The digital economy

A few weeks ago I had the pleasure of being on the panel for Blockchain and the Digital Economy at ADC’s Leadership Forum. The session outline led with three questions:

How has 2020 accelerated the acceptance of the digital economy?
Is blockchain fulfilling its promise as the new universal disruptor?
How real is the role of cryptocurrencies as the new universal store of value?

It’s a large topic and an important one. If we’re to address challenges such as growing inequality then we need to find a way to make the economy work for all of us, rather than just some of us.

Unfortunately, as too often happens, adding ‘blockchain’ as a subtopic results in blockchain dominating the discussion while other (more) interesting ideas are ignored. Blockchain is quickly positioned as the solution and all other ways of framing (and understanding) the problems we face are ignored. This panel was no different in this regard.

Some of the ideas that would have been relevant in a broader discussion are things that I’ve been exploring for a while. Before the panel I’d pulled together an outline of the argument as to why our future is not “the digital economy”, but something much more interesting, and which creates more opportunity and freedom to act in addressing the challenges we’re facing. Rather than let a good outline go to waste I thought I’d build it out a little and publish it here.

Before we start I should probably mention that I approach the digital economy, Bitcoin and blockchain as a technologist who’s formative years building all sorts of massively distributed systems (typically systems that leveraged AI in some way). More recently I’ve collaborated on research projects with sociologists, economists, and experts from various practical domains, where we’ve picked apart trends to understand the underlying factors that are influencing society and the economy. A good entry point to this work is Cryptocurrencies and the trust economy.¹

There is no digital economy, just the economy

There’s an excellent essay that I always recommend to people interested in how technology and society are shaping (and reshaping) each other: Technology and History: “Kranzberg’s Laws”.² The Kranzberg in the title, Melvin Kransberg (1917-1995), was a historian and founder of the Society for the History of Technology, and the laws named after him are the distillation of a lifetime’s work exploring the relationship between technology and society. The essay uses examples from the Cold War, but it’s startling how effective they are at explaining the current circumstances we find ourselves in.

Kransberg’s laws can be seen as an argument against technological determinism,³ the idea that “technology is the prime factor in shaping our life-styles, values, institutions, and other elements of our society”. Technological determinism is rooted in the idea that technology has become autonomous and outrun human control—our machines controlling us rather than us controlling the machines. This is not a new idea (though we often think it is) and it’s easy to find examples in history of a writer railing against the evils of technology and our inability to manage it.

Our relationship with technology is not so simple though, and while the development of new technologies is an important factor shaping society, it is not the only factor, nor even the most important factor. As Kranzberg’s Fourth Law states:

Although technology might be a prime element in many public issues, nontechnical factors take precedence in technology-policy decisions.
Kranzberg’s Fourth Law

This is why VHS won over BetaMax, why our power networks are the shape they are,⁴ the reason why the US civilian nuclear power industry uses a reactor design that works well for a submarine but which is terrible for industrial power generation,⁵ and so on.

This idea—that social factors are more important, which means that we can influence how technology sits in society—is something I’ve been prosecuting for a while: We have choices, we just need to take them up.

Most discussion about the future of the economy has a strong technological determinism bent: we have an “analog economy” which will be supplanted by a “digital economy” when superior digital institutions disrupt the old analogue ones. Bitcoin will make sovereign currencies irrelevant. (More on that later) Or Musiccoin, to grab a random example, will disrupt music distribution by disintermediating existing music distribution institutions. Moving land titles to the blockchain would enable individuals in troubled countries to secure their claim on their family home and the fields that support them. Developing a blockchain-based trade finance solution will eliminate the problems and inefficiencies of correspondent banking arrangements.⁶ And so on. Some firms (and individuals and communities) are thought to be “living in the future”, having already adapted to this economic transformation, taking their salary in Bitcoin and using the platforms and institutions of the digital economy to place themselves beyond the reach of the institutions of the old economy.

This is not the case though.

Take cryptocurrencies. Individuals will generally use the currency they find most convenient, something stable which their counter parties also use. The less convenient a currency is the poorer it’s exchange rate, as we saw in the age of Free and Wildcat Banking. You might think that you’re using the best thought out stable coin pegged to the local currency with a slick consumer app, but if your counter party doesn’t have the app (and isn’t really interested in it) then they will demand a price premium, a surcharge, for accepting your stable coin. The equivalence between that plastic dollar⁷ in your hand (which they’ll accept right now) and the dollar equivalent in the app (which they’re not really interested in) is no longer 1:1. Exchange rates—they’re a thing.

