Bitcoin Utility and Speculation

At the end of 2012 a great deal of data and information concerning the utility of Bitcoin came to light. At that time we had information on the turnover of SilkRoad – total market revenue at $1.2m per month. Reports on profits at Bitcoin based casinos – Satoshidice profits at >$500k in six months and BitZino profits over $160k. Profits from Bitcoin based ecommerce – Bitcoinstore market valued at >$500k in less than 6 months and market reports from Bitcoin based payment processors – BitPay processes $5.2m in March. Additionally the beginning of the year brought news of various markets moving to accept Bitcoin for payment – Reddit, 9flats, WordPress. But much the noise for 2013 related to speculation in Bitcoin as a store of value.

While the speculative nature of Bitcoin brings interesting discussion — we have never seen the worth of FIAT currencies evaluated like a traditional ETF as we have in recent months with central banks QE measures put beside Bitcoin — still filtering utility from the noise helps gauge the growth and health of the market. Speculation can be divorced from Bitcoin no more easily than it can from a traditional ETF like GOOG or AAPL. However, for companies and individuals that truly believe in the future of Bitcoin releasing numbers (profit, loss, volume), both positive and negative, helps in bring a more mature understanding of Bitcoins future and growing pains.

With the data available it is possible to provide different views on utility. In its short history we have already seen speculative value dwarf discussion and crash twice. The buildup and crash of June 9th 2011 was a preview of the recent bubble from April 10th 2013. Prior to the 2011 crash the only dataset available concerning utility was that of SilkRoad. With SilkRoad opening its doors on February 2011 and prior to the speculative bubble, a single Bitcoin was valued around $1. In the April and May lead-up up to the June bubble only minor publicity could be found. At that time Google’s news index had next to nothing on Bitcoin. May 29th saw a nod from the New York Times. June 1st brought SilkRoad into the spotlight via Gawker and by June 8th, one day prior to the crash, The Register reported that two US Senators were calling for a crackdown on Bitcoin. The value took a slow walk from the high of $30 down to $3 by the end of 2011 and hovered in shock around $5 until the middle of 2012.


SilkRoad opened Feb. Walk prior to speculation


Speculative walk prior to 2011 bubble


2011 speculative bubble and slow walk down

This already provides a view on the resilience of Bitcoin to speculation and government intervention. A utility requiring the use of Bitcoin was enough to hold it to pre-speculation value (or more, $4 rather than $1). The 2011 Bitcoin crash should show law makers that harsh intervention on Bitcoin or other p2p crypto-currencies will do nothing more than wipe out the savings of those gambling on it as a store of value, doing little to its actual utility.

Numerologists can use this data to play fortune-teller in dividing speculative movement from utility. One such riddle might go: SilkRoads yearly revenue was $14.4m a year. With 8.25m coins in circulation at the time the utility value of Bitcoin was $1.74 (14,400,000 / 8,250,000 = $1.74). The speculative run up 7 days before the crash to $10 saw a multiplication factor of 5.7. In the height of the bubble, $30, that factor was 17.2. After the crash the multiplier to utility settled at 3 to 4x ($4).

The utility of Bitcoin in 2013 brings more to the space than SilkRoad. It would be difficult to provide a similar narrative of utility without consistent earnings reports from business in the space. However, another view can be provided using the Bitcoin company every couch based developer loves to hate: Satoshidice.

Satoshidice is a casino that uses the Bitcoin blockchain as it’s database, ledger and dealer. Because each block on the chain is limited in size many complain that Satoshidice is spamming the blockchain while others arguing that if Bitcoin cannot handle such a service things should change. In effect arguing that Satoshidice is proving an early measuring stick to test the network for when Bitcoin handles transactions for major financial spaces. What it might also provide is a window into increased utility as it provides a constant flow of transactions. As of now it accounts for more than 50% of blockchain transactions.

As utility has increased in the Bitcoin market with new businesses utilising it and new users Satoshidice has also seen an increase in use. But one would expect the increase use of Satoshidice to be minor in comparison to the other major companies and new users on exchanges that all bring more transactions to the Blockchain. However, data shows only minor divergence of late. The following shows the percentage of Satoshidice traffic per Blockchain Tx (requests) and size in Megabytes. (historical data provided by Dooglus)


Satoshidice still dominates the blockchain and new services are only now starting to make their presence known. Here are the same graphs but looking only at 2013:


Perhaps the situation is changing and it is up to the reader and their inner numerologist to determine if this is a valid view by which to measure utility.


2 thoughts on “Bitcoin Utility and Speculation

  1. The math below is nonsensical. Like dividing number of pirates by number of churches or something.

    > One such riddle might go: SilkRoads yearly revenue was $14.4m a year. With 8.25m coins in circulation at the time the utility value of Bitcoin was $1.74 (14,400,000 / 8,250,000 = $1.74).

    So 14.4M “dollars per year” divided by 8.25M “coins” = $1.74 “dollars per coin-year”. How is a “dollar per coin-year” meaningful as a unit of measure? Why not dollars per coin-month? Why not dollars per coin-hour (since a coin can really be re-circulated within an hour on average).

    Other than that, it’s a pretty good article. When trying to figure out what a bitcoin is worth (other than the obvious “the price it’s worth is between what people are willing to pay and what people are willing to sell them for”), you need to look at two things: Utility-value and Speculation-value (or chance-value).

    Anybody who says “Bitcoin can’t be used as a currency, because it’s too volatile”, doesn’t understand that it will not always be too volatile. Currently, the vast majority of the value of a bitcoin is in “the chance” that it becomes more useful. As that chance becomes reality (or gets ruled out), “the chance” component of the value becomes less of a factor, and the utility part becomes a greater component of the value.

    The “utility-value” increases when more vendors accept Bitcoin, when exchanges can be relied upon more, and when users can more readily understand how to utilize and secure their Bitcoins. The utility as a medium of exchange increases with price stability, but the utility as an investment vehicle decreases as the “chance of a home run investment” becomes realized or ruled out.

    Right now, we’re probably 99.9% speculation-value, and 0.1% utility-value. Or it may be more lopsided than that.

    • I agree the math using SR revenue isn’t a measurement at all. It only looks nice. I mentioned it anyway because it less nonsensical, even if only slightly less, then a lot of the other numbers people throw out. I feel the thrust of the article intends to say that even if we cannot graph utility when more bitcoin businesses provide numbers and data we can get a better sense of it still.

      Your comment is refreshingly sound and I feel the same on the current ratio of speculation vs utility movement.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s