A Healthy US Bitcoin Mining Industry Could Generate Significant Tax Revenue

Crypto supporters were taken aback this past July when the infrastructure bill brought to the U.S. Congress claimed it could raise $28 billion from crypto investors by applying new information-reporting requirements to exchanges and other parties. This projection ended up getting beat down on the internet as the dollar amount seemed to be plucked out of thin air. In reality, figuring out how much taxes crypto investors owe based on their capital gains is incredibly difficult to estimate.

Theoretically, the Internal Revenue Service (IRS) could look through every transaction on every blockchain to see profits and losses in each wallet or account. From there, the IRS could figure out the amount of on-chain gains it could tax. However, that raises the issue of whether those assets were sent from one wallet to another with the same owner, something that may not make it a taxable event. On top of that, there’s the difficulty of getting good information from exchanges to figure out the amount of off-chain gains the IRS could tax. In practice, this collection and estimation process is a mess.

This piece is part of CoinDesk’s Tax Week.

If the U.S. government wants to raise money through taxation on crypto, it could consider encouraging bitcoin miners to set up shop. Doing so could bring in tax revenue inflows from the companies that set up mining operations.

For Tax Week, we wanted to estimate the amount of revenue the U.S. government could stand to gain from bitcoin mining companies. While the result of this exercise is subject to the assumptions underpinning the model, they are worth the look. The fact that this exercise is even possible is a testament to bitcoin mining’s transparency and simplicity.

We built a relatively simple estimate of bitcoin mining profitability using an open-sourced model developed by Galaxy Digital to approximate the cost of mining a bitcoin (the report the model came from is available here), applying simplifying assumptions to represent the entire bitcoin market.

A few caveats before we dive into a brief overview of the methodology are worth mentioning: Right now, there are several publicly traded bitcoin mining companies (which is sometimes referred to as CHARM for Core Scientific, Hut 8, Argo Blockchain, Riot Blockchain and Marathon Digital). Public companies are required to share financial information and those reports show that bitcoin mining companies are, by and large, not paying very many taxes.

In fact, some of the companies book income statement losses and are paying no taxes at all. Start-ups – which bitcoin miners are – are generally unprofitable as they look to spend money building up operations. Our model strips out the business decisions that young companies must make when they are growing, meaning that it only works in a world with a more mature bitcoin mining industry.

We also wanted to normalize for accounting methods allowing companies to minimize tax burdens, mostly through non-cash charges like share-based compensation and some types of depreciation. Doing so makes a company look less profitable on paper than it is in reality.

The last simplifying assumption we make is a big one, in that bitcoin mining profitability will not trend to zero. There is a solid theoretical argument that bitcoin mining economic profit margins will approach zero as new entrants join the relatively low barrier-to-entry market. (The CoinDesk report “Does Bitcoin Have an Energy Problem?” suggests that “bitcoin mining [profit] margins are relatively capped.”) In reality, businesses need to make money over the long term in order to stay open, so we assume that bitcoin miners won’t lose all profitability for at least the near- to medium-term.

Our work relied on the model done by Galaxy Digital for a simple reason. We know roughly how much revenue miners will collect annually in bitcoin terms. The Bitcoin protocol is designed in a way so that a block is mined roughly every 10 minutes, so we can say with confidence that the amount of revenue miners will make annually is 328,500 bitcoin plus transaction fees, which nominally make up about 3% of the current block reward.

As such, the main focus for determining profitability should be on estimating costs. The three main expenses we looked to estimate were cost of revenue (mainly electricity costs for mining); selling, general and administrative expenses (general costs like marketing and rent); and depreciation (a non-cash but real expense that represents the wear and tear on machines used for mining).

In plain English, we looked to estimate the cost of the electricity it takes to power mining rigs and the cost of “keeping the lights on.”

The main cost driver for bitcoin mining is a function of electricity use and price. Galaxy’s model calculates the cost of bitcoin production based on the specifications and performance of 18 different models of ASIC mining machines. These machines draw different amounts of power at varying levels of efficiency. Each type of mining machine operates at a different level of profitability based on the cost of electricity per kilowatt hour. Our base case assumed $0.06 per kwh of electricity cost.

Next, in this machinery-heavy business, the mining companies that buy ASICs have a meaningful amount of depreciation to deal with. Our model assumes 22.5% of revenue as the base case based on the assumption that mining companies will depreciate their ASICs over five years.

These companies then have other costs associated with SG&A which, through public company comparisons and informed by Galaxy’s work, is estimated to be around 12.5%.

Lastly, we included a catch-all for “other expenses” that made up 3.5% of revenues, as a way to curtail the potential overstatement of profitability. We could have easily made the same adjustment in the other direction and recognize this is largely a design choice by our team.

Below we present our results in two-way charts using various scenarios adjusting for bitcoin price, Bitcoin’s total hashrate, the cost of electricity and U.S. share of global hashrate. The numbers in the chart represent the annual federal tax revenue to the government from mining companies, assuming a 21% federal corporate tax rate.

In the event the input is not sensitized against in the chart, the base case is:

In the base case scenario, bitcoin miner pre-tax profitability was estimated at $1.4 billion and a tax bill of $299 million. That scenario shows up in the middle of each table below. All other numbers in the tables are representative of estimated taxes if those inputs were changed. For example, if bitcoin price were $60,000 and hashrate were 250 EH/s, taxes to the U.S. government would be $335 million.

Of course, this exercise was for informational purposes and the results provided are for illustrative purposes only.

We recognize the shortcomings of our model and this exercise. But at the very least, bitcoin mining represents a potentially profitable industry that, when domiciled in the U.S., could provide the government with increased tax revenue. While the specifics of “how much” revenue this could bring the government vary greatly (showcased by the wide range of dollar amounts shown in the two-way tables, in some scenarios even hitting $0), profitable businesses represent tax revenue opportunities for the U.S. government.

Crypto won’t save you from taxes, but it may eventually make them easier to pay, says futurist Dan Jeffries.

Tax guidance lags innovation. So does tax software. Meanwhile, misconceptions abound. If not careful, investors can end up owing more tax than expected and having to unload crypto to pay the bill

Investors in MicroStrategy, Tesla, Block and Coinbase need to consider how wild price swings will affect results, not only directly but indirectly due to complex tax accounting rules.