Navigant Research Blog

The Dynamics of Bitcoin Mining and Energy Consumption, Part III: The Future Is Bleak for Proof of Work

— July 3, 2018

This is Part III of a series. Part I is here, and Part II is here.

In Part II, I wrote about how Bitcoin miners are incentivized to invest in power-hungry facilities as long as Bitcoin’s value is high. I also said that a crypto energy apocalypse is unlikely. It’s time to zoom out from the mechanics and look at the larger trends that shape miner behavior and proof of work (POW)-based energy consumption: local environment, regulations, and technology.

Miners Establish Where Operating Costs Are Low

As I discussed in Part II, being a successful miner is all about minimizing costs, and local environments can have a big impact on a miner’s bottom line. For large-scale facilities, miners prefer cool climates where abundant, cheap electricity exists. In the US, the Pacific Northwest is attractive for its abundant hydropower, which keeps electricity prices in the lower quartile of the national range.

Mining farms are made up of thousands of shoebox-sized units, which allow them to be highly mobile. If changing regulations makes one area unfavorable, miners will move to a new area that meets their needs.

These forces are already at work—crackdowns in China have led miners to search for greener pastures in North America, Iceland, Sweden, and similar locations.

Regulations Force Miners to Move

Since mining is tied to digital currencies, regulations that affect how Bitcoin and other cryptocurrencies are taxed and controlled influence the profitability of mining operations. For example, the US Securities and Exchange Commission could regulate currencies as securities. Governments also target the exchanges where miners convert digital income into real money.

Likewise, countries and municipalities are regulating mining directly, either through outright bans or by forcing miners to pay higher electricity rates. That’s because mining farms with demand in the tens of terawatt-hour range tend to overwhelm local electricity production and drive up rates for residents.

Technology Changes Quickly

The most significant trend is technology evolution. Bitcoin is old technology by blockchain standards—it is notoriously slow and inefficient for most transactions. Newer technologies, like IOTA Foundation’s Tangle, resemble blockchains and can support digital currencies but are faster, more efficient, and do not include mining.

If Bitcoin and Ethereum survive the competition, they may still replace POW. Ethereum is preparing to switch from POW to proof of stake (POS), which eliminates mining. Once POS is deployed, Ethereum’s roughly 18 TWh of annual electricity demand will vanish overnight. Dozens of alternatives to POW exist, and a precious few include mining.

Miners who have invested millions into massive facilities won’t go down without a fight. When Ethereum releases POS, its community will likely fork into two networks: one with POW and one with POS (it happens often).

But because digital currencies are electronic, users can switch networks quickly and easily. A user exodus causes network value (and miner revenue) to crash: one unit of Bitcoin Cash, which forked from Bitcoin in 2017, is now worth 10% of 1 Bitcoin; Ethereum Classic, which began as a fork of the Ethereum network, has since lost 97% of its value. As we saw in Part II, large mining facilities aren’t sustainable in a low value network.

The Takeaway

Outside of cryptocurrencies, POW is already obsolete. Of the more than 160 companies developing energy blockchain applications, not one is building with Bitcoin. And those using Ethereum are poised to switch to a more efficient network.

POW is dying a slow death, strangled by regulations and left behind by more advanced and useful technologies. Its energy consumption will die with it.


The Dynamics of Bitcoin Mining and Energy Consumption, Part II: Mining Incentives and Economics

— June 28, 2018

This is Part II of a series—Part I is here.

Blockchain systems aren’t really single technologies. They are complex architectures with many interacting parts, and hundreds of different architectures exist. For this discussion, the relevant piece of the architecture is the consensus algorithm. That is the group of rules that determine which nodes in a blockchain network are responsible for validating transactions, packaging them into blocks, and storing them in a distributed data structure. One consensus algorithm in particular, Proof of Work (PoW), is the cause of all the hand-waving about blockchain energy consumption.

What Is Proof of Work?

