Matthew Hougan from Bitwise on Why 95% of Reported Bitcoin Trade Volume is Fake


Peter McCormack: Hi there Matt. How are you?

Matthew Hougan: I’m doing great. How are you?

Peter McCormack: Yeah, I’m doing pretty good. So you are in my country and I’m in your country!

Matthew Hougan: Something is wrong here! But that’s right, I’m over in London.

Peter McCormack: Well thank you for doing this at short notice. I saw Jake Chervinsky retweet your research piece and I thought it would be good to get you on the phone straight away, discuss the report and get it out to people this weekend. So it would be useful if you could start by explaining who Bitwise is, why you are important, your role there and the background to this report.

Matthew Hougan: Sure. Bitwise is a specialist crypto asset manager. We created the first cryptocurrency index fund in 2017 and now have over 700 limited partners invested in that fund. We’ve been running a series of cryptocurrency funds and indexes ever since. We’re 15 people based out of San Francisco and we have also been at the forefront of pushing to get an ETF approved by the SEC. Bitwise is a little bit unique in that about half of us come from software backgrounds, expertise in crypto. Half of us come from the financial industry with expertise, particularly in ETFs.

So we’ve been trying to move forward on the ETF space and this presentation was part of that effort. We made a presentation earlier this week to the SEC to support our Bitcoin ETF filing. As part of that, the SEC made our presentation public, so our 226 pages of research got published out to the world.

Peter McCormack: Obviously there have been multiple ETF applications which have all been either rejected or delayed. Can you explain the status of your application and how it differs from the others?

Matthew Hougan: Sure. So there have been multiple ETF applications as you’ve said. They break down into two groups. You can think of them as futures based Bitcoin ETF applications and then direct Bitcoin holding or physical Bitcoin applications. The SEC asked everyone with a futures-based application to withdraw theirs. So they were all withdrawn. In the direct Bitcoin, the really relevant ones there was the Winklevoss proposal, which was rejected last year.

Then there is our proposal and VanEck’s proposal. Those are the two ones that are alive. We have different approaches. I think both are interesting. I have a lot of respect for VanEck as a firm, but those are the two that are currently live. Where they are in the process, if people don’t know how the ETF application process goes, you first file a prospectus and then the exchange where you’re going to list that ETF files, what’s called a 19b-4 and that trigger a sort of statutory review where the SEC has a certain amount of time to review that application, 240 days.

So both VanEck and us are a little over or about a month into that 240 day period. That was the auspices under which we met them to talk about our findings.

Peter McCormack: I have obviously met with Hesta Peirce at the SEC and I also interviewed Gabor Gurbacs from VanEck. So I am kind of aware of the ETF process, but I’ve read your report and the TLDR is that 95% of the reported volume is fake, but lots of good news. So I was like, okay… But I was immediately concerned about the fake volume. I’m used to hearing about billions of dollars of trade volume, but this is just not true, right?

Matthew Hougan: Yeah, this isn’t true. It’s something to walk into the SEC and say, “hey, 95% of the reported volume is fake, so you should approve our ETF”. That’s quite a feat. But it is true and it actually makes sense. So the reported volumes, spot volume of Bitcoin that that’s claimed on sites like Coin Market Cap is about $6 billion a day. That means all Bitcoin would turn over about every two weeks, for something that is designed primarily as a long term store of value, that’s the primary use case right now, that would be absurd.

If you look at gold, gold’s daily turnover is about 0.5% of its market cap, instead of 8% which is the reported Bitcoin figure. What we found is that the real volume for Bitcoin is about $270 million per day on the spot exchanges. We can talk about what else there might be, but on the spot exchanges, $270 million and that it’s turnover roughly reflects that of gold. It’s about 0.4% of its market cap per day.

So you have to get past this echo chamber of nonsense numbers that has been surrounding the crypto industry and get down to the real volume. Once you do, it makes sense and the market is extremely efficient, among the most efficient in the world and you’re left with a beautiful sort of institutional style market that we can build off.

Peter McCormack: Okay. So that is a very interesting and positive way to look at it and I have been in the market for about two and a half years now. While there have been concerns expressed about the size of the fake volume, this is way beyond anything that I or I think anyone else would have predicted. So I guess you suspected something was up when compiling the report and you felt that there was value in taking a closer look at this?

