How to expand without changing Bitcoin?

Argument: Point-to-point cryptocurrency is useful for online transactions: online transactions need to process millions of transactions per day

Conclusion: To be successful, the cryptocurrency must be expanded.

In the past decade, for industry forums and practitioners, expansion has always been an eternal theme. In this article, I will assume that the expansion plans that have been talked about over the past decade have not hit the point, and I will propose another framework. I believe that "institutional expansion" is a neglected expansion direction, and its implementation is likely to not need to damage the reliability of Bitcoin.

What I want to talk about here is the Finney-style bitcoin, in which Bitcoin banks emerge and issue notes with Bitcoin as reserves. If you look closely, the prototype version of this system already exists today. However, in order to ensure that scarcity is not compromised, exchanges and custodians need to make routine proof that their reserves match their liabilities.

Prophet in the mail

Five hours after Nakamoto published the white paper, James A. Donald made his first public comment in the encrypted mail. In response to the white paper, he pointed out:

If there are hundreds of millions of people trading, it is a lot of bandwidth, and everyone must be informed about all transactions or a large part of them.

James's questioning is the same as many people. Bitcoin can only work if everyone keeps the books and keeps them in sync. If the cost of synchronizing the books is too high, then only a few privileges can keep the latest books, which is undoubtedly flattened. The operating system adds a hierarchy.

Interestingly, Satoshi Nakamoto mentioned the SPV proof when answering this question. It seems a bit naive to people's perception of Bitcoin today. The SPV proves that on the surface it is possible for a non-full node to know that a transaction has been confirmed without having to synchronize the entire chain. Using the SPV proof example as a solution to the expansion, it's a bit like the scientist behind the Apollo program said, “Oh, going to Alpha Centauri? It's very simple, we just need to hurry faster than the speed of light.”

It can be said that the SPV proves that it has almost been abandoned as a viable expansion method. In complex real-world situations, it tends to crash because users have to verify the entire chain.

James’s question is correct. He immediately understood that Bitcoin was a single ledger and that all nodes in the network were re-confirmed every 10 minutes. Since the books are to be transparent to everyone, hundreds of millions of traders will kill the system.

But what if the "bitcoin is a global currency for global, online, peer-to-peer transactions" is wrong in itself? Here you need to introduce Hal Finney.

Hal Finney's perspective

In 2010, digital cash pioneer Hal Finney provided a notable case for the institutional expansion of Bitcoin.

In fact, banks can use Bitcoin as a reserve to issue their own digital cash currency and convert it to Bitcoin. Bitcoin itself cannot be extended to allow every transaction in the world to be broadcast to everyone and recorded on the blockchain. There is a need for a more lightweight, more efficient secondary payment system. Similarly, the confirmation time required for bitcoin transactions is impractical for payment.

Banks with bitcoin reserves will solve these problems. They can be like banks before currency nationalization. Different banks have different policies, some are more active and some are more conservative. Some may be partial reserves, while others may be 100% bitcoin reserves. Interest rates may vary, and some banks' cash may be traded at a discount to other banks' cash.

In the perfect vision, Hal knows that the base layer of Bitcoin will never expand to the required level according to the current model. (Unfortunately, many Bitcoin evangelists don't understand this. Their misunderstanding led to the block expansion war of 2015-2017.) In Hal's view, Bitcoin will be a high-performance currency for financial purposes. Large settlements between agencies, not for small cash payments. He realized that Bitcoin's rather slow settlement time (compared to physical cash or credit cards), coupled with the inefficiency of the chain itself, meant that Bitcoin was out of place in the context of daily payments.

What Hal is envisioning is a banking system that can be audited and whose asset-liability ratio is transparent and accountable. In the free banking system, the ratio of reserves/circulating funds is dynamically changing, as depositors can choose banks with different reserve levels to accommodate their risk appetite.

