Line loss factors, explained.

What a line loss factor is

Electricity dissipates as heat in cables and transformers on its way to you. NESO puts the scale of it at around 1.7% across the transmission network and a further 5% to 8% across the distribution networks. Somebody generated that energy and somebody pays for it, so it is measured and allocated.

Distribution losses are allocated using line loss factors, which Elexon defines as multipliers used to scale energy consumed or generated to account for losses on the distribution networks. A factor above 1 grosses a demand volume up: it says that to put 100 kWh through your meter, more than 100 kWh had to leave the grid supply point where the transmission network hands over to your distribution network.

The class, and where to find yours

Factors are not published per meter. Roughly 30 million metering systems sit in the supplier volume allocation sector, banded into line loss factor classes, or LLFCs. Every metering system in a class carries the same factors, and the class the operator assigns also decides which distribution tariff you are billed on.

Yours is in the top block of the MPAN on the bill, and one thing about it catches people out at once. The Retail Energy Code requires the class to be printed at its full three characters, so class 1 appears on an invoice as 001 while the industry files and the operator statements carry it as 1.

Who calculates it, and who approves it

The network operator calculates its own factors under the principles in Elexon procedure BSCP128. Generic factors cover low and high voltage supplies and account for all losses, technical and non-technical. Site-specific factors cover every extra high voltage site and account for technical losses only, and a lower voltage customer can ask for one, which the operator may agree to in writing.

Generic factors come from settled data for a complete twelve-month period at settlement run R3 or later, and the principle fixes that period at three years before the year being priced, so the factors on a 2026/27 bill describe the network of 2023/24. They are split at least between day and night, using the seasonal time of day periods in the methodology, and a value outside 0.750 to 1.250 is flagged for query rather than accepted quietly.

1. By 1 AugustThe method, before the numbers

   The operator submits its methodology and a self-assessment against the BSCP128 principles. Nothing else is approved until this is.

   →
2. By 30 SeptemberThe factors are submitted

   Host operators submit their values. Embedded networks follow by 31 October, or the end of December where they copy their host.

   →
3. October to DecemberElexon audits them

   The calculations are audited against the approved methodology, and metering systems are sampled to confirm each carries the right class.

   →
4. By early MarchTwo committees approve

   The audited values go to the Balancing and Settlement Code committees, and take effect on 1 April.

An approved generic factor does not change mid-year, and the code permits no retrospective change except to correct a material manifest error. The file that carries them is the D0265, sent by the distributor to the data aggregator, the supplier and the settlement agent, and the annual submission must contain every factor for every settlement period of every settlement date in the year. That is 17,520 numbers per class in a 365 day year, once the two clock-change days are counted at 46 and 50 periods rather than 48.

Which charges it grosses up, and which it does not

This is the part that decides whether your arithmetic agrees with your supplier, and the rule is not about the charge. It is about where the charge is levied. A charge on the volume settlement attributes to your supplier at the grid supply point is loss-adjusted, because settlement applied the factor first. A charge on the units your meter recorded is not.

The distribution case reads as a contradiction and is not. Your class does two separate jobs: it names the row of distribution rates you are billed on, and it names the row of loss factors settlement will apply. The Southern Electric Power Distribution statement puts it in one sentence at paragraph 4.1: the adjustment does not affect distribution charges, but is used in energy settlement to take metered consumption to a notional grid supply point. Rebuild a distribution line from grossed-up volume and it reads several per cent high, and the error looks like a supplier overcharge.

The two levies make the same point from opposite directions. The AAHEDC charge is a grid supply point charge by construction, so it is loss-adjusted. The Nuclear RAB levy is defined on electricity supplied, so it is not. They sit side by side on the same invoice in the same units, applied to different volumes, which is why two lines covering one meter and one period can show different quantities and both be right.

One charge is missing on purpose. The Capacity Market supplier charge is calculated from demand in a winter peak window, and we could not find a published statement of which volume that is. We have left it out rather than infer it.

What the gross-up costs, worked through

Southern Electric Power Distribution publishes four factors per class for 2026/27, one per seasonal time of day period. Period 1 is the winter weekday peak, 16:00 to 19:00 from November to February; period 4 is night, 00:30 to 07:30. From March to October only the last two apply, which makes a summer month simple to reproduce.

Two things in it surprise people. The largest factor on every low and high voltage row is the night and the smallest is the winter weekday peak, because the factor is a ratio to the units delivered and the part of the loss that does not scale with load is spread over far less volume at three in the morning. And connection voltage moves it more than anything else about a site, from nearly 7% at low voltage to under half a per cent at 132kV.

Take a half-hourly site on the Low Voltage Network class in that area consuming 100,000 kWh in June 2026, with 80,000 kWh of it between 07:30 and 00:30 and 20,000 kWh overnight. The consumption and the split are assumptions; the factors and the tariffs are published.

Daytime, 80,000 kWh at 1.069

85,520 kWh

Night, 20,000 kWh at 1.078

21,560 kWh

Volume at the grid supply point

107,080 kWh

Effective factor for the month

1.0708

BSUoS at £13.74 per MWh, on 107.080 MWh

£1,471.28

The same line on metered volume, 100 MWh

£1,374.00

Difference, one charge, one month

£97.28

AAHEDC runs the same way and is smaller: 107,080 kWh at the 2026/27 tariff of 0.044269p per kWh is £47.40 against £44.27 on metered units. The distribution unit charges on that invoice are calculated on 100,000 kWh and nothing else. Seven per cent is the commonest reason a careful hand check disagrees with a bill that is correct.

Where it appears, and how to check it

It does not appear. There is no line loss line on a business electricity bill, no separate charge and usually no note. It shows up only as a quantity larger than your meter reading on some lines and equal to it on others, which is why it is often mistaken for an error. It is not one. What is worth looking for is the gross-up in the wrong place.

• Find your class. Read it from the MPAN block, allowing for the leading zeros, or ask your supplier for the line loss factor class or DUoS Tariff ID.
• Look it up. Annex 5 of your network operator charging statement carries the factors for the year, and the full set is on the Elexon Portal.
• Divide, do not guess. Take the quantity on a loss-adjusted line and divide by your metered units. The result should sit inside the published range for your class, and outside 0.750 to 1.250 it is wrong by the industry standard too.
• Check the right lines. Expect the gross-up on balancing, transmission and AAHEDC, and its absence on distribution, on the Contracts for Difference and Nuclear RAB levies, and on the Climate Change Levy.

What is changing

The factors are not changing. The way they are published is. Market-wide Half Hourly Settlement replaces the line loss factor class with a DUoS Tariff ID for migrated meter points and adds a second submission route. Both run during the transition: the D0265 as usual, on the Elexon Portal, and the same values in the Industry Standing Data catalogue. When the transition completes the D0265 stops and that store becomes the only source, so anyone whose bill checking leans on the portal files has a migration of their own to plan.

Deep dives and edge cases

The main path covers the ordinary half-hourly site. These two come up often enough to answer and not often enough to interrupt it.

Questions