The purpose of this research paper is to identify the determinants of regional competitiveness, in order to increase policy effectiveness, as well as to anticipate implications to governments, as the paradigm of regional development policy has shifted from ‘balance’to ‘efficiency’. Regional competitiveness is very important, because it eventually determines national competitiveness. This research consists of three main areas. First, the most competitive regions(SiGun-Gu) have been selected and the characteristics are analyzed. Second, based on Porter's diamond model, the determinants of regional competitiveness have been surveyed and analyzed. The survey and analysis set the priorities for regional policy to enhance  regional competitiveness from these significant factors. In the third area, current regional policy was screened and the implications of this study on current policies are suggested. There are two unique aspects in this research, as compared to the previous literature. One is that this study is not only on measurable variables, such as population, income, innovation etc, but also some immeasurable variables such as location advantage or regional policy. The other is that the Si-Gun-Gus are classified into four groups and analyzed, while most of the previous research has focused on the disparity between Si-Gun-Gus. As a proxy for regional competitiveness, a weighted average value was used for the quantitative analysis including the following factors : population growth rate, employment growth rate, value added in manufacturing industry per capita, the growth rate of value added in manufacturing, and value added in service industry per capita, the growth rate of value added in service. Based on this measurement, the 232 Si-Gun-Gus are classified into three groups: high, med, low. In addition, the 232 Si-Gun-Gus are classified into 4 groups to their size: metro city, hub city, medium-sized city, and rural area. Two typical Si-Gun-Gus in each group have been selected as benchmarks, as shown in the following table. In general, social infrastructure is more developed in metro cities, in comparison to other types of cities, while hub cities are better qualified in terms of the level of education. In industrial factors, hub cities are more specialized in manufacturing industries, while metro cities are more diversified especially in knowledge based industries, 

due to larger population and better industrial environments. Qualitative surveys have been conducted on the experts on regional policies such as entrepreneurs, researchers, government officials, and even professors to examine the determinants of regional competitiveness. The questionnaires are measured on a 5 point Likert scale for dependent variable (which is regional competitiveness) and 40 factors in 4 sectors from Porter's diamond model, as shown in the following table. Since the number of respondents is 760 but the number of regressors is 40, factor analysis was conducted to reduce the number of regressors. As a result, 12 new explanatory variables were created as regressors for regional competitiveness. In this discrete setup, rank-ordered logit regression was used to identify for significant determinants of regional competitiveness. The empirical results show that ‘innovative ability’ in the Enterprise sector, ‘related industries’in the Industry sector, ‘accessibility and amenities’in the Locational advantage sector, and ‘collaboration among enterprises, academies, and research institutes’ 

are crucial and significant factors in regional competitiveness. More precisely, ‘innovative ability’and ‘collaboration’could lead to improved competitiveness in hub cities. However, ‘accessibility’is more important in medium size cities. From both analyses on the determinants and characteristics for regional competitiveness, several implications in regional policy can be derived as follows: long term regional policies are supposed to enhance innovative capability in governments. In other words, local governments and the central government need to play their own roles in contributing to regional competitiveness. The local governments’independent policies to strengthen and address weaknesses through the determinants and characteristics analysis of regional competitiveness are needed. These would be available through additional block grants, which result from the transfer of authority from the central government to local governments. On the other hand, the central government should focus on larger scale policies such as developing related industries or building collaborative networks.