Frequently Asked Questions

In short, all cities were assigned a percentage relative to the other 99 cities, and ranked on that percentage. The lowest scoring city (e.g. the city with the highest metric tons of CO2 per capita) receives a 0% and the highest scoring city receives 100%. All cities in between are assigned a proportional score. The city that is assigned a 100% in the green space indicator does not commit 100% of its space to open parks, rather it has a larger share of its city land committed to green space than the other 99 cities. The indicators were weighted and then each city is given a ranking in the sub-index. Then a city’s ranking across all three sub-indices is averaged to give the city its overall ranking.

In more detail, the Sustainable Cities Index is constructed by a three-stage averaging process. Some of the indicators are composites, meaning these take an average of their component sub-indicators. In most cases this is the simple average, however given the importance of housing costs to household spending the affordability index was weighted 70:30 in favor of housing. The three sub-indices are calculated by taking weighted averages of their component indicators and the overall score is calculated by taking the simple average of the three sub-indices.

Even where there is no weighting system applied, since the number of indicators differs across sub-indices, the weights in the overall index do implicitly differ. The same applies for the sub-indicators: two components which go into one indicator will naturally have half the weight of another indicator within the same pillar which has only one component.

The averaging process demands that the scores be converted into common units, for which we use percentages. Each is scaled such that the worst-performing city receives 0% and the best performer receives 100%. Since the sub-indices and the overall index are simply averages of the indicators, they are also measured in percentage terms.

Several of the indicators have outlying values – these are defined as observations two standard deviations away from the mean. These are given the maximum or minimum score, as appropriate, and the next-highest/lowest value is defined as the boundary observation which is used to calculate the scores of the other (non-outlier) values.

City-level data are used wherever possible, though in some cases only national-level data exist. Where there is no comparable city-level data across countries, the national value is taken, and a national database is used to scale the cities so that they are given a spread around the national average.

In 2018 we have revised the calculation of the indices to give greater emphasis to the digital capabilities of cities. We use the adoption of digital solutions as a proxy measure for the pace at which cities are equipping themselves to meet future needs. Connectivity, mobility, citizen engagement and disaster management are all areas where the adoption of new digital solutions will enable cities to accelerate their sustainable development.

Arcadis commissioned the Centre for Economics & Business Research (www.cebr.com) to conduct the research. They are a well-respected economics consultancy based in the UK. The city archetypes and clusters are the work of Arcadis.

The cities were chosen as a representative sample of 100 leading cities in the world spread across the continents. They represent a wide geographical, political and economic range as well as varying urban sustainability challenges. They were also chosen as cities that have sufficient available data to allow for international comparison.

For the Index, Arcadis commissioned the Centre for Economic and Business Research (Cebr), a recognized and established research firm and their team of senior economists to compile the information. Cebr used their vast databases and connections to compile the data from globally credible sources (i.e. the World Health Organization, CDP, Siemens etc.) Most sources are publicly available, and all are listed in the Appendix of the report.

London is ranked the world’s most sustainable city in 2018 with particularly high scores in the People and Profit pillars. The results highlight that strengths reflected in London’s status can offset challenges associated with affordability and congestion. London’s Planet ranking is lower, but still in the upper quartile, reflecting air quality and waste management issues seen in other large cities.

Interestingly, Seattle, ranked 19^th, is the epitome of a balanced city, ranking around the upper quartile in all three pillars. Sixteen European cities are in the upper quartile. Many of these, Amsterdam and Vienna, for example, are balanced cities where our measures suggest that the needs of citizens, business and the environment are all served well.

These 31 criteria were selected by Arcadis subject matter experts. The indicators used to reflect the criteria were sourced by Cebr who was commissioned by Arcadis to conduct the actual research. The indicators were selected on the basis of the information being available across all cities, the source credibility, and more. All indicators, explanations, and weightings justifications are in the Appendix of the report.

People: Education, Health, Demographics, Income Inequality, Work-life Balance, Crime, Affordability, Access to Public Transport Services, Transport Applications and Digital Capabilities, Cultural Offerings, Cost of Broadband, Access to Digital Public Services – Property Taxes, and Wi-Fi Availability.

Planet: Environmental Exposure, Green Spaces, Energy, Air Pollution, Greenhouse Gas Emissions, Waste Management, Bicycle Infrastructure, Electric Vehicle Incentives, Negative Emission Technologies – Carbon Capture and Storage, Water & Sanitation and Natural Disaster Monitoring.

Profit: Transportation Infrastructure, Economic Development, Ease of Doing Business, Tourism, Connectivity, Employment and University Technology Research.

The Sustainable Cities Index 2018 incorporates an even wider range of indicators than previous editions to capture how well cities are managing this delicate balance. Therefore, we should not seek to compare the results of the 2016 Sustainable Cities Index and the 2018 report as they are different data sets and measurements.

A weighting system is applied to the calculation of sub-index scores. To determine the weightings, the individual indicators are ranked from most important to least important in terms of how appropriate and reflective the indicator is as a metric for urban sustainability and quality of life. The most important metric is then assigned the highest weighting while the least important receives the lowest. The weights for the variables in between are spread evenly. To see the highest to lowest rankings for each indicator, please see the Appendix of the report. After sub-index scores were calculated for each of the 100 cities, a simple average is taken of all three to create the final Index scores and rankings.

Through research including an international urban ethnography, Arcadis has categorized the citizen experience of cities into eight archetypes. Borrowing from Corbusier’s model, we briefly summarize the archetype by outlining what it feels like to live, work and travel within these archetypes. They are not intended to be comprehensive descriptions of cities on their own; rather, when combined to create city clusters. The archetypes give us an interesting and further way to explore aspects of the citizen experience in any given city.

By referring back to our research, we identified links between the qualities of the indicators we measured in the Index and in the archetypes. For example, a common citizen experience in Automated cities was the provision of digital public services, such as online property tax payments, was also measured in the People sub-index. By assigning a relevance to each of the indicators from the sub-indices for each archetype, we built a simple statistical model that provides insight and ‘clusters’ similar characteristics of the city experience—thus the city clusters. We further created petal graphs to help with the visualization of the data and through that visualization, one could see cluster patterns emerging and how certain characteristics against the architypes defined what cluster a city sits in.

The data availability limits to the city limits in most cases.

City-level data are used wherever possible, though in some cases only national-level data exists. Where there is no comparable city-level data across countries, the national value is taken, and a national database is used to scale the cities so that they are given a spread around the national average. Health, Energy, Waste Management and Bicycle Infrastructure are a combination of city and national level data or national data.

If you still have more questions or are a member of the press looking to get in touch, please e-mail us at cities@arcadis.com