Printer Friendly

7. Country stakes in climate change negotiations.

For a comparative assessment of country stakes in climate change negotiations, we focus on summary measures of source and impact vulnerability. We simplify by computing aggregative indices from variables whose measures are compatible. The first index is the weighted sum of nonrenewable energy sources (coal, oil, gas, natural bitumen, oil shale), measured in year-equivalents of current domestic energy consumption and weighted by relative C[O.sub.2] intensity (1.0, 0.75, 0.60, 0.75, 0.75, respectively). The second index aggregates renewable energy sources (solar, wind, hydro, geothermal, biogas, sugar ethanol, tallgrass ethanol, jatropha ethanol), again in year-equivalents of domestic energy consumption. Our two sequestration variables (potential for reduced deforestation and carbon storage) are not measured in compatible units, so we leave them separate. For the same reason, we leave the two impact variables (sea level rise, weather damage) separate.

7.1 Overall Correlations

Our indices all have highly skewed distributions, so we compute rank correlations to obtain robust estimates of their relationships. Table 17 presents the results, with starred coefficients denoting significance at the 95% level. We are particular interested in assessing the interaction of two broad dimensions: source vulnerability (related to factors affecting the response to emissions limits--nonrenewables, renewables, employment risk and sequestration options) and impact vulnerability (sea level rise and weather damage). Part of our interest lies in determining whether correlations suggest reinforcing or offsetting effects within the two vulnerability groups. We are also interested in the overall direction of the relationship between the two dimensions, because this has implications for successful negotiation of a global protocol. In general, countries with high impact vulnerability and low source vulnerability should be the strongest supporters of a protocol, while the converse should be true for countries with low impact vulnerability and high source vulnerability. Country postures in the other two cases (both vulnerabilities high or low) would depend on the relative strength of the two effects.

Within the source vulnerability group, Table 17 indicates that four correlations are significant and relatively large, and that all four moderate overall vulnerability via offsetting effects. Countries with plentiful non-renewable energy resources (relative to energy demand) also tend to have significantly less sequestration storage potential. Countries with more sources of renewable energy tend to have slightly less nonrenewable energy. Countries with relatively high employment vulnerability also tend to have less renewable energy resource options (again, relative to energy demand) and less potential for sequestration through reduced deforestation. As a statistical corollary, countries with plentiful renewable energy sources also tend to have high potential for sequestration through reduced deforestation; and those with higher sequestration (storage) opportunities tend to have less potential for sequestration through reduced deforestation.

Within the impact vulnerability group, the two dimensions have a small but significant negative correlation: Countries with large potential impacts from sea level rise tend to have smaller potential damage from weather events.

Across the two dimensions, overall results are mixed. Countries vulnerable to sea level rise tend to have weaker options for renewable energy and sequestration via reduced deforestation, higher employment vulnerability and greater options for sequestration via storage. Countries vulnerable to weather damage also tend to have lower employment vulnerability. However, they also have greater renewable and non-renewable energy options, less potential for storage sequestration and greater potential for sequestration via reduced deforestation.

To summarize, our correlation results suggest that countries with high source vulnerability in some dimensions tend to have lower vulnerability in others. The same is true for impact vulnerability. Between source and impact vulnerability, the evidence is mixed. While these general results are of some interest, none of the observed correlations is very high. By implication, country cases should be our principal focus because they tend to be unique.

7.2 Country Cases

For a composite view, we combine all of our vulnerability measures into a general index that reflects countries' ability and willingness to participate in an international protocol. We construct an index with high values for low source vulnerability and high impact vulnerability, and low values for the converse case. Again, we ensure robust estimates by using ranks rather than numerical values. We also normalize ranks to the range 1-100 to prevent distortion from differences in data structure. (10)

Our analysis considers seven dimensions that affect country orientation. Five dimensions promote a positive orientation toward a protocol: Three reflect source vulnerability (renewable energy resources and both dimensions of sequestration (deforestation reduction and storage)), and two relate to impact vulnerability (sea level rise, weather damage). Ceteris paribus, the higher a country's measure in any of these dimensions, the greater the relative attraction of a global protocol. Two dimensions, both source vulnerability elements, reflect negative factors: nonrenewable energy resources and employment vulnerability. Ceteris paribus, the greater a country's measure in either dimension, the lower the relative attraction of a global protocol.

To develop an overall orientation index, we compute standard ranks (rank 1 for the largest value) for the five positive dimensions and inverse ranks (rank 1 for the smallest value) for the two negative dimensions. We compute one sequestration measure by averaging ranks for deforestation and storage potentials. We normalize all six remaining rank measures to the range 0-100 and select the set of non-island states that have complete measures for all variables. (11) This yields a computation set of 120 countries. We compute orientation indices as weighted averages of dimensional ranks, with total weights constrained to one (Table 18). We test for robustness by computing indices with widely-varying weights to reflect relative emphasis on energy resources (renewable and nonrenewable), impact vulnerability (sea level rise; weather damage); neutrality (equal weights); positive source factors (renewable energy; sequestration potential) and negative source factors (nonrenewable energy, employment vulnerability).

Table 19 reports the results for countries, by World Bank region. For each weighting scheme, we compute index values for all 120 countries and divided the results into three equal groups with high (1), medium (2) and low (3) values. Then we tabulate the results for each country. We assign countries to the High orientation category if at least three of five index values are 1's and the rest are 2's. We assign them to the Low orientation category if at least two of five values are 3's. Intermediate cases are assigned to the Middle category.

