Big enough to control
“Instead of investing in gold as a hedge against inflation, a hedge against the onerous problems of any one country, or the devaluation of your currency whatever country you’re in – let’s be clear, bitcoin is an international asset, it’s not based on any one currency and so it can represent an asset that people can play as an alternative.”
– Larry Fink, CEO BlackRock, July 2023
While it is encouraging to have Larry switch sides and advocate for bitcoin, I might hypothesise on the change of strategy from the US. It is clear from actions across the rest of the world that bitcoin is being welcomed in some quarters and the US is pivoting, deciding to take a strategic position.
This sentiment shift is reflected in a judgement yesterday from a lower court judge in the US in the ongoing case of Ripple v SEC. Ripple jumped 70% on the news that the token itself is not a security and people trading it are not committing a violation, specifically:
“The court finds that “Programmatic Sales”, eg offering XRP on crypto exchanges, does not constitute the offer and sale of investment contracts.”
The victory was not universal though because the initial issuance of the token (the ICO) was deemed in breach of the Securities Act.
Whatever happens, the shift is meaningful. If the SEC can’t even take Ripple down then it is highly unlikely now they will expend their resources pursuing others. Regulating and moving on is the only sensible approach. I’d say that is where we are going.
Where does that leave the larger assets? Right now, bitcoin is small. 4% of gold’s market cap. The exchanges are disparate and global. The major holders of bitcoin are all over the world. It’s hard to control. Imagine a scenario where the ETF’s get approved; BlackRock, VanEck, Cathie, Grayscale and co. Steadily they acquire 20% of the bitcoin in circulation. That would be a significant and strategic stake and probably well within reach within a few years.
At that point the US would have a large part of the asset within its companies since Coinbase is likely custodian for the proposed ETFs. Imagine legislation at that point “for your safety” that says if you buy bitcoin for your own use and custody there is a special tax you must pay, or a certificate you must get to prove competence, some barrier anyway. If you don’t want to pay those special taxes then buy the BlackRock ETF; on its own that will be sufficient to push people down the ETF route. They do not need to worry about custody or training and it’s done. A large proportion of the supply then becomes a matter of the whim of the US Government. If they can get a handle on bitcoin, they will have a handle on everything else too, it’s the anchor asset of the sector.
So paradoxically, a good strategy for the government might be to let it get big. Using other people’s money to buy bitcoin within institutions you control and then exerting influence over it is easier. Some people will get wildly rich as a result but that will be the price the USG will have to pay.
A $400,000 bitcoin puts the market cap around $8 trillion, smaller than gold, bigger than silver. It would also fluctuate less at that point and consequently might be less interesting to people. That number is not randomly selected. It is a guesstimate of what is not so high as to be an offence to the government while being big enough to be controllable. It will be even more tolerable if you have managed the journey to $400k, which perhaps the USG will do.
One thing in my mind is sure. Larry Fink is not risking his firm’s reputation for a 2x return and neither is he trying to kill off the USD.
Men’s haircuts
One of the few things I recall from university economics is the indicative power of the cost of a men’s haircut. They are a tremendous proxy for estimating the wealth of a nation. Looking at this list it would be hard to argue; despite some anomalies like Finland, which looks low.
The reason is haircuts are not tradeable goods. You might travel overseas for a high cost medical operation but nobody is jumping on a plane for a short back and sides. This theory extends even to the local level. There may well be a cheaper haircut three blocks away from where you currently go but nobody changes barber to save $5. The relationships are sticky. The price simply reflects average wages in the country and then rents in the area you go. Expensive part of town; more expensive but net outcome for the barber is similar.
The profession is even more interesting because there isn’t much you can do to increase your earning power as a hairdresser other than hope that the country you are in either is or gets rich. So, in selecting where you wave your scissors, country choice and then location choice are paramount and once those choices are made you are pretty much stuck with the income it generates. As expensive an area as possible is probably best.
