"Indexed price" does not exactly match other sources of indexed price information

[darteaga]

There are some small discrepancies between the "indexed price" (not the PVPC) as reported by pvpc-next and the indexed price as reported by "Som Energia".

For example, for today (23 Feb), the indexed price reported by pvpc-next is:

And the indexed price reported by Some Energia:

Notice that e.g. from 4h to 5h the price reported by pvpc-next is 0.0672 €/kWh and in contrast the price reported by Som Energia is 0.0385 €/kWh.

I understand that Som Energia adds some margin to the price quoted (actually 0.02 €/kWh), so the two values won't be necessarily identical. However this margin should imply that the value quoted by Som Energia is actually higher than the value quoted by pvpc-next, when in fact is the other way round.

I know that the issue of the electricity pricing is complex, so I might be missing something... I tried to look at the code myself, but couldn't fully understand.

Thank you very much!

[privatecoder]

I have to say the curve looks quite similar but with a small offset.

The issue is, that I just re-activated the old integrations indexed price which is calculated backwards from the PVPC price, i.e. INDEXED = PVPC - ADJUSTMENT (rounded to 5 decimal places).

After a quick look I couldn't find a separate API indicator from ree.es that would allow me to get the indexed price directly.

Let me know if you find something.

[privatecoder]

From what I found:

PVPC contains a futures component since 2024 (RD 446/2023): PVPC is no longer pure spot. In 2025 it was 60% spot / 40% futures; in 2026 the futures share may be even higher. An indexed tariff from the free market is based on the pure OMIE spot price, not PVPC.

Network losses (pérdidas): Hourly loss coefficients (~4–17.5% for low voltage) are applied as a multiplier (1 + Perd_h) on the spot price. PVPC already bakes these in differently than the indexed formula does.

Additional cost components: The indexed formula includes several small regulated surcharges (capacity payments, system operator costs, deviations, etc.) that don't cancel out cleanly in a simple subtraction.

The accurate formula would be:

Indexada_h = (OMIE_spot(10211) + Servicios_ajuste) × (1 + Perd_h) × 1.015 + Peaje_energia[periodo] + Cargo_energia[periodo] + FE + Marge_comercializadora

I need to think about implementing this or not.

[darteaga]

Thank you for your reply. I understand this is a complex issue.

Most Spanish electricity retailers (“comercializadoras”) offering indexed tariffs do disclose the formula they use. However, they generally do not publish the final hour-by-hour results of applying that formula. Instead, they often reference PVPC or the OMIE spot market, stating that the result is “similar,” but without providing the computed values.

So far, Som Energia is the only exception I’ve found that publishes hourly values, although they label them as a “trend” (“tendència”), since they claim that additional correction factors are applied a posteriori.

This makes it difficult to validate an implementation against real billed data, and it remains unclear whether all retailers use the same formula and apply the exact same adjustments (beyond their margin). In particular, the formula applied by Som Energia is not identical (although very similar) to the one you quote.

It is surprising that it is so difficult to obtain precise information about the actual price reflected in one’s electricity bill.

I will reach out to Som Energia to ask whether they can provide either an API or a precise description of how to compute the final price using ESIOS data, and get you updated.

For the time being the values offered by your code are good enough as a trend.