Since the foundation of the Republic of Korea in 1948, the economy has been one of the two pillars of Korean diplomacy, alongside security.¹ Indeed, the country’s trade has virtually always been over 50% of its GDP since 1970s, reaching over 100% in 2010s (see Figure 1).

¹ Korea added “public diplomacy” as a third pillar of diplomacy in 2010.

Show the code

##| label: fig-trade_gdp
##| fig-cap: Trade as % of GDP in Korea

trade_gdp <- WDI::WDI(indicator = c("trade_gdp" = "NE.TRD.GNFS.ZS",
                                    "gdp_growth" = "NY.GDP.MKTP.KD.ZG"),
                      country = "KOR",
                      start = 1960, end = 2022)

trade_gdp %>% ggplot(aes(year, trade_gdp)) +
  geom_line(color = "#2780e3") +
  scale_x_continuous(breaks = seq(1960, 2022, 5)) +
  # add percentages to y axis
  scale_y_continuous(labels = scales::percent_format(scale = 1)) +
  labs(#title = "Trade as % of GDP in Korea",
       x = "Year",
       y = "Trade as % of GDP") +
  ggthemes::theme_pander() +
  theme(axis.text.x = element_text(angle = 45, hjust = 1))

Therefore, Korea’s trade constitute an important data source for studies on Korean diplomacy. In this blog post, I will explore the trade data in the kdiplo R package. There are two main sources for the data.

Korean Statistical Information Service (KOSIS) which has Korea’s trade data from 1965 until 2023 with countries and regions.
Correlates of War (COW)’s Trade Data v4.0 which Korea’s trade data from 1949 until 2014.

In the trade data in the kdiplo package, I share both sets of data (export_kosis and import_kosis referring to export and import data from KOSIS; export_cow and import_cow referring to export and import data from COW) as well as mutated export and import variables that rely mainly on KOSIS data, but imputing it with the COW data where the former is missing. While in my own research, I use other ways to impute the still missing data, in the kdiplo package, I don’t do that, leaving it to the users’ discretion.

Let’s explore Korea’s trade data in the kdiplo package.

I have not yet created a function to get the data from the package, but one can simply get the data and assign it to an object. This is what the dataset looks like:

Show the code

trade <- kdiplo::trade_data

glimpse(trade)

Rows: 16,511
Columns: 18
$ iso3c                  <chr> "ABW", "ABW", "ABW", "ABW", "ABW", "ABW", "ABW"…
$ country                <chr> "Aruba", "Aruba", "Aruba", "Aruba", "Aruba", "A…
$ year                   <dbl> 1965, 1966, 1967, 1968, 1969, 1970, 1971, 1972,…
$ export                 <dbl> NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA,…
$ import                 <dbl> NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA,…
$ total_export           <dbl> 175082000, 250334000, 320229000, 455400000, 622…
$ total_import           <dbl> 463442000, 716441000, 996246000, 1462873000, 18…
$ export_kosis           <dbl> NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA,…
$ import_kosis           <dbl> NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA,…
$ export_cow             <dbl> NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA,…
$ import_cow             <dbl> NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA,…
$ index                  <dbl> 52.9, 54.4, 56.1, 58.3, 60.9, 63.9, 66.7, 68.7,…
$ cpi                    <dbl> 0.1516194, 0.1559186, 0.1607911, 0.1670966, 0.1…
$ export_cons_2015       <dbl> NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA,…
$ import_cons_2015       <dbl> NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA,…
$ total_export_cons_2015 <dbl> 1154746877, 1605542879, 1991584636, 2725369811,…
$ total_import_cons_2015 <dbl> 3056614628, 4594968105, 6195904267, 8754655055,…
$ updated_at             <date> 2024-04-27, 2024-04-27, 2024-04-27, 2024-04-27…

Trends in Korea’s Trade Volume over Time

Let’s mutate the trade volume as the sum of exports and imports, and the total trade volume as the sum of total exports and total imports.

Show the code

trade %<>% mutate(trade = export + import,
                  total_trade = total_export + total_import)

Let’s visualize the trade volume over time.

Show the code

trade_years <- trade %>% group_by(year) %>%
  summarise(total_trade_group_by = sum(trade, na.rm = T),
            total_export_group_by = sum(export, na.rm = T),
            total_import_group_by = sum(import, na.rm = T))

trade %<>% left_join(trade_years, by = "year")

trade %>% ggplot(aes(year, total_trade_group_by/1e9)) +
  geom_line(color = "#2780e3") +
  #geom_line(aes(year, total_export_group_by/1e9), color = "#CF4446") +
  #geom_line(aes(year, total_import_group_by/1e9), color = "black") +
  scale_y_continuous(breaks = seq(100, 1500, 200), labels = scales::comma) +
  scale_x_continuous(breaks = seq(1948, 2023, 5)) +
  labs(#title = "Trade Volume of Korea (in billion $)",
       x = "Year",
       y = "Trade Volume") +
  ggthemes::theme_pander() +
  theme(axis.text.x = element_text(angle = 45, hjust = 1)) +
  geom_vline(xintercept = 1962, linetype = "dashed", color = "#CF4446") +
  annotate("text", x = 1962, y = 200, label = "Park Chung-hee's First\nFive-Year Economic Plan", color = "black") +
  geom_vline(xintercept = 1988, linetype = "dashed", color = "#CF4446") +
  annotate("text", x = 1988, y = 500, label = "Democratization", color = "black")

Figure 2: Korea’s Trade Volume (in billion $)

There is this myth that Korea’s economic development was all thanks to Park Chung-hee’s economic development plans which built on export-led industrialization. Although we cannot deny his influence on Korea’s industrialization and economic growth (see Figure 3), but Korea’s trade volume has grown exponentially since the 1990s following the country’s liberalization and democratization.