It turns out that the killer app for a currency is tax, a government standing over you with their monopoly on violence and demanding that a large portion of your income, your biggest liability, is handled over in the currency of their choosing. The case of the Indian Pound, British Raj, and cotton growing communities is illustrative here: the colonial government required land owners to pay land taxes in Indian Pounds (or have their land confiscated), where the only source of these pounds was the cotton buyers backed by the colonial government. This drove subsistence farming communities to transition to cotton farming, from food to cash crops, bringing them into the Raj’s monetary system in the process.

There are some examples of governments (typically local governments) accepting tax payments in a crypto currency. The key point though is that the tax debts are still denominated in the local sovereign currency, not the crypto, so you, the tax payer, are forced to accept the currency risk should you choose to run your finances in a cryptocurrency.

We might build these new digital institutions, but social factors will determine if they will be used.

The truth likely lies part way between technological determinism and not. This is related to the ever popular trolly problem. The interesting aspect of the trolly problem isn’t the choice you make, but the fact that your presence at the switch forces you to make a choice. The only distinction worth making is whether or not the choice is actively or passively made. We should also note that choosing not to use an imperfect technology (or attempting to ignore it) might be worse than choosing to use it, as an imperfect technology in an imperfect world might be preferable to the imperfect world without the technology.

Kransberg’s Fourth Law can be used to more effectively frame what’s happening here. Technology provides us with new capabilities, new affordances, but it’s society that determines how we might use these capabilities. Or, put a different way: new technology provides new options, creates new choices, enabling us to choose differently, but the same social forces as in the past still determine the shape of the choices we make.

This is similar to the idea that “the medium is the message”, attributed to Marshall McLuhan—that a communication medium, and not the messages it carries, should be the primary focus of study. If we consider a message to signify the content and character of a communication, then a message’s character is determined by the media (technology) used, while the content is itself another (earlier) medium.⁸ Just as the content of the text medium is speech, (and text is the content of radio in terms of the script, and radio is the content of TV in terms of the serial, and so on), the content of our digital economy is our analog economy: the two are not separable. Our new economy will be our old economy repackaged to take advantage of the the affordances provided by the new medium. We can even draw a line back through history to see that:⁹

the globalised economy is the result of digital information management and the global multi-modal container network, a medium, unbundling value-chains from the previous colonial economy;
the colonial economy was the result of steam power and the telegraph, a medium, unbundling production and consumption in the pre-globalised world; and
a pre-globalised, pre-colonial world where production and consumption were in the same place.

This insight enables us to rewrite Kransberg’s Fourth Law:

Although technology might be a prime element in shaping the “digital economy”, the social norms and preferences of the “analog economy” is will have precedence in technology-policy decisions.

Our economy will evolve, not be replaced.

The challenge then is not, as we often assume, to build new ‘digital’ institutions to replace our existing imperfect ‘analogue’ ones—the technological disruption narrative. But to understand how the norms and practices between institutions will evolve as we take advantage of the new medium.¹⁰ Or, putting this another way, rather than trying to find technical solutions to social problems, we’ll find (new) social solutions to overcome technical limitations.¹¹

We’ll wait until the end of this essay to explore what this means for the future of the economy, the question of “How has 2020 accelerated the acceptance of the digital economy”. Before that we need to address the questions of “[i]s blockchain fulfilling its promise as the new universal disruptor?” and “[h]ow real is the role of cryptocurrencies as the new universal store of value?”

Bitcoin has reached maturity

It should be abundantly clear that Bitcoin—and cryptocurrencies in general—are here to stay. If I was to try and locate Bitcoin on an analyst’s s-curve / hype cycle, then I would estimate that we’re coasting onto the plateau of productivity.[Now, in 2023, we appear to have reached this plateau, which Bitcoin and crypto seeming to have reached a stable state.] In the last few weeks we’ve seen both Mastercard¹² and VISA¹³ announce that they’re supporting Bitcoin (though we need to wait before we understand what this means in practice), and Bitcoin ETFs (Exchange Traded Funds) are clearly things.¹⁴ Institutional investors, hedge funds, and so on are buying Bitcoin, as are firms such as Tesla which has changed the title of the CFO to be “Master of Coin”.¹⁵ Bitcoin has clearly arrived as a speculative asset.

While Bitcoin is maturing as a speculative asset, the story is different when we consider it as a currency.

There are a few obvious use cases for Bitcoin. The first is repatriation payments, where immigrant workers send a portion of their wage to family still in the home country as support. Conventional repatriation services were expensive and slow. When Bitcoin emerged as a viable alternative prices dropped and payments were transmitted faster—a good thing. Markets for grey market goods (goods that are legal to buy, but which you might not want your partner seeing on the credit card bill) have also adopted Bitcoin. Grey markets have always had higher chargeback rates from the credit card processors driving up merchant costs—moving to merchant reviews coupled with the payment finality of Bitcoin pushed costs down. Bitcoin has found a role for itself as a complimentary currency: a currency that while not necessarily a national currency, but which supplements or complements national currencies. Think “global, stateless, gift card” and you’re not too far off. Once we look beyond those two use cases though, the argument for Bitcoin as a “universal store of value” is much weaker.

