Update June 2022:
After over a year of hosting with Compass Mining and being mostly satisfied during that time, a lot of recent issues at Compass have come to light involving billing problems, slow support times, and non-transparent business practices. While I still have my machines hosted there, I do not at this time recommend using Compass until such a time as there is renewed clarity about the details of their operations and better communication to customers.
Originally published July 2021:
In recent months, China has cracked down on bitcoin mining, which means other countries, and particularly the US and Canada, are gaining a larger share of the mining network.
This article takes a look at Compass Mining, one of the investment opportunities that exists for investors in that space.
Bitcoin Mining: Coming Full Circle
The Bitcoin network relies on miners (operators of specialized processing equipment) to add transactions in blocks to the blockchain. Once a few more blocks are added on top, a long chain is formed and that is what makes a bitcoin transaction recognized as final.
To add a new block, a miner has to solve a puzzle that the previous block created. So, they use electricity to run processors to solve the problem, and create the next block of transactions. This is called “proof-of-work”. In exchange for expending these resources, the miner earns a certain amount of newly-created bitcoins when they create a block, as well as any bitcoin-denominated transaction fees that users were willing to pay to make sure their transaction gets into the next block.
When Bitcoin’s open source software was released in 2009, participants could mine coins on a typical computer. As more people started doing it, the industry became more competitive, and so it started requiring high-end gaming rigs and graphics cards to mine coins profitably, along with low electricity costs.
Eventually, when the network became big enough, people started making application-specific integrated circuits (aka “ASICs”) to make the most efficient miners possible, and from that point, pretty much only ASICs in facilities with low electricity costs could mine competitively. In addition, miners connected themselves together to form “pools”, so that they could smooth out their bitcoin rewards.
Here’s an example of an ASIC miner:
Image Source: Bitmain
So, the Bitcoin network underwent a division of powers. Individual users could still operate full nodes on a simple device with open source software, meaning they could download and audit the entire blockchain, verify transactions, and reject any noncompliant blocks. Meanwhile, mining shifted into the realm of industrial businesses, where they need low costs of electricity, and sufficient scale with many industrial-grade ASIC miners, to be regularly profitable. The era of solo bitcoin mining diminished.
For many years now, over half of the Bitcoin network has been estimated to exist in China, with dedicated farms of ASIC miners. China has been popular for a few reasons. One reason is that with so much overbuilt hydroelectric capacity from China’s partially top-down economic approach, bitcoin miners could make use of that otherwise-wasted energy and put it towards profitable use. A second reason is simply that, with relatively cheap labor and abundant electronic fabrication capabilities, China is a cost-efficient place to build, operate, and maintain electronic equipment.
China has “banned” bitcoin mining a few times, but it never had much effect. Year after year, China remained the dominant jurisdiction for bitcoin mining, regards of unclear regulations.
However, in the first half of 2021, the country seems to have finally succeeded in banning it to the point where industrial-scale bitcoin miners have been disconnected from their power sources and forced to move their equipment elsewhere if they want to use it. There will naturally be some smaller miners in the country under the radar, but the age of more than half of the network existing in China seems to be over.
As a result, the United States, Canada, Kazakhstan, and other places have been gaining market share in the Bitcoin network in recent months.
A company called Compass Mining caught my eye late last year, but it wasn’t until China banned mining and the opportunity set grew considerably for American miners a couple months ago, did I get interested enough to look into it at a deeper level. Compass provides a marketplace service for individual people to mine bitcoin, by buying an ASIC miner and hosting it in one of the facilities that they partner with. That combines industrial scale required for profitable mining with democratized access to bitcoin mining.
So, I decided to try their service out, and I also reached out to their executives (CEO Whit Gibbs and Director of Content Zack Voell) to gather some details. I also have some contacts elsewhere in the bitcoin mining industry that I’ve been discussing with, to get an idea of what’s happening on the ground throughout the North American bitcoin mining industry.
I get pitched all the time by various companies to promote their products and services, and I ignore all of them. Instead, the relatively small number of partnerships I do are with companies that I use myself and that I reach out to from my end about giving them some publicity, rather than them reaching out to me.
