Bitcoin Forum

So I've been thinking a lot about banking lately, specifically how it could be done within an anonymous, secure, community where fraud is so freaking easy. All-in-all, I don't think it would be that difficult to set up a BitCoin bank. The problem I see is incentive to continue running it.

Traditional banks make money mostly in three ways:

- Loans paid back with interest
- Fees associated with accounts
- Other associated 'services' like a safe deposit box, financial consultations, etc

By far, I'd say the primary driver for a banks revenue stream is probably loans. Loans are done through a series of complex financial manipulations where banks are allowed to lend out roughly 9-10 times what they have in the vault at any one time (called 'fractional reserve banking' http://en.wikipedia.org/wiki/Fractional-reserve_banking) and I see that as a major problem with our current banking industry. I don't think it's a good idea for a BitCoin ban so that leaves only two major ways for a bank to earn money.

Obviously, offering things like safe deposit boxes is complex in an anonymous system (though not impossible) as are financial consultations (they can compromise a customers anonymity and the consultant would need to compromise their anonymity to prove they're not a huckster). So, really, that leaves only fees associated with accounts (holding and withdrawing) as a reliable way for a BitCoin bank to make money.  This is certainly doable but definitely (probably?) won't make a banker wealthy (which might not really be the goal).

What challenges do you think need to be overcome before we can see First Bank of BitCoin open its doors?

Rage

I think a bitcoin bank could only loan the coins it has on deposit ie 100% reserve. lending in a deflationary currency is dangerous imo.

http://www.freelakotabank.com/ (http://www.freelakotabank.com/) bills itself as the worlds only non fractional bank. If anyone would know how to run a bitcoin bank it would be them. They only do silver rounds though at the moment.

More than likely they would store bitcoins offline using truecrypt and other tools and would have to hire security experts and submit to third party audits.

This is one of the reasons I like bitcoin :D

I think the main benefit of using a bitcoin bank in the future will be :
 - Cutting on transaction confirmation time when transfers happen inside the bank
 - Cutting on transaction fees, still when transfers happen inside the bank

Also, there's no reason transaction times and costs would be cut only for transactions within the bank. Established banks will trust each other and do transactions with each other without having to trust the individual depositors (possibly even on a ripple network). Other huge advantages to banks like these are securing the bitcoins themselves, offering insurance in case of lost bitcoins due to a flaw in the backup regime, or a virus, and implementing access control to different wallets, which would be useful in a large company.

The beautiful thing is that none of these are particularly expensive. If backup and security is done right, the insurance costs should be very low. Everything else is pretty simple, with low barrier to entry. It's just not the kind of stuff each individual bitcoin owner should have to do.

Loans can be made on time deposits.  The bank offers me a 1 year CD.  Then they can give someone a 1 year loan.  I cannot use my deposit for that time period, and the person being loaned the money cannot as well.  The bank tries to make a profit by charging a higher interest rate on the loans than deposits.  Sometimes they screw up and lose money on individual deals.  But mostly they set it up to be safe enough not to.

The biggest challenge in giving a loan with BTC is the difficulty in establishing you aren't going to scam the bank and having a way for the bank to come after you if you screw them.  Reputation could be used to try to make sure you have a good track record.  However, to account for all the losses from people not paying you back, the interest rate would likely be fairly high.  This means only the most profitable investments can be made with loans (this is a good thing).  You likely won't see loans made for consumer goods.  Loans with collateral could work, but you need a way to enforce that.

anyone willing to work on a bitcoin bank please reply.  basically we'd be setting up a virtual proxy for community credit.  the way banking works right now is the Fed creates reserves via a double bookkeeping entry where by it simultaneously buys a US Treasury bond and from a primary dealer (think JPM, BofA, Citi, Wells, etc) and then they deposit the money on reserve at the Fed.  This reserve did not exist before the this transaction, it is created out of thin air by the Fed.  Primary dealers can then make loans to you and I off of this reserve base depending on the "reserve ratio".  If for example the required reserve ratio is 10% then $100 in reserves would support $1000 in new credit money ($100/.10).  In this way, banks can create money out of nothing there by controlling the supply of credit.  The only limiting factor is the willingness or ability of the the rest of us to take on new debt.  After the housing bubble this tapped US consumer out which is why the only thing currently keeping the economy afloat is the Federal Government filling in the gap with $5 trillion expansion in its debt.  With the debt limit debate fast approaching, our politicians are hell bent on fiscal austerity that will pull this public support out from under the economy...It is my belief that this will create an even greater shortage of credit and liquidity than already exists cutting off the life blood from small business USA.  there will have to be radical new solutions to dealing with this problem when the next shoe drops unless you are happy with Goldman Sachs or Citigroup as your community banker.  this is where ideas like a bitcoin bank could come in.  it would be 100% reserve banking (honest banking) where pools of global bitcoin savings would be lent out to projects approved by the owners of that particular bitcoin pool.  any profits accruing to these pools would flow to all members of the bitbank in the form of bitdividends.  there would have to be some mechanism to ensure that all interest is recirculated back into the bitcoin community to prevent hording. because there is 100% reserve the value of the bitcoin would not be undermined by adding bitcoin credit into the bitcoin economy.  what the bitbank would not be is a legalized form of usury like debt dollar system we have globally today.  its a massive undertaking and one that would face an enormous amount of push-back but given the gross miss-allocation of stored capital that exists today, it's worth a shot.

