The preliminary feasibility study (PFS) was adopted in 1999 to enhance the objectiveness and transparency of government-run projects under the agenda of fiscal reform. It has received recognition from the World Bank, International Monetary Bank (IMF) and other prominent global organizations for its contribution to the efficient and transparent use of budgets. However, concerns have emerged about certain PFS-related issues attributed to the synchronization of PFS and budget allocation, and due to the growing public attention over PFS results. One notable issue pertains to the adequacy of estimations for land compensation, which accounts for a large share of the cost estimations for analyses on economic value. The competent authorities argue that the costs for land compensation―estimated through PFS―are unrealistically low and thus hinder project execution. The problem may lie within the PFS analysis itself, but it may also be because the gains from development are reflected in the real compensation costs.

As such, this study focused on projects that were subject to PFS in 1999-2012 with available information on project dates and location. It analyzed the movements in land prices according to the project dates to examine whether government-run projects reflect the gains from land development in the compensation and if so, at which point they are reflected. The results were used to draw on the policy implications for land compensation estimations for future government projects. Time-series data was established on the prices of project sites or nearby land (per lot) for 14 government-funded projects from 1996 to 2014. Then, in a panel analysis using the difference-in-difference model, this study showed that the prices of both project sites and nearby land began to surge from a year prior to the initiation of PSF―during which time the competent authorities design a project and prepare for PFS―and continued throughout the PFS, basic design and implementation planning phases. Additionally, the prices of project sites exhibited a higher increase than that of nearby land until the basic design phase. The results indicate that the increase is driven by the fact that information on project design and development planning are made public and expectations are generated over the development gains as the project undergoes PFS. This, in turn, pushes up land prices and, although it can vary from project to project, the development gains during the basic design and implementation phase are reflected. The findings imply that PFS-estimated compensation costs are lower than real compensation costs not because of errors in the PFS estimations, but rather because of the time difference between cost estimation and real compensation, and reflection of the development gains during this period. Furthermore, a separate analysis of ongoing projects that passed PFS and those that did not revealed that the land prices for the project sites of the latter significantly increased far before the project was put forward but declined with the initiation of PFS, recording a lower growth than that of nearby land. However, once a project receives a budget and starts to roll, this pattern is reversed. Given that PFS and budget allocation have become increasingly synchronized, the market’s instant response to PFS results is a sign of a troubling reality.

These findings indicate that to smoothly operate government-funded projects, it is necessary to foster prompt management of the entire process, from establishing the basic plan―including the release of information on project planning, compensation negotiations and project approval―to conducting PFS and developing the basic design. Above all, PFS should ideally take no more than 4-6 months. But, due to the lack of significant proposal contents such as location and the frequent changes in project plans, it can sometimes take 1-2 years before PFS results become available. To avoid this and conduct a prompt and timely study, the system should to be managed to allow only fully completed proposals to be selected as PFS-subject projects by the PFS committee. Also, once submitted, in principle, no changes to the project plan should be allowed. Compensation-related provisions in the PFS guidelines should also be revised to better reflect the current situation. The ratio of assessed values to market values of land has been quite low in Korea, and this is the partial reason behind the rapid hike in project site prices around the time of project establishment and growing gap between compensation costs as projects become delayed. Thus, PFS provisions regarding the land compensation ratio must be updated to reflect at minimum the actual conditions.

Preface
Executive Summary

Chapter 1 Introduction
　Section 1 Research Background
　Section 2 Research Objectives and Scope
　Section 3 Differences from Previous Studies

Chapter 2 Preliminary Feasibility Study and Land Cost Estimation
　Section 1 Overview of the Preliminary Feasibility Study
　Section 2 Land Compensation Cost Estimation in PFS
　Section 3 Land Compensation and Related Regulations

Chapter 3 Empirical Analysis
　Section 1 Analytical Model
　Section 2 Data Overview
　Section 3 Empirical Results

Chapter 4 Conclusion and Policy Implications

References
ABSTRACT