Net asset value means the value of assets net of liabilities. Of course NAV changes over time: unlike the 'Bitcoin Fund' which appeared briefly earlier in this thread, and which charges an up front load rather than tapping into the fund's assets, how else would this trust fund itself?


I wasn't trying to speculate about how the agreement's provisions might change in the future, I was trying to clarify what it actually says right now, which is that shares may be redeemed only in whole baskets, and that whole baskets will be redeemed for Bitcoin.


I'm afraid I'm not seeing the big issue here. Shorting anything in volume may cause its price to fall. You can short Bitcoin right now if you want: open up a Bitfinex account, deposit some collateral, and place your sell order.

As to the ETF being "deemed" to be determined by the ETF, the original premise of the article was merely that it would likely become the principal price discovery mechanism, not that anyone would need to deem it that way in order for it to be so.


The Bitcoin market already allows both 'collateralized' shorting and derivatives on BTC/USD. The impact of a broadly available ETF might change this in quantity, but not in kind (except to the extent that options will be vastly better than futures, IMHO, but hopefully you still get my point).


I agree that it may have a profound effect on Bitcoin valuation, but I do not see any actual argument to the effect that a larger, more liquid, more efficient market with the potential for derivatives vastly superior to what we have now would be other than good news for the Bitcoin economy. In other words, I hear what you're saying, I'm just not grasping any argument along with it.

A couple more clarifications of the clarifications, for clarifications' sake...


The same is true of all ETPs: shares can be traded back and forth between different holders all day long, sometimes adding up to a significant percentage of the total shares outstanding. That need not have any impact at all on the circulation of the underlying entity in the broader economy. Likewise, you and I could trade 1 Bitcoin back and forth between ourselves all day long if we wanted.

Is there a potential confusion here between shares being sold (traded from one holder to another) and shares being redeemed (converted from shares into Bitcoins)?


I wonder whether there might be some confusion created by two completely different uses of the word 'collateral'? In the case of collateral for shorting, the word refers to nothing more than something of value which the short seller provides to the entity who is providing whatever it is they are short selling. The short seller provides the collateral to secure their commitment to later cough up the cash to re-purchase whatever it is they bought to sell short and return it to the entity which provided it in the first place. It would make no sense at all for the collateral ever to be identical to the thing which is being sold short, because if the short seller already had it, they could just sell it, not short sell it.

In the fiat world, collateral is normally the rest of the investor's equity portfolio plus cash.

But this use of the word 'collateral' has nothing at all to do with what backs the ETF itself. If I want to short sell some Google, and my portfolio has sufficient value to collateralise my short sale, nobody complains that I am not putting up some Google before I short Google, and nobody suggests that my short sale is somehow going to leak Googleness out of Google or break the link between Google shares and underlying Google.


Hmm, here again, I wonder whether some readers could find this confusing...

If, by "share value", you mean "the value of a share as set by the market" (i.e., price), then generally speaking there is never any such linkage except that which is enforced by arbitrageurs. (Provided that can be done; in the case of closed end funds, the price may vary by quite a bit relative to net asset value.)

If, however, you mean "the value of the underlying Bitcoins held by the trust", then the share value is, by definition, the net asset value of those Bitcoins per share.


On the contrary, redemption of shares by Authorized Participants always yields Bitcoins delivered to the AP by the trust. The passage in the S-1 which you've referred to previously is describing the process which will be followed if the AP fails to deliver a basket of shares to be redeemed by 9:00 a.m. on a particular day of the week, not the other way around. In other words, the passage about collateralization of that commitment, etc., etc., has nothing to do with failure by the trust to redeem shares in Bitcoin, it has to do with failure of the AP to deliver shares when redeeming for Bitcoin.


Naturally, no Bitcoins need to be borrowed from anyone prior to short selling ETF shares, since the requirement is for ETF shares to be borrowed.


The short selling of ETF shares is very precisely limited by the number of ETF shares available to be shorted.

If what you're referring to is actually the practice of naked short selling -- selling something that you have not first borrowed -- then clearly that has nothing to do with collateral, or Bitcoins, or anything else that is specific to this example. Naked short selling is a broader issue, and abusive naked short selling in particular brings with it a whole raft of legal considerations.


If the problem you're getting at is just that, like any other commodity ETF, the trust must hold some of the underlying entity (in this case, Bitcoins) in order to exist at all, then that seems to me an entirely different discussion than all of the above about collateral and redemption and the difference between trading and redeeming.

