Issue 112

Chipageddon

Delivered on 10 May 2021 by Justin Pyvis. About a 5 min read.

The ABC (an Australian government news organisation) ran an article 'explaining' the global microchip shortage, titled What does chipageddon have to do with climate change?

As you might have guessed by the title, the authors attribute the global chip shortage – chipageddon – to a drought in Taiwan (its worst for 56 years, according to the BBC), which they say "is happening in the context of a changed climate, particularly the extremes". At first glance, that makes sense – chip production requires a lot of water and the ABC cites a 2012 journal article that concludes climate change will lead to "a rise in drought frequencies" in Taiwan (as well as increased floods).

There are two major problems with the ABC's article. First, the paper cited only modelled the years 2080 - 2099. It's a bit of a stretch to say the current drought "is happening in the context of a changed climate", then citing a paper that supports that argument by modelling climate change 59 - 78 years into the future.

But the biggest problem with the article is there's no evidence a lack of water has actually affected Taiwan's chip production.

Due to the drought, the government has been transporting water from less affected regions for use in technology parks (where most microchip factories operate). So while agricultural crop yields might be damaged due to the lack of typhoons in 2020 (around 70% of water usage in Taiwan is for relatively lower value agricultural use), microchip manufacturing has so far been spared.

Don't get me wrong: climate change is something Taiwan's chip producers will have to manage going forward. But it can be managed – Taiwan's biggest issue is not the drought but a chronic underpricing of water. Taiwan has the cheapest water in Asia and one of the lowest water prices in the world. That leads to inefficiencies all the way down the supply chain, from wasteful behaviour by households to entire industries being built around artificially cheap water.

Fix prices, fix the drought.

As an addendum (because you won't learn much about chipageddon by reading that ABC article), the current chip shortage stems from the combination of the US-China trade dispute disrupting supply chains (Huawei reportedly stockpiled six months' worth of inventory) and a huge miscalculation on behalf of manufacturers around the world. According to Taiwanese chipmaker TSMC, "In March, 2020, as COVID paralysed the US, car sales tumbled, leading automakers to cancel their chip orders. So TSMC stopped making them."

Global manufactures panicked when the pandemic hit and cancelled their orders. Suppliers stopped producing. Now they're crying shortage because the rebound in the demand for durable consumer goods has been so intense. Hence, chipageddon.

The Maher effect

Old man yells at Bitcoin
Source: Ben Kaufman/Twitter

Many of you may have seen comedian/talk show host Bill Maher's hit job on Bitcoin last week. He clearly doesn't understand crypto and makes a lot of mistakes in his analysis, but he tends to do that with everything he doesn't understand:

  • On vaccines, "It's presented as this genius medical advancement. No it's actually this risky medical procedure."
  • On camera phones, "It's a phone not a Swiss army knife."
  • On social media, "I already know what the captain of the football team is doing these days: mowing my lawn."
  • On SARS, "Stop scaring us with diseases we will never get... Mysterious Asian diseases don't come knocking on your door."
  • On streaming, "I just think people watching other people playing video games is a waste of f**king time."

However, one valid point he did bring up was Bitcoin's electricity use:

Bitcoin uses more electricity per transaction than any other method known to mankind; just one uses more energy than a million Visa transactions and has the same carbon footprint as 85,000 hours of watching YouTube ... Bitcoin uses more energy than Netflix, Apple, Facebook, Microsoft and Google combined.

