NYT Missing the Point on Flexible Load

The latest talk in the bitcoin world was the NY Times report, The Real-World Costs of the Digital Race for Bitcoin. In our view, this article tells an incomplete story, focusing on BTC mining without also addressing additional relevant information related to utilities and commissions. The authors seemed to select and use only the information supporting their conclusion that mining operations “can create costs — including higher electricity bills and enormous carbon pollution — for everyone around them, most of whom have nothing to do with Bitcoin.”  However, the authors did not realize or acknowledge the fact that bitcoin mining actually can have the opposite impact.

Regarding the ERCOT storm case, they claimed that miners were “exploiting the system,” even though the incentives to reduce load during extreme events are known and available to many users and these incentives were developed because it was believed that market forces can balance the system vs. overbuilding the system with resources. Bitdeer and other miners who curtailed operations during the event, reducing system burdens and constraints. Rather than addressing this perspective, the authors concluded that Bitdeer simply took money from the system instead of operating to the system’s benefit. “Bitdeer would make more than $18 million for not operating, from fees ultimately paid by Texans who had endured the storm.”

Operating a 170MW facility in Rockdale, Bitdeer would pay over $40 million a year in power cost. In a freak weather anomaly where the system valued each MWh at $9,000/MWh (potentially mispriced), the 170 MW load for each hour would be $1.5 million per hour or $36 million per day – yet the company received only $18 million for a five-day event. Their operating agreement was a discount to the price at which the market would actually value their power if they had financially hedged and sold the position.

Bitcoin Miners Get Paid For Benefiting The System, Not Exploiting It

The concept that bitcoin mining facilities are leeches on the power grid is confounding, as this would mean a complete failure of the utilities to understand the power system and the NY Times authors understand the power system better than the local utilities. Many utilities were marketing and trying to attract miners to their systems. The utilities and ISO’s are not powerless: they have many levers, from power contract structure to system studies, to restrict any entity on the system if it truly was detrimental to their rate base or grid integrity. The utilities have no incentive to allow loads that would increase cost to the system, particularly without any benefit. The article states, "They can avoid fees charged during peak demand, resell their electricity at a premium when prices spike and even be paid for offering to turn off. Other major energy users, like factories and hospitals, cannot reduce their power use as routinely or dramatically without severe consequences.”

This sounds like a significant benefit to have on the system, not a detriment causing more cost. If only major users could do this more, renewables would be easier and overall rates could come down, given that peak load represents a large cost to the system, another example of the article telling only a partial story.

The choice of interviewers was also skewed toward academics – professors from University of Houston, Princeton, and Berkeley. They did interview an energy consultant at Wood Mackenzie who clearly is already biased and “who has previously criticized Bitcoin’s dependence on electricity as ‘inessential.’” They did interview a blockchain attorney and an exploration company – both of whom have limited knowledge and understanding of the power system and particularly resource planning. (Note: Author David Bellman was a former American Electric Power Managing Director of Strategic Planning and played an integral part in AEP’s resource plans.)

One of the arguments of the article is that mining is causing a lot of emissions. Even though Riot and other miners have facetiously noted they do not emit, the focus on emissions is more of a function of utility decisions, not a load center. A load center can only influence the utility decision on resource building by its load profile. As noted, a mining load profile is significantly better for a utility to integrate renewables than typical large energy users. However, not one utility executive was questioned. If society really did not want fossil fuels, this can be implemented by local utilities via public utility commissions, state regulators, and potentially federal policies. To blame the load for the outcome of utility decisions is perplexing.

The argument that mining’s dependence on electricity is “inessential” is its own argument. However, the decision tree for “inessential” in the US is driven by markets, not government bureaucrats or consultants. Netflix now represents 15% of internet traffic – would that be considered inessential? At one point (since retracted), the leading economist Paul Krugman said the following in 1997: “The Internet’s impact on the economy has been no greater than the fax machine’s… ten years from now, the phrase ‘information economy’ will sound silly.” Imagine if we had given up on the internet because it used power and was deemed “inessential.”

There have been several New York Times articles over the past few years which have truly reduced their credibility as a source of independent journalism. Power is a tough topic, and the authors should have discussed it with someone who has done resource planning and worked in the utility world if they wanted to focus on emissions as a result of power demand. It is disappointing that the academics failed to point them to the utilities resource planning process. In addition, the authors were unaware that bitcoin miners actually can help the system – or perhaps they did, and intentionally omitted that side of the story.

Miner WoW View

• Mining economics were flat WoW.
• The S19JPro breakeven price is between $70-$80/MWh.

Henry Hub WoW

• Henry Hub was down slightly WoW.

PJM WoW

• For the PJM region, we use PJM-W hub as the benchmark. PJM-W is the most traded power hub in the US.
• PJM saw only minor changes WoW.

ERCOT WoW

• For the ERCOT region, we use ERCOT-North hub as the benchmark. ERCOT-North is the most traded power hub for ERCOT.
• ERCOT saw only minor change WoW.

CAISO WoW

• For the CAISO region, we use SP-15 hub as the benchmark. SP-15 is located in Southern California.
• CAISO saw a big drop in summer prices.

NYISO WoW: NY-G

• This slide uses the NY-G hub as the benchmark for the NYISO region. NY-G is the most traded power hub in NYISO.
• NYISO saw only minor changes last week.

NYISO WoW: NY-A

• This slide adds NY-A for the NYISO region.
• NY-A is up and moving on its own again.