Trade Statistics: Import Volumes and Values by Country

1 Data

財務省貿易統計 - 普通貿易統計（CSV形式の統計表）
普通貿易統計 
library(estatapi)
library(dplyr)
library(readr)
library(arrow)
library(DT)
library(ggplot2)
library(scales)
library(gghighlight)
library(ggthemes)
library(ggsci)
library(ggrepel)
library(ggtext)
library(gt)
library(maps)

outdir <- "../data/trade"
figdir <- "../fig"
dir.create(outdir, showWarnings = FALSE, recursive = TRUE)
dir.create(figdir, showWarnings = FALSE)

appId <- keyring::key_get("e-stat")

1.1 1988–2024

get_stats_data <- function(statsDataId, cdCat01) {
  estat_getStatsData(
    appId = appId,
    statsDataId = statsDataId,
    cdCat01 = cdCat01
  )
}

df_list <- list(
  get_stats_data("0003258346", "081090010"), # 1988-1990
  get_stats_data("0003258347", "081090010"), # 1991-1995
  get_stats_data("0003258348", "081050000"), # 1996-2000
  get_stats_data("0003228183", "081050000"), # 2001-2005
  get_stats_data("0003228184", "081050000"), # 2006-2010
  get_stats_data("0003228185", "081050000"), # 2011-2015
  get_stats_data("0003313966", "081050000"), # 2016-2020
  get_stats_data("0003425294", "081050000")  # 2021-2024
)

df <- do.call(rbind, df_list)

write.table(df, file = file.path(outdir, "import_volumes_and_value_by_country.csv"), sep = ",", row.names = FALSE)
write_parquet(df, sink = file.path(outdir, "import_volumes_and_value_by_country.parquet"))

1.2 Read the Data

# df <- read.csv(file.path(outdir, "import_volumes_and_value_by_country.csv"), header = TRUE, check.names = FALSE)
df <- read_parquet(file.path(outdir, "import_volumes_and_value_by_country.parquet"))
# all <- read.csv(file.path(outdir, "import_volumes_and_value_all.csv"), header = TRUE)
all <- read_parquet(file.path(outdir, "import_volumes_and_value_all.parquet"))

2 Tables

DT::datatable(
  df, rownames = FALSE, filter = 'top',
  extensions = 'Buttons',
  options = list(
    pageLength = 10,
    dom = 'Blfrtip',
    buttons = list(
      c('csv', 'excel'),
      I('colvis')
    )
  )
) |>
  DT::formatCurrency(grepl("value", names(df)), currency = "", interval = 3, mark = ",", digits = 0)

3 Figures

3.1 Import Volume

# table(df$国)
country_map <- tibble::tibble(
  original = c("アメリカ合衆国", "ニュージーランド", "チリ", "大韓民国", 
               "中華人民共和国", "イラン", "カナダ", "ジョージア"),
  english = c("United States", "New Zealand", "Chile", "South Korea", 
              "China", "Iran", "Canada", "Georgia"),
  latitude = c(37.1, -40.9, -35.7, 37.5, 35.9, 32.4, 56.1, 42.3) # 代表的な緯度
)

df_volume <- df |>
  filter(`統計品目表の数量・金額` == "合計_数量2") |>
  mutate(
    year = as.integer(gsub("年", "", `時間軸(年次)`)),
    ton = value / 1000,
    国 = sub("^\\d+_", "", 国)
  ) |>
  left_join(country_map, by = c("国" = "original")) |>
  mutate(
    country = coalesce(english, 国),
    hemisphere = factor(
      case_when(
        !is.na(latitude) & latitude > 0 ~ "Northern Hemisphere",
        !is.na(latitude) & latitude < 0 ~ "Southern Hemisphere",
        TRUE ~ NA_character_
      ),
      levels = c("Northern Hemisphere", "Southern Hemisphere")
    )
  ) |>
  select(-english, -latitude)

country_order <- df_volume |>
  summarise(total_value = sum(value, na.rm = TRUE), .by = country) |>
  arrange(total_value) |>
  pull(country)

df_volume <- df_volume |>
  mutate(
    country = factor(country, levels = country_order)
  ) |>
  filter(year < 2025)

country_list <- c("New Zealand", "Chile", "Australia",
                  "United States", "South Korea", "China",
                  "Georgia", "Iran", "Canada")
color_palette <- setNames(pal_uchicago()(length(country_list)), country_list)

