Cryptocurrencies Revisited: Taking A Step Back

Piotr Kukielka
Jan 11, 2021 · 15 min read

Cryptocurrencies, in their current form, have more than ten years of history already. Have they proved themselves? You will hear different answers depending on who you ask. Some people will tell you that cryptocurrencies successfully demonstrated that a distributed financial system independent from governments and not backed up by any commodity could work. And they will be right. Others will say that BitCoin and other cryptocurrencies let them down because they never lived up to the promise of being a real currency. And they will be right too.

While I applaud the idea of cryptocurrencies, I think that they still miss the mark. They often use smart technological solutions to enhance blockchain technology, but that might not be enough to create real currency. Blockchain forces specific design, which implicitly causes all systems based on it to have similar properties.

I would like to look at these properties, discuss if they are pushing us in the right direction and what that right direction is. If those properties are not what we are looking for, how can we change them? What changes in the core design would that require?

I will try to answer those questions.

Don’t want to trust? Don’t buy the crypto.

Let’s start by revisiting the history of money. You can find a full overview of it in many different sources, like Wikipedia, or Nick Szabo’s excellent article. I will focus on factors that allow something to function as money and cause other people to accept it. I summarised how I see it on the diagram:

As you can see, I believe that all non-commodity forms of money are based on trust. It’s only a matter of who is trusted: individuals, private or governmental institutions, or technical solutions. The tendency is to gradually shift the trust from interpersonal relations in the direction of centralized, global entities, like a nation or government.

Cryptocurrencies are not different in that regard. This topic’s excellent study can be found in the paper “Cryptocurrency and the Myth of the Trustless Transaction” by Rebecca M. Bratspies. She notices that cryptocurrencies significantly expand the number of parties that have to be trusted: protocol developers, miners, market owners, wallets developers, and the ever imperfect technology. They are independent of the governments and maintained by the collective of people. But the individual user does not have too much to say. We are moving from systems where control over the currency was direct (debt, commodity) to the systems where those in power always have the last word. In the case of fiat money controlled by governments, currency can be — for example — debased at any time. In the case of cryptocurrencies, the influential minorities (which have the most aggregated currency ownership or computational power) can decide to rollback any transaction or, in fact, implement any other change.

And it’s not just a hypothetical situation. It has already happened multiple times. In 2017 Bitcoin split into Bitcoin and Bitcoin Cash over a technical disagreement in the community. In 2016 rollback of fraudulent transactions in the Ethereum resulted in the fork. Forbes summarised it simply:

The fact is that Ethereum has compromised its principles in order to rescue a client. Or, in the language of another world, the Ethereum central bank has directly recapitalized the DAO commercial bank by monetizing its debts.

So here we are again, replicating the errors of old systems we tried to fix. We find ourselves in a situation where a minority of the community holds the most power and can decide about the rest. It is possibly even worse with cryptocurrencies because that power can manifest itself only in cases of significant turning points. Massive frauds can sometimes be reverted, mainly if they affect substantial stakeholders. But no one cares or really can do anything about smaller scams or issues. If you get robbed, you are screwed. There is no one to appeal to. After all, cryptocurrencies are cryptographically secure, so if you got cheated, it was your fault, wasn’t it? Enter the rule of the technocracy.

What doesn’t kill you makes you stronger.
But not in the case of cryptocurrencies.

But let’s imagine the majority always decide. It would not prevent frauds, but could it improve other aspects of the system? Realistically many users would probably not understand or care about all complexities of cryptocurrencies. Such democratic cryptocurrency would be prone to either harmful changes caused by ignorance of crowd or stagnation if it stays away from the changes completely. In the case of mature, performant cryptocurrency with all bugs wiped out, would stagnation be such a bad thing, though? Possibly not, as long as everything works as expected and the world remains unchanged.

But cryptocurrencies are based on a shared belief and social consensus. If an improbable event disturbs the equilibrium — if a black swan appears — impact on such a system can be significant. And in a system based on trust, once it has been damaged, it is hard to regain.