The same is not true of women’s haircuts, mostly because the range of options is much wider. It is possible to spend $50 up to $1,000 because there are generally more things happening with a female haircut. It is also more likely because of the time involved that coffee and nibbles will be provided which throws the pure price comparison of the haircut off because the whole experience includes quite a few tradable goods.
All of this is a matter of interest only, but for those of us that don’t really believe government statistics, the haircut proxy might be much more informative than we thought. Namely, Germany isn’t doing all that well at the moment and has probably been struggling for a while, on any official measure it would have haircuts over the $40 mark.
The message in all of this is do not lament the cost of your haircut. Perversely, for all non-tradable goods, the higher the better.
Machine payments
For a machine to make payments to another machine, those payments need to be both programmable and permissionless. That means no counterparties, banks, or KYC. Software only.
This particular query to Open AI’s GPT-4 is asking for a payment gateway to an API. The AI immediately codes for the Bitcoin Lightning Network. It is incredibly simple to do (at least for someone with basic programming skill) and it permits nano-payments. So you can pay 0.000001 cents for an API call, something that is not possible with traditional payment mechanisms.
It shouldn’t be a surprise that capacity in the Lightning Network continues to grow; it is just incredibly useful on so many levels and increasingly easy to do with the help of AI. That the whole thing is growing with so little mainstream fanfare is extremely encouraging.
1 billion tokens
I was sent a rather eccentric video from an AI enthusiast last week. I say eccentric because the presenter is dressed in a Star Trek captain’s outfit for no apparent reason (other than he’s probably a fan).
The video discusses a paper from Microsoft Research called LongNet. Those of you currently using OpenAI will be familiar with the context limits. Currently, you can only give the AI so much context for your query, depending on what you are using, that might be a few pages or an entire chapter of a book, but no more. I use GPT-4, with a maximum token input of 32K.
LongNet permits 1 billion tokens. It is completely insane and suggests the model can consume almost everything that has ever been written as context for a single query. It also does it in a very novel way which our Star Trek hero does an excellent job of explaining.
I would dismiss this as unlikely if it hadn’t come from Microsoft Research. It is potentially enormous and the equivalent of always having the leading experts on every subject at all times on hand to help you. There is standing on the shoulders of giants, and then there is standing on the shoulders of all giants at once.
There is a lot of hype around AI. Personally, because I use it every day I admit to being in the hype camp. A lot of technical prophecies are also coming true as a result and I would remind you (again) of the Isaac Asomov video on the pace of change. Massively accelerated change is possible and actually likely.
Maybe hype, it also might be exactly as a lot of people have in the past predicted.
Euro-Trash
Euro quarterly update time. Our chart is getting too small to read so for completeness the whole thing is provided along with a condensed version beneath.
Suffice to say the Euro had a decent quarter against the Dollar. Rather to my surprise too given the much higher pace of rate rises across the pond. Level pegging.
When measured against Bitcoin, the Euro has now lost some 70% of its purchasing power in the three years since we started measuring its progress. That is despite Bitcoin itself having a peak-to-trough drop of a similar amount over the same period. That’s the thing with steady devaluation, you do not really notice.
Elsewhere, the ECB released their Central Bank Digital Currency proposals. They are extensive. We can conclude from this that this is happening and I am actually excited about it. The most fascinating thing is that the money is issued by the ECB itself, so theoretically holders will have an account with the Central Bank; ultimately it probably won’t work like that but if you were a European retail bank you might have reason to be worried.
What is also obvious is that people still love cash. The feedback to the ECB was overwhelmingly that people value the privacy afforded by cash. Right throughout the documents they hammer home that cash is not being replaced.
The CBDC era is here, we will cover it more next week. FedNow is coming, the Bank of England is also well on the way with its own currency. A new era, that frankly was ushered in by bitcoin. Would they be doing this if they had not seen the innovation in digital payments?
Far from undermining the leading crypto-asset, I expect CBDCs to shine a light on the fundamental weakness of centrally managed issuance. Bring it on.