Redistribution Effect: How Taxes Transform Inequality

Dumbbell chart with {ggplot2} techniques: gradient-colored segments, custom legends, and rich text formatting for income inequality visualization.

#TidyTuesday
Dumbbell Chart
Author

Aditya Dahiya

Published

August 6, 2025

About the Data

This analysis explores income inequality before and after taxes using data processed and visualized by Joe Hasell at Our World in Data. The dataset combines information from multiple authoritative sources including the Luxembourg Income Study (2025), OECD’s Income and Wealth Distribution Database (2024), and population data from UN World Population Prospects (2024) and Gapminder. The core metric used is the Gini coefficient, which measures inequality on a scale from 0 to 1, where higher values indicate greater inequality. The data includes both pre-tax income (market income before taxes and most government benefits) and post-tax income (disposable income after taxes and benefits), with income figures equivalized to account for household cost-sharing. This comprehensive dataset enables examination of how different countries redistribute income through their tax and benefit systems, providing insights into the effectiveness of various redistributive policies across nations and over time.

Figure 1: This dumbbell chart compares income inequality before and after government intervention across countries using their most recent available data (years shown in parentheses). Red circles represent pre-tax Gini coefficients, blue circles show post-tax values, and the gradient segments illustrate the transition. Black percentages indicate how much inequality each government reduces through taxes and transfers, revealing dramatic differences in redistributive effectiveness—from minimal impact in some nations to substantial reductions exceeding 40% in others.

How the Graphic Was Created

This advanced dumbbell chart was created using R’s ggplot2 ecosystem with several sophisticated techniques to enhance visual storytelling. The gradient-colored segments were achieved by breaking each line into multiple small segments using dplyr’s rowwise() and do() functions, then mapping colors based on Gini coefficient values along a common scale from 0.22 to 0.75. The visualization required ggnewscale to implement multiple color scales—one for the gradient segments and another for the custom point legend. Rich text formatting for the value labels was accomplished using ggtext’s geom_richtext() function, allowing HTML-styled text with colored numbers and arrows. Country flags within the data points were added using ggflags package’s geom_flag() function. The custom legend was created by plotting invisible dummy points with scale_colour_manual() and guide_legend(), while hiding the gradient legend with guide = "none". Font management was handled through showtext and sysfonts packages, with Google Fonts integration for consistent typography throughout the visualization.

Loading required libraries

Code 
pacman::p_load(
  tidyverse,            # All things tidy
  
  scales,               # Nice Scales for ggplot2
  fontawesome,          # Icons display in ggplot2
  ggtext,               # Markdown text support for ggplot2
  showtext,             # Display fonts in ggplot2
  colorspace,           # Lighten and Darken colours

  patchwork,            # Composing Plots
  ggridges,             # Making ridgeline plot 
  gghalves              # Half geoms in ggplot2
)

# Option 2: Read directly from GitHub

income_inequality_processed <- readr::read_csv('https://raw.githubusercontent.com/rfordatascience/tidytuesday/main/data/2025/2025-08-05/income_inequality_processed.csv') |> 
  janitor::clean_names()

# Not using the raw data in my graphic here
# income_inequality_raw <- readr::read_csv('https://raw.githubusercontent.com/rfordatascience/tidytuesday/main/data/2025/2025-08-05/income_inequality_raw.csv')

Visualization Parameters

Code 
# Font for titles
font_add_google("Saira",
  family = "title_font"
) 

# Font for the caption
font_add_google("Saira Condensed",
  family = "body_font"
) 

# Font for plot text
font_add_google("Saira Extra Condensed",
  family = "caption_font"
) 

showtext_auto()

# A base Colour
bg_col <- "white"
seecolor::print_color(bg_col)

# Colour for highlighted text
text_hil <- "grey20"
seecolor::print_color(text_hil)

# Colour for the text
text_col <- "grey20"
seecolor::print_color(text_col)

line_col <- "grey30"

