In the modern day, investors care more than ever about the ethics of business. As climate change worsens and social injustice becomes a larger part of the public consciousness, aligning with ethical conduct is a crucial part of multinational firms’ images. However, the challenges of business have changed as well. While firms previously focused on how to enrich themselves, firms now have to think more about how they can enrich themselves and better the world around them at the same time. In this environment, financial instruments known as Use of Proceeds Bonds have become popular, as they dedicate the funds raised to specific positive net impact causes. The two that this essay focuses on are green bonds and social bonds. To consider why the private sector might invest more in one instrument than another, this paper compares the sources of investment in renewable energy (a green cause) relative to sources of investment in water infrastructure (a social cause). This paper will also go into specifics on investment in both sectors in terms of technology and regional variation. The purpose of this essay is to explore if and why one form of UOP bond is more popular than the other. However, it is not meant to be a definitive answer as to why one is more popular than the other. The data used in this piece is limited and would have likely had different results if it used different data sources to represent green investment and social investment, such as conservation investment (green) against transportation infrastructure investment (social). By the end of this piece, there will be a hypothesis as to why one form of investment is more popular than another, not a conclusion meant to represent objective fact.
Code
options(dplyr.summarise.inform =FALSE, pillar.warn.conflicts =FALSE)suppressWarnings({suppressMessages({suppressPackageStartupMessages(library(tidyverse))suppressPackageStartupMessages(library(readxl))suppressPackageStartupMessages(library(here))suppressPackageStartupMessages(library(patchwork))suppressPackageStartupMessages(library(esquisse))library(tidyverse)library(readxl)library(here)library(patchwork)library(esquisse)#Part 1: ICMA Data on Use of Funds Bonds path_to_sheet_1 <-here("C:/Users/adams/Downloads/Data Raw", "ICMA-Export.xlsx") read_ICMA_sheet <-partial(.f = read_excel,path = path_to_sheet_1,sheet ="ICMA_LGX_Data_Table",col_names =FALSE ) sheet_header_1 <-read_ICMA_sheet(range ="B4:O4") ICMA_desc_range <-"B4" ICMA_desc <-read_ICMA_sheet(range = ICMA_desc_range) |>pull() ICMA_info_range <-"B5:B2945" ICMA_info <-read_ICMA_sheet(range = ICMA_info_range) ICMA_desc <-read_ICMA_sheet(range ="B4:O4") ICMA_desc ICMA_info <-read_ICMA_sheet(range ="B5:O2945") the_names_of_columns_1 <-as.character(ICMA_desc[1, ])colnames(ICMA_info) <- the_names_of_columns_1 ICMA_info ICMA_info_2 <- ICMA_info %>%mutate(`Total Issuance (USD bn.)`=sum(`Amount issued (USD bn.)`)) ICMA_info_2 clean_ICMA_info <- ICMA_info %>%mutate_at(vars(matches("NA")), ~replace(., is.na(.), 0))#Part 3: Governmental Investment in Infrastructure Through Global Infrastructure Hub Data path_to_sheet_3 <-here("C:/Users/adams/Downloads/Data Raw", "data (2).xlsx") read_GIH_sheet <-partial(.f = read_excel,path = path_to_sheet_3,sheet ="data (2)",col_names =FALSE ) sheet_header_3 <-read_GIH_sheet(range ="A1:H1") GIH_desc_range <-"A1" GIH_desc <-read_GIH_sheet(range = GIH_desc_range) |>pull() GIH_info_range <-"A2:A1410" GIH_info <-read_GIH_sheet(range = GIH_info_range) GIH_desc <-read_GIH_sheet(range ="A1:H1") GIH_desc GIH_info <-read_GIH_sheet(range ="A2:H1410") the_names_of_columns_3 <-as.character(GIH_desc[1, ])colnames(GIH_info) <- the_names_of_columns_3 GIH_info GIH_info_better <- GIH_info %>%distinct() GIH_info_region <- GIH_info_better %>%select(-c(`Country`, `Income Group`)) %>%filter(!is.na(`Water