I'm considering it (I have a registered business already in the process of accepting bitcoins), however for me any bitcoin bank should honour a charter/code thats in line with the bitcoin ethos. Apart from writing the software, this imho is the first thing that needs defining before going into the bitcoin business.

Currently my proposed banking charter would be

• 100% loan backing : Every loan of 1BTC is backed by a reserve of 1 BTC
• Transparency : Real time view of the banks reserves, so as to prove viability
• Saving separate from Loan/investment : Any savings/investment are in a separate reserve which is protected and paid back in case of collapse (no risk to savers)
• Minimum digital protection guarantee : Numbers of copies stored, encryption levels, access rights, etc.
  

However these conditions will make it much less profitable to be a bank than traditional bank, whilst many may see that as an issue it also means its less a 'gold mine' opportunity so likely to develop fast.

Profits on loans are simple and easy, either traditional interest or another scheme fitting those who cannot pay interest (certain religions)
Savers will expect interest BUT a 100% no touch policy would mean no way to generate interest. I'd suggest an agreed use percentage and a interest level based on that. i.e. a saver agrees to allow 20% to be used by the bank and in return gets a 1% interest return on the entire savings... (number plucked out of the air for example only).
For those who just want to use banking current account facilities, perhaps a small monthly fee to covers running costs?
Bank would also be an ideal place to run community investment and loan systems : Peer to peer with the bank offering certain guarantees and escrow services...

Just my ideas so far, would like to hear what others think and what they would want of a bitcoin bank...

What would be the point of paying interest on people's savings, if you can't use their deposits to earn interest yourself (via lending)?

Indeed which is why I suggested underneath an agreed fixed amount of saving for investment perhaps as a system to allow interest. Or prehaps not paying interest, instead taking a small fee monthly for the service of keeping it safe, quick access etc.

Finding a banking method that works both for the consumer and the bank is tricky which is why I'm asking what others think.

IMO lending for interest can work in a sustainable way with a deflationary currency.

The banking part of bitcoins sounds interesting but I am new to this so I can only speculate, sorry if any of this sounds really dumb but I am learning about this and its going to help to think this through...

It looks like there is no rule of law so perhaps its going to be very difficult to do something like loans. The bank right now shouldn't be much more than accepting deposits and charging a security fee to the depositor. People are scared anyway of something happening to their computer, like a virus or whatever, so perhaps there is a way to alleviate that fear. Good old fashioned bitcoin safe deposit box.

Deposit security box is the first thing a bank would do imho, partly to establish trust and confidence in the bank (obviously being a registered business is required but until we have outside auditors of the banks, they have to earn their trust with potential customers.

I do think loans are viable just different from conventional banks. Much closer to the old traditional loan system,

I as the banker have N deposits of bitcoins (my load reserve)
U wants of loan of M bitcoins (M <= N)
U have to provide I with reason to I to loan you the cash, currently that would likely by the existing otp trust system.
If I think your trust worthy I calculate a interest rate on that risk, I loan you that money and you pay a portion of the loan + interest each month.
U default, I did my risk calculation wrong and take the lose.
U are a good customer, U get the loan, I earn interest and your risk ratio (and therefore interest) get better.

You could possibly use a bitcoin deposit for a fiat money loan. If the bank needs to make a margin call or you late with a payment they have the option of dipping into your deposit. Once the loan is paid off you get your bitcoins back or you can take out another loan.

I see this as part of the risk assessment, this is the closest we can probably get at the moment to a security. If U are willing to put some of you money upfront as a security its clearly reduces the risk, its possible to think of other securities but the lack of physical accountability obvious cuts any many of the more traditional ones.

Another way to lower risk would be to have a current account which demonstrates a good constant flow of bitcoins.

A portion of the fiat loan could also be used to buy more bitcoins and as those coins increase in value they then can be used to pay back the loan. E.g. you may borrow 100k and you use 80k to buy a house
and 20k to buy bitcoins. You reduce each monthly payment but are then liable for a balloon. When your balloon is due you sell your bitcoins to pay off the balloon and and use the remainder for anything you like. e.g. go on holiday, take a sabbatical, start a new company, get married and start a family or so on.

has anyone asked the question yet... how do you enforce a loan?
say i put in bogus information.. i'm anonymous.. i take out a 10btc loan... then I go and decided to forget that account, email, whatever... suddenly you have no way to force me to pay back 10BTC.... even if you had my information, going to court for it would be really really tough.

That's why i think bitcoin bank will never work...

This is why ancient banks based on gold asked for guarantees (ie: stuff you do not need right now, but that are valuable, like cars, jewels, etc...)

exactly, and what is a bitcoin bank going to use as collateral?

There's no reason to think this couldn't be done with the existing legal system.  Reputation is also collateral in some respect (or could be).  If you don't pay your debt, your reputation is smeared and no one will do business with you.

seeing that bitcoin is international... using an existing legal system is going to be a severe pain in the ass...
and reputation is hardly a danger for many people, new name, new face... the internet is anonymous.. and bitcoin is semi-anonymous.

Just because bitcoin is 'anonymous' doesn't mean the institutions that utilise it need to be.

You could set up a bank exactly like any other cash bank is set up now... domestically.

Why does a bitcoin bank have to automatically be international?

For security you could put up title to your house or land, just like when people take out loans for fiat.

if its not an international bank, then its of less importance..
if it is an international bank... then they are going to have alot of hurdles to overcome.

Perhaps, but an Australian bitcoin bank could still be mighty handy for Australians, regardless of whether or not other countries have them too.