Banks with insufficient reserves may fail, but this will be a healthy market signal, and the survival of the fittest can make the entire market stronger. A review of the 2008-2009 disintegration of the banking system will reveal that financial institutions know that they will get help if there is a problem, so they build up leverage. Since the government has made it clear that it will not allow banks to close down, this valuable feedback mechanism has disappeared in the market and the risks have become increasingly difficult to detect.

In the words of Elaine Ou:

Financial institutions make people feel safe by concealing complex risks. The cryptocurrency puts the risk in the most visible places and boasted on the Internet.

In the financial sector, risks are never lost even if they are hidden. And artificially suppressing risks often has unacceptable effects and tends to release them in a more dramatic way.

The risks spread quietly before us. Financial institutions failed one after another in a series of false balance sheets in 2009. When our current currency experiment finally disintegrated, the systemic risk of long-term suppression would also be released.

Can Bitcoin alleviate this problem? Maybe not. But its structure is conducive to creating an alternative financial system that is more transparent and open than the current financial system. This is the Finney-style bitcoin: Bitcoin is a virtual commodity and a provable reserve for financial institutions. It is not a liability for anyone, it is a virtual commodity that a bank can rely on when it verifies its viability.

Capacity expansion

Briefly review the implications of capacity expansion. It is now clear that simply raising the upper limit of the block size does not work. This is because Bitcoin needs to be auditable, and the audit blockchain requires a complete, unabridged ledger.

Fundamentally, Bitcoin needs everyone to know every transaction. Can you expand without affecting this core function? Let us look at this perspective, the expansion can be divided into the following categories:

Deferred settlement/reconciliation (mainly lightning network). The role of Lightning Networks and other delayed settlement models is to give users the ability to create relationships and settle them later. The reliability of the main chain still exists and is available, they are simply not used to guarantee every transaction. However, these models temporarily weaken reliability and make some trade-offs. For example, the final settlement is no longer an instant settlement, and you must stay online to receive payment.

Database model (large scale base layer extension). As mentioned above, simply increasing the size of the ledger will affect the reliability of the blockchain, as this will result in not everyone being able to maintain the ledger. There may be a way to do this with SPV and fraud evidence in a way that trusts are minimized, but we haven't found it yet.

Extend reliability to other chains (sidechain, safe inheritance, merge mining). This model uses bitcoin reliability to protect other blockchains or extend the bitcoin block space. Merging coins like Namecoin, proof methods like Veriblock, and sidechains like Rootstock fall into this series. These represent a compelling potential extension because they extend the settlement reliability of Bitcoin to potentially infinite block space, but it is still being explored. However, reliability may be compromised and there may be a risk that the miner will review the sidechain or otherwise interfere with the sidechain. We've seen products like Liquid choose the alliance, not the PoW.

The organization with the least trust. This approach takes advantage of Bitcoin's reliability: natural auditable, scarce digital cash, and applies it to depository institutions. In short, unlike individual users who are bitcoin users, exchanges, banks, and custodians are the end users, and their users benefit indirectly from the reliability of Bitcoin. The trade-offs still exist. Some characteristics of Bitcoin do not apply to depository. However, if an agreement such as the proof of solvency is implemented, although it is reviewed by an intermediary, part of the security of Bitcoin can still be exerted.

What should Bitcoin Bank look like?

Is it reasonable for Hal to envision the banking system with Bitcoin as its reserve? In a sense, it is the world we live in today, because many users only indirectly contact Bitcoin through custodians and intermediaries. While most exchanges are considered to be 100% reserves, they are usually claimed, but in reality this is not universal. For example, QuadrigaCX clearly maintains a portion of its reserve reserves during most of its operations. I don't need to look back at the dirty history of scams and negligence in cryptocurrency exchanges.

Things as simple as proof of solvency will cause Quadriga to wait until the fall of the day to expose its problems. Because, in the real world (imagine a world where solvency proves to be common in exchanges) Quadriga will refuse to prove their reserves and will be questioned in order to avoid the heartache of user asset loss early.

