Distributed Ledgers Part I: Bitcoin is deadTags: bitcoin, blockchain, btc, distributed ledgers
“What awaits us is never what we were awaiting” French Fintech Investor, circa July 2015, Washington DC.
I uttered the words “Bitcoin is Dead” at FintechStage in Barcelona a few months ago, see previous post here. Strong words for a non-techie who has a limited understanding of the subtleties of all things crypto. BTW, this is your cue to mock me.
Other than the pure pleasure of shock and awe and the weak ego driven desire to win an Oxford debate, what possessed me to utter such words and what exactly do I mean by these words.
Let me clarify.
First, I am in awe at whoever came up with Bitcoin. Simply brilliant in its simplicity and theory. Second I wish the currency itself was less volatile and the protocol tweaked so the blockchain could deliver on its promises. Third, I never lose sight of the facts, see my previous post of First Principles, here.
And the facts are:
– Although proof of work is the best there is out there crypto-security wise, it is very energy intensive and not optimized to scale. It is a massively inefficient protocol by design. Further the forward state of the Bitcoin blockchain, even if using proof of work, is dependent on merchants, exchanges, miners, wallet providers, multi sig service providers… and can be or is being undermined/centralized in different ways (see the latest development with forks in the chain and miners cherry picking which transactions to confirm).
– Block size is an issue and a limiter – there are currently many discussions around this topic.
– Speed of transaction confirmation is an issue and a limiter. I have been told there are a couple of initiatives to deal with speed but I am unsure of status or feasibility.
– Miners who are tasked with both mining for new blocks and confirming transactions, are incentivized via seignorage. They get rewarded when they “discover” a new block and get paid in BTC, betting on the rise of said BTC. Miners do not get incentivized properly for confirming transactions. It is unclear whether a solution to incentivize miners for confirming transactions is being worked out sensibly in the Bitcoin marketplace today. If Miners do not get paid to confirm transactions, the blockchain is at risk in the future. Even if some of the initiatives around transaction fees gain traction they will inevitably require a hardfork and a radical change to the bitcoin protocol which although not impossible may be rather unlikely. As Robert Sams pointed out to me, transaction fees reflect the marginal cost of verifying a transaction so the security of the network will decline as the block award keeps halving and this means that miners will not be incentivized to mine at a sufficient high hash rate when the reward shifts way from seignorage to transaction fees. If you think through this important insight, the bitcoin protocol needs to keep the marginal cost of transaction fees high (the debate around changing the block size limit is relevant here.
– Processing transactions on the bitcoin blockchain is neither cheap nor free. Any business model that banks on this assumption needs to be rethought. “Free” or “Cheap” cannot exist forever.
– Mining concentration is an existential threat to the blockchain. It is unclear how this concentration can be avoided especially if concentration is unavoidable due to how the bitcoin protocol is set up.
– Even though proof of work is the safest crypto security protocol at present, the blockchain is still at risk of being attacked. Suffice it to say that no one can guarantee 100% that the blockchain cannot be attacked which means no one can guarantee a 100% that a transaction that has been validated will remain in its validated state. This has profound implications on the finality of a purchase made on the blockchain. If there is no finality assured 100%, even if it is assured 99.5% for example, many use cases will elude the bitcoin block chain. Probabilistic settlement will definitely not be accepted in the capital markets world where participants require to hold any system they use to the highest standard possible. Other use cases where requirements may not be as stringent might do with probabilistic settlement? Maybe
– The Trustless aspect of the bitcoin blockchain is its greatest asset. It is also its greatest weakness, depending on the use case. Actors known to one another who work off of precisely defined AML/KYC parameters have no need for a trustless paradigm. Put in other words, the regulatory friction of a trustless paradigm is extremely high, especially in the financial services industry where knowing the actors across the value chain of a transaction is crucial.
– The deflationary quality of Bitcoin – by design and not by accident – is a major drawback. Deflation naturally encourages hoarding and delays spending, which is the behavior we are witnessing with BTC holders. Imagine a “currency” that would be deflationary by nature, adopted by a nation. Spending would be delayed. With decreasing spending, investments would be delayed. We know all too well what deflation does to countries. (For that matter we also know all to well what hyper inflation does to a country too but I digress.)
If I distill the above facts I factually conclude that the bitcoin blockchain does have its limitations. What are the implications of these limitations? Do the implications vary from one use case to another, one industry to another? Can these limitations be mitigated now, over time?
Let me now switch to certain use cases I have seen in the market. I limit myself to reviewing use cases in the financial services industry. There is no doubt in my mind there may be interesting opportunities in other industries.
The below use cases listed below apply to the open/permissionless Bitcoin blockchain (I am purposefully not discussing permissioned ledgers nor am I discussing other Bitcoin-like permissionless blockchains) .
Here are the current investment cases that do not make sense to me:
– Mining: I do not understand this investment opportunity. It seems the activity is capital and energy intensive with dubious long term resiliency and investment returns in light of the fixed number of bitcoins.