p <- ggplot(df_volume, aes(x = year, y = ton, fill = country)) +
  geom_col(position = "stack") +
  scale_y_continuous(labels = scales::label_comma()) +
  labs(
    title = "Import Volumes by Year and Country",
    x = "Year",
    y = "Import Volume (t)",
    fill = "Country",
    caption = "**Source:** Ministry of Finance, _Trade Statistics_."
  ) +
  theme_bw() +
  scale_fill_uchicago() +
  theme(
    plot.caption = element_markdown()
  )

ggsave("import_volumes_by_year_and_country.pdf", plot = p, path = figdir, width = 8, height = 6)
p

p <- ggplot(df_volume, aes(x = year, y = ton, fill = country)) +
  geom_col(position = "stack") +
  scale_y_continuous(labels = scales::label_comma()) +
  labs(
#   title = "Import Volumes by Year and Country",
    x = "Year",
    y = "Import Volume (t)",
    fill = "Country",
#   caption = "**Source:** Ministry of Finance, _Trade Statistics_."
  ) +
  theme_bw() +
  facet_wrap(~ country, scales = "free_y") +
# scale_fill_uchicago() +
  scale_fill_manual(values = color_palette, drop = TRUE) +
  theme(
    legend.position = "none",
    panel.border = element_blank(),
    axis.line.x = element_line(linewidth = .25),
    axis.line.y = element_line(linewidth = .25),
    strip.background = element_rect(fill = NA, color = NA),
    strip.text = element_text(face = "bold"),
    text = element_text(family = "Helvetica")
  )

ggsave("figure1.pdf", plot = p, path = figdir, width = 8, height = 6)
ggsave("figure1.png", plot = p, path = figdir, width = 8, height = 6, dpi = 600)
p

p <- ggplot(df_volume, aes(x = year, y = ton, color = country, group = country)) +
  geom_line() +
  scale_y_continuous(labels = label_comma()) +
  labs(
    title = "Import Volumes by Year and Hemisphere",
    x = "Year",
    y = "Import Volume (t)",
    color = "Country",
    caption = "**Source:** Ministry of Finance, _Trade Statistics_."
  ) +
  theme_bw() +
  scale_color_uchicago() +
  theme(
    plot.caption = element_markdown()
  ) +
  facet_wrap(~ hemisphere, scales = "free_y", ncol = 1)
# gghighlight()

ggsave("import_volumes_by_year_and_country_hemisphere.pdf", plot = p, path = figdir, width = 8, height = 6)
p

3.1.1 Map

3.1.1.1 2024

df_volume_2024 <- df_volume |>
  filter(year == 2024) |>
  bind_rows(
    data.frame(
      country = "Others",
      value = all[all$year == 2024, "quantity2"] / 1000 - sum(df_volume$ton[df_volume$year == 2024])
    )
  ) |>
  mutate(
    country = factor(country, levels = c(country_order, "Others")),
    percentage = round(value / sum(value) * 100, 2)
  )

gt(df_volume_2024)
cat01_code 統計品目表(輸入) cat02_code 統計品目表の数量・金額 area_code 国 time_code 時間軸(年次) unit value annotation year ton country hemisphere quantity2 percentage
081050000 081050000 130 合計_数量2 50103 大韓民国 2024000000 2024年 NA 310340 NA 2024 310.340 South Korea Northern Hemisphere NA NA
081050000 081050000 130 合計_数量2 50157 ジョージア 2024000000 2024年 NA 264516 NA 2024 264.516 Georgia Northern Hemisphere NA NA
081050000 081050000 130 合計_数量2 50304 アメリカ合衆国 2024000000 2024年 NA 1128952 NA 2024 1128.952 United States Northern Hemisphere NA NA
081050000 081050000 130 合計_数量2 50409 チリ 2024000000 2024年 NA 1106464 NA 2024 1106.464 Chile Southern Hemisphere NA NA
081050000 081050000 130 合計_数量2 50606 ニュージーランド 2024000000 2024年 NA 112719022 NA 2024 112719.022 New Zealand Southern Hemisphere NA NA
NA NA NA NA NA NA NA NA NA NA NA NA NA Others NA 0 NA 
map_volume_2024 <- map_data("world") |>
  left_join(
    df_volume_2024 |>
      mutate(
        country = case_when(
          country == "United States" ~ "USA",
          TRUE ~ country
        )
      ),
    by = c("region" = "country")
  )

