Planning the Energy Transition With Your Eyes Closed
Ignoring uncertainty doesn’t make it go away — it makes blackouts more likely and the transition less fair
Here is something that should worry anyone who cares about the energy transition: the models that governments use to plan it largely assume the future is predictable. Steady demand growth. Stable fuel prices. Reliable wind and sunshine. Plug in the numbers, optimize, build the plan.
The real world, of course, does not cooperate. The COVID-19 pandemic upended demand forecasts. The war in Ukraine sent gas prices through the roof. Heat waves and cold snaps push electricity consumption to extremes that no trend line anticipated. Each of these shocks reshuffled the economics of which power plants to build, which to retire, and how much backup capacity to keep around.
So what happens when you plan for a predictable future but live in an unpredictable one?
The hidden cost of certainty
In a paper published in Energy Economics with Alicia Bassière, we built a generation expansion model for Germany’s electricity sector through 2040, comparing what happens when planners account for uncertainty versus when they don’t. We simulate shocks to demand, fuel prices, and renewable output, the three big unknowns that keep grid operators up at night.
The results are sobering. Deterministic planning, the approach most regulators actually use, systematically underinvests in capacity. It looks cheaper on paper, but that’s because it ignores the scenarios where things go wrong. When those scenarios materialize (and some always do), the system doesn’t have enough power to go around. In many of our simulations, deterministic planning produces twice as many outage hours as probabilistic planning.
And those outages are not shared equally. Power shortages and rationing hit lower-income households hardest, the same households that already manage their energy consumption most carefully and have the least capacity to absorb disruptions. Planning that ignores uncertainty doesn’t just produce a less reliable grid. It produces a less fair one.
Can policy fix this?
The good news is that the right environmental policies can substantially offset the risks of deterministic planning, even without switching to more complex probabilistic models.
Among the policies we tested (carbon pricing, renewable targets, carbon budgets, coal phase-outs), a carbon budget consistently delivered the best balance between emissions reduction and system reliability. By capping total emissions rather than prescribing specific technologies, a carbon budget incentivizes a diverse mix of investments: renewables for cheap clean energy, but also dispatchable gas capacity as backup. This diversity is precisely what makes the grid resilient to surprise.
By contrast, policies that push a single technology (like aggressive renewable targets without complementary investments) can inadvertently increase outage risk by making the system overly dependent on weather. Coal phase-outs, while important for emissions, need to be paired with adequate replacement capacity to avoid reliability gaps.
The bottom line
The energy transition is happening, and it should. But planning it as if the future were certain is a recipe for underinvestment, more blackouts, and costs that fall disproportionately on the most vulnerable. Acknowledging uncertainty is not an argument against ambition. It is an argument for smarter policy design, one that builds resilience into the system rather than hoping for the best.
Bassière, A. and Benatia, D. (2025). “Moving Forward Blindly: Uncertainty, Reliability, and the Energy Transition.” Energy Economics, 151, 108883. DOI.
