Appendix B: Global Risks Perception Survey and Methodology 2016
Definitions and Changes
The Global Risks Report 2017 is based on an improved methodology; however the results are therefore largely comparable. The Report adopts the following definitions of global risk and trend:
Global risk: an uncertain event or condition that, if it occurs, can cause significant negative impact for several countries or industries within the next 10 years.
Trend: a “trend” is defined as a long-term pattern that is currently evolving and that could contribute to amplifying global risks and/or altering the relationship between them.
The list of risks and trends assessed in the Global Risks Perception Surveys (GRPS) remains unchanged with the exception of the addition of the global risk “Failure of regional or global governance” (defined as the inability of regional or global institutions to resolve issues of economic, geopolitical or environmental importance). As a result, the Report covers 30 global risks this year.
Some of the names of the trends were modified to better reflect long-term pattern characteristic of trends (for instance, the trend “rise of chronic diseases” was edited to “rising chronic diseases”). The definitions were mainly unchanged.
This year’s GRPS included an entire section on emerging technologies. After consultations with experts, 12 critical emerging technologies were identified; selected findings are described in Part 3 of the Report.
The following section describes the survey and methodology in greater detail.
The Global Risks Perceptions Survey
The Global Risks Perception Survey (GRPS), discussed in Part 1, is the main instrument used to assess global risks in this Report. The survey was conducted between early September and mid-October 2016 (from 07 September to 15 October 2016) among the World Economic Forum’s multistakeholder communities of leaders from business, government, academia and non-governmental and international organizations as well as members of the Institute of Risk Management.
This year, the GRPS is a key instrument used as supporting data for the elaboration of the Report. For this year’s Report, the GRPS went through an important review to ensure the quality of the results. This process was performed in collaboration with the Global Risks Perception Survey Review Group on The Global Risks Report 2017, a group composed of experts in survey methodology and risks perception (see Acknowledgements section).
Among the most significant improvements are the changes to the scales of the Global Risks Landscape. Indeed, the impact scale has changed this year from an abstract 1–7 scale, subject to interpretation and thus bias, to a more substantive and meaningful scale of impact measurement (i.e. minimal, minor, moderate, severe, catastrophic). On the likelihood scale, the scale of 1–7 was kept but a particular probability was attached to each number in order to ensure that all respondents had the same understanding of the likelihood being considered. Throughout the survey, the questions were modified and the phrasing was refined to reduce any ambiguity.
Raw responses were cleaned in order to improve overall data quality and completeness. Surveys with a completion rate below 50% were dropped, reducing the number of available responses from 989 to 745. The respondents did not provide sufficient information about their gender or the sector in which they work in 92 and 119 cases, respectively. Similarly, 93 respondents did not indicate the country in which they are based.
Figure B.1 presents the profile of the 745 survey respondents remaining in the sample. To capture the voice of youth, the survey also targeted the World Economic Forum’s community of Global Shapers.1 Respondents under 30 accounted for about one-fifth of total respondents.
The Global Risks Landscape 2017 (Figure 3)
Respondents were asked to assess the likelihood and global impact of each of the 30 risks. For each risk, they were asked, “What is the likelihood of [the risk] occurring globally within the next 10 years?” and “What is the negative impact for several countries or industries within the next 10 years?” For the first question, the possible answers ranged from 1 (“extremely unlikely” with an associated probability of occurrence lower than 5%) to 7 (“extremely likely” with an associated probability of occurrence greater than 95%). For the question on impact, respondents could select one of five choices: “minimal”, “minor”, “moderate”, “severe”, or “catastrophic”. These five alternatives were turned into a 1–5 scale (1 = minimal, 5 = catastrophic). It is worth noting that, as a consequence of the scale modification, the impact results cannot be compared with those of previous years.
Respondents could also choose “No Opinion” if they felt unable to provide an informed answer. Respondents could also leave the question completely blank. For each risk, partial responses – those assessing only the likelihood of occurrence or only its impact – were dropped. A simple average for both likelihood and impact for each of the 30 global risks was calculated on this basis.
Formally, for any given risk i, its likelihood and impact, denoted respectively likelihoodi and impacti, are:
where Ni is the number of respondents for risk i, and likelihoodi,n and impacti,n are, respectively, the likelihood and impact assigned by respondent n to risk i. The likelihood is measured on a scale of 1–7 and the impact on a scale of 1–5. Ni is the number of respondents for risk i who assessed both the likelihood and impact of that specific risk (the answers of respondents who left one of the two questions blank were not taken into account).
To draw the Global Risks Interconnections Map (Figure 4), survey respondents were asked to answer the following question: “Global risks are not isolated and it is important to assess their interconnections. In your view, which are the most strongly connected global risks? Please select three to six pairs of global risks.”
