Thursday, November 15, 2018

Cryptocurrencies' Seven Deadly Paradoxes

John Lewis of the Bank of England pens a must-read, well-linked summary of the problems of cryptocurrencies in The seven deadly paradoxes of cryptocurrency. Below the fold, a few comments on each of the seven.

Lewis' paradoxes are:
  1. BTC transaction fees
    The congestion paradox. As we saw when Bitcoin's "price" spiked last November, transactions are affordable only when no-one wants to transact:
    Bitcoin has an estimated maximum of 7 transactions per second vs 24,000 for visa. More transactions competing to get processed creates logjams  and delays. Transaction fees have to rise in order to eliminate the excess demand. So Bitcoin’s high transaction cost problem gets worse, not better, as transaction demand expands.
    Worse, pending transactions are in a blind auction to be included in the next block. Because users don't know how much to bid to be included, they either overpay, or suffer a long delay or possibly fail completely.

  2. The storage paradox. To participate directly, rather than via an exchange, a user needs to (a) have high enough bandwidth and low enough latency to the community of miners, and (b) store the entire blockchain:
    so an N-fold increase in users and transactions, means an N-squared fold increase in aggregate storage needs. The BIS have crunched the numbers for a hypothetical distributed ledger of all US retail transactions, and reckon that storage demands would grow to over 100 gigabytes per user within two and half years.
    Of course, there are proposals to make this waste of storage be the waste of resources that secures the blockchain itself. But, as I pointed out in Making PIEs Is Hard, they have problems too.

  3. The mining paradox. Since, as Eric Budish points out:
    From a computer security perspective, the key thing to note ... is that the security of the blockchain is linear in the amount of expenditure on mining power, ... In contrast, in many other contexts investments in computer security yield convex returns (e.g., traditional uses of cryptography) — analogously to how a lock on a door increases the security of a house by more than the cost of the lock.
    the cost of mining cryptocurrencies is a feature not a bug, and thus miners have to defray their costs:
    Rewarding “miners” with new units of currency for processing transactions leads to a tension between users and miners. This crystalises in Bitcoin’s conflict over how many transactions can be processed in a block. Miners want this kept small: keeping the currency illiquid, creating more congestion and raising transaction fees – thus increasing rewards for miners facing ever more energy intensive transaction verification. But users want the exact opposite: higher capacity, lower transactions costs and more liquidity, and so favour larger block sizes.

    Izabella Kaminska points out this tradeoff has been *temporarily* masked by capital inflows creating subsidies via the mining rewards system. ... A private cryptocurrency must continually attract more capital inflows to mask the transactions costs (a staggering ≈1.6% of system payment volume). By contrast, most traditional mediums of exchange don’t require such sizeable capital inflows to maintain their transactions infrastructure.
    BTC "price"
    This need for an inflow of funds from investors who believe the "price" must inevitably go up means that if a cryptocurrency's "price" is dropping or flat for a long time the currency's "price" risks a collapse as miners are forced to dump it to pay bills.

  4. The concentration paradox. As I discussed in Gini Coefficients Of Cryptocurrencies, HODL-ings of cryptocurrencies are highly concentrated:
    An asset is valued by the market price at which it changes hands. Only a fraction of the stock is actually traded at any point in time. So the price reflects the views of the marginal market participant. You can raise the value of an asset you own by buying even more of it, as your purchases push the market price up. But realising that gain requires selling- which makes someone else the marginal buyer and thus pushes the market price downwards.

    For many assets these liquidation effects are small. But for cryptos they are much larger because i) Exchanges are illiquid, ii) Some players are vast relative to the market iii) There isn’t a natural balance of buyers and sellers iv) opinion is more volatile and polarised.  High prices reflect cornering the market and hoarding, rather than ability to readily sell to a host of willing buyers.
    The big HODL-ers cashed out around $30B in last November's massive pump-and-dump scheme.

  5. The valuation paradox. Why does a cryptocurrency have a value?
    The discounted cashflow model of asset pricing says value comes from (risk-adjusted, net present discounted) future income flows. For a government bonds it’s the interest plus principal repayment, for a share it’s dividends, for housing it’s rental payments. The algebra of pricing these income flows can get complicated, but for cryptocurrencies with no yield the maths is easy: Zero income means zero value. ... What other sources of value are there? The mere expectation that in future cryptocurrencies will be worth more than they are today and so can be flipped for a profit? The problem is that if, as Paul Krugman argues their “value depends entirely on self fulfilling expectations”, that is a textbook definition of a bubble.
    Why do people think cryptocurrencies have a "value"? Because their "price" is displayed on websites just like "fiat currencies" and equities. But, unlike "fiat currencies" and equities, the "price" of cryptocurrencies is the result of massive manipulation, primarily via the Tether "stablecoin" and massive wash trading, in extremely thin markets.