Show the code

##| label: fig-gdp_growth
##| fig-cap: "Korea's GDP Growth Rate"

trade_gdp %>% ggplot(aes(year, gdp_growth)) +
  geom_line(color = "#2780e3") +
  geom_smooth() +
  scale_x_continuous(breaks = seq(1960, 2022, 5)) +
  # add percentages to y axis
  scale_y_continuous(labels = scales::percent_format(scale = 1)) +
  labs(#title = "Trade as % of GDP in Korea",
       x = "Year",
       y = "Korea's GDP Growth Rate") +
  ggthemes::theme_pander() +
  theme(axis.text.x = element_text(angle = 45, hjust = 1))

Trade Dependence

We live at an age of increasing great power rivalry in the Indo-Pacific region, even this relatively new geopolitical naming of the region causing controversies with China. For quite a long time, region’s two great powers, China and Japan, outsourced great power management to the United States and ASEAN. However, in recent years with growing assertiveness of Chinese foreign policy and the US’s pivot to Asia, the region has seen the rise of US-centric minilateralisms that one way or another encircle China. For long, Korea, despite its iron-clad alliance with the US, has not taken an active part in most of these minilateralisms to avoid antagonizing China. Often two reasons are cited for this: 1) Korea’s economic dependence on China, and 2) China’s influence on North Korea, hence Korean Peninsula.

Let’s visualize Korea’s trade dependence over time.

Show the code

trade %<>% mutate(trade_dep = trade / total_trade_group_by,
                  label = ifelse(year == max(year), iso3c, NA))


trade %>% filter(iso3c %in% c("JPN", "CHN", "USA", "VNM")) %>% #, "SAU", "HKG", "TWN")) %>%
  ggplot(aes(year, trade_dep, color = iso3c)) +
  geom_line() +
  geom_label_repel(aes(label = label), nudge_x = 8, size = 8) +
  scale_x_continuous(breaks = c(seq(1948, 2023, 10), 2023)) +
  scale_y_continuous(breaks = seq(0, 1, 0.1), labels = scales::percent_format(scale = 100)) +
  labs(#title = "Trade Dependence of Korea",
       x = "Year",
       y = "Share in Korea's Trade Volume") +
  ggthemes::theme_pander() +
  theme(axis.text.x = element_text(angle = 45, hjust = 1)) +
  theme(legend.position = "none")

Figure 4: Korea’s Trade Dependence with China, US, Japan, and Vietnam

China has surpassed Japan and the United States as Korea’s largest trading partner since the mid-2000s. Before COVID-19, China’s share in Korea’s trade reached 25%, declining since then to just above 20%, but still remaining as the largest trading partner. The US’s share in Korea’s trade is around 15%, while Japan and Vietnam’s share has been 6% each. The following is the ranking of Korea’s trading partners since its foundation. Below, I visualize the rankings of Korea’s trading partners over time.

Show the code

# give trade data countries rankings based on trade_dep.

trade_decade <- trade %>% 
  mutate(decade = floor(year / 10) * 10)

trade_decade %<>% group_by(iso3c, decade) %>%
  summarise(trade_dep_decade = mean(trade_dep, na.rm = TRUE), .groups = 'drop')
  
trade_decade %<>% 
  group_by(decade) %>%
  arrange(decade, desc(trade_dep_decade)) %>%
  mutate(rank = row_number()) %>%
  ungroup()


top10 <- trade_decade %>% filter(decade == 2020) %>%
  top_n(-10, rank) %>%
  pull(iso3c)

first_appearance <- trade_decade %>%
  filter(iso3c %in% top10) %>%
  group_by(iso3c) %>%
  summarise(first_decade = min(decade[rank < 11])) %>%
  ungroup()

# Join this back to the original data to identify where to place labels
label_data <- trade_decade %>%
  inner_join(first_appearance, by = "iso3c") %>%
  filter(decade == first_decade)


