W3C home > Mailing lists > Public > public-interledger@w3.org > January 2016

Re: How do bank payments actually work?

From: Evan Schwartz <evan@ripple.com>
Date: Tue, 26 Jan 2016 14:56:02 -0500
Message-ID: <CAONA2jXBExKMHZ9uA_xwPXKx+LdGptMtUVh-mNCWqN4DU19KGA@mail.gmail.com>
To: Melvin Carvalho <melvincarvalho@gmail.com>
Cc: Fabio Barone <holon.earth@gmail.com>, Joseph Potvin <jpotvin@opman.ca>, Web Payments <public-webpayments@w3.org>, Interledger Community Group <public-interledger@w3.org>, tomblomfield@gmail.com
Interledger is a protocol for secure payments across different ledgers or
systems that track accounts and balances. It is designed to be used for
cross-currency and cross-asset payments, including fiat currencies,
cryptocurrencies, and other types of transferrable resources.

Regarding the differences between the values and designs of different
ledgers: in the interledger model it is the "connector" that offers to
trade one asset for another. Connectors will have different reasons and
rates for trading between different assets, but the idea is that as long as
there is someone willing to trade the units of one system for the units of
another, there should be a way to route a payment through them.

One of the key points of ILP is that if each of the ledgers -- no matter
how different their design is -- can support conditional transfers based
upon cryptographic conditions, multi-hop payments can be made risk-free for
the sender and recipient.

Hope that helps. Keep the questions coming!

On Sat, Jan 23, 2016 at 7:23 PM, Melvin Carvalho <melvincarvalho@gmail.com>

> On 23 January 2016 at 19:37, Fabio Barone <holon.earth@gmail.com> wrote:
>> Please bear me with me for my ignorance...
>> ..but I would like to understand this a tad better:
>>    - I have seen similar ideas in complementary currency spaces, the
>>    idea being that value can be exchanged over different currency circles
>>    - It never took off because the underpinning values systems differ
>>    too heavily - there's no way to , bridge a local money X with another Y,
>>    there's too much difference in money design and value systems
>> Now, my questions for these interledger exercices, are you guys talking
>> about interledger,
>>    - but based on the SAME currency as value exchange? Or different ones?
>>    - based on fiat currency, and/or bitcoin?
>>    - no currency at all, "just" sync records?
>>    - generic interledger which would work no matter what resource the
>>    ledger is actually focused on?
> Personally, I would like to be able to cover all 4 cases, in different
> layered workflows, that sit at a layer above the ledger technology.  ie
> loose coupling between the ledger and the ledger communicaton ... that's
> what im working towards
>> Maybe my questions are completely off, if I in fact would have understood
>> things completely wrong.
>> Would welcome a brief clarification, thanks.
>> 2016-01-23 10:04 GMT-05:00 Joseph Potvin <jpotvin@opman.ca>:
>>> Two comments on the published description...
>>> 1. "Banks pay a few pence per transaction, although no bank currently
>>> charges customers for this service."
>>> It's worth noting that this charging structure is suitable for
>>> transactions greater then, say, about 2£. It would require a different
>>> transaction fee structure to handle micropayments.
>>> 2. "Three times a day, VocaLink will send a message to all participant
>>> banks informing them of their position. To “settle” the funds, participant
>>> banks have accounts at the Bank of England. They will either make a single
>>> payment to FPS (if money has flowed out of their bank), or receive a single
>>> payment (if the net transfer of funds is in their favour). This payment at
>>> the Bank of England is just another double-entry in a ledger; the bank’s
>>> settlement account is debited and the FPS account is credited with the same
>>> amount."
>>> I think perhaps this is mis-stating the operation by using words "make a
>>> single payment to FPS" and "receive a single payment". I'm fairly certain
>>> these are accounted for as loans, to which the "Bank rate of interest" is
>>> applied. Please correct me anyone thinks I'm wrong about how this
>>> particular settlement system works. I think that while the mechanics remain
>>> true that it's "just another double-entry in a ledger", these show up in
>>> the books as off-setting loaning and borrowing by the Central Bank, and
>>> they include an interest rate which needs to be taken account of.
>>> http://www.investopedia.com/terms/b/bankrate.asp
>>> For those of you creating test environments, taking account of this
>>> factor complicates the model in two ways. First directly, you would need to
>>> attach some sort of index (the interest rate) to the inter-bank
>>> transactions. Second, you would need to create an policy-motivated actor
>>> (agent) who makes decisions about that index.* [I'll proceed a bit
>>> off-topic here just to illustrate...] This is because this  "the Bank Rate"
>>> which is the benchmark by which a Central Bank motivates increases or
>>> decreases for all interest rates of the given currency zone. The 'fun'
>>> start to happen when a central bank decreases an interest rate in order to
>>> simultaneously incentivize capital investments (its cheaper to borrow) and
>>> decrease forex traders' demand for that currency (better returns on static
>>> deposits are found elsewhere) in a period of competitive currency
>>> devaluations, all of which artificially stimulates that country's export
>>> market. Well, they've all been doing that for a while, and have run out of
>>> room at the bottom. But they keep going!
>>> http://www.bloombergview.com/quicktake/negative-interest-rates
>>> <http://www.bloombergview.com/quicktake/negative-interest-rates>  *
>>> [Back on-topic...]  Any model of payment must be a simplication of
>>> complex reality, so this is not a critique of the published description as
>>> far as it goes. I just raise a caution that with a title like "How do bank
>>> payments actually work?", this summary of some of the mechanics of the
>>> system inevitably has to leave out much of how bank payments actually work.
>>> I've c.c.'d the author, Tom Blomfield, in case he'd like to comment.
>>> Joseph Potvin
>>> Mobile: 819-593-5983
>>> On Wed, Jan 20, 2016 at 6:03 PM, Melvin Carvalho <
>>> melvincarvalho@gmail.com> wrote:
>>>> Interesting post on the inter ledger element of banking.
>>>> https://getmondo.co.uk/blog/2016/01/20/how-do-bank-payments-work/
>>>> Im thinking of simulating this on a testnet for people to play around

Evan Schwartz | Software Architect | Ripple
[image: ripple.com] <http://ripple.com>
Received on Tuesday, 26 January 2016 19:56:51 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 26 January 2016 19:56:51 UTC