investment, millions (constant 2015 USD)`)) %>%mutate(region =`Region`) %>%arrange(`region`) %>%group_by(`region`) %>%summarise(Region_total =sum(as.numeric(`Water investment, millions (constant 2015 USD)`), na.rm =TRUE)*sum(as.numeric(`Population`), na.rm =TRUE)) %>%distinct() %>%ungroup() GIH_info_region GIH_info_totals <- GIH_info_better %>%group_by(Year) %>%summarise(total_water_investment_public =sum(`Water investment, millions (constant 2015 USD)`)) GIH_info_totals GIH_info_yearschange <- GIH_info_better %>%select(-`Population`, -`GDP per capita, current USD`) %>%pivot_wider(names_from = Year, values_from =`Water investment, millions (constant 2015 USD)`) GIH_info_yearschange#Part 2: IRENA data for Renewable Energy Investors path_to_sheet_2 <-here("C:/Users/adams/Downloads/Data Raw", "2023_IRENA-CPI_RE_Investment_Trends.xlsx") read_IRENA_sheet <-partial(.f = read_excel,path = path_to_sheet_2,sheet ="IRENA-CPI RE Investments",col_names =FALSE ) sheet_header_2 <-read_IRENA_sheet(range ="A1:D1") IRENA_desc_range <-"A1" IRENA_desc <-read_IRENA_sheet(range = IRENA_desc_range) |>pull() IRENA_info_range <-"A2:A387" IRENA_info <-read_IRENA_sheet(range = IRENA_info_range) IRENA_desc <-read_IRENA_sheet(range ="A1:D1") IRENA_desc IRENA_info <-read_IRENA_sheet(range ="A2:D387") the_names_of_columns_2 <-as.character(IRENA_desc[1, ])colnames(IRENA_info) <- the_names_of_columns_2 IRENA_info IRENA_info_yearschange <- IRENA_info %>%pivot_wider(names_from = Year, values_from =`Investment in USD billion`)#Part 4: World Bank Private Sector Infrastructure Investment path_to_sheet_4 <-here("C:/Users/adams/Downloads/Data Raw", "CustomQuery (1).xlsx") read_WB_sheet <-partial(.f = read_excel,path = path_to_sheet_4,sheet ="CustomQuery",col_names =FALSE ) sheet_header_4 <-read_WB_sheet(range ="A1:F1") WB_desc_range <-"A1" WB_desc <-read_WB_sheet(range = WB_desc_range) |>pull() WB_info_range <-"A2:A420" WB_info <-read_WB_sheet(range = WB_info_range) WB_desc <-read_WB_sheet(range ="A1:F1") WB_desc WB_info <-read_WB_sheet(range ="A2:F420") the_names_of_columns_4 <-as.character(WB_desc[1, ])colnames(WB_info) <- the_names_of_columns_4 WB_info_filled_1 <-fill(WB_info, 1) WB_info_filled_1 WB_info_filled_2 <-fill(WB_info_filled_1, 2) WB_info_best <- WB_info_filled_2 %>%select(-c(`Payment commitments to the government`, `Investment commitments in physical assets`)) %>%arrange(region) %>%group_by(region, `investment year`, `primary sector`) %>%summarise(Region_total_2 =sum(as.numeric(`Total investment commitments`), na.rm =TRUE)) %>%distinct() WB_info_total_water <- WB_info_filled_2 %>%filter(`primary sector`=="Water and sewerage") %>%group_by(`investment year`) %>%summarize(total_water_investment_private =sum(`Total investment commitments`, na.rm =TRUE)) WB_info_total_water totals_past_2007 <- WB_info_total_water %>%filter(`investment year`>2006) %>%filter(`investment year`<2021) %>%mutate(Year =`investment year`) %>%select(-`investment year`) WB_info_final_1 <- WB_info_best %>%filter(!(region %in%c("<strong>Grand Total </strong>", "<strong>Subtotal - East Asia and Pacific</strong>", "<strong>Subtotal - Europe and Central Asia</strong>", "<strong>Subtotal - Latin America and the Caribbean</strong>", "<strong>Subtotal - Middle East and North Africa</strong>","<strong>Subtotal - South Asia</strong>", "<strong>Subtotal - Sub-Saharan Africa</strong>" ))) WB_info_final_1 WB_info_final_2 <- WB_info_best %>%filter(region %in%c("<strong>Grand Total </strong>", "<strong>Subtotal - East Asia and Pacific</strong>", "<strong>Subtotal - Europe and Central Asia</strong>", "<strong>Subtotal - Latin America and the Caribbean</strong>", "<strong>Subtotal - Middle East and North Africa</strong>","<strong>Subtotal - South Asia</strong>", "<strong>Subtotal - Sub-Saharan Africa</strong>" )) %>%mutate(region =case_when( region =="<strong>Grand Total </strong>"~"Grand Total", region =="<strong>Subtotal - East Asia and Pacific</strong>"~"Subtotal East Asia and Pacific", region =="<strong>Subtotal - Europe and Central Asia</strong>"~"Subtotal Europe and Central Asia", region =="<strong>Subtotal - Latin America and the Caribbean</strong>"~"Subtotal Latin America and the Caribbean", region =="<strong>Subtotal - Middle East and North Africa</strong>"~"Subtotal Middle East and North Africa", region =="<strong>Subtotal - South Asia</strong>"~"Subtotal South Asia", region =="<strong>Subtotal - Sub-Saharan Africa</strong>"~"Subtotal Sub-Saharan Africa",TRUE~ region # Keep other values unchanged )) WB_info_final_2# Part 5: Frankenstein-ing the Infrastructure Tibbles Together path_to_sheet_5 <-here("C:/Users/adams/Downloads/Data Raw/CPIAUCSL.xls") read_CPI_sheet <-partial(.f = read_excel,path = path_to_sheet_5,sheet ="FRED Graph",col_names =FALSE ) sheet_header_5 <-read_CPI_sheet(range ="A11:B11") CPI_desc_range <-"A11" CPI_desc <-read_CPI_sheet(range = CPI_desc_range) |>pull() CPI_info_range <-"A12:A938" CPI_info <-read_CPI_sheet(range = CPI_info_range) CPI_desc <-read_CPI_sheet(range ="A11:B11") CPI_desc CPI_info <-read_CPI_sheet(range ="A12:B938") the_names_of_columns_5 <-as.character(CPI_desc[1, ])colnames(CPI_info) <- the_names_of_columns_5 CPI_info <- CPI_info %>%filter(format(observation_date, "%m") =="01") %>%mutate(inflation_rate = (CPIAUCSL -lag(CPIAUCSL)) /lag(CPIAUCSL)) CPI_info CPI_info_past_2007 <- CPI_info %>%filter(observation_date >"2006-01-01", observation_date <"2020-01-01") %>%mutate(`Year`=year(observation_date)) %>%select(-observation_date) CPI_info_past_2007 yearly_total_water_infra <-merge(totals_past_2007, GIH_info_totals, by ="Year") yearly_total_water_infra CPI_total_water_infra <-merge(yearly_total_water_infra, CPI_info_past_2007, by ="Year") %>%mutate(inflation_factor = CPIAUCSL/CPIAUCSL[which(`Year`==2015)]) %>%mutate(adjusted_value = total_water_investment_private * (CPIAUCSL/ CPIAUCSL[which(`Year`==2015)])) %>%rename("Private Sector Investment"=`adjusted_value`, "Public Sector Investment"=`total_water_investment_public`) CPI_total_water_infra final_yearly_total_water_infra<- CPI_total_water_infra %>%select(-c(total_water_investment_private)) %>%pivot_longer(cols =c(`Private Sector Investment`, `Public Sector Investment`), names_to ="column_header", values_to ="values")})})
area_of_green <- ICMA_info %>%filter(`Bond type`=="Green") %>%group_by(`Issuer location`) %>%summarize(sum_by_country =sum(`Amount issued (USD bn.)`)) %>%arrange(desc(sum_by_country)) %>%slice_head(n =10) %>%distinct()area_of_social <- ICMA_info %>%filter(`Bond type`=="Social") %>%group_by(`Issuer location`) %>%summarize(sum_by_country =sum(`Amount issued (USD bn.)`)) %>%arrange(desc(sum_by_country)) %>%slice_head(n =10) %>%distinct()ggplot(data = area_of_green, aes(x =`Issuer location`, y = sum_by_country, fill =`Issuer location`)) +geom_col() +scale_fill_brewer(palette ="PRGn") +labs(title ="Total Amount in USD Billions Issued in Green Bonds in the 10 Countries with the Highest Investment", x ="Country", y ="Total Investment") +theme(plot.title =element_text(size =9), axis.text.x =element_text( angle =45, size =7, hjust =1))
Code
ggplot(data = area_of_social, aes(x =`Issuer location`, y = sum_by_country, fill =`Issuer location`)) +geom_col() +scale_fill_brewer(palette ="RdYlBu") +labs(title ="Total Amount in USD Billions Issued in Social Bonds in the 10 Countries with the Highest Investment", x ="Country", y ="Total Investment") +theme(plot.title =element_text(size =9), axis.text.x =element_text(angle =45, size =7, hjust =1))
Which is Invested in More, Social Bonds or Green Bonds?
