The Bear Case for Crypto, Part II: The Great Bank Run

This is the latest in my ICO mania series of blog posts. I have written a follow-up to this post, The Bear Case for Crypto, Part IV: Bitprime Loans.


A couple of months ago, when the price of 1 Bitcoin hovered around a relatively benign $5,000, I sat down for coffee with an investor who runs a fairly substantial chunk of money. Over the course of that chat, he told me that nobody in his business could foresee how the price of cryptocurrencies would ever fall, given the rapid increases in demand and the way that technology adoption is thought to work.

Hearing something like this come out of the mouth of a professional investor is extremely disconcerting.

First, I’m a “blockchain without Bitcoin” guy, and the longer this cryptocurrency bull run goes on, the longer I have to wait for due attention to be paid to utility-driven applications of blockchain technology.

Second, this is bubble thinking. My spirit animal, the noble woodchuck (a type of marmot), is a solitary critter, not a herd animal. I therefore despise bubble thinking.

Third and finally, being an English lawyer, I foresaw a number of ways that the Bitcoin/ICO bubble could collapse which were regulatory in nature. I sat down and wrote a blog post about it called The Bear Case for Crypto. About 40,000 of you have read it so far.

In that post, I foresaw the day when regulators would crack down on bad actors in the space, causing significant disruption. I named this scenario the “Zombie Marmot Apocalypse,” because I really like marmots and use the word “marmot” at every possible opportunity, whether appropriate or not.

As Bitcoin’s price keeps rising but the number of enforcement actions does not, I see my projection was premature. I would like to now revise it, and propose a more likely collapse scenario. The Zombie Marmot Apocalypse is still going to happen. However, before it happens, I believe we are going to see…

The Great Marmot Bank Run

The “crypto ecosystem,” of course, has no central bank. Nor is it a bank. That does not mean that we should not think of it, in terms of how risk is accumulating as Bitcoin’s price rises, as being something very much like a bank with substantial exposure to a risky asset whose risk increases by the day.

In previous discussions about market misconduct in our space, much has been made of the fact that this exchange or that one may be running a fractional-reserve.

What is not said so often is that the slogan “be your own bank,” which was common among Bitcoiners back in 2013-14 before transaction fees rose and it was rebranded “digital gold,” may be more accurate than we think. *The cryptocurrency space itself* might be seen as a kind of fractional-reserve system writ large.

The price applicable to everyone’s bitcoin holdings is determined by the then-prevailing spot rate. As the price of Bitcoin rises based on relatively thin trading at the margins, Bitcoin hodlers (yes it’s spelled “hodl,” not hold), who measure their Bitcoin wealth in dollars, mark their Bitcoin “deposits” to market on the assumption they can withdraw at any time by exchanging to USD. As their holdings appreciate, they introduce no additional USD into the system.

But if we zoom out and look at this from a systemic perspective, as with a bank, a demand deposit is properly recorded as a liability of the bank. In Bitcoin’s case, holders are effectively like depositors, whose holdings are dollar-denominated and need to be convertible into dollars on demand (given the fact that economic activity unrelated to speculation on the Bitcoin blockchain is, for all intents and purposes, non-existent). This is the case no matter how high the price of a bitcoin soars, and despite the fact that nothing about this “decentralized” arrangement guarantees that there is an increase in the aggregate number of dollars available to satisfy vastly inflated (dollar-denominated) claims, either systemwide or by reference to particular depositaries.

Holders expect that Bitcoin intermediaries/exchangers (exchanges, wallets, traders) will be in a position to provide them with instant dollar liquidity when they wish to margin trade or take their profits (dollar liquidity which, I am reliably informed, is provided in part by mainstream commercial banks on a very short-term basis at very high rates). With some notable exceptions (i.e. Tether, which is expressed to not be redeemable), Bitcoin companies’ ability to provide this liquidity will be dependent on either having sufficient deposits to meet fiat conversion demand, or banks continuing to extend credit so that they can do so.

We should ask ourselves: with the market cap of Bitcoin rising by $10 billion a day, is it realistic to expect that exchanges, trading counterparties and other market participants are prudently hiving away their profits to stave off a liquidity crunch? Or is it more likely they are reinvesting it into their businesses, or even into Bitcoin itself?

The most obvious way, in my view, that Bitcoin mania could turn into Bitcoin panic is when a Bitcoin depositor goes to sell their BTC for dollars – and there are no dollars available to satisfy the sell order.