– “Currency” in Developed World: Using BTC as a means of exchange makes little sense to me in Western Europe, in the US. Why would anyone want to buy BTC by selling fiat currency which is “relatively” stable and which can be used without friction every day? Especially when the future price of BTC is high or volatile or uncertain? What can possible compel an average consumer to buy bitcoins and use bitcoins when it is so easy to use fiat currencies? I can see buying BTC as a speculative investment, holding it in the hope of a price increase but that would defeat the purpose of using BTC as a currency for payments.
– Bitcoin as a Retail Payment Rail in Developed World: Processing of transactions is slow, see the issues with Currency above, I do not get it.
– Merchant/Transaction Acquisition play in Developed World: Merchants can be acquired but if there is no one to purchase in BTC, see above points, then I do not get it… again.
– Bitcoin Exchanges: The opportunity will always be limited by the total amount of BTC in circulation and the less volatility BTC will experience, the less the investment returns will be attractive.
– Reserve “currency”: I had to include this one in the mix even though most Bitcoin investors clearly did not invest with this idea in mind initially. The CEO of Coinbase recently tweeted he thought Bitcoin would supplant/challenge the $ as the reserve currency in 10-15 years. There are reasons why the World left the gold standard behind: short term volatility, deflationary pressure, difficult to implement, constraints to economic growth… These are the same reasons why Bitcoin will never be a/the reserve “currency” – or should we use the term reserve “asset” which is more appropriate. This use case would be more plausible where the deflationary quality of Bitcoin not be present by design and the supply function be elastic but I digress again.
Note that these investment cases, minus the reserve currency one, are all part of the first wave of investments in the bitcoin ecosystem. They are in their totality what some refer to as the plumbing, the infrastructure of the ecosystem. Also note these were built with the premise of fast and wide adoption. We have yet to see any meaningful adoption in the retail world and the institutional world seems quite lukewarm.
Here are recent use cases that I have a hard time making sense of:
– Applications focused on the clearing and settlement of dematerialized securities in the capital markets world where the need for finality of settlement and a 100% proof of ownership is a requirement. THIS IS A HUGE ISSUE, which Robert Sams has astutely pointed recently. See my point above on the risk of attack on the bitcoin blockchain.
– Applications focused on facilitating trade finance or supply finance where finality of legal transfer of ownership from one point to the other along the value chain is also a requirement. Same issue as point above.
– Applications focused on asset registries where one of the stakeholder is a financial institution (buyer or seller) where AML/KYC requirements demand full transparency of other stakeholders. I will hedge myself here as depending on the stringency of AML/KYC requirements this use case may make sense.
– Applications that run counter to a legal framework or regulatory framework where AML/KYC requirements will not tolerate certain actors participating to the blockchain to remain anonymous – or should I say pseudonymous.
I am sure I am missing many more.
Although I do agree the banking and capital markets world will benefit tremendously from distributed ledger technology providers who will work with the current legal, operating and regulatory frameworks, I cannot help to be puzzled by how the bitcoin ecosystem has tried to impose a technology solution without fully understanding if that solution was a good match for existing problems.
Here are some use cases where the bitcoin blockchain and BTC might make sense:
The reason why these use cases may make sense is that some or all requirements for an open protocol that is trustless and where anybody or anything can be connected to it with little regulatory friction are present – lack of alternative, no trust in fiat currency or local/central government, alternatives too costly, regulatory light frameworks…
Put in more sophisticated words and borrowing from Richard Gendal Brown: “Are there use cases where an opened permissionless blockchain – or ledger – is worth keeping an eye on?” The answers may vary based on the very framework supporting these use cases which may not hold constant over time.
– As a retail payment rail in emerging markets: In geographical locations where the local fiat currency is very volatile or unstable. For example Argentina. In geographies where incumbent payment rails are either prohibitive or have low penetration and where financial services infrastructure does not enjoy a high level of trust. Were a bitcoin like protocol emerge with faster settlement and better usability, adoption would happen much faster than with Bitcoin though
– As a b2b payment rail in emerging markets: Across certain currency pair corridors where incumbent solutions are either slow or expensive – read PayPal, Western Union enabling invoice payments for micro SMEs for example. Caveat being there needs to be enough liquidity per currency pairs
– As a remittance rail solution with destination being emerging markets: In geographic locations where the recipient of the remittance has a poor choice or an expensive choice from current incumbent solutions. Again, there needs to be enough liquidity per currency pairs
– Applications for pure bearer assets: Gold is the perfect example imho. Some – and Taulant Ramabaja is one of them – will disagree with me and note that although bearer assets are Bitcoin compatible, this only holds for assets endogenous to the bitcoin blockchain and that Gold certificates, although a bearer asset, still need a counterparty
– Applications that deliver asset registries and transfer mechanism where trustless mechanisms are a must: When actors are not know to one another, where the asset is public or a public good which most times creates “tragedy of the commons” behaviors between actors – in international settings even more so. Examples of such assets or public goods may be carbon trading, international IP rights, water rights… Again, some may push back and stress that all the above examples need counter parties and as such the Bitcoin blockchain is not suited. Interesting debate.
