[bastawhiz]

How would this be better than what we have now? If the answer is "cryptocurrency increases in value," I can assure you that's a terrible reason.

Cryptocurrency is slower than card networks. It's more expensive than practically any other way of sending money. And as far as scale goes, it would need to become orders of magnitude more efficient to handle even a small percentage of consumer transactions. After all, what's the point of a cryptocurrency if the only people who can host the full blockchain (or even acquire the full blockchain) are large banks and the government? I.e., where do you even get an internet connection that can accept, in near realtime, a full record of every monetary transaction performed with such a currency? Unlike card networks, every member of the network needs to process every transaction eventually.

I don't think cryptocurrency is at the point where the scalability concerns can be addressed to be used as legal tender for an economy as large as the US.

If this is meant to replace bonds or other government issued securities, what problem is it solving? I can't think of one.

[paulmd]

> Cryptocurrency is slower than card networks. It's more expensive than practically any other way of sending money. And as far as scale goes, it would need to become orders of magnitude more efficient to handle even a small percentage of consumer transactions.

Cryptocurrency with a centralized authority is not subject to these issues. After all, if there is a trusted authority then what you really have is a database with some API layered on top. You don't need miners, you don't need a blockchain, etc. It's no more difficult to scale a currency like this than it is for Visa. Canada was looking at exactly this idea about 5 years ago, with the MintChip project.

http://business.financialpost.com/news/fp-street/canadian-mi...

Think of this as a bank account that you can interact with in a programmatic/scriptable fashion using a private key. Which is really a lot of what people find desirable about cryptocurrency.

The deflationary monetary policy ponzi-schemes, the waste of energy and data, etc can all go. And in turn, the government gets to eventually eliminate the cash economy and make sure that all of that gets taxed (this is the dark side of cryptocurrency - since everyone gets to see all transactions, it's quite easy to trace the flows of money, and it's only anonymous until you try to do something outside the network, like cash out or exchange it for real-world goods).

[skybrian]

Yes, that's not really cryptocurrency anymore. It's a federal bank that has retail customers.

Maybe not a bad idea, though. It might make some government payments (such as Social Security) more efficient?

[soperj]

>that's not really cryptocurrency anymore

Why not?

[nostrademons]

The "crypto" in "cryptocurrency" refers to the use of encryption primitives and digital signatures to ensure the authenticity of one's balance without needing to trust a central authority.

If you're trusting a central authority anyway, then all this is unnecessary, and you can replace it all with an API over a database, as paulmd's post suggests. But then you've eliminated the whole reason why people use cryptocurrency, and what differentiates it from Visa.

[otakucode]

What differentiates it from Visa is that there is literally nothing behind Visas numbers. They could be pure fiction right now and no one but Visa would know. Also, Visa takes a cut. As more and more of the whole economy flows through payment processors, they have become de facto taxing bodies. Their fees drain the economy, and can be used to manipulate the economy to Visas whim. A government-backed cryptocurrency would be provably legitimate, and would cut Visa out of the loop.

[nostrademons]

Up-thread, the proposal is to fix the scalability & energy usage issues by having a trusted authority such as the U.S. government run a database with a payment API on top.

Under such a system - how do you know that there is anything behind the government's numbers? They could be pure fiction right now and no one but the government would know. Also, the government takes a cut. As the whole economy flows through their currency, they become a de jure taxing body. Their taxes drain the economy, and can be used to manipulate the economy to the government's whim.

It may shock you to learn that not everybody trusts the U.S. government. The growth of cryptocurrencies worldwide is largely fueled by the fact that the set of people who don't trust the U.S. government is larger than the set of people who do - and in fact, they've seen their biggest up-take in societies like Zimbabwe and Venezuela where trust in civil monetary authority has completely collapsed.

[QML]

The FED can still be a central issuer, and a digital USD can thus still be considered a cryptocurrency.

[mrb]

«Cryptocurrency is slower than card networks»

Not inherently. If you compare apples to apples, accepting a credit card transaction is equivalent to accepting a zero-confirmation cryptocurrency transaction, which is just as fast as a CC charge (instantaneous.) And it is technically safer for a merchant to accept a 0-conf crypto tx because a CC charge is trivial to reverse (via a fraudulent chargeback) while a tx in the mempool of thousands of nodes is typically (not always!) hard to make disappear.

