Pengfei Ni ：Regression analysis for GUCI indicators

      In Chapter 1, we conducted detailed analysis on the mechanism and elements that shape the competitiveness of a city. Specifically, how significant is each of these elements in determining the competitiveness of a city? Which are more important? Having an understanding of the roles of these elements proves to be the pre-condition for the development and competition strategies of a city. In this Part, regression analysis is conducted on the basis of the URCI indicator tests, using the explanatory indicators (including the level-I indicators and key level-II/III indicators) as independent variables, and GUCI as the dependent variable. In the meantime, cause-and-effect analysis is conducted using GDP per capita as the dependent variable and the explanatory indicators.

Level-I indicators: industry structure being the most important

       The regression analysis on the 7 level-I explanatory indicators shows that industry structure has the biggest influence on GUCI, with a regression coefficient of 0.8363 and a goodness of fit (R²) of 0.8231. Once again, see Table 3.4 for the regression coefficient, goodness of fit and correlation coefficient of the indicators. General regression coefficients show the extents of impact of the indicators. Figure 3.1 shows the impact of the 7 indicators on the competitiveness of the cities in order of their significance: industry structure > hard environment > global connectivity > human resource > soft environment > enterprise > living environment. Hard environment and global connectivity are important indicators following industry structure. Hard environment includes a number of elements of technological innovation. In today’s world of economic globalization, global connectivity is as important as hard environment to a city. This indicates the importance for cities, as major players in global competition, taking the path of internationalization and building international metropolises. It should be noted that human resource is critical to the competitiveness of a city. However, in this study, the human resource includes a considerable proportion of labor status indicators. As a result, human resource does not seem as important as we previously thought.  

Level-II indicators: enterprise connectivity being the most important

       Among the 152 indicators, 40 are level-II indicators. See the attached figure for the regression coefficients, goodness of fit (R²) and correlation coefficients obtained through the regression analysis. Particularly, see Figure 3.2 for the top 10 indicators, among which, enterprise connectivity ranks No.1. Enterprise connectivity > hi-tech industry > market size > air transportation > information connectivity > manufacturing industry > technological innovation > service sector > financial market > tax burden. Enterprise connectivity describes the ability of individual enterprises to control the global economy. It is a distinct indicator of a city’s competitiveness in the context of economic globalization. Indicators such as hi-tech industry, manufacturing industry and service sector fall into the scope of industry structure, while other indicators fall into the scope of global connectivity. It further proves the contribution of industry structure and global connectivity to the competitiveness of the cities.

       Figure 3.3 shows the 10 elements with the least impact (the largest reverse impact). While labor cost, infrastructures, social security, natural environment, and catering service have the largest negative impact on the competitiveness of cities, enterprise performance, water transportation, status of labor market, culture and entertainment, and land transportation have the least contribution to the competitiveness of the cities. All of the above 10 indicators are basic elements in production and living environment. They are fundamental to the citizens and urban development of the cities. However, these basic elements are less important than higher level indicators in international competition of the cities (the 150 sample cities) as long as there is no significant bottleneck or gaps.

 Level-III indicators: capital market being the most important

       Among the 152 indicators, 105 are level-II indicators. See the attached figure for the regression coefficients, goodness of fit (R²) and correlation coefficients obtained through the regression analysis. Specifically, the 10 most influential indicators include: capital market, number of transnational company headquarters, international patent applications,  number of transnational company regional headquarters, the number of international hotel groups, feedback from government portals, airport handling capacity, the number of renowned universities, the number of transnational business service providers, and the number of transnational manufacturers (See Figure 3.4). These indicators mainly describe the elements of financial capital, technological innovation, economic control and industrial layers, which are critical to the competitiveness of individual cities. 

       Notably, the 10 least influential or most negatively influential indicators are (sequentially) hotel room price, employees’ income, office rental, electric power price, living cost, per capital land area, criminal rate, per capita fresh water ownership, and weather environment (see Figure 3.5). In the regression analysis, the price of restaurant, employees’ earning, office rental, electricity price and living cost have been treated reversely in regression analysis. Therefore, the higher the indicators are, the more competitive a city is. Land and water are the most fundamental elements for the survival and development of human beings and cities.  However, our analysis shows that it is not ’the more, the better.’

——From Global Urban Competitiveness Report(2007-2008)，Pengfei  Ni with Peter Karl Kresl