An ideal bitcoin bank will use a model like proof of solvency to pass the reliability of bitcoin to depositors. Of course, this is not perfect, and can be deceived, but it is difficult to implement. If you are a public company, you can deceive your auditor, but you may be discovered at some point, and you are violating the law. Any audited Bitcoin bank will only decide to audit itself if they believe they can pass the audit. As mentioned above, if the reserve certificate becomes popular, it will classify reputable and trustworthy banks in the bitcoin deposit industry. These banks routinely prove the reserves, while some banks are suspected of being unwilling to accept audits. And thus classified as a bank that is not trusted.

What needs to be clear is that I don't object to IOUs that flow between banks and banks, but they usually don't know the ownership of Bitcoin. What I want to say is that by providing certain guarantees to depositors, these oweds can be made more like bitcoin.

How to expand without changing Bitcoin?

The above diagram shows that if lightning network and other chain or layering methods extend the reliability of Bitcoin elsewhere, exchanges with proof of solvency can do the same. The sidechains (if clarified) and the lightning network and the institutional expansion model are not mutually exclusive, I see them as parallel and complementary ways to expand bitcoin. One thing to note is that the arrears issued by exchanges without a reserve certificate are of little value and are far removed from the Bitcoin base layer.

There are also some issues worth mentioning: Lightning Networks and other Layer 2 solutions are likely to become mainstream methods of capacity expansion, but the reliability of these solutions is still not comparable to Bitcoin. The characteristics in Bitcoin are not established in Lightning Networks. of. Although there are no fundamental errors, these programs do not give you rock-solid settlement guarantees like pure bitcoin, and lightning network enthusiasts and developers will recognize this. Since the precedent for Bitcoin to expand under various trade-offs has been determined, it should be extended to various agencies.

Bitcoin credit creation

Many holders will be horrified to oppose the term "partial reserve fund", even though the first disciple of Nakamoto's disciple, Harvini, will say the objection. However, I believe that if banks are responsible for the free market and disclose their actual reserves, then the risk of some reserves is controllable.

I think the problem with exchanges operating on partial reserves is that they don't have 100% reserves, but they disclose unrealistic risks to depositors. Although this has been a heated debate among Austrian economists, I personally support a free depository market where exchanges can operate at various reserve or capital ratios.

The most important thing is that they need to disclose the reserve status so that users can fully assess the risk of insolvency. As we all know, banks do not actually need 100% reserve because users usually do not withdraw all their deposits immediately. For example, in the United States, larger depository institutions must retain at least 10% of their reserves. I don't know what this number should be for Bitcoin, but I believe the market can find this number. Obviously, the popularity of lending institutions like BlockFi suggests that some users prefer interest-bearing accounts, so they can tolerate more risks in their banks.

What does the proof of solvency actually prove?

So far, I have always believed that proof of solvency and proof of reserve are basically homogeneous, but this will damage their own meaning. In fact, I should use terminology more accurately. Proof of reserve can prove the goods you actually own, but if there is no corresponding proof of liability, the proof of reserve is usually meaningless and can only prove the goods you claim to own. The two are published together to prove the condition of solvency.

How to expand without changing Bitcoin?

The first method to prove solvency is the paradigm of Greg Maxwell and Peter Todd, which we call the Merkle approach. According to Zak Wilcox, the Merkel method allows exchange users to verify that their balance is included in the list of all customer balances posted in the certificate's certificate. This process has two parts that prove what you owe and show what you have. As Greg described:

First, it's easy to show you how much actual money you have on the chain by signing a message.

Then you need to prove how much you should have. There is a small problem here. For example, you can post each person's balance according to the account ID, but for privacy and business reasons, this is unacceptable.

