Nadia Yuning Xing
April 17,2024
This study focus on analyzing the trends, capital allocation and efficiency of green bond issuance in Latin America, a choice based on several key considerations. First, as a region rich in biodiversity and facing significant environmental challenges, Latin America is in dire need of innovative financial instruments to balance its economic development with ecological preservation. Second, the region has demonstrated leading global innovation in green finance and sustainable development policies, particularly in carbon pricing and the development of green bond standards. In addition, Latin America’s economic structure and stage of development make it an ideal case study for examining how infrastructure finance can contribute to sustainable development. By exploring Latin America’s green bond market in depth, it will not only provide insights into the region’s own sustainable development, but also provide valuable references for the development of similar markets globally. The selection of Latin America as the region of focus for this study therefore aims to enhance our understanding of the role of green financial instruments in promoting environmental protection and economic growth, while also providing support and insights for global climate action.
This analysis seeks to provide an overview of green bonds in LaTAm regions by answering the following questions:
What is the trend in green bond issuance over time in Latin America and the world?
What is the primary investment focus of green bonds?
What is the primary investor of green bonds?
To do this analysis, I combined five data sets from GBTP: https://www.greenbondtransparency.com/support/resources/
The data sets are respectively containing:
(1) bond’s general information
(2) allocation of proceeds to project categories using international standards (GBP, CBI)
(3) measurements of impact and outcome KPIs
(4) individual bond-tranche information
(5) detailed information of the projects receiving disbursements
In the first paragraph, let us first explore the green bond issuance trend in Latin America from 2016 to 2022 and compare it with the global green bond issuance trend. From the line chart below, we can find that generally speaking, Latin America and the world have maintained the same trend, both growing first and then declining. The difference is that global green bond issuance has risen rapidly in 2021, and this has not also happened in the Latin America. In terms of growth rate, the growth rate of green bond issuance in Latin America and the world is basically the same, except for 2021.
library(tidyverse)
library(readxl)
library(plotly)
library(ggplot2)
library(scales)
comparison <- read_csv("/Users/xingyuning/Desktop/susfin/final project/01_data_cleaning/comparison.csv")
data_long_comparison <- comparison |>
pivot_longer(cols = c("LAC_volume_usd", "global_volume_usd"), names_to = "volume_type", values_to = "issuance_volume")library(dplyr)
data_growth <- comparison %>%
arrange(issuance_year) %>%
mutate(
LAC_growth = (LAC_volume_usd - lag(LAC_volume_usd)) / lag(LAC_volume_usd) * 100,
Global_growth = (global_volume_usd - lag(global_volume_usd)) / lag(global_volume_usd) * 100
)
library(tidyr)
library(ggplot2)
data_growth_long <- data_growth %>%
pivot_longer(cols = c("LAC_growth", "Global_growth"), names_to = "growth_type", values_to = "growth_rate")
p1 <- ggplot(data_long_comparison, aes(x = issuance_year, y = issuance_volume, fill = volume_type)) +
geom_col(position = "stack") +
scale_fill_manual(values = c("LAC_volume_usd" = "#377eb8", "global_volume_usd" = "#4daf4a")) +
theme_minimal() +
labs(title = "Issuance Volume of Green Bonds", x = "Year", y = "Issuance Volume (bn)", fill = "Volume Type")
p2 <- ggplot(data_growth_long, aes(x = issuance_year, y = growth_rate, color = growth_type)) +
geom_line() +
geom_point() +
scale_color_viridis_d(begin = 0.3, end = 0.9, option = "D") +
theme_minimal() +
labs(title = "Annual Growth Rate of Green Bond Issuance", x = "Year", y = "Growth Rate (%)", color = "Growth Type")
p1_ly <- ggplotly(p1)
p2_ly <- ggplotly(p2)
subplot(p1_ly, p2_ly, nrows = 2, margin = 0.05, shareX = TRUE) %>%
layout(title = "Green Bond Market Trend",
xaxis = list(title = "Year"),
xaxis2 = list(title = "Year"))The following graph is showing the issuance volume among Latin America countries:
library(leaflet)
library(countrycode)
library(rnaturalearth)
bonds_new <- read_csv("/Users/xingyuning/Desktop/susfin/final project/01_data_cleaning/bonds_new.csv")
bonds_new$issuer_jurisdiction <- ifelse(bonds_new$issuer_jurisdiction == "Supranational", NA, bonds_new$issuer_jurisdiction)
bonds_new$ISO_code <- countrycode(bonds_new$issuer_jurisdiction, origin = "country.name", destination = "iso3c", warn = FALSE)
bonds_new <- bonds_new[!is.na(bonds_new$ISO_code), ]
bonds_summary <- bonds_new %>%
group_by(ISO_code) %>%
summarise(volume_usd_total = sum(volume_usd, na.rm = TRUE))
world_map <- ne_countries(scale = "medium", returnclass = "sf")
latin_america_caribbean_map <- world_map[world_map$region_un == "Americas" &
world_map$subregion %in% c("Central America", "South America", "Caribbean"), ]
latin_america_bonds <- merge(latin_america_caribbean_map, bonds_summary, by.x = "iso_a3", by.y = "ISO_code")
leaflet(latin_america_bonds) %>%
addProviderTiles(providers$CartoDB.Positron) %>%
addPolygons(fillColor = ~colorQuantile("Blues", volume_usd_total)(volume_usd_total),
color = "#BDBDC3",
weight = 1,
opacity = 1,
fillOpacity = 0.7,
highlight = highlightOptions(weight = 2,
color = "#666",
fillOpacity = 0.7),
label = ~paste(iso_a3, ":", formatC(volume_usd_total, format = "f", big.mark = ","))) %>%
addLegend("bottomright", pal = colorQuantile("Blues", latin_america_bonds$volume_usd_total),
values = ~volume_usd_total,
title = "Issuance Volume (USD)",
labFormat = labelFormat(big.mark = ","))In this paragraph, I want to show in which categories these green bonds are allocated. From the chart below, we can see 57% of green bonds invest in energy category, followed by water, buildings and Land-use.