In the current environment, there are a number of ways such a shock could arise. To begin with, I seriously question the intermediaries’ and traders’ ability to top up their USD holdings quickly enough to catch up with their depositors’ and counterparties’ paper gains in Bitcoin. There is also the possibility that, in the event of a correction or an enforcement action, a risk-averse bank to a major service provider withdraws either credit or banking services to that provider, compromising that service provider’s ability to convert BTC into dollars, provide margin lending, or even to hold fiat deposits at all.  Or an exchange could get shut down for trading unregistered securities, a possibility ever since the SEC classified “DAO Tokens” as securities over the summer.

Whatever sets off the panic, the consequence will be that the market may not have confidence that the Bitcoin reserves held by fiat on/offramps will be able to be sold on that platform at any price. This will place pressure on other platforms and their diminishing pools of liquidity as investors run for the exits.

This is different from a mere fall in price, although a liquidity crunch will precipitate that. This is a question of a wholesale withdrawal of dollar liquidity so severe that it may become impossible to shift Bitcoins – I say again – at any price, much as it was impossible to secure an interbank loan at any price during the peak of the 2008 credit crunch brought on by the collapse of the subprime mortgage bubble.


A thread on Hacker News reveals that people think everything in Bitcoin is bilateral and balances out, so there’s no liquidity risk. Incorrect. Borrowing from that thread, most people currently think Bitcoin works like this:

  • Day 1: Alice buys 1,000 Bitcoin from Bob for 1 USD on a street corner
  • Day 2: Market price on the street corner is 1 Bitcoin for 1,000 USD
  • Day 3: Alice sells 1000 Bitcoin to Charlie for 1,000,000 USD on a street corner

This only works as long as Charlie keeps showing up with new dollars. In a panic, the “street corner” or open bazaar suddenly turns into little pools of jealously guarded liquidity trying to stay open. Charlie does not show up. Instead, the opposite occurs. e.g.

  • Day 1: Alice buys 1,000 Bitcoin from Bob for 1 USD per BTC in Fred’s Bitcoin Emporium
  • Day 2: Market price on the street corner rises to 1 Bitcoin for 1,000 USD
  • Day 3: Diana deposits 1,000 BTC she bought in 2010 on Reddit at Fred’s Bitcoin Emporium
  • Day 4: Diana and Alice ask Fred, or Bob, to convert their Bitcoins to USD and withdraw
  • Day 5: *liquidity shock, such as conversions being halted, exchange getting shut down by the SEC, or very sharp correction*
  • Day 6: Panic

Unless Fred’s Bitcoin Emporium has cash of its own or a bank liquidity facility, the platform only has Bob’s $1,000 to meet $2,000,000 in Bitcoin deposit demand being targeted at its platform, assuming Bob has not withdrawn it via bank wire and is still willing to trade it.

When a shock pushes enough people towards the exit at the same time, that is when people will realize that this is the case, and there are not enough dollars in the Bitcoin economy to redeem their balances. Delusionally, they may not be willing to part with it for pennies on the dollar per BTC, either, so bid-ask spreads will widen. As this phenomenon spreads, individual platforms will be placed under pressure as their sources of liquidity are tapped out or flee.

A loss in confidence in one platform would lead to loss of confidence in others.

And that’s what the beginning of a liquidity crisis looks like. This one will be glorious.

… and Consequences

Bitcoin only works as a conspiracy: we HODL together or we die together. The higher its price rises, the greater the conspirators’ incentives are to cheat. The more Bitcoin soars, the greater the dollar/BTC notional mismatch, easing the requirements to trigger a collapse while making it more likely that one will occur.

The only things of which we can be certain are the immorality of promoting such a scheme and the fact that when it ends it will end badly. The $64 billion question is where the end will begin. If not through regulatory intervention, then we could expect some weak hands to fold in the form of:

  • ICO startups sitting on $3-10 billion of Bitcoin, or
  • third party services that hold customer BTC and USD balances, which can be freely exchanged without backing out their exposure to traders on an exchange like GDAX,

who possess enough BTC under sufficiently central control to kick-start a liquidity crunch. In a panic scenario these businesses will be put under pressure to sell, whether to preserve their coin fundraises (in the case of ICOs) or to allow their users to withdraw USD they’ve been promised (in the case of a service which allows seamless switching between USD and BTC and hasn’t managed its liquidity risk properly).

Companies which hold two sets of balances that aren’t backed out, in particular, will be faced with the choice of either borrowing, selling assets, or halting withdrawals. Those are not good choices.

In a falling-knife scenario which becomes more likely with each passing day, liquidity will likely be draining out of exchanges as well. Any such run following the usual pattern will be self-reinforcing and highly contagious; sell pressure from liquidity-starved companies will drive sell pressure among long-term holders, and vice-versa.

