ظهور و تکامل خوشه های همگرایی منطقه ای در بازار انرژی چین

Employing the new regression tests for Convergence, Club Convergence and Clustering proposed by Phillips and Sul (2007), this paper models and analyses the behaviour of China's energy sectors. Energy market ‘convergence clusters’ are identified using new price data, and their regional spatial distributions are mapped for four major fuel types: coal, gasoline, diesel and electricity. The findings are as follows: i) as yet, there are no fully integrated national energy markets in China, as more than one convergence cluster is identified for all four fuels; ii) some regional energy markets can be regarded as ‘quite mature,’ as evidenced by the existence of some highly concentrated convergence clusters connected geographically; iii) some regional markets remain in a ‘state of transition,’ as witnessed by convergence clusters that are scattered geographically and that are growing in membership; iv) it seems that there is more regional-based integration for coal and electricity than for gasoline and diesel, as more convergent clusters were identified for coal and electricity than for gasoline and diesel; and v) overall, China still appears to be in the process of energy market integration, as demonstrated by the number and evolution of convergence clusters over time.

Although China has been praised for promoting market competition among the state-owned, collective and private sectors (Qian and Xu, 1993), assessments of its performance have been mixed. Many authors who have studied China's market economy have concluded that some of China's markets are integrated. Rozelle et al. (1997), for example, investigated market integration in China's rural sectors in the late 1990s and found evidence in favour of such a transition. Zhou et al., 2000, Park et al., 2002, Huang and Rozelle, 2006 and Awokuse, 2007 investigated market integration in China's agricultural commodity markets, and all found that the grain markets were well integrated. Fan and Wei (2006) conducted tests for price convergence amongst 96 commodities and concluded that prices had converged to the ‘Law of One Price’ in China for an overwhelming majority of goods and services. However, some authors argue that changing patterns of provincial economic structure suggest that China's markets have become less rather than more nationally integrated during much of the reform period. Young (2000) argues that China's economic reform has actually led to the fragmentation of domestic markets. Poncet (2005) measured China's domestic market disintegration and identified its determinants and concluded that China is a ‘fragmented economy’. Therefore, whether China's reforms have led to a more integrated internal market remains inconclusive ( Lan and Sylwester, 2010). However, perhaps this is not surprising given that substantial differences in regional, location-specific advantages and central government preferential policy have resulted in economic development disintegration in China ( Demurger et al., 2002). Based on the above discussion, it might be reasonable to expect that China is not yet a completely integrated market economy and that it is still essentially in a ‘state of transition’. In fact, empirical studies have suggested that regional markets are highly fragmented due to interregional protection in China (Li et al., 2003). The final outcome may be the formation of powerful, geographically disconnected (or partially connected), regionally based growth zones. However, a rejection of convergence for the country as a whole does not imply there is no evidence of convergence within regional subgroups. Examples include the possible existence of ‘convergence clusters’ around separate points of equilibria or steady state growth paths and cases in which there may be both convergence clusters and divergent members in the full panel of regions or sectors. Demurger et al. (2002) demonstrated the relative contributions of location and preferential policies in China's regional development; as expected, these convergence clubs will be geographically linked, i.e., coastal, east–west, north–south, or natural resource based. If local equilibria or club convergence clusters exist, then it is of considerable interest as to where they are, what sectors they are in and whether they have/are evolving over time. In separate work, Ma et al., 2009b and Ma and Oxley, 2010 have investigated the convergence of major energy fuel price series using traditional unit root and panel cointegration methods. They found that energy price series are not convergent as a whole and regional energy price series display a differing convergent pattern; this finding implies that there might be some regional energy markets in China. Given the above discussion of China's market performance, it seems important to investigate whether there is any empirical evidence for convergence clusters for specific commodities and how the clusters evolve over time. Therefore, several questions and issues are worth considering. Firstly, testing for the existence of convergence clusters and determining how regional markets have evolved in China can help us understand the process of China's market economic development and may provide useful insights for policy makers and academics. Secondly, given the concerns and doubts about China's economic reforms (Demurger et al., 2002, Poncet, 2005 and Young, 2000), such tests can add to the evidence regarding China's market development and integration. Thirdly, although some studies have investigated China's market integration, most have focused on agricultural commodity markets (Awokuse, 2007, Huang and Rozelle, 2006, Park et al., 2002 and Zhou et al., 2000) and aggregate commodity markets (Fan and Wei, 2006 and Poncet, 2005). Thus, few regional commodity markets have been investigated and identified, particularly the potential convergence clubs that might exist in regional energy markets. Finally, and perhaps most importantly, given the substantial differences in regional, location-specific advantages, central government preferential policy and economic reforms, it may be advantageous to investigate regional market formation and evolution rather than to simply determine whether China has a market economy as a whole. A limited number of studies have identified economic growth and the development of convergence clubs in China. For example, Maasoumi and Wang (2008) investigated regional economic reforms, economic growth economic development and convergence using a metric entropy-based measure. Their results show that there exist many small economic development convergence clubs in both the pre- and post-reform periods in China. Weeks and Yao (2003) discovered system-wide income divergence during the reform period (1978–1997) because the coastal provinces do not share a common initial technology progress rate with the interior provinces. However, these studies do not relate to specific commodity market convergence cluster tests; thus far, there have been few econometric tests of ‘convergence clustering’ for important, specific commodity markets in China. Therefore, the main objective of this paper is to identify the existence of regional convergence clusters and critically investigate the transitional dynamics of their formation using the new testing procedures of Phillips and Sul (2007) as they apply to China's four major fuel price series (coal, electricity, gasoline and diesel). Energy price reforms have perhaps had the most significant impact on the income transfer between the interior and coastal areas given that the interior provinces are the main suppliers of raw industrial materials. The paper is organised as follows. Section 2 introduces China's major energy reforms to provide the historical background necessary to enable sensible interpretation of the results of the tests. Although econometrically powerful, the testing approach of Phillips and Sul (2007) is atheoretical because it requires no prior, specific inputs or assumptions regarding potential regional convergence club locations or associations. Section 3 outlines the testing approach and explained how it is applied to the data that are discussed in this section. Section 4 presents the empirical results and discussion. The final section presents some conclusions and possible policy implications.