– Applications that deliver asset registries for privately held securities: where by definition the finality of settlement and ownership does not exist. VC investors will be familiar with this given the recourse they always request in closing documents when investing in startups – indemnification clauses, reps and warranties. Financial institutions may be involved here when the AML/KYC issues have been mitigated. The pilot NASDAQ is running with private securities on the bitcoin blockchain is a perfect example here.
– Applications that supports financial applications that deliver financial smart contracts: to users that do not have access in real life to capital markets or specific financial assets due to their geographical location and where said users do not know one another. Again, argument that another more efficient protocol may gain faster traction is valid.
– As a platform to manage an individual’s identity and/or data: as an individual interacts with others and financial institutions.
I am struck by a common thread here. Except for identity management use case and asset registries for private securities, all address marginal applications in the financial services industry where certain weaknesses or gaps can be exploited positively and more than compensate for potential friction. Core use cases come with too much friction. To me there are two types of core use cases in financial services that make up the bulk of all activities: a) retail driven which address payments or wealth management and b) institutional driven which address either b2b payments, asset management or the plumbings that allow incumbents to operate in capital markets, lending, insurance…
Getting back to the title of this post and what I meant: Thhe first waves of investments in the Bitcoin ecosystem have focused on mining, retail payments, merchant/transaction acquisition, store of value, mobile wallets, exchanges. None of these have wide application with the public at large or financial institutions nor do I see wide application in the future based on current technology limitations or current regulatory constraints. I also do not believe in wide applicability across many work flows and processes in financial services and with incumbents to be more precise. Further, tacking onto a unique Blockchain many different use cases, many different apps may end up weakening said blockchain for that matter – a point I have come to realize while discussing with Tim Swanson.
My neutral/negative stance may be erroneous. To be wrong, several things will have to happen. First the technology limitations of the bitcoin blockchain would have to be removed – or some of the main ones mitigated. I cannot predict whether this will happen. My bet is such an outcome will be very difficult to produce in due time. Second the legal framework organizing banking or insurance in general and the capital markets or ownership of financial assets in particular would have to be tweaked. This would be a massive undertaking requiring global collaboration between many actors. Not impossible but very slow to achieve. Third, the regulatory framework would have to evolve and deepen its understanding. Again not impossible and again would require much time and effort. Time is of the essence here and I rate as a higher probability the advent of new crypto platforms and technology. I assume very bright young engineers are hard at work building the next generations of distributed ledgers, shared ledgers or blockchains – ones that will presumably not come with some of the bitcoin limitations and that will be built to specifically address real problems.
In hindsight, I should have uttered the words “In light of limited applicability in the financial services industry, the brave “new-world” promise of Bitcoin is dead” so that my intent would have been clearer and possibly less provocative. Bitcoin cannot solve any and all problems in the financial services industry. There cannot be a one size fits all solution here.
The more fascinating question to me is, if I am right in stating there is no one size fits all paradigm, what other protocols, distributed/shared ledgers are of promise and for which use cases? The sky may very well be the limit here and the financial services industry will be profoundly impacted by many different distributed ledger technologies in the near future – not by Bitcoin or just Bitcoin. This needs to be the subject of future posts. Stay tuned.
ps: Were Tinkerbell wave her magic wand and instantly solve Bitcoin’s volatility, long term deflationary conundrum and bitcoin’s technology drawback, its future would be quite rosy indeed. Then again, it would not be bitcoin anymore. Let’s take a practical example and assume Tinkerbell were to magically do away with the Bitcoin supply cap and adopt a more stable mining award – anathema to Bitcoin purists for sure – the resulting protocol might be much more resilient and could compete on better footing with the likes of Visa or PayPal – for this idea I have Robert Sams to thank.
pps: I would like to thank Tim Swanson, Richard Gendal Brown, Robert Sams and Taulant Ramabaja for their comments and advice, questioning my sanity to publish this post or suggest I change the title lest I get assaulted by haters – they are in no way responsible for my thought process and conclusions. I am grateful for their support.
ppps: For full disclosure, At R66 we have invested in two crypto companies to date: Ripple (a permissioned ledger play) and Mirror (a smart contracts platform for financial assets currently based off of the Bitcoin blockchain). We recently submitted a term sheet to a Bitcoin remittance company and a Bitcoin B2B SME invoice payment company. We lost both as each company signed with other investors at valuations we could not digest. Although both companies used the bitcoin blockchain I believe both could and may switch to another protocol in the future, should the move prove to be warranted. The R66 team continues to learn about the crypto space every day and continues to be excited by the wide world of investment opportunities which we are convinced will revolutionize many aspects of the financial services industry.