p <- ggplot(map_volume_2024, aes(x = long, y = lat, group = group, fill = region)) +
  geom_polygon() +
  gghighlight(ton > 0) +
  labs(title = "Import Volumes by Country (2024)", fill = "Volume (ton)") +
  theme_void() +
  scale_fill_nejm() +
  theme(
    legend.position = "none",
    text = element_text(family = "Helvetica")
  )

ggsave("import_volumes_by_year_and_country_map_2024.pdf", plot = p, path = figdir, width = 8, height = 6)
p

3.1.1.2 2015

df_volume_2015 <- df_volume |>
  filter(year == 2015) |>
  bind_rows(
    data.frame(
      country = "Others",
      value = all[all$year == 2015, "quantity2"] / 1000 - sum(df_volume$ton[df_volume$year == 2015])
    )
  ) |>
  mutate(
    country = factor(country, levels = c(country_order, "Others")),
    percentage = round(value / sum(value) * 100, 2)
  )

gt(df_volume_2015)
cat01_code 統計品目表(輸入) cat02_code 統計品目表の数量・金額 area_code 国 time_code 時間軸(年次) unit value annotation year ton country hemisphere quantity2 percentage
081050000 081050000 130 合計_数量2 50103 大韓民国 2015000000 2015年 NA 59519 NA 2015 59.519 South Korea Northern Hemisphere NA NA
081050000 081050000 130 合計_数量2 50304 アメリカ合衆国 2015000000 2015年 NA 342344 NA 2015 342.344 United States Northern Hemisphere NA NA
081050000 081050000 130 合計_数量2 50409 チリ 2015000000 2015年 NA 1835615 NA 2015 1835.615 Chile Southern Hemisphere NA NA
081050000 081050000 130 合計_数量2 50606 ニュージーランド 2015000000 2015年 NA 76410521 NA 2015 76410.521 New Zealand Southern Hemisphere NA NA
NA NA NA NA NA NA NA NA NA NA NA NA NA Others NA 0 NA 
map_volume_2015 <- map_data("world") |>
  left_join(
    df_volume_2015 |>
      mutate(
        country = case_when(
          country == "United States" ~ "USA",
          TRUE ~ country
        )
      ),
    by = c("region" = "country")
  )

p <- ggplot(map_volume_2015, aes(x = long, y = lat, group = group, fill = region)) +
  geom_polygon() +
  gghighlight(ton > 0) +
  labs(title = "Import Volumes by Country (2015)", fill = "Volume (ton)") +
  theme_void() +
  scale_fill_nejm() +
  theme(
    legend.position = "none",
    text = element_text(family = "Helvetica")
  )

ggsave("import_volumes_by_year_and_country_map_2015.pdf", plot = p, path = figdir, width = 8, height = 6)
p

3.1.1.3 2005

df_volume_2005 <- df_volume |>
  filter(year == 2005) |>
  bind_rows(
    data.frame(
      country = "Others",
      value = all[all$year == 2005, "quantity2"] / 1000 - sum(df_volume$ton[df_volume$year == 2005])
    )
  ) |>
  mutate(
    country = factor(country, levels = c(country_order, "Others")),
    percentage = round(value / sum(value) * 100, 2)
  )

gt(df_volume_2005)
cat01_code 統計品目表(輸入) cat02_code 統計品目表の数量・金額 area_code 国 time_code 時間軸(年次) unit value annotation year ton country hemisphere quantity2 percentage
081050000 081050000 130 合計_数量2 50105 中華人民共和国 2005000000 2005年 NA 320513 NA 2005 320.513 China Northern Hemisphere NA NA
081050000 081050000 130 合計_数量2 50304 アメリカ合衆国 2005000000 2005年 NA 595506 NA 2005 595.506 United States Northern Hemisphere NA NA
081050000 081050000 130 合計_数量2 50409 チリ 2005000000 2005年 NA 4172253 NA 2005 4172.253 Chile Southern Hemisphere NA NA
081050000 081050000 130 合計_数量2 50606 ニュージーランド 2005000000 2005年 NA 54346610 NA 2005 54346.610 New Zealand Southern Hemisphere NA NA
NA NA NA NA NA NA NA NA NA NA NA NA NA Others NA 0 NA 
map_volume_2005 <- map_data("world") |>
  left_join(
    df_volume_2005 |>
      mutate(
        country = case_when(
          country == "United States" ~ "USA",
          TRUE ~ country
        )
      ),
    by = c("region" = "country")
  )

