Comment by lkrubner | Hacker News Reader

Comment by lkrubner | Hacker News Reader

For historical perspective, consider the last great panic to sweep the USA before the Federal Reserve was established:

http://en.wikipedia.org/wiki/Panic_of_1907

What is (so far) missing in the world of Bitcoin is a single actor powerful enough to play the organizing role that Morgan played during that previous crash:

"Morgan summoned the presidents of the city's banks to his office. They started to arrive at 2 p.m.; Morgan informed them that as many as 50 stock exchange houses would fail unless $25 million was raised in 10 minutes. By 2:16 p.m., 14 bank presidents had pledged $23.6 million to keep the stock exchange afloat. The money reached the market at 2:30 p.m., in time to finish the day's trading, and by the 3 o'clock market close, $19 million had been loaned out. Disaster was averted. Morgan usually eschewed the press, but as he left his offices that night he made a statement to reporters: "If people will keep their money in the banks, everything will be all right""

This news went up around 2 hours ago and market prices haven't really budged. In bitcoin, these kinds of shocks are almost always instant, and usually happen days or weeks in advance with insider knowledge trading.

Everyone with a brain knew Gox was dying a month ago. In relation to the panic you quote, this isn't a panic over the price falling - it is equivalent to one stock exchange back then stealing your money and going insolvent. Functionally, it is a private company failing, with the collateral that anyone stupid enough to have given them money is pretty much sure to lose it.

But that just means that the business failed, and it doesn't reflect much on the overall bitcoin economy at all. Hell, I was into bitcoin 2 years ago and knew right away MtGox was an incompetent lot that were bound to crash and burn. I also always kept my coins in local wallets and only had USD on an exchange long enough to wire it out or buy BTC that I could then move to my own wallets.

I applaud you on your foresight and you should pat yourself on the back extra good tonight.

But clearly others trusted MtGox. I haven't traded a bitcoin since 2011 so I don't know much about the community past what I'd absorb about anything that's on HN daily. But isn't the claim that MtGox lost like 6% of all BTC in existence? If you think this is going to sail by without impacting the larger community, I think the foresight that led you away from MtGox is failing you now.

People are now $500+ million dollars poorer than they thought they were. That has ripples. That will erode confidence. And my bet is that it will erode confidence.

But isn't the claim that MtGox lost like 6% of all BTC in existence?

If so, then that would further imply that MtGox has lost ~3.5% of all BTC that will ever exist.

Perhaps, however I suspect bit-coin is going to be forced to increase the long term cap at some point in the future as they keep being destroyed. Which might not seem like an issue but it adds a lot of instability as nobody knows how many bit-coins are still in circulation.

PS: Don't forget holding bit-coins does not get you a vote it's the miners that decide how things evolve.

A bitcoin can be divided into many pieces.

There's no reason to issue more bitcoin. The forces of supply and demand will keep the value stable, regardless of the exact number of bitcoins outstanding. Issuing more bitcoin would only serve to invite fears that it would be issued in shady ways, or that such issuances might become routine and inflation would take place. This would only further undermine the currency's perceived staying power.

Why do you think "forces of supply and demand will keep the value stable"? Bitcoin is much more susceptible to huge valuation fluctuations than fiat currency, where central bank can step in to shift demand/supply.

Unfortunately humans are famously loss averse[1] which leads to downwardly sticky nominal prices[2] in practice. In the face of deflation a lot of people just stop buying and selling and hope that prices go back up, or hold out for a nominal offer as big as what they think they "should" get even if the real value is the same.

If your bitcoin economy is full of nerds who really understand the subjective value of money than that's fine. But if you were to start having lots of normal people using Bitcoin and denominating prices and contracts in Bbitcoin than deflation would be a huge problem. As long as most of the surrounding economy isn't using Bitcoin it still wouldn't have the positive feedback loops that caused the Great Depression under the gold standard, though.

[1]https://en.wikipedia.org/wiki/Loss_aversion [2]https://en.wikipedia.org/wiki/Sticky_prices

So can gold. It's not the question of dividing it up.

And weighting, assaying, and using the tiny divisions of gold has a steep upwards cost as the amount of gold gets smaller.

I remember reading that it's only divisible by 10^100.

only

I meant only, because if Bitcoin effectively becomes the leading world currency 10^100 might not be enough. However it shouldn't be hard to change that limit, or to find a way around it.

You have no idea how big 10^100 is.

That is 1.4 * 10^90 (atomic units of bitcoin) for everyone on earth. There are only around 1.2 * 10^12 actual US dollar coins and notes in the world, and only 1.2 * 10^13 dollars if you count all of M2 money supply.

So can a dollar. Admittedly not as many pieces - only 100 of them - but the principle is exactly the same.

