The Project
Max Moore - Student Number: 1914324 - 10/01/2022

European Energy Consumption: Are The Trends Clear?

Energy consumption is critical to the everyday function of todays economies. With influence ranging from geopolitics, climate change, and the welfare of a nation, its important to look to the future and question how Europe’s consumption habits may change, and why.

Specifically, this project aims to answer the following:
- What do the past 20 years of energy consumption look like in Europe, and which countries are curbing their consumption habits the most.
- What are the sources and impact of energy production.
- By observing the primary influences in energy consumption, can we predict the future trends of European energy consumption.

20 years of consumption

The following chart compares the relative changes in final energy consumption since 2000, and evaluates each countries consumption trend from the past 5 years.

Here, we see large variation with un-clear trends for many countries. The majority of countries jump around an index score of 100, highlighting how time trends are a poor predictor for energy consumption. This is supported by the OLS regression results which say Norway have performed the best at curbing their energy consumption in the past 5 years, but when looking at the full 20 years, they have been very inconsistent at reducing their consumption.

Energy! Energy! Energy!

The following charts break down each countries final energy consumption, energy production by source, and greenhouse gas emissions by industry.

From this, we see how energy consumption is the biggest influence behind greenhouse gas emissions in Europe, with the larger economies polluting the most. However, growth in energy from renewable sources suggests that over time, we should expect the size of energy's role in GHG emissions to fall.

Influences of energy consumption

The following section examines trends in primary factors of energy demand. In the Journal of Energy Policy, S. Thomas and J. Rosenow argue that the biggest drivers behind recent changes in European energy consumption are activity levels and geographic factors.

The following two charts show how economic growth correlates with employment rate and transport activity, all of which are indicators for activity levels in the transport, industry, and services sector.

From this it can be concluded that, with consistent economic growth in the future, countries can expect increased energy demand from all three sectors.

The following two charts explore how average winter temperatures and population levels are likely to change in the future, both of which are indicators of energy demand in the residential sector.

In particular, the cold northern countries of Scandinavia can expect warmer winter temperatures as climate change worsens, leading to lower energy demand from heating. As well, falling populations imply that in the long term, energy demand will weaken.

Project Conclusion

With greater public focus on climate change and energy consumption, warmer winters, and falling population levels, long-run energy consumption should fall. However, it is generally very difficult to predict each countries trend in energy consumption due to number of inputs that influence demand, and a lack of clear predictors as to why some country’s consumption trends differ from others. With data such as technological efficiency improvements and government climate stance, OLS regressions could be run to better predict how each countries energy consumption may change in the future.

Data: Access, Automation, Cleaning, and Analysis

The data for this project was sourced from multiple agencies and websites, using GitHub and API URLs to access the raw data via python. In particular, the following sources were used:

- API URLs from the Energy information administration and Eurostat were used to obtain energy and economy data.
- OECD, Worldbank, and United Nation datasets were uploaded to GitHub.
- Scraped Wikipedia webpages and image-samples from Copernicus were used as sources for temperature data.

Using the provided Google Colab code, each charts data and analysis can be replicated, as all sources of data are publicly available via URLs.

Each chart presented its own difficulties when sourcing data. In particular, APIs that wouldn’t allow for multiple countries to be requested at the same time (EIA API), and who’s format was complex and difficult to work with (OECD API) were tricky to work with. This was solved by looping API requests, and directly downloading data sets when APIs weren’t appropriate (such as historic yearly data).
The trickiest data to source was historic weather, with most APIs and datasets hidden behind paywalls. To resolve this, I scraped average monthly temperature figures from Wikipedia, and used gray-scaled color samples from the Copernicus produced ‘Weather anomaly’ maps to estimate relative anomaly differences in Europe.
To better analyze trends in energy consumption, moving averages, indexes, and OLS time regressions were calculated. This was achieved using the Pandas, Numpy, and Sklearn packages in Python, and by doing so, better comparisons can be made between the relative performance of each country.