Saturday, 17 April 2021

Crypto-mania is back: Crypto vs. Fiat, and why newer isn't always better

After a multi-year hiatus, cryptocurrency has recently stormed back to the fore, with Bitcoin breaching new highs above US$60k, along with a resurgence in a multitude of other crypto currencies, whose aggregated market capitalization has now run up to some US$2tr.

While opinions vary on Bitcoin, we have seen particularly enthusiastic endorsement emanating from the technology/software industry, including many high-profile industry figures. Some Silicon Valley companies like Square and Tesla have purchased Bitcoin at the corporate level, and Paul Graham even went as far as likening Turkey's recent crypto ban to banning the microprocessor in 1976. To these folk, Bitcoin is obviously the way of the future. This zeitgeist recently prompted someone to ask on Twitter, why are so many intelligent & capable software people so quick to embrace Bitcoin and the "future of money" with little in the way of critical thinking and only a cursory understanding of the issues?

The answer, I believe, is that these folk have - based on many decades of experience - a strong predisposition to believe that anything that is both *new* and *digital* must be superior to prior "old world" solutions. And that is understandable, because that's usually been the case with most areas software has touched. However, there is no inevitability that will always be the case in all situations. In addition, these folk understand software, but often their understanding of economics, monetary systems, and markets is much more lacking, which creates blindspots.

Instead of uncritically embracing the hype, let's take a step back and think through the issues. There are really only two potential applications/needs cryptocurrency might hope to fulfil, aside from functioning purely as instruments of speculation: (1) to function as a currency/means of exchange/payment, displacing/acting as a (superior) alternative to fiat currencies; and/or (2) to function as a "store of value". The default presumption is that - owing to putative problems/shortcoming associated with fiat currency (some true, some imagined) - Bitcoin et al are unquestionably superior - indeed a natural evolution towards next-generation solutions. Unfortunately, the reality is a lot more complicated than that.

Let's start with #1. The primary limitation Bitcoin enthusiasts attribute to fiat currencies is the ability for central banks/governments to increase the supply over time and thereby depreciate their value. However, despite what many merely assume, it is not necessarily desirable for a currency to have the attribute of sustaining 100% of its purchasing power over a very long period of time, and it is especially undesirable - nay disastrous - for currencies to increase in value (for the reasons explained below). Fiat's currencies' greatest weakness (their ability to be depreciated over time) is therefore also their greatest strength. Undeniably, they are at risk of government abuse and excessive depreciation, but the tendency for them to not appreciate, and at least slightly depreciate over time, is in fact a feature, not a bug.

Fiat currency has no intrinsic value, and in functioning as a currency, its primarily role is simply to facilitate economic activity/commerce/exchange by *temporarily* holding/transferring value long enough - and with sufficient predictability - to facilitate productive and efficient economic exchange. Without such an exchange medium, people would be forced to resort to barter, which is highly inefficient. Fiat currencies are a *tool*. They have no fundamental value - they instead facilitate the more efficient exchange of things that do have real value (labour, capital) in the real economy.

For a currency to function well, it is absolutely essential that it is neither too abundant nor too scarce. Either situation can destroy the efficient fulfilment of a currency's role with attendant negative economic consequences - some very severe. If a currency becomes too abundant and it leads to hyperinflation, it fails to hold its value long enough to facilitate efficient exchange. People/merchants begin to refuse its acceptance for goods and services, as they are no longer able to predict what goods and services they will be able to receive back in exchange. Often in these situations, local merchants start to refuse the local currency and resort to using more stable foreign currencies, like the USD or Euro.

This is fairly well understood. However, what is less well understood is that a currency being *too scarce* is just as potentially damaging, because it can lead to hording and a reluctance to engage in economic transactions. In economic systems, one person's expenditure is another person's income. Consequently, if somebody decides to save instead of spend, they deprive somebody else of income, and lower the level of aggregate demand in the economy. If enough people do that at the same time, the outcome will be a depression, as the loss of income from higher savings feeds on itself in a "doom loop" (people cut their spending in response to lower incomes; businesses lay off employees and cut investment in response to weaker demand and lower profitability, etc). This is the classic "paradox of thrift" Keynes identified as the cause of the Great Depression, and the downward spiral it can lead to in the absence of stabilizing Keynesian fiscal stimulus (the government acting as the "spender of last resort") can result in unspeakably bad depressions with devastating consequences.