library(dplyr)
library(ggplot2)
allocations_new <- read_csv("/Users/xingyuning/Desktop/susfin/final project/01_data_cleaning/allocations_new.csv")
distinct_allocations <- allocations_new %>%
distinct(bond_name, .keep_all = TRUE)
allocation_counts <- distinct_allocations %>%
group_by(category) %>%
summarise(count = n())
allocation_counts <- allocation_counts %>%
mutate(percentage = count / sum(count) * 100)
color_palette <- c("#6baed6", "#fd8d3c", "#74c476", "#de2d26", "#9e9ac8", "#c6b0a4", "#e377c2", "#17becf", "#bcbd22")
ggplot(allocation_counts, aes(x = reorder(category, -percentage), y = percentage, fill = category)) +
geom_bar(stat = "identity", color = "black", size = 0.5) +
geom_text(aes(label = sprintf("%.2f%%", percentage)), vjust = -0.3, size = 3.5, color = "black") +
scale_fill_manual(values = color_palette) +
labs(title = "Proportion of Green Bond Allocation By Bonds Number",
subtitle = "Green bond allocations categorized by usage",
x = "Category",
y = "Percentage (%)") +
theme_minimal() +
theme(plot.title = element_text(face = "bold", size = 14),
plot.subtitle = element_text(size = 12),
axis.title = element_text(size = 12),
axis.text = element_text(size = 10),
legend.position = "none",
plot.background = element_rect(fill = "#f0f0f0"),
panel.grid.major = element_line(color = "#e0e0e0", size = 0.5),
panel.grid.minor = element_blank()) +
coord_flip() 
The third paragraph analyzes the investor type of these green bonds. The chart shows the majority are issued by Non-financial corporation (64.4%), followed by financial corporate (25.6%) . The sovereign green bonds only account for 3.13% in the total number of green bonds.
bonds_new_issuer <- bonds_new %>%
distinct(bond_name, .keep_all = TRUE)
issuer_counts <- bonds_new_issuer %>%
group_by(issuer_type) %>%
summarise(count = n())
issuer_counts <- issuer_counts %>%
mutate(percentage = count / sum(count) * 100)
library(ggplot2)
color_palette <- c("#6baed6", "#fd8d3c", "#74c476", "#de2d26", "#9e9ac8", "#c6b0a4")
ggplot(issuer_counts, aes(x = reorder(issuer_type, -percentage), y = percentage, fill = issuer_type)) +
geom_bar(stat = "identity", color = "black", size = 0.5) +
geom_text(aes(label = sprintf("%.2f%%", percentage)), vjust = -0.3, size = 3.5, color = "black") +
scale_fill_manual(values = color_palette) +
labs(title = "Proportion of Issuer Type of Green Bonds in LAC",
subtitle = "Green bonds distributed by issuer type",
x = "Issuer Type",
y = "Percentage (%)") +
theme_minimal() +
theme(plot.title = element_text(face = "bold", size = 14),
plot.subtitle = element_text(size = 12),
axis.title = element_text(size = 12),
axis.text = element_text(size = 10),
legend.position = "none",
plot.background = element_rect(fill = "#f0f0f0"),
panel.grid.major = element_line(color = "#d6d6d6", size = 0.5),
panel.grid.minor = element_blank()) +
coord_flip() 
From the above analysis, we can see that there was a huge increase in green bonds in the Latin American market in 2019. The main countries driving the overall growth of the Latin American green bond market include Brazil, Mexico and Chile. Around 2019, some Latin American countries began to implement or improve policies and regulatory frameworks on green finance and green bonds. For example, countries such as Colombia and Mexico have introduced policies to encourage green investment, which have provided legal and policy support for the development of the green bond market. From the perspective of investment allocation, up to about 60% of the green bonds are invested in the energy sector; from the perspective of investor types, 64% of investors are non-financial institutions, and the smallest participants are sovereign bonds and local government bonds.