Hacker News new | past | comments | ask | show | jobs | submit login

It's interesting to think about how the role of central banks might change in the age of crypto currencies.

Governments can print Euros and Dollars. But not Bitcoin and Ethereum. Popular theory is that governments have a good grip on currency usage because they can decide which currencies they accept for tax payments. And because they can force merchants to accept certain currencies. It will be interesting to see if this grip holds up.




Since there is no fractional reserve and the blockchain ensures that you have the equivalent of fully correspondent banking with each node in the network there is no need for a central bank on the other hand the blockchain itself can be viewed as is the central bank with the large mining pools acting as its board of directors.

That said since no blockchain currently offers a credit system it’s not really possible to test this out fully since there is no true equivalency.

A better question would be not if we need central banks but rather should the economic system be based on credit and interest or not.


Bitcoin exchanges certainly can "print more bitcoin" in the same sense as fractional reserve banking - the exchange doesn't have to have a stash of coins that exactly corresponds to the sum of bitcoin-denominated customer accounts. Mtgox blew up by being a fractional reserve "bank" destroyed by a "bank run".

Heck, bitcoin exchanges can even print ""dollars"" in the form of USDT.

> should the economic system be based on credit and interest or not

This is almost unavoidable - remember that every not yet paid invoice is a form of credit.


Ofc they can’t which is why I said that the crypto needs to support it for it to work properly otherwise the only thing you are left with is some odd options based scheme or a MtGox type ponzy racket.


Not really.

With most cryptos, there is really a strict pre-established monetary policy baked into the algorithm with really no flexibility at all.

"A better question would be not if we need central banks but rather should the economic system be based on credit and interest or not."

I think a better way to say it: "Do we need monetary policy or not" - and I think the answer is 'yes' - so whether it's crypto or not ... there will have to be some kind of group/committee making those rules, and that's de-facto the central bank.


There is a cosensus protocol that can be used to propose and approve a change the crypto, this is no more difficult than changing a monetary policy it’s just different. The code changes isn’t the hard part the policy itself is.


"hard part the policy itself is" which is ultimately the point of a 'central bank' really.


Except they keep scamming...


There's no fundamental reason why Bitcoin shouldn't also have fractional reserve. I mean, cash also has a fixed supply! (from the Bank's POV) It's not like Chase Bank (or whatever non-central bank) can just print more dollars.

If you just sit on your Bitcoin it collects 0% Bitcoin-denominated interest. Maybe you'd prefer to make a loan to a bank and have them pay you interest. For the term of the loan that money isn't sitting in an address you control, it's been given to the bank and they're consequently free to lend it out to other people. Viola, Bitcoin have been created, just not on-chain.

EDIT: I was referring to physical dollars. As in, dollar bills and actual coins. I'm aware that Chase creates money, but it cannot literally print money.


>It's not like Chase Bank (or whatever non-central bank) can just print more dollars.

Yes, it can, that's what fractional reserve means. Banks create (most of) the money they lend. This is regulated by the central banks but carried out via ordinary retail banking. When you swipe your credit card, buy a car, etc. money is being created out of thin air. The universe remains balanced because it corresponds to your obligation to repay; as you do, the money ceases to exist.


I don't think you understand what I was trying to say. I'm well aware that they create money, but they're literally not printing dollars. They do not cause M0 to increase.

Nobody but the Treasury can cause physical dollars to be created, but that doesn't prevent fractional reserve from occurring.

So, anybody who claims Bitcoin doesn't support fraction reserve based solely on the fact that banks can't create new Bitcoin is incorrect.


Okay.

I'm Chase Bank.

How do I issue 600 BTC given I have 20 BTC in my account.

There is zero allowance in the protocol for this to happen.


Easy. As an exchange almost all your volume is “off chain” transactions between two traders. And since you aren’t audited or beholden to any financial regulation there is nothing stopping you from having one side of the trade be your own bot. All that bot does is sell newly created fake bitcoin and keeps the dirty USD fiat for the exchange owner. Again since the system isn’t audited, the exchange can have massively more “bitcoin” in its database than it actually has on the blockchain. The system (Er, scam) can work just fine until a bunch of folks start withdrawing bitcoin.

Since virtually all of the exchanges are unaudited, sketchy “businesses” I wouldn’t be surprised if they all run this way. It might come to pass the almost all of the exchange volume consists of trading fake bitcoin that is backed by almost no real bitcoin.


sure but these "off chain" transactions are not really Bitcoin, but the exchange's promises of Bitcoin (or whatever cryptocurrency under discussion). As long as custommers succeed in end-to-end conversion the distinction is moot, until they fail to succeed in this end-to-end conversion.