My suspicion (I am not an expert on finance) is that (if it remains successful long enough) Bitcoin is going to end up being fractionally reserved by the exchanges—I mean openly so, with the consent of the depositors. That should generate a large increase in the apparent supply of Bitcoins. I mean, gold is already heavily fractionally reserved...

They would absolutely have to.

The impending cap on bitcoin production would dry up the already recession-like liquidity, making it worse than gold as a currency.

And as we saw with the banks, fools who hand over money to unregulated banks don't get their money back. Why would any bank not steal all the money, or get robbed, when it costs nothing to do so?

Nobody knows how many dollars are still in circulation. They have good guesses, but nobody knows about the millions in cash that was lost/burned/buried in the desert. It doesn't matter. Supply and demand doesn't rely on the total number of bitcoins in existence; only the number that are available on exchanges. And with blockchain analysis, we can have a very clear picture of bitcoin circulation - much better than with traditional currencies.

With fiat money, this is not a problem, because that's only one of the available parameters to determine the relative value of a currency. Also, inflation constantly pushes fiat currency value towards zero, so your stashed 1970 dollars are now worth much less. Bitcoin, being deflationary in nature and not linked to anything, is a different ballgame.

The funds were likely stolen, not sent into the void.

Even if they were sent into the void, bitcoins are highly divisible. Just shift the decimal point over a few places and you've got convenient units once again.

Right, but with the blockchain being public, there's a certain parallel between stolen BTC and stolen art.

market prices haven't really budged

Say what? They've fallen 10-15% over the last hour or so (depending on whether you look at Coindesk ot bitCoinTicket) and they're still going down.

In relation to the panic you quote, this isn't a panic over the price falling - it is equivalent to one stock exchange back then stealing your money and going insolvent.

The challenge here is that the Bitcoin system is built on faith. If enough faith disappears, so does the value of Bitcoin. In this manner, it is similar to a bank run.

Note that the 1907 panic and the present Bitcoin situation are fundamentally different in nature. BTC exchanges operate as, well, exchanges and stores -- more akin to a security deposit box with the ability to trade with other boxholders -- not lending institutions.

The other piece of the puzzle is that stock exchanges would allow trading on margin, often as much as 10:1 margin or more (an error repeated in the lead-up to the 1929 crash). Because of transactions, local banks were further implicated.

The trust lost wasn't in the exchanges' and banks' fiduciary responsibility but in their creditworthiness, and what Morgan accomplished was to say "don't trust their credit, trust ours". Not a foolproof gamble, but one which worked. The Federal Reserve takes this one step further by being a lender of last resort: if you're a bank and you need a loan, you can always go to the Federal Reserve's lending window to get more cash (though the FDIC may end up owning you). Stockholders can be wiped out (bank stockholders), but depositors are insured (subject to limitations).

The problem with Mt. Gox is, apparently, one of fiduciary trust or negligence -- it's not clear who's made off with the coin, but it's not creditworthiness that's in question.

Why would a depositor care about the difference between trustworthiness and creditworthiness?

If 50%+1 of the Bitcoin miners agree, couldn't they just mint 700000 extra bitcoin to make the depositors at MtGox whole? The confidence it inspires would probably be worth more than the impact of the inflation, and with Bitcoin prices dropping miners need to support the price to stay above break-even.

Technically speaking, yes. That would kill Bitcoin, though, because the community depends on it having an (eventually) fixed supply rather than it being having "an eventually fixed supply, except when we have a really good reason to do quantitative easing."

Does it really? Could it not just as easily help Bitcoin to become a currency with real-world controls?

I already remarked once that Bitcoin is remarkable democratic in a way; perhaps this is a good time to leverage that...

I know this is somewhat tangential to your point, but a bailout isn't remotely the same as quantitative easing.

Quantitative easing is buying financial assets at marginally above-market price with printed money in order to boost inflation and lower real wages. It's a useful policy (assuming you believe in Keynesian stimulus) because the distributional effect is close to nil, unlike fiscal stimulus or mailing money to specified individuals.

The distributional effect is small because the bank buys a bond worth $9,000 for $9,003 or so. This puts $9,003 in circulation but the person receiving $9,003 in cash receives only $3 of profit. So you've increased the money supply by $9003 but you've given the beneficiaries of QE only $3.

QE is a far more efficient way to increase inflation than fiscal stimulus since fiscal stimulus nearly always results in malinvestment/malconsumption (e.g., building bridges to nowhere, etc) and usually has highly non-stimulative side effects (raising nominal wages of government workers).

A bailout is highly targeted. Inflation is a side effect. The people receiving the money trade nothing of value in return for the money. It's totally not the same policy.

> It's a useful policy (assuming you believe in Keynesian stimulus) because the distributional effect is close to nil, unlike fiscal stimulus or mailing money to specified individuals.