To protect from such events, the system should be designed with an additional focus on antifragility. Something is antifragile when it can become better or stronger when suffering from negative environmental inputs. Great examples could be immune systems, muscles, or evolution. The immune system learns how to fight new threats after encountering them, muscles grow stronger after being micro damaged when exercising, and evolution through survival of the fittest makes sure new generations of organisms constantly adjust to the changing environment.

Those systems are not indestructible. On the contrary — they consist of many fragile parts. But damage or destruction of any individual part gives valuable feedback to the whole system, allowing it to improve. They have to be fragile on the micro level to be antifragile on the macro level.

Most of the cryptocurrencies seem to have the opposite design. They might look antifragile at first glance — they are guaranteed to be worth the same amount for everyone, and no one can take away tokens from the users. But that strength is at the micro level. No single cryptocurrency owner can fail independently, but cryptocurrencies are susceptible to all kinds of environmental, social, and software disasters. When those happen, loss of trust cannot be contained at the local level. Whatever happens to cryptocurrencies happens globally. Responses to such events must also be global — and those tend to be either too slow or dictated by an ignorant crowd, or influential minorities.

Despite that, the creator of the “black swan” and “antifragility” terms, Nassim Nicholas Taleb, is a cryptocurrencies fan. As you can read in his article:

In the complex domain, expertise doesn’t concentrate: under organic reality, things work in a distributed way, as Hayek has convincingly demonstrated. But Hayek used the notion of distributed knowledge. Well, it looks like we do not even need that thing called knowledge for things to work well. Nor do we need individual rationality. All we need is structure.

It doesn’t mean all participants have a democratic sharing of decisions. One motivated participant can disproportionately move the needle (what I have studied as the asymmetry of the minority rule). But every participant has the option to be that player.

Somehow, under scale transformation, emerges a miraculous effect: rational markets do not require any individual trader to be rational. In fact they work well under zero-intelligence –a zero intelligence crowd, under the right design, works better than a Soviet-style management composed to maximally intelligent humans.

Which is why Bitcoin is an excellent idea. It fulfills the needs of the complex system, not because it is a cryptocurrency, but precisely because it has no owner, no authority that can decide on its fate. It is owned by the crowd, its users. And it has now a track record of several years, enough for it to be an animal in its own right.

For other cryptocurrencies to compete, they need to have such a Hayekian property.

But the free market isn’t so efficient because some people are “motivated enough to disproportionately move the needle” for everyone else. The point is that everyone knows one’s local market conditions better than any external decision-maker. And everyone can individually act based on that knowledge for his own self-interest. Information about those decisions and valuations is passed to the market in the form of prices, which can be used by other market players to aid their own choices. The system is so efficient because it propagates local knowledge without any centralized effort. As Hayek wrote in “The Fatal Conceit”:

Adaptation to the unknown is the key in all evolution, and the totality of events to which the modern market order constantly adapts itself is indeed unknown to anybody. The information that individuals or organisations can use to adapt to the unknown is necessarily partial, and is conveyed by signals (e.g., prices) through long chains of individuals, each person passing on in modified form a combination of streams of abstract market signals. Nonetheless, the whole structure of activities tends to adapt, through these partial and fragmentary signals, to conditions foreseen by and known to no individual, even if this adaptation is never perfect. That is why this structure survives, and why those who use it also survive and prosper.

So it’s not only about who decides but also in what way. Collective voting wouldn’t bring the same result as bottom-up self-organization. It cannot because no single entity has a full picture.

For a self-organizing system to emerge, you need a framework that gives all participants some degree of freedom. They need to be able to experiment and make their own decisions.

Later in the same book Hayek says this explicitly:

One revealing mark of how poorly the ordering principle of the market is understood is the common notion that `cooperation is better than competition’. Cooperation, like solidarity, presupposes a large measure of agreement on ends as well as on methods employed in their pursuit. It makes sense in a small group whose members share particular habits, knowledge and beliefs about possibilities. It makes hardly any sense when the problem is to adapt to unknown circumstances; yet it is this adaptation to the unknown on which the coordination of efforts in the extended order rests. Competition is a procedure of discovery, a procedure involved in all evolution, that led man unwittingly to respond to novel situations; and through further competition, not through agreement, we gradually increase our efficiency. To operate beneficially, competition requires that those involved observe rules rather than resort to physical force. Rules alone can unite an extended order.