As a final note: this is not a Bitcoin obituary; Bitcoin will be fine. However, its price may change a bit. And if those price changes are violent and in a downward direction, it is not difficult to see how institutional service provision to the Bitcoin space (which is proceeding with seeming alacrity) could both trigger a liquidity crisis in Bitcoinland when withdrawn, and introduce systemic risk to the mainstream financial system as well. Banks, if they are being even remotely prudent, should be in the process of modeling the conditions under which they are willing to continue to be exposed and/or continue to extend dollar liquidity to the sector.

How the next 12 months will play out

I therefore modify my prediction from September. The Great Marmot Bank Run will happen first. When and how, and what shock triggers it, no man can say. But happen it will.

A veritable orgy of regulatory intervention, a scenario I call the “Zombie Marmot Apocalypse,” will follow. It is long overdue. The introduction of millions of ordinary retail investors to the market (300,000 last week on Coinbase alone!) brings Bitcoin something absent from the strictly nerd-only crashes of the past: seemingly helpless and innocent victims.

Often described in cryptocurrency circles as “Grandma,” she is the most terrifying litigant/complainant in the entire Universe. This is because Grandma has the power to band together with other Grandmas to become an unstoppable and very angry mob of Grandmas, pitchforks in hand, baying for blood.

And blood she shall have. Grandma, qua depositor or investor, will call down swarms of attorneys to sue everyone in sight and claw back what losses they can. Frauds will unravel. Law enforcement will finally get political backing for additional resources to investigate. Promoters’ reputations will lie in tatters. Years of chaos, legal wrangling and Congressional hearings will follow.

If we’re lucky, the mainstream financial sector will have had the good sense to limit its exposure, so the crisis will not spill into the real economy.

After all that, fifteen-year technology adoption curves, rather than insane, short-term speculative bubble curves, will once again rule the day.

Just remember that you heard it here first.

EDIT: A good point from my Twitter buddy Bruce Kleinman:

Additional accelerant: more sophisticated folks might *try* to move their Bitcoin off of imperiled exchanges into their own wallet app. That it is a very narrow pipe, as you know, capable of handling ~250 transaction per minute. Transaction backlog (mempool) would explode higher, with transaction fees going ballistic. That would add to the panic.


Over the past 48 hours, with the Bitcoin price continuing its exponential rise, transaction activity on the actual Bitcoin blockchain has been modest. Takeaway? Virtually all transactions are taking place in exchange dark pools. This reinforces your thesis, as those represent “ponds” of liquidity rather than a pool of liquidity.

EDIT 2: Good thread from Craig Wright although I sense some hints that his answer to the problem of scaling is a fork of Bitcoin called “Bitcoin Cash.”


I conclude this post, as I conclude all my posts about ICOs, with a picture of some adorable Vancouver Island marmots.




49 thoughts on “The Bear Case for Crypto, Part II: The Great Bank Run”

  1. this is utopic. it happens only when all people withdraw cash at the same time.
    It happened in Greece already and in some special conditions.

  2. We don’t need fiat in order to transact thus there will be no bank run. Crypto currency users will just exchange their bitcoin for another coin if it plummets. The current fractional reserve system ensures that there are not enough dollars going around so why is this a problem for bitcoin? It’s not.

  3. Very interesting commentary, thanks.

    I have a question about how the exchanges handle trades, that relates to your thesis. I thought that, on your run of the mill exchange, honest, no margin, that they were simply a place to trade with other users of the exchange. So to convert your bitcoin to dollars, at a given price x, there has to be a buyer at price x. If not, then the offer price has to lowered until it meets a real bid.

    Presumably a potential buyer had to at some point put real currency on the exchange. If the structure I’ve outlined reflects reality, all liquidity is provided by other users, and the exchange is not taking on liability for users accounts. Now, maybe the exchange does do some market making, that should be kept to a minimum, and balance out, if there is an active market.

    Your description makes it sound like the exchanges have make the trade directly with the user, at market price, if someone wants to convert bitcoin to dollars. Is that really the case?

    1. On an exchange, no, at least not in the beginning.
      The mismatch is systemic in that it is almost certain that the notional amount of BTC in dollars is substantially higher than the actual amount of dollars presently prepared to buy BTC. Even if you had a trade where a user deposited dollars and bought Bitcoin, say, six months ago, as the price of Bitcoin goes parabolic the crypto ecosystem as a whole begins looking a whole lot like a fractional reserve (in that the “deposits” of Bitcoin which people expect to be able to redeem are substantially higher than the sum of all available dollars which could redeem them).
      Exchanges do enter into contracts with their users for e.g. margin trades. They may also trade on their own accounts. Ultimately we can’t know what their exposure is due to the opacity of the market.

  4. Say a friend had deposited some USD on an exchange and is waiting for the bitpocalypse – how would your thesis affect their attempts to buy cryptocoin in the event of liquidity drying up?

    Would their exchange allow them to buy the crypto-currency off chain in one of their darkpools, or would they have already stolen all the deposited money to cover other users transactions?

    1. Good question. Would depend on the exchange/intermediary’s liquidity management. I am more concerned with systems that are mainly focused on USDBTC rather than crypto to crypto as it’s likely institutional USD lending which greases the wheels, so to speak.

  5. You don’t ask Coinbase or GDAX to “convert” your BTC into USD. You sell your BTC to someone else offering you fiat in exchange. You’re not relaying on Coinbase’s liquidity like a traditional bank. It’s an exchange, not a bank!

    1. Changes little. In a full blown liquidity crisis contagion will spread to traders as well.

      Have made some amendments to make clearer what I think the bank is for the purposes of the “Great Bank Run”: Coinbase isn’t the bank, the ecosystem is the bank. Coinbase and other venues are sources of dollar liquidity for that bank.

    2. I go to Coinbase, I say I want to sell x BTC, Coinbase quotes me a price that’s good for about a minute, and I can transact against that price and Coinbase transfers $y to my checking account for my x BTC on deposit. Coinbase may turn around and flip the x BTC on Gdax, but my counterparty is Coinbase. “You sell your BTC to someone else offering you fiat in exchange,” correct, Coinbase is offering me fiat in exchange.

  6. In the 4 years I have traded bitcoins, not once did I use an exchange. Instead there are many P2P markets where Bitcoin holders can exchange BTC for Fiat directly.
    Maybe you are too focused on how things work in the US, but I can assure you that Bitcoin works perfectly fine without exchanges in most places around the world.
    Check out this site to get an idea: .
    You can also do it at which is a well known market in Europe.

    1. I agree. Bitcoin works fine without exchanges. It’s just very unlikely that Bitcoin would be worth $10,000 a coin this morning without them and the direct US dollar on/offramps they provide.

  7. You seem very focused on withdrawals to USD. Here’s a mindfuck for you: when the time comes when a long term holder wants to redeem profit, they wont even need USD. If everything we want to buy is available in cryptocurrency, what need do we have of ‘old’ money?

  8. Preston, I’m wondering if you have any opinions as to whether a sudden bearish turn in the financial markets might trigger a run on the bank as well in the crypto market, and whether the correlation between crypto-returns and regular market returns will show a less than attractive correlation this time around. If more insti and retail investors are allocating a piece of their portfolio to crypto, should the stock market crash, would that also cause a liquidity run in the crypto space?

  9. The unspoken rule of the crypto world is: “do not convert back to legacy tokens everyone at once”. There will never be a bank run back to the old system. The real bank run into the crypto will continue happening… we don’t care if there is not enough fiat backing for all the invented money that arose from nothing.

  10. Well written thoughtful article. Same thing could happen to many currencies and assets. It’s always about confidence and matching buyer and seller

  11. I think your article is stating the obvious. If all users want to sell their btc at the same time, it’s obvious that price will collapse and it is very difficult to find a buyer. What’s the news?

  12. Hi, Thanks for the fascinating yet horrifying read.

    I read Bitfinexeds articles on Tether roughly 1 week ago, after which I began to wonder if something like a bank run could happen. I then came across this article which eloquently described my worst nightmare.

    Kraken states the following on their practices page.

    Financial Security
    We maintain full reserves — a “bank run” is an impossibility.
    Customer funds reside in a bank account separate from our operations account, and fees are transferred on a daily basis.
    Customer funds cannot be borrowed to fund operations, nor can they be lent, even for margin trading on our own platform.
    We have solid relations with our bank, and an agreement is in place allowing us to wind down our account in an orderly fashion, should our relationship ever come to an end.
    We are pursuing multiple banking partnerships to build in some financial redundancy — so that even should the above situation come to pass, our daily operations will not be interrupted.

    I’m slightly reassured by this due to their track record with Mount GOX. However, what are your thoughts on these statements? Are they worth the pixels they’re displayed on?

    One other factor I can see that may alleviate a bank run or make it worse is that they will only allow the withdrawal of 25k per day and 100k per month. This is the same across many exchanges.

    Any wisdom is greatly appreciated.


Comments are closed.