  6. The anonymity paradox. Most cryptocurrencies provide pseudonymity, not anonymity, and unmasking the person behind the pseudonym is easy in practice. But even users of cryptocurrencies that claim anonymity such as Monero need exceptional opsec to avoid exposure. But Lewis makes a different point:
    The (greater) anonymity which cryptocurrencies offer is generally a weakness not a strength. True, it creates a core transactions demand from money launderers, tax evaders and purveyors of illicit goods because they make funds and transactors hard to trace. But for the (much bigger) range of legal financial transactions, it is a drawback.

    It makes detecting nefarious behaviour harder, and limits what remedial/enforcement actions can be taken. Whilst blockchains can verify a payment has been received and prevent double spending (albeit imperfectly), many other problems are unsolved.
    Auer and Claessens demonstrate empirically that developments which help establish legal frameworks for cryptocurrencies increase their value. Keep a cryptocurrency far from regulated institutions and you reduce its value, because it drastically restricts the pool of willing transactors and transactions. Bring it closer to the realm of regulated financial institutions and it increases in value.
    Given the likely advances in forensic technology over time, and the authorities' long memories, recording your illegal transactions in an immutable public ledger is probably foolish.

  7. The innovation paradox:
    Lewis' last paradox is the most entertaining:
    Perhaps the biggest irony of all is that the more optimistic you are about tomorrow’s cryptocurrencies, the more pessimistic you must be about the value of today’s.

    Suppose bitcoin, ethereum, ripple et al are just the early flawed manifestations of an emergent disruptive technology. Perhaps new and better cryptocurrencies will arise to overcome all of the intrinsic problems of today’s. ... Whereas goods derive worth from their value when consumed, currency derives worth from the belief that it will be accepted as for payment and/or hold its value *in the future*. Expect it to be worthless in the future, and it becomes worthless now. If new cryptocurrencies emerge to resolve the problems of the current crop, then today’s will get displaced and be rendered worthless.
    Cryptocurrencies are mechanisms for transferring wealth from later to earlier adopters. That is the importance of "number go up" as David Gerard puts it. Thus, unlike "fiat currencies", there is a continual demand from new wanna-be early adopters for new cryptocurrencies to be created. There are already a couple of thousand of them. There is a real chance that the intensive research into blockchain technology will result in a "new and better" cryptocurrency.
Go read the whole post and click on the links, you'll be glad you did.


David. said...

Jemima Kelly reports Forking hell, crypto is collapsing!:

"Here's a one-month view of the king of crypto, bitcoin, which at pixel was down almost 13 per cent over the last 24 hours. Having managed to stay above $6,000 for the whole of 2018 (albeit down from around $20,000 last December), it has now broken through that level and is trading at its lowest in more than a year, at around $5,500 (screenshot from industry news site Coindesk):"


"In the top 50 cryptocurrencies listed on CoinMarketCap, just one was showing any kind of stability: USD Coin, a dollar-based coin issued by Goldman Sachs-backed crypto wallet company Circle and the centralised overlords of the decentralised future, Coinbase. Stability in the sense that traders were so desperate to get out of their positions in free-floating cryptos that they were willing to pay a 2 per cent premium for the luxury of holding something they can redeem for dollar. (It was trading around $1.02 — a bit like a bond yielding minus 200 basis points.)"

David. said...

Timothy B. Lee's Bitcoin and Ethereum fall to lows not seen since 2017 is already out-of-date. As I write BTC is trading around $4.8K, down 14% in the last 24 hours, and ETH is around $148, 15% down. These are prices from last October 10, at the start of the massive pump-and-dump.

David Gerard wrote yesterday:

"The other big news this week was the price of Bitcoin dropping $1000 in one fell swoop — someone appears to have dumped a massive pile of coins at around 1600 UTC on 14 November. Nobody knows who, or has a solid reason. I would guess that someone got squeezed and had to sell up fast."

It looks like another whale just cashed out today.

David. said...

The End Of Trust, a special issue of McSweenney's developed with the EFF, has a chapter in which Edward Snowden explains to his lawyer, Ben Wizner. It is worth reading, but with skepticism. Snowden is completely accurate about the underlying technology, but far too credulous of its ability to deliver on its promises.

The special issue is free to download, and at first glance contains lots of worthwhile information, so go read it.

David. said...

Cryptocurrency "prices" are still in free-fall. As I write BTC is down over 11% in the last 24 hours and ETH is down over 15%.

David. said...

Dave Michels' You may not actually own your Bitcoin – legal expert adds to the fun by pointing out that cryptocurrencies don't meet either of the two common law categories of things you can own:

"Personal property includes rights over two categories of things. First, there are “things in possession”. These are tangible items which you can physically possess and transfer to another. The £20 note in your pocket is a thing in possession.

Second, there are “things in action”, a mixed category of rights that can only be claimed or enforced by legal action. This includes debts, rights under contract, and intellectual property. The £20 you have deposited at a bank is a thing in action, because the bank owes you a debt of £20. That debt is intangible, but, if necessary, could be enforced through legal action."