p <- ggplot(map_volume_2005, aes(x = long, y = lat, group = group, fill = region)) +
  geom_polygon() +
  gghighlight(ton > 0) +
  labs(title = "Import Volumes by Country (2005)", fill = "Volume (ton)") +
  theme_void() +
  scale_fill_nejm() +
  theme(
    legend.position = "none",
    text = element_text(family = "Helvetica")
  )

ggsave("import_volumes_by_year_and_country_map_2005.pdf", plot = p, path = figdir, width = 8, height = 6)
p

3.1.2 Monthly

df_monthly <- df |>
  filter(
    grepl("数量2$", `統計品目表の数量・金額`),
    !grepl("合計", `統計品目表の数量・金額`)
  ) |>
  mutate(
    year = parse_number(`時間軸(年次)`),
    month = as.integer(sub("月_数量2", "", `統計品目表の数量・金額`)),
    ton = value / 1000,
    国 = sub("^\\d+_", "", 国)
  ) |>
  left_join(country_map, by = c("国" = "original")) |>
  mutate(
    country = coalesce(english, 国),
    hemisphere = factor(
      case_when(
        !is.na(latitude) & latitude > 0 ~ "Northern Hemisphere",
        !is.na(latitude) & latitude < 0 ~ "Southern Hemisphere",
        TRUE ~ NA_character_
      ),
      levels = c("Northern Hemisphere", "Southern Hemisphere")
    )
  ) |>
  select(-english, -latitude)

df_monthly <- df_monthly |>
  mutate(
    country = factor(country, levels = country_order)
  )

p <- ggplot(df_monthly, aes(x = month, y = ton, group = interaction(year, country), color = country)) +
  geom_line(alpha = 0.3, linewidth = 0.5) +
  scale_x_continuous(breaks = 1:12, labels = month.abb) +
  scale_y_continuous(labels = scales::label_comma()) +
  labs(
    title = "Monthly Import Volumes by Hemisphere and Country (1988-2024)",
    x = "Month",
    y = "Import Volume (t)",
    color = "Country",
    caption = "**Source:** Ministry of Finance, _Trade Statistics_."
  ) +
  theme_bw() +
  scale_color_uchicago() +
  facet_wrap(~ hemisphere, scales = "free_y", ncol = 1) +
  theme(
    strip.background = element_rect(fill = "gray90"),
    strip.text = element_text(face = "bold"),
    legend.position = "right",
    plot.caption = element_markdown()
  )

ggsave("monthly_import_volumes_by_hemisphere.pdf", plot = p, path = figdir, width = 8, height = 6)
p

label_data <- df_monthly |>
  filter(ton == max(ton), .by = country)

p <- ggplot(df_monthly, aes(x = month, y = ton, group = interaction(year, country), color = country)) +
  geom_line(alpha = .4, linewidth = 0.5) +
  scale_x_continuous(breaks = 1:12, labels = month.abb) +
  scale_y_continuous(labels = scales::label_comma()) +
  geom_text_repel(data = label_data, aes(label = country),
                  size = 3,
                  color = "black",
                  family = "Helvetica",
                  segment.color = "black",
                  point.padding = 1,
                  # United Sates, Chile, Iran, Canada, China, South Korea, New Zealand, Georgia
                  nudge_x = c(1.3, -3.5, 1.5, -.5, -.5, 2.8, 2, 3.2),
                  nudge_y = c(100, 6000, 300, 500, 150, 300, 400, 1000),
                  segment.curvature = c(.2, 0, 0, 0, 0, 0, .2, .1),
#                 segment.curvature = 0.2,
                  segment.ncp = 3) +
  labs(
#   title = "Monthly Import Volumes by Hemisphere and Country (1988-2024)",
    x = "Month",
    y = "Import Volume (t)",
    color = "Country",
#   caption = "Source: Ministry of Finance 'Customs Statistics'"
  ) +
  theme_bw() +
# scale_color_uchicago() +
  scale_color_manual(values = color_palette, drop = TRUE) +
  facet_wrap(~ hemisphere, scales = "free_y", ncol = 1) +
  theme(
    legend.position = "none",
    panel.border = element_blank(),
    axis.line.x = element_line(linewidth = .25),
    axis.line.y = element_line(linewidth = .25),
    strip.background = element_rect(fill = NA, color = NA),
    strip.text = element_text(face = "bold"),
    text = element_text(family = "Helvetica")
  )

ggsave("figure2.pdf", plot = p, path = figdir, width = 8, height = 6)
ggsave("figure2.png", plot = p, path = figdir, width = 8, height = 6, dpi = 600)
p

