The value of money

Coffee

Real coffee is Italian. Espresso, of course. I learned to drink it when I was 17, during my first visit to Italy.

And it’s not expensive. In a real Italian café, you walk in, ask for a shot of espresso, and pay €1. In a Starbucks, if you are so inclined, you pay between €2 to €5 for different products, some of whom use the name coffee.

And the most expensive cup of coffee is a cup of Kopi Luwak or Civet Coffee[1], which will set you back a handsome €50.

All coffee – sort of. All different prices. Why?

Money

Money is an extraordinary versatile invention that can be used for many functions.

One function is to express value setting in a transaction. Another is return on investment, or rent paid on a loan. It expresses both historic value and future claims. And it also reflects power relationships – money can compel people to do things.

From a behavioral perspective, money is always based on trust[2]. There is no more intrinsic value in gold than in e.g. iron – it’s just that there’s a lot less gold, and it is much more inert (i.e. it does not chemically react as well as iron, making it less useful in daily life). So whether our paper money is backed by gold or not does not actually make any real difference, other than in our heads. Which is where the real value of money is stored anyway.

In our modern economies, money is supposed to represent the value of our production. Counting the monetary value of our production gives us a handy number to indicate our prosperity: the Gross Domestic Product, or GDP.

As the Economist recently pointed out, there are two core problems with our current way of calculating GDP.

The first is that GDP misses out on any value creation that we choose not to express in money. You may think this is limited to grandparents babysitting their grandchildren, but technology is recently having a major impact on this problem, through the recent rise of open source, creative commons and genuine peer-to-peer economic relations[3].

The second is that a lot of value destruction (pollution, accidents, over-extraction, financial risk-taking) that is monetized is defined as value in classical GDP calculations[4], instead of cost.

But that is not all. Not only is the way we define and calculate GDP problematic.

Also the way we understand, use and even define money is under stress.

Classical economists tend to focus on the power aspect of money.

But money is also a unit of information – which is what I want to focus on here.

How is money information?

Money, when expressed as a price, carries information.

It carries information on the kind of coffee we drink, where we drink it, and whether or not the coffee beans were previously digested by an animal.

Money allows us to express how many apples it takes to pay for a car. And not just any kind of apples, or any kind of car.

This is because the price of apples carries a lot more information than just the price of a piece of fruit. That price carries with it a lot of contextual information.

Fresh apples are more expensive than wrinkled ones. Sweet eating apples are more expensive than sour cooking apples. Apples in small quantities are more expensive than apples in large quantities. This price difference reflects important contextual information about apples: they go off. There is a time-related value shift in apples.

The same is true for the price of the car. A car loses about 20-25% of its re-sale value the moment you first drive away with it right after your purchase. The relative value of electric cars is currently changing vis-à-vis classic internal combustion cars. That reflects not only the quality of the car, but also future policies. If governments get serious about tackling climate change, they will want to encourage adoption and use of electric cars, and discourage the use of internal combustion cars – especially the older, less efficient and more polluting ones.

All of that information, and a lot more, is captured in the money value – the price – of a specific car.

The same is true for markets. Markets are a collection of snippets and bits of information by the different actors on the market[5]. Buyer A bids more because he knows something about the transaction that buyer B doesn’t. But by performing the action of offering a higher price, buyer A does convey information. It may not be exactly the content of the information bit that caused him to offer more, but it does convey that he does know something that causes him to offer more. That kind of information will tell buyer B, or a seller, to adapt their position, based on their own privately held information combined with the knowledge that buyer A knows “something” that causes them to offer more.

The price of, say, a pork belly futures contract is based on a lot of information. The price will include aspects of current and forecast economic growth, taking into account interest rates and, again, expectations of their movement, aspects of employment figures and current and expected inflation, the anticipated influence of global warming on the production and price of feed products, the level of trust in the global financial system, and many, many more factors.

Money and prices even include information that we have only recently started to uncover, e.g. the fact that, under conditions of scarcity (say, when we are really poor), our brains become much less efficient at dealing with money, and we enter into a tunnel vision approach that is actually harmful to us[6]. It is no coincidence that rich people obtain lower prices than poor people for the same transaction – the information that they are rich, and even, in a rough measure, how rich they are is captured in the better price they obtain.

