Hacker News Reader

What you say is true and applies to ANY commodity market (or in fact any non-centrally regulated market): FX, precious metals, oil, textiles, fruits, etc.

I can set up my own independent exchange to trade, say, gold in the same way there are thousands of online gold exchanges. Then I can artificially inflate the gold prices on my exchange, using fictitious money (like the willy bot did), leading customers to come to me to trade their gold for high prices, and therefore impacting the worldwide market price of gold... at least a tiny little bit for a tiny little while... until customers find out that they can't withdraw their fictitious money from my exchange, so my sham and its impact ends right there. This is what happened to MtGox in February 2014.

The bigger the market you are attacking is, the less time it takes for the sham to be discovered, because the tiny percentage of people who try to benefit from the fictitious high prices quickly overwhelm your ability to fake solvency. A huge market like gold would pretty much suffer no effect due to its size. And as the Bitcoin market continues to grow, the hypothetical impact of such a sham should diminish over time.

I wanted to add that naturally it is very costly to run this sort of attack. Because as you are artificially manipulating the prices, you have to keep paying out the customers' arbitrage profits. This is why willy caused MtGox to lose tens of millions of dollars in the end. Mark Karpeles was simply clueless or naive in thinking willy would help him. So don't do it, or you will lose money :)

In summary, I do believe regulation can and already does help BTC. They bring finance professionals in the market who have a clue (unlike Karpeles). They bring trust. They bring more trading volume. These are all things that make BTC less manipulable, hence less manipulated.

Is there a difference when it comes to Bitcoin that makes it unique? In your example the majority of gold exchanges are regulated, arent they (if not, lets use a stock exchange as an example instead)?

In that situation the bad actor is unregulated, and the others are all regulated. With Bitcoin the opposite is true - only one exchange is regulated. Does this matter?

1) it's not unique to bitcoin, happens in all markets.

2) it's very expensive to do. For example if you inflate prices at an exchange at 10% of market price, then you must pay that rate sooner or later when inevitably people sell at a 10% premium and cash out. If you run a ponzi, it'll collapse inevitably.

3) honest regulated exchanges that interface with other honest regulated aren't affected. The market is affected, but not the individual exchange. i.e. if Mt. Gox was still alive and this new Winklevoss exchange Gemini didn't trade with them, it wouldn't affect Gemini if Mt. Gox collapsed. Individual traders who traded at both Gemini and Mt. Gox would be affected, but that's the risk of trading on a crappy unregulated exchange. But the honest exchange itself isn't affected, it cannot result in a catastrophic failure like you say. (feel free to elaborate on that)

tl;dr: not unique to bitcoin, doesn't happen often because manipulating the market has a price tag and a risk tag, and while markets are affected by manipulation (obviously) one bad exchange cannot spawn failure or security problems or ponzis or insolvency at other honest exchanges from arbitrage efforts of its users. It just doesn't work that way.

An exchange's shadiness can only affect other exchanges in proportion to its market share. If there was price manipulation on Mt. Gox and it spilled over to all other exchanges then that was because Mt. Gox was the dominant platform for trading bitcoin.

FX in general is very difficult to regulate and there are a lot of shady firms and exchanges out there. However, I think it's hard to argue that the price isn't reflective of the true exchange rate. Over $5 trillion/day changes hands at a tight spread and very low volatility. Finance is based on trust and eventually legit venues get reinforced through a feedback loop.

>>One exchange's shadiness will always affect every other exchange.

I don't think that is correct. A shady exchange will be avoided by market participants if it significantly deviates from a fair price.

Has everyone already forgotten the LIBOR and FX price fixing scandals?

Markets are inherently shady, and there is no such thing as a 'fair price.'

If you can fix a market, you will. That goes double if you can fix it even more effectively with a price-fixing cartel - especially a state-sponsored one.

(Arguably this is what QE really does, only QE is quasi-legal.)

The only protection is good regulation. Markets on their own tend naturally to criminality and/or instability.

Karpeles was typical in that he was playing a variation on the usual games. But he wasn't professional enough to play them well.

Also, I'm not quite seeing the logic of promoting a currency system that has been proven to be trivially hackable.

(Although I suppose if you can get enough people to believe that next time will be different and your exchange will be the exception, there's certainly money to be made.)

> Markets are inherently shady

> The only protection is good regulation. Markets on their own tend naturally to criminality and/or instability.

Regulation like how Argentina's government mismanages their peso to disastrous effect (people have to trade currency on the black market to get by)? Or how some governments have pulled the "print money" lever until the currency is hyperinflated to zero value?

Governments are inherently shady.

> Also, I'm not quite seeing the logic of promoting a currency system that has been proven to be trivially hackable.

That's false. An exchange going down due to criminal activity is in no way Bitcoin being hackable. Bitcoin was not hacked, and MtGox criminal activities are a reflection on them, not on Bitcoin.

Historically, the common reason for a large discrepancy in price between exchanges is that one of them is insolvent. So the price is fictional in that case anyway: you can deposit funds there and "trade", but your principal and your profits are already gone by that point, anyway.

