The DATA Act offers short-term relief from policy-driven energy costs, but could weaken grid efficiency and raise long-term costs for consumers.
It is no secret that years of perverse federal and state policies promoting wind and solar electricity as replacements for natural gas- and coal-fired power have yielded large increases in costs and decreases in reliability. That is why, for example, California has the nation’s highest rates and a serious reliability problem. For wind and solar power, the two main drivers of higher costs are the costs of backup power (from natural gas turbines) to avoid service interruptions, and the costs of much-longer transmission systems.
Such policy-driven costs—the result of ideological imperatives and interest-group pressures—provide no additional value to consumers of electric power. Furthermore, permitting delays for power plants and transmission systems are so large as to represent a serious threat to the nation’s power infrastructure, a state of affairs that cries out for Congressional actionyielding faster and more predictable paths for infrastructure investment and integration. And for large consumers of power, the increasing risk of service interruptions raises the possibility of some form of rationing, which inevitably would be driven by political pressures.
Grid Exit and Self-Generation Trends
Accordingly, it can surprise no one that large demanders that can avoid such costs by exiting the larger grid system are availing themselves of that opportunity. This is the case particularly for data centers, which are increasingly installing their own generation capacity on their sites. Over the longer term, an expansion and modernization of the electric grid serving most or all customers would serve to reduce aggregate costs and to spread the fixed costs of the system over a broader customer base, that is, to exploit the large-scale economies that are a feature of a modern electric power system.
But in the near term, the inefficient policy-driven costs are a fact of life analogous to the high taxes and oppressive regulatory environments imposed by some state and local governments. Such policies provide incentives for individuals and businesses to move from high-tax, high-regulation states to geographic alternatives with more efficient policy environments, even though doing so leaves the fixed costs of the high-cost state public sectors to be spread across a smaller number of taxpayers. Both the ability of large power users to provide their own capacity and the ability of individuals and businesses to vote with their feet impose important constraints on the ability of governments to impose perverse policies.
For power consumers who remain on the grid system—most industrial, commercial, and residential users—this dynamic has both upsides and downsides. On the positive side, the constraints on the imposition of inefficient costs are real. On the negative side, the fixed costs—large capital costs in particular—that are already in place can similarly be spread across fewer consumers, each of whom must bear a larger share in the here and now.
The DATA Act and Deregulated Power
Senator Tom Cotton (R-AR) has introduced Senate Bill 3585, the “Decentralized Access to Technology Alternatives (DATA) Act of 2026,” which, in brief, would exempt from federal regulation all new power generation facilities not connected to the grid. This bill would have the effect of addressing at least in part the large near-term cost problems created by policy favoritism for wind and solar power. The ability of important segments of consumer demand to avoid such costs by providing their own capacity provides a crucial long-run constraint on the ability of bureaucrats and politicians to engage in perverse policymaking. Political pressures for states to impose inefficient regulation of such facilities might remain, but competition among states for data center investments would impose sharp limits on such meddling.
At the same time, Cotton’s bill creates a longer-term problem: less exploitation of the scale economies characterizing the electric power infrastructure. In the short term, generation capacity needed to serve peak demand is a good example because, for the most part, the US structure of electricity prices does not charge rates based on daily or seasonal peaks. Such costs simply are imposed on all consumers, and with fewer consumers those remaining must bear higher costs. Note that data centers, which use large amounts of power at a more-or-less constant rate, largely are “base-load” rather than peak consumers. (As outside temperatures fluctuate, cooling needs for the servers might fluctuate somewhat.)
This means that as data centers drop off the grid, the costs of peak capacity and generation must be spread across fewer consumers, each of whom must bear a higher burden. Time-of-use pricing could be implemented given the installation of “smart” meters in homes, businesses, and factories, but the privacy and civil-liberty implications of a system in which the government knows each consumer’s power consumption on a minute-by-minute basis are not salutary.
Cost Shifting and Ratepayer Burden
The larger cost-allocation problem has been recognized by President Donald Trump and many others explicitly: Mr. Trump announced in the State of the Union address a “ratepayer protection pledge” under which companies would pay increased electricity costs in areas where new data centers are built. And in recent news, the large technology firms are reported to have promised: “to bear the cost of new electricity generation to power their data centers.” The precise meaning of that promise is murky: Does that mean that they will subsidize the power consumers remaining on the grid so that the higher average costs will be covered? That is doubtful; the cost problem created by data centers exiting the grid would remain.
Grid Fragmentation and Long-Term Efficiency Risks
Sen. Cotton’s bill would reduce the problem of inefficient costs, but in the longer run, it also will serve to “balkanize” the grid, increasing system-wide average fixed costs that would have to be spread across a smaller consumer base. What is needed is legislation ending federal meddling in the physical architecture of the electricity system, combined with state-level efforts to allocate peaking costs by broad consumer segments correlated with peak demands, rather than through individualized monitoring of power meters. Sen. Cotton would be well-advised to amend the DATA Act to end such federal interference with cost efficiency in the electric power sector.
About the Author: Benjamin Zycher
Dr. Benjamin Zycher is a senior fellow at the American Enterprise Institute, where he works on energy and environmental policy. He is a former senior economist at the RAND Corporation, a former adjunct professor of economics at the University of California, Los Angeles (UCLA) and at the California State University Channel Islands, and is a former senior economist at the Jet Propulsion Laboratory, California Institute of Technology. He served as a senior staff economist for the President’s Council of Economic Advisers, with responsibility for energy and environmental policy issues. Dr. Zycher has a doctorate in economics from UCLA, a Master in Public Policy from the University of California, Berkeley, and a Bachelor of Arts in political science from UCLA.