In a way, pricing – and money – is the result of multiple hash functions. A price represents a lot of other data, which can be of any size, into a data set that is universally understood and accepted. So much dollars/euros/yen/renminbi represents a vast amount of underlying data and metadata, and their contextual relationship.

We accept these hash functions because we apply them ourselves all the time. It starts with learning, as a child, how many sweets your weekly pocket-money will get you (and the frustration that your favorite snack is more expensive than your friend’s – who gets a euro more every week anyway, it’s not fair[7]), over valuing a real Italian espresso, to endless dinner party discussions on the value of houses in your city.

We intuitively understand money, and its hash functions, but only because we apply them all the time ourselves.

One could say that reverse engineering those hash functions of money is the core activity taking place on Wall Street and other financial markets. We call it analytics and data mining, but it is always based on the information carrying function of money.

And that information carrying function must be as efficient as possible[8].

But, today, that information carrying function of money, and the efficiency of the hash function, are under threat.

The threat comes, unsurprisingly, from information technology. Information technology contains all aspects of hardware, software, AI, data analytics – it is a general purpose technology that affects, and changes, the way we define and use information.

And information technology’s threat to the information carrying function of money comes in two forms.

They are, first, indirect competition, from de-monetization of value, new ways of communicating value and zero marginal cost production, and, second, direct competition, from new forms of money.

Indirect competition

Indirect competition comes from two sources: Open Source and its related phenomena, and the emergence of the zero  marginal cost economy.

Open Source is, from an economic perspective, a vastly under-estimated phenomenon. The reason is very simple: most of it is not monetised, so it is not visible as money or prices.

But consider a startup like seats.io, in which I work on business development. It was set up by two developers in their spare time. They worked together for almost two years to create a really clever product that solves a particularly difficult and specific problem: how to easily, flexibly and interactively integrate reserved seating into the technology of a ticketing purchase process. And it cost them almost nothing in expenditure – cash out – to set it up. Ignoring the effect of sharing the hardware and the deflationary effect of Moore’s law, the key observation is the possibility to use almost all of the key technology for free.

The Internet runs mainly on Open Source software, which is free for everyone to download, install and run. Operating systems, development tools, SDKs, plug-ins, libraries and parts of applications are available everywhere, for free.

10-15 years ago, the development and technical setup of a startup like seats.io would have cost easily 150.000 euros, in connection costs, license costs to software, storage costs etc. Today, it is virtually free.

Yet, the value has not disappeared. It has only been demonetized. That means that those “150.000 euros” are still there, but they are no longer expressed in money. Which means the hash function is no longer available – the tag is missing.

As an IP strategist, people often asked me about the business models around Open Source. How can you make money, when you cannot charge for the software you develop?

The answer is simple: a balance sheet and a profit/loss account have two sides. When the cost savings from pulling in Open Source outweigh the reduced license revenue from pushing it out, and the quality is good enough, then business will use it and contribute to it.

Open Source causes value to shift. From the technical commodity to understanding the client’s needs. From a focus on delivering something that addresses one specific need to delivering something that works in an open environment, and can address many more needs. From the basic technical structure to the much higher-end aspects of supporting the client’s business model and reducing their costs.

But here’s the catch. All those contributions, representing genuine value creation, efficiency and productivity gains, are no longer expressed in money.

Why? Because, thanks to Information Technology, alternative hash functions, other than money, about value in software and IT technology have become available.

Within developers’ communities, concepts such as reputation, trust and recognition by peers have become important hash functions of value and quality of open source components. The reaction by the crowd to work submitted by people is transparent and traceable, and most developers can understand whether someone else has contributed something of value.

No longer is the price an indication of the value of a component or library, but the traceable and transparent background and history of a developer or a team, and the reactions of their peers to their work.

The value of genius and innovation (or indeed just simply solid good work – never quite as simple as it sounds) is recognized as such – but without any monetization attached.