For this reason, a currency should *never* be something that causes people to regret using it to purchase goods and services because it subsequently appreciates in value. If it does, then it is failing to fulfil its role as a facilitator of commerce and is creating the risk of triggering unnecessary depressions. And yet this is precisely the problem with Bitcoin. A well known meme has circulated for years about a guy who bought a pizza with Bitcoin in circa 2010. Had he instead held on to his Bitcoin, it would now be worth hundreds of millions of dollars.

A currency is functioning well if when you want to buy a pizza, you buy a pizza and never think twice about the currency you used to pay for it. The currency is practically invisible - you think about the pizza, not the means of exchange. The only reason you should ever regret buying that pizza is if you observe your waste line growing. The *pizza* is the focus of the economic transaction, not the currency which is used to pay for it. How many people bought a pizza in 2010 using USD, and today regret doing so? A "currency" that punishes people for using it to purchase goods and services is one of the most disastrous and dangerous currencies you could create. In all likelihood, it's adoption would swiftly plunge the world into an economic depression the likes of which we have never before seen. 

For this reason, fiat currencies are deliberately not designed to appreciate in value. At the very least, the money supply needs to be increased in line with GDP to avoid this outcome, lest a growing level of economic activity chase a finite amount of currency, entrenching 2-3%+ annual price deflation (and more if currency is horded, which is likely). The world experienced deflation, repeated depressions and recurring bank crises during the 19th Century, because we adhered to "hard money" ideas and had not yet figured out that it is essential currencies are not too scarce and do not appreciate over time. The utter human carnage that was the Great Depression was the final straw that ushered in a more enlightened period of monetary understanding (subject to abuse, yes, but which overall has been far more successful than the monetary regimes of the pre Great Depression era, in promoting economic stability, growth and human welfare). We learned through hard and painful experience that the hard money approaches now being enthusiastically embraced by the crypto community are economically disastrous.

The other important role of currencies is that they have *predictable exchange value*. While holding their value in the very long term (over decades) is less important, holding their value over shorter periods of time is absolutely critical. If you sign a contract to sell your house for $1.0m that settles in six weeks time, it is essential that you have confidence that the amount of goods and services you can buy with that $1.0m will be approximately the same in six weeks time as it is today. If it could only be $0.5m, it exposes you to tremendous, and potentially life-changing risk. The unpredictability of the exchange value (in terms of goods and services) of such a currency would render it far too risky and impractical to use for transactions of consequence. This is why in less developed countries with unstable local currencies, large transactions are often denominated in USD. Bitcoin and crypto also fail spectacularly in this regard, due to their hyper volatility. 

In short, for a currency to function well, you need it to be not too scarce, not too abundant; be widely accepted; and to have predictable exchange value. This facilitates commerce. The ideal situation is fairly close to what central banks in developed markets have achieved in recent decades - largely stable prices with a very modest pace of deprecation of perhaps 1-2% a year. This depreciation discourages people from hording cash, and yet the pace of depreciation is very modest such that it does not impair the predictability of cash's exchange value. It is ideal/optimal for encouraging commerce and exchange.

Though new and digital, Bitcoin is therefore an inferior currency to pre-existing systems, with many structural problems/flaws. Yes, I know it seems difficult for many to grasp that something new and digital might actually be an inferior solution to an already solved problem, but with the context properly understood, that is in fact the case. Bitcoin could function as a sideshow supplementary currency, but it would be utterly disastrous if it became the primary unit of account and means of exchange/contracting in modern economies, and would almost certainly precipitate a devastating depression.

The second potential role Bitcoin could fulfil would be to function as a long term store of value, and here some of the arguments proffered are (at least superficially) somewhat more plausible. Right now, it is not possible for somebody to earn money, store it in cash - either physically in a box, or in a bank account given negative real (if not nominal) deposit rates - and have the same purchasing power available to them in 20-30yrs as they have today. There is no mechanism of long term storage via means of cash. Bitcoin and cryptocurrencies might therefore be argued as as providing as solution to this unmet need. 

Fiat currency does not store value in the long term because that is not its primary purpose. Fiat currency is intrinsically worthless, and is therefore not a repository of value, but merely an efficient means of exchanging real world value. It functions as a "suspense" account of sorts - I will exchange item X of value for fiat with Y, because I predict that when I walk down the road to store A, they will exchange item Z of value for that fiat. The fiat itself is worthless and does not store value, but it is predictable others will accept it in exchange for value, because others will likewise do the same. 

Nevertheless, people do want to save and store value over time. On a system basis though, how is it that value is actually stored, and indeed, what exactly is value? When analysing these issues, one cannot content themselves with the micro, but must look at the macro, because fallacies of composition emerge if one extrapolates the micro to the macro (such as Keynes' paradox of thrift already discussed).