It's a legitimate concern, but also one that was answered last week by the Harvard Business Review:

In December 2019, one report suggested that 73% of Bitcoin’s energy consumption was carbon neutral, largely due to the abundance of hydro power in major mining hubs such as Southwest China and Scandinavia. On the other hand, the CCAF estimated in September 2020 that the figure is closer to 39%. But even if the lower number is correct, that’s still almost twice as much as the U.S. grid, suggesting that looking at energy consumption alone is hardly a reliable method for determining Bitcoin’s carbon emissions.
...
Almost all of the energy used worldwide must be produced relatively close to its end users — but Bitcoin has no such limitation, enabling miners to utilize power sources that are inaccessible for most other applications... Bitcoin miners from North Dakota to Siberia have seized the opportunity to monetize this otherwise-wasted resource, and some companies are even exploring ways to further reduce emissions by combusting the gas in a more controlled manner.
...
Many journalists and academics talk about Bitcoin’s high “per-transaction energy cost,” but this metric is misleading. The vast majority of Bitcoin’s energy consumption happens during the mining process. Once coins have been issued, the energy required to validate transactions is minimal. As such, simply looking at Bitcoin’s total energy draw to date and dividing that by the number of transactions doesn’t make sense — most of that energy was used to mine Bitcoins, not to support transactions.
...
Because Bitcoin's energy footprint has grown so rapidly, people sometimes assume that it will eventually commandeer entire energy grids... [However,] the energy mix of Bitcoin grows less reliant on carbon every year... [and] miners are unlikely to continue expanding their mining operations at the current rates indefinitely.

TL;DR: Bitcoin's electricity use and environmental impact is "a lot less alarming than you might think".

Merkel nails it

For those who missed it, US President Joe Biden jumped on the 'abolish intellectual property (IP) for COVID-19 vaccines' bandwagon last week, "amid pressure from Democrats in his party to remove patent protections". In other words, yet another policy directed at virtue signalling rather than accomplishing anything useful.

The thing is, IP is not the handbrake on vaccine production and distribution. IP holders have been more than willing to licence it to whomever has, or is willing to develop, manufacturing capacity. Moderna (the creator of one of the two mRNA vaccines) even went so far as to suspend enforcement of its IP way back on 8 October 2020:

...while the pandemic continues, Moderna will not enforce our COVID-19 related patents against those making vaccines intended to combat the pandemic. Further, to eliminate any perceived IP barriers to vaccine development during the pandemic period, upon request we are also willing to license our intellectual property for COVID-19 vaccines to others for the post pandemic period.

It's not just Moderna, either. All vaccine makers are licencing their products to whomever wants them. Take AstraZeneca, which for all its problems has built a supply network with 25 manufacturing organisations in 15 countries to produce three billion doses this year.

Thankfully the world still has a handful of politicians that possess some common sense. In response to Biden's support of abolishing IP, German Chancellor Angela Merkel (via a spokesperson) nailed it:

The US plan would create "severe complications" for the production of vaccines. "The limiting factor for the production of vaccines are manufacturing capacities and high quality standards, not the patents. The protection of intellectual property is a source of innovation and this has to remain so in the future.

If you want to make the 'scarring' from the pandemic even worse, then abolishing vaccine IP would be a great way to do that, as it would undermine the incentive to quickly innovate during the next crisis.


Issue 111

China's nationalising data

Delivered on 03 May 2021 by Justin Pyvis. About a 3 min read.

To be honest, I'm surprised it took this long:

China said thirty-three applications, including map navigation software provided by Baidu Inc. and Tencent Holdings Ltd., violated regulations, with most collecting personal information that weren’t relevant to their services.
...
In March, the administration published rules that banned mobile app operators from declining to offer basic services to customers who refuse to provide "unnecessary" personal information, and it vowed to "concretely protect" the rights of internet users. The rules went into effect on Saturday.

China has started to rein in its internet companies on concerns over their growing influence in every aspect of Chinese life, as well as the vast amounts of data they’ve amassed through providing services like online shopping, chatting and ride-hailing.
...
The government is also said to have proposed a state-backed venture with some of the country’s biggest e-commerce and payments platforms that would oversee the lucrative data they collect from hundreds of millions of consumers.

