Paraguay has cheap and abundant power, largely thanks to the Itaipú dam. Bitcoin mining, an energy-intensive process, therefore has an interesting economic base here.

Whereas previously ordinary computers were used to perform calculations for Bitcoin software, since 2012 specialised processors specifically designed for maximum efficiency have been used: so-called Application Specific Integrated Circuits (ASICs). These ASICs have significantly increased the computing power of the Bitcoin network.

Today, the integrity of the Bitcoin ledger is secured by a global, distributed network of independent ASIC miners. These miners have no task other than verifying Bitcoin transactions and solving proof-of-work puzzles.

In Paraguay, the electricity rate for consumers is just above 3 US cents per kWh. For Bitcoin mining, or rather ‘data centres’ since Bitcoin itself is unregulated (a whole story which for the sake of convenience I won't go into for the length of the blog), a special rate of 5.4 cents per kWh applies for medium voltage mines. If you additionally include costs for hosting, maintenance, security and insurance, you arrive at about 7 cents per kWh.

Investor

An investor last year had become excited about a Dutch party that could supply and house ASICs for him at a mine in Villarica, Paraguay. He made a test investment of 25 devices, worth about $60,000, with the hope of quickly recouping this investment.

In the spring of this year, this investor approached us. The uptime (the time the devices were actively mining) turned out to be only around 15%, leading to minimal returns. He was looking for an alternative with better uptime.

Fortunately, we have several contacts in the Bitcoin mining industry. And after investigating the situation, (for convenience I'll spare the details for the length of this blog) and observing the numbers, we decided after consultation to move the equipment to a mine of friends with 99% uptime.

Everything seemed fine. And then on 20 April last came the halving. Every 210,000 blocks, about four years, the revenues for Bitcoin miners halve. This process is baked into the code and will continue until all 21,000,000 Bitcoins will be mined. That will be around 2140.

Half as much revenue at the Bitcoin price then AND now meant, for the project with 25 devices, that it cost money every month instead of making money. No bueno.

Basically, in very simplified language, the Bitcoin price would have to double after the halving to get this project properly profitable again. On top of that, the computing power of ASICs is rapidly being further developed with even more efficient power consumption.

Technical developments in the ASIC industry also played a role. Whereas in June 2023 the Whatsminer M50 at 120th/s was still cost-effective, newer models, such as the Antminer S21 pro at 234th/s, offer much more computing power at the same power consumption. As a result, the older equipment was rapidly becoming obsolete.

The investor decided last month that this investment adventure had bought him enough frustration and money. So yesterday I picked up the devices for him. Given the rapid pace at which ASICs are succeeding in development, the residual value per device is nil.

Calculations (simplified and with assumptions):

The investor bought 25 devices in June 2023 at a Bitcoin price of about $30,000, accounting for 2 Bitcoins. He settled in Bitcoin.
The first two months were spent shipping from China to Paraguay and installing the devices.
From August to March, mining was done at an uptime of 15%, which hardly yielded anything.
From April, it was mined at 99% uptime, but due to halving, money had to be added monthly.
By August 2024, the residual value of the devices was zero.
Result: a loss of 2 Bitcoins.

If this investor had kept his 2 Bitcoins and not entered this project he would now have had $127,200 at the current daily price of $63,600.

Conclusion

Bitcoin mining can only be profitable for those who have access to power at rates significantly lower than the global average of 14 cents per kWh. In Paraguay, power is indeed cheap, but rapid technological advances in ASIC equipment and Bitcoin's structural halving make it difficult to guarantee long-term profits. Competition in the mining industry is ruthless, and only those parties that can mine with power below 5 cents per kWh will be able to make a consistent profit.

Future prospects

Bitcoin mining is likely to continue to focus on locations with surplus power and a low opportunity cost. Despite the challenges, Bitcoin remains a unique digital asset with absolute scarcity, which makes it attractive to many investors.