And the banks could do business with each other based on reputation.

i meant its less important due to the the distribution of bitcoin users.. only a certain few could use that bank regularly...
Its hard to say if a bank would want to do buisness with an individual in another country.. the inter-country laws are sticky. but for the moment lets assume that each country gets its own bitcoin bank (or 30)... which i think would be the most awesome thing ever mind you :P
this would probably be a good model..
however, i was under the impression people were starting banks as external entities as bitcoins are.

Each bank will have its own risk assessment... reputation, a current account and other details are likely to be the minimum for a loan.

Its no different from the current banking/loan system, Can you prove you are working? Do you have a bad credit history? Have you lived at your current location for several years? Can you show ID?

No loan is going to be given anonymously, as a standard business I can profile a customer quite well based on digital information. As banks grow this profile will likely get better as well.

I'd also expect credit rating agency to appear to track bad credit just as they do now.

Some loans will go bad of course, but that risk factor goes into everybody loans interests rates.

It's much different than the current system.  If I default on my car loan, they take my car.  If I default on my mortgage, they take my house.  If I default on my credit card, they come after me for it and can win a judgment against me.

This is because (in Australia at least) they hold the title to it.

So what's different?

If you default on your bitcoin mortgage from a bitcoin bank where you've used your house as security, they take your house.

The first two are secured loans, and require some way of securing which usually is by default local. I can't in country A secure credit in country B using my house/car etc. because they can't get it, if I default (easily).

Secured loans just affect the risk ratio, an easy "takeble" item so lowering the rate you pay interest as collateral.

But take most unsecure loans or a credit card, in those cases essentially they have 3 outcomes
1) You pay
2) You default and they call ballifs to take goods upto the value
3) They wipe it out as bad debt, and make sure you can never get credit again.

3 is for many people what happens, the real reason you want to keep paying your credit bills is so you CAN get credit in future.

As soon as the first bitcoin bank is setup I fully expect a bad debt protocol will come with it, perhaps even a name and shame system (public bad debt record as used in England in the last few centuries).

My thoughts in being a banker, is that I'd accept bad debt will happen sometimes, its part of the business and that risk will be calculated in the interest rates people would pay. Want cheaper interests rates, prove to me your less of a risk (complete loans etc.)

#3 happens, but people often need to file for bankruptcy, and their assets get cleaned out for it.  It's not a trivial process.  There is the arm of the government legal system that they have that they can use as well.

I think the shame/good name system could work, although proving yourself initially is going to be quite costly.  No one will loan to you without really high interest rates (credit cards have quite high interest rates).  But you won't want a high interest loan unless you are very profitable, trying to prove yourself and will pay the cost, or are a scammer.  It's also easy to create a new identity from scratch, so you could just go through that process a lot.  Someone will be creative and find a way to make this work if it's a good idea, though.

I think it could work with the correct KYC polices in place,
but of course this is at cost of anonymity.

Bitcoin transactions are anonymous, but that doesn't mean banking systems using bitcoins need to be. I can imagine a bitcoin bank would not lend anonymously for obvious reasons. The solution is simply that the bank needs to know who/where the person is, and have a reasonable expectation that the loan will be repaid. Some banks will have higher thresholds for "reasonable expectation" than others, and you'll see this risk reflected in the interest rates they offer. A bank which required proof of identity and/or collateral will be able to offer a lower interest rate and the market will decide what model works best, or perhaps multiple types of banks would exist concurrently.

Also, regarding fractional reserves... there is nothing wrong with them SO LONG AS bank losses are not insured by a government agency AND so long as the bank makes this fractional policy known to its depositors. A fractional bank can offer higher interest rates than a 100%-deposit bank, and individuals can choose their risk/reward accordingly.

I get how it's possible to create a fractional fiat bank, because you can just make $ out of thin air to lend out, but how do you create a fractional reserve bitcoin bank?

Okay Bank lending is one aspect of a Bank, which seems doable with some caveats about size of loan and trust but not the only one. The other two personal banking areas are saving and current accounts.

So what does that mean in a Bitcoin bank?

A saving account traditional holds your money in trust returning you an interest rate bonus for doing so. The interest is generated via the bank being able to use your held money for loans, investments etc. The problems are saving guarantees, currently most countries guarantee consumer savings up to some value even if the bank fails, that of course can't happen for bitcoins, so do would we need a guarantee? and if so how would it be achieved?

My approach would be that saver specifies how much of their savings can be used for interest, from this the amount of interest s generated but also is potential lose if Bank closes.
So I save 10BTC and decide to allow 5BTC to be used for a guaranteed interest rate return of 1% per month. So assuming no component interest (could be the savers choice, which pool interest goes into), I earn 0.5BTC interest per month.
A Bank may set a minimum amount to disallow free savers (0 used for interest), on the other hand it might just let it happen (free saving account) as a goodwill / PR thing. I'd also suggest that a 0% interest saver account is really a wallet backup account and may be clearly to call it a backup account rather than savers to show it earns nothing.

A current account, the most used part of a current account is paying bills.

For a BTC current account i'd suggest

Instant transfer internal to the bank and to other banks with a data transfer agreement (A Bank could fast track bitcoin transaction time because it knows both ends, if they are both customers)

Traditional bank to bitcoin bank transfers - probably local country only but transferring money from Cash Bank to and from bitcoin
bank would be safe and cheap for the Banker.

Statements and tracking - allowing you safely see what you have, where you spent it

Secure - you wallet and bitcoins safely stored and protected from local effects

Overdraft - A bank may choose to allow a customer so head room, a predefined loan space...