«It's more expensive than practically any other way of sending money»

Far from true. Average remittance fees are around $7 per $100 sent. Meanwhile Ethereum has fees typically under $0.50 per tx: https://bitinfocharts.com/comparison/ethereum-transactionfee... Of course there are periods of fee surges (over $5 per tx!), but it's not common.

«a full record of every monetary transaction performed with such a currency»

That's technically not needed. There have been proposals to implement what we call UTXO commitment sets, which is basically a way to revamp a blockchain so that it can discard old transactions and just keep track of current balances. (This is different, but roughly the same principle as "pruning.")

[bastawhiz]

> accepting a credit card transaction is equivalent to accepting a zero-confirmation cryptocurrency transaction, which is just as fast as a CC charge (instantaneous.)

Except my charge is guaranteed to happen (or be rejected) within a reasonable amount of time. If you scale up anything to the size of, say, Visa, the transaction times are going to grow to be unbounded. I know my credit card charges are going to settle in about a day. There's no guarantee when the miners will work through the backlog of transactions and actually give me my money in any time frame.

If you're a small business, this matters a LOT.

> Average remittance fees are around $7 per $100 sent. Meanwhile Ethereum has fees typically under $0.50 per tx

My bank charged me $10 to send a 50k wire transfer last week. What service, exactly, is more expensive than a 7% fee?

Ethereum, meanwhile, swings in price violently enough that in the time it takes for an ACH transfer to complete, the value could have shifted enough to negate the entire savings on fees. What benefit does low fees have on a currency if the currency is worth 5-10% more or less day to day?

There's no reason to assume a government issued cryptocurrency will be any less volatile than cryptocurrency.

> There have been proposals to implement what we call UTXO commitment sets, which is basically a way to revamp a blockchain so that it can discard old transactions and just keep track of current balances.

And here we are, talking about creating a new legal tender for the second largest economy on the planet. As I said in my original post, it doesn't matter if it's on its way. Speculative fixes for a real problem don't make the problem go away. Major cryptocurrencies currently don't do it, so we shouldn't talk about them as if they do.

Even still, the number of balances will grow over time. Unlike a real bank, you can't just close an account when someone dies. Anyone can create as many balances as they want. And losing your private key means there's a permanent record of the money you lost.

[mrb]

«If you're a small business, this matters a LOT.»

With a CC it takes 1-3 business day for the merchant to receive spendable funds. With a crypto tx it takes minutes (the time for 1 confirmation). This is a huge advantage for cryptocurrencies.

«What service, exactly, is more expensive than a 7% fee?»

It's the worldwide average of all remittance companies. See http://remittanceprices.worldbank.org/en which claim 7.13%.

«Major cryptocurrencies currently don't do it, so we shouldn't talk about them as if they do.»

You spoke as if the problem was unfixable. I am clarifying it can be fixed. The reason most cryptocurrencies do not implement UTXO commitment sets is simply that having to store all txs is not (yet) an intolerable pain point.

[splintercell]

> Cryptocurrency is slower than card networks.

Two cryptocurrencies come to mind:

a) Nano (formerly RaiBlocks) which achieves 7000 tps, compared to Visa's capacity of 25k-65k tps (but VISA rarely spikes over 4000 tps), its wallets and network is already live

b) Solana (still in dev stages), which claims to be able to achieve 710k tps without network partitioning. This actually knock centralized systems out of the water. Of course, the centralized systems can always mimic the behavior and achieve similar throughput.

But the main point is, because a lot of money is in blockchain, a lot of it is being spent on finding ways to make it faster (which is why, innovation for faster throughput is happening in the blockchain field).

[vinniejames]

I just had to pay $8 to convert USD to another currency. The fee is less than $1, and often less than $0.10 to make an exchange like this via cryptocurrency.

In order to move money out of my own bank account, digitally, into another account, it is impossible to do instantly. Many banks charge a substantial fee, even for a 3 day turnaround[1].

Looking only at card networks, while ignoring the greater banking system is an incomplete view of the potential here.

1. https://www.nerdwallet.com/blog/banking/ach-transfers-costs-... M

[vkou]

> I just had to pay $8 to convert USD to another currency.

That's the cost of the spread.

If you tried to convert $10000 BTC to USD today, you would pay anywhere between -$200 and $200 (Or ????$, if the market decides to go crazy), depending on the exchange rate when your transaction would clear. With a forex transaction through a bank, the bank fixes the exchange rate (And charges you money for it.)

[vinniejames]

That was just the “fee” without even including the spread, which was much more, often +20% depending on the exchange.