So still a commercial bank can not fractional reserve bank BTC, only fractional reserve bank promises for BTC. For a while people may fall for such a scheme but players learn...


Say the reserve requirement is 10%. Wikipedia states [1]:

> The minimum reserve is generally determined by the central bank to be no less than a specified percentage of the amount of deposit liabilities the commercial bank owes to its customers.

If someone gives Chase $100, Chase is allowed to loan out up to $90 of it, and must keep the remaining $10 in reserve.

1: https://en.wikipedia.org/wiki/Reserve_requirement

The protocol allows someone with 100BTC to send someone else 90BTC.


What is created is an IOU. A bank says that hay, I will pay 1 usd to you. And you can transfer that IOU to the coffee shop. Now, there is literally nothing stopping the "bank"[1] to make a bitcoin IOU and you to transfer that to the coffee shop owner. In that case you have increased the monetary supply of "bitcoins", as it is irrelevant for the coffee shop owner if he is receiving a bitcoin or an IOU of one bitcoin from a credible institution. There is no difference to the case where it is irrelevant if you get a dollar bill or a IOU from a bank. Both are called dollars for a very good reason, and if you just swap dollar to bitcoin, both can be called bitcoin.

[1] Well, except current regulators, but they obviously do not exist in bitcoin economy...


This is mostly correct but you have one thing backwards. When you pay the coffee shop (with a debit card) you're not giving them an IOU, you're giving them the same kind of dollars which you originally deposited into the bank.

The system does include IOUs but those don't move around. When you give your hard cash to the bank it creates an IOU, that's what the balance you see when you login is. Your bank then lends out the hard cash to other people.

When you pay with your credit card the bank is still giving the merchant real money, the merchant would receive actual on-chain Bitcoin, but you're simultaneously creating a Bitcoin IOU which you promise to pay the bank.


I am quite confident I do not have this backwards. (I may be bad at communicating this, though...)

When I pay coffee shop with a debit cards, exactly what happens is that the bank just tells me that hey, we owe you now less, and we owe the coffee shop a bit more. At that point there is nothing else happening. I am definitely not giving any "real" dollars because I do not have them left, but I have lent them to the bank so that I have only the IOU from the bank left.

What you see in your bank account is at the same time real money (as usually talked about), but really really nothing but a bank's promise to pay you some later day real real dollars.

And because it is so convenient, most people are perfectly happy with moving around these IOUs, the coffee shop owner can then walk to the bank and request that the bank actually pays the dollars (or bitcoins) that the bank owes, but it is quite rare that happens - and that is the whole point of fractional reserve banking.


Okay, I'm not a banker so there's a good chance I'm the one who's wrong, and I bet the answer changes depending on the specific transaction.

> What you see in your bank account is at the same time real money (as usually talked about), but really really nothing but a bank's promise to pay you some later day real real dollars

We completely agree on this part.

> When I pay coffee shop with a debit cards, exactly what happens is that the bank just tells me that hey, we owe you now less, and we owe the coffee shop a bit more

I don't think this is true. Or, I don't think this is always true.

I can see it being true as a special case: If both you and the merchant use the same bank then sure, probably all that happens is two different IOU tallies are updated.

But I'm currently visiting Turkey, and when I swipe my card Chase definitely does not "owe the coffee shop a bit more", Chase has absolutely no relationship with the Turkish coffee shop I'm sitting in.

And maybe a temporary IOU is created between Chase and whatever Turkish bank the coffee shop is using, but I expect that IOU to be quickly resolved in a few days, when settlement happens. Real money is moved from your bank to the acquiring bank, and an associated IOU shows up in the bank account of the merchant.


> But I'm currently visiting Turkey, and when I swipe my card Chase definitely does not "owe the coffee shop a bit more", Chase has absolutely no relationship with the Turkish coffee shop I'm sitting in.

Admittably I made some shortcuts. Obviously chase does not likely have a relationship with that Turkish coffee shop, but it does have a relationship (which may be complex, involving visa, mastercard etc.) with the bank of that coffee shop.

So practically Chase says that now that you paid your coffee in turkey, we owe you a bit less, but at the same time we as a bank have agreed to owe that amount of dollars to the Turkish bank, who the on owes to the coffeeshop owner some turkish lira. (and someone makes also the currency conversion there). The banks have between themselves some credit/limit arrangement that they do not need to actually start moving "real" central bank money [1] every time I buy a coffee. At the end of the day, most of the time the customer payments should not have a systematic drift (if there were, it would be called bank run), so banks just work within these credit limits.