There was talk that Bitcoin would replace sovereign currencies, a point of view that started with the gold bugs but gained wider acceptance, as they see governments’ ability to print money as problematic. Gold bugs like Bitcoin’s deflationary monetary policy, where the total number of Bitcoins that will ever be issued is capped (and then the total number of coins will slowly drop as they get lost in locked wallets and the like, driving the value of a single coin up), as it rewards those who hoard the currency (and the government can’t print money and so devalue your savings). This has led to something of an obsession with the exchange rate between Bitcoin and US dollar, enthusiastically watching it grow to the moon as if this proves some point. It has also led to the tendency to focus on Bitcoin’s total market cap, the aggregate value of all Bitcoins issued based on the current market price. While both of these practices make some sense if we consider Bitcoin speculative asset, they make no sense if we want to consider it a currency.

Currencies are said to have three uses: as measures of value, stores of value, and means of exchange. Bitcoin is not really good for any of these three uses as it’s exchange rate is too unstable—it’s easy to forget that governments put a lot of effort into maintaining price stability, and hyperinflation in the places like the Wiemar Republic is what happens when these efforts either fail or are abandoned. Similarly, it’s unlikely that a government would denominate its finances in Bitcoin as this would have the government forgo an essential tool in managing the economy, which is no easy job at the best of times. Greece’s troubles and the Euro is a cautionary tale here.¹⁶ And given that tax is the killer app for a currency, we’re unlikely to see wide spread adoption of Bitcoin as a currency. It was currency instability and the problem of fluctuating exchange rates that drove the governments to shutdown Free Banking and force their economies to adopt a single sovereign currency.

There is still some talk of digital sovereign currencies: governments creating and issuing their own cryptocurrencies. The rational is that we need to move to digital money if we’re to have a digital economy. This is a bit confused though, as we’ve had digital currencies for at least a few decades. Any savings or debt you have with a bank is denominated in that bank’s digital currency. If I pay you via my banking app, and we use different banks, then our banks talk and do a currency exchange which is reconciled via their shared relationship with the government (denominated in a third, wholesale government issued digital currency). Government regulation forces the exchange to be at 1:1. (The same is not true for complimentary currencies, which can be devalued at the whim of their owner.) Today, in Australia, these transactions are effectively instant, thanks to the new payments platform, so there’s not much point in developing a digital Aussie dollar.

The actual effect of creating a digital sovereign currency would be to bring governments back into consumer banking, as all of the currency’s users would need to have a government managed account. It’s not clear that this is a good thing.

When governments were in the business of running consumer banks they weren’t that great at it, and their banks had a higher cost-to-serve than non-government banks.¹⁷ This is not a reason to not create a digital sovereign currency, but it is a problem that must be acknowledged. The other reason would be if a government wants to capture as much in-market payment data as possible to both deal with crime (money laundering, tax evasion and the like) and see (and shape) citizen behaviour.¹⁸ There’s precedent for this, such as the direct payroll integration many governments are rolling out: they know what you’ve been paid before you do. Again, this is not a reason not to create a a digital sovereign currency, but it is a problem that must be acknowledged.

Finally, we should acknowledge that as Bitcoin matures a number of problems will crystallise. This is the “Bitcoin is full of fraud” vs “Most fraud has nothing to do with Bitcoin” debate. Both sides are correct. We can see from the public transaction data that all the bad habits of an unregulated market are present in Bitcoin: front running, wash trades, money laundering, and so on. However, the volume of value and transactions passing through Bitcoin is small when compared to other digital platforms. A money launderer, for example, is much more likely to target an online game that enables them to trade digital goods, or to crowdfund themselves,¹⁹ as these public platforms have transaction volumes required to enable money launderers to hide their activities. Any internet accessible service that enables users to get value into, and out of, it will be the target of organised crime. Bitcoin is small in comparison. As Bitcoin matures and sees broader adoption, volume will increase and regulators will need to step in, but then that’s already happening.²⁰

So, in summary:

Bitcoin as a speculative asset: mature, or close to.
Bitcoin as a complimentary currency: mature.
Bitcoin as a “universal store of value”: no, not a thing.

Blockchain has failed in the market

This is probably the most controversial point of view in this essay, but it’s also likely to be the most misunderstood. It’s should be clear from the previous section that I don’t mean “Bitcoin has failed in the market”, though some folk can’t seem to seperate the two in their heads. Nor am I imply that “blockchain doesn’t work”, as it clearly does, as we can see with Bitcoin. It’s just that blockchain is not to be the universal disruptor folk thought (or perhaps, hoped) it was.

