaton is the four-letter Hebrew theonym יהוה‎, the name of God in the Hebrew Bible. The four letters, written and read from right to left, are yodh, he, waw, and he. The name may be derived from a verb that means “to be”, “to exist”, “to cause to become”, or “to come to pass”.

It’s a very appropriate description of his quiet interviewing style, in which he allows his guests a lot of space “to be” –  to define and describe their territory. He also lets the listener make up their own mind about what’s happening.

He always has super interesting guests. The one I listened to this week was with Chris Dixon, author of the book “Read Write Own: Building the Next Era of the Internet “ Chris is also a partner at venture capital firm Andreessen Horowitz.

This podcast is about his book and how he sees Web 3.0 developing in a decentralised environment, away from the centralised sandboxes of Web 2.0. Of course, he sees blockchain as playing an essential role in this process. His book is worth reading, but the podcast perfectly introduces his thinking. You can listen to it on Apple or Spotify.

With the Tour de France having finished in Nice over the weekend, I thought I would share these two clips:

This year, the tour went up the Cime de la Bonette. A 23km climb at 6,8%, topping out at 2800 meters. Of course, there would be a crazy Dutchman who attempted it on a kid’s bike the morning the Tour went over the climb.

It’s hilarious, but it also gives you an idea of how long, high and challenging climbs of this sort are. And how enthusiastic the crowds are!

You can watch it here (it’s in Dutch, with subtitles)

The Tour also went over the Bonette in 2008, the year South African neo-pro John-Lee rode for Team Barloworld. He led the peloton over the top of the climb but missed a corner on the descent. Crashing over the low retaining wall, he landed on some steep scree and had to be rescued, as he was in danger of falling a long way down the mountainside. Fortunately, he wasn’t injured.

You can watch the drama unfold here.

Amanda and I will be going to Nice in September, and I hope to ride the Bonette and see it for myself. I am unsure of a lot of things, but I’m pretty sure that will not be the only climb on our itinerary.

That’s all for this week!

Except to say: remember to be very, very careful out there!

Piet Viljoen
RECM