Bitcoin and Energy @ Texas A&M

Hello Bitcoiners!

Last year I helped launch the Texas A&M Blockchain and Energy Research Consortium. This is an interdisciplinary group of business and engineering faculty members dedicated to researching the interface between energy and Bitcoin. Our first workshop took place last Wednesday where we hosted 50 people from all over Texas to introduce our research group. Here are the highlights:

Voltage bridging

We first presented a tutorial on voltage bridging problems in the Texas grid. Occasionally, power irregularities occur throughout Texas. For example, a tree may fall on a power line or a wind turbine at a wind farm may fail. The question is how to respond to these business interruptions.

This is important for Bitcoin mining because an industrial miner uses a Large Flexible Load (LFL). They're big because industrial Bitcoin mining in Texas is no walk in the park: publicly traded companies like Riot have facilities across Texas capable of generating up to 1 gigawatt. What's special about these loads is that they're flexible: they can turn on and off in a snap. Historically, this has been a major benefit to the Texas power grid. As demand increases, prices begin to rise, and Bitcoin miners will rationally stop mining when their costs rise. This is a “demand response”, a rational change in behavior compared to a market price.

The downside is that LFLs, like Bitcoin miners, can amplify these disruptions across the network. Because their load is large, a 600 megawatt power interruption can have a domino effect. If the disruption lasts longer, miners may temporarily shut down their machines, which also destabilizes the network as it can quickly knock a large load offline.

One of the Ph.D. Students in our group conducted several experiments using our S19 ASIC, which the Texas Blockchain Council donated to our lab last year. He found that a single ASIC can “survive” a power interruption on its own. However, the question remains how to tie hundreds of ASICs together. There may be a hardware innovation here that could help Bitcoin miners survive larger and longer power outages.

Miners vs. data centers

Other large consumers on the grid, such as data centers and hospitals, have solved this problem by installing backup generators for precisely these power interruptions. This makes sense for them because their end users (internet consumers and patients) demand uninterrupted service. But miners do not serve Bitcoin users directly, but rather the Bitcoin network, which operates on a global scale. Even though 17% of the global Bitcoin hash rate comes from Texas, a power outage for industrial miners in Texas will not significantly change the security of the blockchain.

To be more clear, miners today receive the majority of their revenue from the block subsidy provided by the protocol, rather than directly from Bitcoin users. The protocol allocates its block subsidy to whoever mines the next block on the global Bitcoin network. If a Texas miner loses power for a moment, they will temporarily stop participating in the global Bitcoin lottery or temporarily stop providing hash power to their mining pool. In any case, the miner can go back online and continue hashing with little market loss.

One might think that the power outage penalty would increase as the miner gets a larger share of its revenue from transaction fees, which will happen over time as the block subsidy decreases. Nevertheless, the protocol only awards the block reward to the miner (or mining pool) with a successful block. This pay-for-performance price is a feature, not a bug. This ensures that the sender of the Bitcoins only pays the transaction fees when the miner appends the block to the ledger. The mining lottery ultimately protects the user of the Bitcoin network from the details and vagaries of mining. To put it simply, a power outage is the miner's problem, not the user's problem.

This is the main difference between a miner and a data center or hospital. A power interruption could cause an operation or a complex AI calculation on an AWS server to fail. So the downward costs are much higher. However, since miners do not face such a disadvantage, they themselves have no incentive to pay for expensive backup generators.

Policy response

The concern for ERCOT, the Texas grid operator, is that these flexible loads can increase the impact of power interruptions and impose negative externalities on the rest of the grid. Therefore, ERCOT is considering requiring miners to install backup generators. This would be a mistake because it would impose an onerous regulation that would penalize the miner for acting in his own rational self-interest without clearly stating the costs of the negative externalities.

The standard economic solution would be to accurately measure this externality and implement it through a pricing mechanism so that miners can internalize any externalities they impose. This would at least use prices to encourage behavior rather than imposing regulations that impose certain costs. No matter what policy ERCOT chooses, the economy will slightly tip in favor of off-grid mining over on-grid mining.

Stay tuned as this new political agenda unfolds.