This paper employs the new testing approach of Phillips and Sul (2007) to model and analyse China's energy sector price transition behaviour in the presence of common movement characteristics. Price convergence clusters are identified, presented and analysed. The spatial distributions of convergence clusters are mapped and discussed based on four sub-periods for each of the four fuels. The paper presented some evidence that a number of regional convergence clusters have appeared for the major energy markets in China as reforms have been gradually introduced. The following conclusions can be drawn. Firstly, there is no statistical evidence in favour of a single nationally integrated energy market in China, as there is more than one convergence cluster for all four fuels, particularly during the latest sub-period. Secondly, some regional energy markets have become more mature as some convergence clusters have become more highly concentrated and geographically connected. Thirdly, some regional markets still seem to be evolving and developing as the convergence clusters identified remain geographically scattered. Fourthly, it is consistent with the characteristics of the various types of energies tested, more regional integrated markets exist for coal and electricity than for gasoline and diesel. Fifthly, the dynamics of convergent clusters are different for each energy type. Thus, convergence clusters for the electricity market are the most dynamic, while those for the gasoline market are the least dynamic. Finally, China is still in the process of energy market integration as evidenced by the continuing evolution of the number, composition and coverage of the various clusters; however, China's petroleum market is fairly well integrated even though more than one convergent cluster has been identified. According to the conclusions presented above, some policy implications can also be proposed: Firstly, as more regional energy markets are identified, China's national energy policies and regional development strategies will likely remain disconnected. Therefore, creating a single, market-oriented economy continues to be a great challenge for China. Secondly, coal production is a regional activity in China that creates regional coal markets. Thus, the process of market integration for the coal sector is likely to be slower and more difficult than, for example, the petroleum sector. Such a situation is also likely to significantly affect the dynamic evolution of the national electricity market due to China's dependence on coal to produce electricity. Therefore, the reforms within the coal and electricity sectors will play a crucial role in the development of a market-oriented energy economy in China. Thirdly, coal price adjustment has led to some changes in electricity prices; however, the changes do not appear to be completely transferred to electricity prices because the dynamic of convergence clusters for electricity market lagged behind those of the coal market (compare the numbers of convergence clusters for coal and electricity in the last row of Table 1 and Table 2). In fact, electricity price adjustments not only lagged behind but were also less in magnitude than that of coal in China (Hang and Tu, 2007, Lam, 2004 and Wang, 2007). Therefore, electricity price reforms might also significantly influence coal market price formation because coal and electricity prices are highly correlated. Thus, electricity price reform might be another key factor that affects the formation of an integrated national energy market. Finally, because a market economy can play an important role in the efficient allocation and utilisation of resources and given the conclusions above, it is not surprising that China's energy efficiency is much lower than that of the developed economies. Therefore, to improve energy efficiency, reduce energy intensity and reduce greenhouse gas emissions, China must reform its energy pricing mechanism to integrate regional markets into a nationwide energy market. These reforms are particularly necessary for the coal and electricity markets.