futures trading service

[killerstorm]

hi

I'm looking for some BitCoin-related project to develop (web site/service), and one of things which comes to mind is currency futures trading.
As far as I can tell there are no things like this right now, neither there are prediction market/betting services. (It looks like BitBet have died.)

There is a lot of currency exchange sites for Bitcoins, but currency futures allow easier hedging against exchange rate swings and easier speculation on exchange prices.

Besides that, these currency exchange sites are huge PITA to deal with when you need to work with USD or other currency, as they require verified accounts linked to identity, bank wires and so on.

While futures trading could be 100% pure BitCoin, and thus fully anonymous and hassle-free. The trick is to do settlements in bitcoin rather than through shipment of actual currency. So it is more like prediction market than futures exchange.

So we can take, for example, MtGox spot prices as a base (as it has largest volume, as far as I can tell). At delivery date seller must pay the difference between futures price and current spot price on MtGox if MtGox price is higher; otherwise, buyer pays the difference to seller.

There is, of course, counterpart risk, which is especially acute with anonymous payments. That's why we will have margin calls -- sellers/buyers are required to post collateral when price goes up/down, so there are interim partial settlements each day (for example) instead of a huge pay day. If counterparty does not answer margin call his position is eliminated.

Thus while there are still counterparty/credit risks they are limited to daily fluctuations via margin calls.

So, the question -- would anybody use this service? (I need to know before I start developing it.)
The benefit is that you can easily hedge or speculate on currency price changes.

The downside is that service itself acts as counterparty for contracts, so you need to trust service. I guess it is better than trusting each individual market player. Although it is possible to handle some parts in P2P manner, and service can settle balance as soon as it receives margins, so risk can be minimised

I thought about doing options too, but options are way more tricky w.r.t. anonymous payments as counterparties can just walk away if it is not to their favour. Service can hold option fee in escrow until it is settled, but still I think there is a huge risk of abuse. With futures it is much more honest.

If you do not get futures terminology please check wikipedia: http://en.wikipedia.org/wiki/Futures_contract

[mpkomara]

I am interested in this service.  Maybe a poll would be an easier format for this thread.  bitcoinsportsbook.com had the closest thing to a futures contract, but I haven't seen activity there since Feb. 

[killerstorm]

I am interested in this service.  Maybe a poll would be an easier format for this thread.

Ouch, I didn't know that this forum can do polls. Ok, let's say I'm collecting opinions about how futures exchange should work like. There are many possible options about settlements, spot prices, margins, defaults etc. Maybe somebody knows how to make "physical delivery" which won't be PITA.

bitcoinsportsbook.com had the closest thing to a futures contract, but I haven't seen activity there since Feb. 

Thanks. I've skipped it because it says sports in the name.
I see significant activity in MtGox price bets so I guess there is a significant demand for betting on exchange rates.

[mndrix]

I might be interested in using your service.  I like your idea to avoid physical delivery and to have regular margin calls.  I suspect that miners acquiring new hardware would be particularly interested in futures since prices dramatically affect their profitability.

Most participants won't know a lot about futures trading and will only want a simple way to hedge against price fluctuations.  If there's any way to make futures trading easy for them, I suspect the service would be popular.

[hozer]

I was about to ask for someone to quote me (in BTC of course), what it would take to develop a solution to accept payment in Bitcoin for domains and webhosting through the OpenSRS reseller API: http://www.opensrs.com/integration/api

Being able to hedge exchange rate volatility would be necessary to be able to offer consistent prices in BTC for domains. I would also prefer an exchange that required a full cash position to cover any margin risk vs margin calls.

I would also ask you to think about the anonymity implications a little... Do you want an anonymous futures trading service, or a low-barrier-to-entry futures trading service? They are very different design goals, with very different legal implications. I'm interested in a transparent, low-barrier-to-entry trading service, and I will want to know the physical address of the exchange, and the management team running it, and I have no problem providing the same information.

[Nefario]

Futures trading (on almost anything) will be a major feature of BitcoinGlobals(my company) stock market, which will be launching in the next couple of weeks.

[killerstorm]

Being able to hedge exchange rate volatility would be necessary to be able to offer consistent prices in BTC for domains. I would also prefer an exchange that required a full cash position to cover any margin risk vs margin calls.

I would also ask you to think about the anonymity implications a little... Do you want an anonymous futures trading service, or a low-barrier-to-entry futures trading service?

Can you please elaborate on "full cash position" and "low-barrier-to-entry"?

I really don't see how non-anonymity can help unless we can prosecute people who do not fulfill their obligations. I dunno, maybe that would make registering fake accounts too much hassle, but in theory it doesn't solve anything. Otherwise some security deposits might cover potential loss from abuse.

I understand that in your case you'd like to minimize counterpart and credit risk. Maybe it can work via insurance service provided by exchange?

They are very different design goals, with very different legal implications. I'm interested in a transparent, low-barrier-to-entry trading service, and I will want to know the physical address of the exchange, and the management team running it, and I have no problem providing the same information.