# Custom Colours for dots
custom_dot_colours <- paletteer::paletteer_d("nbapalettes::cavaliers_retro")

# Custom size for dots
size_var <- 14

# Define Base Text Size
bts <- 80

# Caption stuff for the plot
sysfonts::font_add(
  family = "Font Awesome 6 Brands",
  regular = here::here("docs", "Font Awesome 6 Brands-Regular-400.otf")
)
github <- "&#xf09b"
github_username <- "aditya-dahiya"
xtwitter <- "&#xe61b"
xtwitter_username <- "@adityadahiyaias"
social_caption_1 <- glue::glue("<span style='font-family:\"Font Awesome 6 Brands\";'>{github};</span> <span style='color: {text_hil}'>{github_username}  </span>")
social_caption_2 <- glue::glue("<span style='font-family:\"Font Awesome 6 Brands\";'>{xtwitter};</span> <span style='color: {text_hil}'>{xtwitter_username}</span>")
plot_caption <- paste0(
  "**Data:** Joe Hasell at Our World in Data", 
  " |  **Code:** ", 
  social_caption_1, 
  " |  **Graphics:** ", 
  social_caption_2
  )
rm(github, github_username, xtwitter, 
   xtwitter_username, social_caption_1, 
   social_caption_2)

# Add text to plot-------------------------------------------------
plot_title <- "Redistribution Effect: How Taxes Transform Inequality"

plot_subtitle <- "Red dots represent market-driven inequality (pre-tax Gini coefficients), blue dots show government-adjusted outcomes (post-tax), with percentage reductions revealing redistributive effectiveness. Countries are ranked in descending order of their success in narrowing income gaps through taxes and social programs." |> str_wrap(110)

str_view(plot_subtitle)

Exploratory Data Analysis and Wrangling

Code 
# Check which year has the most countries
income_inequality_processed |> 
  drop_na() |> 
  count(year, sort = T) |> 
  arrange(desc(year))


# ============================================================================
# DATA PREPARATION AND TRANSFORMATION
# ============================================================================
plot_data <- income_inequality_processed |> 
  # Remove rows with missing values in any column
  drop_na() |> 
  # Group countries to get most recent year data for each
  group_by(entity) |> 
  slice_max(order_by = year, n = 1) |>  # Keep only the latest year per country
  ungroup() |>  # Remove grouping for subsequent operations
  
  # Calculate improvement metrics for visualization
  mutate(
    
    # Calculate proportional improvement (reduction in inequality)
    improvement = (gini_mi_eq - gini_dhi_eq)/gini_mi_eq,
    
    # Convert to percentage and round for clean display
    percentage_reduction = round(improvement * 100, 1)
  ) |> 
  
  # Add country codes for plotting flags
  mutate(
    country_code = str_to_lower(
      countrycode::countrycode(code, "iso3c", "iso2c")
    )
  )



# Create a common gradient data set for colouring the line segments
gradient_data <- plot_data |>
      rowwise() |>
      do({
        n_segments <- 20  # Number of small segments per line
        x_vals <- seq(.$gini_mi_eq, .$gini_dhi_eq, length.out = n_segments + 1)
        # Use actual Gini values for common color scale (0.22 to 0.75)
        color_vals <- x_vals[-length(x_vals)]  # Use x-position for color mapping
        
        data.frame(
          entity = .$entity,
          x = x_vals[-length(x_vals)],
          xend = x_vals[-1],
          color_position = color_vals,  # Color based on actual Gini values
          improvement = .$improvement,
          gini_mi_eq = .$gini_mi_eq,
          gini_dhi_eq = .$gini_dhi_eq,
          percentage_reduction = .$percentage_reduction
        )
      }) |>
      ungroup()

# A small custom dataframe for making a custom legend
legend_df <- tibble(
  entity = c("Canada", "Lithuania"),
  gini_value = c(0.685),
  type = c("Pre-tax\nGINI", "Post-tax\nGINI"),
  dot_colours = custom_dot_colours
)

The Plot

Code 
# CREATE THE DUMBBELL CHART
g <- plot_data |> 
  
  # Initialize plot with countries ordered by improvement on y-axis
  ggplot(
    mapping = aes(
      y = reorder(entity, improvement)  # Order countries by redistribution effectiveness
    )
  ) +
  