Matthew Hougan: Yeah that’s right. We sort of went down the path… Our ETF approach all along has been to gather data from as many real exchanges as possible to create the equivalent of what we call a crypto consolidated tape. In the equity market, the consolidated tape is the aggregate market for every stock and we wanted the same thing in crypto, just to capture the full market.

So we started this process of evaluating all of these exchanges and the data doesn’t lie. Once you pull down the trade data and the order book data, these exchanges aren’t even doing a good job of faking volume! I went through the same process that I think a lot of people will go through when they hear this, which is sort of the seven stages of grief. Where first it’s shock and disbelief, then it’s maybe anger and horror and gradually you start to accept it and then you realize that it’s a positive thing.

So it is a process, but it’s just true that data, pretty comprehensively shows that this is the real spot market and I think it’s the market where we have to reset to and build from there.

Peter McCormack: But your report is out there, as of now it has been retweeted 471 times. This is just your version, but there will be other people who’ve shared the article or retweeted it. But I haven’t seen any impact on the price.

Matthew Hougan: Yeah, that’s because I think this isn’t news. I think there’s been this broad feeling in the industry that there’s a lot of fake volumes out there. Maybe it’s not 95% that people thought, maybe they thought it was 50 or 60%, but I think there’s almost a cathartic relief that we’ve demonstrated with hard data what the reality is. Like you said at the beginning, in the report, there is a lot of good news in the report, once you get away from that top line story. So I think that’s actually not news. The news is that the underlying market is extremely efficient. It’s just you have to get past that first sort of shocking headline.

Peter McCormack: But it’s almost like the fake volume isn’t having any impact on price, as they aren’t real buy or sell orders in spot markets that are being executed. Therefore the fake volume does not do anything but convince people that these exchanges have volume. But there’s no real impact on the market price.

Matthew Hougan: It doesn’t impact the market price at all and it’s just a way for those exchanges to get listing fees. It’s like in the equity market, if someone put up a random website and started printing numbers from IBM, no one would pay attention. Or if you see those crazy ads on TV where they say, “I bought an iPhone for one penny”, that doesn’t change the real trading price of an iPhone or what you pay at the store.

These are numbers that the reported volume and the prices and the spreads being quoted on these exchanges are numbers that just no one who is really playing in the institutional crypto market pays attention to. They’re a cacophony of meaningless numbers, so they don’t influence the price at all. At least that’s what the data shows.

Peter McCormack: is the reporting of fake volume illegal?

Matthew Hougan: That’s a great question. Most of these exchanges that report fake volume have unknown domiciles. They don’t have hard and fast headquarters where you can pin them down, so I don’t know. I would think, it sounds wrong, but it depends on where those exchanges are located. If they even have locations to be served enforcement actions. I’d note that most of these exchanges that are reporting real volume, it’s easy to find out where they’re located.

A lot of them have raised venture capital, a lot of them are regulated by FinCEN, by the BitLicense, by the FCA in Europe. So is it illegal? I don’t actually know the answer to that because I don’t know what jurisdiction most of these reported exchanges exist in.

Peter McCormack: What if it was a US exchange?

Matthew Hougan: That’s a good question. I would think it probably trips over certain laws, but I’m not an expert in that space.

Peter McCormack: Okay. Are these all Bitcoin trades, which are fiat to Bitcoin or do they also include fiat to altcoin?

Matthew Hougan: We didn’t include Bitcoin to altcoin. It just includes fiat and stable coins. We didn’t look at altcoin trading and some of these exchanges very well may have legitimate altcoin trading, that was just beyond the scope because we were focused on our Bitcoin ETF.

Peter McCormack: You have said that there are 10 exchanges which make up most of the trading. I expect that these are the obvious ones, Coinbase, Kraken, Bitfinex etc. You said that most are regulated?

Matthew Hougan: They’re regulated in one way or the other. So they’re not regulated as exchanges, but they’re regulated as money services businesses, which actually carries a fair number of obligations in terms of reporting and structure. 5 of them have BitLicenses and the BitLicense has a lot of requirements around AML/KYC and other regulatory statutes. So one exchange has neither of those, but nine of the other ones have the money services business license, and five have the BitLicense.

Peter McCormack: So I’m guessing that for your ETF application, you can isolate the real trading, which is relevant to your ETF application.

