Could the next driver of blockchain adoption be central bank backed digital cash?
“I don’t play with formulas. I play with pictures.”
— Benoit Mandelbrot
Most people look for certainty before they make a decision.
Fortunes are made investing in this.
Before it becomes this.
Hesitating, because they don’t trust their inferences, and trapped, cycling through feelings of excitement and disappointment, as they look for certainty through the opinions of others.
Today, it’s easier than ever to do your own research, yet when it comes to investing, in assets like cryptocurrencies, why is it that so few people do?
It’s what happens when you fire up google and try and find out something you don’t know. This behaviour drives the advertising business, and almost everything in between, right down to cheating in your local pub quiz.
The problems arise when the answers aren’t straightforward.
If you want to know who was the 26th president of the United States, or who was the next monarch after Henry VIII? Or who won what, or when something happened? You’ll be rewarded with the answer, microseconds after typing the question.
But what about when the question isn’t that simple?
What happens when the answer is multidimensional?
If you’ve spent time trying to find out what cryptocurrencies are, what they do, and how they can benefit you, then, unless you are interested in tech, or in how computers work, the chances are, you’ve experienced it.
Information gaps are a glitch in your wiring. You’re hard-wired to find out the answer. George Loewenstein, a behavioural economist at Carnegie Mellon University, wanted to know what makes people interested in something, and, in 1994, he ran experiments to find out.
Loewenstein discovered something he wasn’t expecting. When your interest in a topic is piqued, you are driven to find out the answer, and when the answer is not simple or easy to find it causes you to experience pain.
The pain of not knowing is what made Google — Google, and why Netflix shares have gone up over 5,000% since 2013.
Why Netflix? Have you ever binged watched a show? If so, you’ll know the answer.
Do you remember when you first became aware of cryptocurrencies, and what was it that got your attention?
If you don’t have a technical background, it was probably a story where the protagonist, an ordinary Joe, or Jane, did something unusual, contrasting against what the majority would do, and whatever they did, changes their lives. It’s either a tragedy, where the hapless person gains something of value and then loses it or a comedy, where their lives are changed forever.
It’s a story as old as time.
Like the programmer who, in 2010, swapped 10,000 Bitcoin for two pizzas, or the guy who threw out his computer containing the private keys to his Bitcoin investments, and only realising his mistake months later, guessing the laptop was probably buried in a local landfill site, he was trying to get permission, (and the money) to dig it out, because the value lost was in the millions.
When did you first hear of Facebook? What year did you start using it?
Some people like to be avant-garde and start using something new before it’s anywhere near mainstream, and by the time the product is being used on buses, trains, and seen everywhere, they are long gone and have moved on to the next big thing.
Is it possible to place human behaviour behind product adoption and usage into a framework?
In the Smart Dust series of articles, podcasts, and videos, we discussed Gartner, a leading consultancy company, who are in the business of putting a probability on the emergence of new technology.
The Gartner hype cycle or hype curve is a useful visual representation of the stages a new idea, product, or service moves through, from initial introduction to mass market use, allowing you to factor in the background conditions and estimate the product’s most likely location on the hype curve.
The Gartner curve moves through five stages.
Peak of inflated expectations.
Trough of disillusionment.
Slope of enlightenment.
Plateau of productivity.
In December 2018, Bitcoin and the leading alt-coins made a low, after losing 80-95% of their value. If you look at the December 2018 lows, ask yourself what is the likelihood of this low being the third point of the Gartner hype cycle, the trough of disillusionment?
In modern financial markets, most of the volume is driven by quantitative algorithms. The efficient-market hypothesis is a financial theory that states that all asset prices fully price in all known information, and EMH implies that attempting to time the market is impossible.
Burton Malkiel, a professor of Economics at Princeton University popularised the theory that financial markets follow a random walk, again implying that trying to predict the next move of any financial market was as useful as reading your horoscope.
A random walk is a mathematical stochastic that describes a succession of random steps. There are many types of random walk stochastic processes, but the one used most used in finance is the Markov chain. The problem with Markov chain stochastics is they are memoryless.
The visible result of outliers in financial market performance is “fat tails.”
Fat tail events, like the financial crashes of 1929, 1931, 1937, 1974-75, 1987, 2000-2002, 2007-2009, and the collapse of Long Term Capital Management in 1998, would be next to impossible if financial markets followed the laws as expected.
And yet they happened.
A recent academic study by ex-Oxford Professor of Physics, Neil Johnson suggests, that instead of being random and memoryless, financial markets follow a branch of science called complexity theory.
Johnson defines complex systems as, “the study of phenomena which emerge from a collection of interacting objects.”
If Johnson’s hypothesis is correct, the implication is financial markets have a memory.
How do you react if you lose? If you are speculating in a market and you’ve just lost the previous three trades, will you feel no psychological effects from the losses, or will you hesitate, or trade less aggressively? If the answer is yes, your decision in the present has just been influenced by the past. In short, your actions were influenced by your memory of past events.
In path dependent systems, as new information becomes available, you readjust your beliefs, depending on the updated results. The math behind the updating process is called Bayes theorem.
The Voight-Kampff Test
In the movie Blade Runner, the world consists of humans and replicants. Replicants, human-like androids, were designed to replace humans in off-world colonisation. Replicants were designed to do different tasks, and are optimised to perform the role of soldiers, miners, and even pleasure models. The problem is, the replicant design is so perfect that they’ve become self-aware. As a fail-safe, the designers built in a short four-year lifespan, and because of this, a group of replicants have come back to earth to find their creator, and remove the four-year life fail-safe.
Blade Runners are bounty hunters who track down replicants and ‘retire’ them. People accused of being replicants must pass the Voight-Kampff test; a series of questions designed to monitor response times and emotional tells. Over a series of questions and responses Blade Runner’s keep a running score; a probability that the person being questioned is a replicant.
Of course, the Voight-Kampff test is fictional, but, by replacing humans and replicants with new information entering financial markets, the Voight-Kampff test becomes a way to demonstrate how path dependency and likelihood work together.
The test uses Bayesian analysis.
Before we get to Bayesian analysis, let’s look at the method of statistical inference you might be familiar with.
To test your own idea (alternative hypothesis), you test a different hypothesis called the ‘null’ - the commonly excepted idea. The null hypothesis can never be accepted. The best you can do is reject it.
This kind of statistical analysis is called frequency analysis, or frequentist inference. It entails using a sample of a larger population of data and attempting to reject the “null,” the commonly excepted idea.
But, without large amounts of data to work with, how can you make decisions when the data you have is limited? How can you make probability assumptions when you only have a handful of results, not a data set of thousands or millions of records?
This is where Bayesian analysis, also known as inferential statistics can be used.
Traders talk of and use probability, but it’s more useful to use likelihood. Probability is bounded to between 0 and 100. Likelihood is not. Likelihood is proportional to a probability and does not have to add up to one. Likelihoods don’t mean anything on their own. Their power comes when you compare them.
Let’s use a coin toss to show the usefulness of using likelihoods.
Let’s say you want to find out if the coin you are using is fair or biased. With a fair coin, the probability of getting heads or tails on each flip is equal at 50%.
After flipping a coin ten times, the results are six heads and four tails.
Using Pascal’s triangle, you can look up the probability of getting this result: The probability of six heads and four tails, with a fair coin, is 21%.
After getting a 21% probability using a fair coin, let’s say that new information comes in and you adjust the bias of the coin in response, so instead of 50%, you adjust the assumed bias of the coin to 75%.
Using Pascal’s triangle, with the bias of the coin adjusted to 75% heads, the probability of six heads and four tails is 15% using a 75% bias.
The next step is to compare the probabilities to give a likelihood ratio.
21% / 15% = 1.4
A 1.4 likelihood ratio means six heads and four tails is 40% more likely under a fair coin than under a 75% biased coin.