3.2 Import Value

df_value <- df |>
  filter(`統計品目表の数量・金額` == "合計_金額") |>
  mutate(
    year = as.integer(gsub("年", "", `時間軸(年次)`)),
    ton = value / 1000,
    国 = sub("^\\d+_", "", 国)
  ) |>
  left_join(country_map, by = c("国" = "original")) |>
  mutate(
    country = coalesce(english, 国),
    hemisphere = factor(
      case_when(
        !is.na(latitude) & latitude > 0 ~ "Northern Hemisphere",
        !is.na(latitude) & latitude < 0 ~ "Southern Hemisphere",
        TRUE ~ NA_character_
      ),
      levels = c("Northern Hemisphere", "Southern Hemisphere")
    )
  ) |>
  select(-english, -latitude)

country_order <- df_value |>
  summarise(total_value = sum(value, na.rm = TRUE), .by = country) |>
  arrange(total_value) |>
  pull(country)

df_value <- df_value |>
  mutate(
    country = factor(country, levels = country_order)
  )

p <- ggplot(df_value, aes(x = year, y = value, color = country)) +
  geom_line() +
  scale_y_continuous(labels = scales::label_comma()) +
  labs(
    title = "Import Value by Year and Country",
    x = "Year",
    y = "Import Value (thousand JPY)",
    color = "Country",
    caption = "**Source:** Ministry of Finance, _Trade Statistics_."
  ) +
  theme_bw() +
  scale_color_lancet() +
  facet_wrap(~ hemisphere, scales = "free_y", ncol = 1) +
  theme(
    plot.caption = element_markdown()
  )
# gghighlight()

ggsave("import_value_by_year_and_country.pdf", plot = p, path = figdir, width = 8, height = 6)
p

3.2.1 Monthly

df_value_monthly <- df |>
  filter(grepl("金額$", `統計品目表の数量・金額`)) |>
  filter(!grepl("合計", `統計品目表の数量・金額`)) |>
  mutate(
    year = parse_number(`時間軸(年次)`),
    month = as.integer(sub("月_金額", "", `統計品目表の数量・金額`)),
    ton = value / 1000,
    国 = sub("^\\d+_", "", 国)
  ) |>
  left_join(country_map, by = c("国" = "original")) |>
  mutate(
    country = coalesce(english, 国),
    hemisphere = factor(
      case_when(
        !is.na(latitude) & latitude > 0 ~ "Northern Hemisphere",
        !is.na(latitude) & latitude < 0 ~ "Southern Hemisphere",
        TRUE ~ NA_character_
      ),
      levels = c("Northern Hemisphere", "Southern Hemisphere")
    )
  ) |>
  select(-english, -latitude)

df_value_monthly <- df_value_monthly |>
  mutate(
    country = factor(country, levels = country_order)
  )

p <- ggplot(df_value_monthly, aes(x = month, y = ton, group = interaction(year, country), color = country)) +
  geom_line(alpha = 0.3, linewidth = 0.5) +
  scale_x_continuous(breaks = 1:12, labels = month.abb) +
  scale_y_continuous(labels = scales::label_comma()) +
  labs(
    title = "Monthly Import Value by Hemisphere and Country (1988-2024)",
    x = "Month",
    y = "Import Value (thousand JPY)",
    color = "Country",
    caption = "**Source:** Ministry of Finance, _Trade Statistics_."
  ) +
  theme_bw() +
  scale_color_npg() +
  facet_wrap(~ hemisphere, scales = "free_y", ncol = 1) +
  theme(
    strip.background = element_rect(fill = "gray90"),
    strip.text = element_text(face = "bold"),
    legend.position = "right",
    plot.caption = element_markdown()
  )

ggsave("monthly_import_value_by_hemisphere.pdf", plot = p, path = figdir, width = 8, height = 6)
p