Tables 20, 21 and 22 provide useful information about general relationships. Table 20 indicates that correlations among the six dimensional indices are all relatively low, except between renewable energy sources and employment vulnerability and weather damage. Nevertheless, Table 21 shows that alternatively-weighted combinations of these indices are highly correlated, with one exception: high weights for positive source vulnerability factors (index D: renewable energy; sequestration potential) vs. high weights for negative source vulnerability factors (index E: nonrenewable energy, employment vulnerability). Table 22 shows that many countries retain their orientations as the index weights change. Overall, 35 of 120 countries rate 'High' because they display positive orientations across all five weighting schemes. Conversely, 48 countries are consistently 'Low'. The results suggest that the Latin America / Caribbean region has the strongest regional orientation toward a protocol, with 16 countries scoring 'High', 5 'Middle' and 1 'Low'. However, sub-regions are distinctly different. Non-Andean South America and Central America are almost completely in the 'High' Category, while the Andean countries (except Peru and Ecuador, which are 'High') are 'Middle' and the Caribbean is mixed.

In contrast, Eastern Europe / Central Asia (except Georgia) and South Asia are predominantly 'Low' regions: 18 of 21 ECA countries and 2 of 4 SAR countries, including India, are in the 'Low' category. The Middle East / North Africa region is also unbalanced in this direction, with 2 'High', 3 'Middle' and 4 'Low'.

Sub-Saharan Africa and East Asia / Pacific are more evenly balanced. In SubSaharan Africa, 12 states are 'High', 14 'Middle' and 8 'Low'. Within the region, however, sub-regions differ markedly. Among Coastal West African states, 8 are 'High', 1 'Middle' and 2 Low. In contrast, all Central African states are 'Middle' (5) or 'Low' (3). The Sahelian, Eastern and Southern Africa sub-regions exhibit more diversity.

Unfortunately, our evidence also suggests that, among the World Bank's partner countries, total emissions are much greater from the 48 'Low' states than from the 35 'High' states. Overall, 'High' states account for 21.0 % of emissions, while 'Low' states account for 50.1% (Table 23). China and India, both 'Low' states, account for nearly 80% of that group's emissions. However, even among states other than the four greatest emitters (China, Indonesia, Brazil, India), 'Low' states account for the greatest share of emissions. This unfortunate result provides a suggestive indicator of potential difficulty in attempting to extend an emissions-reduction protocol to low- and middle-income countries in the next round.

To summarize, our results suggest that the geographic distribution of protocol orientation is far from random. Among 35 states with 'High' scores, almost two-thirds are in three sub-regions: non-Andean South America (8), Central America (6), and Coastal West Africa (8). States with 'Low' scores also display some concentration, particularly in Eastern Europe and Central Asia. Unfortunately, carbon emissions are also heavily concentrated in World Bank states with 'Low' scores: China (4.9 Gt), India (1.8 Gt) and 46 others (4.2 Gt) (Table 23).

7.3 Two Dimensions of Country Vulnerability

It is also useful to position countries two-dimensionally, according to their relative source and impact vulnerability. We focus on non-island states, and for this exercise we include states with some incomplete data. Using normalized ranks (1--100), we compute source and impact vulnerability indices as means of the available dimensional indices for each country (four for source vulnerability (nonrenewable energy, renewable energy, sequestration, employment); two for impact vulnerability (sea level rise, weather damage)). Then we divide the source and impact vulnerability index values into equal 'Low', 'Medium' and 'High' groups.

Table 24 and Figure 7 present our results for non-island states. Ceteris paribus, 'High' source vulnerability should discourage participation in a global protocol, while 'High' impact vulnerability should encourage it. The most positively-oriented states should therefore have 'Low' source vulnerability and 'High' impact vulnerability. In Table 24, 26 states are in this category. All are low- or middle-income countries, and they are heavily concentrated in three regions: Latin America, Sub-Saharan Africa and Southeast Asia.

Table 25 indicates that these 26 states account for about 11% of tabulated emissions. Table 24 identifies 32 states with the greatest negative orientation ('High' source vulnerability, 'Low' impact vulnerability). These are middle- and high-income countries, concentrated in three regions: Eastern Europe, Central Asia and the Middle East. However, a number of these negatively-oriented countries are major emitters. Table 25 shows that together, they account for about 11% of total emissions (dominated by C[O.sub.2] emissions from the Russian Federation, see Table 3).

Table 25 shows that two groups with the greatest volume of emissions are subject to conflicting pulls in the two dimensions of vulnerability. The countries with 'High' source vulnerability and 'High' impact vulnerability account for about 19% of all emissions. Of particular interest in the 'High'/'High' group are China and India, which together account for 68% of the greenhouse emissions from countries in the lower half of the international income distribution (Table 3). The 'High'/'Medium' group, accounting for 27% of emissions in Table 25, is dominated by the United States and the Ukraine.
COPYRIGHT 2007 The World Bank
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2007 Gale, Cengage Learning. All rights reserved.

 Reader Opinion




Article Details
Printer friendly Cite/link Email Feedback
Author:Buys, Piet; Deichmann, Uwe; Meisner, Craig; That, Thao Ton; Wheeler, David
Publication:Country Stakes In Climate Change Negotiations: Two Dimensions of Vulnerability
Date:Aug 1, 2007
Previous Article:6. Impact vulnerability.
Next Article:8. Summary and conclusions.

Terms of use | Copyright © 2016 Farlex, Inc. | Feedback | For webmasters