Fee - A fixed monthly charge would seem to be the simplest and fairest system. The customer pays a small fee per month for the current account to be maintained.

Would anybody find such a thing actually useful?

I think I've never understood this fractional reserve banking 'cos I've seen obviously informed people make statements like this.  My understanding is that our current banking system is about 5% reserve - a bank can load out $100*0.95=$95 of every $100 it receives.  When the $95 is redeposited, then another $95*0.95=$90.25 can be loaned, then $90.25*0.95=$85.73 and so on.  Creating vast amounts of "non-existent" credit based money, from a minimal initial investment.

What noagendamarket is describing here, that is, a bank that loans the coins it has on deposit (I implicitly understood "can loan *all* the coins it has on deposit"), is NOT a 100% reserve bank, it is a 0% reserve bank - it is *way* worse than a 5% fractional reserve because an initial $100 deposit can turn into an infinite amount of credit.

An actual fully reserved bank, i.e. a 100% reserve bank, could not make any loans at all.  That's the only way to ensure that all deposits are backed 100% by reserves.

Shoot me down.

You tell someone they have 100BTC in their account, and give a loan out for 90BTC, keepign 10 BTC in reserve.  Repeat this times a lot of customers, say you have 100,000 BTC in account credits, loan out 90,000 BTC, and keep 10,000 BTC.

If someone who deposited wants their money, you take out of the 10,000 BTC.  If enough customers try to take out the money, they are SOL.

Pixie, tomcollins et al, thanks for your input.  I think THE most important role bitcoin could play is as a mechanism for social credit.  I think its important to not get lost in the model of traditional banking.  let's not even call it a bank for a moment.  lets assume these are social lending pools. tom you are right on the money in that there doesn't need to be an enforcement mechanism greater than Ebay's reputation watch.  since the lenders (bitcoin savers) create the pool, they can set the terms, and rules for enforcement.  Because they are taking that risk they get a portion of the future bitcoin stream that results from the project (interest).  If the borrower fails to repay the loan that is a bad outcome for the lenders but not bad for the economy because those bitcoin will be spent and redistributed throughout the bitcoin economy. BECAUSE THERE IS NO FRACTIONAL RESERVE LENDING (ie lending out bitcoin notes that promise to deliver real bitcoins), NOTHING IS CREATED OR DESTROYED.  The lenders loss, is the producers gain.  The borrower reputation goes down but even he can get back in the game eventually.  The lenders learn their lesson and in the process become better underwriters.  This cuts out the idea of a bank entirely. Because we can safely store our own bitcoins, we don't need the middleman who was only ever really offering the secure storage of money anyway.  My idea is merely to link borrowers (entrepreneurs) with willing savers (investors).  In fact every transaction (bitcoin loan) can have its own custom terms depending on the parties involved.  THIS IS WHAT FREE MARKET CAPITALISM REALLY IS.  Not the banking cartel monopolized system of usury that we currently live under that the monopolists would have you believe is free market capitalism.  to me the road block to this will always be the establishment, and they will use taxation to try to control any free credit system that emerges from this clear monetary revolution that is bitcoin.

Again I am really interested in working on this project real time.  Feel free to contact me via email if you know of people already working on such a product.

some entities may be interested in it.
already distributed enough on Globe and have both financial and political interests.
both legal and not/acused in illegal.
for example - number of charity orgs in 1st case and wilileaks/other freedom-f-orgs.

Tell me about investors who invest in this social lending pool.  Suppose I deposit BTC100 in your lending pool. Do you then issue me with a credit note (i.e. bitcoin notes, "promises-to-pay") saying I have the right to recover BTC 100 from the pool?  Do you then lend my BTC100 to someone else?  Well, now I have 100 "virtual" bitcoins, and someone else, the borrower, has 100 "real" bitcoins.  That's 200 bitcoins.  Where before there were only 100.

If you want this to work, without your investors being really pissed off, then their deposits can be recovered only on condition the loans get repaid.  That way there's never extra bitcoins "created".  Good luck finding investors under those conditions.

If an investor has a track record of investigating his investments and deriving Bitcoin profits from them, I might be interested in lending him Bitcoins in return for more Bitcoins after a specific date.  Naturally, the Bitcoins would be inaccessible to me while they are being used.

If I wish for someone to expend effort storing my Bitcoins securely, and make them available to me via a trusted ID system, it's reasonable that I pay for this.  One way I can pay for this is to allow some of my Bitcoins (any arbitrary percentage) to be used in investment by said entity, along with the explicit statement of repayments, inaccessibility time, risk and guarantees.  Alternatively, I can simply pay a small percentage of the Bitcoins stored in fees.

As long as the system has reputation to be lost, shouldn't the market choose between different risk profiles?

It might make sense for companies producing something to be the first lenders.

Like an automobile manufacturer, for example. "I will loan you bitcoins to buy my automobile" kind of thing.

Basically, any case where really it is not in effect actual bitcoins that are being loaned, really what is being loaned is "whatever it is that the borrower had been thinking of borrowing bitcoins to buy".

In other words loans *denominated* in bitcoins but not involving actually putting any actual bitcoins into the borrower's possession.

Two of the blockchain-based currencies my IRC bots support are already doing this; it has also the advantage of causing the loans to be at least to some extent "secured" (the extent being a combination of the effectiveness of the repossession facilities in use and the re-use value of the used merchandise, and this can be priced into the cost of the loan aka the price of the merchandise when purchased "on credit" in this way).

-MarkM-

Yes, it's called fractional reserve banking. Here is a documentary called Money as Debt about fractional reserve banking: http://www.youtube.com/watch?v=Dc3sKwwAaCU

Fractional reserve banking would not be possible with Bitcoin as I see it, unless the bank would start to give out credits and demand interest for those credits, but that's not possible with Bitcoin as I understand it.

It is quite possible, but also ridicously risky and might lead to massive bankrupcies and defaults.

You mean making Bitcoin credits is possible? Yes, if a Bitcoin bank knows that only a small percentage of its customers' bitcoins stored at the bank will ever be demanded by the customers, then they can lend out the other part and demand interest for that. lol. That will create a similar situation as today with the risk of runs on banks which means that the banks have lent out more money than they actually have in deposit.

exactly i hate the fatcats too :)

It's reasonably possible to orchestrate a bank run day with Bitcoin banks.  It could happen every February third.
Banks could protect themselves in their contracts by charging for high liquidity.  If you wish to withdraw more than 50% of your savings, you must give 30 days warning or pay a penalty, for example.

A bank can provide valuable services in return for the fees or hassles.  I might bank with a Bitcoin banker who; launders my Bitcoins, provides Biometric security access (blood, hashed SNPs?), and has branches or deals with other Bankers around the world.

Tell me about investors who invest in this social lending pool.  Suppose I deposit BTC100 in your lending pool. Do you then issue me with a credit note (i.e. bitcoin notes, "promises-to-pay") saying I have the right to recover BTC 100 from the pool?  Do you then lend my BTC100 to someone else?  Well, now I have 100 "virtual" bitcoins, and someone else, the borrower, has 100 "real" bitcoins.  That's 200 bitcoins.  Where before there were only 100.

If you want this to work, without your investors being really pissed off, then their deposits can be recovered only on condition the loans get repaid.  That way there's never extra bitcoins "created".  Good luck finding investors under those conditions.

Yes this transaction would create a new asset that is a claim on "BTC100 + interest" contract owned by the pool.  But that is not the same thing as 100 bitcoins, its just a contract.  Again the lenders/investors are receiving a BTC100 + x in the future.  Whatever this x is compensates for the risk of parting with the BTC100 today and possibly permanently.   In deciding what x is depends on what the lenders demand for return on their bitcoin savings is and what the borrowers demand is for bitcoin financing and ultimately, the potential (perceived potential) of the venture to produce a rate of bitcoin profit that is greater than x. 

This is much different than how lending works today.  A metaphor is that the bank would make you a BTC100 loan which you then use to purchase the collateral securing the loan. The seller of the collateral deposits the BTC100 coins back in the bank. You now have the real asset and a liability to the bank to repay BTC100 + interest.  The back has a claim on BTC100 + interest x secured by the real asset, and a liability to the seller of the asset of BTC 100 + y.  The BTC100 didn't exist before this transaction and neither did the interest x.  The bank pockets the difference x-y and now holds a claim on the real asset involved.  If you default and the BTC value of that asset falls below the BTC 100 deposit liability, the bank owes the depositor BTC100 - real asset value.  With the financial crisis today, this shortfall is being picked up by the taxpayer through an artificially wide difference between x-y (US 10Y 3.2% - fed funds 0-.25) and the ability of the banks not to have to account for this loss all at once.  Therefore, all that the Financial Crisis did was make explicit what was the implicit tax payer backing of the banks' ability to create BTC100 when they did not exist before.  This is credit money inflation and since 1980 has run at about 9.8% annually in the US.  Thanks to this system, the owners of the finance industry gets to keep the x-y profits it made along the way paid out in dividends.  So ultimately lending today is just a scheme by the lenders to gradually siphon off a bigger an bigger share of the overall pie of wealth.  Now that the scheme has reached its mathematical limit (ie the borrowers cant take on more debt), these holders of the newly concentrated wealth can wield it to restructure the World in whatever way they see fit. 

OF COURSE THIS IS A HUGE OVERSIMPLIFICATION, but the point is that we wouldn't even be talking about something like Bitcoins if people weren't waking up to the fact that something serious has changed with our debt based money system.  That is why Bitcoin is such a brilliant way to start a new community based system of trade.
In the community based economy, the lending pools or investment trusts would be an integral way to keep the currency circulating and put the benefits of lending where it belongs, with the savers not the bankers themselves.

For those interested, money as debt 2 is a great video for exploring this topic:

http://www.youtube.com/watch?v=rCu3fpg83TY

It could be a nice idea offer something like 3.9% APR current accounts and offer loans and such quite like a normal bank but you'd have to have contracts so that people HAD to pay for it and they should upload photo ID like a passport so you can prove it in court etc :)

Essentially you are entering into a contract with a loan and with a thorough screening process you could have enough information to go after the person in court fairly easily (assuming they were in your country). The fact that the contract is in bitcoins doesn't matter, as a court would rule on any sort of contract, whether someone promised to pay in Euros or gold or chicken, so long as you could prove that the value of the dispute is greater than $20 (in the US, at least 7th Amendment).

I think the concept of a bank is a fairly reasonable one. Individuals deposit their BTC in a "wallet" account setup through the bank site. Loans come from individuals looking to purchase mining hardware and present a proposal of what they need, the cost, the estimated Mhashes, current mining group and number of hours per day the rig would run along with identification information, passport, driver's license, ss card, proof of income, record of any collateral, btc and otherwise. Based on this, the current and future estimated difficulty and exchange rate, an interest rate and repayment plan would be devised that allows for a generous margin for both profit and potential risk. Depositors have basically invested their money in future mining expeditions and the bank profits by assuming the risk and collecting extra overhead as a service fee.