Converting $10k BTC to USD would not cost $200 in transaction fees. If that’s how much you’re paying, find another exchange as you’re getting ripped off.

Converting BTC to another crypto currency would barely incur any cost at all.

[mavdi]

You've only scratched the surface of cryptocurrencies and come here to make bold, completely untrue statements about their properties.

Having a central actor, like a government, would solve a lot of scaling problem, as well as transaction history issues and also cost associated with making transactions.

[foepys]

> Having a central actor, like a government, would solve a lot of scaling problem, as well as transaction history issues and also cost associated with making transactions.

Then why Bitcoin, Ethereum, etc. instead of USD? Why the need for a crypto currency if everything is already in place? In place with laws and regulations.

[Toine]

Exactly this.

The first line of the Bitcoin paper is : "A purely peer-to-peer version of electronic cash would allow online payments to be sent directly from one party to another without going through a financial institution".

The whole point of cryptocurrencies was to get rid of the third party in the first place...

[gwbas1c]

> If this is meant to replace bonds or other government issued securities, what problem is it solving? I can't think of one.

It's a legal pump and dump scheme!

[snissn]

> Cryptocurrency is slower than card networks.

How long does it take you to set up a merchant account? With cryptocurrency, it's practically instant.

[thedudemabry]

While a valid point, setting up a merchant account also imbues the merchant with some trust properties that are not available with a cryptocurrency wallet. Namely, the ability to procedural and legal recourse if something goes wrong with any new transaction.

[ac29]

24 hours or so. Its a non-issue for business operations.

[snissn]

How long does it take to set up a business?

[ojr]

there are new blockchain tech releasing this year that will scale to thousands of transaction per second and provides free transactions for end users. Companies like Consensys provides infrastructure so you don't have to run a full node but just connect to a full node with a light client. You need USDC to escape from volatility. The crypto market needs a stable currency with the assurance the peg doesn't break. Digital currency solves the problem with inbound/outbound wire fees in which cost me a few hundred last year...

[rbreve]

You live in a bubble, your argument is not valid for millions of people that don’t have bank accounts, have you seen western union fees? Africa and Latin America will embrace cryptocurrencies because it will be faster and cheaper and will not require people to open a bank account, only a cell phone with data which most people already have

[dang]

Probably this got downvoted because of the "you live in a bubble" part, which crosses into personal attack. Could you please (re-)read https://news.ycombinator.com/newsguidelines.html and edit those bits out of what you post here? This comment would be just fine without it.

[rbreve]

Oh sorry, I wasn't attacking.

[toomuchtodo]

Payments in developing countries is already fast through the use of mobile phones, no cryptocurrency overhead required.

[rbreve]

Fast but with very expensive fees, and the problem is also corruption, privacy, and abuse by banks and financial institution. With the blockchain beeing decentralized you wont get ripped off and you won't get tracked either (in a way) Also, imagine if Trump decides to ban or tax transfers from USA to LatinAmerica/Africa, then crypto will be the solution.

[QML]

Which apps or companies are these? I have not heard of widespread use of mobile payments in developing countries besides China.

[toomuchtodo]

Africa: https://www.pymnts.com/news/mobile-payments/2016/mobile-paym...

Latin America: http://www.idgconnect.com/abstract/10231/latin-america-warms...

[skybrian]

If you know more, could you share? I haven't read anything about cryptocurrency being successful in Africa or Latin America. (Versus, say, M-Pesa in Kenya.)

[rbreve]

look it up, there are a lot of articles about it

[pshin45]

To add to this point, M-Pesa has been a huge success in Kenya, but has had trouble expanding to other countries because of some inherent limitations of M-Pesa's model.

A cryptocurrency could enable the next evolution of mobile payment services like M-Pesa. I touched on this in a past blog post:

https://medium.com/@petershin45/hate-bitcoin-this-might-chan... (You can scroll down to the section titled "How Bitcoin can help the poor and unbanked")

[powera]

"Cryptocurrency is slower than card networks." - not really, on a per-transaction basis. If you buy something at a store with a credit card, it takes weeks to get the money. With Bitcoin, you can spend the money in an hour. For small transactions, there's generally no need to wait past the point that the transaction is broadcast; and for online transactions it's not inconvenient to wait an hour before providing the product.

(of course, overall throughput is still atrocious)

[icehawk219]

Waiting that long before the merchant gets their money is a feature, not a bug. It forces the merchant to float a balance that can be used to issue refunds and charge backs back to the consumer without the bank, card network, or consumer taking the hit.