It is important to remember that both fiat and crypto currency are a figment of our imagination, and nothing tangible of value actually exists or is stored. Value exists in the real world of atoms, and currencies - fiat or crypto - are merely means of allocating, transferring, and co-ordinating that value. At a system level, value/savings/capital ultimately derives from human effort that has been expended in the past that has a replacement cost, and which has accumulated/been organized in a manner that allows us to more effectively meet human needs and desires than was the case prior to those efforts.

That's a mouthful and is perhaps best illustrated by way of example. Say for instance I were to build a hydroelectric dam. Billions of dollars worth of human effort would go into building it, from detailed engineering, to direct materials production such as concrete and steel, to complex engineered products like turbines and all their components, as well as considerable amounts of physical construction labour. Deriving further up the chain, effort has gone into discovering and digging raw materials out of the ground, like iron ore, and processing and refining those inputs into finished intermediate products like steel.

All of this takes tremendous human effort - what we often call "work" - and the outcome is a hydro dam that can produce useable electricity very efficiently, which can be used to increase human welfare/productivity, and better meet our needs and wants. This represents accumulated value, or "installed capital". It is a store of value because replacing all that effort entails a real cost, and the installed capital generates real value/utility to human beings from low-cost electricity generation. Furthermore, it is highly predictable that that utility will exist in the future (so long as humans want and need electricity).

At a system level, *that* is how value is stored, and wealth accumulates. Paper money or crypto currency that merely represent 1s and 0s on distributed computers, in and of themselves store nothing. It is the *real world stock of accumulated capital* that backs financial claims that legitimately store value. Without them, both fiat and crypto currencies are utterly worthless, and no value is stored aside from the inter-temporal exchange value discussed in the prior section of this article (intermediating between goods X and Z). Large scale storage of value, at a system level, therefore must derive from real world things that have a predictable ability to generate real world value for human beings in the future. 

Nevertheless, it is true to say that people will always demand human labour, and will be willing to exchange labour for something they predict will allow them to cash it in in the future for other humans' labour, and at present people are currently unable to store and time shift an hour's worth of labour on a risk-free basis over multiple decades by holding cash. They will instead be chiseled by inflation over time, and maybe even negative deposit rates. The closest approximation they have is lending money to other people - either directly (via purchasing bonds etc) or indirectly via financial system intermediation, who will (hopefully) pay you back by expending remunerated labour and repatriating you the proceeds. This exposers savers to credit risk/default. Does Bitcoin not potentially plug this gap?

At a theoretical level, it is arguable it indeed could, but then so could anything. I could designate a magic toilet roll as a store of value. As long as people are willing to exchange labour or items of real world value for this magic toilet roll, because they believe others will likewise accept it, it could function as a store of value. But it is important to remember that the magic toilet roll itself does not store any value. In order for you to take money "out" of the magic toilet roll, someone else must be willing to simultaneously put it in. For this reason, in reality there is no value stored in the some US$2tr market cap of digital currencies whatsoever. Nothing actually exists that can be taken out, anymore than if my magic toilet roll was valued at US$2tr. In order for any of this US$2tr to be taken out, other investors must put an equivalent amount in. But, it is also true to say that so long as people are willing to do so, it can de facto function as a store of value.

While granting this theoretical point, however, there is the much more difficult pragmatic point for the Bitcoin bulls to overcome: how many people are actually buying Bitcoin to store a stable amount of value for 20-30yrs, and how much human demand actually is there for this outcome? And are there not already other and potentially superior means of doing this?

I am yet to met someone who - saving/investing on a 20-30yr time horizon - is targeting a zero real return. Most people, when they put money away for the long term and to save for retirement, are desirous of obtaining a return on their investment and compounding their money over time. This can be achieved by buying income producing real assets, such as the hydroelectric dam I mentioned (via, for instance, purchasing shares in a company that owns a hydroelectric dam/s), and reinvesting the cash flows. It cannot be done owning an asset that functions purely as a putative store of value that generates no income (aside from speculative profits, which are an altogether different thing).

Moreover, what fraction of the people that are actually buying Bitcoin today are doing so because they hope/think it will simply preserve purchasing power for 20-30yrs, without generating a meaningful return? The truth is that the vast majority of people buying Bitcoin and other cryptocurrencies are doing so because they hope/expect to reap massive windfall capital gains. So the rationalization proferred to justify Bitcoin's growing presence bears no resemblance to its actual de facto usage.