In China, the Communist Party always comes first. The rapid rise of Chinese Big Tech companies appears to have caught President Xi Jinping somewhat by surprise. The establishment – particularly the heavily state-controlled banking and finance sector – was being rocked and it was a completely spontaneous, market-driven disruption. Certainly not ideal for a 'centrally planned' economy, and not a good look for Xi Jinping when Jack Ma, the founder of Alibaba, was being paraded around the world as a capitalist symbol of success in deeply communist China.

Having cut down tall poppies such as Jack Ma (who has made one public appearance since Xi Jinping was personally involved in the government's cancellation of his planned public offering of Ant Group), the focus is now on regaining control of a new commodity: data.

Yes, China's government has a 'great firewall' that at least in theory prevents its citizens from being exposed to ideas from the outside world (in reality that's impossible). It also controls most of the domestic scuttlebutt by monitoring and censoring every conversation in essential social apps such as WeChat.

But where the Communist Party let its guard down was in finance. In the past it was easy: China's banks were all either state owned or private in name only, with the government exercising control over them directly or via the heavy hand of regulation. But with the emergence of Alipay in particular, over half of China's population decided it didn't need to use the traditional banking system as much, and:

"Once people's money moves from conventional bank accounts into their virtual wallets, much of it doesn't return."

Given that Alipay was privately owned and operated as a non-bank payment company, it was effectively unregulated. That meant it could out innovate and compete the state-owned banks, which "did much of their work with state-owned enterprises, ignoring small firms that could use more capital".

So here we are. Last week we saw that China's new digital currency was not a disruptor to be feared but a platform upon which the likes of Alipay will soon have to operate. Now there's a "state-backed venture" being considered, which will further tether China's biggest e-commerce and payments platforms to the Communist Party. Xi Jinping will once again be able to track every transaction and exert full control over to whom credit is allocated, with the stated aim of "concretely protecting" the rights of internet users nothing but a facade.

China's nationalisation of data and the creation of a digital currency are not a risk to dominant fiat currencies such as the US dollar, nor are they a risk to cryptos. If anything, the opposite is true and Xi Jinping should be careful not to suffocate one of the most vibrant sectors in China with his latest power grab.


Issue 110

The end of Clubhouse

Delivered on 27 April 2021 by Justin Pyvis. About a 3 min read.

Clubhouse is probably still the latest hot app to come out of Silicon Valley. As I wrote back in March:

Clubhouse quickly attracted the who's who of celebrity life, including Bill Gates, Elon Musk, Mark Zuckerberg and even Oprah. It's still invite-only – what fad isn't? – but since day one I've had concerns about its potential longevity.

Well here I am again, bringing you another story about why Clubhouse – which only last week secured Series C funding at a valuation of $US4 billion – could well become vapourware. The inspiration of today's issue was an article by Ed Zitron, with which I generally agree. Ed notes that Clubhouse "assumes that listening to people talk is fun, and that talking to people entirely on a mobile app is also fun". He continues:

Twitch succeeded because it created an actual social network that did something truly new, and enabled people to enjoy watching people play games, which is a thing that people do in person and really enjoy doing. What Clubhouse thinks they're doing is recreating the experience of watching a live panel, or a fireside chat, or, indeed, listening to a podcast, when the actual general experience of Clubhouse is more akin to being at a weird party with too many conversations. And frankly, people enjoy watching someone play a game - there are lots of different games and lots of different people to play them. There are only so many voices, only so many subjects, and people only have so much patience to listen to them.

Do read the whole thing. Perhaps Ed and I are wrong – perhaps it's my inherent dislike of the very thing Clubhouse is trying to scale, the dreaded conference, that has skewed my view. After all, plenty of people seem to enjoy conferences (or was it always more about the free travel and accommodation?).

But I just don't see the appeal to what is essentially a permanent, rolling, extremely broad conference with infinite rooms to which I have to listen in real time at normal speed (turning a podcast up to 1.25x speed is a game changer – try it sometime). But outside of the big, heavy hitting Clubhouse events featuring celebrities and other influential persons, who has time to listen to random people dribble on, unfiltered, for hours on end? And if your only real selling point is the occasional appearance by a big name (their time is incredibly scarce/expensive), what does Clubhouse have to offer over any other platform/medium once the venture capital dollars are exhausted and those VIPs stop showing up?