It may sound like turtles all the way down, but that is exactly what the modern financial system is. Credit all the way down. There is very little else than credit in the system.

[1 ]which the turkish bank does not even have access but that is one more complexity layer I ignore here...


Okay, I've done some more research and I'm much less sure than I previously was. I mean, I'm now pretty sure you're right: at the time of the transaction the only thing which is transferred is an IOU.

It still seems to me, however, that at the end of the day there's a net settlement process which happens at the central bank and moves real central bank money between the banks as a result of your transaction. (I'm can't figure out what the international equivalent is but assume there is one)

Do you have any links I could read which say otherwise?

> At the end of the day, most of the time the customer payments should not have a systematic drift (if there were, it would be called bank run)

I don't think this would constitute a run, I think it just results in the banks shipping central bank money to each other to settle the difference.

I'm finding it surprisingly difficult to Google this, would love any pointers you can give.


> It still seems to me, however, that at the end of the day there's a net settlement process which happens at the central bank

There is a daily settlement process via central bank, but you do not need to settle daily. Imagine only transaction today was me sending 10 bucks to your bank account. Your bank would likely say that hey, pay that 10 bucks (plus interest) some day later when there is a bigger payment coming. And by tomorrow someone from your bank paid to my bank 10 bucks, so then there was no need to do transfers. Banks do run credit lines against each other. Obviously, if I were to send one hundred million bucks, your bank would cough and say that okay, let's see when the money is at our central bank account....

> I don't think this would constitute a run,

If there is a long term systematic and significant outflow of money from a bank, that is definitely a bank run. Banks can't create central bank money, so if my bank needs to send daily money to your bank and is not receiving anything back, it needs to start liquidating some of the assets.

> I'm finding it surprisingly difficult to Google this, would love any pointers you can give.

Unfortunately I have found very little good information on money, these thoughts are just based on lots of thinking on the fascinating thing called money... Most sources belong to two camps:

1. Basic , naive economics that is saying that money is created when I deposit money and then bank lends it to you so that you have the cash in hand and I have the money in the account so the amount of money is doubled in society

2. The ones who think they are smart and say that hey, that naive theory is false. Actually banks just create money with double accounting wizardy and loans. Or something like that.

The first one fails to capture the technical details of money creation and the second ones fails to capture the economics.


If you want a good overview of how money works overall this PDF from the Bank of England is pretty approachable:

https://www.bankofengland.co.uk/-/media/boe/files/quarterly-...

There are a lot of misconceptions about money around, many based on older ideas of physical or commodity-backed currencies.


I didn’t say it should or shouldn’t I said that currently they do not offer any sort of a built-in credit system.

And if you do fractional reserve off the blockchain then your crypto isn’t currency but rather closer to say what gold was during the gold standard era.

And yes Chase can “print” dollars that’s what the fractional reserve system means for every dollar they store in the central bank they can loan X dollars. When you ask for a mortgage Chase essentially creates most of those dollars out of thin air.


> currently they do not offer any sort of a built-in credit system.

Indeed, but what does credit mean in a trustless system?


Well, every crypto currency has it's own characteristics. And therefore different USPs. I don't expect currencies that involve blockchains to be the dominant payment channels of the future. For payments, I expect simpler constructs. Lightning Network already handles most transactions without blockchain interaction. And Iota is a completely blockchainless currency that got traction already.

I think there will be many more approaches before one becomes the 'everyday' crypto currency.


We have been dealing with exactly these issues at the Intercoin project (https://intercoin.org)

Credit is a voluntary thing. Both sides can agree to create credit lines (trustlines) out of thin air without any third party ledger or permission. (Well maybe except Usury laws.)

Now the problem with credit is that you don’t know how solvent the debtor is and many debts they have. Credit agencies have sprung up to try to address this somewhat.

But the whole POINT of value based money is to introduce a third party representation of real assets, and that real world scarcity requires solving the double spend problem.

So the whole thing with every technology is remembering that A paid B, and not forgetting it (eg make collusion really infeasible).

Many blockchains are just append only databases stored on every node. They elect a leader based on PoW or PoS and it’s horribly inefficient. There are far better ways.

But anyway back to credit. Money is a social phenomenon that benefits from a network effect. A casino’s chips are worthless as a medium of exhange outside the context of the casino and the same goes for JPY, EUR etc.