In truth, cryptocurrencies' primary function today - and in my opinion, in the future as well - is simply to function as vehicles for speculation. Human beings like to speculate/gamble - always have and always will. Wanting to get rich - and ideally as fast as possible and without a lot of physical effort - is a virtual human universal, and humans are drawn like moths to a flame to domains they feel offer this possibility. This is the primary driver of crypto enthusiasm, even if all manner of rationalizations are proffered to justify the crypto ecosystem. The real justification is good old fashioned human greed. Few, if any, holders are buying it to store value. They are buying it to try to get rich quick.

This is also not to mention the fact that less-speculative (though still occasionally speculation-prone) long term store-of-value alternatives to fiat currency already exist, such as gold. Gold - a genuinely rare, beautiful and lustrous metal - has fulfilled this role and been desired by humans for thousands of years. Someone genuinely motivated by risk aversion and storing value would stick to a tried and true method rather than a speculative and novel approach. And if one insisted on digitization, there would be nothing to stop a gold-backed cryptocurrency assuming the throne and displacing Bitcoin, which would marry the best of both worlds. But there is limited interest in this, because it would remove the possibility of rampant speculative profits, which is the primary motivation for crypto enthusiasm. 

There are also a menu of different fiat currencies to choose from - the world does not have just one fiat currency - and investors can substitute currencies with higher inflation risk for those with lower risk, as often happens in high-inflation less-developed countries, where their economies and banking systems become partly/highly dollarized. Solutions already exist, and for most people most of the time, they are good enough.

So what could be the boom's undoing? It's hard to say. The typical downfall of most speculative manias is the capacity to add supply. IPO mania was felled by a flood of new junk IPOs that eventually sated and overwhelmed demand. Something similar is happening with SPAC IPOs at the moment - so long as there is appetite, the financial industry will continue to manufacturer more SPAC IPOs until the market is glutted and prices collapse (the irrationality of SPACs in not in the primary issuance, but in the secondary buying; without enough secondary buyers, the whole game is up).

Non-fungible tokens will inevitably collapse because of the capacity to add supply. New NFTs will continue to mushroom, while artists often take a 5-10% cut every time the asset turns. A good way to envisage this is buyers of NFT walking into a room with suitcases of cash, while artists are walking out another door with suitcases full of cash representing their 5-10% turnover cut. The whole system functions like a Ponzi scheme because there must be a continuous influx of new money in order to offset the accelerating leakage, and with new NFT supply being rapidly added, it is only a matter of time before supply overwhelms demand.

For crypto in general, the same thing will happen, as more and more crypto currencies are manufactured. But Bitcoin seems to have (so far) succeeded in entrenching itself as the gold standard and differentiating itself from the long tail of other currencies, and the supply of Bitcoin itself cannot be increased. This could result in the Bitcoin bubble inflating for a lot longer than would otherwise be the case in many bubbles. That being said, Coinbase is clipping 0.5% every time Bitcoin turns, and has a market capitalization approaching US$100bn - investors current estimate of the present value of future clippage, or suitcases of cash walking out of the door. But that may not be enough to slow things down.

How high could prices go? It's anyone's guess, because prices are driven purely by demand and supply, and demand is partly a function of price increases. However, the total amount of global wealth was estimated by Credit Suisse to be some US$360tr at 2019 year-end 1. It has probably increased since then - let's say to US$400tr. The total market cap of crypto, at some US$2.0tr, is therefore now already about 0.5% of total global wealth. 

Could it go to 5%? Anything is possible. History cautions against trying to call the peak of human speculative excess - things can be taken to utterly unimaginably absurd extremes. But in the long run, I must say it is very difficult for me to imagine crypto settling out even at 2.5-5% of total global wealth levels, let alone anything higher. US$2tr in combined market cap also also represents about US$250 per global capita - including children and people from less developed countries in Africa etc. There is still room to run further in the short term - there always is - but on the long sweep of things I wouldn't be investing today with the expectation of making 100x your money.

I used to believe Bitcoin would likely eventually go to zero. I no longer believe that. My prediction now is that we see repeated waves of speculative excess - huge giddy run ups, followed by spectacular collapses; long periods of disinterest/sideways action; and then renewed eco booms. People will always like to speculate, and the smaller the market cap gets after any bust, the less buying will be required to support and push up the price. After a bust, at some level the price will get low enough for a number of people to step up to the plate and bet on another boom that windfalls them 20x. And the cycle will repeat. It doesn't make a lot of rational sense, but then again nor does Vegas.


PS I have disabled the comments section to this blog as it has been overrun by spam. Apologies for the inconvenience to legitimate commenters.