The issue with shady exchanges, they all seem to have the issue of collapsing overnight. So even if not playing by the rules will raise the prices, that might be counteracted by the major risk of you losing everything on the exchange. Or having everything seized by some gov't.

There's a reason they're called pyramid schemes

> There's a reason they're called pyramid schemes

Just because someone call something a pyramid scheme, it doesn't mean it is one.

This is completely wrong.

There have been large periods where the degree of shadiness on certain bitcoin exchanges has led to large, and persistent price discrepancies. For example, during the huge bull run of late 2013, there was a large and consistent price premium for bitcoins on MtGox. The reason was that MtGox was not processing USD withdrawals in a timely way, but was processing bitcoin withdrawals. If you had tried to arbitrage the price on MtGox vs. Bitstamp or BTC-e, you would have found your money tied up in 3 week limbo while the withdrawal was processing.

Your point is true, but unrelated to mine. I was talking about price manipulation and arbitrage in a scenario where there is no delay on withdraws/deposits.

To put it another way, if price manipulation is happening, you typically won't notice a price discrepancy. The reason is because of arbitrage. Since price discrepancies are lucrative, people will use arbitrage to take advantage (and hence eliminate) any price discrepancy. The end result is that the price was manipulated across all exchanges, even though only one exchange is being shady.

You're starting from the assumption that the manipulator will be effectual.

It is reasonable to expect that arbitrage will tend to make manipulation harder. The basic goal of arbitrage is to recognize open bids that are 'wrong' according to some estimate of the broader market, and to then act on the profitable ones. So in the presence of arbitrage, the manipulative bids that are against the market will quickly disappear.

If a large segment of the market is simply fraudulent, all bets are off. Fraud may well be a better description of what was going on at Mt. Gox than price manipulation (for instance, it isn't really simple price manipulation if their stated customer (dollar) account balances were 10x greater than the paper currency that had ever been deposited).

Theoretically, the market should "price in" the risk at different exchanges.

If I can sell a bitcoin for $200 on a perfectly stable exchange, or for $202 on an exchange where there's a 1% risk I will get nothing, then by switching from the stable exchange to the risky exchange there's a 99% chance of +2 and a 1% chance of -200, for an expected benefit of 0.992+0.01-200=-0.02 so it wouldn't make sense to use the risky exchange. The exchange with a 1% risk would have to offer a price of at least $202.02 for it to make sense to switch to them.

Of course, knowing the relative chances different exchanges will collapse is the hard part!

Your logic should also apply to real international financial markets, which seem to work well enough, so clearly this worldview is missing something

One of the things I'm really looking forward to is USD/EUR denominated sidechain assets eg. https://tether.to

At present, arbitrage between different exchanges is easy if you are moving bitcoins, but moving USD is very time consuming and requires all sorts of verification. When the exchanges start accepting certain dollar denominated crypto assets you will be able to move USD between exchanges as easily as moving bitcoins. Liquidity and market efficiency and prices will all become extremely similar across markets. Any bad exchange will rapidly be shown up as a fraud as they will collapse within hours as any price differences or manipulation will be attacked by speculators. If anyone stops paying out deposits they will be called out within hours (vs. months and months when mtgox stopped paying out USD) and will rapidly lose credibility and collapse.

I once looked into a pegged side chain type coin to make trading easier and it's impossible to do with certain currencies, especially USD. You become responsible for all trades with your token/coins, so if you sell 20,000 USDcoins to user A and they sell the coins to User B without following KYC/money laundering rules the feds come after you demanding ID for User B.

Same goes for CADcoin or GBPcoin, maybe EURcoin would be different but unlikely. You would have to set it up as a centralized service where all transactions req proven identity paper trail which in the end is just as cumbersome as using regular banks.

There's tons of other regulations like preventing your USDcoin from being used for online gambling, which is why Paysafecard must sell crippled PINs in US.

You're definitely right, especially right now as we're dealing with a sprawling market consisting of multiple exchanges with no laws or regulations governing them. However, I believe the Winklevoss twins (given past history) are looking to transform a burned-out shell into prime real estate. It may not be an appropriate example, but I look to GitHub: Is Git really that much better than Mercurial? If GitHub was called MercurialHub, would we all be using Mercurial right now?

(note: this wasn't an invitation to argue about whether Git or Mercurial is better...just trying to point out that sometimes, one company can have a massive effect on a community, especially a community that's small and devoted like Bitcoin but has no real banner to follow)

I chose Git over Mercurial because Linus made it, not because of some hosting startup that can be pwned by random guy with no previous experience in hacking.

You can try to arbitrage on an unregulated exchange, but then you're assuming the greater risk of getting Goxxed.

There are plenty of examples in financial markets where lower risk is correlated with lower returns, and this may be another.

http://www.thebtcindex.com/

Look at the spread there. Bitcoin is not short of coders who can automate obvious arbitrage opportunities, so spreads like that directly indicate just how damn hard it actually is to get ActualMoney out of them.