The challenge here is define ‘blockchain’. Indeed, most essays that mention blockchain start by trying to define it and it’s various subtypes: private, permissioned, and so on. Depending on how narrow (or broad) this definition is, they might also define Distributed Ledger Technology (DLT). I’m going to take a slightly different approach.

I started my career in distributed systems, so I have two buckets in my head full of useful techniques for building distributed systems. There’s the synchronised state machines, where you run a programme in multiple places, feed them the same inputs, and then share their outputs to tweak (and so align) their operation. This group includes PAXOS, RAFT through to tricks like log shipping.²¹ The key is that the same programme runs on each node. The second group is where the programmes are all different, and our focus is on the interactions. We develop distributed patterns of messaging that enable a heterogeneous (rather than homogenous) group of software programmes to collaborate. Distributed AI is the obvious example here.

When Bitcoin emerged back in 2009, what was most striking about the initial essay was that it didn’t mention any new technology. The newest technology in the stack was proof of work, which emerged as an email anti-spam approach, and that was from the late nineties, ten years earlier. Other than that it was a synchronous state machine approach based on log shipping, with the next state for the system determined (effectively) by open lottery (i.e an unbounded rather than bounded set of participants).

Blockchain was a solution to a particular technical problem—that of creating a stateless currency (i.e. a currency with no issuer)—a solution that leveraged known technology. What was fascinating was how the author solved the problem of who issues the currency by making issuance a side effect of processing transactions, and randomising which server received the new currency. The challenge of creating a “digital native” currency was always issuance, and this nicely steps around the problem as it’s not possible to determine, prior to issuance, which server will receive the new currency.

So the bucket in my head labeled ‘blockchain’ has a very lose definition of blockchain: any solution that runs the same software on every server, processes records to create a transaction log, uses the log to synchronise the servers, and relies on an open lottery to determine which transaction log to use going forward. This fuzzy definition has enough room in it to enable us to use it as the foundation of a general computing platform (think of the transaction log as a paper tape²² and it’s obvious), as we can see with Ethereum. Interestingly, it’s also specific enough that we can predict its performance capabilities, and they’re terrible.

It was easy, early on, to predict that blockchains will struggle to support high, or even modest, transaction volumes. The assumptions this combination of technologies implies about inter-node communication and consistency guarantees have well understood implications. There are some obvious things you can do to improve performance, but there are known limits to these approaches. I wrote about this quite a while ago. For example, we could tweak the size of the blocks and the time between when they’re generated (optimise parameters) to make communication between nodes more efficient and realise a modest gain. The team behind Ethereum did this when it was developed. We can off-load processing from the servers, make them do less by combining transactions, running something like a bar tab that is reconciled every so often. This is what layer-two applications do, which includes everything from Lightening Networks²³ through exchanges, and now VISA and Mastercard. We can also federate, breaking our one big blockchain into many smaller islands that talk to each other. (This is commonly called ‘sharding’, a term adapted from the web-application world.) Federating would enable us to process a lot more transactions, though each account is effectively restricted to only transacting with accounts on the same island, as communication between islands quickly becomes much more expensive than communication within an island.²⁴ Rather than one large blockchain, we have thousands of little ones that talk to each other. A lot of work has been done over the last 10 years to make blockchain scale, and we’ve very little to see for it.

It was clear that if we wanted to build solutions for many of the use cases that were being thrown up for blockchain, that we were going to have to move away from blockchain as “the platform under Bitcoin”. This is the motivation behind Distributed Ledger Technology (DLT). If we keep the idea of a shared transaction log—a chain of blocks—but ditch the rest of blockchain, then we can build much more performant solutions, though solutions which don’t have all the strong security, ‘trust’, and consistency guarantees that blockchain has. For example, I was at the demonstration for one ‘blockchain’ which was for, all intents and purposes, a traditional single-server database application that happened to emit a transaction log as part of its operation. It wasn’t surprising that it could support high transaction volumes as it didn’t provide any of the guarantees (or incur the associated overheads) that blockchain does. (More about DLT, and the opportunities and challenges associated with it, later.)

Blockchain failed in the market, not because it didn’t work, but because it a solution to a particular problem, rather than being the general-purpose technology as was commonly assumed. We have been unable to find any use cases where blockchain brings something new to the table, other than hosting a cryptocurrency such as Bitcoin, despite over ten years of investing in startups and proof of concepts in the search for the killer app.

It’s useful to think of the proposed use cases for blockchain as being divided into two groups.

The first group contains use cases that we already had solutions for, but which we forgot about.

Provenance might be the poster child here, with a number of provenance solutions were put into production in the 2000s and earlier. Meat and Live Stock Australia (MLA), for example, deployed a national beef provenance solution around then, to support the development of export markets by tracking product through the supply chain (so that problems could be identified and fixed, and quality improved) as well as proving quality to external markets. Now BeefChain is also a thing.

