Results of Efficiency Frontier

The efficiency frontier consists of the most efficient regions for each particular year.

Regions on the frontier are assigned an efficiency score of one. Regions with scores approximating to one are those who are closer to the frontier. Compositions of efficiency frontiers without and with environmental factors during 1997 to 2001 are shown in Table 4.3.

Generally speaking, about one-sixth of the regions in the sample are on the frontier at least once for the time period from 1997 to 2001 when environmental factors are not considered. With environmental factors, about one-third of the regions are on the frontier.

With or without environmental factors, Shanghai (09), Hunan (18), Guangdong (19), and Tibet (26) are on the frontier every year. Fujian (13) is on the frontier in some years without environmental factors and is on the frontier for every year with environmental factors.

Heilongjiang (08), Jiangsu (10), and Hainan (21) behave most efficiently after taking the environmental factors into account. Two municipalities, Beijing (01) and Tianjin (02), are on the frontier for some years with environmental factors. Most of these best performers are in the highly developing areas of China.

Composition of the efficiency frontier sorted by areas of China is in Table 4.4. The east coastal regions are on average in a better position no matter with or without environmental factors. Taking into account environmental factors makes the number of regions on the frontier increase. The total amount of regions gained on the frontier mainly results from the east area. The efficiency frontier derived from technical efficiency is a relative concept.

We cannot conclude that those east coastal regions in the frontier have absolutely good environmental conditions. However, these provinces perform better than their inland peers when both economic and environmental factors are concerned.

Table 4.3 Technical Efficiency Score of Region for Variable Returns to Scale 08 Heilongjiang C 0.851 0.962 0.797 1.000 0.823 1.000 0.857 1.000 0.871 1.000 09 Shanghai E 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 22 Chongqing W 0.411 0.428 0.417 0.565 0.416 0.562 0.420 0.423 0.430 0.430 23 Sichuan W 0.834 0.855 0.876 0.890 0.855 0.876 0.828 0.839 0.845 0.849 31 Xinjiang W 0.777 0.777 0.789 0.801 0.752 0.767 0.787 0.787 0.775 0.775 Number of regions on the frontier Numbers of regions on the frontier 7 9 7 11 5 9 4 9 4 9

Note:

(1) ^a Technical efficiency of the region during the period 1997-2001 without environmental factors.

(2) ^b Technical efficiency of the region during the period 1997-2001 with environmental factors.

(3) E is the abbreviation for east area, C is the abbreviation for central area, and W is the abbreviation for west area.

Table 4.4 Composition of the Efficiency Frontier for Variable Returns to Scale

Without environmental factors With environmental factors

1997 1998 1999 2000 2001 1997 1998 1999 2000 2001

East 5 4 3 2 2 7 7 6 6 6

Area of China Central 1 1 1 1 1 1 2 2 2 2

West 1 2 1 1 1 1 2 1 1 1

Total 7 7 5 4 4 9 11 9 9 9

Note: The numbers in this table are the number of regions on the frontier

4.5 Results of Productivity Change

In the above analysis, the efficiency frontier for each year is constructed from the efficient regions of the given year. This is a kind of static analysis that disregards movements of the frontier, and regions on the frontier have the same efficiency score of one.

Geometric means of the Malmquist productivity change summary indices and the components of growth for each sample region are listed below.

Malmquist indices comparison among regions without/with environmental factors is displayed in Table 4.5. On the left side of Table 4.5, the Malmquist indices and its components without environmental factors are listed. The average Malmquist index is 0.955, with 4 regions’ indices exceeding unity, implying that they have positive production growth.

The east regions generally perform better than inland ones. The sources of productivity growth for those east regions are technical change rather than efficiency change. Most west regions and some central ones lie in the rear of the list. This result is consistent with the developing disparity in China (World Bank 2001) whereby the east areas have better economic conditions.

After incorporating the case of the three undesirable and costly emissions as inputs, regional performance rankings on average do not change: The Malmquist indices and its components with environmental factors are listed on the right side of Table 4.5. The average

Malmquist index is 0.957, with 6 regions showing a positive productivity growth. The overall rankings of Malmquist indices change slightly with and without environmental factors.

Productivities of three big cities, Shanghai (09), Beijing (01) and Tianjin (02), improve for a large extent when environmental factors are considered. The regions which improve their rank of position for more than 5 positions are: Hainan (21) in the east; Heilongjiang (8) in the central. The regions which regress more than 5 places are: Jiangsu (10) and Zhejiang (11) in the east; Jiangxi (14) and Hubei (17) in the central.

In order to examine whether an association exits between the two rank lists without/with environmental factors, the test of Spearman rank correlation coefficient is used for this purpose. It is a nonparametric rank correlation procedure for making inferences about the association between two rank series. The Spearman correlation coefficient for the Malmquist indices is 0.9108 with 1% significant level which strongly reject null hypothesis that there is no association between the two rank lists. Therefore, it can be generally concluded that those regions with higher productivity while GDP is solely concerned still rank superior when both GDP as well as environmental factors are considered.

