The curves of power
Curveballs?
Europe is arguing over the power price again. Italy’s latest move matters because it goes straight at the cost stack before bids are compared. That is why the merit-order debate is live again.
What matters here is not so much the rhetoric around market design as the decree behind it. By proposing to reimburse gas-fired producers for ETS costs, Italy was not simply offering relief after the market cleared. It was trying to weaken carbon-cost passthrough inside the wholesale price mechanism itself.
Lion Hirth’s formulation is the right place to start. The merit order is not a peculiar legal rule that somebody once chose for electricity. It is the short-run supply curve. Cheaper supply comes first. Dearer supply comes later. Demand cuts through the stack somewhere. The politics begins when the marginal unit becomes too expensive to tolerate.
That is also why so much public discussion goes wrong. The argument is often framed as if the whole issue were a single auction rule: the highest accepted bid sets the price, everyone else gets paid that price, therefore the rule itself must be the scandal. But payment rule and market logic are not the same thing. Change one auction from pay-as-clear to pay-as-bid and participants change how they bid. The underlying marginal logic does not simply disappear.
The day-ahead market is where most people meet the issue. It is the headline price. It is the political object. But it is not the easiest place to see the underlying structure. The official crossing is visible on EPEX. The richer book is not. That is why reserve is such a useful side door. In German balancing procurement, the awarded edge can actually be seen. Ranking becomes concrete there in a way that public wholesale discussion rarely is.
That is the purpose of the new lab in our interactive mode. Not to settle the argument. Not to reproduce the hidden order book. To make the structure visible enough that the argument becomes more precise. It moves from the current political fight into public reserve curves, then into day-ahead clearing, then into a transparent benchmark reconstruction, and finally into the differences between ranking, selection, activation, and payment across products.
The reconstruction matters because price is visible, but the stack behind it is not. Public load, wind, and solar can get you part of the way. A stylised supply stack can get you a little further. The gap is where the omitted structure starts to matter: outages, constraints, block bids, portfolio behaviour, hydro opportunity cost, cross-border congestion, all the things that turn a clean diagram into a real market. That is not a bug in the exercise. It is the point of it.
FCR and aFRR then help with a second confusion. They show that the same underlying ranking problem can appear in different institutional forms. Day-ahead and FCR both end with a common price for successful bids. aFRR capacity keeps the ranking logic but changes the remuneration rule. aFRR energy makes price-ranked activation more explicit again. Seen together, the examples stop the debate from collapsing into slogans about one villainous mechanism.
That is where this becomes interesting again. Not because the merit order is sacred. Not because intervention is unthinkable. But because once you can actually see the curve, you can finally say what kind of intervention is being proposed: after the market clears, inside the market before it clears, or in place of the mechanism altogether.