Proving that the reserves are actually very easy, if the exchange uses all of their UTXO signatures for a transaction, then everyone can see that the exchange has n BTCs. Of course, the exchange may borrow bitcoin for this purpose. That's why only continuous proof is effective, and should be combined with cash flow analysis (imagine, an exchange habitually borrows 10,000 BTCs in the week before the quarterly reimbursement test, and falls off the second day of the test. it!)

The challenging part is how to prove your debt, that is, what is your debt to the depositor. This is where the Merkel tree comes in. It allows users to verify that their accounts and balances are included in the final hash without revealing private information such as balances and account information for everyone. Like group immunization, if a sufficient amount of verification is performed on their balance, the user can relatively strongly believe that the exchange is not lying.

A bad exchange can of course cheating. For example, the balance of a hidden account that claims that they do not wish to perform an inspection is zero, but they have a great risk of doing so. Even if one of the zero-balance accounts is checked, the fraud will be exposed. .

As Zach said, the Merkel method

Providing a way to convince yourself to check whether the exchange recognizes your liability in a public statement. If these numbers are different from the actual situation, you can make an informed decision about whether to continue using their services.

Blockstream's Steven Roose has determined the process of proving the reserve ratio through an example from BIP and Github, but requires proof of liability (as described above) or a reliable auditor's cooperation.

The problem with the Merkel method is that it exposes the liabilities of the exchange, and many exchanges may be reluctant to do so. Therefore, in 2015, Dagher, Bunz, Bonneau, Clark and Bohne issued the “Rules: Privacy-protected solvency for Bitcoin exchanges” prove". In response to the shortcomings of the Merkel method, Dager et al. pointed out in the Rules:

To enable Exchange X to publicly prove that it has enough Bitcoin to cover the balances of all its users, it also ensures that:

All users' accounts are completely confidential

No account has a negative balance

Exchanges do not need to disclose their total liabilities or total assets

Exchange does not need to disclose its bitcoin address

The Rules include three agreements:

Proof of Assets (Reserve): The exchange uses some Zero Knowledge Proof (ZKP) techniques to prove that it has a certain number of BTCs without revealing the number.

Proof of Indebtedness: The sum of the balances submitted by the Exchange and the need to ensure that the balance submitted by the Exchange is correct without revealing the privacy to the depositor.

Solvency proof: In the case of a zero-knowledge proof, the exchange proves that its assets and liabilities are balanced.

This is an improvement over Merkel's and signed message methods because it does not disclose the exchange's balance, but instead outputs a simple 1 or 0 to prove whether the exchange is repayable.

In short, among the Merkel methods and the various zero-knowledge proof methods that have been proposed, there are a large number of tools that enable Bitcoin banks to prove their solvency. Now, they have no reason not to do so.

Where is Bitcoin Bank?

So if Finney-style Bitcoin banks can help with bitcoin expansion, do they really exist? Large exchanges and custodians are just a few other trusted third parties. As the gatekeepers of Bitcoin, they have to say that they often do more harm than good, weakening the ability to open access and free exit.

Coinbase, Bitfinex, and Xapo's BTC owes do not give users the same trading guarantees as the original commodity bitcoin, but they still represent a direction of expansion, even if it is less popular. Those who claim to believe in institutional expansion, such as Xapo's CEO, Wences Casares, mention:

A lot of transactions happen in Xapo, because Xapo internal transfers don't need to go through the blockchain, so they are instant and free. So so far we have seen that in Xapo's business, the ratio of internal transfer to blockchain transfer is about 20:1.

It is easy to see how bitcoin banks issue notes based on deposits, which is equivalent to a heterogeneous chain. To make it credible, you need convertibility. This is the same problem as Tether, and for some time no one believed they could redeem the USDT. A sustained reserve certificate will help resolve the issue.

As noted above, the Reserve Audit certificate allows depository institutions to demonstrate that they hold a certain amount of reserves and then use the help of a trusted auditor to prove that their liabilities match their reserves.