As I described at length in the separate derivatives article that was linked from the original article, businesses currently do not have any particularly effective ways of hedging Bitcoin currency risk. Let me go right out on a limb here and assert that pretty much any businessperson who wants to deal with any serious volume of Bitcoins is going to appreciate a more robust hedging mechanism than the BTC/USD futures traded on ICBIT.

As I explained in that separately linked article on derivatives and counterparty risk, at the moment this is a huge gaping hole in the Bitcoin landscape, and it acts as a strong disincentive for real businesses to do much of anything involving Bitcoins as an actual currency.

Oh, hey -- that's great (and entirely unexpected), many many thanks for that! 

(Like many folks, I generally use any given receiving address only once, so off I go to swap in a new one...)

I'm not sure whether this will be helpful or not -- so if it isn't, just skip this post entirely -- but as the OP, I feel some obligation to try and summarize a few of the various threads that have emerged here so far...

To recap the original article (Winklevoss Bitcoin Trust May Become THE Price Discovery Mechanism for Bitcoin), if the Winkelvoss trust is approved, there are strong arguments for believing it could wind up becoming the principal price discovery mechanism for Bitcoin, supplanting existing exchanges. In addition, much of the speculative and investment volume currently on existing exchanges could wind up migrating to trade in the trust. The article outlines several reasons why the inadequacies of existing exchanges, combined with the advantages offered by an ETF, could bring this about. The article used gold ETFs as an historical example; representing around 6.5% of global gold demand, gold ETFs do not set the price of gold, but they do exert significant influence.

The original article does not address any questions at all about price manipulation or potential malfeasance on the part of the fund issuer or how it might all be exploited by central banks to harm Bitcoin. The original article's focus was on arguments about price discovery in the first instance; the article also touched on the relevance of the ETF for the eventual introduction of standardized options, and it touched on the primitive nature of trading network-style exchanges as compared to 'real' exchanges with an identifiable counterparty. If there was any speculation in the article, it was focused on that last issue, and the idea that the ETF may incentivize someone to step up and create a 'real' exchange.

Through the course of the thread, we've covered the area of share basket creation and redemption via Authorized Participants and revisited the fact that the trust will be backed by real live Bitcoins; we've also covered why individual investors won't be cashing in their shares and getting Bitcoins in return, and why that isn't a problem.

We've touched on what differentiates the Winkelvoss trust from the Exante fund, and also from the recently announced 'Bitcoin Fund', and touched on the distinction between a hedge fund and a currency fund.

'NewLiberty' in particular has introduced the idea that the ETF could provide an efficient means for central banks or other potential enemies of Bitcoin to cause problems, and if I have followed the thinking correctly, has also suggested that there could be some way for the ETF to break the link between shares being traded and the underlying entity backing those shares.

So, I think that's roughly where we are now.

In the latest few posts, if I've understood correctly, 'notme' has wondered about shorting the ETF with USD collateral, 'halfawake' has commented about derivatives as distinct from ETFs themselves for risk management, and both 'molecular' and 'notme' have brought us right back to the original topic again and wondered whether the ETF really could become the principal price discovery mechanism anyway.

For my part, my thoughts on all of those last bits are closely connected. As I see it, a great advantage of an ETF would be that it opens up buying and selling and short selling to a vastly wider audience, and it provides an underlying security suitable for the introduction of standardized derivatives. (It makes little difference what is used as collateral in ordinary short selling, by the way: collateral is merely what you are putting up to show you have the resources to buy back the share when the time comes, and the share still has be borrowed from someone before it can be sold. I.e., the share being sold short is not the collateral. Even in the relatively rare case of naked short selling, someone still has to be pay the piper at the end of the day; analogously, you can write naked calls all day if you want, too, but you will still be forced either to buy them back or to cough up the underlying should that call expire in-the-money.) And assuming that standardized options eventually became available on the ETF -- as tends to happen for anything with significant volume -- the Bitcoin economy would immediately gain hedging mechanisms far surpassing the capabilities of anything we currently have available.

For my part, I would have imagined that folks concerned about market manipulators would be more concerned about the relatively low volume exchanges we have at the moment, where just a few million dollars here or there can exert quite an influence on exchange rates. Generally speaking, I think boosting volume and increasing market breadth makes it harder, not easier, to fiddle with free markets.

Hmm, I'm pretty sure I wouldn't agree with all those agrees...but we can at least agree to disagree on that. 

