Edible Oils Import by India

India, one of the world’s largest consumers and importers of edible oils, has seen significant evolution in its import basket over the last decade. An analysis of import data from the financial year (from Nov to Oct) 2010-2011 through 2023-2024 reveals a dynamic volume growth, shifting consumer preferences, and geopolitical influences on sourcing.

The data clearly indicates a robust long-term growth trend in the total quantity of edible oil imported by India, with quantities often exceeding 15,000,000 tonnes in recent years.

The code structure

## Load required libraries
library(data.table)
library(ggplot2)
library(ggiraph)  ## For interactive plots
library(scales)
library(gridExtra)
library(gt)
library(ggthemes)
library(viridis)

## Custom theme for plots
my_theme <- theme_bw(base_size = 14) +
  theme(
    text = element_text(family = "sans", color = "#333333"),
    plot.title = element_text(size = 18, face = "bold", hjust = 0.5,
                              margin = margin(10, 0, 10, 0)),
    plot.caption = element_text(size = 10, color = "#666666", hjust = 0),
    axis.title = element_text(size = 14),
    axis.text = element_text(size = 12),
    legend.position = "bottom",
    legend.title = element_blank(),
    panel.grid.major = element_line(color = "#EEEEEE"),
    panel.border = element_blank(),
    axis.line = element_line(color = "#333333"),
    axis.text.x = element_text(angle = 90, vjust = 0.5, hjust = 1),
    strip.background = element_rect(fill = "#F2F2F2", color = NA)
  )

## Function to create interactive stacked bar chart
create_interactive_stackchart <- function(data, x_var, y_var, fill_var,
                                          label_var = NULL,
                                          title = "",
                                          y_lab = "Value",
                                          caption_text = "Data source: Ministry of Commerce, India | Analysis: Pawan") {

  ## Create base plot
    p <- ggplot(data, aes(x = data[[x_var]], y = data[[y_var]],
                          fill = data[[fill_var]])) +
        geom_bar_interactive(
            aes(tooltip = paste(data[[fill_var]], ": ",
                                format(data[[y_var]], big.mark = ","),
                                " (", ifelse(!is.null(label_var),
                                             paste0(data[[label_var]], "%"),
                                             ""), ")", sep = "")),
            data_id = data[[fill_var]],
    stat = "identity",
    position = "stack"
    ) +
    scale_fill_viridis(discrete = TRUE, option = "D") +
    my_theme +
    scale_y_continuous(labels = comma) +
    labs(
      y = y_lab,
      x = "Year",
      title = title,
      caption = caption_text
    )

  # Add labels if provided
  if (!is.null(label_var)) {
    p <- p + geom_text(
      aes(label = ifelse(.data[[label_var]] < 5, "",
                         paste0(data[[label_var]], "%"))),
      position = position_stack(vjust = 0.5),
      size = 3.5,
      color = "white",
      fontface = "bold"
    )
  }

  return(p)
}

ANNUAL EDIBLE OIL IMPORT (VALUE IN USD)

## Read data
palm <- readRDS("/home/pawan/edibleoils/palmoil_import.rds")
soyabean <- readRDS("/home/pawan/edibleoils/soyabean_import.rds")
sunflower <- readRDS("/home/pawan/edibleoils/sunflower_import.rds")

## Process data
palm$Edibleoil <- "Palm Oil"
soyabean$Edibleoil <- "Soyabean Oil"
sunflower$Edibleoil <- "Sunflower Oil"

## Format years
format_year <- function(year_str) {
  year_num <- as.numeric(substr(year_str, 1, 4))
  paste0(year_num, "-", year_num + 1)
}

palm$Year <- format_year(palm$Year)
soyabean$Year <- format_year(soyabean$Year)
sunflower$Year <- format_year(sunflower$Year)

## Combine data
edibleoil <- rbind(palm, soyabean, sunflower)

## Calculate shares for total country
oil_value <- edibleoil[Country == "Total",
                       .(ImportYear = sum(ImportYear, na.rm = TRUE)),
                       by = .(Year, Edibleoil)]

oil_value[, Total := sum(ImportYear), by = Year]
oil_value[, Share := round((ImportYear / Total) * 100)]

## Create interactive plot
p5_interactive <- create_interactive_stackchart(
  data = oil_value,
  x_var = "Year",
  y_var = "ImportYear",
  fill_var = "Edibleoil",
  label_var = "Share",
  title = "Edible Oil Import by India (Value in USD)",
  y_lab = "Value in US $ Million",
  caption_text = "Data source: Ministry of Commerce, India | Analysis: Pawan | Financial Year: April-March"
)

## Save interactive version
girafe(ggobj = p5_interactive,
       width_svg = 12,
       height_svg = 8,
       options = list(
         opts_tooltip(css = "background-color:white;color:black;padding:5px;border-radius:3px;"),
         opts_hover(css = "fill:orange;"),
         opts_zoom(min = 0.5, max = 2)
       ))

## %>%
##   girafe_save("annual-edibleoil-import-interactive.html")

PALM OIL IMPORT ANALYSIS

Palm oil remains the single largest imported oil, but its proportional share has decreased. It accounted for a high of 79% of total imports in 2010-2011, but this share has trended downwards, settling at 56% in 2023-2024.
India maintains a near-exclusive on its Southeast Asian neighbours for palm oil, with Indonesia and Malaysia consistently dominating the supply chain. While the proportional distribution between the two varies year-to-year—with Malaysia often holding the slight edge in the mid-period (e.g., 71% in 2013-2014)—the combined share of these two nations routinely exceeds 90%, highlighting a concentrated supply risk.

## Read quantity data
palmoil_qty <- readRDS("/home/pawan/edibleoils/palmoil_importqty.rds")
soybeanoil_qty <- readRDS("/home/pawan/edibleoils/soybeanoil_importqty.rds")
sunfloweroil_qty <- readRDS("/home/pawan/edibleoils/sunfloweroil_importqty.rds")

## Data cleaning function
clean_import_data <- function(dt) {
  ## Remove commas from numbers
  dt$ImportMonth <- gsub(",", "", dt$ImportMonth)

  ## Convert month abbreviations to numbers
  dt$Month_num <- match(dt$Month, month.abb)

  ## Create financial year (Nov-Oct)
  dt[, Fin_Year := ifelse(Month_num > 10,
                          paste0(Year, "-", as.numeric(Year) + 1),
                          paste0(as.numeric(Year) - 1, "-", Year))]

  return(dt)
}

palmoil_qty <- clean_import_data(palmoil_qty)
soybeanoil_qty <- clean_import_data(soybeanoil_qty)
sunfloweroil_qty <- clean_import_data(sunfloweroil_qty)

## Aggregate data by country and financial year
palm_agg <- palmoil_qty[Country != "Total",
                        .(Import = sum(as.numeric(ImportMonth), na.rm = TRUE)),
                        by = .(Country, Fin_Year)]

## Group countries
palm_agg[, Country_Group := ifelse(Country %in% c("INDONESIA", "MALAYSIA"),
                                   Country, "Other Countries")]

## Calculate totals and shares
palm_final <- palm_agg[, .(Import = sum(Import)), by = .(Country_Group, Fin_Year)]
palm_final[, Total := sum(Import), by = Fin_Year]
palm_final[, Share := round((Import / Total) * 100)]
palm_final <- palm_final[Fin_Year >= "2010-2011" & Fin_Year != "2024-2025"]

## Interactive palm oil plot
p1_interactive <- create_interactive_stackchart(
  data = palm_final,
  x_var = "Fin_Year",
  y_var = "Import",
  fill_var = "Country_Group",
  label_var = "Share",
  title = "Palm Oil Import by India",
  y_lab = "Quantity (Tonnes)",
  caption_text = "Data source: Ministry of Commerce, India | Analysis: Pawan | Oil Year: Nov-Oct"
)

girafe(ggobj = p1_interactive,
       width_svg = 12,
       height_svg = 8)

## %>%
##   girafe_save("palm-oil-import-interactive.html")

SOYBEAN OIL IMPORT ANALYSIS

Soybean oil has maintained a volatile but stable proportion, fluctuating between 15% and 30% over the period.
The import of soybean oil is heavily dictated by South American supply, with Argentina acting as the primary source. Argentina consistently provided over 70% of India’s soybean oil needs for much of the period. Brazil plays a significant, though secondary, role, with its share showing an upward trend in certain recent financial years.

soybean_agg <- soybeanoil_qty[Country != "Total",
                              .(Import = sum(as.numeric(ImportMonth), na.rm = TRUE)),
                              by = .(Country, Fin_Year)]

soybean_agg[, Country_Group := ifelse(Country %in% c("ARGENTINA", "BRAZIL"),
                                      Country, "Other Countries")]

soybean_final <- soybean_agg[, .(Import = sum(Import)), by = .(Country_Group, Fin_Year)]
soybean_final[, Total := sum(Import), by = Fin_Year]
soybean_final[, Share := round((Import / Total) * 100)]
soybean_final <- soybean_final[Fin_Year >= "2010-2011" & Fin_Year != "2024-2025"]

# Interactive soybean oil plot
p2_interactive <- create_interactive_stackchart(
  data = soybean_final,
  x_var = "Fin_Year",
  y_var = "Import",
  fill_var = "Country_Group",
  label_var = "Share",
  title = "Soybean Oil Import by India",
  y_lab = "Quantity (Tonnes)",
  caption_text = "Data source: Ministry of Commerce, India | Analysis: Pawan | Oil Year: Nov-Oct"
)
girafe(ggobj = p2_interactive,
        width_svg = 12,
        height_svg = 8) 

        ## %>%
                ##   girafe_save("soyabean-oil-import-interactive.html")

SUNFLOWER OIL IMPORT ANALYSIS

The most striking trend is the increasing reliance on sunflower oil. For the better part of the decade (2010-2011 to 2020-2021), sunflower oil constituted a minor share, typically between 6% and 14%. However, in the latest period (2023-2024), its share has surged to a high of 22%, indicating a significant structural shift in consumer demand and import policy favouring softer oils.

sunflower_agg <- sunfloweroil_qty[Country != "Total",
                                  .(Import = sum(as.numeric(ImportMonth), na.rm = TRUE)),
                                  by = .(Country, Fin_Year)]

sunflower_agg[, Country_Group := ifelse(Country %in% c("UKRAINE", "RUSSIA"),
                                        Country, "Other Countries")]

sunflower_final <- sunflower_agg[, .(Import = sum(Import)), by = .(Country_Group, Fin_Year)]
sunflower_final[, Total := sum(Import), by = Fin_Year]
sunflower_final[, Share := round((Import / Total) * 100)]
sunflower_final <- sunflower_final[Fin_Year >= "2010-2011" & Fin_Year != "2024-2025"]

# Interactive sunflower oil plot
p3_interactive <- create_interactive_stackchart(
  data = sunflower_final,
  x_var = "Fin_Year",
  y_var = "Import",
  fill_var = "Country_Group",
  label_var = "Share",
  title = "Sunflower Oil Import by India",
  y_lab = "Quantity (Tonnes)",
  caption_text = "Data source: Ministry of Commerce, India | Analysis: Pawan | Oil Year: Nov-Oct"
)
girafe(ggobj = p3_interactive,
       width_svg = 12,
       height_svg = 8)

## %>%
##   girafe_save("sunflower-oil-import-interactive.html")

TOTAL EDIBLE OIL IMPORTS (COMPOSITION)

# Get total imports for each oil type
total_palm <- palmoil_qty[Country == "Total",
                          .(Palmoil = sum(as.numeric(ImportMonth), na.rm = TRUE)),
                          by = Fin_Year]

total_soybean <- soybeanoil_qty[Country == "Total",
                                .(Soybeanoil = sum(as.numeric(ImportMonth), na.rm = TRUE)),
                                by = Fin_Year]

total_sunflower <- sunfloweroil_qty[Country == "Total",
                                    .(Sunfloweroil = sum(as.numeric(ImportMonth), na.rm = TRUE)),
                                    by = Fin_Year]

# Merge and reshape
total_oil <- merge(total_palm, total_soybean, by = "Fin_Year")
total_oil <- merge(total_oil, total_sunflower, by = "Fin_Year")

total_oil_long <- melt(total_oil, id.vars = "Fin_Year",
                       variable.name = "Oil_Type", value.name = "Quantity")

# Calculate shares
total_oil_long[, Total := sum(Quantity), by = Fin_Year]
total_oil_long[, Share := round((Quantity / Total) * 100)]
total_oil_long <- total_oil_long[Fin_Year >= "2010-2011" & Fin_Year != "2024-2025"]

# Create summary table
summary_table <- total_oil_long[, .(Total_EdibleOil = sum(Quantity)), by = Fin_Year]
gt(summary_table) %>%
        tab_header(title = "Total Edible Oil Imports by India") %>%
        fmt_number(columns = Total_EdibleOil, decimals = 0) %>%
        cols_label(Fin_Year = "Oil Year", Total_EdibleOil = "Total Imports (Tonnes)")

Total Edible Oil Imports by India
Oil Year	Total Imports (Tonnes)
2010-2011	7,200,810
2011-2012	9,703,219
2012-2013	10,351,294
2013-2014	11,397,056
2014-2015	13,818,424
2015-2016	14,310,720
2016-2017	14,897,076
2017-2018	14,030,226
2018-2019	15,345,150
2019-2020	13,317,101
2020-2021	13,563,657
2021-2022	13,897,542
2022-2023	16,720,932
2023-2024	16,040,413

# Interactive total oil plot
p4_interactive <- create_interactive_stackchart(
  data = total_oil_long,
  x_var = "Fin_Year",
  y_var = "Quantity",
  fill_var = "Oil_Type",
  label_var = "Share",
  title = "Edible Oil Imports Composition by India",
  y_lab = "Quantity (Tonnes)",
  caption_text = "Data source: Ministry of Commerce, India | Analysis: Pawan | Financial Year: Nov-Oct"
)
girafe(ggobj = p4_interactive,
       width_svg = 12,
       height_svg = 8)

SELF-SUFFICIENCY ANALYSIS

self_sufficiency <- fread("/home/pawan/edibleoils/oil_aval.csv")
self_sufficiency[, Domestic_Share := round(Domestic_Availability * 100 / Total_Availability_LMT)]

# Interactive self-sufficiency plot
p6_interactive <- ggplot(self_sufficiency,
                         aes(x = Year, y = Domestic_Share,
                             tooltip = paste("Year:", Year, "\n",
                                             "Self-sufficiency:", Domestic_Share, "%"))) +
  geom_col_interactive(fill = "#2E86AB", width = 0.7) +
  geom_text(aes(label = paste0(Domestic_Share, "%")),
            vjust = -0.5, size = 4, fontface = "bold") +
  my_theme +
  ylim(0, 60) +
  labs(y = "% Self Sufficiency",
       title = "India's Edible Oil Self-Sufficiency",
       caption = "Data source: Ministry of Agriculture, India | Analysis: Pawan") +
  theme(plot.caption = element_text(size = 10, color = "#666666"))

girafe(ggobj = p6_interactive,
       width_svg = 12,
       height_svg = 8)