  # ---------------------------------------------------------------------
  # DUMBBELL SEGMENTS (connecting lines with gradient along each segment)
  # ---------------------------------------------------------------------
  # Plot the gradient segments
  geom_segment(
    data = gradient_data,
    mapping = aes(
      x = x,
      xend = xend,
      
      # Color varies based on Gini coefficient values
      colour = color_position  
    ),
    linewidth = 3.5,
    lineend = "round"
  ) +
  paletteer::scale_colour_paletteer_c("pals::kovesi.rainbow_bgyrm_35_85_c71") +
  # paletteer::scale_colour_paletteer_c("viridis::turbo") +
  ggnewscale::new_scale_colour() +

  # ---------------------------------------------------------------------
  # ADD COUNTRY FLAGS INSIDE THE DOTS (which we will plot later)
  # ---------------------------------------------------------------------
  ggflags::geom_flag(
    mapping = aes(
      x = gini_mi_eq,
      country = country_code
    ),
    size = size_var * 0.9
  ) +
  ggflags::geom_flag(
    mapping = aes(
      x = gini_dhi_eq,
      country = country_code
    ),
    size = size_var * 0.9
  ) +

  # --------------------------------------------------------------------
  # DUMBBELL POINTS (pre-tax and post-tax markers)
  # --------------------------------------------------------------------
 
  # Pre-tax inequality points (red)
  geom_point(
    mapping = aes(
      x = gini_mi_eq,  # Position at pre-tax Gini value
    ),
    colour = custom_dot_colours[1],    
    size = size_var, 
    pch = 21, 
    stroke = 4,
    fill = NA
  ) +
  # Post-tax inequality points (blue)
  geom_point(
    mapping = aes(
      x = gini_dhi_eq, # Position at post-tax Gini value
      y = reorder(entity, improvement)
    ),
    colour = custom_dot_colours[2],    
    size = size_var, 
    pch = 21, 
    stroke = 4,
    fill = NA
  ) +
  
  # -------------------------------------------------------------------------
  # COUNTRY NAME LABELS
  # -------------------------------------------------------------------------
  geom_text(
    mapping = aes(
      x = 0.58,         
      label = paste0(entity, " (", year, ")"),   
      y = reorder(entity, improvement),
      
      # Special alignment for South Africa (right-aligned), others left-aligned
      hjust = ifelse(entity == "South Africa", 0.8, 0)
    ),
    
    size = bts / 2.8,        
    colour = text_hil,
    family = "body_font"
  ) +
  
  # -------------------------------------------------------------------------
  # RICH TEXT LABELS (values with colored formatting)
  # -------------------------------------------------------------------------
  geom_richtext(
    mapping = aes(
      x = (gini_mi_eq + gini_dhi_eq) / 2,  # Center labels between points

      # Create multi-line HTML formatted text
      label = paste0(
        "<span style='color:",custom_dot_colours[2], "'>", 
        round(gini_dhi_eq, 2), "</span>  ← ",    # Pre-tax in red
        "<span style='color:",custom_dot_colours[1], "'>", 
        round(gini_mi_eq, 2), "</span><br>",   # Post-tax in blue
        "<span style='color:grey10'>", 
        percentage_reduction, "%</span>"      
      )
    ),
    fill = NA,
    label.color = NA,   
    size = bts / 6, 
    vjust = 0.5,  
    label.padding = unit(0.05, "lines"),
    lineheight = 0.5,
    family = "title_font"
  ) +
  
  # -------------------------------------------------------------------------
  # SCALES + COLOR SCALE + LABELS
  # -------------------------------------------------------------------------
  guides(colour = "none") +
  
  # Creating a Custom Legend ------------------------------------------------
  geom_point(
    data = legend_df,
    mapping = aes(
      x = gini_value,
      y = entity,
      colour = dot_colours
    ),
    size = size_var,
    pch = 21,
    fill = NA,
    stroke = 4
  ) +
  geom_text(
    data = legend_df,
    mapping = aes(
      x = gini_value,
      y = entity,
      colour = dot_colours,
      label = type
    ),
    size = bts / 2,
    vjust = 0.5,
    hjust = 0,
    nudge_x = 0.02,
    family = "caption_font",
    lineheight = 0.3,
    colour = text_col
  ) +
  scale_colour_identity() +
  
  
  scale_x_continuous(
    limits = c(0.22, 0.75),
    expand = expansion(0)
  ) +
  labs(
    title = plot_title,
    subtitle = plot_subtitle,
    x = "Gini Coefficient (0 = Perfect Equality, 1 = Maximum Inequality)",
    y = NULL,
    caption = plot_caption
  ) +
  theme_minimal(
    base_family = "body_font",
    base_size = bts
  ) +
  theme(
    
    # Legend: Not needed as colour does not convey any additional information
    legend.position = "none",
    
    # Overall
    text = element_text(
      margin = margin(0,0,0,0, "mm"),
      colour = text_col,
      lineheight = 0.3
    ),
    
    # Labels and Strip Text
    plot.title = element_text(
      margin = margin(5,0,5,0, "mm"),
      hjust = 0.5,
      vjust = 0.5,
      colour = text_hil,
      size = 2 * bts,
      family = "body_font",
      face = "bold"
      ),
    plot.subtitle = element_text(
      margin = margin(2,0,5,0, "mm"),
      vjust = 0.5,
      colour = text_hil,
      size =  bts,
      hjust = 0.5,
      family = "body_font",
      lineheight = 0.3
      ),
    plot.caption = element_textbox(
      margin = margin(8,0,5,0, "mm"),
      hjust = 0.5,
      halign = 0.5,
      colour = text_hil,
      size = bts * 0.8,
      family = "caption_font"
    ),
    plot.caption.position = "plot",
    plot.title.position = "plot",
    plot.margin = margin(5,5,5,5, "mm"),
    panel.grid.minor.y = element_blank(),
    panel.grid.major.y = element_blank(),
    panel.grid.minor.x = element_blank(),
    panel.grid.major.x = element_line(
      colour = alpha(text_hil, 0.7),
      linewidth = 0.3,
      linetype = 3
    ),
    
    # Axes Lines, Ticks, Text and labels
    axis.line.x = element_line(
      colour = text_hil,
      linewidth = 0.3,
      arrow = arrow(length = unit(5, "mm"))
    ),
    axis.line.y = element_blank(),
    axis.ticks.x = element_line(
      colour = text_hil,
      linewidth = 0.3,
    ),
    axis.ticks.length = unit(5, "mm"),
    axis.ticks.length.y = unit(0, "mm"),
    axis.ticks.y = element_blank(),
    axis.text.x = element_text(
      margin = margin(1,1,1,1, "mm"),
      colour = text_hil
    ),
    axis.text.y = element_blank(),
    axis.title.y = element_blank(),
    axis.title.x = element_text(
      colour = text_hil,
      margin = margin(2,0,0,0, "mm"),
      size = bts
    )
    
  )

ggsave(
  filename = here::here(
    "data_vizs",
    "tidy_income_inequality.png"
  ),
  plot = g,
  width = 400,
  height = 500,
  units = "mm",
  bg = bg_col
)

Savings the thumbnail for the webpage

Code 
# Saving a thumbnail

library(magick)

# Saving a thumbnail for the webpage
image_read(here::here("data_vizs", 
                      "tidy_income_inequality.png")) |> 
  image_resize(geometry = "x400") |> 
  image_write(
    here::here(
      "data_vizs", 
      "thumbnails", 
      "tidy_income_inequality.png"
    )
  )

Session Info

Code 
pacman::p_load(
  tidyverse,            # All things tidy
  
  scales,               # Nice Scales for ggplot2
  fontawesome,          # Icons display in ggplot2
  ggtext,               # Markdown text support for ggplot2
  showtext,             # Display fonts in ggplot2
  colorspace,           # Lighten and Darken colours

  patchwork             # Composing Plots
)

sessioninfo::session_info()$packages |> 
  as_tibble() |> 
  dplyr::select(package, 
         version = loadedversion, 
         date, source) |> 
  dplyr::arrange(package) |> 
  janitor::clean_names(
    case = "title"
  ) |> 
  gt::gt() |> 
  gt::opt_interactive(
    use_search = TRUE
  ) |> 
  gtExtras::gt_theme_espn()
Table 1: R Packages and their versions used in the creation of this page and graphics

Links