Lending through the technology lifecycle: Strategies for information search by banks in renewable energy project finance

Published: 30 April 2024| Version 1 | DOI: 10.17632/df5zf9jhgs.1


The code replicates the results of the manuscript: Lending through the technology lifecycle: Strategies for information search by banks in renewable energy project finance. The data for the code is sourced from Bloomberg New Energy Finance (BNEF), and the Bloomberg Terminal. The authors do not have the right to share the raw data. Readers are recommended to download the raw data from Bloomberg New Energy Finance and the Bloomberg Terminal before running the scripts. For any questions and requests, please contact Anurag Gumber (corresponding author).


Steps to reproduce

Note: There are two projects required to run the results: 1) BIS Organizations, followed by 2) Banking Information Search. For each project, the top-file to execute in respective projects is “Headline.R”. Data: The files contain sample data format that must be matched before the code is executed. The following data must be downloaded from BNEF: 1. Global organizations database: Extracted 29 Feb 2024 for manuscript 2. Global transactions database (for Solar PV, Wind Onshore, Wind Offshore) : Extracted 23 Jan and 31 Jan 2024 for manuscript The following data must be extracted from the Bloomberg Terminal: 1. Bloomberg Industry Classification System Level 1,2,3,4 classification for organizations in BNEF organizations database. Templates for all files containing column names are provided by the Authors. Manual classifications of BICS levels and sub activities used to classify organizations into company categories (types) contain confidential information that authors are not authorized to share. The following steps can be applied to replicate the results: 1. BIS Organizations: The data downloaded from BNEF and Bloomberg terminal must be placed inside “Data Files” (or path must be changed in “Read Data.R”. The downloaded data must comply with template format and column names. Thereafter, “Headline.R” should be executed. The output files are stored in “Data Files”. The classification files, namely for BICS and subactivities, is classified. 2. Banking Information Search: a. The data from BNEF (Transactions database) should be downloaded and placed in “data files”. The path for the data must be corrected in “Read Data.R”. b. The output data from BIS Organizations should also be read within “Read Data.R” at the bottom of the file. Path should be corrected by user. c. Thereafter, “Headline.R” (within Banking Information Search) should be executed. The Headline file executes fractional response (frm) and logit models and saves the output automatically. d. The outputs of models (frm and logit) are stored in “out files”. The tables present in the manuscript are also stored in “out files”. The author provides excel template for quick replication of manuscript tables. e. The output plots are stored in folder “plots”.


Eidgenossische Technische Hochschule Zurich


Banking, Financial Economics, Clean Energy Finance


HORIZON EUROPE European Research Council