Hacker News Reader

Comment by ghiculescu | Hacker News Reader

> In private discussions I have heard three counter-arguments, none of which I accept

I think this article skips the most compelling (to me) counter argument.

If there’s a 51% attack, the currency loses all value. Nobody will want to trade Bitcoin or use it as a currency once this attack is exploited. So a 51% attacker is disincentivized to perform said attack because doing so would make the prize lose its value.

Now I write this, I suppose if your worry is a government trying to destroy Bitcoin, then that’s exactly how to do it…

View on HN

> a 51% attacker is disincentivized to perform said attack because doing so would make the prize lose its value

North Korea and Russia heavily use crypto. If either does something super stupid, the Congress could create a credible commitment to repeatedly launching attacks on their payment networks, including Bitcoin. This isn’t some far-fetched threat.

I think the U.S. would be reluctant to do so (although I'd love it if they did), because they hold several billion dollars of bitcoin

> the U.S. would be reluctant to do so (although I'd love it if they did), because they hold several billion dollars of bitcoin

And billions more in private losses. It should be something we’re reluctant to do. But things break in war.

Devaluing bitcoin reduces the incentive to mine it, which I consider a net positive all around

You're right that the article doesn't address that counter-argument, but the suggested means of profiting from the attack is immune from it:

> A successful 51% attack, indeed even a credible threat of such an attack succeeding, would almost certainly sow fear and uncertainty in a wide range of public markets. An attacker could leverage this because they would have a certain amount of control over when news of the attack broke, so they could (for example) take a short position on a portfolio of financial stocks before launching the attack.

> take a short position on a portfolio of financial stocks before launching the attack

This is subject to detection and reversal. FX, credit and rates, on the other hand, could be massively exploited in a risk-on trade.

>This is subject to detection and reversal

Why would it be reversed?

> Why would it be reversed?

Massive short placed right before a sudden financial disaster? It may not be reversed, but it will definitely be frozen and investigated.

I think one could argue that there are probably easier ways to manipulate a market than a 51% attack on Bitcoin.

If someone tried to buy 10s of billions of dollars of Bitcoin shorts, who would take the other side of that deal?

A 51% doesn't get you coins, it lets you spend coins twice. Imagine:

1) borrow a bunch of BTC, 2) use it to buy stuff you want, 3) 51% attack to fork the chain and reverse your spend, 4) pay back the borrowed BTC

What does it matter if the price of BTC falls through the floor afterward?

Of course depending on what you buy, what your lender knows and cares about, and more, you might still face repercussions that make it unreasonable to expect to benefit, but that's enforced outside of BTC rather than by BTC, relying on the systems BTC proponents often explicitly distrust.

You can already spend coins twice, it's just that one of the addresses you send the coins to ultimately won't get them. So people have to wait some time to ensure the transaction fully processes before they can give you the goods in full confidence that they have received payment.

At some point, if cryptocurrencies become more established, it might be financially beneficial for one blockchain if another blockchain becomes non-viable.

The point of a 51% attack is to destroy any such thing as "fully processes".

Okay sure. At best you can do that once, ever. So you’d better buy something worthwhile.

Yeah, "ideally" the cost of the attack can be paid with borrowed BTC; otherwise, it would certainly have to be a very large haul to be worthwhile.

I think this is not a realistic worry for potential attackers. If someone gets control of 51%, there will be a lag time for when people start to truly believe that the currency lost its value. So there will definitely be a huge value gain for any attackers that manage to pull this off.

> a 51% attacker is disincentivized

You missed this:

"An attacker could leverage [a market crash] because they would have a certain amount of control over when news of the attack broke, so they could (for example) take a short position on a portfolio of financial stocks before launching the attack."

> I suppose if your worry is a government trying to destroy Bitcoin, then that’s exactly how to do it…

Andreas Antonopoulos - 51% Bitcoin Attack

https://www.youtube.com/watch?v=ncPyMUfNyVM

Did you read the article?

Please don't comment on whether someone read an article. "Did you even read the article? It mentions that" can be shortened to "The article mentions that".

One approach would be to just rent the hardware, and short the coin.

> short the coin

Unacceptable counterparty risk and scale limits.

Even when buying puts on a regulated exchange like LedgerX, which custodies all the cash that would pay out on the bet?

> a regulated exchange like LedgerX, which custodies all the cash that would pay out on the bet

Bitcoin crashing would take out LedgerX. At that point, you’re an unsecured creditor. There is also legitimate question to them paying out a massive short after something like this.