Similarly, for the Risks-Trends Interconnections Map 2017 (Figure 1), respondents had to identify up to three trends that they consider important in shaping the global agenda in the next 10 years and the three risks that are driven by each of those trends. For completeness, the two questions read “Which are the three most important trends that will shape global development in the next 10 years?” and “For each of the three trends identified in the previous question, select up to three global risks that are most strongly driven by these trends.” The information thereby obtained was used to construct the Risks-Trend Interconnections Map 2017.
In both cases, a tally was made of the number of times each pair was cited. This value was then divided by the count of the most frequently cited pair. As a final step, the square root of this ratio was taken to dampen the long-tail effect (i.e. a few very strong links, and many weak ones) and to make the differences more apparent across the weakest connections. Out of the 406 possible pairs of risks, 167 or 41% were not cited. Similarly, out of the possible 377 trend-risk combinations, 33 or 9% were not cited. Formally, the intensity of the interconnection between risks i and j (or between trend i and risk j), denoted interconnectionij, corresponds to:
where N is the number of respondents.
Variable pairij,n is 1 when respondent n selected the pair of risks i and j as part of his/her selection. Otherwise, it is 0. The value of the interconnection determines the thickness of each connecting line in the graph, with the most frequently cited pair having the thickest line.
In the Global Risks Landscape and Risks-Trends Interconnections Maps, the size of each risk is scaled according to the degree of weight of that node in the system. Moreover, in the Risks-Trends Interconnections Map, the size of the trend represents the perception of its importance in shaping global development (answer to the first part of the question on trend, as explained above); the biggest trend is the one considered to be the most important in shaping global development.
The placement of the nodes in the Global Risks-Trends Interconnections Map was computed using ForceAtlas2, a force-directed network layout algorithm implemented in Gephi software, which minimizes edge lengths and edge crossings by running a physical particle simulation.2
The Emerging Technologies Matrix (Figure 3.1.1)
For the first time this year, the GRPS included questions on emerging technologies. The first question asked in this section was on the consequences of emerging technologies. For each of the 12 emerging technologies identifies, respondents had to answer the following questions: “How likely is this emerging technology to bring significant benefits within the next 10 years?” and “How likely is this emerging technology to bring severe negative consequences within the next 10 years?” and finally “How confident are you about your responses for this emerging technology?” For the first two questions, respondents could answer from 1 (extremely unlikely) to 7 (extremely likely). Similar to the likelihood questions used to build the Global Risks Landscape 2017, probabilities were attached to each selected risk. For the question on the level of confidence, respondents could select an answer ranging from 1 (extremely low confidence) to 7 (extremely confident).
Here again, respondents were given the option of choosing “No Opinion” if they felt unable to provide an informed answer. Respondents could also leave the question completely blank. A simple average of responses to the benefits, negative consequences, and level of confidence questions was calculated. Formally, for any given emerging technology i, its benefits and negative consequences, denoted respectively benefitsi and neg.consequencesi, are:
where Ni is the number of respondents for emerging technology i, and benefitsi,n and neg.consequencesi,n are, respectively, the benefits and negative consequences assigned by respondent n to the emerging technology i and measured on a scale from 1 to 7. Ni is the number of respondents for the emerging technology i who assessed both the benefits and the negative consequences of that emerging technology (the answers of respondents who left one of the two questions blank were not taken into account).
Other Emerging Technologies Questions (Figure 3.1.3)
After the questions on the consequences of emerging technology, the respondents had to select the three emerging technologies that need better governance. The exact question is: “Please select the three emerging technologies where you believe better governance is most needed. By ‘governance’ we mean the rules, norms, standards and/or institutions that allow stakeholders to take effective decisions that maximize the benefits and minimize the negative consequences of a technology.” The computation for each emerging technology i is:
where N is the number of respondents to the survey, and variable governancei,n is 1 when respondent n selected the pair of risks i and j as part of his/her selection. Otherwise, it is 0. As a result, governancei (the score) measures the percentage of respondents selecting the emerging technology i.
The respondents had to then answer a question about which emerging technologies exacerbate each of the five categories of global risks. The question reads: “For each question, please select the three emerging technologies that you believe will most significantly exacerbate global risks within the stated risk category. By ‘exacerbate’ we mean increase the likelihood and/or impact of those risks.” For each risk category, the results are computed as:
where N is the number of respondents to the survey and, for emerging technology i for the risk category a (economic risks, environmental risks, geopolitical risks, societal risks, or technological risks), variable exacerbatei,a,n is 1 when respondent n selected the pair of risks i and j as part of his/her selection. Otherwise, it is 0. As a result, exacerbatei,a is the score assigned to emerging technology i for risk category a and measured as a percentage of respondents selecting this emerging technology.
Jacomy, M., T. Venturini, S. Heymann, and M. Bastian. 2014. “ForceAtlas2: A Continuous Graph Layout Algorithm for Handy Network Visualization Designed for the Gephi Software.” PLoS ONE 9(6): e98679. doi:10.1371/journal.pone.0098679