Provenance solutions are logistically challenging to deploy as its difficult to capture a chain of evidence, a chain of signed digital documents, that connects producer, processing and logistics steps to the product in your hand. If we want to certify that an organic steak is in fact organic, to certify its provenance, then we need the local organic farming body to attest that the framer is certified and audited as organic, and the farmer to attest that they bred and grew the animal as organic, and the butcher to attest that the stake was taken from the cow sold to them by the farmer, and so on. Even then we can’t be sure of the steak’s provenance, as people are not always reliable—provenance is all or nothing, either the data is complete and correct, covering every step of production and distribution, or it tells us very little,

Provenance solutions succeed when we can cover off all these logistical challenges: enduring that all touch points are certified as required, distributing digital keys for signing, auditing participants, and so on. They fail when we cannot. A peak industry body, such as MLA for example, can cajole members into participating in a provenance solution for the good of all. On the other hand, a grocery chain can only force their suppliers onto the solution, but not their suppliers’ suppliers, making the solution worthless.

The problem with applying blockchain to this first group of problems, the problems that we forgot we already had solutions for, is that it brings no new technology to the table, no new capabilities. The solution will succeed or failed due to other factors—as we saw with the peak industry body vs grocery chain. The question of using a blockchain or not is irrelevant. Provenance is the example we used here, but there are many others include managing property titles, and identity.

The second group are use cases that we don’t have a solution to.

Trade finance is a good example of this second group. The global trade finance network is something that evolved rather than being designed, and it could be more efficient. If we look in our distributed technology bucket there’s all manner of interesting tools that we could use to make it more efficient. The reason that we haven’t already solved this problem is because there is no formally defined “trade finance network”. Trade finance is actually a huge number of informal industry norms and practices, regulation in different regions, and firms and institutions large and small. Improving trade finance means establishing new norms and practices, ones native to distributed digital technology, across the entire ecosystem. However, we have the same situation as with the grocery store and provenance—there is no single controlling body that can cajole or force participants in the trade finance network to adopt these new digital norms and practices.

What we have is a collective action problem, not a technology one. The problem was previously unsolvable due to structural and social challenges that we have no solution to. Adding blockchain into the mix—or, indeed, any other technology—doesn’t change this. Blockchain isn’t the solution to problems like trade finance, settlement, modern slavery, and so on, as the challenges these problems present are one of joint action, not data synchronisation.

Technological determinism lends us to the false belief that (simple) technical solutions can address (complex) human problems. Bitcoin, for example, doesn’t require us to trust a currency issuer to not issue more currency that they committed to, ergo blockchain is a technology for trust, ergo blockchain can be used to address the problem to ensuring property rights in an environment where citizens don’t trust the government. Life is not so simple. Your property rights only exist as someone—in the West, the state with its monopoly on violence—is willing to enforce them. If we want to ensure property rights in uncertain environments, then we need to address the uncertainty, not deploy a new database.²⁵

This leaves us with the question of: why did Blockchain emerge when it did? The technology it uses was at least 10 years old when it emerged. Why didn’t Blockchain appear 10 years earlier?

Blockchain emerged in the late 2000s because that was when bandwidth dropped to the point that the solution became economically viable. A one megabyte block every 10 minutes just wasn’t practical on consumer internet connections before about 2008-2009. Bitcoin emerged when it did because there was not point releasing it earlier. For all we know Satoshi Nakamoto was sitting on the Bitcoin from 2000, quietly improving the solution, until they thought the time was right.

Blockchain didn’t failed in the market because it doesn’t work. We might see new blockchain solutions in the future, solutions with the impact of Bitcoin, though it’s unlikely. Bitcoin is an incredible act of creativity which will be very hard to replicate.

Blockchain failed because it is a very particular solution, a very creative solution, to a particular problem. More over, it’s also only part of any solution. The catch cry that “Blockchain is more secure”, for example, ignores the fact that security is an attribute of the system and not the individual technologies. The reason that Bitcoin’s ledger has never been hacked has more to do with the many other soft targets around the ledger than any inherent invulnerability in the ledger itself. If we look at fraud and theft in the banking sector, it would be challenging to find a crime due to someone hacking a bank’s general ledger, or account databases even.

Blockchain failed because it was a solution, and not a new technical capability. The use cases proposed for Blockchain—other than Bitcoin—were typically already solved (making Blockchain an expensive while worse performing alternative), or represented unsolved joint-action challenges that technology can’t help us with.

Ultimately Blockchain failed as it was positioned as the universal disruptor, and it clearly isn’t.

The future economy

At this point we can get back to the question of the future of our digital economy. We’ve argued that there’s is no “digital economy”, just “the economy”, and that while Bitcoin has an established role in this economy as a speculative asset (and some role as a complimentary currency), Blockchain is largely irrelevant (other than as the thing under Bitcoin).