Except that we don't really see QE causing the hyperinflation many feared or the lowering of real wages. If it is intended to do these things, it is a miserable failure.

The fact is that when you dump a whole lot of money into what Steve Keen called "the slowest part of the system" it isn't going to actually have much effect. Velocity in money matters if you want the money to actually circulate.

That's simply incorrect. According to Krugman (back in 4/2013, back when data was not yet released) 2013 will be a test of Market Monetarism (QE to target NGDP), since the QE3 will be offset by fiscal austerity:

...we are in effect getting a test of the market monetarist view right now, with the Fed having adopted more expansionary policies even as fiscal policy tightens.

http://krugman.blogs.nytimes.com/2013/04/28/monetarism-falls...

The results are in:

http://econlog.econlib.org/archives/2014/01/the_parrot_is_s....

By the criteria laid out by QE's detractors, it worked.

Technically speaking, no.

patio11: that is not how the Bitcoin protocol works.

Even if 99% of the Bitcoin miners agree, they cannot create extra bitcoin; it is simply impossible because that is not something that miners control. Every full node (client) on the Bitcoin network verifies the validity of each block. A block that created more bitcoins than allowed by the protocol would be rejected by the clients (and the miners who follow the original rules). The "extra" bitcoins would be on a hard fork and thus invalid to anyone on the main (original) blockchain.

More info: https://en.bitcoin.it/wiki/Weaknesses#Attacker_has_a_lot_of_...

But the protocol can be changed.

If the most common bitcoin clients are updated to allow this "bonus", and at least 51% of the network installs this update, their blockchain will become the authoritative one. Older or other clients will have to accept it if they still want to participate in the chain.

The authoritative blockchain only determines the order of valid transactions, not which transactions are valid.

A transaction that conjures Bitcoin out of nothing is certainly invalid. Patio11 got this wrong.

The validity of transactions is checked by the individual bitcoin clients. After all, something, somewhere must check if they are valid according to the protocol rules! And this client code can be changed to anything at all, as long as most miners agree to update to it.

Transactions that make up bitcoin out of nothing are already used as rewards for mining a block.

Individual clients don't have to use the same client that miners use.

If 51% of miners start using an inflationary version of Bitcoin "Bitcoin-QE", but the clients (exchanges, merchants, wallet providers, etc) don't approve, they won't update the client and use the partition of the network that is secured by the remaining 49% of miners. People wouldn't want to accept the Bitcoin-QE mined on the inflationary branch and the investment wouldn't pay off.

Of course Bitcoin-QE miners could start an attack on Bitcoin in other ways, but that also woudn't be necessarily economical.

It's very hard to tell what 51% will do, but it is clear that you don't have full control of the currency.

You're right, the defecting miners would have to convince people to download a new client, or convince the sites which those people access Bitcoin through to use the new client. Sorry, should have mentioned that, but I thought the thrust of the concern was "Could we just rewrite the rules of Bitcoin to make this retroactively not happen?" And the answer to that is "Yes, but the community would explode."

What you are describing has already happened once, with the rollback and the 0.7 -> 0.8 version upgrade.

No. See https://news.ycombinator.com/item?id=7296376

This is actually impossible. Bitcoin has two security points, the miners (proof of work) and the nodes (clients such as the reference client).

Gaining 51% can allow you to double spend your own transactions, but you can't spend other peoples money or make more coin, because the clients will reject those transactions even if 100% of the miners put it through. The nodes verify what the miners do. Even 100% control does not give you God power. The best you can really do is deny all transactions and thus destroy Bitcoin.

You want a bailout of the (once) rich? Talk about remaking the financial system in the image of the existing one!

Well, what will happen is that there's now going to be great pressure for all exchanges to prove they are not in a fractional reserve state (one of the advantages of bitcoin is that it makes such a proof possible.)

It'll be interesting to see how many of them can provide such a proof... I'm guessing most of them can (but only time will tell)

If they successfully provide this proof, there's no reason to expect a bank run.

Planet Money had an interesting show about this:

http://www.npr.org/blogs/money/2013/12/20/255839292/episode-...

Note that these pledges were CREDIT, not MONEY. A distinction without a difference when it comes to post-Federal Reserve days.

You can't do this with BTC, because there is no credit market at this point.

Everything in BTC is full-reserve, all the time; there is no fractional reserve banking of BTC - it is in your wallet, or it is in my wallet.

What makes you think that online wallets (including exchanges) keep 100% of their reserves? Without inside information, I find it extremely hard to believe that they would just eschew that advantage.

And that was the birth of the Federal Reserve system. I thought bitcoin people were against exactly that...

If bitcoin needs a savior, then it is pointless.

Reply on news.ycombinator.com