Social media play a similar role: although they live of advertising, they also vastly undermine the efficiency thereof. People trust their friends much more than they trust any advertisement. As marketing tries to grapple with communities, ambassadors and other new concepts, a large part of the monetisation of the relationship between brands and their audience disappears.

In a first time, this pushes up profits of course, but as competition slowly starts to kick in[9], part of the B2C relationship becomes demonetised.

Again, the hash function of money is replaced by the hash functions of personal relationship, trust and recognition. The value is still there, but it is no longer expressed as a number or value in a currency.

This movement is reinforced by other economic developments, such as the peer-to-peer economy, crowd funding, and other value exchange market places where non-monetary value creation and exchanges are becoming a key part of the transactional process. Adding value, but hiding it from monetisation.

So what we see is that Information Technology is creating new hash functions, that allow actors in society (and the markets) to express, transfer and share value, but without monetizing it.

In addition, there is the rise of the zero marginal cost society. Zero marginal cost means that the cost of creating an additional copy of something is zero.

Again, this is mainly a function of information technology: as Moore’s Law continues to apply, we see that the cost of creating an additional aspect of value (be it a digital copy of a song, or a photo on your Facebook account, or indeed most digital data or knowledge, even, arguably the cost of an additional Watson of IBM) is effectively zero.

And the problem with zero cost value, is that it is impossible to provide a money value to such a copy.

The tautological nature of that observation is meant to reveal the real problem: if we can’t monetise something, we have no good instrument to determine its value – there’s no good hash function available.

And that is ever more becoming problematic. Most of our economic system, and indeed most of our policies, are based on the assumption that whatever is valuable, is also scarce enough to carry with it a monetary value. I.e. we assume that when something is free, it is not valuable.

But that is no longer the case. The hash function of money is simply failing in a zero marginal cost society.

We can no longer rely on pricing to understand what is valuable and what is not.

And that problem is growing significantly year on year.

Direct competition

Yes, we’ll be talking about Bitcoin again.

Currently, almost all money is issued under rules of legal monopolies by central banks and the financial institutions authorised by the central banks.

Private banks create money mainly through the creation of debt.

This worked well when the creation of debt was linked to productive investment, but since the financialisation of business that has taken place during the last twenty years, this link has been severely cut. Most debt today is speculative, and looks for rent rather than value as return on investment.

However, technology is coming up with viable alternatives to debt created through centralised fiat money.

The best known example is of course Bitcoin. Bitcoin, which uses the Blockchain as its underlying technology, is a node-network based virtual currency.

Bitcoin has copied[10] a key feature of money: the unicity of its information.

Money is, in practice, almost impossible to counterfeit. Just like coins and notes are very, very hard to counterfeit (so hard it is economically not really viable), so are classical digital bank accounts very hard to hack. Different systems of encryption and identity verification ensure that it is, in practice, very hard to counterfeit money.

Bitcoin has replicated this unicity of information on the cheap: rather than verifying the unicity of a piece of payment information through expensive encryption and security arrangements, the unicity of a piece of information in the blockchain is verified by the network itself. Since all nodes of the network contain and copy sufficient relevant pieces of information required for the validation of a transaction, the network becomes unbreakable unless someone controls more than 50% of the nodes. Something that is, because of the way the blockchain is set up, very unlikely to happen.

For a hash function to work, it is essential that the hash itself does not overlap – in money terms: money should not be easy to counterfeit.

Bitcoin has been able to replicate this characteristic of money. Therefore, to the extent it is sufficiently trusted by a sufficient number of people, Bitcoin has the capability of becoming a parallel currency.

And that’s a big deal, because, until the blockchain, it was deemed so hard to create a parallel currency that the risk was considered negligible.

But if Bitcoin or an alternative private source of the unicity of money’s hash function survives, and becomes a credible alternative to classical money, then the floodgates of darwinistic competition will open.

New currencies will arise. They will focus on the information technology aspects of money (many of which are very poorly understood, if recognised at all).

And were alternatives can arise, the economics change.

Many of money’s inefficiencies are related to its monopolistic position: there is no good alternative (“TINGA”).