PoW functions by pitting special nodes against one another in a race to solve a mathematical puzzle that requires a huge amount of computational power (energy) to solve but is trivial for others in the network to verify. Miners are willing to invest resources into the race because the winner gets a prize (12.5 Bitcoin, or ~$90,000 USD). Like race cars, more powerful systems are more likely to win the race and the prize. Miners are incentivized to keep investing in compute equipment as long as they win often enough to net a profit—resulting in massive IT facilities like the one in Mongolia.

In the Bitcoin network, a new block is up for grabs every 10 minutes. To keep this interval constant, the difficulty of the PoW puzzle dynamically adjusts based on the total mining power of the network. In simple terms, as cars get faster, the racetrack gets longer—and drivers have to burn more gas to finish.

Proof of Work Difficulty and Mining Power Trends: 2016-2018

(Sources: Navigant Research, Blockchain.Info)

Mining Activity is Ruled by Economics

We can think about mining economics in terms of a simple profit equation, with the implicit assumption that miners will operate as long as profit is greater than zero:

Profit = Revenue – Costs

Costs (hardware, facilities, electricity, personnel) are the most stable component of the miner profit equation. This is because revenue is comparatively volatile: miners are rewarded in Bitcoin, and successful miners receive 12.5 Bitcoin, but the corresponding value in “real money” changes on a minute to minute basis. Miners must convert their rewards to real money eventually to pay their bills, and almost no one accepts payments in Bitcoin.

Economic Complications

A key point is often overlooked: in the Bitcoin network, miner rewards halve every 4 years. For a given mining facility to remain profitable—all else held equal—the value of Bitcoin must double over the same period. If it doesn’t, the revenue opportunity for miners shrinks and they have to reduce costs to remain profitable.

As the value of Bitcoin increases, more miners are incentivized to join the network or invest in new, power-hungry equipment. If the value drops, the incentive flips, and miners will leave the network or shrink their operations. As miner compute power leaves the network, PoW adjusts its difficulty downward, reducing energy consumption.

What Does This Mean for Future Energy Consumption?

Unfortunately, it doesn’t mean that energy consumption isn’t a problem. It just means that energy consumption is unlikely to continue growing at the prodigious rates seen over the last few years. There is only one doomsday scenario: the value of Bitcoin continues to double every 4 years, pulling more miners into increasingly difficult PoW competitions.

As we’ll see in Part 3, that scenario is highly unlikely (unless you’re this guy). Now that we have a basic framework in place, we can discuss the various factors that will affect PoW mining and energy consumption in the future, and what this means for utilities that have to make plans today.


What Exactly Is AI? Does It Matter?

— June 19, 2018

The more I read about artificial intelligence (AI), the less clear it becomes what people mean when they talk about AI. And while semantic arguments can be a waste of everyone’s time, a loose definition of Al presents an opportunity for vendors to package software into something that it isn’t. Blockchain and AI vie for position as the most overhyped term in technology today. But blockchain at least benefits from a common understanding of the underlying technology, even if its potential uses are severely overstated. Any buyer of AI-powered software should beware of strangers bearing gifts: always make sure that AI products are capable of what you want them to do, not what the vendor claims.

AI Beats Expectations to Be Simultaneously Everything and Nothing

Everyone has a definition of AI. Your perspective, knowledge of analytics, and desperation to sell some technology have a great bearing on what constitutes AI. I have seen definitions that range from “any code that includes an IF statement” to “nothing currently in existence.”

Technology marketers can never be accused of reticence when it comes to latching on to the latest industry buzzword. AI’s loose definition means it can be applied to virtually any piece of technology. An IF statement in a computer code means that a computer will make a decision based on some form of data input. At a very basic level, this is an intelligent decision.

Many disagree and claim the boundary for what constitutes AI lies in more sophisticated analytics processes. Essentially, people look for examples of a computer mimicking human thought. The big question is: How close to human thought does a computer have to get before it is considered capable of “thought?” Indeed, each time a new development happens in the field of advanced analytics, such as AlphaGo beating a human at the notoriously complex game of Go, someone will always say that it’s not true AI. This trend is summed up in Larry Tesler’s theorem that AI is “whatever hasn’t been done yet.”