The exchange is not long Bitcoin, and it holds (edit: typo) all trader assets separately from its own, so no, a crash in bitcoin wouldn’t “take out” LedgerX. You’re just making stuff up now.

> exchange is not long Bitcoin, and it hold me all trader assets separately from its own, so no, a crash in bitcoin wouldn’t “take out” LedgerX

It would take Herculean effort for any crypto business to survive, well, crypto wiping out. Deposits would become worthless, cash would be pulled, credit would vanish as would investment. If markets don’t take them out, their banks and regulators will. It’s the crypto equivalent of Treasuries defaulting.

It doesn't matter -- LedgerX doesn't own those Bitcoin or the customer deposits and never did. Even if the business fails (from e.g. being in debt and the fees drying up), customer assets are held separately from their own assets and would be available for withdrawal.

Since you pretty clearly didn't research before posting, you may not realize that FTX bought LedgerX (in late 2021), and that owner did become insolvent, and regulators did look into the books, and even that didn't prevent options contracts from being settled and assets withdrawn. Guess we found Hercules!

If you're not willing to admit when you're wrong, you'd be doing the site a favor to stop posting out of ignorance -- it's not helping.

> customer assets are held separately from their own assets and would be available for withdrawal

This protects cash and crypto at LedgerX. (Assuming it has few senior creditors.)

Shorts are different. If you place 1 BTC short with LedgerX, it doesn’t hold 1 BTC of cash. Instead, it maintains claims on others. Those claims become unlikely to pay in a crash. That’s LedgerX’s counterparty risk, which does it in, which does you in. Shorts are inherently leveraged in a way longs are not.

> that owner did become insolvent, and regulators did look into the books, and even that didn't prevent options contracts from being settled and assets withdrawn

Yes, standard bankruptcy priority was followed. Also, no leveraged positions needed to be unwound way out of the money.

There is plenty online about why shorting coins on crypto exchanges is stupid. It’s a simplified version of why FX shorts, when done with large anticipated gains, are placed offshore from the target currency.

>If you place 1 BTC short with LedgerX

LedgerX doesn’t support short-selling or margin trading. If you buy a put (the suggestion I made in my original comment), the counterparty sets aside the cash to honor it.

(I think you’re confusing short-selling with a short position, but that doesn’t sound like mistake someone with your bio would make.)

Are you done being confidently incorrect, or can I expect you to pollute more discussions with baseless claims you know nothing about?

Bookmarking to warn future HNers what they’re dealing with.

It’s fine to make mistakes. It’s not fine to be that wrong, with that level of confidence.

I don’t know anything about LedgerX or Crypto trading but do have a fair bit of fx experience so have a few questions about what you are describing.

Doesn’t this just move the counterparty risk away from the exchange and onto another participant in the crypto ecosystem? One that is even more systematically long crypto and one with less regulatory protections around senior debt?

Why would someone be a counterparty to this trade? It seems like they are trading a lot of downside risk to get cash equivalents upside.

I’m sure I’m not understanding something about this setup but it certainly seems worse in all ways than an exchange actually enabling more traditional shorting.

Thanks! You're showing a lot more curiosity and honesty than JumpCrisscross.

>Doesn’t this just move the counterparty risk away from the exchange and onto another participant in the crypto ecosystem?

They have to set the money aside for the (cash-secured) put they're writing, out of their control, so no, you don't depend on any later solvency of that counterparty to honor the put. And, as above, customer access to these assets survived even the FTX bankruptcy, since they're not recognized as exchange assets in the first place.

>Why would someone be a counterparty to this trade? It seems like they are trading a lot of downside risk to get cash equivalents upside.

Same reason anyone else sells cash-secured puts.

>I’m sure I’m not understanding something about this setup but it certainly seems worse in all ways than an exchange actually enabling more traditional shorting.

Except that you, the person taking a short position (by buying puts) actually get a payout in the worst-case scenario, and have effectively no counterparty risk. That's at least one way that it's better.

Also, I don't know what's "non-traditional" about buying puts as a means of establishing a short position. It seems that, like JumpCrisscross, you're being sloppy about the difference between short-selling (borrow an asset to sell) and a short position (any position that increases in value as the underlying decreases).

Edit: Also, one thing that bothers me here is the lack of a coherent model behind the objections. On the one hand, you both want to claim something is impossible, but when I show it is, you want to insist that that way is "worse in every way". What?