Of the three questions put to the panel, the last two are beside the point as the global pandemic will have little influence on their fate. We can, however, tweak the first question to provide something like:

How has 2020 accelerated the ~~acceptance~~ penetration of ~~the~~ digital technology into the heart of the economy?

To tackle this question it’s useful to think of the economy as having four layers:²⁶

Market: how consumers find and obtain the products and services they need
Institutional: the assumptions, practices, norms, and regulation that shape how organisations work together in and across industries to service market demand
Organisation: norms, practices, processes, and so on inside a firm that determine what work is done
Group: the norms, preferences of individuals and teams that shape how work is done

The first and last—market and group—are the easiest to deal with so we’ll tackle them first.

Evidence is starting to emerge that consumer adoption of the various forms on online shopping (online ordering with delivery and click-n-collect) has accelerated by 3-5 years. We’re buying more digitally than ever before. This is particularly true for the elderly. It’s not clear, though, how big this acceleration will be in the mid-term as there is a “two steps forward, one step back” factor, and we don’t know how big a step that backward one will be.

The uncertainty during the year did drive up savings as people hoarded cash, and many of their leisure activities were shut down. However, it doesn’t appear that the pandemic has done much for the adoption of cryptocurrencies: either as a speculative asset or as a complimentary currency. Bitcoin was already maturing by the time the pandemic started, so it’s difficult to tease apart effects. If there was increased adoption due to the pandemic then it’s unlikely to be significant outside trend mid- or long-term trends.

Similarly, the need for many people to work from home has had a dramatic affect on how we work, and we’re working more digitally than before. City centres are not crowded and many people would like to spend more time working from home where they can walk to the pie shop at lunch time, despite regional governments trying to cajole us back into the office. We can also tie this to the previous point by observing that the need to work from home has made much of our spending more local. The pattern of flow of value in the community has changed, though it’s not clear how much and what the longer term implications are.

We should also note that we haven’t seen a shift from working in the office to working from home, but a shift from working in the office to working anywhere. If we refer back to the medium is the message, then we can see that the future of work is not a digital workplace: a location on a digital platform such as SecondLife. The new (digital) workplace is the network of digitally mediated relationships between workers in a team, and not a ‘place’ per se (either physical or digital). It used to be worker goes to work, now work goes to worker. Making work digital has made the work fungible. It also means that techniques such digital twins enable us to bring more work, such as diving fork lifts, into the digital workplace. The digital workplace is rapidly evolving, consuming our old concept of physical workplace in the process. It’s unclear how much the global pandemic has accelerated this trend, but it’s looking like the acceleration of group has been more significant than with market.

As work now goes to worker (rather than worker to work) some workers are even heading out to the country, intending to trek into the city every week or so while spending the rest of their time in the great outdoors. Many commentators have pointed out that this enables workers in the city to have a tree or sea change, but it also enables telemigration and globotics.²⁷ Telemigration has been a building as a trend for some time, the pandemic is likely to have accelerated this and consequently made work more global, but it’s not clear by how much. We might consider this the impact on the organisation layer, with COVID-19 accelerating a trend to unbundle the physical locations where firms have employers. We’re doing the same work but we’re arranging it differently. This might be the next unbundling (building on Baldwin’s model, mentioned earlier) in the history of globalisation. It’s unclear where this trend will go, but it is both an opportunity and a problem. Telemigration could be used to offshore work, contributing to the growing problems in society. Or it could be used to enable more flexible work, and so making work more equitable by helping workers juggle their home and professional commitments. We have a choice, it’s up to us to take it up and decide wisely.

The last level—industry—is the most interesting. This is also the area where that idea that “technology will disrupt institutions” plays out. Our future economy is not an economy where the existing institutions have been replaced by better and more ‘digital’ alternatives. Trade finance will not be disrupted by a blockchain trade finance solution that replaces existing institutions with new digital ones. Assuming this makes the mistake of considering the financial mechanics of an industry as seperate from its productive work. Digital technology is changing all aspects at once, finance, production, work, and engagement with the community and customers. The end result being that we’re likely to see the institutions endure, but what they do, the norms and practices around them, will change.

Consider the DFMA construction process, which is something I write about far too much.²⁸ Putting a digital model of building at the heart of the business enables construction firms to take a radically different approach to construction, one that is better, faster, cheaper, safer and greener. The firms at the front of this transition are some of the most digital things you can find. They blend manufacturing and construction, and the distinction between building and building product is blurred.