As far as I'm aware, the existing ways of managing volatility are pretty weak -- and lack the liquidity necessary for hedging any kind of larger position. I took a look at this awhile back in an article called Bitcoin Derivatives, Liquidity and Counterparty Risk and concluded that the situation at the time was severely limited. But I'd be very happy -- eager, in fact -- to learn of some new addition to the fray.

(The last I knew, when the likes of Coinbase or BitPay referred to "managing volatility", what they meant was "avoid volatility by not holding Bitcoins any longer than necessary".)

The eventual availability of options on something like the proposed Bitcoin ETF would be an absolute godsend in this respect.

Yes, I'm sure you're right on this: this is not aimed primarily at the folks who are already buying and selling Bitcoins anyway (and who would not want to hand over a fee to someone else for the privilege!).

Just to clarify first: the proposed ETF provides an initial ratio of 5 shares to 1 Bitcoin, and this ratio will change very slowly over time, as the fund's fees are deducted from the Bitcoins it holds.


Since the S-1 -- remember, it is only an S-1 -- specifies just that the AP, Sponsor and Trustee must agree on appropriate collateralization, it seems to me premature to speculate on the nature of the collateral they might agree upon. Possibly more to the point, though, is that to my mind, this proviso sounds more like a mechanism intended to promote the orderly operation of the fund -- and to prevent it becoming bogged down by Trustee nitpicking over the precise time of day or day of the week that a Basket of shares is to be delivered -- than a mechanism intended to provide a disguised back door for market manipulation. What we're talking about here is the AP promising to deliver the shares even if they may wind up being late, and the Trustee being willing to remove Bitcoins from the trust and discharge them to the AP on the understanding that the corresponding shares will in fact be delivered to it. As far as I can tell, qualitatively speaking, this is little different from an arbitrageur shorting the shares of any ETF while buying the underlying on the spot market.

And many thanks for promoting such a friendly environment on the forum -- I appreciate it! 

You're very welcome -- just 'Greg' is fine by me, though. 


The S-1 doesn't specify who the Authorized Participants might be, or what requirements might be in place for becoming one, but in theory, the answer to your question seems to be yes, given this from page 4 of the S-1:

"The creation and redemption of Baskets require the delivery to the Trust, or the distribution by the Trust, of the number of Bitcoins represented by the Baskets being created or redeemed, the amount of which will be based on the combined NAV of the number of Shares included in the Baskets being created or redeemed. The initial number of Bitcoins required for deposit with the Trust to create Shares is [10,000] per Basket."


Without peeking at the specific futures contract you mentioned, I can't say whether the analogy holds, but I can say that this creation and redemption mechanic is exactly what makes the intermediaries well positioned to provide market making services and to engage in arbitrage that ensures the price of Bitcoins via shares closely tracks Bitcoins purchased directly via the same currency as the shares. It's the job of the Authorized Participants to gauge market demand and create or redeem accordingly, and it's very much in their interests to keep everything operating smoothly and efficiently. Far from being a bad thing, as some folks might infer due to their privileged role in being able to swap shares for the underlying entity, their involvement is actually what make the whole thing possible.


Yep, I agree -- although there might be plenty of risks associated with the whole thing, I don't think the design itself is flawed in such a way that it would contribute to those risks. In other words, yes, there might be all kinds of problems and all kinds of unforeseen consequences, but those won't be the result of someone's having just set it up crazily in the first place.

Without speculating about who might want to do what to which currency pair and for what sort of reason, I'd just like to clarify some of the mechanics of how ETFs operate.

From the standpoint of individual investors, shares or units in ETFs are traded -- i.e., bought and sold -- rather than created or redeemed. For present purposes, 'creation' refers to the issuance of a share and the placing of the underlying entity into the issuer's safekeeping to back that share, while 'redemption' refers to the reverse process of removing the underlying entity from the issuer's hands.

In the case of index-based ETFs and commodity ETFs, creation and redemption only occurs in relatively large baskets. The sheer size and value of those baskets means that individual investors never actually wind up redeeming them. The creations and redemptions still occur in the background, however, courtesy of 'Authorized Participants', as they're called in the case of the Winkelvoss trust.

For a real world, happening-right-now sort of example, consider the SPDR Gold Trust: it really does add gold to its custodian's vaults when investors add money to the fund, and it really does liquidate gold from the vaults when investors "cash out". But this process is intermediated typically by large financial institutions who agree to perform this role. When an individual buys one unit in the gold trust, it does not mean that one specific piece of gold is immediately added to the vaults. What actually happens is that the individual is really buying not from the trust itself, but from an intermediary, and they are buying a unit which has already been backed with gold added to the vault. (Otherwise, the intermediary wouldn't have had the unit to sell in the first place.)

If it is approved, investors will find a very similar process in the Winkelvoss trust as the size of the fund expands or contracts in response to demand. The Winkelvii will not be rushing out to buy or sell .2 Bitcoins each time an individual investor buys or sells one share in the trust, and they will not be swapping out .2 Bitcoins in response to individual investors saying they'd like their one fifth of a Bitcoin now please. However, the APs will be causing exactly that sort of process to happen on a larger scale as they handle baskets of 50,000 shares at a time and either distribute them to buyers or collect them from sellers.

Without the involvement of these intermediaries providing a buffer between the issuer and the trade in ETF shares, it would be difficult for the thousands of different ETPs out there even to exist.

It leaves quite a bit to be desired in its current form, but the ICBIT exchange also provides futures on Bitcoin vs. USD. Protection of a merchant's cashflows from currency exchange risk is a textbook usage case for futures and forward contracts.

Seeing the "Bitcoin Market Capitalization" described instead as the Bitcoin monetary base expressed in USD could make a nice improvement. 

Hmmm, I may have missed something, but with one exception, I have a hard time reading much value into this particular setup. It's designed as a straightforward currency fund, not a hedge fund, and thus all the comparisons about how great their fees are, relative to hedge funds, are apples to oranges comparisons. How could there possibly be a performance fee, for example, given that there is no performance whatsoever that is internal to the fund: 1 BTC = 1 share, always and forevermore. The QP restriction eliminates huge swathes of ordinary investors, while the prohibition on BTC redemptions means arbitrageurs cannot force the price of a 'share' (in quotes deliberately) to track the BTC/USD exchange rate. The principal marketing message I take away is hand-holding: we'll act as caretakers for something so complicated that you really shouldn't worry yourself about it...i.e., we'll buy Bitcoins for you and keep good backups, and we'll even call and wake you at 3 a.m. (their example, not mine) if we think you should know that the exchange rate is moving significantly.

The one value I could identify is one offered by any currency fund: denomination in the native currency provides balance sheet transparency for businesses or individuals who want or need exposure to BTC. (They introduced a clever bit of FUD referring to IRS audits and how they want to just blend in so nobody takes any notice of them.)

Again, maybe I missed something, but this seems to me like it has many disadvantages and very few of the advantages of an ETF. On the contrary, it strikes me as cherry picking designed to part the very rich with .5% of their 100K -- plus the BTC/USD bid/ask spread, which lurks in the background...

I think it's important to distinguish the Exante fund from an ETF. Among other things, the Exante fund does not trade on an exchange and cannot have standardized derivatives built around it. It does nothing to enhance liquidity in the USD/BTC trade, it cannot be shorted, and it is not available for arbitrage. By contrast, an ETF is open to all comers and is suitable for the eventual introduction of standardized derivatives. It can be shorted, it is available for arbitrage, and it provides a paradigm example of adding liquidity to the trade between the quoted currency and the underlying commodity.

Relative to the existing fund, the proposed ETF seems to me like a very different kettle of fish, and if approved seems likely to have a qualitatively and quantitatively different impact on the Bitcoin economy.

Wow, thanks for this -- it's exactly the kind of discussion I'd hoped for with that anonymity & privacy article. (Not sure whether it's better to start a separate thread for that?)


And even for those of us who do have access to the modern accoutrements of electronic banking, I think Bitcoin still offers tremendous value -- people can actually do what they want with their money, rather than waiting on permission from a bank behaving as if it's their money.


Touché.


I certainly agree that it could make it more efficient, and it could be great it if it did...but I'm not yet convinced that it will.


I couldn't agree more.

That's strange, since SpeedyBitcoin.co.uk uses Lloyds as their receiving bank. Maybe their gripe was with using a personal bank account for business transactions at all, rather than using a personal bank account for Bitcoin business transactions?

In case it's of any help, there's a separate list of banks reported to be hostile to Bitcoin; although that doesn't answer your question, perhaps it could save you some time pursuing avenues that won't work out?

Yep, I see what you mean. I think the person who wrote the zxcvbn checker works for Dropbox, and he just intended it to illustrate some of the pitfalls of common ways of measuring password strength, ways which could inadvertently give users bad advice. As you've just demonstrated, this seems like a case where it could do exactly that -- give bad advice.

As I understand it, this is the strength guesstimator which they are now using on the Dropbox registration page. (See his original article for more details.)