Elnätsavgift — What Determines Your Bill

Every Swedish electricity bill is really two bills. One is for the electricity itself — the commodity, bought from whichever supplier you’ve chosen on the open retail market. The other is the elnätsavgift (network fee), paid to the company that physically owns and operates the wires reaching your building. The two are easy to conflate because they usually arrive as separate line items on the same invoice, but they work on completely different logic: you can switch electricity supplier in minutes, at no cost, among dozens of competing retailers. You cannot switch your nätbolag at all. It is assigned to you by geography — one company per address, holding a regional nätkoncession (grid concession) that functions as a legal Natural Monopoly. Sweden has 168 elnätsföretag, and which one you fall under is simply a function of where you live.

That asymmetry is the reason the elnätsavgift gets regulated so differently from a normal market price, and it’s the reason this page exists: the fee is not something you can shop around on, so the only real lever a customer has is understanding what determines it and whether the regulator thinks it’s reasonable.

What you’re actually paying for

The elnätsavgift covers the cost of building, maintaining, and operating the physical grid: cables, transformer stations, metering, fault repair, and the capital cost of the assets involved. In Sweden it is commonly structured into up to four components — energiavgift (a per-kWh charge covering losses), effektavgift (a forward-looking, often time-differentiated capacity signal — see the aggregate-load principle governing how it must be designed), kundspecifik avgift (metering and administration), and fast avgift (a residual fixed charge). Not every DSO uses all four; as of early 2026, just over 30 of Sweden’s ~171 DSOs applied an effektavgift (Source - Ei R2026-04 Förhandsprövning Avgifter (2026)) — around 13% of households had actually received one as of spring 2025 (Source - EIFS 2026-8 Nätföretags Information till Elanvändare (2026)).

This is distinct from the anslutningsavgift (connection fee) — a one-time charge paid when a new facility first connects to the grid, or when its capacity is increased. The anslutningsavgift is a separate reform track (EIFS 2026:10, in force 1 January 2027 — see Anslutningsavgift 2027 — What Changes for DSOs and Customers) governing a single upfront payment; the elnätsavgift is the recurring charge that continues for as long as you’re connected. A customer facing a large anslutningsavgift can sometimes trade it down via a villkorat avtal (accepting a lower guaranteed capacity for a cheaper connection) — but that trade-off doesn’t apply to the ongoing elnätsavgift itself.

Why the bill has grown faster than inflation

Two separate government-commissioned reports, asking different questions from different legal starting points, arrived at the same headline finding within weeks of each other in 2026.

Ei’s R2026:05 (June 2026), commissioned to assess the outcome of current tariff regulation, found household network fees rose +115% between 2005 and 2026, against general consumer-price inflation (CPI) of +49% over the same period — more than double. Larger-fuse customers (mostly firms) saw +85%.

Konkurrensverket’s competition-law review (July 2026), looking specifically at the 2024–2027 tillsynsperiod from a different angle, found average villa network tariffs rose +70% between 2012 and 2025, against KPI of +30% — again, well over double.

The two figures aren’t directly comparable (different start years, different customer segments), but the direction and the reason are the same, and neither report frames it as isolated company-level pricing behaviour. Swedish DSOs are natural monopolies regulated in advance through four-year intäktsramar (revenue caps, see RP5 Revenue Cap Methodology (2028–2031)). The capital-cost component of that cap values a DSO’s entire asset base — not just newly built assets — at nuanskaffningsvärde: what it would cost to build the equivalent grid today, rather than what the company actually paid for it historically. Since 2012, construction costs (byggkostnadsindex, BKI) have risen almost twice as fast as general inflation (KPI). Because a DSO’s actual capital financing competes in the broader capital market — not the construction market — Konkurrensverket argues KPI, not BKI, is the more accurate proxy for what capital genuinely costs a grid company. Valuing the whole asset base at BKI-linked replacement cost therefore risks crediting DSOs with capital costs well above what they actually pay to finance those assets — a mechanical, structural effect built into the valuation method itself, not a matter of individual company conduct. Ei’s own discount rate (kalkylränta) for local networks rose from 3.39% to 4.53% between the 2020–2023 and 2024–2027 periods, compounding the effect.

The same period shows DSO finances comfortably strong by conventional measures: median soliditet (equity ratio) 46–51%, profit margins 16–23%, return on equity 9–13% — all above Sweden’s more market-risk-exposed electricity retail and generation sectors. Roughly 80 of Sweden’s ~165 local/regional grid companies were net payers of koncernbidrag (group contributions to a parent company) in most years 2020–2024, totalling around 20 billion SEK cumulatively — money that, structurally, comes from the same revenue cap that determines your bill, even though it isn’t spent on the grid itself.

One further, less visible driver noted by grid companies themselves: the energy tax is invoiced through the network bill rather than the electricity bill, which inflates the visible size of the “nätavgift” line without any of that increase actually being network cost.

What’s being fixed, on three tracks at once

The two reports above are the empirical backbone for regulatory action already underway — see Swedish DSO Tariff Reform — Three Parallel Tracks (2025–2027) for the full detail. In brief:

• A new capital-valuation method for RP5 (2028–2031) — Ei intends to move from kapacitetsbevarande (replacement-cost) to förmögenhetsbevarande (historical-cost) valuation, the method most other European regulators already use. This is prospective only: it will reshape tariffs from 2028 onward but does not revisit the already-decided 2024–2027 caps.
• Förhandsprövning (pre-approval) of tariff methods — Ei has formally proposed (Source - Ei R2026-04 Förhandsprövning Avgifter (2026)) that it gain the power to approve a DSO’s charging methodology before it takes effect, rather than only reviewing it after the fact through tillsyn. ACER’s 2025 review found Sweden and Finland were the only two EU member states where the regulator does neither.
• Active tillsyn (supervision) right now — in April 2026 Ei opened a formal investigation into five named DSOs’ effektavgift designs (Source - Ei Tillsyn Elnätsavgifter (2026)), checking compliance with ellagen’s objective/non-discriminatory/reasonable standard and Art. 18 of the EU Electricity Market Regulation. Konkurrensverket, separately, says it will keep reviewing actual 2024–2027 prices against actual DSO costs, with particular attention to how companies price the final year of the period (2027) — pricing plans no company would disclose in advance.

None of this changes what you’re billed today. All three tracks converge on 2027–2028.

Where an effektavgift applies, a home battery can in principle cut the charge by cutting the monthly peak, but Swedish studies find the saving small unless the control targets the peak. One Karlstad study of a house (peak charge from the average of the three highest hours of the month) found that a battery used only for self-consumption did not pay off, and a RISE and Energimyndigheten study found limited effect of batteries on peak power tariffs; both point to revenue stacking as essential (Source - Swedish Household PV-BESS Studies (Rinio 2025, Ollas et al 2026)). See Power Tariffs and Home Batteries for the quantified Energiforsk simulation. A Dalarna study of Sälen holiday homes likewise found that a battery adds clearly to PV only where the demand charge is high and the load peaks across the year (Source - Fiedler Holiday Homes PV Battery Demand Charges (2022)).

How to read your own bill, from 2027

From 1 January 2027, a new rule (EIFS 2026:8) requires any DSO applying an effektavgift to show the specific effektavgiftsgrundande mätvärde — the actual metered values your capacity charge is computed from (e.g. the highest hourly mean power in the month) — on mina sidor, with timestamps and an explanation of how they feed the charge, and to reference them on the invoice itself. Separately, Ei had clarified (ställningstagande Ei2025:06, repealed 17 June 2026 along with the EIFS 2022:1 framework it was grounded in) that a time-differentiated effektavgift must be set from the aggregate load on the DSO’s whole grid, not solely from your own peak hours — a DSO cannot set “expensive” pricing windows purely around when you happen to use the most power, without checking whether those hours actually coincide with system-wide peak demand. That principle is no longer a live Ei position on paper, but its substance is expected to carry into the replacement effektavgift model due 12 April 2027.

Data gaps

• Whether Ei’s R2026:05 recommendation for annual public transparency (named lists of DSOs with high internal-loan costs, profits, or dividends) has actually started being published on ei.se
• Konkurrensverket’s stated 2027-pricing follow-up review — outcome not yet public