The new grid infrastructure needed for the much-publicized data center buildout is being unevenly subsidized by other ratepayers through minimally publicized utility agreements, a report from the Harvard Electricity Law Initiative charges.
This has the effect of foisting upon the public billions of dollars worth of upgrades to benefit a handful of very wealthy corporations, legal fellow Eliza Martin and Electricity Law Initiative Director Ari Peskoe wrote March 5 in their announcement of the paper.
The authors of “Extracting Profits from the Public: How Utility Ratepayers are Paying for Big Tech’s Power” said they reviewed nearly 50 regulatory proceedings to reach their conclusions and devise their recommendations.
Their focus is utilities funding discounts to Big Tech and its electricity-hungry data centers by socializing the cost across other ratepayers, then redacting the details of those agreements in public utility commission filings in the name of trade secrecy.
“Utilities tell PUCs what they want to hear,” the report says: “that the deals for Big Tech isolate data center energy costs from other ratepayers’ bills and won’t increase consumers’ power prices. But verifying this claim is all but impossible. … The subjectivity and complexity of ratemaking conceal utility attempts to funnel revenue to their competitive lines of business by overcharging captive ratepayers.”
Because big data centers have a big economic impact, the authors add, there is political pressure on PUCs to not endanger their construction in a particular state or district by rejecting proposed data center contracts.
The report notes the oft-cited predictions that data centers will drive soaring near-term U.S. power demand, and it notes an oft-cited observation about regulated utilities: When they build more infrastructure, they are in line for more regulated profits.
So they have incentives to be optimistic about future growth, the authors say, and as monopolies, they do not face competitive pressures that would push them toward less expensive or more efficient solutions.
The report lays out reasons PUCs may deviate, intentionally or not, from the “cost causation” principle that guides ratemaking:
-
- Attributing the utilities’ costs to various ratepayer classes depends on contested assumptions and disputed methodologies; different approaches to cost allocation will yield different results.
-
- PUC commissioners, whether elected or appointed, may feel political pressure to favor a certain ratepayer class.
-
- The utility may exploit its informational advantages and intentionally provide false information.
With the data center market, utilities are competing for a profitable chance to serve an energy-intensive customer that bases siting decisions in part on power costs — so they have incentive to offer low prices that shift cost to other ratepayers, the authors write.
They focus on three mechanisms through which that shift can be carried out:
-
- Special contracts containing secret terms are approved through opaque regulatory processes, often in short and conclusory orders with only cursory analysis.
-
- Gaps between federal and state regulations allow costs for data center infrastructure to be left to ratepayers; saddle ratepayers with stranded costs that arise; and allow data centers to reduce their share of regional charges by reducing their energy use a few hours per year.
-
- Data centers contract directly with co-located power generation, disrupting power market pricing and delivery rates.
Collectively, these factors create problems, the authors argue: “Without systematic changes to prevailing utility ratemaking practices, the public faces significant risks that utilities will take advantage of opportunities to profit from new data centers by making major investments and then shifting costs to their captive ratepayers. The industry’s current approaches of luring data centers with discounted contracts or lopsided tariffs are unsustainable.”
They offer several recommendations to protect consumers:
-
- Establish more rigorous guidelines for reviewing special contracts — many states now give PUCs broad discretion but no particular standard of review, and these special data center contracts seldom are rejected.
-
- End special contracts and require new data centers to take service under tariffs.
-
- Amend state law to require retail competition and allow for “energy parks” that bring generation, storage and connected customers together either isolated from a utility network or with just an export-only interconnection.
-
- Require utilities to disclose data center forecasts to foster competition.
-
- Allow new data centers to connect only if they commit to flexible operations that can reduce system costs.
Finally, the authors reject the idea that hiding subsidies for data centers in utility rates is a valid policy tool. Utility rates always have been a means to achieve economic and energy policy goals, they write, and this allows policymakers to avoid the unpopular move of raising taxes to pursue these goals.
But data center subsidies fail the traditional cost-benefit analysis applied to such subsidies, they say, and they interfere with needed reforms in the power sector.
To meet data center demand, utilities propose more of the central power generation and transmission expansions that they always have relied on, the report says, rather than using advanced technology and improved planning and operational policies. These revised policies would extract more value from the existing infrastructure but would not carry the same profit margins, and the option is being ignored.