In that light, I’m happy to add Compass to my short list of partnerships, since I’m the one that reached out to them and I think they deserve the attention.
How Compass Mining Works
Compass Mining is a marketplace and intermediary that connects together entities with hosting facilities, entities with mining hardware, and entities who want to buy hosting space or miners or both.
Compass makes deals with facilities that have spare capacity to host mining equipment. On their facilities page, for example, they list a variety of facilities that they have partnered with. This is just a subset of them as an example:
Chart Source: Compass Mining
Many are in the United States and Canada, but they also have facilities in other countries. Each facility also lists its hosting price and whether its energy source is classified as renewable or not.
Most of the facilities charge around $0.060 to $0.065 per kWh for their hosting price, which includes the slot on the rack in the facility, the electricity (which makes up the bulk of the cost), and a technician’s time for setting it up and monitoring it in the facility.
For context, that all-in cost is less than half of the electricity price I pay per kWh for my home, which is why I can’t just buy the miners and hook them up in my home. These hosting facilities have commercial rates and set up in regions with cheap and abundant electricity.
Compass also has an ongoing selection of ASICs for sale. Sometimes they have more available than other times, and the prices change over time based on supply and demand. Zack Voell, the Director of Content at Compass, has explained that ASIC prices tend to appreciate quickly when the bitcoin price appreciates, but tend to depreciate more slowly when bitcoin’s price falls, since they are a bit “stickier” once they are set up and mining bitcoin. I’ve monitored this trend as well.
Earlier this year, there was an acute ASIC shortage. The price of bitcoin was going up which increased demand for the ASICs, but due to global semiconductor shortages, only so many new ASICs could be made at a time. As a result, each available ASIC became quite valuable. As bitcoin has corrected in price, and as China pushed bitcoin miners out of its country, the ASICs themselves have come down in price and have become easier to get.
However, now the bottleneck is with hosting facilities. The industry can’t just snap its fingers and instantly create proper facilities with good electricity contracts for a few billion dollars worth of ASIC equipment, to absorb the hundreds of thousands of ASICs that are coming out of China. A number of companies in North America and elsewhere are building new facilities this year, and Compass still has available space at certain contracted facilities. Compass is growing their capacity by a ton this year and next, according to their CEO.
Compass has a bunch of different types of customers/partners, including those selling miners, those who are facility hosts and want to access a broader market, and people who want to mine. This article focuses on that latter group, those who want to mine.
The cool thing about Compass, is you can use it as a turnkey solution to start mining. You can buy an ASIC on Compass, and host it at one of their contracted facilities. That facility will hook it up for you, and once you select your mining pool of choice and provide a public bitcoin address to send your earnings too, you’ll start earning fractional bitcoin. It’s not a cloud mining service or anything like that; you own the physical ASIC that you bought, and rent electricity/space in a facility where you host it, and earn the rewards. You can resell your ASIC if you want to.
It took me a few minutes to purchase the ASICs and facility plan. Once the hardware was ready some weeks later (there is a bit of a turnaround), it took another few minutes to contact Compass about setting up my connection to a mining pool of my choice, and to start receiving the earnings to the public bitcoin address I designated. And now, it’s simply working. I can log into my mining pool account and check the uptime of my miners at any time, and check to ensure that the fractional bitcoins are arriving in my address.
In addition, I took time to do research and run through the risk/reward analysis, which changes over time and is open to multiple variables. Overall, the purchase and setup process itself was pretty painless, and how much time I put into the research was up to me. Naturally, I put in a lot of research time and modeled various scenarios.
Profit Estimates and Risk Analysis
There are publicly-traded bitcoin mining stocks, and they have their advantages in a portfolio. These companies have easy access to capital, and rather low costs of power. But with Compass, you own the miner directly and at book value, which can make for a pretty high return on investment.
Here’s an example for some rough math.
Let’s say you buy an Antminer S19j Pro for $7,500, and have a $150/month hosting fee for it which covers electricity and other normal costs at $0.062 per kWh. These are roughly the rates for one of my example miners, but like I said, the numbers change over time.
There are websites that calculate ASIC profitability. For this one, I rounded up to $0.07 kWh expenses:
Chart Source: ASIC Miner Value
From there, we can run through some models. There are certain types of environments where bitcoin is more or less profitable than others. The best conditions are when the bitcoin price is pretty high, but the global hash rate is held down, either by ASIC supply bottlenecks or hosting facility bottlenecks. Right now, that’s the case, so mining is rather profitable at this time.
USD-Denominated Example
The ASIC that cost $7,500 is set to earn $9,669 per year using current bitcoin prices and network hash rate after paying for electricity, assuming you convert any bitcoin you earn into dollars. I’ll round the earnings down to $9,000. With that mining profitability, you earn the cost of your machine back within the year, and everything after that is profit.
A top-of-the-line ASIC miner can last 3-4 years or more. How long it lasts precisely depends on how ideal the conditions are in terms of heat and maintenance, and how much new hash rate and better ASICs come into the market to make this one obsolete enough to no longer be profitable.
How long you mine is up to you. You could mine for a year and then re-sell the machine, for example. Or you could hold it for a few years and mine with it until it’s out-of-date, unprofitable, and depreciated to near zero value.
Let’s say you mine for three years, starting with $9,000 for the first year, but the second year is only 75% as profitable ($6,750), and the third year is only 50% as profitable ($4,500). In this case, you earn $20,250 in operating profits, which already factors out hosting/electricity costs. The initial $7,500 device would be scrapped or sold for a very low value, so let’s call that zero.
Then of course, you pay income or corporate taxes on those earnings, which vary depending on your location.
Overall, that’s a really good rate of return over a 3-year period if it works out that way (and it might not work out that way). You turned $7,500 into $20,250 in that hypothetical scenario, which is a 170% return on your initial investment. In addition, since a lot of those profits are front-loaded, you could take the first year of profits and buy a second machine, and start earning with that.
BTC-Denominated Example
We can also run the numbers in bitcoin terms. The device cost $7,500 initially, and $150/month. It earns maybe 0.0008 gross BTC/day with the current global hash rate as it currently is.
Let’s assume it earns 0.0008 BTC/day this year, 0.0006 BTC/day next year, and 0.0004 BTC/day in the third year, as more hash rate comes onto the network. And then you cease operating it and get rid of it for a negligible value that we’ll call zero.
You can buy an ASIC with 0.25 BTC at current prices ($30k USD). If all goes well, you would have earned 0.65 BTC during those three years using those assumptions. You also pay some BTC throughout the year for electricity costs, and that will vary throughout the year but at current prices would average 0.06 BTC per year.
So, you obtained more bitcoins than you would have if you had used the initial investment to buy bitcoins instead, although it comes with increased risk of something preventing you from earning those bitcoins over the next 3 years. That’s why it’s an investment: it contains risk and reward. The good news is that you should likely earn back a large portion of your initial investment in the first year, so the investment quickly de-risks itself.
If you can squeeze a fourth marginally profitable year out of it, it would enhance the return. If the miner breaks or becomes unprofitable after a year or two, you’d likely recoup the initial cost but not earn much compared to having just bought some bitcoins.
Hybrid-Denominated Example
Let’s say you use dollars to pay all the expenses, and hold whatever bitcoins you earn. So you pay $7,500 up front and $150/month for three years. That’s $12,900 in investment, spread out over three years but we’ll assume you set it all aside up front. You then earn 0.0008 BTC/day for 1 year, 0.0006 BTC/day the next year, and 0.0004 BTC/day in the third year.
You’ll have 0.65 BTC by the end, if it works out that way. At current $30,000 BTC prices, you’ll have turned $12,900 into $19,500, during three years. You’ll owe some income taxes as well, but you get to factor out the depreciation and other expenses.
However, if bitcoin soars to $100,000 by the end of that three-year period as an example, then your retained 0.65 BTC would be worth $65,000. You turned $12,900 into $65,000 during three years. That’s the power of mining if bitcoin has a bullish outcome.
The breakeven price would be somewhere around $20,000. If bitcoin ends the period below that value, then in terms of how many dollars your retained bitcoin are worth, you would not have recouped your investment.
Risks
From those rough models, we can then factor in various risks.
The first key risk is obviously the bitcoin price. If the bitcoin price crashes, then the investment in a bitcoin miner won’t do very well, and could fail to recoup the initial investment. Naturally, only someone who is reasonably bullish for the long run on bitcoin should buy and host bitcoin mining equipment.
The second set of risks relates to network difficulty. The more ASICs that connect with the bitcoin network, the fewer bitcoin you earn per day with your miner. Right now, there is the equivalent of about 1one million high-end ASIC miners hooked up to the network, so I’m earning about one millionth of the bitcoin generated each day from block subsidies and transaction fees per high-end ASIC that I own. If most of the disconnected ASICs get hooked back up next year, along with newly-created ones, I could be earning half that much or less per ASIC. A model should assume a degrading in BTC earning power during the lifetime of the machine, but it can be hard to determine how steep that decline slope should be.
The third set of risks relates to physical disruption. For example, the hosting facility might cease to operate for one reason or another, or lose its supply of inexpensive electricity, and you’d have to move your miner to another facility with downtime. This could be due to market reasons, physical damage, or regulatory changes. Alternatively, the ASIC could break down and require warranty repairs, or break down outside of the warranty period and cease being able to mine bitcoin earlier than the base assumption. Both of these are moderately low probability outcomes, but certainly possible. These risks can be mitigated if you own multiple machines at multiple different facilities through Compass. Also, if you have a problem with your machine that the facility caused, you can reach out to Compass and work with them to make sure you are treated well. The CEO explained that as the intermediary and brand involved in this process, Compass has a “happy customer” policy and wants to ensure customers are treated fairly.
The fourth risk is more of a hassle than a risk. Unlike simply buying bitcoin, in order to mine with Compass you need several thousand dollars to buy at least one ASIC miner, and you need to know a bit of “Bitcoin 101”. You don’t need to know too much technical stuff, like you don’t need “Bitcoin 401”, but you need some basics. For example, you need to crunch some numbers like I did in this article to see if it makes sense to you from a risk/reward perspective to buy bitcoin, mine bitcoin, or do a bit of both (or if you’re bearish, to avoid the space). You have to know what a mining pool is, and you need to have a bitcoin address that the pool can send the fractional coins to, such as a hardware wallet, or an address in an account somewhere.
Final Thoughts
Ever since early 2020 when bitcoin was under $7,000 per coin, I’ve been bullish and long on bitcoin, and continue to be so with a long-term perspective. That view of course has risks associated with it, and I treat it as part of a diversified portfolio.
Here in 2021, I’m specifically interested in bitcoin mining, due to this one-time event of China banning bitcoin mining, and the Bitcoin network’s hash rate no longer being China-dominated for the first time in many years. This creates physical bottlenecks in terms of shipping and hosting those offline miners, and so bitcoin mining is currently more profitable than average and is set to be like this for a decent amount of time.
In other words, there is a good probability of getting more bitcoin out of mining, than the bitcoin-denominated investment costs of doing so. And it’s pretty time sensitive to this year; that’s where the opportunity mainly is, since eventually more miners will be made and many of the existing lightly-used Chinese ASICs will find a new home in another hosting jurisdiction.
I like the fact that Compass makes bitcoin mining accessible to individual investors. Thanks to the the marketplace effect, you get economies of scale that are similar to an industrial-grade mining business, but with a manageable amount of capital and effort on your end. You won’t be the most competitive miner out there, but you’ll be reasonably competitive, and you’ll hold the equipment at book value rather than paying a valuation markup for a public mining stock. Plus, individual mining helps to further decentralize the global Bitcoin network hash rate.
It’s not without risks, but for those who are bullish on bitcoin, the risk/reward opportunity set is attractive in the current environment. Buying an ASIC and setting aside cash to pay the electricity is like committing to dollar-cost average for a while, so it’s a disciplined way for those who are bullish on bitcoin to accumulate more coins at a discounted price.
If you’re a high-roller, Compass also has VIP plan that lets you buy 12 Antminer S19J ASIC miners for the equivalent of $4,500 each, which are installed one per month over the course of 12 months.