Alternatively, you can let users determine that their internal balance matches their own deposits, and if there are enough users to do so, your exchange is taken for granted as solvency. It should be noted that the proof of reserves is not a cure. Coindesk's Danny Bradbury pointed out that Bitcoin reserves and French currency operations should all be verified, and snapshots are far less effective than continued reserves.

Historically, many exchanges have made reserve certificates, and it seems that the bankruptcy of Mt Gox has intensified the process. Interestingly, historical reserves have proven to be largely broken promises. Some exchanges have completely removed their previous reserve certificates, and some have promised to provide ongoing reserve certificates.

June 2011: Mark Karpeles used the famous 424,242 BTC transactions to generate a rough proof of solvency.

February 2014: Coinkite released a certificate of reserve audit that has been removed.

February 2014: After the bankruptcy of Mentougou, executives at Coinbase, Kraken, Bitstamp, BTC China, Blockchain.info and Circle jointly issued a joint statement promising for auditing and transparency. Only Kraken and Bitstamp proved the reserve and no one continued to prove it.

February 2014: Coinbase asks Andreas Antonopoulos to review their storage, although this is not a formal review. He then deleted the content in his blog.

March 2014: Bitstamp released an external proof of its solvency and had to create the largest transfer transaction at the time.

March 2014: Kraken uses the Merkel method to prove reserves, claiming that they "expect to continue regular audits," but not.

April 2014: Coinfloor, the UK exchange, publishes its first report on solvency. Unlike other existing Bitcoin exchanges, they will consistently publish a report every month, and last month they released the 60th report, far more than the sum of all other exchanges.

August 2014: Stefan Thomas announces that his reserve audit certificate for OKCoin has been successfully completed. However, in a now-removed reddit post, the outgoing OKCoin CTO subsequently claimed that OKCoin misled Thomas and forged some of the audit content. An article by CCN entitled "OKCoin passed the reserve audit certificate" was subsequently deleted.

I can confirm that OKCoin deleted some accounts (used by OKCoin robots) in August 2014 to pass the reserve audit. In fact, these robots trade in the case of partial reserves. Stefan Thomas was deceived during the audit. This is an unavoidable shortcoming for the reserve audit certificate.

August 2014: Huobi publishes a reserve audit certificate authorized by Stefan Thomas.

June 2015: Bitfinex releases a press release claiming that using Bitgo's multi-tap software, they will abandon their integrated account model and store the user's currency in a quarantine account so depositors can verify their deposits in real time. In August 2016, Bitfinex was attacked by 119,000 BTCs, and they abandoned the method of isolating multiple tokens. Bitfinex subsequently announced the cold wallet addresses of BTC, EOS and ETH for public review.

November 2018: Tether issued a proof of class reserve, and their partner banks Deltec Bank and Trust Limited proved their cash balance. Although the skeptics are not very satisfied, the published figures match the number of USDTs in circulation.

The coincidences have emerged: exchanges and depository institutions often only issue proof of reserves when they are under extreme coercion. A series of panic in 2014 was triggered by the bankruptcy of Mt. Gox. Although the exchanges claimed that the reserve certificate would be a continuous and routine job, no one except Coinfloor fulfilled this promise.

Maybe things are changing. New depository institutions such as Fidelity Digital Assets, Square Crypto, Bakkt and ErisX are entering the market, and some of them have announced plans to be more accountable to Bitcoin users. As regulations become more complex, one day they may abandon the encrypted audit of Bitcoin banks. Now, QuadrigaCX has been exposed not to accidental key loss, but the actual situation is their bankruptcy and potential fraud, 2019 may be the appropriate time for some exchanges to re-examine their reserve certification agreement, otherwise, their cake May be taken away by a group of new people.

in conclusion

Bitcoin is a technology related to institutions, a country without an army. Perhaps we should do something realistic and not forcefully bring it to an unsuitable field. Yes, a large number of custodians and banks have emerged, many of whom are drooling about user deposits, and more than one billion dollars in these honeypots have been stolen or misappropriated.

In the real world, how to change this situation to adapt to the nature of Bitcoin? Although there is a voice of "do not master your currency without mastering the private key," Bitcoin banks still exist, and the convenience under the trade-off is compelling. Let's admit one thing: as long as they are providing valuable services, the company will be running, let us focus on getting them to the reliability of Bitcoin.

Ten years have passed and Bitcoin has entered adolescence. Maybe just look at the essence of it: a promising tool for a small number of application scenarios, it can be more calm in its own field. By introducing Bitcoin's natural transparency into a range of institutions responsible for Bitcoin, we can fundamentally improve the current state of the Bitcoin deposit industry.

Opposed voice

1. Bitcoin banks are essentially incompatible with Bitcoin

There is a nihilistic view in Bitcoin that strongly denies the importance of exchanges and custodians as if they do not exist. In my opinion, this is usually a nostalgic mood for the 2010-2012 era, because the network at the time was very flat and had no grade. Of course, you can't suppress the free development of business and business, and smart entrepreneurs decide to provide valuable services to currency holders, such as exchanges, custody and banking services.

Most authorities believe that this is a dark irony, and I think it is a completely natural evolution. Banks make sense throughout the network and we must be with them. It's undeniable that running a node to verify a transaction is really important, but in fact, some nodes are more important than others, especially switching nodes, which make blockchain browsers, blockchain APIs, business services, and large lightnings one day. Routing is possible. There is nothing wrong with this, and Bitcoin has no compromises.

Although "do not master your private key, you don't know your bitcoin" is the trend, and it is absolutely correct, but ignores one point: How do we view those who have decided to give up the private key in exchange for the bank's debt? Do we laugh at them because they don't keep private keys (which are still unrealistic for most ordinary people)? Or do we sympathize with them and try to improve their situation by letting the exchange take responsibility? I highly recommend choosing the latter.

2. Why are people still beginning to prove their reserves despite their reluctance?

The cryptocurrency industry has an anomalous feature that can be called a transparency paradox. In short, the more transparent you are, the more attack points you expose, the more likely your critics will destroy you. Therefore, openness and transparency are disadvantageous. Since the industry has so far been only slightly regulated, most successful projects are very opaque in their operations. For an established project, there is no annual report like 10-K, and there is no prospectus like S1 for the newly issued currency.

The same is true for bitcoin depository institutions, at least in the United States, where they are subject to vague policies because there is no specific area of ​​regulation. In this context, it is advantageous to disclose their operations as little as possible. In addition, the reserve proves to be costly. In the past three years, the exchange has only differentiated from competitors in terms of liquidity and the number of coins, and has not pursued its advantage in credibility.

I believe that letting the exchange begin to prove that its reserves may have several fuses:

The growth of self-regulatory organizations (SROs). Without any new legislation or active regulatory agencies, self-regulatory organizations may play a bigger role in the United States and other developed countries, and Japan is already ahead. Self-regulatory organizations need to propose to their national governments to set standards for exchanges, so that members of the organization can prove that solvency becomes a simple matter.

The chain reaction of QuadrigaCX. The full details of the scandal have not yet been announced, but it is increasingly impossible to lose the private key. Forensic evidence shows that this is a deliberate event, and the exchange has been operating with partial reserves for several years. This kind of deception is unprecedented in the bitcoin world, and the door-to-head is hacked, rather than deliberately stealing funds from depositors.

The division of the grey/black market and the compliance exchange. The split is coming, a series of complex, embraced exchanges will emerge and will be completely broken with unregulated exchanges. This group of newcomers tries to reflect their own differentiation, not based on the number of tokens on the shelves, but in terms of credibility and reliability. The difference is naturally reflected in the audit that introduces the reserve certificate.

3. Part of the reserve will completely destroy the value proposition of Bitcoin

There is a common misconception that Bitcoin banks will partially damage the reliability of Bitcoin by taking part of their reserves. What is certain is that the period when the bank takes part of the reserve will generate inflation. The QuadrigaCX is like this, they don't have enough reserves, and if you owe Bitcoin to Bitcoin assets, it means they secretly increase the supply of Bitcoin.

However, some of the hidden reserves are unsustainable because they are often found, as experienced by exchanges such as Mentougou and Quadriga. When this happens, the bitcoin broad money supply (or "M2") shrinks as fraudulent behaviors occur, and those oweds become non-convertible. Part of the reserve is equivalent to adding leverage, while deception is de-leveraging. Therefore, only in the period when part of the reserve is actually running, there will be hidden inflation.

Large Bitcoin banks such as Coinbase, Bitfinex, etc. have a strong incentive to guarantee their solvency because they want to protect their reputation, and if they do fraud, executives will face jail. As the industry matures and more and more regulated banks, most of the managed funds will come together with the most responsible banks.

4. Part of the reserve is essentially evil.

Compared with positivism, this is more like a philosophical position. I just believe that part of the reserve bank of Bitcoin is inevitable. Just because it is inevitable, we should promote it as responsible and transparent as possible. I didn't think that part of the reserve for Bitcoin Bank was not good because of any problems with the reserve itself, but because it distorted the solvency of the exchange. 100% of the reserve can be exchanged at any time, and some of the reserves will be unpaid.

If I lend Bitcoin to a friend for a month, I add M2 for Bitcoin and I created credit. Genesis, BlockFi, and Unchained Capital all do this and are bigger. Institutional bulk brokers are about to emerge, and they are the same concept, but larger. When banks run part of their reserves, they are doing the same. They create credit by lending a portion of user deposits and make money by the difference in interest rates between the loan interest rate and the deposit interest rate. Therefore, in order for some reserve skeptics to know what they are, they must oppose all lending activities related to Bitcoin. I have actually seen this view, but it looks very harsh and unrealistic. Loans and lending are very clear needs.

Just as I treat the custodians, I have a similar attitude towards some of the reserves: they are inevitable, so we might as well make them as transparent as possible. I recommend giving Bitcoin banks the same treatment as Bitcoin: the free market. Now that we have a regulated market, everyone is naive to think that this is fair, but fraud still occurs on a regular basis, shocking and plaguing people. Why not provide a free market and allow different reserve ratios to exist more transparently?

5. Effectively auditing Bitcoin banks is impossible

One of the toughest critics of the Bitcoin reserve currency model is Eric Voskuil. In an article on the Libbitcoin wiki, he reviewed the economist Saifedean Ammous's view that bitcoin is a sound reserve for commercial banks or central banks, just as it used to be gold.

Eric dismissed the view that the bitcoin deposit certificate was credible and pointed out:

The ratio of issued Bitcoin debt to BTC reserves cannot be effectively audited.

Eric’s criticism seems to depend on some beliefs:

Commercial banks will be supplemented by the state. In fact, banks are an extension of the country.

Proof of reserves will never provide sufficient protection for depositors.

The maintenance of the deposit reserve ratio must rely on trust and therefore cannot be enforced.

These depository institutions will reoccupy the entire bitcoin market through restructuring, which will offset each other's debts.

I can't explain this in detail here, and frankly, I only disagreed with Eric on a few key issues. I believe that the reserve certificate can provide the depositor with a guarantee of solvency if it is the responsibility of a reputable auditor. I also don't believe that the government will immediately control the money supply in the bitcoin deposit market. Commercial banks are independent and will continue to remain independent in the illegal Sith countries.

Let us review it. To believe that the Bitcoin banking system can get rid of the problem of the gold reserve system, you must believe that Bitcoin has some advantages over reserve gold as compared to gold. Specifically:
Bitcoin can be audited by design. What individuals do when they run a full node is not only to continuously review the supply, to ensure they meet the rules, but they also review the entire historical transaction sequence to ensure that each pen is legal and conforms to the rules.