But if virtual currencies like Bitcoin start to compete with money, based on superior information qualities, then we will see an even bigger syphoning away from classical monetisation.

Bitcoin itself is a good case in point. The number of Bitcoins is restricted by the original algorithm setting up the blockchain. No more than 21 million Bitcoins can ever be mined. The consequence on the information carrying aspect of Bitcoin, as currency, will be interesting.

The most likely consequence is that smaller and smaller bits of Bitcoin will represent the same value, leading to nominal deflation – which is another word for destroying the information carrying capacity of money.

Conclusion

As stated, the information carrying function of money is under threat. And the first signs of impact of this problem are becoming visible.

It is likely that the changing value of money as an information carrier is affecting the efficiency (or rather, the lack thereof) of monetary policies. Although central banks have been printing money like crazy, it does not seem to have the effect their theories predicted.

Is that caused by the fact that the efficiency (and therefore the value) of money is diminished as a result of technological developments? I think so.

In that context, it is particularly worrying that the monetary authorities do not seem to spend a lot of time on the way technology is affecting their models and theoretical approaches.

Not that the 2008 crisis has left us in any doubt on the reliability or applicability of those theories and models.

But as we move into a time where the information carrying function of money is undermined, both by the indirect action of demonetization of value and the direct action of competing currency hash functions, it is not certain whether the classical functions of money still work as they did before.

Given the relationship between money and power, the fact that most of government activity is based on the assumption of the ability to tax value creation because it is monetized, and investment and pension funds and other classical instruments of capital investment that are based on monetized return on investment are based on the information carrying efficiency of debt-created fiat money, it seems like interesting times lie ahead.

I’m not sure which way this will all go, so I won’t be placing any bets.

The real challenge, of course, is that it is no longer sufficient to know what to bet on, and what the odds would be. It is also no longer certain what to bet with.

Luckily, one thing remains certain. Italian espresso will still be the best value for – well – whatever value expression you prefer.

The real challenge is that central banks that do not have a deep and useful understanding of how technology is fundamentally affecting their intellectual tools, and sometimes even rendering them inadequate, may well be wasting literally trillions of dollars or euros.

 

 

 

Notes:

[1] Yes, that’s where the coffee beans acquire a special aroma from travelling through the digestive system of the civet cat.

[2] Which implies that the very existence of money contradicts the assumption of people acting in their “rational self-interest”. As Daniel Kahneman confirmed in “Thinking, Fast and Slow”, the percentage of the population acting in their rational self-interest is equal to zero. If you have not yet done so, please go and read that wonderful book.

[3] What is often called the “sharing” economy, such as Uber or Airbnb is not really “sharing” economy in this respect, these are classical monopolistic platforms that try to extract rent from other operators, by shielding their price-setting algorithms. By reducing pricing transparency, they actually reduce the value of money.

[4] Check out Rutger Bregman’s inspiring “Utopia for Realists”.

[5] Perfect information transparency is, of course, perfectly impossible. This is because no piece of information can meaningfully relate to all the context that is required to fully understand it. Information transparency is one of those dogmatic figments of classical economists’ imagination. The status of information transparency and completeness of markets is really an expression of a Heisenberg-light principle of uncertainty. The fact that we now generate as much information in two years as we did in the last 10,000 years does not help this problem.

[6]Scarcity” by Sendhil Mullainathan and Eldar Shafir.

[7] The “unfairness” of money is another fundamental characteristic that goes under-researched.

[8] Incidently, this is the reason why insider trading was banned. Not because it is immoral (it was not deemed immoral until it was made illegal, and there is no general moral rule that using private information to benefit from a deal is wrong), but because insider trading at market level hinders the information carrying efficiency of money and markets. Insider trading is a bug, not a feature.

[9] As a side note, the increases of profitability of business indicate less efficient markets. This is correlated to slower innovation and decreases in R&D investment, but also to stronger Intellectual Property rights and more monopolistic and rent-seeking behavior.

[10] All real innovation starts with copying. “Copycats” by Oded Shankar.



Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>