Do you mean that exchange should take counterparty and credit risks and you will trade with exchange as with a trustworthy entity?

I'm not sure that any low-barrier-to-entry scheme can be viable to exchange operator.

But I'll think about this...

[hozer]

Can you please elaborate on "full cash position" and "low-barrier-to-entry"?

They are very different design goals, with very different legal implications. I'm interested in a transparent, low-barrier-to-entry trading service, and I will want to know the physical address of the exchange, and the management team running it, and I have no problem providing the same information.

Do you mean that exchange should take counterparty and credit risks and you will trade with exchange as with a trustworthy entity?

I'm not sure that any low-barrier-to-entry scheme can be viable to exchange operator.

But I'll think about this...

What I'm meaning by the term 'full cash position' is that I have to have a positive balance (in bitcoin and/or other currencies) sufficient to cover the *entire* risk of the trade I'm wanting to make, so that there is never a margin call. (I don't want to wake up one day and find out all my trades got canceled because the market moved from ... oh, say $0.8 to $1.6 in a week). I suppose you could have margin calls be an option, but then I want the *exchange* to guarantee that even if the counterparty defaults on the margin call, the trade still goes through.

This way I'm just trading with an exchange, with a physical address, and known simple contracts, and by doing this, we make the legal system our allies. You (the exchange operator) *could* take people to court who default, but that's up to you and your own fraud prevention. The physical address and knowing the jurisdiction also allows me some protection for putting a lot of cash in to cover margins, because if you get hit by a huge fraud wave and have to file bankruptcy, I'll be likely to at least attempt to be able to recover something.

I guess this isn't really 'low-barrier-to-entry' though, but this is what I think I need when I go try and explain this thing to my attorney and tax preparer.

[xf2_org]

Bitcoin futures and options marketplace is needed badly. Do it.

+1 agreed

[killerstorm]

I've started with this, but it turns out that there are many possibilities of what to do in case of default of one party and they all suck a bit.

I first thought about resolving problems via the market itself but it is rather complex to implement and has many weird corner cases, so I decided to leave this and go with something simpler.

Two simplest options are:

1. Cap maximum possible profit/loss and require full reserves. This way default cannot happen.
2. Settle contract immediately when margin requirements are not met.

For example, let's consider situation where parties agreed to buy/sell 100 BTC at 1 BTC/USD.

In the first case they need to agree on exchange rate caps (which determine reserve requirements), let's say they choose 0.5 BTC/USD to 2.00 BTC/USD.

If it goes to 0.5 BTC/USD then BTC-buyer needs to pay seller 100 BTC (100/0.5-100 = 100), thus he is required to allocate 100 BTC until contract is settled.
If it goes to 2 BTC/USD then BTC-seller needs to pay buyer 50 BTC (100/2-100 = -50), thus he needs to allocate 50 BTC.

This is nice because defaults are not possible, but it requires traders to have large reserves (or cap their contracts). But this can be ameliorated in case trader engages in multiple contracts in different directions -- exchange can calculate worst possible outcome and demand only as much margin as needed to cover that case. I.e. if you've bought 100 BTC at 1 BTC/USD and sold same amount at 1.1 BTC/USD then you get profit no matter what and so you need no reserve at all.

In the second case, if seller has only 50 BTC in reserve and price shot up to 2 BTC/USD contract is settled immediately -- seller pays 50 BTC to buyer and that's it, no matter where price goes next. Then buyer can issue another buy order to get position back, with profit he got covering price difference. But this is not possible if current futures price is far from spot price, etc. Also it opens a room for abuse -- if trader anticipates significant price move he might issue both sell and buy order from different ('fake') accounts. Then when price moves he loses only one of his initial margins but wins on opposite contract. E.g. price goes 20% up while initial margin is 5%. Then sell contract loses 5% in initial margin but buy contract gets 20% profit and asshole trader wins.

[FreeMoney]

I think there is place for all kinds of systems, but I'd start with capped loss and full reserve. That will still be useful and will be easiest to implement and keep free of fraud.

[topynate]

Any broker or exchange that allows 5% initial margin is crazy. It might be OK in the extremely liquid forex markets that deal in well established currencies, but not in a market like Bitcoin/USD where the market depth can be measured in thousands of dollars. Rule of thumb, your margin shouldn't be of the same order of magnitude as the spread...

[dacoinminster]

I am completely convinced that bitcoin options and futures (betting on bitcoin and on other currencies, stocks, bonds, commodities, etc) are going to be the backbone of bitcoin's future value. You can read all about it in thread linked in my signature.

While you guys are working on stuff like this, I'm trying a very simple "futures" experiment: http://forum.bitcoin.org/index.php?topic=10008.0

[BillyHiggins]

I like your procedure for picking the right trade, its very good.I hope it will provide good help to all the traders.

[killerstorm]

Preview version running on testnet done. See reddit post http://www.reddit.com/r/Bitcoin/comments/hve8n/bitcoin_currency_futures_exchange_preview/ for details.