So, starting with a prior probability of 50%, and adjusting the bias up or down you can build likelihood models of your analysis.
Have cryptocurrencies bottomed? Has the bear market ended? Instead of reading articles that are written by writers mirroring the price action, bullish when the market moves up, shifting to bearish if the market moves down, what if you could put a number on the likelihood of your hypothesis?
If you believe December 2018 is the low, what does the prior evidence suggest? What is the likelihood of this assumption being true?
Start with an assumption. In this case, the belief the cryptocurrency market hit stage three of the Gartner curve, the trough of disillusionment, in December 2018.
Then, use an initial prior probability of 50% and adjust from there as new information becomes available.
If your hypothesis, cryptocurrencies have bottomed, ending stage three of the hype curve, is true, what is the likelihood of the new information, given your belief is true?
Or, from the opposite perspective, if the assumption is false, and the cryptocurrency markets have not yet hit bottom, what is the likelihood of this new event?
By listing the events in the cryptocurrency markets since the low, ask yourself what is the likelihood of these events, given your assumptions, and by keeping a running score of the likelihood, you’ll be able to take note of mispricing in the market. For example, prices move higher, while your likelihood moves lower.
When was the last time you withdrew cash you expected to need for the week ahead from a cash machine? More and more of us have stopped using cash. Our habits have changed. Yes, there are still some old schoolers out there with gold money clips, but they are in the minority.
Next time you’re in a queue, waiting to pay for your lunch, take notice of how other people pay.
You might think that cryptocurrencies are now Main Street, heard of by everyone, even if they don’t quite know what they are.
Ask around, next time you’re socialising. You might be surprised by the answer.
Let’s say using Bayesian analysis you have estimated, with a likelihood of 70%, that the cryptocurrency markets hit stage three of the Gartner curve in December 2018.
If your assumption is true, the next stage after stage three is stage four, the slope of enlightenment.
Ask yourself, what is going to be the driver? What’s going to power cryptocurrency from the minority product it is now into the mainstream, used by everyone across all age groups?
Technically minded readers may scoff at cryptocurrencies being a minority product, but compared to visa transactions in the retail domain crypto is still in diapers.
Boosting transactions speeds, hard forks, and bolt-ons, like the lightning network with atomic swap capability, are all well and good, but try and explain this to someone in their late fifties who first posted on Facebook in 2017, and you’ll see the problem.
Most of us use smartphones. But do you know how they work? Have you ever asked yourself how your smartphone deals with picking up Bluetooth, WIFI, and 4G?
Each technology uses a different frequency range, so why doesn't your smartphone need several separate antennas, one for each frequency range?
The answer is the fractal. Only by using fractal designs with self-similar properties, can internal wide-band antennas be designed to pick up all the frequencies needed by the technology stuffed into your cell phone.
99.999999% of the population don’t give this any thought and just use their phone.
For a product to be mass-adopted, and used by everyone, across all age ranges, it needs to be easy to use.
A contactless debit card is easy to use. Beep, transaction approved.
Central Banks in the Age of Blockchains
This week, the Financial Times reported that the IMF, (International Monetary Fund) and the World Bank, are launching a “Learning Coin” on a private blockchain to gain a better understanding of how blockchains work.
In a statement, the IMF said:
“The development of crypto-assets and distributed ledger technology is evolving rapidly, as is the amount of information (both neutral and vested) surrounding it. This is forcing central banks, regulators and financial institutions to recognise a growing knowledge gap between the legislators, policymakers, economists and the technology. This project begins to bridge that gap and form a strong knowledge base of the technology among IMF and World Bank staff.”
On 3rd April 2019, the World Economic Forum published a white paper titled: “Central Banks and Distributed Ledger Technology: How are Central Banks Exploring Blockchains Today?”
The white paper lists 10 projects Central Banks are actively pursuing.
Retail central bank digital currency.
Wholesale central bank digital currency.