And let’s be honest: cryptocurrencies are not designed with self-organization in mind. Because every user shares the same protocol and ledger, the degree of freedom is minimal. Rules are strict, and changing them requires changing the code. This, in turn, needs common consensus and participation of the majority in the upgrade process.

Where is the Hayekian property in that?

Not a fan of wealth redistribution?
I have bad news…

Another severe problem worth consideration is the issue of money supply and money creation. Traditionally fiat currencies were always originating from the commodity money, which was giving them a history of value. When the baseline was established, the commodity’s link could have been cut, transforming commodity or representative money into fiat money. Since that point, its amount just had to be scaled to reflect market size changes (otherwise, inflation or deflation would occur).

In the case of cryptocurrencies, it’s harder because they start from zero, both in terms of the money value and volume. Let’s look at this problem a bit closer.

Most cryptocurrencies use one of the three modes for the issuance of their tokens:

  1. The fixed pool created at the start by the owner(s)
  2. Mining new tokens by spending some computing power (cryptocurrencies based on Proof of Work algorithm)
  3. Paying dividends based on the already held pool of tokens (cryptocurrencies based on Proof of Stake algorithm)

In each of those models, owners and early adopters benefit from it much more than later users. Of course, that incentive helps with the adoption, but it’s hard to justify in the long run. It is designed that way only because most cryptocurrencies are close to useless in the initial growth phase. So in the absence of present benefits, they need to promise big wins in the future.

But in the face of the wide success of any specific cryptocurrency, that mechanism means a massive reallocation of wealth that would need to happen. We create money out of thin air, and we need to make enough of it to cover all transactions in the economy. Critics of fiat money owned by the government often rightfully point out that governments cannot help themselves but have to inflate the currency for the sake of profit. And inflation is wrong because it causes a redistribution of goods from everyone to the currency issuer. But how are the cryptocurrencies different? Their introduction also requires redistribution of goods — and a lot of it. And those goods are also redistributed to the small group of beneficiaries.

Another serious monetary problem of cryptocurrencies is that the tokens’ amount is growing independently from the market demand. As a result, cryptocurrencies very rarely hold their value comparing to any chosen basket of goods or other currencies. They must be either inflationary or — almost always — deflationary.

Which is better? There is no general agreement in this area. One of the moot points is the question if money amounts should be adjusted to the market growth. Without that, prices would drop naturally with an increase in productivity and efficiency. Most, if not all, central banks existing today follow inflationary policy, while many Austrian Economics, for example, believe that mild deflation is acceptable. Moderation is the key, though — you will be hard-pressed to find any respectable economists who say that either high inflation or deflation is good for the economy. Probably either school would advise to keep it no higher than the annual market growth rate.

That is simply impossible with the cryptocurrencies today. Most of them have some mechanism for growing the number of tokens, but they are often focusing on keeping that growth constant. They have no synchronization mechanism to adjust the number of tokens depending on the market growth or — which is more critical in the development phase — demand for the given cryptocurrency. It would be tough to retrofit such a mechanism into existing systems as it requires centralized input. But markets that have information about the crypto supply and demand are private enterprises. Depending on their data would be against the idea of decentralization and independence. Getting independent, not centralized details on the whole economy seems to be even more challenging.

Also, the stabilization of the price of the cryptocurrencies is not in the interest of many stakeholders. After all, the growing value of the tokens attracts many people hoping for easy profit, which in turn boosts price even more.

To fix that problem, we would have to guarantee the tokens’ prices from the start. But in a distributed system without central entities that could be trusted, there is no obvious way to achieve that purely on a technological basis.

In search of a solution

Let’s stop there for a moment and check what problems we’ve discussed so far:

  • Dilution of trust, which creates an image of a trustless system by hiding real issues
  • Loss of control of single participants in favor of influential minorities with most capital or computational power
  • Fragility caused by design, allowing various social and technical problems to affect cryptocurrencies on the global scale
  • The inability to keep both inflation and deflation on the low level
  • Redistribution of wealth into the hands of early adopters or cryptocurrency creators

Could those issues be fixed? Or maybe properties that money needs to have to be money are mutually exclusive with being genuinely owned by people, antifragile, and created ethically? I think they are not.