In Lovell et al. (1995) on OECD countries, the inclusion of two environmental indicators did change the ranking, reflecting that the environment is a decisive variable when assessing a nation’s relative performance. However, this is not to say that environmental factors are not of importance to Chinese regional comparison, because of this unchanged productivity ranking. It is rather a warning of the extreme developing disparity in China, whereby the non-coastal areas are frail in economic growth as well as in environmental protection. We call this phenomenon the ‘double deterioration’ of regional development in China.

Table 4.5 Decomposition of the Malmquist Index without/with Environmental Factors

Without Environmental factors With Environmental factors

ID Region Area

(1) All Malmquist index averages are geometric means.

(2) E is the abbreviation for east area, C is the abbreviation for central area, and W is the abbreviation for west area.

(3) The Spearman rank correlation coefficient for the Malmquist indices is 0.9108 with p-value less than 0.01.

The double deterioration in China can also be clearly observed through the regional indices changes without/with environmental factors summarized in Figure 4.3. Figure 4.3 presents the decomposition of the Malmquist index by area. There appears to be an obvious difference between the east and the inland-central-west areas: The productivity growth (MALM) of the east area dominates those of the central and west areas without/with environmental factors. With respect to technical changes (TECHCH), the east area still leads the central and west areas without/with environmental factors. For efficiency changes (EFFCH) without environmental factors, the east area performs worse than the central and west ones. However, this gap gets narrowed after taking environmental factors.

0.860 0.880 0.900 0.920 0.940 0.960 0.980 1.000 1.020 1.040

MALM EFFCH TECHCH MALM EFFCH TECHCH

Without environmental factors With environmental factors

East Central West

Figure 4.3 Decomposition of Malmquist Index without/with Environmental Factors by Area

One may wonder whether or not the industry composition creates the disparities since the pollution emitted is mainly from the secondary industry. Recall Figure 4.2, which presents the industry composition of the three areas in section 4.2: The percentage of

secondary industry in the east area is higher than that of the other two areas. A postulate that an area with a higher percentage of secondary industry performs even worse under environmental concerns is definitely not supported. A possible explanation is that the secondary industry in the inland area is pollution-intensive, such as basic metals and chemicals. Their production equipment and environmental control skills are less developed, hence inducing higher pollution. ‘Double deterioration’ is a consequence of inefficient funds to replace dirty equipment and fuel for the poor regions.

4.6 Sub-Conclusions

Two decades of rapid economic growth have brought about a steady deterioration to the environment in China. Air pollution alone contributes to the premature death of more than a quarter of a million people each year (World Bank, 1997). With the threat of Asian Brown Clouds, this problem is starting to prompt global attention. In this chapter we have provided an evaluation of the performance of those regions responsible for the conduct of economic development and environmental problems in China.

The empirical results can be summarized as follows: First, the fast developing east coastal regions experience comparatively higher technical efficiency and productivity growth than the other inland regions when GDP is solely considered as a region’s output. Second, in static analysis, taking into account environmental factors makes the number of regions on the frontier increase. The total amount of regions gained on the frontier mainly results from the progress of east area. Third, in dynamic analysis, the ranking lists without/with environmental factors change just slightly. This result is statistically significant which provide evidence that these two rank series without/with environmental factors are highly related. The possible interpretation for this phenomenon is that for those regions with inferior productivity suffer from costly environmental problems at the same time. In this

study, it is called as a ‘double deterioration’ in China. Fourth, in the comparison of the Malmquist index and its components, the east area performs better than the inland central and west ones after the adjustment adding into environmental factors. The above phenomenon should be attributed to highly-polluting production processes rather than the industrial composition.

Receiving $45 billion in 1998, China was the largest FDI (Foreign Direct Investment) host country among the developing Asian economies (United Nations, 1999). However, per capita FDI in the west area is only eight percent of that in the east (Hu, 2001). Traditional rules, such as ‘economy first, environment later’ or ‘the coastal first, the inland later,’ still dominate the national development policy. Furthermore, China open up for all industries without discrimination after it entered the WTO in 2001. People in China, especially in the areas with lower income, may welcome dirtier industries so as to increase their income.

China hence faces a dilemma of economic growth versus environmental protection.

Our empirical findings are consistent with EKC theory: while the poorer inland areas are on the increasing stage of per output pollution, the richer east is on the decreasing stage of per output pollution. Better environmental performance has been accompanied with economic achievement for the fast-developing area. On the other side, double deterioration of the inland area is indeed a warning for China to pursue balanced regional development.

The inland regions may produce and mine using a lower grade of equipment that is highly polluting, and they still cannot afford better equipment to treat the pollutants. According to EKC theory, with persistent economic growth, the environment of the inland China will sooner or later improve. However, before this turning point occurs, they are now suffering from a double deterioration of economic performance and environment.

The following principles may serve as some inspirations to speed up the development of the inland China: The first is to diminish transportation expenses in these areas. Most west regions are relatively disadvantageous in not only having a longer distance to market, but also

higher transportation costs, which are also obstacles to import the latest pollution abatement technologies and information. The second is to ask for domestic and international assistance in financing, local environmental policy reforms, and education. In the long term, growth without environmental protection could lead the industry to be less competitive under pressure from a world that needs to adhere to environmental protection. Our warning of a

‘double deterioration’ may be beneficial in promoting sustainable development of China’s economy as well as that of the global village.