Bitcoin Market Update 4/4

Hi Everyone,

Since my last update on 3/31, markets have been quite volatile trading between $420/BTC and $500/BTC.

While third party-payment processors in China have been disallowed from servicing bitcoin exchanges there, there has not been any official ban of bitcoin by the central bank and there have been no exchanges cut off from payment processing through their respective banking partners.  This seems to be more of a crackdown on third-party payment processors than bitcoin itself.

Given the IRS rulings on bitcoin as "property" some big winners in the bitcoin markets are undoubtably selling off part of their stake to pay out taxes owed before the filing deadline in April.

Another interesting idea I read from Vinny Lingham, CEO of Gyft, is that the bitcoin price is currently weak because Bitcoin merchant adoption is outpacing consumer adoption. When merchants cash out of the bitcoins their customers give them, this depresses the price faster than normal broad-based demand for bitcoin can buy it up.  This makes a lot of sense and could play a big part in explaining why the price has dropped and stayed depressed recently.  You can read Vinny's full article here:

I was originally planning to talk more about the more tactical aspects of orderbook dynamics today, picking up where we left off last time but I'll do that next week instead.  This week I wanted to address the new Michael Lewis book, Flash Boys: A Wall Street Revolt and the recent New York Times article about it which you can find here:

The premise is that HFT shops were "front-running" orders on the market by reading orderflow on one exchange and issuing orders to other exchanges faster than the rest of the market.  For example, Brad sends an order to buy 100k shares of ABC through his trading platform and routing service for execution on BATS (50k shares) and NYSE (50k shares).  BATS is located closer to Brad's trading desk so BATS receives his order first.  An HFT shop sees Brad's order on BATS and quickly sends an order to NYSE to buy some shares of ABC and repost them for sale at a penny or few pennies higher.  By the time Brad's order reaches the NYSE, he can no longer buy 50 shares at the price he intended to since the HFT shop already bought them and now has to pay a penny or a few pennies higher to the HFT shop to get his 50 shares.  The HFT shop has effectively taxed Brad a few pennies per share.  Rinse and repeat throughout the day and the HFT shop reaps in the profits.  Or so the story is told.

Unfortunately, the story is much more complicated than Michael Lewis suggests.  Lewis would have you believe that the markets are rigged and that the big guys are out to get the little guys.  While it's true that the "retail guy" on average loses money to the "professional", his narrative is disingenuous.  In the capital markets, it's not that one side is AGAINST any other side; it's that every side is FOR themselves (I'll come back to this distinction later once we get to the topic of microwave and laser communication networks).  It just so happens that retail flow and clunky institutional flow are the least sophisticated participants who end up paying out a premium to their faster and shrewder counterparts.  Some of this premium is simply the price for execution, the price for liquidity.  Whereas once upon a time, tick sizes were 1/8 and 1/16 of a dollar, and retail flow had to pay up to the human market makers, human prop traders, and retail brokers to express their opinion, now tick sizes are a penny, spreads are smaller and retail flow plus the rest of the gang, to some extent, have to pay up to HFT shops.  In other words, retail flow has always been getting "screwed" in the sense that they have to pay someone.  It's just that now, that "someone" has changed and that "someone" also collects a few premiums from the other players.  Also, back in the day, a firm could pay an exchange a large sum of money for a seat which would give them direct market access, and trade much faster than any retail person could through their broker.  Now, a firm can pay a telecoms company a large sum of money to use their fiber optics line to the exchange which is a few milliseconds faster than the next guy and trade faster than the rest of the market.  It's gotten to the point where fiber optics (fast, high bandwidth, not affected by the weather) are too slow and companies have been building out microwave dish networks (faster, low bandwidth, affected by the weather) and laser grids (faster, high bandwidth, not affected by the weather) to go even faster (See  At that point, it's just a bunch of HFT firms trying to outdo each other.  In other words, it's not that the HFT firms are against retail folks or the rest of the market, they are effectively attacking each other and trying to eat each other's pie too.  In fact, the real winners of this whole market shift toward speed are the telecoms and technology companies building out these maximum speed communication networks.  Effectively, these telecoms companies are "extorting" HFT firms in the sense that they can approach HFT firms and say, "Hey why don't you pay us to use our network which is slightly faster than anything else that exists?  Oh, by the way, we are also renting it out to all of your competitors.  If you don't use us, you will lose to your competitors and get pushed out of all your speed-sensitive strategies."  This leads to an arms race to the bottom where the arms dealer has all of the leverage.  Also note that the game here isn't about being faster than retail flow since everyone has always been faster than retail even with fiber optics, it's about being faster than each other.  As such, profit margins for HFTs have been shrinking and in a free and open market, profits will continue to be driven down to near zero as they crowd each other out and pay out operating costs to the telecoms companies which own the communications networks.

Also, there is a problem with Lewis' examples of "front-running" an order on one exchange by getting to the next exchange faster.  Market-makers by their natural behavior almost always quote more than the size they are actually willing to do, distributed over multiple exchanges with the intention of canceling their excess orders once their desired size is filled assuming they don't get filled everywhere at the same time.  The reason they do this is to ensure they do not miss an opportunity to trade regardless of on which exchange an opposing order arrives.  For example, Joe is a market maker and is willing to sell 100 shares of ABC.  He puts 70 shares for sale on BATS and 70 shares for sale on NYSE.  Once he gets hit for 70 shares on BATS, he cancels part of his ask on the NYSE so that it now only shows 30 shares.  This is very normal behavior, does not involve any type of "front-running", and would result in vanishing liquidity on one exchange after an order fills on another, the same outcome as what Lewis says is the result of malicious activity.

There's another problem with Lewis' example: how do the HFT firms which "front-run" orders know the full size of the order?  If an HFT firm sees Jim buy 50 shares of ABC on BATS, how does he know if Jim intends to buy 50 more shares at the same price on NYSE or 20 shares or 100 shares or no shares?  By not knowing Jim's full size, the HFT firm cannot with certainty know how much to buy on NYSE and repost at a higher price.  All that being said, of course the HFT firm still has an advantage in that he can buy up some amount on NYSE and repost it and it would be positive EV even though it exposes the HFT firm to some risk.

All that being said, I am not strongly in favor of HFT.  I'm simply making an observation that the question at hand is more complex than at first glance.  To keep it fair, here's one argument against HFT.  From an efficiency standpoint, HFT is wasteful.  Like I mentioned earlier, HFT firms are in an arms race to the bottom as they pay up more and more to stay on the save footing as each other.  They are all trapped in the Nash equilibrium defect-defect outcome of a Prisoner Dilemma situation.  Everyone would be better off if everyone was slower but each individual player has an incentive to be faster which causes everyone to try to be faster which causes everyone to be worse off.  If there was some enforcement mechanism whereby each party would be slower and be guaranteed that everyone else was as slow as him, everyone wins.  Unfortunately, in practice, such enforcement is difficult.  Government regulations are not hard restrictions in the same sense that the laws of physics are hard restrictions.  People will always find a way to game the system to be slightly faster.  For example, even if the government mandated that exchanges could only accept orders through a copper wire, distance to the exchange still matters.  If all HFT servers were mandated to be equidistant from the exchange servers by a 100 meter wire, individual server processing speed still matters (how about ASICs and FPGAs?) and, in a minor and pedantic way, even the coiling/straightness, temperature, and quality of the wires matter.  It's hard to guarantee a completely fair and unbreakable system in general and especially through regulation.

I'd like to end with an analogy to sports.  Most baseball fans are rather indifferent to the use of steroids in the sport.  I've even heard the argument that it makes for a more exciting and watchable game.  But much beyond that, virtually no one complains about the fact that professional baseball players can buy better gloves, shoes, and bats than other professional baseball players.  No one complains that a tennis player's racket costs thousands of dollars or a golfer's club is too well-designed, precise, and light-weight.  Trading is also a game where those with better equipment are at an advantage.  So long as the rules are clear and no one is violating them, all is fair in the game.  We do have the option of changing the rules of the trading game, which I am not against, but if we do, we should take care to lay out the rules carefully and with every consideration taken into account lest we create loopholes which make things worse and also ensure that those rules are enforceable in as precise and all-encompassing a way as possible.

Kevin & Team Buttercoin
Bitcoin Trading Made Easy | 

Buttercoin currently in Private Beta, if you'd like access right now apply here

    P.S.: Whether trading should be called a game or not is purely a semantic distinction. I don't mean that trading is a game in the frivolous sense; I mean that it's a game in the game-theoretic sense.  And if an activity or system can be aptly represented in a game-theoretic way, we should consider the incentives and payoffs of its participants and in its architecture and keep them in mind should we try to design a better system.  In game theory, "mechanism design" and "implementation" are the sub-fields which study these systems and their construction.

    Bitcoin Market Update 3/31

    Hi Everyone,

    This week bitcoin traded between $560/BTC and $580/BTC before tumbling to $440/BTC-$460/BTC amid rumors of another crack down by the People's Bank of China.  Although the rumors have not been confirmed, certain news sources claim to have seen a PBoC document circulating which effectively bans bitcoin transactions in China.  Some argue that the rumors could very well be true while others say this is just more of the cyclical ban/no-ban FUD from China.  There are yet others who say that this is a ploy by government and media insiders to manipulate the bitcoin price for their own portfolios.

    There was another rumor earlier this week that gox have, in their control, 670k of their "lost" BTC.  If that is the case, it would put further downward pressure on the markets as unexpected supply gets reintroduced to the markets.  This kind of market news, whether it be that coins are lost or found, always has two countervailing forces.  First, a supply shock directly affects the price up/down depending on whether coins are lost/found respectively.  Second there is a loss/gain of confidence in the bitcoin ecosystem when coins are lost/found which drives the price down/up.  While the first effect is symmetric between positive and negative supply shocks, the second effect is not.  The gain in confidence when coins are found is much smaller in effect than the loss of confidence when coins are lost.  This suggests that the markets are overly optimistic about exchange counterparty risk likely because people are used to the ecosystem and infrastructure of mature exchanges and project those expectations onto this relatively fledgling industry.

    Today I'd like to talk a little bit about the sort of information a person could glean from looking at the structure of the limit orderbook, particularly the dispersion of the limit order sizes at the same price level.

    • First measure: Given a total size of 100 bid at some price level, it is a stronger "up" signal to see two bids than one bid e.g. (50,50) is better than (100).  In general it is stronger to see multiple bids than a single bid given that the total size is the same e.g. (33,33,34) is better than (50,50).
    • Second measure: Given a number of bids at the same price level, it is a stronger signal to see equally dispersed sizes rather than lopsided sizes e.g. (50,50) is stronger than (99,1).  A simple mapping function that captures this relationship quantitatively and gives an ordinal ranking to size structures with the same number of elements is the geometric mean (as opposed to the arithmetic mean).  For example f(50,50)=sqrt(50*50)=50 and f(75,25)=sqrt(75*25)~=43.3 and f(99,1)=sqrt(99*1)~=9.95 hence f(50,50)>f(75,25)>f(99,1).
    As you can see, there's quite a bit of useful information in the orderbook a trader could consider to forecast short-term price fluctuations.  To go further down the rabbit hole, consider the following:
    • How do we create a tradeoff between a size structure which is strong by the first measure but weak by the second or vice versa.  For example compare (10,10,80) with (50,50).  (10,10,80) has more elements than (50,50) (first measure) but is also more disperse in its sizes (second measure).
    • How does queue ordering affect short-term expectation?  How does a size structure of (25,75) compare to (75,25) assuming the first number represents a bid/ask higher up the queue at the same price level than the second number?
    • How about size structures across multiple exchanges?  How does (50,50) on a single exchange compare to (50) on one exchange an (50) on another exchange?
    These are all good questions that I hope to discuss further in one of my later Markets newsletters.

    Kevin & Team Buttercoin
    Bitcoin Trading Made Easy | 

    Buttercoin currently in Private Beta, if you'd like access right now apply here

      Bitcoin Market Update 3/21

      Hi Everyone,

      This weeks markets fell from the $630-$640/BTC range to around $570/BTC.  The sell-off started earlier this week on the Chinese exchanges amidst rumors that the government would be banning all bitcoin transactions.  It turns out the rumor was false and the subsequent headline was removed from the news site.  Instead, the Chinese central bank will be issuing new regulation on bitcoin which, although better than a ban, is still negative news for the short-term.  Hopefully, the markets can rebound within a few days.

      In other news, close to 200k of gox's bitcoins have apparently been "found".  Some believe that gox always had those coins.  Even before the announcement, through blockchain analysis, some internet sleuths found sizable bitcoin holdings in an address known to belong to gox.  I mentioned in my 3/7 update, a bifurcated pattern emerged from this address produced by a 60-40 recursive splitting function coinciding with leaked gox code.  It's likely that gox withheld the "finding" of this wallet for two weeks before making a public announcement.  Why?  There's a lot of interesting theories which you can read about on reddit or bitcointalk if you are interested.

      In any case, finding these some 200k coins is certainly good for those who lost coins on gox but the actualization (as opposed to remaining probabilistic) of this supply shock can also explain the recent dip in price as 200k "new" coins will enter the market sometime in the future.

      Following Warren Buffet's scathing comments about bitcoin, someone has made a 1000 BTC bet on bitbet that BTC will outperform Berkshire Hathaway over the next year:  A few things to note here.  Bitbet functions on a parimutuel system (like in horse racing) whereby all bets are pooled together and then the odds are determined based on the proportion of those betting "yes" versus those betting "no".  Contrast this with the fixed-odds betting you might find with a Vegas sports bookie.  Parimutuel systems protect the bookkeeper from having to make the market on the odds of a bet by letting the market make it for him.  The downside for the bookie is that he will likely not get as much action since there is little incentive to bet until the last minute.  Bookies can correct for this incentive problem by giving earlier bets more weight in the payoff and later bets less.  Still, in total, action is shallower when a parimutuel system is used instead of a fixed-odds type system even with a time-weighted incentive structure.  Ultimately, parimutuel systems are used only when the book keeper has little to no idea where the true odds lie.  In horse racing, parimutuel systems are used because the bets are not boolean in the sense that there are multiple horses.  While it might be possible for a bookie to give odds on two-horse or three-horse race, multiple horses make the odds difficult to lay.  Since the bookie takes no risk in a parimutuel system, the "juice" (also called vigorish) or premium to the bookie is usually smaller.  Compare 10-20% juice for a fixed-odds bet versus 4% juice for a parimutuel bet.

      Kevin & Team Buttercoin
      Bitcoin Trading Made Easy | 

      Buttercoin currently in Private Beta, if you'd like access right now apply here

        Bitcoin Market Update 3/14

        Hi Everyone,

        This week markets have been trading between $620/BTC and $650/BTC.  A relatively calm period after the mtgox storm.  Given the current market sentiment, I see a rally in the near future, perhaps the very beginnings of a fourth great bitcoin bubble.  The type of news coming out about bitcoin this week suggests signals organic growth and awareness of bitcoin and reminds me of earlier periods of bitcoins history which were followed by strong market rallies.  The news has a flavor of being "disperse" in the sense that there are many things happening in many unrelated areas.  For example,
        • Xapos is coming out with a bitcoin vault with built-in insurance.
        • Bitcoin surpasses Western Union in transaction volume.
        • emailed a pro-bitcoin article to its users.
        • Congressman Jared Polis also wrote a pro-bitcoin email that's been getting passed around.
        • New York is in the final stages of creating Bitlicense regulation.
        • Conan O'Brien mentions bitcoin on his show.
        • Warren Buffet says bitcoin is a mirage. (negative news)
          • Arguably all news is good news and as normal Americans see more about bitcoin and cryptocurrency in the media, they become more comfortable with the idea.  At least it begins to seem less and less foreign/alien.
        • Wikipedia founder, Jimmy Wales, receives a good deal of bitcoin donations after publishing his personal bitcoin address.
        • Ebay files a patent for a bitcoin currency exchanger.
        • Rumors of a big wall street firm hosting an internal (employees-only) event about bitcoin that gets completely booked within 10 minutes.
        • Dorian Nakamoto's story continues to hit the mainstream news.
        This pattern of news where it is very disperse and sort of self-propagating has historically signaled a change in market sentiment from neutral to positive.  Also, the boom and bust cycles of the bitcoin markets have gotten shorter and less severe (i.e. the cycles have shorter periods and lower amplitude).  The market now overreacts both up and down to a lessened effect and also more quickly "resets" back to normal market conditions.  In less than a week after the largest exchange in the world goes bankrupt, the market has settled on a relatively stable price in the current range and volatility has almost dropped to normal levels.  All of this leads me to believe that we are at the beginning of a steady rally.

        Today I also wanted to talk about bitcoin storage solutions backed by insurance (e.g. Xapos, Elliptic Vault).  To me, this doesn't make sense.  While it's great for marketing and plays well into the consumer's perspective, logistically, any company which underwrites insurance on a bitcoin vault is taking on a type of risk that requires the insurance premium collected for underwriting it to be unreasonably high.  This is due to a few reasons.
        • The information is strongly asymmetric between the vault storage company (VSC) and the underwriting insurance company (UIC).  What expertise does the UIC have on cryptography and best practices in this emerging industry?  The VSC knows much more about just how secure their vault is.  So for the UIC to price the risk premium of underwriting the vault, they have to add an additional premium for adverse selection.  What they lack in expertise, they must pass on as extra cost to the insurance buyer.
        • The actual tail risk of losing bitcoins in a vault has historically been significant.  A VSC like Xapos charges 16 basis points per year which implies a risk-neutral expectation of losing the whole vault (assuming the whole vault was insured) of 1 in 625.  So if the actual risk of losing the vault was 1 in 625 per year (is it?) and Xapos paid their underwriter the entire 16 bps premium (took no profit margins themselves) and their underwriter was fine with that (the underwriter also has no profit margins), then they all break even in the expected value sense which brings me to...
        • Considering expected value by itself is not how companies subject to limitless downside, like UICs, think.  When you make a million every day and lose $364m once a year, your expected value is near zero and slightly positive but your skew risk is very high.  If you have $100m in assets, you would much rather prefer making a million plus a thousand every other day and losing a million every other day than the previous situation since, in the previous situation, if you go bankrupt, you cannot continue business.  So the UIC has to charge an additional premium for skew risk over the risk-neutral premium, and their profit margins.
        • UICs for retail markets like life insurance and health insurance can collect premiums from largely independent parties thus diversifying their risk.  UICs for bitcoin storage vaults cannot since there are only a few VSCs that exist.
        • Finally, if bitcoin truly goes mainstream and the vaults get huge, the insurance loses effectiveness due to the counterparty risk of the insurance company.  If the vault is lost and the losses are more than the equity of the insurance company, the insurance company defaults and the claims are not fully paid out.  What insurance company could underwrite the gold at Fort Knox?
        To summarize, (1) the insurance company knows a lot less about security than the vault company causing there to be asymmetric information in the underwriting process, (2) the actual risk of losing the vault implies a rather high insurance premium (would you pay, say 10% yearly, for that kind of insurance?), (3) insurance companies have to be compensated for skew risk causing the premium to be even higher, (4) insurance companies cannot diversify since there are very few vault storage companies, (5) if the vault gets big enough, who insures the insurance company?

        Ultimately, Elliptic Vault does not insure the whole sum of their customer deposits, only a fraction for a hefty cost.  Rumors are that Xapos also is only insuring up to a certain limit and that they are underwriting it themselves which would then explain why the premium is only 16 bps.  In that case, the structured product these companies provide is a hybrid between full insurance and no insurance.  There might be a market for this; I just happen to think the premiums will be too high for the amount covered.

        Kevin & Team Buttercoin
        Bitcoin Trading Made Easy | 

        Buttercoin currently in Private Beta, if you'd like access right now apply here
          After re-reading my last update, I feel compelled to clarify one point.  I may have come across as against the idea of insured vault storage solutions but that isn't the case.

          A fully insured vault storage solution would be good for the bitcoin ecosystem, and any company that can make the numbers work will do very well. The solution isn't obvious to me, and anyone looking to use such a service should make sure to do their own due diligence.

          That said, Wences Casares is one of the smartest and most capable people in the bitcoin space. Of companies that could figure out a good solution to the insurance problem, Xapos would be at the top of the list.

          Bitcoin Market Update 3/7

          Hi Everyone,

          This week markets rallied from about $570/BTC to over $700/BTC before settling back down at the $630/BTC to $670/BTC.

          As I'm sure you've all heard, Newsweek found themselves a fake Satoshi Nakamoto.  The unfortunate fellow has the media chasing him around as he adamantly denies he is the real Satoshi.

          In other news, an mtgox address with 180,000 BTC has been splitting its coins into smaller wallets in a binary tree structure (60-40 randomly-weighted branches).  This is behavior is in line with leaked gox code showing a 60-40 recursive splitting function.  The intended use for this function is to create smaller wallets to minimize the amount of unconfirmed change which gets sent back every time coins are disbursed.  This suggests that they are preparing to disburse their coins back to the customers.  Whether it plays out this way remains to be seen.

          Recently, I've been getting pretty deep into a paper which describe how to calibrate a Hawkes process (a self-exciting process) to market data as a way to separate price changes due to exogenous events from those due to endogenous events.  Trades can be modeled after this random process whereby some trades are triggered by exogenous events like news or earnings while other trades are simply triggered by trades themselves (endogenous).  This captures a phenomenon that we observe in real markets which is that trades tend to cluster together.  You can think of it this way: some trades are real expressions of opinion while other trades are merely reactions to those real trades or reactions to reactions etc.  Sort of like a rock thrown into a pond creating ripples.

          So why is this interesting?  The ratio of "reactive trades" to "total trades" is called the branching ratio and gives insight on how saturated the market is with game-playing strategies as opposed to fundamental trading, event-driven trading or even purposeful hedging.  With the advent of high-frequency trading, this branching ratio has risen dramatically over the years from .3 in 1998 to over .7 in 2012 for E-mini S&P 500 futures, the most traded futures contract.  This means that over 70% of all trades on this asset are reactions and ripple effects to real market sentiment.  As a side note, the effect of HFT can also be seen in the decreasing average volume per transaction.  For example the average size of a WTI crude transaction dropped from 26.6 contracts in 2005 to 1.7 contracts in 2012.

          In any case, the trading game has changed a great deal over the years.  A higher branching ratio means that it's more important now to outplay the other traders and algo bots than it is to interpret relevant information and also, a great deal of profit can simply be made by reading order flow without regard to the fundamentals of an asset so long as we trade quickly.

          And that finally that begs the question, how "reflexive/reactive" are the bitcoin markets?  My guess is that the branching ratio is very low and that the market is not at all saturated with these game-playing strategies.

          I've attached the paper referenced as well as another one describing a trading strategy based on a Hawkes calibration.  Also see:

          Kevin & Team Buttercoin
          Bitcoin Trading Made Easy | 

          Buttercoin currently in Private Beta, if you'd like access right now apply here



            Recently, I've happened upon a machine learning algorithm called Random Forest based on a randomized "growing" of Decision Trees. Among machine learning algorithms, it currently rivals support vector machines in predictive accuracy but also trumps it in interpretability. The Random Forest algorithm is also extremely fast without parallelization and absurdly fast with parallelization (since all trees of the forest can be grown simultaneously). Also, with most models, variable selection is particularly important to reduce the effect of noise. However, the Random Forest algo automatically filters irrelevant selector variables so any dataset, no matter how filled with arbitrary or irrelevant variables, can be passed in without compromising the accuracy. And since it is so computationally fast, the slowdown due to large numbers of selector variables will be negligible.

            You can read more about Random Forests here:

            Closely related to Random Forests is the concept of Boosting, but they are not exactly analogous due to Random Forests having the additional step of using random selector variables per tree of the forest. You can read more about Boosting here:

            You can read about Decision Trees here:


            Mt. Gox is rapidly losing market share to Bitstamp and others. Gox is down from 48% to 43% and Bitstamp is up from 18% to 20%.  Also, Gox withdrawals for USD are backlogged at least a month, so people are artificially bidding up the USD/BTC price there with their dollars in "limbo". Bitstamp could be reflecting the "truer" price of USD/BTC. BTC-e has the reverse problem of not processing USD deposits so their USD/BTC price is bid down to artificial lows.

            In other news, Bitcoin difficulty is about to skyrocket at the next readjustment: Given the past week where the average hash rate was much greater than the difficulty, this implies strong incoming demand for Bitcoin because the price remained stable.  We can think of the excess supply introduced into the BTC economy beyond 25 BTC every 10 min as the integral between the hashrate and the difficulty stepwise function.  Given that supply is flooding into the BTC economy and the price has remained stable and gone up a little, I surmise that either demand has followed pace with an upward trend or that miners are hoarding their minted coins even more than they were already.


            India's currency, INR, has devalued 2.1% against the US Dollar yesterday in the largest drop since 2011. With inflation at 6%, demand for Bitcoin should be rising in India where there is a fear of another devaluation. The Central Bank has been raising interest rates causing a contraction of the money supply further slowing the economy with the intention of keeping the currency value afloat (to little effect). Bond markets have dried up as foreign investors have taken out over $10 billion. This market seems prime for disruption by Bitcoin.

            In other news, the Fed has barred Germany from expatriating their own gold supposedly held in Fed vaults.  It's not surprising that paper gold is starting to lose value as physical delivery becomes increasingly important. If there is ever a run on the bank for physical gold, it’s possible that some people might find themselves empty handed. And hopefully the gold bars that are actually delivered to people aren't filled with tungsten: In either case, there's always Bitcoin.

            Bitstamp now has narrower spreads and a deeper top of the orderbook than Mt. Gox. As long as Gox fiat withdrawals are backlogged and delayed, Bitstamp will continue to take Gox's market share.

            Bitcoin holding strong at $115+. Hashing rate continues exponential growth with difficulty adjusting periodically.

            A few interesting IPOs happened on the current de facto Bitcoin stock market. Will be watching those closely.

            I've also been doing some research on this idea of measuring market efficiency in capital markets with Kolmogorov Complexity / Lempel-Ziv Complexity. Kolmogorov Complexity is an information theoretic measure defined as the shortest algorithm required to specify an object (e.g. string). Unfortunately, Kolmogorov Complexity is incomputable and serves more as an abstract concept. Lempel-Ziv Complexity, however, is computable and is implemented by many file compression methods in software. A paper I read which used this measure on price returns of various capital markets found some interesting results. Their conclusions were that the S&P 500 is 99.1% efficient, Dow at 95.4%, NASDAQ at 75.4%, while a market like the Colombo Stock Exchange is only 10.5% efficient. I’d be interested in what the LZ measure says about Bitcoin price returns and market efficiency.



            Last Thursday, NASDAQ halted trading for a few hours due to technical difficulties. It just goes to show that even the biggest players aren't immune to hiccups.

            More gold bars filled with tungsten found in New York.

            Bitcoin trading volumes broke record highs last Wednesday in tandem with a jump upward in price. Looks to be due to a large buyer (possibly institutional) rather than many small buyers.

            FINTRAC (Canada) declares that Bitcoin services are not Money Transmitters. Seems like Canada, Germany, Hong Kong, and Singapore are good targets for Bitcoin businesses.

            Labcoin on has gone up almost 200% in the past week. It seems the market is agreeing that this ASIC production house is legitimate and can deliver.

            Hashing rate and difficulty has continued to grow exponentially throughout the past week until today when the hashing rate peaked and then fell slightly.


            Bitcoin is at new highs since the crash in April.

            Tradehill has shut down and the IAFCU has discontinued banking services for Bitcoin companies due to regulatory pressure.

            Crypto Financial on Havelock Investments is doing surprisingly well.  Since their meager IPO, some private investors have come forward to buy out the remaining public shares and the price has been holding strong.

            New IPO on Havelock. wants to be a quote-driven, in-person Bitcoin exchange (compare this with order-driven Valuation is less than $150k so this might be a good investment opportunity solely because of the low valuation. Their business model is somewhat questionable since they are just charging 17$ per year to be a dealer on their site. They also charge 1% of notional for Bitcoin escrow but even with the escrow service it's still not completely safe for both sides of the trade since the fiat is transacted in person, without escrow.  At least this will be interesting to watch just because of the low cap.

            I've been running some analysis on how large trades affect local, rolling volatility and in turn the width of spreads in the Bitcoin markets. This is to sort of "proxy" how long it takes for the markets to "reset" to "normal conditions" after a large move. Markets seem to reset after about 2-3 hours after a trade of size $10,000 given a Gox-like liquidity environment.  This can be a pretty good estimate of the upper bound of how much a market-making strategy could absorb as well as some indication of how much fiat can be remitted through bitcoin for a given time period.


            A flash crash happened today, likely created by interlocking algo bots or two parties triggering each other’s stop losses in cascading succession.

            Last Wednesday, Bitcoin tanked from mid-$140s to mid-$120s. Many believe this is due to the Tor network experiencing lots of lag and many hidden services on Tor going down (DDOS?). Some people suspect this is related to the doubling of Tor nodes a few days ago. Most likely a government or some botnet.

            Activemining jumped 70% in under 30 minutes this week on news that they contracted 28nm ASIC chips from eASIC and are currently in production. Stock tanked eventually settling at pre-announcement levels. Rigs should be able to produce over 24 THashes/sec.

            Poland coercively moved half of private pension funds to state-owned funds. This was to pad the government's balance sheet so the debt-ridden country can borrow even more.

            Whistleblower reveals 16 top banks and hedge funds get key economic information 10 minutes to 1 hour ahead of everyone else.

            Dealcoin jumped 20% after IPO on Havelock. Now back down to about 7% up.

            BTC Growth first report due for Sept 16. Will be interesting to see this.

            Apparently this site,, allows users to code their own algo trading bot and backtest it on historical Gox/Bitstamp data. Mostly based on technical analysis and no real tools for HFT.

            I read an interesting paper on the clustering of trades in time in Bitcoin markets.  Rather than use a Poisson process for order arrival intensity (used for things like call arrival times at a call center), they use a Hawkes process which is an extension of the Poisson process but allows for a clustering effect (used for things like seismology, crime epidemia, wildfires). Their explanation for trade clustering is that some trades are based on actual exogenous information while other trades are simply reactions to the aforementioned trades. By what they call the branching ratio (65%), we can see what portion of trades are "real" rather than reactionary. Also interesting is a paper they cite as a reference which goes into detail on how to apply this Hawkes fit to trading strategies, particularly market-making strategies.  One hurdle is that the Hawkes MLE is of O(n^2) time-complexity. Even speedup techniques (memoization) do not shortcut the MLE by much. Some research has been done on applying GPUs (throwing hardware at the problem, which is ok for polynomial time problems) to the MLE.  Another alternative is to use GMM (generalized method of moments) which is less accurate but faster.





            Some evidence of tape painting in the markets.

            Bitstamp to require verification for all types and amounts of withdrawals as of Sept 30, 2013.

            Armory raises $600k.

            Labcoin had some huge swings (down then up) over the weekend.

            Activemining trending down this past week.

   and Crypto Financial resettling at IPO levels.

            Many BTC denominated IPOs happening in China.

            BTC-Growth had it's first monthly report showing 1.403% growth. The fund manager, Greg Mulhauser, made a good long play on Ukyo.Loan on Bitfunder. Ukyo.Loan is a callable (at face) and putable (at 110% of past 30-day average price) bond issued by Bitfunder's founder which yields .05% daily. Greg sees the bond as an indirect way to invest in Bitfunder - which I agree - and I see Bitfunder doing well into the future.

            NEOBEE, a Cyprus bank and payments processing business, will have their first tranche virtual IPO tomorrow at discount price for XBOND holders who can convert their XBOND bonds to shares of NEOBEE based on a face value of .001 BTC to a discounted .0025 BTC share price. The second tranche is available to the public on Friday at .003 BTC per. Very likely to sell out immediately in first tranche and likely to sell out immediately in second tranche. This is due to the XBOND market trading at well above callable levels (XBOND is callable at 110% of face, as in .0011 BTC and is currently trading at .0012 BTC). This seems initially dangerous because the bonds could be called before the conversion to NEOBEE shares occurs, thus incurring losses of .0001 BTC per share for any XBOND holders. However, the market believes that this is unlikely since the XBOND issuer is unlikely to screw the bond holders in the first ever XBOND conversion option given that the XBOND issuer and NEOBEE have this agreement. I agree with the market in this diagnosis. Ultimately, this implies a high demand for NEOBEE shares as people are willing to pay a premium to get a discount (strange but exactly as it sounds).  Essentially, the implied price of a NEOBEE share is .003 BTC plus the BTC equivalent of the call risk of XBOND from now until tomorrow's IPO. In either case, this is a stock to watch. NEOBEE is raising $3mm for 60% of their equity.

            Coinbase reduces instant buying to 10 BTC per week. They possibly got into some trouble recently since hedging and rebalancing those instant buys are harder in fast moving markets.  Yet, this risk management method is still certainly better than requoting their customers.


            Markets rallied strongly on Bernake's "no tapering" message about QE. Subsequently, USD dropped and gold gained. In the BTC markets, the price stayed horizontal but since it is priced in USD, BTC effectually lost some modest value.

            Aggregate volume on the Chinese exchanges has surpassed the volumes of USD/BTC on Mt. Gox, Bitstamp, and BTC-e combined.

            $300k+ forced refunds from BFL clear.

            Winklevii talk about Bitcoin at a hedge fund conference to mixed reactions.

            Dec13 and Mar14 BTC futures on ICBIT showing strong contango. Since the cost of carry on BTC is negligible, this behavior is strange. 30% returns over 6 months seems like free money. It could be that the market is pricing-in either the devaluation of USD or some default risk on the part of ICBIT. In the former case, nominal gains on BTC denominated in USD over this period are equal to real gains of 0 from a purchasing power perspective. In the latter case, contracts into Mar14 may never settle if ICBIT shuts down before then (possible) or is a scam (unlikely). Insolvency is unlikely since ICBIT does not function as a clearing house even to unwound, margin-called orders which get closed out randomly against winning positions on the other side.  Also there is some consideration for the cost of carry on BTC being positive in the sense that most people do not know how to properly secure their bitcoins. Most likely, the contango is pricing-in all some of the above reasons in some proportion with most of the weight on ICBIT possibly shutting down or being forced to shut down. Otherwise, there is just free money here.

            Labcoin had a scandal where they missed their expected date to start mining. Labcoin posted some pictures of their chips and boards which looked a bit off. People are wondering if they are a scam and so their stock took a big hit. Some people are saying that Labcoin is intentionally looking suspicious to drive the price down so insiders can buy-in at cheaper prices.

            Some members of the board on Activemining resigned and sold their stake in a public announcement lambasting the CEO and citing irreconcilable differences, particularly regarding his approach to PR. Resigning board members say the 28nm eASIC chips are still great but the CEO is driving the company to the ground. Stock tanked pretty hard.

   IPO on Havelock was a record-breaking success with IPO shares selling out within 3 minutes. Price jumped 50% almost immediately on the open market.

            BTCTC closing down due to regulatory issues. All stocks on BTCTC tanked following the announcement. Many issuers froze trading on their symbol.


            Steady appreciation in the BTC markets.

            BTC China has temporarily waived trading fees.

            Labcoin running into serious issues. Big controversy about insider's selling shares during the lock-in period as a large block was sold that did not come from public shareholders. Also, management has ties to a former scammer and people believe the whole company is just floating on vaporware. Share price has been obliterated dropping 90% in a week sitting at 1/4th of its over-subscribed IPO price.

            Second Market releases its Bitcoin Investment Trust to accredited investors and institutions. Fees are 1.5% for depositing after Jan 1, 2014, 1.5% for withdrawing, and 2% management fee per year.  Makes sense only if the convenience is worth the high fees.

            Dealcoin not doing well. Trading at 42% of its IPO value.

            CryptoFinancial not doing well either. Trading at 65% of IPO value.

            Casinobitcoin dropped 25%, now at 3x IPO value.

            NEOBEE sitting at close to IPO price.

            Bond markets in Bitcoin doing much better with Ukyo.Loan sitting at close to callable level and XBOND still sitting at above callable.

            BTC-Growth second report due tomorrow.

            Some are saying Atlantis was a scam as no funds have been returned since closing down.

            Cyprus-style wealth confiscation through bail-ins may start happening in other countries if and when banks fail.

            US CDS is ticking up.


            Silkroad was seized and DPR was taken in by the Feds. BTC markets crashed about 20% before rebounding about 10-15%. Altogether, a little surprising that the price recovered so quickly and so strongly. This implies that Silkroad is not a major factor for Bitcoin's value which is a good signal. 26k BTC was seized and if the Fed decides to liquidate it, BTC price is expected to drop significantly.

            Activemining has moved from BTCT to Bitfunder.

            Labcoin value continues to evaporate as it becomes clearer that they are a scam.

            Bitfunder no longer allows US traders to participate. All other traders must get verified. Bitfunder stocks tank due to massive selloff by US traders.

            For the first time ever, short-term LIBOR is under short-term T-Bill yields. The market is implying that the default risk of the US government is higher than the default risk of banks. This has some validity as the debt ceiling deadline approaches. Given Congress' inability to negotiate a budget causing governmental shutdown, the market thinks it's possible that Congress will be locked on the debt ceiling debate as well and cause a technical default. Personally, I think Congress will make the deadline and not cause technical default. Technical default has more far-reaching consequences than the government temporarily shutting down, as technical default could throw the world economies into an unknown area. Anyone holding USD or US debt will be affected.  Arguably, default and inflation are two flavors of the same deal.  Ultimately, it makes no difference to a lender whether he doesn't get paid back or he gets paid back in something worthless.

            For the past few months, front-month gold futures have been in backwardation as persistently as weeks on end.  Given that there are positive carry costs to gold, it makes no sense that spot gold is trading higher than future gold...unless people think that there is real counterparty risk to a futures contract not settling on settlement. But futures contracts are cleared by clearing houses so if any single counterparty defaults, the clearing house will settle with the other side of the contract. So the fact that gold is in backwardation means the market thinks that even clearing houses might fail to settle which means we’re in bad shape. Physical gold today can trade at a premium to paper gold tomorrow because of two reasons that I can see: 1) There is a productive use for gold now due to a temporary shortage of products which use gold as an input to its production or 2) paper gold might become worthless as gold fails to be able to be delivered on expiration. Since the market value of gold is orders of magnitude higher than it's use value, case 1 is never the situation since not even severe shortages in products which use gold would move the price enough to put gold in backwardation. If front-month gold futures continue to stay in backwardation for an extended period of time or deepens, and if back-month gold futures go into backwardation as well, negative consequences like gold-hoarding and fiat inflation could appear.


            The recent run up in price has been massive.  This could be attributed to several reasons:

            • A division of Baidu has started accepting Bitcoin for its services. To be fair, Baidu is a huge conglomerate of small divisions and the larger company has made no statement for or against Bitcoin.

            • There has been huge buying pressure in China and price movements recently have been happening there first, before they are reflected in the USD markets shortly after. Fierce competition among exchanges in China has caused some exchanges to offer promotional, fee-less trading.

            • SilkRoad shut down and the price recovered very quickly from a small crash. This points to strong fundamentals in Bitcoin as the market is suggesting that bitcoins have value beyond black market use. This signal, in turn, might cause an even stronger run up in price. Also, if the Feds do not liquidate the seized bitcoins, the supply shock to the Bitcoin economy should cause the price to go up at least a few dollars.

            • SecondMarket has been constantly buying up BTC for it's investment trust. Opening BTC as an investment to accredited investors may have increased overall demand to a new untapped market; that is, a market of wealthy investors who don't know how to acquire bitcoins for themselves.

            • Political deadlock about the debt ceiling might have fueled fears that the US has a non-negligible chance to undergo technical default. I personally think this is unlikely as it is in both the Democrats' and Republicans' interest to not default. Ultimately, the problem is more about debasement rather than technical default.  Inflation and debasement of the US dollar is the final consequence whether it be from default or from staving off default through loose monetary policy. Already there is talk among the BRICS countries to trade oil in something other than USD. If the US loses its petrodollar status, demand for USD will take a huge hit.

            In the past two days, there has been a correction downward from the recent run-up.  My guess is that given the knowledge of the last two bubbles and given that the most recent prior bubble is still in fresh memory, this current run-up was likely to pop at a lower proportional peak than the last two bubbles.  This is because people are now more aware that bubbles happen and everyone will be trying to duck out before the pop. This regresses into a "guess 2/3rds of the average" type of game since each person wants to duck out - not before when they think the peak will be, but before they think everyone else thinks the peak will be.  See:

            SilkRoad coins have been moving around. The Feds might be thinking of liquidating them after all, possibly through an auction. They have also seized another $20m worth or so.

            Top hedge fund manager Novogratz from Fortress Investment Group has personally endorsed Bitcoin.

            CryptoFinancial, a Bitcoin stock that IPO'd on Havelock, has bought out Havelock.  Given recent troubles with Bitfunder and BTCT, I'm sure Havelock was happy to sell while they are still operating. CryptoFinancial, having strong ties to the Panamanian banks, will be able to get all the licensing and compliance done to make Havelock a legitimate stock exchange. Altogether, this seems like a win-win provided that CryptoFinancial can carry out their side of the plan.


            The markets have been rallying very strongly this past week breaking the all-time high price of $266.  Currently, BTC/USD sits at $288 but has gone as high as $297 on Bitstamp and $320 on Mt. Gox.

            • Some of this rise could be due to all the positive press from the Dublin conference.

            • Some of this could also be attributed to the buying frenzy in China right now. BTC China has recently passed both Gox’s and Bitstamp’s 30-day trading volume.

            • Bitcoin made the front page of the Money Section of the Wall Street Journal yesterday. (This could translate into at least another few percent of rallying in the next few day.)

            • For the first time the Federal Reserve has acknowledged Bitcoin.

            • IBM Executive Architect thinks Bitcoin could be as transformative as the World Wide Web.

            • People have been buying and selling over $100,000 through a Vancouver Bitcoin ATM which has only been live for a few days.

            • SilkRoad 2.0 launched.

            In the Bitcoin equity markets, nothing terribly exciting:

            • back down to IPO levels.

            • Havelock confirmed to be bought out by CryptoFinancial which itself is back up to IPO levels.

            Recently, there's been an academic paper going around about a potential attack based on selfish-mining:

            I thought I'd take a moment to address this.

            Mining cartels could theoretically mine a disproportionately larger number of blocks per hash rate than their honest counterparts. There is also incentive for individual selfish miners to join the cartel to maximize their own rewards. The paper has diagnoses this problem situation.

            The good news is that this sort of theoretical attack is rather impractical in reality. First, this attack happens over a long timespan and, as the cartel starts growing in size, there might be significant backlash from the community against cartel members. Plus since there is plenty of warning time as this begins to happen, the community can react properly. As an example, if BTC Guild started doing this, many miners would leave the pool since some of them would be ideologically against anything that would harm or set back Bitcoin. Furthermore, BTC Guild itself benefits more from the continued survival of Bitcoin rather than some small excess profits in the short run.  Also, some small changes to the Bitcoin protocol by the head developers would make this much harder (e.g. when two chains of same length are found, randomly select which one to mine on instead of mine on whichever one is received first.

            So while this attack is technically possible, it still requires coordination on a large scale (e.g. large mining pools) and, even then, the incentives are not entirely conducive to making the attack practical. Ultimately, I'm not that worried. In any case, if it begins to happen, we will have plenty of warning and more than enough time to react.


            Markets continue to rally this week breaking all-time highs and currently sits at $408/BTC. Buying pressure from China continues to push the price up with some hiccups and small crashes here and there.

            In one of my last updates, I talked about the diminishing proportional size of Bitcoin price bubbles. Today, I want to talk about the fractal nature of bubbles. Essentially, there can be small bubbles forming within a larger bubble which can, in turn, be a part of a series of bubbles in an even larger bubble. My reasoning for why this can and does happen is because different types of investors and speculators have different timeframes by which they judge whether the price is "too high" or "too low". For an investor/speculator with a long time-perspective, we can consider the 2011 bubble up to $36 and the 2013 bubble up to $260 to be inflection points at big bubble peaks. Alternatively, for extremely short term investors/speculators, price movements even slightly beyond the bid-ask bounce range can be considered an ultra-mini-bubble. And for investors/speculators with time-perspectives in the middle, they perceive mid-sized bubbles.

            With that said, I believe we are in the rising part of a big bubble with some room to grow. There's a couple of factors I would consider in attempting to call the top (keep in mind that this is all speculation). First, the second great bitcoin bubble peaked at a over 7 times the price of the peak of the first great bitcoin bubble price over a time period of 2 years. The latest "bubble" if it is a bubble is starting just 6 months after the peak of the second bubble. Going by that and and going by the exponential adoption of Bitcoin, it seems reasonable that the price of Bitcoin should peak between $450 and $500 if bubbles are of the same proportional magnitude. If we believe the bubble should peak at a lower proportional magnitude, we can guess at something like $350 to $450. And then, there should be some consideration for the block halving event in Nov 2012 whose effect is not obvious from just the price. Going further on that train of throught, we could argue that that market cap rather than price should be the chart we are looking at in analyzing bubbles since it factors in the block halving effect on Bitcoin supply. Also, consider that if buying is constant and there is a supply contraction, price increases more quickly generating more press about Bitcoin which, in turn, drives greater demand due to people buying who wouldn't have found Bitcoin otherwise. One could also argue that since Bitcoin's supply schedule was determined from the very beginning, all expectations of block halving would have been factored into the price from the very start and that block halving does not constitute even a discounted supply shock.

            But there is still something to consider where this bubble differs from earlier ones. For the first time, a country besides the US is at the forefront of pushing up the price - namely, China. Given that China currently is a place where there is much capital but few good options for investment, many people are betting on Bitcoin for lack of alternatives. Since much of the fresh money from China does not have recent memory of the previous bubbles, these investors could push the bubble up to a greater proportional peak than previous bubbles while Americans push back on the price by selling for fear of the bubble popping. "Meta-reasoning" would suggest that these sellers would realize that the Chinese do not have recent memory of the last two bubbles and, in turn, not actually selling into the buying pressure.

            This, of course, is all speculation on my part.

            Hedge fund manager Raoul Pal puts Bitcoin long-term price target at $1,000,000 and I can't tell yet if this overly optimistic or overly pessimistic.


            Earlier this week, the markets were highly volatile. After reaching a peak of $900 in the US markets, the price crashed down to $600 and was followed by non-stop fluctuations. Only today and yesterday has there been some semblance of stability, with a steady rise up to $810 on Bitstamp at the time of writing this.

            In Chinese markets, BTC hit an all-time high of $1150 in CNY. This rise has largely been fueled by two things. The predominant cause is the huge speculative buying pressure in China. Secondary causes include the Senate hearings going better than expected and generally positive news about Bitcoin over the past weekend. Apparently, there were people from CCTV (China's news network) at the Senate hearings so the effects of all these causes might have compounded on each other.  Also worth noting is that the second Senate hearing had a more leveled tone as compared to the mostly positive tone of the first Senate hearing.

            On Monday there was an FEC deadlock (3-3) on deciding whether to allow campaign contributions in Bitcoin. As an aside, I have a suspicion that one of the reasons the senators were careful not to be against Bitcoin was because they realized bitcoins could be used for campaign contributions.

            Also, earlier last week, there was a strong buzz on the Chinese message boards asking how to buy Litecoin. Litecoin is apparently difficult to buy in China since there are few to no established LTC exchanges there. One of the most common ways to buy LTC was by exchanging for it with BTC. One would expect this to be a trigger of the big crash on Sunday but the timing was different. Also another reason for LTC's popularity and jump in price is that, as with many speculators, late comers might think the BTC gravy train has already left so since all the good spots were already taken in BTC, they decide to speculate on LTC instead (especially since the price is lower and a lot of these speculators have no concept market cap; they just see a lower price and pile on).

            Also interesting to note, although BTC is trading higher in the CNY markets than the USD markets, for the past two days BTC was trading even higher on the KRW markets.

            Arbitrage on the price differences across different exchanges in different countries is difficult due to strict capital controls in China for moving money out of the country. The limit is $50,000 per year per person. That, combined with the problems Bitstamp was having with their software, ensured that the price difference was not arbed away. However, there is still an opportunity for sellers of Bitcoin to liquidate on the Chinese exchanges for a higher price so long as they don't mind holding CNY and have a bank account in China.


            As you may already know, markets have been extremely volatile this week, largely due to news coming out of China. The most glaring piece of news is that the Chinese government has banned banks and payment processors from dealing with Bitcoin.  Price crashed to a low of $650 today but has since rebounded to $800 after a fall earlier this week from $1200 to $850.  While many see this selloff as warranted fear that Bitcoin demand and adoption in China will slow or reverse due to the recent ruling, others have seen it as a buying opportunity.

            Today, there's also been a single large buyer at $830 buying up to $880 every time it touches that price over a period of 30 minutes.  I suspect that this is a large institutional trader/firm who has been waiting to gain exposure to Bitcoin at a decent price.  ”Smart money” entering into the market during crashes is usually a good sign for medium/long term growth.

            In the US, news came out that Bank of America is dedicating an analyst to Bitcoin, David Woo.  In his piece, David claims that Bitcoin has a maximum value of about $1300 if 1) it captures 10% of the e-commerce business, 2) becomes one of the top three players in the money transfer business, and 3) achieves a store of value reputation equivalent to silver. I think this is a gross underestimation of Bitcoin's potential as well as faulty math.  See original report here:

            First, regarding 10% of the e-commerce business. Not only is 10% a low estimate of Bitcoin's maximum potential market share in the e-commerce business but David's numbers do not factor in inflation. If the dollar loses value, the price of goods will go up accordingly so using Bitcoin to buy those goods would capture a larger $ value even for the same 10% of goods on the market. Also, in his calculations for consumer spending, he only considers cash and checking accounts but not credit cards.  And going further on that, he implicitly assumes that the rest of the world will behave in the same way as the US, but the rest of the world has much less consumer spending on credit cards; since he leaves out credit from his calculations, this would further bias his results to a smaller total world consumer spending number for e-commerce.  I can understand some reason for leaving out credit since Bitcoins are irreversible but leaving out 100% of credit can’t be right. Given that different people have different appetites for how much they value each side of the "ability to chargeback" and "low fee" tradeoff, even 50% of credit spending is a conservative estimate to include.

            Second, becoming one of the top three players in the money transfer business is a misleading comparison. He is comparing Bitcoin to Western Union, MoneyGram, and Euronet. This is like comparing the invention of the airplane to the top three automobile companies of the time. He is comparing a protocol to a business, like comparing the internet to internet companies.

            Finally, he draws a comparison to silver. He says that Bitcoin is 5 times more volatile than gold and so its ability to store value gives it an eventual market cap of 1/5 that of gold’s and then he tacks on another 1/60 since the gold to silver price ratio is 60:1.  First, why not compare Bitcoin to gold directly?  Second, how does the volatility calculation directly convert to a market cap anyway?  Shouldn’t returns also be considered as a factor which might make one store of value more appealing than another? But even going along with his volatility to market cap methodology, let's say Bitcoin's volatility goes down as it becomes more widely adopted and exchanges become more liquid; how would that affect market cap?

            All-in-all, it's good news that BofA is interested in Bitcoin which is a much different tune from last year or even earlier this year. My suspicion is that David is not comfortable giving more than a super conservative estimate of Bitcoin’s value since there is reputational risk for him of seeming like one of those Bitcoin loons to the rest of his company. It could also be the case that BofA is planning to enter into the Bitcoin space through an ETF or other structured product and wants to keep the price low before it does so it can start at a cheaper price.

            On another topic, there was a paper recently about improving on Bitcoin's block times to 1 second per block; you can find it here: The premise is to use a blocktree instead of a blockchain and consider valid the longest chain determined by the heaviest sub-tree at each node. The central idea is that orphaned blocks add legitimacy to deeper blocks even if they are off the main chain at the moment or even forever.  What about the wasted computing power miners would expend finding orphaned blocks? This ultimately doesn't matter since lower profitability would adjust the difficulty down and miners would be able to internalize the cost of finding orphaned blocks as additional difficulty. In other words, a higher probability of finding orphaned blocks at a lower difficulty yields the same payout as a lower probability of finding orphaned blocks at a higher difficulty.


            Markets are beginning to stabilize after some turbulence at the start of the week. Mixed news continues to stream in as Fidelity has enabled its customers to invest in the Bitcoin Investment Trust by Second Market through their individual IRA accounts before quickly disallowing it the day after the news broke.

            Barry Silbert, Second Market's CEO, is saying Wall Street will enter into Bitcoin over the next 3 to 6 months in 3 waves. The first wave will be IRA money and asset management funds. The second wave will be hedge funds and institutional investors.  The third wave will be Wall Street banks themselves.

            I broadly agree with his assessment and ordering of the waves with some caveats.  First, I think the first two waves will have a great deal of overlap. Second, I think this will take closer to a year or two rather than 3 or 6 months. Third, retail adoption will happen in parallel to all three of these waves. Finally, I'd like to add that following the Wall Street waves, we will likely see university endowments, company pension funds, and state-managed pension funds for public employees. I expect this final wave to take the longest (possibly 3-4 years from now).  

            David Marcus is expecting the price to double by the end of 2014. Going by futures markets on ICBIT, it looks like 2.5x the price by Dec 2014 is what the markets expect, so this is pretty close to Marcus' prediction. Yet historically, the futures markets have consistently under-predicted the price of Bitcoin even though it is strongly in contango and Bitcoin has no obvious or explicit carry costs.


            This past week has been tumultuous for Bitcoin in light of the Chinese central bank's ruling that financial service companies are no longer allowed to touch Bitcoin. As banks and payment processors are severing relationships with the Chinese exchanges, BTC China - the largest of the exchanges - currently has no fiat deposit method available, finding itself in a Gox-like situation. Since the rumors on 12/5 and the confirmation of the Chinese ban on 12/18, price has been dropping and volatility has been rising. Price went as low as $382/BTC (Bitstamp) on Wed before rebounding to the $600/BTC (Bitstamp) range where it is today.  

            While many Chinese speculators have been dumping, many non-Chinese speculators and investors have seen this event as an opportunity to buy cheap bitcoins. Also, although the original rally in Bitcoin was due to speculators from China, in the process of speculating and discovering bitcoins, I'm sure that at least a small fraction of initial speculators have been converted into Bitcoin true believers. So it's reasonable that the price at its recent lowest ($382/BTC) is still higher than the price before the Bitcoin craze started in China. As with all of these bubbles, the highs and lows are both successively higher as awareness and understanding continue to increase, one bubble after another.

            In retrospect, one of the reasons the Chinese government might had decided to come down on Bitcoin is to stop real money laundering. Many wealthy Chinese have a very difficult time moving money out of their country due to strict capital controls of allowing no more than $50,000 to be transferred out of China per person per year. Many wealthy Chinese may have seen Bitcoin as an opportunity to circumvent this capital control whereby they buy bitcoins in China with CNY and sell them overseas for USD.

            Long term prospects for Bitcoin are still looking good. What people and governments decide to call Bitcoin (e.g. legal/illegal, good/bad) does not affect the Bitcoin protocol's usefulness and long-term viability as a payments system and store of value.

            On another note, I'd like to share a thought I had about hashing: Scrypt (LTC) versus SHA-2 (BTC).  Scrypt was chosen for LTC because Scrypt hashing is more resistant to specialized hardware for mining (e.g. GPU, FPGA, ASIC).  In other words, with Scrypt, the efficiency gains from switching to a higher tier of specialized hardware is much lower than if SHA-2 hashing was used. The main reason this was touted to be desirable was that, this way, mining would be more decentralized or at least if mining were to be centralized, it would happen more slowly. But I wonder if that is actually true? If Scrypt is FPGA/ASIC-resistant, this would give botnet operators disproportionate hashing power on the network and would foster centralization in a different way. Ultimately, Scrypt trades away oligopolistic mining pools for oligopolistic botnet operators. Intuitively, it seems Scrypt is slightly preferable since most botnet operators cannot output huge amounts of hashing power since botnets usually consist of old and slow machines.

            Going further, there are Scrypt-based cryptocurrencies like Yacoin whose software requires greater and greater memory as the hashing rate goes up rendering static, customized circuit boards useless and making it not just ASIC-resistant but ASIC-immune.

            Quark and Securecoin are another batch of interesting altcoins. They both use 6 different hashing algorithms (e.g. Grøstl, Skein, BLAKE, BLUE MIDNIGHT WISH, JH, SHA-3), each one wrapped up in another. So if any one gets cryptographically compromised, the protocol stays secure.


            Bitcoin price this week has been consolidating around the $600s to $700s. As overall volatility is shrinking, the price has risen a bit. My interpretation for what's happening is that most Chinese speculators have exited the market and some recent good news has pushed the price up.

            Most notably:

            • Li Ka-Shing, Hong Kong's richest man, has invested into Bitpay though his venture capital arm, Horizons Ventures.

            • plans to accept Bitcoin in 2014.

            • The Monetary Authority of Singapore has decided to not to regulate the use of Bitcoin for purchasing goods and services.

            Bitcoin price would probably have gone up further if it wasn't for the bad news coming out of India. Immediately after the Reserve Bank of India (RBI) issued a warning to the public about the potential dangers of using Bitcoin, multiple fixed-price exchanges in India have shut down. The Enforcement Directorate (ED) even went so far as to raid the offices of and in search of evidence of money laundering. This is surprising since the RBI warning had a very "wait-and-see" tone toward Bitcoin rather than an adversarial one.

            Dogecoin may have started off as a joke but recently shot up to a top 10 cryptocoin position based on market cap before sliding down to 12th where it sits now. On Christmas, a Dogecoin online wallet service was hacked for over 20 million DOGE which amounts to about $12,000.  If hacks are being done to acquire it, it must have value.

            It begs the question, can anyone create an altcoin and create market value? It seems, at this point, that any tweak or variant of Bitcoin can create at least some value until the market for variant cryptocurrencies hits saturation much like the story of the internet boom of the late 90's. For now, these altcoins provide investors and users with another choice or flavor of cryptocurrency and it also gives speculators hope (possibly false hope) to get rich quick if they feel they've already missed the Bitcoin train. Ultimately, a free market of competing cryptocurrencies will bestow growth on the good and useful altcoins while culling the bad and pointless ones.

            But do these altcoins actually cannibalize each other's market caps? It seems that they do currently compete against each other to a large degree. The decision process of a person who wants to get into cryptocurrencies first answers the question, "Do I prefer crypto or non-crypto?" and then answers the question, "Do I prefer Bitcoin or one of the altcoins?" Indeed many people currently buy Bitcoin for the sole purpose of converting it to Litecoin.  But I think this mentality and decision process will change in the near future. Just as USD competes not just against EUR, JPY, etc. but also gold, oil, equities, and all other assets which can be bought with USD, each of the cryptocurrencies will compete not only against each other but with everything else on a more individualistic basis. The question will then be, "Which asset (e.g. USD, BTC, oil, S&P 500, LTC) do I prefer."

            Also worth mentioning, ETFs like Bitcoin Investment Trust, the Winklevii Bitcoin ETF, and, tangentially, NEOBEE are hot in the Bitcoin world right now. Going into the future, I expect not only ETFs which track a single cryptocurrency but ETFs which represent baskets of cryptocurrencies. So, instead of the S&P 500 which tracks a basket of 500 equities, there might be a Crypto 500 which tracks a basket of 500 cryptocurrencies. This could then be followed by inverse ETFs and leveraged ETFs.


            Bitcoin markets have been rallying this week from just under $750 to about $800 on Bitstamp. Notable news from this week:

            • BTC China (and China in general) has been losing significant market share in the exchange world. BTC China is down to 7,000 BTC daily volume compared to Mt. Gox at 9,500 BTC daily and Bitstamp at 14,500 BTC daily.

            • An SF hedge fund is looking for an execution trader for Bitcoin:

            • Wedbush - a prominent Wall Street firm - published a favorable report on Bitcoin:

            • Fortress Investment Group is forming a Bitcoin fund:

            • Paul Krugman says, in a half-joking way, that Bitcoin is evil:

            It’s looking like 2014 will be the year Wall Street joins Bitcoin.

            Most notably, I'd like to speculate on the impact of Fortress starting a Bitcoin fund on Bitcoin price. If they are doing an ETF in the same style as Second Market, I see them absorbing about a similar amount of capital, so something like $60m in 3 months. Given that Fortress has a more professional reputation in the financial world, they could pull even bigger numbers from Wall Street institutional investors. Also, the news and announcement of the fund will generate more awareness and credibility on the street. I see the price going up at least 10% just due to this new fund absorbing capital. Considering secondary and tertiary effects and being slightly optimistic on Wall Street's reception of Bitcoin, we could see over 100% increase in price over the next 3 months just from this event. Add on more buying pressure from the other bullish news coming out of Wall Street and it’s looking like a good year.


            Bitcoin markets continue to rally as volatility falls. Price rose from around $800/BTC to $880/BTC with a spike around $1000/BTC in the middle of the week. Notable news from this week:

            • is now accepting BTC for purchases and has done over $120k in Bitcoin sales in its first day.

            • Zynga is now accepting Bitcoin for some of its online games.

            • Major adult-content processor, Verotel, is now accepting Bitcoin as well.

            • New York Department of Financial Services have announced a hearing in late Jan about the possibility of issuing BitLicenses.

            •, the largest mining pool, got dangerously close to 51% of the network.  As panic spread through the forums, many miners left the pool bringing it down from about 45% to 38%.

            As we begin to see major merchant adoption and Wall Street interest, this year is looking to be a big year for Bitcoin.

            I think it is likely that 2014 will be even better for Bitcoin than 2013. Even with Chinese speculation largely out of the market, price has almost caught up to pre-ban levels. Now with growing merchant adoption, a distinctly different type of market demand will start to develop for Bitcoin. Instead of Bitcoin as an inflation-protected vehicle for capital or speculatory instrument for profit, there will be increasingly more people who buy Bitcoin so they can get a discount at a number of large online merchants. Merchants have an incentive to offer discounts because some of the value they save on transaction processing can be passed onto the customer based on the how the elasticity of demand and the elasticity of supply for a product matches up with each other. Also, this type of demand for Bitcoin is more stable than the speculatory kind which should lead to lower overall volatility in Bitcoin price. This, in turn, should attract more demand for Bitcoin as a vehicle of storing wealth. Demand of one type begets demand of another as Bitcoin bootstraps itself into the mainstream.


            This week markets have fallen from $850/BTC to around $800/BTC after an initial rise last weekend up to $900/BTC. Notable news from this week:

            • Silk Road seized coins will be sold through a US Marshal auction.

            • Wells Fargo calls for a Bitcoin "rules" summit to discuss how Bitcoin fits in with the traditional banking landscape.

            • Bitcloud is an aspiring distributed anonymous corporation (See which seeks to create decentralized versions of Youtube, Facebook, Dropbox, Spotify, and Comcast (ISP).

            • Ethereum, a new, turing-complete, digital contract system inspired by Bitcoin will be launching in about a week. You can read more about it here:

            Regarding Bitcloud and Ethereum: they are very ambitious projects and it is unclear whether they will take off. I think they are worth keeping an eye on, particularly Ethereum. Ethereum is also, to my knowledge, the first implementation of the Greedy Heaviest-Observed Sub-Tree (GHOST) protocol I mentioned in one of my previous newsletters which allows for faster block confirmations in the blockchain/blocktree by realizing that orphaned blocks give it's parent chain legitimacy even if never included in the final “blockchain”. See:

            On an unrelated note, I also wanted to touch upon stealth addresses. Recently some folks have described a way for payees to publish a single address whereby any payers and third-parties will not be able to figure out what other payments have been made to that address. It works by creating a Diffie-Hellman shared secret (see between the payee and payer and then offsetting the publicly published payee address by some function of the shared secret. The payer then uses the private key associated with the published public address and offsets that private key by the same function of the shared secret. This allows payers to send their payment to a different addresses and it allows the payee to have the ability to retrieve that payment from those different addresses thus disassociating all payers from each other. Unfortunately, payers can still be linked if the payee does not take extra precautions. For example, if the payee decides to send out an amount of bitcoins from his wallet which taps into the bitcoins sent to him by multiple payers, those payers can then be linked.  To completely avoid this issue, a customized wallet would have to be built or we could just use something like Zerocoin.


            This week markets have fallen slightly from about $800/BTC to $785/BTC.  All in all, a slow week in the markets. Notable news from this week:

            • The Argentinian peso was devalued to 8 ARS/USD recently as the central bank loosened its capital controls on the struggling currency. The "blue dollar" (black market rate) trades even lower at about 12 ARS/USD.

            • TigerDirect is now accepting bitcoins.

            • JPMorgan CEO, Jamie Dimon, says that Bitcoin is “a terrible store of value.  It could be replicated over and over.”  Unfortunately, it didn't seem like Jamie did his homework.

            • Marc Andreesen explains why Bitcoin matters: He even gets into the Byzantine Generals Problem.

            Today I'd like to touch upon fee pricing for maker-taker markets. I've talked about this topic in one of my earlier market updates but to refresh: In a maker-taker market, between the liquidity provider and the liquidity taker, one side pays a fee and one side pays a negative fee. Usually, the liquidity provider has the negative fee but this is not always the case. So what are the tradeoffs here? Each liquidity taker issues a market order which represents an opinion he has about the market. The stronger the opinion, the more he is willing to pay up as a premium over the current market price (e.g. suppose market price equals the mid price).

            So what is the real premium he must pay? He must pay the exchange a fee for taking liquidity and the market maker half of the spread to hit the order. So given a fixed propensity to pay a certain premium to express a certain opinion, it matters little to the liquidity taker whether the premium goes to the exchange or the market maker. Therefore, in equilibrium, if we charge the liquidity provider a higher fee, the spreads will be larger and vice versa. So to ensure tight spreads, we give the liquidity provider a negative fee. But there is a limit to how small the spread can be. Suppose the smallest increment in price is a penny. Then, if we already have a market where the spread is a penny and the fee to the liquidity provider is -.1%, improving that fee for the liquidity provider to -.2% does nothing to the spread. Instead, such an action would simply transfer value from the exchange to the liquidity provider and action dries up, hurting the exchange even further.

            Another strange effect of too negative a fee for liquidity providers is that the liquidity providers at the top of the order book are actually liquidity takers in disguise. If you wanted to buy BTC and the market was $99.99-$100 but had to pay 2% as the liquidity taker and the negative fee to the liquidity provider is -1%, you might decide to quote on the bid in hopes of someone else with more immediacy than you hitting your quote.  So instead of most liquidity providers functioning as market neutral participants, many liquidity providers at the top of the orderbook are simply liquidity takers without high immediacy. This is bad because any apparent gains made in price discovery from having a narrow spread are illusory since the true price could be anywhere in the range of $98-$102, and markets don't move as liquidity takers without strong immediacy just pile on the bids and asks at the top of the book.

            So what is the best solution? On one hand we want narrow spreads for both price discovery and cosmetic reasons. On the other hand, we don't want stagnant markets. Well, we can check a limiting case. The minimum spread in a market with no fee frictions is $.01 (or whatever the smallest denomination is). If markets are completely random and efficient, liquidity providers would be willing to take any marginal amount in spread + negative fee to provide liquidity. So in the limit case, a zero fee would be enough to attract these liquidity providers to quote a $.01 spread. In other words, you can virtually guarantee that you (as the exchange) are not leaving money on the table by charging the liquidity provider a zero fee. Given that the markets are not perfectly efficient, you could probably get away with a small negative fee. You could give a negative fee to liquidity providers of up to one half of the natural spread which is the spread which liquidity providers would have quoted in the zero fee world.

            Also, bear in mind that this analysis is done with the assumption that third party intermediaries to these capital markets are not present.  Given the presence of brokers, rebates for either providing liquidity or taking liquidity create strange incentives if the rebate is paid to the broker rather than to the client for whom he executes.  For example, brokers might fill their client’s orders on the exchange which provides the highest rebates instead of the best price for his client.


            This week markets have been pretty stable around $800/BTC after recovering from a crash midweek down to around $730/BTC.

            Notable news this week:

            • Charlie Shrem, CEO of Bitinstant, was arrested on charges of money laundering. He is accused of knowingly collaborating with a SilkRoad currency exchanger to skirt anti-money laundering procedures and help SilkRoad users convert their dollars to Bitcoin. If the evidence in the indictment is accurate, Shrem will likely face a difficult legal battle.

            • Visa CEO says he does not see Bitcoin as a threat since traditional payments systems are still safer for consumers. Arguably he's right in the sense that the most common automobile of the early 1900’s was probably safer than the first airplane.

            • BTC China has started accepting bank deposits again. Price and volume charts suggest bot behavior immediately after deposits opened: Some people have argued that this is fake volume generated by BTC China hitting its own quotes but that's unlikely to be the case. More likely is that there is a dominant bot which is keeping the price in a narrow band as fresh deposits hit the exchange and trade with it or there are two bots interlocked with each other, oscillating the price.

            • ANXBTC - a Hong Kong exchange - gave away approximately $65,000 in Bitcoin in a PR stunt that involved giving away red packets filled with $10 in Bitcoin each.

            • Mint adds BTC support.

            • Overstock CEO reveals he holds millions of dollars worth of Bitcoin.

            • New York Department of Financial Services held hearings about Bitcoin.

            • US Treasury issues guidance that Bitcoin miners and investors will not be regulated.

            Today I'd like talk about the functions of money and the properties associated with those functions. To begin, we should ask ourselves why a system with money might be desirable in the first place. Systems with money are largely superior to barter systems because of a phenomenon coined the "double coincidence of wants". In a barter economy, the only time a trade happens is when A wants what B has and B wants what A has. In other words, there needs to be a "double coincidence" for a trade to happen and that is rather rare. In money-based systems, a single coincidence is enough to guarantee a trade since the party receiving money knows he can, in turn, exchange it for a good he wants at a later time.

            So money is useful because it serves as a medium of exchange, a unit of account, and a store of value.

            For money to function well in these three ways, it must have certain properties:

            • Scarcity

            • Divisibility

            • Fungibility

            • Portability

            • Durability

            First, money must be scarce in the sense that it cannot be counterfeit so as to retain a stable value. Second, divisibility is important because different goods have different values in the market so money which can be divided up very finely is better at capturing precision in value.  Fungibility is also important for the same reason: it allows money to be recognized and interchanged without regard to special properties of one monetary unit versus another of the same denomination.  Since the velocity of money will be higher than the velocity of any other good, portability is important to mitigate the frictions of transport and transfer. For this reason, money almost always has a high value to weight/volume ratio. Finally, money must be durable in that it should not decay or sustain wear while one carries it. For this reason, perishables don’t make for good money (e.g. apples, oranges).

            Given these considerations, is Bitcoin sound money?

            To quote Satoshi Nakamoto (8-27-2010):

            “As a thought experiment, imagine there was a base metal as scarce as gold but with the following properties:

            - boring grey in colour

            - not a good conductor of electricity

            - not particularly strong, but not ductile or easily malleable either

            - not useful for any practical or ornamental purpose

            and one special, magical property:

            - can be transported over a communications channel

            If it somehow acquired any value at all for whatever reason, then anyone wanting to transfer wealth over a long distance could buy some, transmit it, and have the recipient sell it.

            Maybe it could get an initial value circularly as you’ve suggested, by people foreseeing its potential usefulness for exchange. (I would definitely want some) Maybe collectors, any random reason could spark it.”


            Markets have been tanking these past two days after an earlier week of stability at the $800/BTC level. Currently, the price has fallen to about $707/BTC on Bitstamp.

            One explanation for the steep drop is that Mt. Gox has some issues with Bitcoin withdrawals. Gox claims that their proprietary wallet software has a bug which caused them to lose track of which addresses hold how many bitcoins thus accidentally triggering double-spend flags in the blockchain when they issue instructions to spend already-spent coins. At the time the withdrawals were failing, many people were concerned that Gox did not actually have any coins left and was completely insolvent due to friendly fraud, the government/banks freezing over $10m of their assets, hacks, etc. This caused a huge panic sell as Gox's price premium over Bitstamp's dwindled.

            It's also interesting to note that yesterday there was a spike in Bitcoin days destroyed (BDD) higher than anytime in Bitcoin history. BDD is a hybrid measure of how many and how old coins are that just moved. In other words, to get a high number for BDD, we would need to see very old and unmoved coins all of a sudden move in a large quantity. This usually signals a higher probability of selloff as it could represent an early adopter or early miner deciding to move old coins into an exchange and liquidate his BTC position.

            It's also possible that the Gox event and the BDD event are linked. It's possible that Gox decided just to issue transfer instructions to every wallet it has to reconsolidate all their bitcoins to a single address. It makes sense for them to do this because if you didn't know which of your wallets held how many coins, the only way to recalibrate the ledger is to tell each address to send coins to a consolidation address expecting some of the transactions to fail. Some of those funds moved may be very old coins which have not moved for a very long time thus creating a huge spike in BDD. Block 284349 had over 137,000 BTC in output:  Would be curious to see someone do some blockchain analysis to see what’s going on.

            Other notable news:

            • Apple bans's wallet app from their App store. It reminds me of a Ghandi quote that goes “first they dismiss you, then they laugh at you, then they fight you, then you win”.

            • Dogecoin continues to gain traction. Tipping DOGE on the Dogecoin subreddit becomes commonplace.

            • Looks like Ethereum will be launching later this month. I strongly recommend learning about Ethereum as it could be a better foundational protocol than Bitcoin. I still have some reservations about it, particularly how the initial funding phase will work and whether the project is becoming politicized, but it is certainly innovative enough to deserve some attention.  See


            This week’s markets have been falling with a sharp crash from $700/BTC to $550/BTC on Monday largely due to the Gox transaction malleability fiasco. On Tuesday, multiple Bitcoin exchanges were under DDOS attack whereby the attacker used transaction malleability to morph the hashes of many transactions on the network. Exchanges which used these ambiguous hashes to update their ledger had to halt withdrawals. Bitstamp and BTC-e were both affected. As of today prices have started to recover as Bitstamp announced they would resume BTC withdrawals after fixing their transaction malleability problems.

            Mt. Gox BTC are currently trading at a 40% discount to Bitstamp's BTC. The market is suggesting that Gox only has 60% of their customer’s coins and are likely insolvent at this point. I think this is a slight overreaction. While I do suspect that Gox has much fewer coins than their ledger requires to pay off all customers, I think 40% is too heavy a discount. Insolvency is still quite possible depending on whether they have enough of their own money to pay off customers should they decide to do so. Here is a case where if they are missing a smaller amount of coins, they would be more likely to pay off the difference, while if they are missing too much, they would probably decide to default and close shop.

            Also interesting to note is that at the very beginning of this fiasco with Gox freezing BTC withdrawals, Gox's price crashed taking Bitstamp and BTC-e’'s prices with it, as there was widespread confusion in the market and an ongoing assumption that the exchange prices should be correlated. Once the market realized that this was a Gox specific issue, Bitstamp and BTC-e rebounded quickly. In one particular trade during the crash on BTC-e, there was a huge market order sold into very shallow liquidity causing some coins to trade at $106 (what a bargain). However, then the DDOS attack came online, causing Bitstamp and BTC-e to freeze BTC withdrawals and drop again.

            This week demonstrates the massively profitable opportunities that exist in a wild market. In these situations, highly accurate and privileged information is king. If you were able to discern that the issue was Gox specific before the decorrelation happened, you could have bought up cheap coins on Bitstamp and BTC-e but without facing Gox counterparty risk.

            Also, there is a market handicap in favor of insiders. Bitstamp operators could easily have bought up a bunch of BTC right before announcing withdrawals were resuming. And there are rumors that Gox is doing exactly this: buying up their own cheap coins as they scramble to fix their ledger's transaction malleability issue.

            In other news,

            • Standard Bank in South Africa decided to do a Bitcoin pilot program with Switchless before changing their mind a few days after it was announced at Finovate.

            • Bing added Bitcoin currency conversion rates.

            • SilkRoad 2.0 was hacked and about $2.7m was stolen. Operators claim transaction malleability was the cause of this theft but there are suspicions that the heist was an inside job.

            • Multiple Bitcoin price tracking sites have removed Gox from their trackers.


            This week markets have been in chaos with Mt. Gox ($103/BTC) trading well below Bitstamp ($573/BTC) and BTC-e ($554/BTC).

            Deciding whether to buy up cheap Gox coins is a hard problem because it involves asymmetric information. Gox knows exactly how many coins it actually has while everyone else is left in the dark. Often these asymmetric information situations create perverse incentives like insider trading. There has already been some evidence of this: 


            Right now the Gox market is estimating an 83% chance of Gox default.  On the other hand, is a site which allows you to trade real bitcoins for Gox bitcoins and vice versa and the price is currently .45BTC/GoxBTC implying a 55% chance of Gox default. So it is likely that the true value of GoxBTC is between its current price of $100/GoxBTC and $260/GoxBTC. Also note that Bitcoinbuilder has done a 24h volume of about 5,300 GoxBTC while Gox has done a 24h volume of 69,000 GoxBTC, so the markets are suggesting that the true price is closer to the Gox price. Also, the ask wall is very shallow right now so the amount of cash it would take to move the price up is very little.

            Ultimately, the question is how many phantom bitcoins were created on Gox's internal ledger as Gox transferred out the real coins during the transaction malleability exploit? The greater the number of phantom coins on their system means the more coins they have to buy up themselves to retain solvency for their customers. If the loss they must take to do this is greater than their liquid assets, they are insolvent and can only resolve the issue in one of three ways:

            • Default on all customer funds, face lawsuits later. (Already there are class-action lawsuits against them being filed.)

            • Discount all Gox bitcoins by how much they can afford e.g. pay out 50% of each customer's Bitcoin balance and call it a day and hope customers don't sue as much. (This happened during one of the Bitcoinica hacks where they paid out 50% of the owed coins. It seemed like people were much more ok with that solution than expected.)

            • Market manipulation: 

              • Hope that asks fill up in the orderbook and buy up their own BTC at a massive discount, delete those coins from their ledger and take the loss.

              • Buy up what is left of their real coins on their own exchange and sell them on Bitstamp, then use that money to buy up phantom coins as much as possible, delete those coins from their ledger and take the loss. (My guess is that they are doing this but face the problem of a thin ask wall so they can't buy up enough phantom coins to regain solvency. This would also explain the discrepancy between Gox's price and Bitcoinbuilder's price.  And it would also partially explain the dip in Bitstamp and BTC-e's price since Gox would be putting extra selling pressure on those markets.

            There have been rumors that Gox was processing refunds automatically during the transaction malleability fiasco: . If that's the case, they could be missing a huge chunk of customer bitcoins.

            On the other hand Roger Ver, a prominent Bitcoiner has been buying up a lot of GoxBTC possibly due to faith in the exchange or insider information. And as another piece of relevant/irrelevant information, Gox CEO Mark Karpeles seemed to be in good spirits from the live feed of Gox protestors outside of Gox's office in Japan. There are also theories that Gox is solvent but purposefully spreading fear and panic to keep the price low so they can buy up cheaper coins and buy back their losses to transaction malleability at a cheaper price.

            There are even more far-fetched theories like the following:  Who knows?

            Ultimately, the information is inconclusive and insiders will have a huge advantage in this market. I would also like to note that selling GoxBTC for GoxUSD may not matter if they are truly insolvent, so personally I would buy GoxBTC if I had GoxUSD if only to capture the potential upside suggested by Bitcoinbuilder's market. I suppose litigation over USD is easier than litigation over BTC, but it seems to me not to carry that type of premium. Also, to clarify, I'm not saying I would spend real USD or real BTC to move into the Gox market, only that I would buy GoxBTC with GoxUSD if I had any sitting on the exchange.


            As you may have all heard, gox is filing for bankruptcy.  On one hand, my condolences go out those who have lost significant holdings on gox.  On the other hand, the fall of gox paves the way for a new generation of exchanges with professional infrastructure and high-end security.  The end of gox marks the beginning of a new era for bitcoin.

            Markets have stayed in the $500/BTC to $600/BTC range with the exception of a crash going as low as $420/BTC in the middle of the week.

            In world news, Ukraine's currency the hryvnia (UAH) has been crashing amidst all-out revolt in the country.  The currency has fallen from 8.13 UAH/USD to 11.15 UAH/USD.  Bank runs in Ukraine have already started and the currency does not seem to be slowing in its freefall.

            Overshadowed by the gox news, NEOBEE a banking and payments processing bitcoin company in Cyprus launched its flagship branch.  NEOBEE's stock price on Havelock doubled from .003 BTC/share to .006 BTC/share over two days.  NEOBEE currently trades at .0052 BTC/share.

            Also, if you haven't heard already, we are launching very soon.  To reserve a spot for our early access program, sign up and spread the word: