Ef­fi­cien­cy bench­mark­ing

• The minimal regulation
• Efficiency benchmarking methods
• Input and output parameters
• Background
• Efficiency scores
• Expert report for the third regulatory period
• Further information

As part of the process to determine revenue caps,nationwide efficiency benchmarking is carried out before the beginning of a regulatory period. Efficiency benchmarking serves to establish individual efficiency scores for the network operators. Inefficiencies identified must be fully resolved within one regulatory period.

Participation in efficiency benchmarking is required for all operators with more than 30,000 connected customers (standard procedure).

TSOs are subject to special efficiency benchmarking arrangements under section 22 ARegV.

The minimal regulation

Network operators with fewer than 30,000 connected customers can participate in a minimal regulation procedure (see section 24 ARegV) and receive an average efficiency score derived from weighting and averaging all efficiency scores throughout Germany from the previous regulatory period.

Regulatory period	Efficiency score (%)
Fourth regulatory period	97,01 %
Third regulatory period	96,69 %
Second regulatory period	96,14 %
First regulatory period	87,5 %

Efficiency benchmarking methods

The basis for carrying out efficiency benchmarking for DSOs and determining individual efficiency requirements is sections 12 to 16 in conjunction with Annex 3 ARegV.

As required by Annex 3 ARegV the methods to be applied in efficiency benchmarking are data envelopment analysis (DEA) and stochastic frontier analysis (SFA).

Data Envelopment Analysis – DEA

DEA is a method providing optimal combinations of input and output from a linear optimisation problem. It determines an efficiency frontier from the data of all the benchmarked companies and their positions relative to this frontier. DEA is a non-parametric method, meaning that no functional connection between input and output can be inferred. (See Annex 3 ARegV.)

Stochastic Frontier Analysis – SFA

SFA is a parametric, stochastic method, meaning that a functional connection between input and output must be created a priori. This method uses regression analysis so that a statistical connection between input and output can be identified and the strength of this connection can be determined. Network operators' inefficiency is then established on the basis of the deviations between the actual costs and the costs estimated by means of regression analysis.

Input and output parameters

The Bundesnetzagentur is also to use input and output parameters in its efficiency benchmarking. Input parameters are firstly the total costs minus the permanently non-controllable costs. In addition, the capital costs included in the total costs are standardised with regard to the asset lifetimes and acquisition dates to even out these differences.

Output parameters are to be determined in accordance with the requirements of section 13 ARegV.

On the basis of a cost driver analysis the model for establishing efficiency scores is determined using the input and output parameters.

Background

The audited and transferred cost data are included in efficiency benchmarking to determine the model for efficiency benchmarking. As in the past, the expert consortium of Swiss Economics, Sumicsid and IAEW provide support to the Bundesnetzagentur in this work. Consumers and representatives of the economic sectors affected are to be consulted in a timely manner as a part of efficiency benchmarking (see consultation meeting).

In preparation for the consultation, the Bundesnetzagentur publishes all parameters used in the cost driver analysis and model selection (cost and structure data as well as area structure data). During the consultation, the consumers and the economic sectors affected are presented with the result of the model design for efficiency benchmarking based on what is currently known about cost driver analysis.

In addition to the reconciliation accounts from the cost examination, all network operators participating in the standard procedure receive four separate data receipts in succession to check for completeness and correctness. This is part of their obligation to cooperate.

Efficiency scores

The requirements call for the calculation of four efficiency scores (two methods, each with two input parameters), whereby the highest efficiency score is always used.

Efficiency scores for network operators must be published in accordance with section 23b EnWG.

Current scores (as from 2021) for electricity network operators are published in the data table on this page.

Expert report for the third regulatory period

Final expert report (April 2019)

Company-specific efficiency scores were determined as part of nationwide efficiency benchmarking for electricity distribution system operators.
The methodology and the results of the advisory project for determining the efficiency scores were documented in a report from 4 April 2019.
Efficiency benchmarking for electricity DSOs (report from 4 April 2019)

Further information