So I suppose I still don't see the appeal of Clubhouse. To be honest, I'm not really sure it has an appeal, once the celebrity novelty wears off. It's just a directory of live podcasts but with no editing, no way to economise on my time and seemingly few monetisation options in the future. It may not be worthless but I suspect the best case for investors is to be bought out by a desperate Microsoft (which recently failed to buy an audio business with an actual product and social network, Discord) and rolled into LinkedIn (also owned by Microsoft).

The Cellebrite hack

The below has been cross-posted from last Friday's Brekky Wrap. If you haven't already signed up for the best daily morning wrap of the latest market, economic, political and tech news from Australia and beyond, you should!

Whoops, wonder what Apple has to say?
Whoops, wonder what Apple has to say? Moxie Marlinspike/Signal

The CEO of encrypted chat app Signal, Moxie Marlinspike, claimed that Cellebrite – "an Israeli digital Intelligence company that provides tools for collection, analysis, and management of digital data" – gives "very little care" to its own software security. In a blog post, Marlinspike exposed several vulnerabilities and some code that might be copyrighted by Apple in Cellebrite's equipment, which he acquired when in "a truly unbelievable coincidence... I saw a small package fall off a truck ahead of me". 🤣

Stepping back: Signal and Cellebrite have an ongoing feud, after Cellebrite falsely claimed in December 2020 that it had "cracked" Signal's chat encryption. The timing of Marlinspike's attack is surely deliberate, given that Cellebrite recently announced that it would soon go public using a special purpose acquisition company (SPAC) at a valuation of around $US2.4 billion.

Legal ramifications: Cellebrite is used by law enforcement, especially in the US, to collect evidence from suspects' electronic devices (apparently with complete trust – no internal security audits seem to have been conducted). If Cellebrite can be compromised as easily as Marlinspike suggests, it could open such evidence up to legal challenges and seriously compromise Cellebrite's credibility.


Issue 109

The crypto carnage

Delivered on 20 April 2021 by Justin Pyvis. About a 4 min read.

There was a bit of a perfect storm for Bitcoin (and almost all cryptocurrencies) over the weekend, culminating in rapid double-digit price declines. Bitcoin fell almost 15% in 24 hours, Ethereum fell 17% and Litecoin was down some 24%. All have since paired some of those losses but as of Monday, the bitcoin price was still ~10% lower.

Bitcoin's crash.

The major carnage stemmed from a rumour that the US Treasury will soon "tackle financial institutions for money laundering carried out through digital assets". That's a big deal – one reason for Bitcoin's meteoric rise over the past year has been its growing legitimacy, with Mastercard bringing crypto to its network, crypto exchange Coinbase going public last week, growing talk of bitcoin exchange traded funds (ETFs), and then who could forget Elon Musk's $US1.5 billion bitcoin purchase.

Any moves by the US Treasury to undermine that new-found legitimacy will have a detrimental effect, at least in the short-term, on crypto prices. Let's not forget that Turkey last week banned crypto payments entirely, albeit for different reasons (crypto is a threat to the Turkish lira – Turkey's President, Erdoğan, believes that high interest rates cause inflation and has sacked multiple central bank chiefs who have tried to rein in the country's rampant inflation).

Finally there are the ongoing attacks on crypto – Bitcoin in particular – claiming it's nothing more than an "open Ponzi scheme", most recently by Nassim Taleb (of Black Swan/Antifragile fame). We summarised the argument in yesterday's Brekky Wrap (if you haven't signed up, you should – it's free), which boils down to a misunderstanding of what Bitcoin is: a store of value, not a currency.

Taleb's criticism of Bitcoin.
Taleb's criticism of Bitcoin.