From a risk perspective, it would be best to initially limit investors deposits to only pay interest if they were invested in a CD-type format, for a period of 3-months, 6-months, 1 year etc with those withdrawing early receiving a substantial reduction in paid interest. Once the bank has a substantial amount of guaranteed capital, faith in the bank would be increased and it could eventually offer interest on regular "checking" accounts based on average daily balance. A limit to account size could be put in place to ensure no single investor could create a bank run. Other restrictions, such as limiting the number of new accounts offered in the event of low loan demand could be established. Interest paid can be adjusted based on exchange rates and difficulty values.

I think it's a fantastic idea. The bank itself would need a fairly substantial capital backing when it first opens, in BTC and probably a "real" currency like USD. I'm very interested in this and if anyone has the programming skills to create such a thing, I'd be happy to pay you, in American greenbacks.

As stated earlier, I'm seriously thinking and exploring the idea... As a business man, however the first thing is to have a concrete business plan.

As such that is what I'm doing with now, trying to balance the potential books, how much investment, how to generate profits, share-holders or not?, advantage to users, etc.

Ironically given the current state of main economy banks, banking is all about trust... And that is a hard thing to get, for a bank to succeed, at the beginning it has to inspire confidence that it can keep you money safe and knows how to grew it (for interest and share-holders).

One possibility I've considered is to follow the model of the co-operative banks (AKA building societies) in the UK. On these there are no share-holders or owners except the customers. Your deposit with the bank, entitles you to a dividend each years based on the profits made.

The problem with a Bitcoin bank is that the bank would have to lend out their customers' bitcoins to other customers, because in contrast to ordinary money, bitcoins cannot be created out of thin air, plus bitcoins are inherently deflationary so for the bank to buy bitcoins just to lend out would require huge interest rates in order to become profitable. Also, while this has worked for ordinary banks, people today would perhaps demand that there can be no run on a bank dealing with digital currencies, i.e. that the bank is able to pay back all their customers' money if they all should demand it all at once.

You avoid runs completely if accounts are not on demand accounts. Mainstream banks used to do that with time deposits that paid interest and demand accounts that were free or cost a small fee. No reason a bitcoin bank can't do the same.

On the other hand a bitcoin bank would have a tough time making sense. Bitcoins will always increase in value so a business borrowing bitcoin from a bank would have to estimate that increase and add it to the cost of money when creating its business plan. It just doesn't make sense to borrow in bitcoins unless there is absolutely no alternative.

This is unfortunate because the people holding wealth are not always the people with the great business ideas. If you don't have a mechanism in place to utilize the wealth to create business you are running at a much lower efficiency. This is why I've come to believe bitcoin will be a great store of wealth but never a unit of account (for loans in this case.)

Ah, yes the bank can demand that the bitcoins their customers have saved will be frozen for a year for example in order to give interest. But even then customers could demand all their bitcoins all at once which would cause a run on the bank.

The deflation rate of bitcoins may not be a problem on a second thought, because when the bank lends out bitcoins it expects the customer to also pay back the loan in bitcoins (or the equivalent value). But as you said, a borrower would also take into account the deflation rate of the bitcoins, and that may make the customers reluctant to borrow money from the bank.

Pixie, I think your idea of a depositor owned bank is the only way.  That is why I don't like to use the word bank to begin with. What I am really looking to create is more like a collection of one-off Bitcoin investment trusts or venture capital funds depending on the structure of the loan.  Really all the business itself would do is help to facilitate the transactions.  Such services could include connecting investors with borrowers, bookkeeping, evaluating creditworthiness, standardizing contracts, and legal services, etc.  For these services, the Bitbank would charge fees.  These fees then flow back to the member's of the Bitbank community in the same way a thrift works. This not only captures the spirit of community lending and investment, but ensures that their is no conflict of interest between those running the bank and the "depositors", because the two are one-in-the-same.  Again, I'd like to work with anyone to help set this up. Is there a good way of getting in touch via email?

The problem with bitcoins from a banking perspective is that it is difficult to create Bitcoin credits. With ordinary money creating credits is not only easy, it's a must.

What circumstances force people to borrow bitcoins instead of, for example, borrowing enough fiat currency to buy such bitcoins as they desire or need?

Doesn't it make more sense to borrow something else and buy bitcoins than to borrow bitcoins and use them to buy something else?

(Unless you expect the "something else" to appreciate in value faster than bitcoins.)

-MarkM-

Why? Interest and loans and a banking system existed under more or less fixed supply non-fiat money systems for hundreds of years.

If there's a difference in prices for a given good, borrowing in bitcoins could be favorable. If you have a $100 USD good and the seller accepts either $100 USD or 10 BTC when the exchange rate of BTC is 1BTC to 7 USD, you could have pressure to use BTC directly. If the seller feel bitcoins (deflationary) are going to hold their value or appreciate in value better than the inflationary USD, the seller may price their goods based on what they feel the currency they receive will be worth when they go to use it, that is, better than the current exchange rate.

Yes, loans, but I mean credits as in money created out of thin air. Today when banks create a loan they only need to have 10% of that loan as real assets to back it up. The rest, the 90%, is thin air. Bitcoins cannot be created out of thin air, only through massive comping resources.