Interbank securities settlement.
Payment system resiliency and contingency.
Bond issuance and lifecycle management.
Know-your-customer and anti-money-laundering – Digital KYC/AML.
Information exchange and data sharing.
Cash money supply chain.
Customer SEPA Creditor Identifier (SCI) provisioning.
Of the ten projects listed above, the top three, retail and wholesale digital currency, and interbank payments suggest a possible to replacement for cash.
If that sounds a bit of a stretch, then here’s the World Economic Forum’s continuation of the first item in the list — the retail central bank digital currency:
“Central bank-issued digital currency that is operated and settled in a peer-to-peer and decentralised manner (no intermediary), widely available for consumer use. Serves as a complement or substitute for physical cash and alternative to traditional bank deposits.”
Instead of sophisticated bolt-on technologies, like the lightning network and atomic swaps, what is the likelihood that the driver of cryptocurrencies up the slope of enlightenment and into the mainstream will be something everyone can understand?
What if your bank account was upgraded? The only difference being, you wouldn’t have access to physical cash.
Our habits have already changed, we’ve begun to stop using cash, replacing the use of a money clip for a beep.
One way to figure out if this is more likely than not is to research the possible motivations for Central Banks to do this. By asking what’s in it for them, you might uncover some insight that will build you an edge.
One possible motivation could be to change the nature of cash money. With digital cash it would be possible to impose negative interest rates on your account, forcing you to pay for the “privilege” of having access to the services your digital account offers.
The United States currently has over $22 trillion of debt. And that, in case you’re wondering, is a lot. One important metric is a country’s Debt to GDP ratio — a country’s debt compared to its output. The current Debt to GDP ratio of the United States is around 108%. Any figure above 100% raises an eyebrow, because as this ratio goes up, so does a country’s ability to issue new debt.
Because the higher the number, the less likely a country is to be able to pay back its debt. And the less likely a country is to be able to pay back its debt the higher the risk, and the higher the risk, the higher the return demanded by investors. As investors demand more, the more expensive it is for a country to roll over and service its debt.
Previously, in times of crisis, capital works were started to put a nation back to work and stimulate the economy. In recent times, interest rates have been used to control the economy, and in the past, when a country was tied to the gold standard, gold was confiscated and then revalued higher effectively devaluing the underlying currency.
One way to stimulate the economy would be to control how cash works. If Central Banks are successful and ban the use of physical cash, forcing everyone to use digital money, another “feature” of digital cash could be time erosion.
One way to “encourage” the public to spend and stimulate the economy is to have a digital currency that erodes in value if it is not spent.
Think this can’t happen? One feature of financial options is time decay. Options (when buying) give an investor the ability to leverage their investment with the security of fixed and known risk; however, this advantage comes with a considerable disadvantage. Time decay.
Each day, as the option ticks towards expiry, a portion of its value is removed. This is called theta decay.
If all money was digital, Central Banks could add a time decay component, forcing you to spend it and not save. This would help stimulate the economy, and help reduce a nation's debt.
If digital cash with a time decay component was one hypothesis, then as new information becomes available, you could figure out the likelihood of your assumption being true, given the data.
If cryptocurrencies did hit the bottom of the trough of disillusionment in December 2018, ask yourself what are the most likely drivers of cryptocurrencies, up the slope of enlightenment and into the mainstream?
Could it be complicated bolt-on tech understandable only to those with tech backgrounds, or will it be something more simple, something you’re already used to? Something like digital cash?
If financial (and cryptocurrency) markets are complex systems and are path dependent, then as new information arrives, ask what is the likelihood of your assumption being true, given the latest information. Like a chain reaction, is new information reducing the odds in favour of your hypothesis being correct, or is it random, like a drunkards walk, with some data backing and other data refuting your assumptions?
The majority though, don’t do the work, preferring instead to listen to the opinions of others, forever trapped in the cycle of reacting to the news, entering late, out of position, and with no concept of risk.