> have to set the money aside for the (cash-secured) put they're writing, out of their control, so no, you don't depend on any later solvency of that counterparty to honor the put

There is no legal mechanism for them to insulate that cash from creditors.

I guess I can't say I'm surprised by yet another confidently wrong claim on your part. This is one of the most high-profile cases of bankruptcy, where very determined creditors are hellbent on pulling out every penny from a very obvious fraud. And in this case, we already know that -- right in line with my claim -- depositors at one arm of the debtor did in fact have their deposits legally insulated from the (very extensive) creditor claims.

And yet you are claiming, in this very instance, that that is not possible, even as it already happened. You couldn't be more wrong on the facts here.

Perhaps you could have limited yourself to a lesser claim that, under specific conditions, that you enumerate from actual knowledge, depositors can't step ahead of other creditors. By being open and honest, and revealing the full basis of your belief, you could have said something at least true, even if you were (very honorably) wrong about the applicability to this particular case.

You didn't take that route. You instead went with the tried-and-true "if I say it confidently enough, I have to be right". Sigh.

In the other thread[1], you claimed I'm upset at you for your views on crypto. This is childish and false. I love to hear meaningful corrections to my worldmodel. Your comments aren't doing that. You're instead finding the most solid parts, and baselessly asserting things about them that are demonstrably false. I can't actually learn from that sophistry or meaningfully update my beliefs. No one can.

[1] https://news.ycombinator.com/item?id=35559836

You are really missing what he's saying. That this particular default was subject to a bankruptcy judge means a few things that would bring alarm bells to any professional risk manager. Its not a _good_ thing, its just not the worst thing.

A few off the top of my head:

It went to bankruptcy at all. With more forceful regulatory regimes (eg banking/brokerages) they skip bankruptcy proceedings for the most senior debt holders entirely. The regulator steps in and disburses the funds much faster.

The judge had discretion, there was no legal insulation to LedgerX's creditors. They just had maintained enough arms distance to satisfy a bankruptcy judge to rule in one set of creditors favor over another's.

His point is that unlike in other regulatory regimes, nothing in the law, says that deposits at a CFTC exchange are more senior to any other liability. And the precedence on clearing house default is not extensive.

Most risk managers would view the fact that this judge ruled in their favor as the saving grace to a mistake on the risk management side.

Also, you are on tilt here. Statements like "You couldn't be more wrong on the facts here." are pretty embarrassing when you are misconstruing things so dramatically.

Ah, I misunderstood what you were saying. I thought you were suggesting that the counterparties were passing money between each other. Instead, this _is_ a traditional short position. The counterparty is setting aside the cash, with LedgerX, who is actually a registered CFTC clearinghouse so does have oversight and is (theoretically) watching their operations and they are therefore using properly applied risk management for "other peoples money". But the counterparty risk remains with LedgerX. It hasn't gone anywhere and is still something that would need to be managed.

Its been a few years since I did risk management for a CFTC derivative exposed firm but at the time clearing house default procedures were very much up in the air. Dodd-Frank allowed for clearing houses to be declared as systematically important but only CME, ICE and OCC were designated as such. Has that changed?

Other than that, I'll be honest I don't remember what the debt claims were with regard to clearing houses. Is there some special provision of the bankruptcy law that makes "LedgerX doesn't own those Bitcoin or the customer deposits and never did." true? Because typically this would be up to a bankruptcy judge to decide (as opposed to say custodial assets at a brokerage or deposits at a bank which have senior claims by law). That doesn't look to be what has happened in this case by the way. In this case it looks like a bankruptcy judge looked at the books and said "yep these assets/liabilities balance appropriately so its an easy win to just sell the clearing house to someone else". Which is precisely what 'JumpCrisscross said.

So, I'm not put off by this, because its a message board. But maybe you should look at some of your statements to see if they meet your own standards. You've made a few claims that set my alarm bells ringing as someone who has actively worked in this space.

A few examples "actually get a payout in the worst-case scenario, and have effectively no counterparty risk" which is not true. "LedgerX doesn't own those Bitcoin or the customer deposits and never did" which is up to a fairly complicated legal disposition to figure out. "customer assets are held separately from their own assets and would be available for withdrawal" which is an article of faith.