Isn't this more or less what the global bitcoin stock exchange does?  glbse.com

I'd be interested in helping in this, i could be the graphic designer (make logos, advert banners, videos etc)

And as this forum has pretty much only computer nerds i think we would be the only un-hackable bank xD

That's not how fractional reserve works at all...

Check out this documentary called Money as Debt: http://www.youtube.com/watch?v=Dc3sKwwAaCU

It's a simplified explanation but that it how fractional reserve banking works. What do you think the word 'fractional' means?  ;)

Sadly, that really is how it works.  Except that it is actually 100% that is created out of thin air, not 90%.  But then the bank can borrow money (from depositors, other banks, or the fed) until they have sufficient reserves.

And even worse, very often the 10% which is supposedly "real assets" can often consist of very shaky assets - the kind of mish-mash securitized sub-prime mortgage bundles that get artificially high ratings and so can count fully to a bank's reserves, when in fact, they were worth ~10¢ on the dollar when the reckoning came.

My concern is more how will you assure the money the bank holds?

Lets say a bank has 1.000.000 BTC from his account holders and then all wallets are destroyed by a software fault, corruption, virus or even a Solar Storm that destroys the bank-wallet. How do you want to manage the insurance?
1.000.000 out of the total 21.000.000 BTC ever created is equal in value to
  952.380 out of the left 20.000.000 BTC.
So the bank needs this amount of BTC to pay the account holder for the lost BTC. How will the bank pay the insurance premium for this huge insurance sum? Who will be the insurer?
The bank may not loose this amount, but if a normal bank burns, the paper money can be printed again for a much smaller fee - not so the BTC.

How do you want to manage this?

Hi deti, yes it's impossible to regulate banks so it will effectively be a "free banking" system, banks can suffer a bank run or even, if they are as bad as you suggest, effectively disappear. Presumably some of them will be real legal organisations with a contract where they provide you with some USD instead if they run out of BTC or even lose them as you suggest. Also I hope no one bank would be as big as all that, it would be quite scary.

I think a bitcoin bank can not work.

A bank earns money by collecting money from depositors and offering them interest in return. The bank uses this money as a fractional reserve, and lends more money to creditors and expects interest in return. The interest for the depositors is being paid for by the interest of the creditors.

In a strongly deflationary environment it's not easy for the creditor to pay back the loan, because by the time the credit falls due for payment the money is already worth more. For the creditor to be able to pay a positive interest rate, his business needs an roi > deflation rate. Many businesses will not be able to provide this when the deflation rate becomes high. Then lending out money for a bitcoin bank will hardly be possible which also makes paying interest to depositors hardly possible.

So, in my opinion, there is no incentive to put bitcoins in a bitcoin bank.

If bitcoins had a stable value, this problem would not exist.


alexk

Lots of big banks these days don't pay much in the way of interest, half a percent or so.  And their loan business is crap.  They make money on credit cards and transaction fees.

So credit cards are not loan business?

Yeah consumer credit isn't technically a loan.

http://en.wikipedia.org/wiki/Credit_%28finance%29

That page doesn't say consumer credit isn't a loan.

I don't see anything in that wikipedia article supporting your argument. Can you point it out please?

http://en.wikipedia.org/wiki/Loan

According to this article though, credit card debt is an unsecured loan.


The discussion if a credit card debt is a loan or not is irrelevant to my first argument anyway. A Bitcoin Bank can not give out interest at all, because there is no way it can profitably lend money in huge amounts to debtors. So there is no financial incentive to put your Bitcoins in a Bitcoin Bank and i think this prevents Bitcoin Banks from having any success, as long as the bitcoin money supply scheme stays deflationary.


alexk

Sheesh, you can't read the first paragraph?


Consumer credit is vendor finance.  The point is to get people to purchase stuff, not to pay interest.

Ok, i think i understand now. You are mistaking credit cards for consumer credit. These are not the same. Consumer credit means, that someone has to pay for a product at a later point in time and is charged interest in return. Credit card debt is a loan, which also costs interest.


alexk

Well, in the US, credit card debt is normally a loan of goods or services, when it's not a cash advance.  So it's consumer credit.  I'm not sure what the definition is wherever you live.

Regardless, loaning Bitcoins isn't required for such services; and they can support a Bitcoin bank perfectly well.

Ok, you can continue making up your own definitions. According to http://en.wikipedia.org/wiki/Loan credit card debt is an unsecured loan, which i already pointed out. This will be my last post regarding this subject, i'm not here to argue about definitions.

If you can show me a way how a bitcoin bank can pay
1. interest to its depositors or
2. earn money by lending
i will be happy to continue with the discussion.


alexk

Fractional banking works as well with bitcoins as with any other asset.

Person A deposits 100 BTC at the bank and gets interest. He can withdraw them at any time.

Person B loans 90 BTC from the bank and pays interest. He have to pay back within a year.

The bank only has 10 BTC but have promised to be able to give A back his full 100 BTC at any time. This makes it a fractional reserv bank with a 10% reserv. If A want to withdraw more than 10 BTC the bank will have to take a loan to cover it, but as it isn't sure the bank can get this loan A can lose his money.

Some argue that the total ammount of BTC in this example is now 190, and that the bank has fraudulently created 90 BTC. If this is the case or not is a matter of definition but as long as the depositer knows about the fractional nature of the bank, I would not call it fraud.

I can show it:

A deposits 100 BTC and is promised to get 110 back in a year.