Bitcoin "doesn't do anything economic" in the same way that the vast majority of the world's gold "doesn't do anything economic". It's a hedge against people like Erdoğan and is a lot easier to store than physical gold. If you're looking for a blockchain that does a lot of "economic" things (smart contracts, non-fungible tokens, etc), that's most likely going to be Ethereum.

But just because Bitcoin doesn't 'do' anything economic (although there are 'layer 2' developments that might eventually change that) doesn't mean it's "economically equivalent to an open Ponzi scheme".

Regulating the competition away

The big Australian banks – CBA, NAB, Westpac and ANZ – are growing increasingly frustrated at financial innovation that threatens to undermine their monopoly rents:

Last week’s Coinbase float put bitcoin on the radar of the parliamentary committee overseeing the major banks for the first time and National Australia Bank CEO Ross McEwan responded on Friday by urging policy makers to set out a plan to protect investors who trade digital assets.

"We need to think, as a country, [about] what is crypto, how could it be used, and do you trade it – if it is something that should be traded – safely, and so we make it safe. Those are the issues we need to think about – and quite quickly – because people are making a livelihood out of trading cryptocurrencies."

Australian banks are protected but also handcuffed by the implicit backing of the federal government and central bank and many layers of costly regulation. That limits their ability to take risks and innovate, which is why they're only now talking crypto some 12 years after it was invented. They also recently woke up to a more imminent threat – the popularity of buy now, pay later:

"I acknowledge the work and innovation [Afterpay has] undertaken to build such large and successful company – and avoiding all of that regulation is quite a feat," Mr Comyn [CBA CEO] said.
...
"I would suggest the line around innovation in this area is skewed to a complete absence and lack of regulation in a number of areas," he said.

"The government does and should have a posture towards new players to facilitate and encourage innovation, but my point is based on size and scale – they are beyond the point where the legislation framework that applies to that sector needs to be comprehensively reviewed."

Buy now, pay later is eroding banks' interest margins and non-interest income from credit cards, which is basically a heavily regulated version of the former. Buy now, pay later is not so much an 'innovation' in the truest sense of the word – it's an innovation in regulatory arbitrage, or an ability to exploit a loophole the banks didn't see until it was too late.

I suspect it won't be long until regulators come down hard on buy now, pay later. The banks have too much influence in Canberra and there's too much at stake in terms of perceived financial stability:

"We see roughly buy now, pay later users having twice as much credit on their credit facilities, and typically on their credit cards [and] they have more credit products," Mr Comyn [CBA CEO] said. "That is what we can see. But a number of buy now, pay later providers don't contribute to 'comprehensive credit reporting'."

The knives are out and if regulators decide to burden the likes of Afterpay with even a fraction of the existing banking regulation it will hurt their profitability (added costs) and growth potential (less differentiation from traditional credit cards). Tread carefully.


Issue 108

The future of money

Delivered on 12 April 2021 by Justin Pyvis. About a 4 min read.

China has a digital currency. Or more precisely, the People's Bank of China (PBOC – its central bank) has a digital currency, the e-CNY:

A thousand years ago, when money meant coins, China invented paper currency. Now the Chinese government is minting cash digitally, in a re-imagination of money that could shake a pillar of American power.
...
Cryptocurrencies such as bitcoin have foreshadowed a potential digital future for money, though they exist outside the traditional global financial system and aren't legal tender like cash issued by governments.

China's version of a digital currency is controlled by its central bank, which will issue the new electronic money. It is expected to give China's government vast new tools to monitor both its economy and its people. By design, the digital yuan will negate one of bitcoin's major draws: anonymity for the user.

There has been a lot of fearmongering surrounding the news (it's not actually news – the PBOC has been working on a digital currency since at least 2014) – including from the influential Niall Ferguson and Peter Thiel.

Writing for Bloomberg, Ferguson worries that:

Not only are the American monetary authorities underestimating the threat posed to dollar dominance by China’s pioneering combination of digital currency and electronic payments. They are also treating the blockchain-based financial innovations that offer the best alternative to China’s e-yuan like gatecrashers at their own exclusive party.