Generally speaking, "no counterparty risk" is impossible in finance. With traditional finance there is a lot of law, regulation and precedence to help mitigate it but even then actors in the space manage counterparty risk as a matter of course, including by managing their clearing risk. When crypto is involved I think it would be charitable to be even more patient with people who are incredulous about that statement, even in cases like this where it is falling back on normal finance.

Sorry for the long delay to reply, these are good points that are worth addressing.

Yes, you're right, counterparty risk is not completely gone, and it never is. Even for a stock share in Schwab cash account, for all I know, it's some massive, elaborate fraud where the share never existed and I've never compensated for it not being there. Yet most of us would roll our eyes at someone insisting that you "can't profit from future [unexpected] successes at Google" -- even by owning shares -- because "lol counterparty risk".

I thought I was careful about hedging my remarks to allow for cases like this, and it feels like special pleading to scream bloody murder about that (very tiny) risk, when a) it applies to everything, even that Google share above, and b) that clearly wasn't the kind of counterparty risk that JumpCrisscross (JCC) was warning about to support his thesis that you can't short crypto.

Rather, he was claiming that the loss in value of crypto would cascade to your counterparty being unable to honor their side of the short position -- a position, mind you, that is fully backed by cash no longer in their control, and regarded as customer assets by the exchange. He's clearly working from a model in which any such counterparty is using margin, and where the resolution of the contract depends on later sales, with certain price minima, despite my painstaking re-clarification(s) that the short position's model and the exchange do not work like that.

Note in particular his comment "If you place 1 BTC short with LedgerX,", which was never part of my model, and not something LedgerX supports to begin with!

There are many things you can say about that kind of reply, but not "this is a meaningful attempt to engage with SilasX's point", so of course, I'm going to call that out.

(From all I've read, the status of one's deposits at LedgerX is the same as the stock at Schwab -- even if the stock market crashes and Schwab unable to pay its bonds, that Google share is still yours. If you have actual information on that topic, that would be useful to know -- I have a hard time parsing the regulatory filings myself -- but JCC certainly didn't, beyond re-parroting assertions about how LedgerX would definitely crash and use your money to pay off creditors.)

Bottom line -- JCC had many opportunities too root-cause the basis of the disagreement and narrow it down to substantive, resolvable claims about the world, but at every point elected to simply assert increasingly irrelevant points with greater confidence. I've never seen that lead to a meaningful dialogue. And so I wasn't surprised when, the following day, he played the same unproductive game.[1]

As for your point about bankruptcy: I have to remain skeptical that the hyper-savvy risk manager is worried about the Google long blowing up because of the risk that some general market crash would tear down Schwab, reveal that they weren't really holding shares, and then put your claimed Google shares junior to Schwab's creditors -- especially when we have a very recent example of a bankruptcy court electing not to do that (for FTX). If you disagree, I'm happy to hear why.

[1] https://news.ycombinator.com/item?id=35558582

Savvy risk managers don’t tend to use retail brokerages not because of the risk but because other venues suit their operating model better.

That said if they were using Schwab for a long equity position they would not be worried about their shares being used for other creditors payback because the law says it can’t. Share holders have more senior claims than other creditors. Further, even if Schwab was insolvent there is a long chain of custody on equity shares managed by a too big to fail entity (cede) to fall back on.

Other kinds of tradable assets don’t have such protections (such as fx futures or even fractional shares) so when trading these risk managers actively do look at the balance sheet of the clearing partners. In many cases that is a too big to fail entity like the CME but clearing default was a hot topic the last time I worked in the space. I don’t know the specifics of LedgerX held crypto but given it’s a cftc regulated exchange and a bankruptcy judge actually having to rule on this, my guess is those tokens do not have the same protections as a google share at Schwab.

None of this of course matters for a single btc, that’s not enough to get professional risk management involved and if the crypto space went away tomorrow you’d just move along on losing that. But! For someone wanting to put on a large short (think a Soros style bet) figuring out how to manage the counterparty risk is a big part of the trade and it would be very common to not put that trade on with an exchange or otc desk headquartered in the country whose currency you are shorting. Precisely because you’d be worried that the basis for your short thesis would cause contagion that collapsed your counterpart.

I can see how that thinking could be extended to an exchange whose business was focused on crypto. Again, I don’t know the crypto space well enough to take have a meaningful opinion.

Not enough rentable hardware exists on the planet. The article's suggestion that someone would throw billions of dollars into a black hole to "buy up" all hardware globally, get nothing in return, and outpace those who act in their own rational self-interest is just silly.

Reply on news.ycombinator.com