Matthew Hougan: Yeah, what it does for our application is, it leaves you with a market that is, as I said earlier, extremely efficient. One interesting fact, on Coinbase at least, the median spread on Bitcoin is a penny on a $4,000 price. That means it has the tightest quoted spread of any financial instrument in the world, tighter than any stock, tighter than any futures.

So you’re left with this market that’s extremely efficient and you’re also left with a market where the CME regulated Bitcoin futures looks pretty big. There’s about $90 million in regulated Bitcoin futures. That’s bigger than the second largest real spot exchange. So what we’re trying to do and what we tried to articulate in our presentation is that we’ve met or we’re trying to meet the SEC requirements to list an ETF, which include that the market is either uniquely resistant to market manipulation or there is a big futures market. We tried to articulate that we’ve met those and hopefully the SEC agrees, but we’ll find out.

Peter McCormack: Why is an efficient market so important?

Matthew Hougan: Well, so the SEC is concerned about the Bitcoin market because they’re interested in protecting investors. They don’t want investors exposed to a market where there is rampant and serious manipulation, that can’t be surveilled and corrected, and that’s right. The SEC approving an ETF is to some degree a seal of good housekeeping and we want to protect investors or at least have them know the risks that they’re getting into.

Having a more regulated, more efficient market gives, not just the SEC, but also me more confidence, that this could be a market that it could make sense in an ETF. But still, there are many miles to go before the SEC comes to a decision there and many additional factors as well.

Peter McCormack: So when you started your research, how did you identify fake versus real volume? What were the patterns you were looking for and what were the patterns you noticed?

Matthew Hougan: That’s a great question. So we pulled all the spread and order book and trade data from all 81 of the largest exchanges. Then there were some easy top line things that stand out and then some more rigorous data things. So as a top line thing, I mentioned that, Coinbase Bitcoin trades at a one penny spread. On Coinbene, which is the largest reported exchange, which claims to have 18 times the volume of Coinbase Pro, the median spread is around $15. It wasn’t clear to us why someone would trade on an exchange where the median spread was 1500 times larger than it is on Coinbase Pro.

But the more rigorous analysis was to look at things like what we call trade volume histograms. Trade volume histograms is a fancy way of saying how much Bitcoin trades on these exchanges at various sizes. When you look at a real exchange like Coinbase Pro, you see a natural pattern. Most trading takes place in small amounts of Bitcoin and then there’s spikes at whole Bitcoin. So you’re more likely to trade one Bitcoin than you are to trade 0.9 or 1.1. That’s just a natural behavioural response.

When you look at exchanges that are faking volume, what you see is neither of those behavioural patterns. So there’s often the case that they claim more people are trading say eight or nine or 10 Bitcoin than are trading one Bitcoin and there’s no spikiness around a whole Bitcoin. Someone on Twitter made a very cogent comment and said that “often times when people are faking data, they have a hard time faking human behaviour because either they make it totally random, which is not how humans act or they make it totally programmatic, which is not how humans act either”.

We have these behavioural nuances that jump out in the data. So that was just one example. There were plenty of other ways that we proved it. We proved it multiple times for all the exchanges. But that’s a pretty common example.

Peter McCormack: But humans can be pretty resourceful when they need to be. So have you not just now provided them with the method and the patterns for faking volumes, so it appears real?

Matthew Hougan: I totally expect exchanges to up their game on faking volume, which is why we published screenshots of some of these exchanges and why we memorialized this data. In fact, it is the case that we’ve ever seen exchanges sort of change the way they’re faking volume over time. But hopefully by getting out ahead of it and sort of establishing it, what we want is the industry to set standards on what exchanges they’re counting as real volume and sort of reverse the burden of proof. Up until now, we’ve been accepting everything as real. Maybe it’s time to start making exchanges prove that their volume is real before we start counting them.

Peter McCormack: Do you think there are a set of standards for which exchanges could comply with, to improve the industry?

Matthew Hougan: Oh, that’s an interesting question. We’re not at that point yet. I do I think the industry is maturing and as it moves into a more regulated phase and more open phase and we start seeing more people like Bitwise analyzing the volume, I think it will take care of itself. My honest sense is that this will wash out of the crypto ecosystem in the next 12 months and we’ll all be dealing with the shared set of data and responsibilities.

It’s not surprising, anytime there’s a big boom, you get excesses. We saw excesses in the ICO market, which are getting cleaned out through regulatory action and improvements by the industry. We’re seeing the results of excesses on the data side as well and this is just a natural part of the process of cleaning out. So I think it’s going to happen. I think some of it may be regulatory on down and some of it may just be the industry improving from the bottom up.

Peter McCormack: Okay. I’ve only just got to your report, only to skim the report and reach out to you straight away. There are some quite serious allegations against these exchanges. Have you seen any response from them? Any negative feedback?

Matthew Hougan: Mostly the feedback has been overwhelmingly positive from the community and the reason is that I think most of the people in the community have already come to a similar set of conclusions and they just loved seeing it laid out with clean and clear data. People who are serious about building the crypto industry want to see that industry be built on truth. So almost all the comments have been positive.

Now we do expect, we haven’t seen, but I wouldn’t be surprised if some of the exchanges come back with different points of view or different arguments. We’ll see if that transpires in the coming days. We feel very confident about our analysis and feel that it’s very tight. But we look forward to more engagement. If part of our analysis turns out to be wrong, we’ll happily update it. But I think we feel very confident and so far the response from the industry has been very positive about the work we’ve done.

Peter McCormack: Yeah and I have the feeling that there is a healthy argument for Bitwise to have a Coin Market Cap competitor on your website. You may be doing this, you may not be, you might not want to talk about it. Is this something that is in the works?

Matthew Hougan: Yeah, so we actually launched a website, funny you raise that, It’s actually not a commercial enterprise. We’re just creating it as a service to the industry. We’re in the business of being an asset manager and an index provider. Those interests will be served if we’re building on real data. So we have created this so people can see and journalists can see what the real volume is.

Part of what has got us so fired up about this, is we saw these fake volume numbers being reported in places like the Wall Street Journal, the Financial Times and the New York Times and we just realized journalists, the industry, investors all need a better real source to go to and so we created that as an offshoot of this research.

Peter McCormack: Okay. I wasn’t aware. I’m just looking now. It’s quite interesting, but it’s limited on the data. Are there any plans to expand it?

Matthew Hougan: We will l see. We just wanted to get something up, so after this report came out, people had a way to access it. So maybe over time, we’d love to build it out. We’ll see.

Peter McCormack: Let’s get to the good news then because there’s plenty of it. You’ve said the arbitrage has improved significantly?

Matthew Hougan: Yeah, absolutely. So the whole of the crypto industry, the market has sort of professionalized over the last 18 months, particularly the launch of futures. The launch of institutional lending in the crypto market and the entry of some major institutional market makers like Jane Street has brought the arbitrage between these real exchanges down to an almost perfect efficiency.

If you look at these exchanges over the course of the last year, there’ve been been a few flashes, but almost no moments any exchange has been 1% away from the global integrated price for more than a hundred seconds, virtually 24/7, 365 days a year. These exchanges have been trading at exactly the same price around the clock. It is a nearly perfectly arbitraged market and really impressive the degree to which that is true.

Peter McCormack: So Bitmex isn’t listed here, but this is the place where most people I know trade. What were the reasons for considering listing it or not listening to?

Matthew Hougan: Oh, that’s a great question. Bitmex we definitely hear about people trading on that. That’s a legitimate question. It’s a real exchange. This was focused exclusively on the global spot market. Bitmex, as you know, is a swap market and it does leverage swaps as well. It’s a little bit hard to compare volumes when you’re doing highly leveraged swaps. I come from the traditional equity world and in the equity world, the biggest traded futures contract is called the S&P E-mini futures contract and its traded at high leverage. The theoretical volume on that every day is well over a trillion dollars, it’s a few trillion dollars.

So when you add leverage into the equation, the volume comparisons get very difficult. So we were focused on the spot Bitcoin market because that’s where we’re going to be drawing prices for our Bitcoin ETF. We’d love to do research on Bitmex and we’d love to work with Bitmex to figure out where they fit into this system. But it was outside of the analysis. We tried to be very clear in the paper that we were talking just about spot Bitcoin exchanges.

Peter McCormack: Okay, just a question there. I’m a proven bad trader. What is the difference between the spot and the swap markets?

Matthew Hougan: So when you trade on a spot exchange like Coinbase or Kraken, you’re actually trading for Bitcoin. You buy Bitcoin and you have a personal right to individual Bitcoin. When you’re trading on Bitmex, what you’re buying and selling is called a bilateral swap. The way to think of a bilateral swap is like you and I agree that if Bitcoin goes up, you’ll give me money and if Bitcoin goes down, I’ll give you money. It’s just a two-way relationship that Bitmex manages and Bitmex’s job is to handle some collateral and then close out that relationship if one of us is going to go broke. So you’re not actually dealing with Bitcoin itself. You’re dealing with a one to one trading contract with someone else. It’s just a different market. That’s the technical side.

From the more practical side, Bitmex, as you know, offers between 1x and a 100x leverage. So we could bet that if Bitcoin goes up a dollar, you’ll have to pay me $100, if it goes down a dollar, I’ll have to pay you $100. It gets a little bit tricky to compare that sort of notional volume. Is that a $1 bet or is that a $100 bet versus the volume you see when I’m just buying a Bitcoin one for one. So it’s something that’s worth researching. It just wasn’t relevant directly to this question. But it’s an extremely fair point

Peter McCormack: Yeah because I guess a lot of people trade on Bitmex for the same reason as on other exchanges. I understand that there is a technical difference, but it does feel like real volume. So another thing you said, you focus on the exchanges with real volume, nine of the ten of regulated by FinCEN. I guess the one missing from that is Binance and I guess there’s probably no plans for them to change this.

Matthew Hougan: No, that doesn’t seem to be in their cultural DNA!

Peter McCormack: Is that a problem, as it is the highest traded volume of all the exchanges you listed?

Matthew Hougan: Yeah, I’m generally a pro-regulatory guy, so I would love to see all the exchanges come in within the regulated system. But that said, it doesn’t throw out the analysis because the balance and the majority of the volume is within more regulated exchanges. I actually think one of the nice things about the crypto industry is it’s so heterogeneous.

So another thing we talk about in there is market manipulation surveillance. The SEC is very concerned about these markets being manipulated and the truth is five of the exchanges have detailed market manipulation software in place and they’re all different. It actually gives me great confidence that they’re different because there’ll be looking at different things and can learn from each other. So in an ideal world, in my mind, Binance would be inside the regulatory bucket, but that’s not going to happen.

We still think they are very legitimate exchange. The volume is very real and the price discovery that’s going on there is very real. It doesn’t interfere with the overall action and health of the market in any way that we can find.

Peter McCormack: Are you under the belief that there is no real manipulation going on in the Bitcoin markets or is it just resistant to that now?

Matthew Hougan: So the argument we’ve made in the presentation is that the Bitcoin market has certain features that make it uniquely resistant to market manipulation and that the futures market is large and can be surveilled. The question of is there any manipulation is a very hard one to answer. Any manipulation… It’s like proving a negative.

But do I think on a day to day basis, the health of Bitcoin reflects the real interests of economic traders in the market? I do. On a microsecond basis in maybe isolated scenarios, I think more studies may be needed, but on a big picture basis, I think it’s an extremely healthy, extremely efficient market and the price discovery that we’re seeing is very real.

Peter McCormack: One of the benefits is when you remove fake volume, the futures market now is a significant percentage of the market, whereas when there was so much fake volume, the futures market look quite small in comparison.

Matthew Hougan: Yeah, that’s right. That that helps protect against market manipulation in a number of ways. One of the things the SEC looks for is if you were going to manipulate this market, would you have to act in the futures market and would you see your footprints there? It is good news in my opinion, that the regulated side of this market is becoming more significant and I think that’s a trend that will only accelerate from here.

Peter McCormack: Okay. So given that, is it going to change the way you calculate your daily NAV?

Matthew Hougan: We have a detailed process in place, in terms of how we calculate our daily NAV.

Peter McCormack: We should probably explain to people what the NAV is because not everyone will know.

Matthew Hougan: We have a detailed process in place for how we calculate our daily NAV, which is the net asset value. So an ETF has to strike a net asset value at the end of each day that says how much, in this case, the Bitcoin in that ETF is worth per share. Our process for doing that is somewhat complex. It’s designed to ensure that even if manipulation were to occur, it wouldn’t harm the shareholders in the fund. We think it’s a useful methodology.

It’s modelled somewhat off the Chicago Mercantile Exchanges futures methodology and it’s important that we get those details right for investors. I think on a day to day basis, most investors won’t have the specifics of the NAV in mind, but it is very important that they’re carefully designed and that the community understands what they are and that we disclose how they work.

Peter McCormack: So is this presentation with the SEC now or is it something you’ve give in person and work through as a Q&A?

Matthew Hougan: So it was part of the regulatory process. We go down there as part of what’s called the 19b-4 process and we presented this to them. As a result of that, they published the deck that we presented to them. So we had that conversation with them last week and we’ll go forward from here.

Peter McCormack: Do they give you feedback at the time? Is it something you can report on or is this confidential throughout the process?

Matthew Hougan: Yeah, everything’s confidential during the process, but I can say that in our historical experience with the SEC, every time we’re in there, they ask extremely good questions. They’re very knowledgeable about crypto. They’re asking the right questions and we just hope that our answers meet their standards. But I think they’re doing the right thing in crypto and I think they’re very smart about the space.

Peter McCormack: There were multiple applications, right? You mentioned VanEck and Winklevoss. Are these competitive applications or are you stronger together? Do you understand what I mean by this question?

Matthew Hougan: I do. All of us want to win and get to the market first. But anyone who’s interested long term in the crypto industry is generally behind the idea of there being an ETF. So I love the idea that VanEck is out there with an alternate plan, talking to the SEC about alternate ideas. We definitely studied their plan and see what we can learn from it. Similarly, I assume that they love that we’re out there with our ideas. It’s a different approach. I’m sure they study our approach and want to learn from it.

So we definitely want to win. We would love to be first to market, but really I would love for there to be a Bitcoin ETF. So people who want that institutional calibre exposure to Bitcoin has a way to get it and I don’t know that any of us will succeed, but I’m hopeful that we can and at least happy that we’re having the dialogue.

Peter McCormack: So essentially if there is a positive report, it is positive for everyone or is this so independent that it is so focused on your ETF application that it makes no difference to other applications?

Matthew Hougan: I actually don’t know the answer to that. I think it’s positive for the industry as a whole and I think the industry will be better off when we clear out the issues around fake volume because I think that’ll give investors more confidence. Part of our business, we serve primarily institutional investors, financial advisors and other professional investors.

A lot of them look at the crypto market and wonder about the data. I think when we get past the point of them wondering about the data, there’ll be more likely to come on board. So I think that’s generally positive. As to whether it helps other crypto ETF issuers. I have no idea. I have no idea what their conversations are like with the SEC, but I think it helps the industry in general.

Peter McCormack: So what are the next steps for the application? What are the important dates and what should we be keeping an eye on?

Matthew Hougan: So the 19b-4 filing process, there’s a series of deadlines in place. Those deadlines extend out to up to 240 days, at which point the SEC has to make some sort of decision. That 240-day window for both us and VanEck ends in October. But of course anything could happen in between there and we expect to get feedback from them and we look forward to the dialogue. But that’s the kind of calendar that both firms, as far as publicly is known, are on.

Peter McCormack: Okay. This has been great Matt. A couple of things to finish up on. Please tell me how people can keep an eye on what Bitwise is up to. What are the important dates, who you want to hear from and also feel free to give a shout out to the people who worked on this report?

Matthew Hougan: People who want to keep up with what Bitwise is doing on the research front, can sign up for our investor letter. You just go to our website and put in your email address, and we’d love to hear from you. You can also follow us on Twitter at @bitwiseinvest or me @matt_hougan. We’d love to hear from you.

It would be great to send a shout out. I know I’m getting to do this podcast, which is a real treat. But this was a real work by the entire Bitwise team; Hong Kim, Hunter Horsley, Joe, Micah and Phil, it was really a team effort. So a lot of work went into this presentation and a lot of people deserve credit and if people have questions or concerns about the report, feel free to email me as well [email protected]. I love to hear from people.

Peter McCormack: Great. Well, listen, thank you for doing this at short notice. I wish you all the best with this. Hopefully the next time I’m in San Francisco, I can come to the Bitwise office and we can meet in person.

Matthew Hougan: That sounds great. I look forward to it!