Germany's Sabotaged Grid Exposes the Cost of Neglected Infrastructure Defense

On September 1st, improvised explosive devices were used in a sabotage attack at the Preilack substation near the Jänschwalde coal plant in Brandenburg, Germany^[1]. At least one device detonated; others failed. No one was hurt, and 50Hertz reported that power was restored quickly^[2]. But the incident laid bare a pattern that electricity regulators across Europe and North America have been slow to name: infrastructure that has been allowed to age, consolidate, and concentrate becomes both critical and fragile, turning it into a target.

The Preilack substation is no ordinary switching point. It is one of the region's most critical high-voltage nodes, connecting Brandenburg's largest power plant directly to the 50Hertz transmission grid^[4]. That concentration of function, married to visible deferred maintenance and sparse physical security, creates both the incentive and the opportunity for attack. Germany's Interior Minister noted that rocket-like devices with conductive material were fired into extra-high-voltage lines, causing two short circuits^[4]. The substation sits about five kilometers from the main facility and distributes power generated by one of Germany's fourth-largest power stations, serving several million households^[6].

This is not the first time. In January, far-left arsonists attacked high-voltage cables in southwest Berlin, leaving tens of thousands of homes without power for nearly a week in winter^[1]. The pattern is building: a string of attacks on energy infrastructure, with far-left activists claiming responsibility for some and authorities unable to rule out foreign involvement in others^[2][4]. Each attack prompts the same regulatory response: emergency meetings, calls for "better protection," and assurances that the grid will be hardened. What does not happen is the structural shift.

Here is the mechanism that makes infrastructure vulnerable: transmission and distribution operators collect revenue for maintenance, security, and renewal across a decade or more of rate cases. That money gets diverted to shareholders via dividends, while the actual assets age. When a storm or sabotage then forces a repair, the utility requests a "resilience surcharge" or "critical infrastructure protection rider," and the same ratepayers fund the fix a second time. In Germany's regulated environment, that dynamic is harder to hide than in the U.S., but the principle is identical. The substation that was attacked should have been hardened years ago, its perimeter secured, its control systems segregated, its redundancy built in. Instead, Brandenburg's largest power generation chokepoint remained a single point of failure, accessible enough to be found by activists and vulnerable enough to be damaged by improvised devices.

The remedy is not more hardening riders after the fact. It is performance-based regulation that puts revenue at risk against actual security and resilience outcomes, with symmetric penalties and incentives. Grid operators need to be paid for outcomes (availability, recovery speed, equity of service restoration) not for the megawatts of capital they deploy. When 50Hertz spent revenue on visible undergrounding of critical feeder lines, on distributed switching stations that reduce single-point-of-failure risk, on intrusion detection and response teams, it should have earned a return; when it neglected those investments in favor of dividends, that neglect should have cut its revenues. The current framework rewards neither and punishes both.

For ratepayers in Brandenburg and beyond, the immediate cost is clear: a week without power in January, a brief outage in September, and coming bills that will be padded with "critical infrastructure investment" charges for work that should have been done a decade ago. The long-term cost is governance captured by the utilities themselves, in which "resilience" becomes a marketing term for rate increases rather than a structural outcome you can measure and hold operators accountable for.