This China_Global_Power_FDI_readme.txt file was generated on 2020-12-07 by Zhongshu Li, Kevin Gallagher and Denise Mauzerall


GENERAL INFORMATION

1. Title of Dataset: Dataset for China's foreign direct investment in the global power sector 

2. Author Information
	A. Principal Investigator Contact Information
		Name: Denise Mauzerall
		Institution: Princeton University
		Address: 3-C-12 Green Hall and E412 Engineering Quad
		Email: mauzerall@princeton.edu


3. Date of data collection (single date, range, approximate date): 2018-02-01-2018-09-01

4. Geographic location of data collection: Boston University, Boston, MA, USA 

5. Information about funding sources that supported the collection of the data: The data collection efforts receive funding from the Princeton School of Public and International Affairs, the Climate Works Foundation, the Charles Stewart Mott Foundation and Rockefeller Brothers Foundation. 


SHARING/ACCESS INFORMATION

1. Licenses/restrictions placed on the data: N/A

2. Links to publications that cite or use the data: https://www.sciencedirect.com/science/article/abs/pii/S0301421519306433

3. Links to other publicly accessible locations of the data: https://docs.google.com/spreadsheets/d/1jPyn4unb06lx0trME4X-BhksiUUJGPSIoLpzLDAelps/edit#gid=0

4. Links/relationships to ancillary data sets: N/A

5. Was data derived from another source? yes/no
	A. If yes, list source(s): Yes. World Electric Power Plant database

6. Recommended citation for this dataset: 

Zhongshu Li, Kevin Gallagher Denise Mauzerall. Database for China's foreign direct investment in the global power sector (2020)

DATA & FILE OVERVIEW

1. File List: 

China's_foreign_direct_investment_in_the_global_power_sector.csv

2. Relationship between files, if important: N/A

3. Additional related data collected that was not included in the current data package: N/A

4. Are there multiple versions of the dataset? yes/no N/A
	A. If yes, name of file(s) that was updated: 
		i. Why was the file updated? 
		ii. When was the file updated? 


METHODOLOGICAL INFORMATION

1. Description of methods used for collection/generation of data: 

Methods is described in method section in the paper (https://www.sciencedirect.com/science/article/abs/pii/S0301421519306433)

2. Methods for processing the data: 

Our objective is to create a comprehensive plant-level dataset of all the power plants in which Chinese firms have invested around the world.  For each power plant, we provide basic information regarding its name, location, fuel type, capacity, year it came online, primary Chinese investor, investment amount and percentage of shares owned by Chinese firms. The core approach we use to compile our database is to adapt the Platts World Electric Power Database to account for the foreign ownership of China’s overseas power plants. Platts is an electric power unit-level database and its first global version was published in the early 1990s. We use the 2017 version.  Platts includes a comprehensive list of all power plants in the world derived from a variety of sources including direct surveys, power company financial and statistical reports, vendor reference lists, and trade and business press. For every power generating unit, Platts provides information regarding its name, operational status, capacity (MW), year-on-line, primary fuels, pollution control equipment, owner company, location data and additional attributes (Platts, 2015). While Platts does include the name of the company that has the majority ownership in a particular plant, it does not indicate the country of origin of the firm. In addition, there is a variety of subsidiaries or affiliates of the same holding companies, especially for China, and Platts makes limited effort to record them in a clear and consistent manner. Therefore, in order to better understand the reach of Chinese influence in the global power sector, we expand the ownership attributes of the Platts database and identify all power plants outside of China which involve Chinese foreign direct investment. Our approach is a five-step process which we describe below. 

a. Generation of Chinese company list
We first generate a list of Chinese power companies that may invest globally by surveying a wide variety of online databases and public reports. Four online databases are intensively utilized, including FDiMarket (2017), Dealogic (2017), Coal Plant Tracker (2017), and Global Coal Exit List (2017). Combining these sources, we create an initial list of more than 100 Chinese companies that we consider to be potential Chinese investors in foreign power markets. We supplemented this effort with internet searches, and found several additional companies that were lacking from these formal sources. We acknowledge that there could be additional missing Chinese companies, especially small to medium-sized private companies. Our complete list of Chinese energy companies operating abroad is provided in the supplementary information.

b. Direct matching of Chinese companies within Platts 
After compiling a list of Chinese firms that invest globally in the power sector, we then match the names in our company list with the companies listed in the Platts database. We use a simple matching algorithm that searches the Platts database for names of firms in our company list and variations thereof. For every Chinese company, we generate one to three keywords that are representative of all possible names that Platts uses for the company. We tag power plants with the name of a holding company when at least one of the keywords found in the Platts database indicates it is the original owner. We then put all of the power plants obtained via this matching process together into a new list. Examples of keywords are included in the supplementary information. After the direct match, we also check the list of Chinese overseas power plants and remove the false-positive matches: plants that are not Chinese-owned but were wrongly captured by matching keywords.  Coverage of renewable power plants in Platts is much less comprehensive than coverage of non-renewable power plants. Therefore, the top down approach based on Platts that we use may under-estimate Chinese investment in the global renewable power sector.