The DFMA approach has been proven and firms from both manufacturing and construction are getting involved. The challenge is that we understand DFMA construction as organisations but not as an industry, and many of the old institutional arrangements no longer work and we need to develop new ones. For example, construction is based on debt financing, and the risk models that underpin this financing rely on quantity surveying, models which do not work in the DFMA world. A bank wants to pay for the land and then provide funds in tranches, sending out a quantity surveyor at the end of each tranche to check that the money was well spent. A DFMA builder, though, wants some of the money up front to start modelling and manufacturing, and then they’re only on site for a short period of time (as the build is so much faster). Quantity surveying doesn’t work, the risk model prevents lending, and finance is consequently unavailable. The DFMA process needs a new approach to debt finance, one that works for the new work practices begin developed. The firms at the front of the charge have been working with other industry members to develop new approaches to quantity surveying—inspecting work in factories etc, and not just at the site—to overcome this problem. The point here being that if we approach “construction in the digital economy” as a question of replacing existing institutional arrangements with digital versions (on a blockchain!) then we have at best an opportunity missed, at worst what we did would be an actual hinderance to progress as we’ve baked old practices in hard to change digital solutions. Instead we need to understand what the firms want to achieve, how they want to collaborate, and then consider what options the new generation of technology provides.

DFMA construction is currently having its River Rouge moment. While firms has been addressing point challenges, such as quantity surveying, they’ve been forced to vertically integrate as there is no solution for a DFMA firm to connect digitally with the economy. Should, for example, a firm want to buy doors, what they would like to do is send a model describing the door they want along with a description of the contract, and receive a door, made to tolerance, with supporting building product documentation, for quality assurance. This is not possible. Their response is to buy a turret lathe and do it themselves, to vertically integrate. The opportunity is to understand how organisations in the DFMA industry want to work together (rather than how they’re currently forced to work together) and then enable them to do this. A successful approach would see the DFMA firms unbundling, driving the creation of a new digital native industry. One might say that the new digital economy would be unfolding in front of us.

This is a general pattern that would be valuable to explore.

Take land titles. A typical pattern when a developing country manages to obtain a significant external internet connection is for experts from the West to push for the country to “leapfrog years of problems and put all the titles on blockchain”, as the blockchain is more secure (it isn’t, as discussed), more efficient (it isn’t, a database would be cheaper), and so on. A prototype, based on a possible workflow, is built with nice user interfaces (as it’s easy to make a prototype work at the scale of a proof of concept). Done!

This doesn’t solve the problem though. The real challenge with capturing property rights is not the title management process (which is simple forms computing) but the GIS challenge of mapping the country, creating land parcels, and documenting them (that is, the challenge of first creating and then issuing titles). There’s no point of owning or trading a land title if you don’t know where it is and its extent. Usually the pilot will grind to a halt here, as it’s all too much to fund the mapping and data capture. If not, and the project moves forward, we can actually create a worse problem. The GIS / property title solution is likely based on a Western concept of property rights, were there’s one owner for each title and boundary changes are rare. This is likely not how property rights work in traditional communities. Indeed, it’s not how property rights worked in the West prior to enclosure. The rights are likely communal, multilateral, and renegotiated constantly as circumstances change. Unless we can develop and implement a data model that accurately represents these rights then all we’re doing is forcing a Western concept of property onto a community and erasing their heritage. Even worse, in a diverse country such as Papua New Guinea, the details about the nature of property rights and how they are managed might vary significantly community by community. Creating a property rights model that encompasses all this diversity might be impossible, and imposing a single model (no matter how well research) will have us picking winners and losers.

The change in this institutional layer is fascinating, presenting both huge opportunities and problems. Progress in understanding and exploring it, however, was not influenced by the global pandemic.

Conclusions

So, in conclusion, the reason the framing of the panel was an opportunity missed is because it:

put the cart before the horse by putting the technology in front of the economy, and
was consequently too narrow, focusing on particular aspects of what technology could do rather than the possible choices we have and what we want the economy to be.

Rather than narrowing our options to focus on what we think technology is driving us toward, we need to open them up and explore where we can go with the help of technology.

This is where DLT is interesting, as it gives us a wealth of options that were not economically viable before. It’s not that the technology has raced ahead, but that we now have a network and digital environment where distributed computing techniques that have been maturing for decades have flipped from technically possible to economically viable.

We’re being presented with new opportunities, new choices are opening up. It’s up to us, however, to recognise this and take advantage of them.