When looking at a breakdown of all the types of ESG-linked use of proceeds bonds, green bonds make up a clear majority of issuance (in terms of the USD value of the issued bonds rather than the literal number of bonds published), while social bonds are only around a third of green bonds' size. This paper neglects to discuss sustainability/ sustainability-linked bonds since their guiding principles are too opaque, as they can cover any of the ESG catagories. Germany, the United States, and France have the three largest green bond issuances, while France and Supranational organizations (Ex: the World Bank or EU) have the two largest social bond issuances. Note that France, the largest social bond issuer, issues less in social bonds than it does in green bonds by a wide margin, and it is only the third largest green bond issuer. Based on this data, and the assumption that bonds issued for projects represents the market interest in these projects, we can establish that green investment is in higher demand than social investment. The rest of this paper is based around theorizing why this is the case.
Code
#Part 7: Initial Data Analysis of IRENA Datanames(IRENA_info)
[1] "Category" "Product"
[3] "Year" "Investment in USD billion"
Code
IRENA_focus_1 <- IRENA_info %>%filter(Product %in%c("Private", "Public"))sector_investment <- IRENA_focus_1 %>%group_by(Year, Product) %>%summarise(total_investment =sum(`Investment in USD billion`)) %>%ungroup()total_investment_year <- sector_investment %>%group_by(Year) %>%summarise(total_investment_year =sum(total_investment)) %>%ungroup()sector_investment <-left_join(sector_investment, total_investment_year, by ="Year")percentages <- sector_investment %>%mutate(percentage = total_investment / total_investment_year *100)ggplot(data = sector_investment, aes(x = Year, y = total_investment, fill = Product)) +geom_col(position ="stack") +geom_text(data = percentages, aes(label =paste0(round(percentage), "%"), y = total_investment), color ="black", size =3, vjust =1.5) +labs(title ="Renewable Energy Investment in USD billion by Year and Product",x ="Year",y ="Investment in USD billion",fill ="Type of Investor")
Code
IRENA_focus_2 <- IRENA_info %>%filter(Category %in%c("Technology"))sum_by_tech <- IRENA_focus_2 %>%group_by(`Product`) %>%summarise(`Total Spending on Energy`=sum(`Investment in USD billion`))percentages_2 <-with(sum_by_tech, `Total Spending on Energy`/sum(`Total Spending on Energy`) *100)label_with_percentages_2 <-sprintf("%s (%.1f%%)", sum_by_tech$`Product`, percentages_2)label_colors_2 <-c("gold", "purple", "red", "brown", "turquoise", "darkgreen", "lightblue","gray", "maroon" )pie(x = sum_by_tech$`Total Spending on Energy`,labels = label_with_percentages_2,col = label_colors_2,main ="Amount of Investment in Each Type of Renewable Energy Tech Between 2013 and 2022",cex.main = .9,cex = .7)
Renewable Energy Analysis:
Every year between 2013 and 2020, annual private sector investment in renewable energy has been higher than public sector investment. Around 80% of overall investment has been dedicated to solar and wind energy. It is clear that renewable energy is a lucrative venture for the private sector to invest in, for one reason or another. External research validates this claim, as market forces are seeing the opportunity for profit found in a clean energy future. Public policy initiatives around the world alongside concerns of energy insecurity from the Russia-Ukraine war have driven overall investment in renewables to grow at a higher rate than fossil fuels(IEA, 2023). Policy initiatives incentivize energy investment by lowering costs to the firm, such as the tax shields provided by the US Inflation Reduction Act. Demand for clean energy technology is rising rapidly as energy infrastructure is replaced across the world, meaning early investors in clean technology firms could reap considerable returns in the future. Solar energy has attracted