c. Indirect matching for Chinese companies within Platts
In terms of capacity, approximately two thirds of Chinese overseas power plants in our database are found to directly match power plants in the Platts database.  For another one third of power plants, Platts fails to record their Chinese ownership. For instance, after merger and acquisition deals when Chinese companies acquired overseas assets, Platts does not update the original corporate owner with the new Chinese owner in a timely manner.  There are also cases where Platts incorrectly recorded a local partner as the owner when the primary investment in the power plant is in fact a Chinese company. To generate a more inclusive list of power plants that includes not only those plants indicated by Platts to be Chinese owned, but also those Chinese plants that Platts missed, we conducted an internet search for all Chinese companies in our original list.  We looked for those company’s overseas power generating assets and then identified and tagged the corresponding records in Platts.  We call this indirect matching. The internet resources we accessed included annual reports of publicly listed firms, official websites of companies and governments, and trade and business press. For instance, Huanneng’s website and business press show that Huaneng acquired a 50% share of Ozgen from InterGen in 2003 via competitive bidding and now holds a majority stake in  an 1800 MW power generating asset in Australia in which they invested $227 million USD. In Platts, as we expect, there are also records of power plants with a total capacity of 1800 MW under Ozgen but they are still recorded as solely owned by InterGen.  Therefore, we label these missing matches where Platts fails to update as Chinese plants. In a similar fashion, we complete the list of Chinese owned power plants overseas. There are two types of activities that we do not include: (1) Equity investment, defined here as a financial investment made by entities with limited technology capacity in power project development, and (2) projects with EPC companies as a minor shareholders. While we have not included any projects involving equity investment or EPC contractors that are minor shareholders, future research in this area could be illuminating. All links for websites we used for indirect matching are available in the SI. 

d. Expand project level detail
In addition to identifying and validating all Chinese plants outside of China that are included in Platts, we also expand the record of power plant attributes with additional details, including the type of investment (greenfield vs. merger & acquisition), percentage of Chinese ownership, and amount of Chinese investment in USD. Using the same Huaneng example as in section 2.3, we record Chinese ownership as 50% and the deal completion year as 2003 in separate columns. A total investment of $227 million USD is also recorded. However, for many other deals, the investment amount is unavailable. There is general consistency among sources. If there are contradictions or inconsistencies, we give more weight, in order, to government websites, company websites and annual reports, trade and business press. 

e. Project status verification
A power plant project could take as long as a decade from its announcement to the day it starts operating.  Many announced projects end up canceled or delayed and an incorrect recording of Chinese outward investment could wrongly include these “zombie deals”. The most famous example is the Myitsone hydroelectric power project in Myanmar. In 2009, $3.6 billion USD of investment was announced but it was never realized as the project was suspended in 2011. Inclusion of such zombie projects would lead to an overestimation of Chinese investment. For merger and acquisition investment, we only include deals that are registered as completed by Dealogic or by multiple trade and business presses. Deals that were announced but have not been confirmed as completed are not included. For greenfield projects, we use the project status information provided by Platts, which records power plants as in operation, under construction, under planning, deferred, canceled or retired. In our analysis, we only include greenfield projects that are either in operation or under construction as valid investments. Investments at other project stages are excluded to avoid zombie deals. Since every investment is traced down to the power plant level, double counting is avoided. Similarly, the investment itself could also take years to finish. For greenfield projects, it is difficult to track the exact date of initial investment.  Platts, however provides comprehensive data on the year in which projects start operating (commission date). Therefore, although investments precede commission dates by years, due to lack of a viable alternative we use the year in which projects commence operation as the year of the investment. 

f. Partial Ownership
In most cases, Chinese investors are one among many other shareholders in power projects. Therefore, it is important to track not only the total capacity of power projects receiving Chinese investment but also the capacity directly owned by Chinese firms. In our analysis, we use “total capacity” to describe the total capacity of power plants in which China has made an investment of any size and “net capacity” to describe the power capacity China directly owns.
For example, for a 600MW coal power plant in which China has 50% share, Chinese net capacity in this project is 300MW and its total capacity is 600MW. If the capacity number is not specified as net capacity, it represents the total capacity involved with Chinese FDI, rather than the capacity directly owned by China.



3. Instrument- or software-specific information needed to interpret the data: 

N/A

4. Standards and calibration information, if appropriate: N/A

5. Environmental/experimental conditions: N/A

6. Describe any quality-assurance procedures performed on the data: N/A

7. People involved with sample collection, processing, analysis and/or submission: 

N/A

DATA-SPECIFIC INFORMATION FOR: [FILENAME]


1. Number of variables: 10

2. Number of cases/rows: 951

3. Variable List: 

Capacity (MW): Project capacity, a measure of the size of the power projects receiving Chinese foreign direct investment (unit: MW)

STATUS: The status of the projects

Online Year: The year in which the power project starts commercial operation

Technology: The technology used in the power projects to generate electricity

Chinese share: The share of Chinese ownership in the power projects

Type of investment: The type of investment Chinese companies made to acquire ownership in the power projects. Greenfield refers to a greenfield investment. M&A refers to a merger/acquisition transaction

Country: The country where the power project is located

Region: The region where the power project is located

Major Chinese Shareholder: The largest Chinese shareholder in the power project

Webpages or other sources: Sources of information on the power project

4. Missing data codes: 
N/A

5. Specialized formats or other abbreviations used: N/A