Endnotes

1
Evans-Greenwood, P et al. 2016, Cryptocurrencies and the trust economy, Deloitte, Australia, <https://www2.deloitte.com/au/en/pages/technology/articles/the-future-exchanging-value.html>.
2
Kranzberg, M 1986, ‘Technology and History: “Kranzberg’s Laws”’, Technology and Culture, vol. 27, no. 3, pp. 544–560, <http://www.jstor.org/stable/3105385>.
3
The concept of technological determinism is typically attributed to Thorstein Veblen (1857–1929), an American sociologist and economist, though it was Carl Marx who embedded it into the popular consciousness.
4
Hughes, TP 1993, Networks of power: electrification in western society, 1880 – 1930, Softshell Books ed, John Hopkins Univ. Press, Baltimore, Md.
5
Dotson, T & Bouchey, M, ‘Democracy and the Nuclear Stalemate’, The New Atlantis, no. 62, Fall 2020, pp. 14–45, <https://www.thenewatlantis.com/publications/democracy-and-the-nuclear-stalemate>.
6
Saadati, M, ‘Has trade finance pulled the plug on correspondent banking?’, Trade Finance Global <https://www.tradefinanceglobal.com/posts/has-trade-finance-pulled-the-plug-on-correspondent-banking/>
7
This being written in Australia, where the money is plastic rather than paper.
8
This, of course, implies that the current medium will be a message in a future medium, which this is the source of “the medium is the message”.
9
Adapted from Baldwin’s “three cascading constraints” view of globalisation, and simplifying a little. See Baldwin, RE 2016, The great convergence: information technology and the new globalization, The Belknap Press of Harvard University Press, Cambridge, Massachusetts.
10
With new institutions emerging in any gaps created.
11
Kranzberg’s Second Law—”Invention is the mother of necessity”—tells us that new technology is developed to support these new social arrangements, rather than the other way around.
12
Nelson, D 2021, “Mastercard Will Let Merchants Accept Payments in Crypto This Year”, Coindesk <https://www.coindesk.com/mastercard-accepts-crypto-payments>.
13
Jones, T 2021, “Bitcoin Is Probably Coming to Visa”, Gizmodo <https://www.gizmodo.com.au/2021/03/bitcoin-is-probably-coming-to-visa/>.
14
Mitchell, E 2021, “Bitcoin ETFs: What They Are and How to Invest (in 2021)”, Bitcoin Market Journal <https://www.bitcoinmarketjournal.com/bitcoin-etf/>.
15
Hull, D & Melin, A 2021, “Musk Adds ‘Technoking of Tesla’ Title; CFO Is Coin ‘Master’”, Bloomberg <https://www.bloomberg.com/news/articles/2021-03-15/musk-adds-technoking-of-tesla-title-cfo-is-master-of-coin-kmahgu33>.
16
Greece’s inability to manage its exchange rate only compounded the country’s problems, preventing it from inflating its way out of debt.
17
We’ll go with ‘non-government’ as ‘public bank’ and ‘private bank’ have particular meanings that are not what we want to convey.
18
This is likely the driver for China’s efforts to create a digital sovereign currency. Most trade is denominated in US dollars and most trades are cleared via US banks, even though any bank handling USD could clear the trade, as it’s assumed that (as has happened in the past) the US government will step in if the banks are in trouble. This means that the US has access to a huge volume of trade transaction data, data that countries like China would prefer to have themselves.
19
Thomas, G 2012, “Thieves Launder Money by Crowdfunding Themselves”, Vice <https://www.vice.com/en/article/xyyyqq/thieves-launder-money-by-crowdfunding-themselves–2>.
20
For example: “First Bitcoin ‘Mixer’ Penalized by FinCEN for Violating Anti-Money Laundering Laws”, Financial Crimes Enforcement Network <https://www.fincen.gov/news/news-releases/first-bitcoin-mixer-penalized-fincen-violating-anti-money-laundering-laws>
21
Which we can consider a continuation-based approach.
22
i.e. as the paper tape in a Turing machine
23
Hussey, M, Copeland, T & Phillips, D 2020, “What is Lightning Network? And How to Use It in 2020”, Decrypt <https://decrypt.co/resources/bitcoin-lightning-network>.
24
As a side note, it’s possible that exchange rates would emerge between these islands due to the inconvenience bridging them represents, as with the payments example above.
25
This tendency to assume that technology can solve non-technology problems is not unique to blockchain. The same dynamics as with blockchain and property titles can be seen with AI and healthcare. See Aggarwal, A 2021, “Beware hype over AI-based healthcare in lower-income countries”, Financial Times <https://www.ft.com/content/f4dd834c-4835-4ee0-8737-ff98626fa010>.
26
This four-layer model is drawn from research into creativity. See Ford, CM 1996, ‘A Theory of Individual Creative Action in Multiple Social Domains’, Academy of Management Review, vol. 21, no. 4, pp. 1112–1142, <https://journals.aom.org/doi/10.5465/AMR.1996.9704071865>.
27
Baldwin, RE 2019, The globotics upheaval: globalization, robotics, and the future of work, Oxford University Press.
28
Evans-Greenwood, P, Hillard, R & Williams, P 2019, “Digitalizing the construction industry”, Deloitte Insights https://www2.deloitte.com/us/en/insights/topics/digital-transformation/digitizing-the-construction-industry.html>.