B loans 100 BTC and promise to pay 120 back in a year.

After one year the bank has earned 10 BTC.

This example is not fractional reserve as the bank got assets covering all its promises. The bank is just a boker of loans between it's customers. A could of course lose his deposit if B never pays and this forces the bank out of business, but even if the bank just stored the BTC they could be stolen by a hacker. Deposits are never 100% safe.

Thank you for your reply.

Even if there was a fractional reserve in this example i would not have a problem with it. Fractional reserve banking is not the problem with bitcoins, it has been done with precious metals since the middle ages.

My problem with this example is the following:
Bitcoins are highly deflationary. Let's assume 20% deflation in bitcoins and 0% interest that you have to pay to the bank. This means you borrow 100BTC at the beginning of the year and have to pay back 100BTC at the end of the year. By the end of the year, these 100BTC will be worth 20% more than at the beginning. To make up for this 20% difference, the debtor needs a return on investment of at least 20% for his business. If the debtor also has to pay interest, the required return on investment would be even higher. This makes it impossible for many debtors to use borrowed capital and thus makes many businesses economically unfeasible.

20% deflation is a rather conservative estimate of bitcoin deflation. This is the reason why bitcoin banks can't earn money by lending out capital, which makes it impossible for them to pay interest to depositors.


alexk

With regards to the discussion of collateral.  It might be possible for the bitcoin bank to hold title to the asset being purchased with a loan and then when the loan is paid off, turn the title to the asset over to the payee of the loan.  Until the loan is paid in full, the bank owns the asset being borrowed so if the payee breaks his/her agreement with the bank, the bank keeps the asset.  If all borrowing records and deed records are public, this could potentially be enforable by the community, but it would not be as private as not taking the loan or not using the bank. 

I think the more serious question is business banking.  The operation of a board of directors differs with juristiction.  I know that there is a BTC Stock Exchange being started and that is fantastic.  If companies are going to IPO on it before we have a bitcoin banking system, I would assume that they would be previously incorporated in other juristictions.  This is fine.  PB is listed on stock exchanges in America and the UK and has business units all over the world.  Do we hope to one day have a legal system within in the global bitcoin community?  Otherwise even if our money is not tied to any government, is our bank (if we choose to use it) or our Stock Exchange?

A bank always considers the expected inflation or deflation when determining interest rates. With a high deflation interests for both saving and lending would be low. A saver would accept a low interest as he also gains the increase in value from deflation. The one ledning can only pay a low interest as he also have to pay the increase in value.

Deflation do not by itself pose a problem for a bank, but it can indirectly: As you say the return on investment of any venture backed by a loan has to be bigger than interest plus deflation. This could limit the number of such ventures. But the problem is not the bank as it only adds a little to the return on investment you need to break even compared to if you used your own saved money (for example 20% deflation and 2% interest).

A much bigger problem today is the unpredictability of the future value of BTC. Both depositing and lending BTC is very risky for all parties includning the bank.

Right, but if some day the deflation becomes equal to the increasing worth of the economy, lets say 2% p.a., then a bank can take some percent too and everyone will be able to pay this back with some efford.

Honestly I'm not sure what you're here to argue about.

You keep wanting to assume that banks have to make loans and pay interest.  And I'm telling you that this is emperically false.  It's not even the way banks operate today.  Go sign up for a savings account.  See how much interest you get.  Apply for an unsecured loan.  See if you get one.

Banks today operate by enabling trade, not savings or investment.

FRB is not applicable to savings and loans banking. And a deposit bank for bitcoin is pointless...

I think a bitcoin bank loan would be possible using a mining pool.  That is to say, a pool's owner could set up code that allowed a loan to take place (using whatever comfort zone they have for a loan for a particular user.  If you're a newbie, probably a small loan, if you've been with it a while, especially if you're a huge contributor, then perhaps a large loan).
 
Then the payback could be hard coded such that some percentage of their contribution is pulled out automatically (to prevent someone from never paying it back) and then would keep pulling out that percentage until the loan is payed off (plus interest if desired), then they could go back to getting their normal payouts.
 
e.g.
The collateral would be the potential of a particular debtor:  ie if I've been with slush for, let's say a month, and I get about 15 coins per month, maybe he thinks I'm worth 10btc, and is willing to risk 10btc but only if I pay back 75% of my earnings each time there's a pay-out back to him until I've paid off 11btc.  Does this make sense?  So once I hit the 1btc mark (my current payout mark), then 25% of it would go to me, and 75% back to slush until it's paid off.
 
But maybe a year from now I've put together an awesome machine and earn myself 100 btc during that month, then maybe slush decides he'll risk an 80btc loan at a similar payback.  etc.

Someone could still cut and run, but that's up to the pool's owner to decide who is worth what, and at what risk.
 

Hello all,


I notice that there are a lot of misconceptions about fractional reserve banking flying around here. There is a great confusion between the current government central banking fiat currency system (which HAS and DOES print money out of thin air) and sound fractional reserve banking. They are not the same and FRB need not be inflationary. In fact, when done properly there is very little difference between a fractional reserve bank and a 100% reserve bank. A 100% reserve bank can earn just as much money and get just as much interest as a fractional reserve bank. Yes, possibly even more. I describe how this can be done in the following article:

http://onarki.no/blogg/2011/04/fractional-reserve-banking-vs-pure-gold-standard/


However, due to the hyperdeflation in bitcoins at the moment banking is currently impossible. A proper bitcoin bank must wait until the value stabilizes.