Ferguson rightly points out that blockchain-based financial innovations are a threat to governments and central banks – cryptocurrencies such as Bitcoin and Ethereum are indeed "stateless money". And he's right that one of China's motivations for moving into the digital space is to one day usurp the mighty US dollar.

However, he quickly veers off course, drawing the bold conclusion that China has got it right – that "the future of money... [will arrive] in the form of a widely adopted e-CNY", all while the US twiddles its thumbs:

This new Chinese system not only defends the CCP against the twin threats of crypto and big tech, while ensuring that all Chinese citizens' transactions are under surveillance; it also includes an offensive capability to challenge the U.S. dollar's dominance in cross-border payments.

Ferguson assumes far too much in terms of both the outlook for China's new digital currency and its threat to the US dollar's hegemony. This is a country that, while large, has a GDP per capita only marginally above Kazakhstan – i.e. it's poor. In terms of wealth, China is about where Japan was in 1985 – which, coincidentally, was around when American economists were worried about "a growing role for the yen, at the expense of the [US] dollar".

It was also when influential US government officials, such as Treasury Secretaries Blumenthal, Baker and Bentson, "suggested that the dollar was too high against the yen... [and] attempts to talk the dollar down were accompanied by intense trade negotiations aimed at forcing the Japanese to open or share this or that market".

If that all sounds very familiar, it should. It's exactly what the US government has been attempting for the past 4 years – just replace Japan with China and yen with yuan.

Back then, people were also worried about a 'Yen Bloc' that Japan was forming in the Asia Pacific. Now, it's "the Multiple Central Bank Digital Currency Bridge, a project exploring cross-border payments using distributed ledgers... [a] step towards Beijing's long-term goal of internationalising the yuan at the expense of the US dollar".

As we all now know, the fears about Japan's yen displacing the US dollar in the 80s were unfounded. But maybe you think it'll be different this time around because, as Ferguson notes, China's government now has a big, scary weapon called a digital currency (something the Japanese could only imagine), while the US Federal Reserve has sat back with a "What me, worry?" approach.

Allow me to allay your fears – all one has to do is take a quick look at the information we have available on China's new digital currency – or rather, don't have – to realise it's unlikely to become dominant outside of China.

For one, we don't even know if it's blockchain based. That means we have no idea if it's secured by proof of stake, proof of work, or privately (i.e. authoritatively). However, the odds that China opts for proof of stake (e.g. Cardano, Stellar) or proof of work (e.g. Bitcoin, Ethereum) are low, because to do so the e-CNY would have to have to be somewhat public and decentralised (i.e. censorship-resistant). Those words don't exist in China's vocabulary.

If it's blockchain-based – that's a big IF – it's probably going to be private and centralised. In other words, it'll be exactly the same as the existing CNY, only more efficient from the government's point view (spying on its people). If anything, the PBOC's national digital currency would be a blow to Alipay and WeChat Pay, not the US dollar. To use Ferguson's own words:

In 2020, some 58% of Chinese used mobile payments, up from 33% in 2016, and mobile payments accounted for nearly two-thirds of all personal consumption PBOC payments. Banknotes and credit cards have largely yielded to QR codes on smartphones.

China is already digital. The problem is, from the Chinese government's point of view, is that most of the transaction data are held by large private companies run by 'capitalist' billionaires. Moving transactions to a government-controlled private blockchain solves the problem of the private sector's growing financial influence without completely dismantling what they've built.

Hyman Minsky famously claimed that "everyone can create money; the problem is to get it accepted". China can force its own people to accept its new e-CNY, but it can't force anyone else. To an outsider, the new e-CNY offers no advantages over the existing CNY (more data and control for the Chinese government, yaayyy...). It will not be the "future of money", nor will it come close to challenging the US dollar's global dominance in global financial transactions.


View more