the most investment due to its predicted CAGR of 15% between 2023 and 2030(Forbes). In addition to this, contemporary solar and wind technology are also highly efficient. The two have the lowest generation costs relative to fossil fuel and non-fossil fuel energy sources(IEA 2, 2023). The private sector has been enticed into investing in renewable energy, especially solar and wind, due to the profitability of the sector. The public sector is encouraging and engaging in this investment, creating a financial environment where renewable energy projects are in high demand and thus green bond issuance is high.
Code
GIH_info_better %>%ggplot(aes(y = Region, x = Year, fill =`Water investment, millions (constant 2015 USD)`)) +geom_tile() +scale_fill_gradient(low ="white", high ="red") +theme_minimal() +theme(legend.key.size =unit(0.5, "cm"),legend.title =element_text(size =8), axis.text.x =element_text( angle =45, size =7, hjust =1)) +labs(x ="Year", y ="Region", title ="Water and Sewerage Investment by the Public Sector in USD Millions", subtitle ="Divided into Regions")
Code
WB_info_final_1 %>%filter(`primary sector`=="Water and sewerage") %>%ggplot(aes(y = region, x =`investment year`, fill = Region_total_2)) +geom_tile() +scale_fill_gradient(low ="white", high ="blue") +theme_minimal() +theme(legend.key.size =unit(0.5, "cm"),legend.title =element_text(size =6), axis.text.x =element_text( angle =45, size =7, hjust =1)) +labs(x ="Year", y ="Region", title ="Water and Sewerage Investment by the Private Sector in USD Millions", subtitle ="Divided into Regions", fill ="Water investment, millions (not adjusted)")
Code
sector_investment_2 <- final_yearly_total_water_infra %>%group_by(Year, column_header) %>%summarise(total_investment =sum(values)) %>%ungroup()total_investment_year_2 <- sector_investment_2 %>%group_by(Year) %>%summarise(total_investment_year =sum(total_investment)) %>%ungroup()sector_investment_2 <-left_join(sector_investment_2, total_investment_year_2, by ="Year")percentages_3 <- sector_investment_2 %>%mutate(percentage = total_investment / total_investment_year *100)ggplot(data = sector_investment_2, aes(x = Year, y = total_investment, fill = column_header)) +geom_col(position ="stack") +geom_text(data = percentages_3, aes(label =paste0(round(percentage), "%"), y = total_investment), position =position_stack(vjust =0.5), color ="black", size =3) +labs(title ="Annual Worldwide Water Infrastructure Investment in USD Millions (Adjusted to be in 2015 dollars)",subtitle ="Subdivided into Private and Public Sector Contributions, but not adjusted for overlap",x ="Year",y ="Investment in USD Millions",fill ="Private vs Public")
Water Infrastructure Analysis:
A noteworthy aspect about social bonds is that they often target lower income groups, meaning finding meaningful data to represent social bonds is a challenge, as infrastructure investment data does not always isolate low income groups. For example, it is much easier to find global data on real estate investment than it is to find data on affordable housing investment, and it is even more challenging to find data that breaks down private sector investment in affordable housing versus public sector investment worldwide. Water and sewerage infrastructure investment is a good proxy for social bonds because it fundamentally serves a “social” purpose, as it is almost always publicly accessible, while not all real estate investment serves this purpose. In order to engage with this subject, three different data sources had to be consulted. The World Bank provided data on private sector investment, the Global Infrastructure Hub provided data on public sector water investment (but only in 2015 dollars), so CPI data from the Federal Reserve was used to adjust the private sector data in order to make the data comparable.