Auditing bitcoin's M1 is cheap, and running nodes cost only a few dollars a month. In contrast, gold verification is expensive, and XRF spectrometers are expensive and difficult to operate. A fully reliable gold supply chain is very expensive, with only a few in the world, and London is by far the largest. In practice, in a private gold market, the cost of verifying a piece of gold is very high, so there is already a whole reliable supply chain, so gold is only circulated in a closed environment and does not require repeated verification. If you are curious about this, please read the LBMA article shipping rules. Currently, London holds $300 million worth of gold, and the central bank only keeps a lot of gold, not mobile gold.

Although it is impossible for gold to assess the size of its M2, it seems at least reasonable to assess the size of Bitcoin M2. M2 refers to the value of Bitcoin including the credit economy. As most exchanges are doing, if their issued debts can be converted into bitcoins, we have the tools to verify if they are lying.

How to expand without changing Bitcoin?

In short, Bitcoin offers much better audit assurance than gold offers, eliminating the need for a reliable supply chain, expensive storage costs or expensive warehousing verification. The cryptographic nature of Bitcoin can be extended with simple proof of reserve, which is why Bitcoin can withstand scrutiny while minimizing trust on the host.

6. Why do you need an intermediary? Why not push everyone to use Bitcoin directly?

I know that my approach may be seen as a compromise, but I think an era of bitcoin without an intermediary is long gone. It is understandable that ordinary people have greedy needs for the custodians and banks, just as we do not keep our stock certificates. Custody of this matter is a challenge for us. The added benefit of banks is that they can earn interest, make people feel at ease, etc., which makes them extremely popular.

According to Coinshares, approximately 2.9 million BTCs are currently held by companies such as Coinbase, Xapo, Greyscale, and Binance. Coin Metrics tells us that in the past 5 years, about 14 million bitcoins have been activated (total circulation of 17.6 million, but a large part has been lost or slept), so Bitcoin has a 20% effective supply. In the hands of third parties.

I don't believe that one day we will collectively awaken and decide to keep it. I think this industry has two directions: one is that the custodians continue to live up to our trust and lose the user's deposits, and the second is that we make them more responsible. Wanting the latter to happen, we need to admit that they are an important part of the Bitcoin economy, good or bad. If the existence of an intermediary means that Bitcoin fails, then the dogmatist should give up the project.

And, frankly, even if you don't like the idea of ​​Bitcoin banks, asking them to prove their reserves is not a bad thing. Ordinary financial institutions are strictly regulated because the consequences of failure are very serious. Bitcoin depository institutions need an alternative regulatory system, and we may wish to persuade the exchange to conduct an audit.

7. What if the bitcoin does not work well? Can this be promoted?

The framework I propose applies to any auditable digital asset. This is the difference between gold and virtual currency/commodities: they are inherently auditable, and auditing and verifying gold is extremely cumbersome. Privacy currencies are more challenging, but they can still be audited through viewkeys or selective disclosure.

8. Bitcoin bank loans effectively increased the supply of BTC

If the bank's reserve ratio is low, the bank can create a new currency that far exceeds the deposit amount.

This can also happen if there is an active part of the reserve bitcoin bank. In fact, if you think a little, the reality is that. Bitcoin derivatives exchanges such as Bitmex have far more transactions than Bitcoin spot exchanges. The amount of Bitcoin traded there far exceeds their bitcoin deposits, precisely because Bitmex provides loans to users in the form of margins, which is to create credit.

I don't think there is any mistake in creating credit itself, because it is the most basic part of finance. If we create credit in a transparent way and use our savings to create credit rather than misappropriate others, that is a major improvement in our current system, and I think it is worth pursuing. (blockchain base camp)