Let’s look at the economy in its simplest form — when it’s based on interpersonal debt (first one from the top of the diagram).

Despite the severe limitations, it has several interesting qualities:

  • It’s up to the involved parties to define what ‘owes’ really means and what form repayment should take.
  • Debtors’ behavior can give lenders an insight into the local market’s economic condition (regularity of payments, size, frequency of requested loans, etc.).
  • If the debtor cannot pay his debts, both parties can renegotiate the repayment conditions to find a satisfying solution.
  • The default of a single debtor does not impact the system as a whole.
  • If the debtor is obliged to pay back (give back) the same commodity he lent, there is no exchange rate risk involved.
  • There is no problem with currency supply and demand.

That system promotes local negotiations and adjustments which naturally propagate themselves without any central planning or coordination. Financial problems of some debtors make lenders more careful, effectively decreasing credit supply. That forces other borrowers (like entrepreneurs trying to grow their business) to limit the loans (investments) size. So in worsening market conditions, investors receive early warnings about the situation by observing credit supply and adjusting their investments accordingly.

Theoretically, in a free market economy based on fiat money, the interest rates could play a similar role. In practice, however, it’s often the other way around — interest rates are used to force the market into a given direction rather than inform participants what the current market condition is.

The bottom line is that the system is self-adjusting based on the local information effectively propagated to all the users. That propagation is an inherent property of the system, not something that any particular user needs to willingly do.

Such a system is also free from the risk of debasement, exchange rate changes, inflation, and deflation. Of course, a lot depends on what given IOU (“I owe you”) certificate is redeemable to. If it’s an existing currency (like the Dollar), it, of course, inherits all benefits and problems connected with it. But it can also be based on any commodity or chosen basket of goods (for example, a set of industrial metals). In that case, it could be hugely independent of government control, as fixing prices of sufficiently large baskets of goods is not realistic.

In such a system, money creation is not a global problem anymore. Tokens are just certificates of debt that anyone can issue. They attest to the existence of a loan or sale of goods that remains to be paid for. In a sense, every issued IOU is new money entering the market. The more optimistic the market is, and the more goods are produced, the easier it is to get IOU accepted as the means of payment. If the market condition worsens, lenders are more reluctant to give loans and the “money pool” shrinks. It might seem weird to think about money as a debt, but some people argue that’s what money really is. One of them is David Graeber, who wrote in his book “Debt: The First 5000 Years”:

Recall here the little parable about Henry’s IOU. The reader might have noticed one puzzling aspect of the equation: the IOU can operate as money only as long as Henry never pays his debt. In fact this is precisely the logic on which the Bank of England — the first successful modern central bank — was originally founded. In 1694, a consortium of English bankers made a loan of £1,2oo,ooo to the king. In return they received a royal monopoly on the issuance of banknotes. What this meant in practice was they had the right to advance IOUs for a portion of the money the king now owed them to any inhabitant of the kingdom willing to borrow from them, or willing to deposit their own money in the bank-in effect, to circulate or “ monetize” the newly created royal debt. This was a great deal for the bankers (they got to charge the king 8 percent annual interest for the original loan and simultaneously charge interest on the same money to the clients who borrowed it) , but it only worked as long as the original loan remained outstanding. To this day, this loan has never been paid back. It cannot be. If it ever were, the entire monetary system of Great Britain would cease to exist.

If we look at the money from such a perspective, a system based on debt just reflects its true nature. It allows every person to create money, but only to the extent that other people are willing to accept. In opposition to the central banks, they have skin in the game, so taking temporarily profitable but reckless decisions is not in their interest. Such a system regulates itself on the lowest possible level. Other monetary systems require artificial changes in the money pool, which always lags compared to market needs and creates unexpected side effects on the way.

If it’s so great, why does no one use that system? What is the elephant in the room?

The system based on mutual debt limits you to trade only with people you trust. You cannot pay one person with another person’s debt. Unless you deal with the same people back and forth, it is useless.

We need to be able to deal with strangers and use one person’s obligations to pay other people. That cannot happen unless we can establish trust between people who otherwise do not know each other. Is there any way to do that?

We will figure it out in part II of this series.


Virtus Lab company blog