The Paradoxical Definition of AI

I have no idea what AI is supposed to be, and I believe that this uncertainty stems from the blurred boundaries of its definition. The sorites paradox explains this well: Starting with one grain of wheat, how many more grains must be added before one has a heap? Similarly, there is no clear answer to the questions around what constitutes intelligence, when AI takes over from basic tech-based processing, or how close to human thought tech processing must get before it can be deemed even artificially intelligent. Some may propose specific demarcation; others immediately disagree.

What Matters Most Is That None of This Matters

We often start reports on technology with a definition to help the reader better understand the boundaries of what we write about. I have spent far too long trying to come up with a definition of AI with which I am comfortable. I can’t come up with a sensible boundary for what does and doesn’t constitute AI, and even if I did, more people will disagree than agree. But who cares? AI really means the latest and shiniest analytics product to hit the market.

What buyers should bear in mind is whether this technology does the job for which it was intended at a competitive price. Just make sure that anything painted in AI colors is going to make a decent ROI for your business.


The Dynamics of Bitcoin Mining and Energy Consumption, Part I: A Problem in the Here and Now

— June 5, 2018

I’ve written quite a bit about use cases for blockchain, the technology that supports applications like Bitcoin, outside the world of digital currencies. I have spent much less time diving into the impacts of digital currencies themselves. This is because the most exciting and potentially transformative applications for blockchain in the energy world have little to do with them, or with the proof of work consensus algorithm.

The downside of taking a future-focused view, however, is that it tends to gloss over the real challenges faced by utilities and other energy stakeholders today. The rapid rise in the valuation of proof of work-based digital currencies like Bitcoin and Ethereum has created a new, power-hungry digital mining industry that utilities have to confront.

Digital Currency Farms Require Huge Amounts of Power

Take the example of this bitcoin farm in inner Mongolia, which demands 40 MW of electricity per hour and, at the time Digiconomist’s article was written, earns the equivalent of $250,000 per day in bitcoin. At $0.09 per kWh, that’s about $86,000 per day in electricity costs. Even after taking hardware, maintenance, oversight, and facilities costs into account, the farm is turning a substantial profit, and a percentage goes back to the power provider. This farm is one of many around the world that collectively accounts for 65 TWh of electricity consumption annually.

At first glance, this might not seem like a bad deal for power providers. If the deal is structured properly, they get a cut of a profitable and fast-growing business, provided they can support the miner’s power needs. That’s likely why the Chinese government offered them a rate discount as an incentive for setting up shop.

But mining farms aren’t a typical client. The value of digital currencies like bitcoin in real money is highly volatile, which makes continued profits uncertain in the future—and that has serious implications for utilities that are considering developing new capacity to account for a surge of new facilities like the one in Mongolia. If the bubble bursts and mining becomes unsustainable, these farms could disappear as quickly as they sprang up.

A Rock and a Hard Place

Utilities with decades-long planning horizons revolving around rate cases should want little to do with an industry that thrives on short-term volatility, and whose loads could be here today and gone tomorrow. But in the short term, the energy consumption associated with proof of work gives them little choice. Worse yet, not all mining companies bother to speak with utilities directly. Some simply truck in their equipment, plug in, and get to work. They’ve been blamed for rate spikes and other problems in distribution networks around the country.

Examining the Cost of Digital Currencies

Over the next several weeks, look for my blog series on energy consumption associated with bitcoin and other digital currencies and its impacts on utilities and energy systems. The series will build off of work by others in this area—notably the folks over at Digiconomist and the recent paper (no paywall) published by Alex de Vries in the peer-reviewed journal Joule earlier this May.

My intent is to add a systems-level perspective to the conversation that will help build an understanding of the main drivers of mining activity, how it might change in the future, and what utilities can do to make the best of a bad situation that many analysts expect to get worse. The second part of this series will look at the situation from the miner’s perspective, exploring why farms like the one in Mongolia arise, what determines where those facilities pop up, and what causes them to move or shut down.


Blog Articles

Most Recent

By Date


Building Innovations, Clean Transportation, Digital Utility Strategies, Electric Vehicles, Energy Technologies, Finance & Investing, Policy & Regulation, Renewable Energy, Transportation Efficiencies, Utility Transformations

By Author