In a sharp contrast with renewable energy, almost all investment in water infrastructure between 2007 and 2020 has been public sector investment. Most private sector water infrastructure investment has been concentrated in Latin America, the Caribbean, East Asia, and the Pacific Islands. In a sharp contrast to this, most public sector water infrastructure investment has been in Western Europe and North America. Based on this data, it is clear that water infrastructure is not the focus of private sector investment. Investment in water infrastructure, unlike many other utilities, lacks clear sources of revenue despite being a necessary public good. Often revenues collected from water infrastructure do not cover the lofty capital expense associated with the investment, and even if the investment does derive consistent revenue, the payment of the original investment can take 20-30 years to finish. On top of this, the risks of water investments are hard to quantify, as they are often idiosyncratic in a particular project and financial institutions have little experience with quantifying them (OECD, 2.2.2). This value proposition is not particularly attractive to private sector financial institutions, so it makes sense as to why this social good is financed primarily through public spending. If water infrastructure investment is a strong proxy for social bonds, then it is possible that social projects are largely not very lucrative and thus do not invite private sector investment.
It is noteworthy that private capital was used to develop water infrastructure in emerging markets with well-developed government bodies more frequently, while public capital was largely spent in developed markets. This may have to do with municipalities having fewer public finances to spend, leading more emerging markets to turn to public-private partnerships (PPPs). Through the PPP model, the private sector builds and operates the plant using its’ own capital, but sells the water to the government to disseminate publicly. As a result of PPPs, the private sector can be crowded-in to investment in water infrastructure(KPMG). This is interesting because the private sector can have an incentive to engage in social investment if there is a steady cash flow from it through a PPP, which may explain why there is any social bond investment at all. It is also important to note that the original data used does not separate out PPPs, so there could be more private sector investment in water infrastructure than the worldwide water infrastructure investment chart shows.
Conclusion:
It bears repeating; this paper is meant to illustrate a long-form hypothesis rather than a definitive argument. The data used does not paint a full picture of green bonds or social bonds, but is meant to propose a reason why overall investment in the two types of bonds are so different in quantity. Because this is the case, the purpose of this piece is to present a reasonable hypothesis based on the data analysed. However, this any and all claims in this paper would require rigorous analytical work before being provable.
The hypothesis that this paper ends with is that the reason why green bonds attract more investment than social bonds is because the private sector is more interested in investing in green causes than in social ones, while the public sector invests in both causes.
Global Infrastructure Hub. Data Tool. Link: https://www.gihub.org/infrastructure-monitor/data-tool/
The World Bank. PPI Data. Link: https://ppi.worldbank.org/en/ppidata
The Federal Reserve Bank of St Louis. Consumer Price Index for All Urban Consumers: All Items in U.S. City Average. Link: https://fred.stlouisfed.org/series/CPIAUCSL#
IEA. Solar PV. IEA, 2023. Link: https://www.iea.org/energy-system/renewables/solar-pv
Simeon, Bastien. Delivering Water Infrastructure using Private Finance. KPMG International, 2011. Link: https://www.nh.gov/water-sustainability/publications/documents/water-infrastructure-using-private-finance.pdf
