Uncertainties in energy statistics data – Causes and effects of deviating energy balances

Energy balances pose one of the most important input parameters for energy system modelling. Yet, in the data preparation and processing phase of our project eXtremOS , we discovered that a number of national energy balances (e.g. the German energy balance issued by AGEB ) deviate significantly from the national energy balances issued by international institutions such as the European Statistical Office (Eurostat)  or International Energy Agency (IEA).

Differences occur mainly with respect to final energy consumption, and result predominantly from methodical differences between these energy balances. While slight differences between nationally issued and Eurostat energy balances have existed in the past, we discovered more noticeable deviations in the most recent Eurostat data. These differences result from the balance building method based on the most recent Eurostat balance, which was released in early 2019. Such high deviations can have a strong impact on the outcome of energy system models.

This website article provides an overview on the most frequent causes for deviations, as well as some resulting consequences. The examples provided below refer to Germany. We compare the AGEB energy balance to two different versions of the German energy balance issued by Eurostat. All balances show annual consumption data for 2016.

Possible causes for deviations in energy balances

To begin with, frequent causes for diverging energy balances are presented. Similar, yet less current analysis have been performed in /EEFA‑01 18/ or /EEFA‑01 12/. To understand the root of the problem, the different energy balancing procedures for Germany are presented in a simplified overview in Figure 1. The figure shows the process for the German national energy balance as constructed by AGEB, and the IEA and Eurostat balancing procedure. As displayed, each institution gathers data from the same source: the national statistical office. In a next step, an institution-specific method is used to build the energy balances. The process of building a multinational energy balance involves the additional step of reporting a joint annual questionnaire (JAQ), which is created via a standardised format for all purposes beyond national borders. As all energy balances are based on the same national statistical data, the total sum of accounts when balancing the primary energy consumption is identical. Deviations occur with respect to the allocation between energy carriers and the final energy consumption and transformation input categories.


Process of building national and multinational energy balances

Figure 1: Process of building national and multinational energy balances for the case of Germany as described by /EEFA‑01 18/

Based on the data analysis and complemented by /EEFA‑01 18/, table 1 summarizes the major causes for deviations between national and multinational energy balances. Root causes for deviations are balancing methodologies, publication dates and organisational causes.


Table 1

Table 1: Most prominent causes for deviations between energy balances /EEFA‑01 18/


Case study: Comparison of energy balance data for Germany

Focus of the analysis lies on data for 2016, which presents the latest available figures for which the AGEB energy balance as well as the Eurostat energy balance for the previous and the new balancing method exist. Figure 2 compares the values for final energy consumption differentiated by main energy carriers.


Final energy consumption by energy source

Figure 2: Final energy consumption by energy source and total (right) in Germany 2016 for all three energy balances

Attention should be paid to the consumption of solid fossil fuels and oil and petroleum products. Especially the values of the new Eurostat balances display substantial deviations in the magnitude of 230 PJ (58 %) and 360 PJ (9 %) in the year 2016.

The final energy consumption of the two main energy sources solid fossil fuels and oil and petroleum products is assessed in more detail in figure 3, where each bar shows the difference between two of the discussed energy balances.


Deviations in final energy consumption

Figure 3: Deviations in final energy consumption in the industry (left) and transportation (right) in Germany 2016 between the three analysed energy balances

The left side of the figure illustrates that the total industrial consumption differences mainly result from the energy carrier solid fossil fuels. The same is true for the case of oil and petroleum in the transportation sector. It is noticeable that the old Eurostat balance accounted for more oil and petroleum products consumption than AGEB, whereas the new Eurostat balance reports significantly less consumption for this energy source and sector. We suspect that these two cases result from difference in accounting methods. To verify this suspicion, a more detailed analysis of the energy balances is performed for the industry sector. Causes for differences in the transportation sector are more complex and therefore not covered in detail at this point.

Figure 4 shows the energy consumption details for two different usage categories: final energy consumption (right) and transformation input (left).The bars shown on the right hand side of figure 4 depict the difference in final energy consumption of solid fossil fuels in blast furnaces, which is almost equivalent to the solid fossil fuels consumption differences shown in Figure 3. The left part of figure 4 shows the differences between all balances by pairs with regard to the transformation input of solid fossil fuels in blast furnaces.

Deviations in transformation

Figure 4Deviations in transformation input for solid fossil fuels in blast furnace (left) and in final energy consumption for a part of solid fossil fuels (right) in Germany 2016 between the three energy balances

It is apparent that by subtracting the left hand side of figure 4 from the right hand side, each bar of the same colour would (almost) equal zero. This observation translates to the following finding: the share of solid fossil fuels used in blast furnaces is accounted for as final energy consumption in industry in the energy balances of AGEB and in the old Eurostat energy balance and is allocated to the transformation input section in the new Eurostat method.

Bottom line and outlook

A change in the balancing method of the new Eurostat energy balance is the main cause for significantly increased differences between Eurostat and AGEB. In older Eurostat balances, major differences in final energy consumption compared to AGEB were as high as 50 PJ. Using the most current Eurostat balance and comparing it to AGEB as well as the old Eurostat balance leads to differences in the range 200 to 400 PJ and in some cases 600 PJ.

For the specific example of final energy consumption of solid fossil fuels in the industry sector, the new Eurostat balance value exceeds the AGEB value by 61 %, making these two balances almost incomparable for the year 2016. It remains to be seen whether AGEB and other national institutions adapt themselves to the Eurostat method. The range of differences caused by the new Eurostat building method as well as by the old method has yet to be determined and allows for much more detailed analyses concerning this issue. Furthermore, the effects of such statistical deviations on the results of energy system models has to be analysed in detail and poses an idea for further research.

EEFA‑01 12

Buttermann, Hans Georg et al.: Bestimmung des „Bruttoendenergieverbrauch“ nach den Vorschriften der EU-RL/2009/28/EG auf Basis der Daten der   AG-Energiebilanzen (inklusive revidierten Daten für die Jah-re 2005 bis 2009) - Kurzstudie im Auftrag des Bundesministerium für Wirtschaft und Technologie (BMWi). Münster: EEFA – Energy Environment Forecast Analysis GmbH & Co. KG, 2012.

EEFA‑01 18

Buttermann, Hans Georg: Der Beitrag der amtlichen Energiestatistiken zum Abbau von Divergenzen zwischen nationaler und internationaler Energiestatistik - 2. Nutzertagung Energiestatistiken. Wiesbaden: EEFA – Energy Environment Forecast Analysis GmbH & Co. KG, 2018.

Um unsere Webseite für Sie optimal zu gestalten verwenden wir Cookies. Durch die weitere Nutzung der Webseite stimmen Sie der Verwendung von Cookies zu.
Weitere Informationen zu Cookies erhalten Sie in unserer Datenschutzerklärung.