Inside a community currency where everyone began to accept it, you can have the ability to print more money same as a Harvard Facebook has the ability to add features you don’t like but you keep using it. The only difference is that these decisions can be done democratically. So like living in a city whose policies you don’t all agree with.

They can use this money for Basic Income and other things like public infrastructure.

They can peg to an outside currency AND print/dilute it gradually to redistribute wealth in a voluntary way that even anarchco capitalists will accept.

That’s one of the features of Intercoin.


> There are far better ways.

Citation needed since PoS / PoW is the way Bitcoin uses to solve the many generals problem.


Algorand's solution is in fact pure genious, however there is no implementation yet (that I know of, if someone knows of an implementation, please let me know!).


Wht is so great about it specifically?


The fact that proof of work is proven unnecessary to perform sortition/rate limiting the right to propose blocks, even without needing to resort to an interactive protocol, i.e. player-replacability.


So far there does not seem to be any serious threat to any national currency because of cryptocurrency.



At best you can say that people become interested in the same sort of circumstances they're they're interested in conducting transactions in foreign reserve currencies rather than their own.


If you have multiple competing cryptocurrencies that seems pretty equivalent to a system of free banking[1] which, well, worked pretty well in Scotland before the Bank of England took over.

[1]https://en.wikipedia.org/wiki/Free_banking


Bitcoin so far doesn't scale. Not only in terms of transaction fees and processing times... but also at the most basic of levels.

What will happen when the blockchain doesn't fit in a median harddrive?

The decentralization narrative will go away. Only a few people will have enough storage for the blockchain and you will go back to having to trust someone.


> The decentralization narrative will go away. Only a few people will have enough storage for the blockchain and you will go back to having to trust someone.

With a pruning client you only need to store a small subset of the blockchain and there are no major changes in regards to trust. Your client still verifies every single transaction, but it removes data from the chain that it won't need anymore in the future (e.g., an old transaction that already has been spent).

With a SPV client your client only stores the chain headers and then only requests the required data from a server. SPV clients need a small amount of resources and can run on mobile phones. At the same time most of the important functionality (transaction signing, transaction verification, ...) happens still locally and the level of trust that you need to put into the servers is quite small.


Even beyond simple pruning, there are techniques that currently don't seem to be in use yet and should make chain-size a non-issue for the near future. If written about it in another comment here before[0].

[0]: https://news.ycombinator.com/item?id=16929524


That depends if the Bitcoin blockchain size is growing at a faster rate than the storage available on a median hard drive.

The blockchain is currently at 160~GB, and I believe it will grow around <60GB a year. [1] In 10 years it may be ~760GB, but the median hard drive shouldn't be too far from this.

[1] https://charts.bitcoin.com/chart/blockchain-size


Satoshi's white paper talks about this problem and why it's not an issue. It's only 3 pages.


Governments are never going to adopt a currency controlled by foreign entities.


Really? In many parts of the world (outside of the US, e.g. Georgia), dollar is used/accepted even for taxes.


Yes, but accepting a foreign currency as your defacto currency is IMO insane. US sets monetary policy for the US and the US only. So if US raises interest rates to combat inflation that is occurring in the US. Georgia by default accepts this position regardless of what is actually happening in Georgia. If Georgia has deflation occurring, then this is bad for Georgia.

If you want to be a "Big Boy" country, you have to have your own monetary policy and currency. The problem is, managing your monetary policy and currency is really difficult. Countries that experience hyper-inflation are usually ones where monetary policy isn't done right. The other side of it is, when you do have the right monetary policy and currency controls, it allows to survive and possibly thrive in the financial world.


Georgia is trying hard, but Russia isn't making it easy for them.


What does this have to do with anything Nothing


No country has adopted a foreign currency to make their central bank impotent.


Seems like at least Ecuador did:

> The present currency of Ecuador is the United States dollar.

https://en.wikipedia.org/wiki/Currency_of_Ecuador


"No country has adopted a foreign currency to make their central bank impotent."

There are a number of countries who's currencies are pegged to the US dollar at fixed exchange rates. Notably Hong Kong, as well as gulf states such as the UAE, Saudi Arabia, Qatar, etc.

This means their central banks have very limited scope to control domestic money supply and interest rates via monetary policy.


Several small countries have "dollarised". Ireland for decades ran an "independent" currency that had the same value as the pound and decimalised on the same day.


The US Dollar is an accepted global currency for international trades.

So it kinda does make sense they accept it, since they can ultimately use it.




Consider applying for YC's Spring batch! Applications are open till Feb 11.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: