[Re-Running] Bitcoin Fair Value: A First Assessment


We model this assessment on David Woo's December 2013 report: "Bitcoin: a first assessment" (Bank of America / Merrill Lynch). Mr. Woo begins with an extensive and apt summary of the bitcoin ecosystem, and we highly recommend reading his report. He goes on to analyze bitcoin's fair value using an equity-equivalence framework. We re-run Mr. Woo's value analysis (and then some) below, using what we consider more appropriate methodology and comparative metrics for a currency/commodity hybrid such as bitcoin.

We agree with Mr. Woo that bitcoin is both a medium of exchange and a store of value. We also agree that it's easiest to assess fair value by treating these uses separately. Additionally, we would like to note that the below analysis represents various bitcoin success-scenarios which may or may not occur, and that fair present value is therefore some weighted discount to the below success-case values. We treat this further in our conclusion.

Value as a medium of exchange

To assess bitcoin's potential value as a medium of exchange, we will look to existing fiat money flows and supplies, and ask what total value of all bitcoin is necessary to support various fractions of that economic activity. We do this as a starting point for rational, data-driven determination of bitcoin's fair value, and will leave it to the individual reader to decide whether our underlying assumptions and reference points are sensible.

Economic Activity, Money Supply, and Velocity

To determine the supply of money necessary to sustain a certain amount of economic activity, we must look at how quickly that money changes hands. This concept of monetary velocity is a key value for modern economies. Let's look at velocity in the bitcoin ecosystem.

We can see from blockchain data that average daily transaction volume for bitcoin over the last 6 months is roughly 200,000BTC:


while total supply of bitcoin averaged ~10,000,000BTC over the same period:


That yields an annual velocity of roughly 7.3*. This is much higher than the monetary velocity of dollars, which averages under 2.0. In our view, this seems a natural consequence of bitcoin being a far more frictionless medium of exchange compared to fiat money such as Dollars or Euros. Since bitcoin requires no third-party settlement intermediary, it's simply easier and cheaper to send/receive; thus, one would indeed expect it to move around the economy more quickly than other currencies.

The other side of the equation when determining value of a single unit of something is supply. Due to the algorithmic nature of bitcoin, we know the supply exactly; there are currently 12,151,075 bitcoins in circulation as of this writing (see chart above, or visit blockchain.info/charts for real-time data). Additionally, we know that there will never be more than 21,000,000 bitcoins ever issued. We will use this latter supply figure in value calculations below.

* Money Velocity, V, is defined as: V = T / M, where T is the aggregate value of all transactions over a given period, and M is the avg total money supply over that same period. In our discussion, T = (200,000 * 365) = 73,000,000, and M = 10,000,000. Thus, V = 7.3.

Bitcoin as 10% of remittance market

In contrast to Mr. Woo's comparative equity/enterprise-value analysis, we think it's more relevant to analyze bitcoin's potential share of the remittance market as a currency, not as a single corporation/equity. The equity comparison Mr. Woo undertakes is ultimately vague and difficult to justify since bitcoin is not a company. Luckily, since bitcoin is used as a medium-of-exchange (ie, currency), we can look directly at existing demand for currency and determine what value bitcoin must have in order to satisfy various slices of that demand. Using such methodology, the calculation is simple:


Chart source: World Bank

Global remittances in 2012 totaled $478 Billion according to the above World Bank data. If bitcoin accounted for 10% of that activity, the equivalent of $47.8B worth of bitcoin would need to be transacted. Using our monetary velocity of 7.3 from above, the supply of bitcoin would need to be $47.8B / 7.3 = $6.55B.

Table 1. Bitcoin Money Supply Summary 1
Domain Domain Penetration Supply Required
Remittance Market 10% $6.55B

In our view, it seems reasonable that bitcoin eventually becomes widely used in the remittance market due to its frictionless, borderless, and near-zero fee properties. For these reasons, we see 10% of this market as a conservative reference point.

Bitcoin as 10% of "black-market" transactions

Total black market transactions have an estimated annual value of $1829B. 10% comes to $182.9B, yielding a total necessary supply of bitcoin, at our money-velocity of 7.3, to cover this economic activity of $25.05B.

Table 2. Bitcoin Money Supply Summary 2
Domain Domain Penetration Supply Required
Remittance Market 10% $6.55B
Black Market Transactions 10% $25.05B
Total: $31.60B

Like Mr. Woo, we offer no opinion on whether or not bitcoin is suited to black-market transactions, but merely provide the above analysis as a reference point for potential data-driven valuation discussion.

Bitcoin as comparable to one medium-sized nation

We'll (subjectively) use a GDP of $200B as the definition of "medium" for our purposes here. That roughly corresponds to the GDP of countries such as Algeria, Peru, Kazakhstan, and the Czech Republic.


For comparison, as can be seen in the above chart, the GDPs of France, the UK, Brazil, and Russia are all more than 10 times higher, and the GDPs of Estonia, Uganda, Zambia, and Afghanistan are all 10 times lower.

Once again using our velocity of 7.3, we calculate that a money supply of $27.4B worth of bitcoin would be required to support $200B in economic activity; ie, the equivalent of a single "medium" sized country.

Table 3. Bitcoin Money Supply Summary 3
Domain Domain Penetration Supply Required
Remittance Market 10% $6.55B
Black Market Transactions 10% $25.05B
M2 of "Medium" Size Country 1 Country $27.4B
Total: $59.0B

Bitcoin as 10% of global eCommerce

Total eCommerce can be approximated as the sum of Business-to-Consumer (B2C) and Business-to-Business transaction volume. B2C is somewhat readily tracked; estimates for 2013 are about $1.3T. B2B figures are more difficult to reliably determine, but some studies (pdf) suggest B2B eCommerce volume is more than double B2C volume.

In our analysis, we'll assume equal volume for B2C and B2B transactions, yielding total eCommerce volume of ~$2.6T for 2013. To support 10% of that economic activity in bitcoin, the total supply of bitcoin would need to be $35.61B

Table 4. Bitcoin Money Supply Summary 4
Domain Domain Penetration Supply Required
Remittance Market 10% $6.55B
Black Market Transactions 10% $25.05B
M2 of "Medium" Size Country 1 Country $27.4B
Global eCommerce 10% $35.61B
Total: $94.61B

Bottom-Line: Estimated total value for bitcoin as medium of exchange = $94.61B

As the astute reader will note, the above figures are not necessarily fully additive, but we approximate to provide a reasonable upper-bound on bitcoin's medium-of-exchange value if bitcoin garners 10% of the markets noted above plus the general economic equivalent of one medium-sized country.

Value as a store of value

We further agree with Mr. Woo that bitcoin as a store of value most closely resembles precious metals (specifically gold) and cash balances since it generates no interest or dividends in and of itself, and is credibly scarce. We will treat each separately:

Bitcoin as 10% of gold's store-of-value demand

Here we will diverge from Mr. Woo, and instead look at bitcoin garnering 10% of gold's investment demand. While Mr. Woo instead looks at silver, arguing that bitcoin's volatility makes it inappropriate as a gold substitute, we note that we're evaluating potential success-cases and medium/long-run possibilities. As bitcoin gains further adoption and liquidity, its volatility will lessen. Thus, for the purposes of this valuation-grounding exercise, we believe it's both fair and interesting to look at bitcoin's value if it were to achieve 10% of gold's investment demand.

What's the total value of gold as an investment? It's estimated that 50% of gold is used as jewelry and about 10% in industrial processes, leaving 40% of gold consumption for investment purposes.

With ~174,000 tons having been mined, that yields a total above-ground investment-gold supply of $2.7 trillion.

Ten-percent of total investment gold is $270B.

Table 5. Bitcoin Money Supply Summary 5
Domain Domain Penetration Supply Required
Investment Gold 10% $270B

Bitcoin as 1% of "offshore accounts"

A headline-grabbing research paper from 2012 estimates that the world's high-net-worth individuals have hidden between $21T and $32T in untaxed "offshore accounts". While it's difficult to know much about the nature of this money or validity of the analysis, it's worth noting that bitcoin could eventually be seen as a potential vehicle for discretely storing liquid wealth.

Bitcoin is extremely confiscation-resistant, and essentially impossible to block from a flow-perspective. It is not anonymous, but its pseudo-anonymous/user-defined-anonymity properties seem likely to hold privacy advantages relative to traditional money flows. With the recent precedent of the Cyprus bail-ins, Portugal's pension seizure, and the IMF's musings on global wealth confiscation, we believe the world's wealthy may come to look toward bitcoin as a safe-haven for a small percentage of their non-public holdings.

Using the $21T lower-bound from the above report, we see that 1% of such wealth amounts to $210B.

Table 6. Bitcoin Money Supply Summary 6
Domain Domain Penetration Supply Required
Investment Gold 10% $270B
HNW "Offshore" Holdings 1% $210B
Total: $480B

Bottom-Line: $480B in potential store-of-value demand if bitcoin captures small fractions of existing store-of-value assets.

Final Tally

When we add our estimated market capitalization of bitcoin if it captures 10% of common medium-of-exchange uses to our estimate if it captures up to 10% of store-of-value uses, we arrive at a total market-cap of ~$574.6B.

Table 7. Bitcoin Money Supply Summary 7
Domain Domain Penetration Supply Required
Remittance Market 10% $6.55B
Black Market Transactions 10% $25.05B
M2 of "Medium" Size Country 1 Country $27.4B
Global eCommerce 10% $35.61B
Investment Gold 10% $270B
HNW "Offshore" Holdings 1% $210B
Total: $574.61B

At full-issuance of 21,000,000 bitcoins, each bitcoin's fair-value would therefore be $27,381.


Our analysis determines that one bitcoin must be worth $27,381 in order to support 10% of common medium-of-exchange economic activity, plus well under 10% of existing common store-of-value demand. Whether bitcoin will achieve on the order of 10% market-share for those activities is another analysis entirely; one that is far more subjective, and requires a framework for evaluating both micro and macro economic dynamics, viral technology adoption, network effects, and other domains that are beyond the scope of our analysis.

That said, the following conclusions can be drawn from our analysis above:

1) If bitcoin achieves close to 10% market-share of the above uses, it is undervalued by more than an order of magnitude today.

2) The bitcoin market is either inefficient, or is assigning a probability of less than 5% for bitcoin to achieve the success-scenarios outlined herein.

Is Bitcoin a bubble?

The answer to this question depends entirely on how much bitcoin adoption, both as a medium of exchange and a store of value, grows in the future. From our analysis, it's clear that there is significant upside if bitcoin does continue to gain market-share.

To arrive at a fair value for today, one must determine the probability that the above 10% market-share "success scenarios" will occur. If we say there's a 10% of this success itself, then a simple analysis (ie, ignoring discount rates and duration), puts a fair value of one bitcoin at (0.10)($27381) = $2,738 today.

14 responses on “[Re-Running] Bitcoin Fair Value: A First Assessment

  1. Craig

    This entire analysis hinges on the assumption that the velocity of BTC will remain constant at 7.3. This assumption seems unjustified to me, and I suspect it is quite false. I would not be at all surprised to see the BTC velocity change by orders of magnitude in either direction.

    You seem to be treating velocity as something fundamental to the economy. Perhaps it’s just a useless statistic with no important connection to meaningful economic reality.

    As a thought experiment, suppose I could snap my fingers and cause all dollar-denominated prices in the economy (including salaries, the price of labor) to be doubled without changing the nominal money supply or anyone’s expectations about future inflation rates. Just suddenly double the price for every good & service that is priced in dollars. I assert that this would not drastically affect the consumption or production activity of most of the populace. After all, the typical person’s spending habits are driven primarily by their amount of income, not their amount of wealth, and the purchasing power of their income is the same as it was before I snapped my fingers. People have the same jobs, producing and consuming the same goods and services, but everyone is getting paid twice as much and spending twice as much (in nominal terms). The money supply is the same, but the nominal transaction volume is doubled, so the velocity is also doubled.

    (Granted, by snapping my fingers, I have also effected a vast transfer of real wealth from creditors to debtors. All dollar-denominated debts, including bonds, bank accounts, credit card balances, etc., while nominally unaffected, are half their previous real value. What about stock prices? Well, companies with roughly equal amounts of debt and cash on hand would see their share prices double because all the lines of their income statement would approximately double and the value of all of their non-cash assets would double. Companies with high debt and low cash on hand would see their share prices increase to more than double, while companies with low debt and lots of cash on hand would see their share prices increase to less than double. Bank stocks would be in trouble. Their balance sheets would be nominally unaffected, so they would not be insolvent, but their income statements would not look good. Their fixed costs would double, but their (interest) income would remain the same. So most stock prices would approximately double, and any individual who invests most of his wealth in the stock market would see his wealth purchasing power approximately unaffected by my finger snap. Keep in mind that the vast majority of the world’s wealth is not in currency and bank accounts. It is in non-cash assets. For example, estimates of the aggregate wealth of US households are about 10x higher than USD-M2 and about 30x higher than the monetary base of USD. So the real wealth destroyed by my magic inflationary finger-snap is only a few percent of the aggregate wealth of US households — negligible for the purposes of a thought experiment.)

    Have you looked at the velocity of stable currencies other than USD? Are they also in the same order of magnitude (1-10)?

    What about velocity data from unstable currencies such as hyperinflation scenarios?

    I just don’t understand why, as you say, “This concept of monetary velocity is a key value for modern economies.” To me it seems like it’s maybe just a useless statistic.

    1. Dan McArdleDan McArdle Post author

      Hi Craig,

      Thanks for the detailed comments. I agree that there can potentially be large variation on the velocity, though I doubt it could be “orders of magnitude”. My reasoning is simply by comparison to existing mature, well-functioning economies.

      “Have you looked at the velocity of stable currencies other than USD? Are they also in the same order of magnitude (1-10)?”

      Yes. EUR and JPY are both in the low single digits (currently under 2).

      “What about velocity data from unstable currencies such as hyperinflation scenarios?”

      I have not looked at these. Might be interesting, but I don’t see the immediate relevance. There are a lot of other dynamics going on in hyper-inflating economies.

      In short, I understand your objections. You’re pointing out that V is just an accounting relationship, and you contend that it doesn’t convey any real meaning; ie, that it’s a derived value, not some fundamental measure of anything.

      I think the truth is in the middle. Developed economies appear to find an equilibrium of money supply and economic-output that causes V to show up in a certain range. Here’s the flipside to your thought-experiment:
      Suppose we’re just starting an economy, and one person (Joe) has 1 BTC, everyone agrees to use BTC as a medium of exchange, and there are several people wanting to trade goods. How does price get set? Probably *very* arbitrarily at first; this is very much your point. Maybe Joe decides to trade his apples for 0.5BTC. Maybe he thinks 0.001 BTC is fair. It doesn’t matter where he starts. Assuming the economy grows, if Joe had started prices low, other actors will set prices higher so that Joe has to release more BTC in order to buy the things he wants. If Joe set it high, other actors will have to lower prices to get their goods sold. Some equilibrium eventually emerges, independent of the first price-point.

      Building our economy from nothing on up, I would suggest that some underlying human preference for holding period, ability to execute economic transactions over certain time periods, some comfort-level of wealth/income versus spending, ease of actually executing a transaction, etc, all aggregate to form a balance/equilibrium between ideal money supply and the activity that money is used for. In your thought experiment, yes, your finger snap would cause a meaningless immediate change in the accounting identity. I contend that, over time, the prior equilibrium would be reached again. I find it unlikely that mature economies tending to exhibit similar figures is a coincidence. V seems an ok gross proxy for all the intricacies that we can’t model.

      So I’ll agree that V can be significantly different in either direction due to how fuzzy this stuff is, but I don’t see the likelihood of it being orders of magnitude off. While it’s true that Velocity is a simple accounting identity, it also seems reasonable that it tends to be a fuzzy reflection of some natural human-economic equilibrium, a conclusion that (admittedly cursory) data from mature economies supports.

      1. Craig

        Hi Dan,

        Thanks for your response. It is interesting that EUR and JPY have similar velocities, although I am not convinced that it isn’t just a coincidence (or maybe the result of intentional engineering of the money supply by central bankers “just in case velocity is important”.) Perhaps it is relevant that when the Euro was created, it was engineered to have a value very close to the value of a dollar.

        I’m afraid your “flipside” example didn’t make any sense to me. I see no reason why an equilibrium will emerge at all, much less an equilibrium with velocity in the single digits. Perhaps you could expand this example with a lot more detail. Or, even better, maybe you could describe what forces would “un-double” the velocity from my example of our (complicated) economy rather than a toy economy consisting of monopolies and monopsonies on apples and BTC.

        As for why unstable currencies might be relevant, my contention is that the price level of a monetary system is typically stabilized only by the efforts of a central bank. Indeed, maintaining a small, positive inflation rate is the primary raison d’être of central banks. Lacking a central bank, I don’t see any stabilizing forces for the value of Bitcoin. Like gold, its value is free to fluctuate arbitrarily. Thus, velocity data for unstable currencies might be a little relevant, although I agree that the more extreme examples will probably have too many other crazy variables at play to be useful.

        Anyway, hopefully you will consider writing a post (or maybe write me an email) explaining the forces that you think exist to stabilize a price level in the absence of central bank efforts. In it, I hope you will be very careful defining and explaining the rationale for “demand for money”. I have always found the meaning and justification for that term very ambiguous. I’m concerned that, when it comes to monetary theory, maybe economists tried to apply the concepts of supply and demand where they are not really applicable. Supply and demand are very handy tools for micro (where they typically measure flows, not stocks), but it’s not obvious to me that the mathematics of supply and demand have immediate applicability to monetary theory. This concern stems from what I see as a critical difference that the goods and services that are the subject of microanalyses have intrinsic utility, whereas money does not.

        Thanks again for your thoughts.

  2. Craig

    I thought of a slightly different way to explain my point.

    You seem to be assuming that velocity is fundamentally meaningful and relatively constant. You also seem to be using velocity as a proxy for “the aggregate rate of commerce in the economy” or some similarly-worded concept that actually has a useful meaning.

    Rather than assuming that velocity is a good proxy for anything useful, let’s look at the definition of velocity, V = Tn / M = P * Tr / M. As in your post, V is the velocity, M is the money supply, and Tn is the nominal value of all transaction occurring within the specified time frame. I then made the substitution Tn = P * Tr, where P is the price level and Tr is the real value of all transactions occurring in the time frame. If any of these variables can be said to be fundamentally representative of the state of an economy or the amount of economic activity that is occurring, surely it’s Tr, not V. Holding Tr and M constant in the rewritten equation, we see that, by definition, velocity moves in lockstep with inflation/deflation (as illustrated by the thought experiment of my previous comment). V & P are both arbitrary. They can each be increased or decreased by an equal factor without affecting anything meaningful about the real rate of economic activity in the economy.

    (For anyone reading this comment who is unfamiliar with the concept of price level, it can be thought of as follows. Consider a well-defined set of goods and services that includes a wide variety of items: some consumable commodities like a half a pound of salt, 2 gallons of drinking water, and a gallon of gasoline; some more durable commodities like a pound of aluminum, a pound of granite, and a 2×4 piece of lumber; some manufactured goods, like a screwdriver, a bicycle, a computer, and an airplane; and also services like a car wash, a haircut, the preparation of a meal. Once this set is defined, add up the total price (in your currency of choice) of everything in the set. This is the price level, P. It gives us a way of thinking about economic value in a way that is independent of the inflation/deflation behavior of a particular currency. We can convert between “real” values and “nominal” values by multiplying by P. In the event of inflation of the currency, all the prices in the set will increase by an equal percentage (the inflation rate) and thus P increases by that same percentage. Because the items in the set are numerous and diverse, changes in the prices of the items in the set relative to each other have little effect on P. In practice, actually defining which items comprise an appropriate set of goods and services is difficult and debatable, but luckily we don’t need to define exactly what’s in the set to understand the economic concepts related to inflation and money, we just need to assume that such a set exists.)

  3. Owe

    Maybe it would help to split the bitcoins into subsets -used for transactions and used for storage. I suppose it is reasonable to expect that the velocity will be stable within each subset. But in the aggregate I would expect that the velocity would decline substantially with increase of usage as store.Btw, where do the 10% come from, how would that compare to existing currencies? As an aside – bitcoin cannot work independently of a fiat currency. what i mean is, if bitcoin would be the only currency in the world, we would run into the mother of all deflationary scenarios.

  4. Doge wow

    Your huge unstated assumption is that Bitcoin is and will forever be 100% of cryptocurrency.

    Your assumption is already wrong.

    What will happen over time is that improved cryptocurrencies, with improved technology and improved appeal to consumers, will take a larger share of the cryptocurrency market that you estimate.

    Your estimate is one of the entire distributed cryptocurrency market, not bitcoin.

    1. Dan McArdleDan McArdle Post author

      Doge Wow,

      To be fair, I think bitcoin will hover around ~90% of the crypto market for the foreseeable future, as it has to date. Plenty of other experiments will boom and bust in the remaining 10% of crypto market-cap, as we’ve seen happen with hundreds of alts already.

      I explored why here: http://honestnode.com/debunking-debunking-bitcoin/
      and Erik Voorhees eloquently addresses this topic in the second half of his open letter to Peter Schiff here: http://www.reddit.com/r/Bitcoin/comments/1rxmk3/my

      The core misunderstanding in this domain is that most alts are just tweaks/clones that don’t offer enough (if any) improvement to get over bitcoin’s network effects. Further, it’s unlikely that an alt will come along with a significant enough breakthrough. Bitcoin’s blockchain consensus alt is a decades-in-the-making breakthrough in computer science. It’d probably take a similar breakthrough in how to do global consensus to generate a sufficiently new way of doing digital cash. Changing constants/mining-algs in the bitcoin codebase obviously doesn’t qualify.

      That said, I actually like Dogecoin, but not for any technical reason. I think it’s a fascinating (and initially hilarious) experiment in viral marketing. It certainly could capture a non-trivial portion of tipping/micro-trans use cases online (which I don’t include in my analysis above, as they’re essentially non-existent domains today). I kinda think Doge eventually eats Litecoin’s market-cap.

      And then there’s the “Bitcoin 2″ class: Mastercoin, Protoshares, Ethereum, etc. Ethereum seems the most interesting and potentially, long-run, can capture other entirely new domains not included in my analysis above (smart contracts, distributed markets, etc). The huge caveat here, though, is that being Turing-complete, Ethereum will have *many* more security and scalability questions than bitcoin, and will take quite a while to build confidence around. It’s likely that if it works at all, people will just move in and out of it as quickly as possible for many years, as opposed to being comfortable parking significant wealth in ether.

      In all these contexts, bitcoin likely remains “the gold standard” of crypto, even if other coins eat into some specific use-cases.

      1. Doge wow

        So in four years of distributed cryptocurrency, we’ve already reached “the gold standard”? Highly, highly unlikely. Unfortunately it’s difficult to have this discussion rationally because nearly everyone discussing bitcoin fervently is on the Buy side–so you’ll see many well-written articles telling you all about how bitcoin’s future is so rich, but little on the other end. Unfortunately, like articles favoring some stock, they’re heavily biased.

        Let me offer a couple examples of where your assumptions regarding Bitcoin could easily go wrong, regardless of technological competition:

        - You believe Bitcoin will gain wide acceptance, I believe it will not. One fundamental problem ignored by Bitcoin advocates is that most people do not own Bitcoin and the complexity makes it difficult for an average person to own. Bitcoins are heavily concentrated in a tiny portion of the population, at a far greater concentration than normal global wealth, even in the most inequal countries. A new cryptocurrency with superior distribution characteristics will overwhelm bitcoin. The worst part about bitcoin’s distribution is how much lies in the hands of thieves. It’s thieves who own the largest shares of bitcoin, not the people who sunk their resources into mining–when average people find this out, will they want to use bitcooin? Imagine how compelling a cryptocurrency that more people had a reasonable stake in, and was easy to use and secure, would be!

        - The heavy fluctuations in bitcoin price despite its high velocity, combined with a complete lack of regulation and a mismatched market indicates that it’s very susceptible to being hacked to death by financial powerhouses. Imagine what kind of pump-and-dump a large investment bank could do on a completely unregulated market. Imagine what sorts of hacks can go beyond what we’ve seen happen to flexcoin and mtgox. Think of how easily a sophisticated hacker or government could make 51% attacks and freeze the bitcoin economy.

        Look outside the bitcoin echo chamber and think for yourself. Imagine how bitcoin looks to someone who is sophisticated but holds 0 BTC.

        1. Dan McArdleDan McArdle Post author

          As stated, the point of my original analysis was to explore how to calculate success-case fair-value. Also as stated, discussion of whether we get to the success case, and what it takes to get there, are beyond the scope of the article.

          But to briefly address your comments… Bitcoin is very young. The issues you cite – volatility, distribution, online-wallet security, ease-of-use, etc – are no surprise for a completely new domain in its infancy, and should moderate over time.

          And of course it could fail! It’s always been an experiment… Just realize that nearly ever criticism you’ve laid on bitcoin applies equally or more so to alt coins….and most alts have additional problems as well.

  5. Freedboy

    Dear Dan,

    This article has come closest to any I have found that approaches the valuation of bitcoin in the way that I think is correct: back calculating its required value from estimates of total transaction value. I have done something quite similar in this article: http://www.scribd.com/doc/217902157/Valuing-Bitcoin

    There are a couple of areas where we differ in approach.

    Velocity. You calculate the velocity of bitcoin by looking at transactions on the blockchain, and compare this to the velocity of US M2. Firstly I would say that looking at transactions on the blockchain is not the correct base. When economists calculate velocity, they do so by taking the inverse of the proportion of money supply, M1 or M2 say, to GDP. GDP is not equivalent to the value of all transactions, it is equivalent to the value of all final transactions – intermediate transactions are not counted. Estimates of total global transaction volume are ca.$600trn, while global GDP is more like $60trn. Moreover, simple transfers of money are not even included in the $600trn. The 200,000 bitcoin per day ‘transaction volume’ you quote is for all transfers of bitcoin, be it for people trading bitcoin, transferring between accounts, making remittances etc. It therefore grossly overestimates the value of bitcoin transactions, and therefore grossly overestimates bitcoin velocity. The only transactions that should count if we are comparing to US velocity are ones for actual real world goods and services, e.g buying something on Overstock.com. In my article I put a very rough estimate on this number today as $1bn for the year – this is the ‘bit coin GDP’. Recalculating bitcoin’s velocity using your method, the correct answer is therefore more like 0.1.

    The second area where we differ is I compare to M1. I see the merits to using M2, but I reasoned that all the speculative investment in bitcoin is more akin to investing in gold or houses, and should be stripped out of money supply completely, leaving us with only those coins that people keep at hand for actual spending. Consequently, I assumed that as of today there are 2 million ‘M1′ bitcoins in use, with an approximate velocity of 1. This I compare to the velocity of USD M1 of around 7.

    The rest of our analysis was largely the same, and actually I assumed that bitcoin velocity will move towards 7 when it does start to become widely used. Consequently my estimated value of one bitcoin today was $3,544, which is very close to yours!



Leave a Reply to Dan McArdle Cancel reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>