A common argument against CRE/BYOP private grids is that they will compete with the public grids for the limited supply of parts. That’s a valid concern in a zero sum world, but we don’t live in a zero sum world. Economic growth and expansion happen everyday when demand increases.
If the consumers using the private grids were on the public grid, much of the same equipment would be needed to provide them power.
CRE/BYOP will increase the demand for supply. When the demand increases, the supply will follow. That’s Econ 101.
Traditional equipment suppliers are accustomed to long delays (e.g., look at the stop/start nature of ERCOT’s 765KV lines) so they don't expand supply as quickly as one might expect. Since CRE/BYOP developers won't have the stop/start nature to their projects, traditional equipment suppliers will be more aggressive with expansion plans.
CRE/BYOP developers will expand the number of suppliers. They won't be tied to the traditional suppliers. We're seeing this play out in real time as projects have been announced using Caterpillar engines and fuel cells among other options. These types of suppliers weren't being considered by anyone just a year or two ago. Moreover, we’re seeing new gas turbine manufacturers enter the picture. For example: https://finance.yahoo.com/energy/articles/elon-musk-gas-turbines-could-040029538.html.
A recent substack post spoke to this issue more boldy: "Another core issue is power, and again we think the issue is largely solved. We’ve been the first to talk about all issues on power, since March 2024 and subsequently given the grid headroom dips negative at the end of 2027. This ended up driving AI Labs and hyperscalers to build their own power plants, better known as Behind-the-meter generation, bypassing 5-10 year grid interconnection timelines. And while turbine incumbents like GEV quickly faced production bottlenecks, the industry quickly reacted and we’ve since seen a surge in new OEMs and new solutions supplying the market. From fuel cells to truck engines to refurbished turbines, the power generation problem has broadly been solved." https://newsletter.semianalysis.com/p/datacenters-2027-the-year-of-acceleration
Regulated utilities have some artificial incentives that cause them to suboptimize which CRE/BYOP developers won't share. For example, a regulated utility is happy to have equipment custom built because (a) they don't face competition so there's no pressure to do otherwise and (b) they actually make more money if the equipment is more expensive. That's a weird incentivefor a business, right? A CRE/BYOP developer will ruthlessly cut costs and standardize equipment. That means they won't directly compete with the utilities for equipment.
CRE/BYOP developers will show us how a 21st century grid should work and legislators can ultimately take those lessons and incorporate them into their public grids.
This is a globally competitive business. If the U.S. isn’t the first to make it easier to build 21st century grids, someone else will.