# using ggbump to visualize the rankings of trade partners over time.

trade_decade %>% filter(iso3c %in% top10) %>%
  ggplot(aes(decade, rank, color = iso3c)) +
  ggthemes::theme_clean() +
  scale_x_continuous(limits = c(1935, 2025)) +
  scale_y_continuous(limits = c(1, 10), breaks = seq(1, 10, 1)) +
   geom_bump(size = 1.5) +
  geom_point(size = 6) +
  geom_text(data = label_data,
            aes(decade = decade - 0.1, label = iso3c),
            size = 5, hjust = 1.3) +
  geom_text(data = trade_decade %>% filter(iso3c %in% top10 & decade == max(decade)),
            aes(decade = decade + 0.1, label = iso3c),
            size = 5, hjust = -0.3) +
  scale_color_brewer(palette = "RdBu") +
  theme(legend.position = "none") +
  labs(title = "Ranking of Korea's Trading Partners",
       x = "Year",
       y = "Rank")

ggsave("trade_ranking.png", width = 10, height = 6, dpi = 300)

Figure 5: Ranking of Korea’s Trading Partners: 1940s to 2020s

Concluding Remarks

Korea’s trade data is an important source for studies on Korean diplomacy, helping us understand trends and tendencies in Korean foreign policy. In this blog post, I explored the trade data in the kdiplo R package. I visualized the trade volume, Korea’s trade dependence, and the rankings of Korea’s trading partners over time. The data is available in the kdiplo package. In a future blog post, I will explore Korea’s economic diplomacy bringing together the trade data with my other data on Korean presidents’ bilateral diplomatic visits. Feel free to add comments next to this blog post, or make requests using the issues tab in the kdiplo package’s Github.

In line with my post here, I invite you to explore and interact with this data. Let’s craft stories together using these datasets and connect through hashtags #kdiplo, #kdiploviz, #kdata, and #kdataviz.

--- title: "Exploring Korean Trade Data on `kdiplo` R Package" subtitle: "Trade Dependence with China" author: "Kadir Jun Ayhan" format: html: embed-resources: true code-fold: true code-summary: "Show the code" code-tools: true date: "2024-04-23" editor: visual echo: true warning: false image: trade_ranking.png fig-cap-location: top draft: false comments: hypothesis: true categories: - Korea - Korean dataset - dataset - R - R package - diplomacy - kdiplo - kdata - kdiploviz - kdataviz --- ```{r} #| include: false #devtools::install_github("kjayhan/kdiplo") library(kdiplo) library(tidyverse) library(magrittr) library(ggrepel) library(ggbump) ``` Since the foundation of the Republic of Korea in 1948, the economy has been one of the two pillars of Korean diplomacy, alongside security.[^1] Indeed, the country's trade has virtually always been over 50% of its GDP since 1970s, reaching over 100% in 2010s (see [@fig-trade_gdp]). [^1]: Korea added "public diplomacy" as a [third pillar of diplomacy in 2010](https://ayhan.phd/research/chapters/2016%20Introduction%20-%20Korea's%20PD/). ::: {#fig-trade_gdp} ```{r} ##| label: fig-trade_gdp ##| fig-cap: Trade as % of GDP in Korea trade_gdp <- WDI::WDI(indicator = c("trade_gdp" = "NE.TRD.GNFS.ZS", "gdp_growth" = "NY.GDP.MKTP.KD.ZG"), country = "KOR", start = 1960, end = 2022) trade_gdp %>% ggplot(aes(year, trade_gdp)) + geom_line(color = "#2780e3") + scale_x_continuous(breaks = seq(1960, 2022, 5)) + # add percentages to y axis scale_y_continuous(labels = scales::percent_format(scale = 1)) + labs(#title = "Trade as % of GDP in Korea", x = "Year", y = "Trade as % of GDP") + ggthemes::theme_pander() + theme(axis.text.x = element_text(angle = 45, hjust = 1)) ``` Source: [WDI](https://github.com/vincentarelbundock/WDI) Trade as % of GDP in Korea ::: Therefore, Korea's trade constitute an important data source for studies on Korean diplomacy. In this blog post, I will explore the trade data in the [`kdiplo`](https://kdiplo.com) R package. There are two main sources for the data. 1) [Korean Statistical Information Service (KOSIS)](https://kosis.kr/statHtml/statHtml.do?orgId=360&tblId=DT_1R11006_FRM101&conn_path=I3) which has Korea's trade data from 1965 until 2023 with countries and regions. 2) [Correlates of War (COW)'s Trade Data v4.0](https://correlatesofwar.org/data-sets/bilateral-trade/) which Korea's trade data from 1949 until 2014. In the trade data in the [`kdiplo`](https://kdiplo.com) package, I share both sets of data (export_kosis and import_kosis referring to export and import data from KOSIS; export_cow and import_cow referring to export and import data from COW) as well as mutated export and import variables that rely mainly on KOSIS data, but imputing it with the COW data where the former is missing. While in my own research, I use other ways to impute the still missing data, in the [`kdiplo`](https://kdiplo.com) package, I don't do that, leaving it to the users' discretion. Let's explore Korea's trade data in the [`kdiplo`](https://kdiplo.com) package. I have not yet created a function to get the data from the package, but one can simply get the data and assign it to an object. This is what the dataset looks like: ```{r} trade <- kdiplo::trade_data glimpse(trade) ``` # Trends in Korea's Trade Volume over Time Let's mutate the trade volume as the sum of exports and imports, and the total trade volume as the sum of total exports and total imports. ```{r} trade %<>% mutate(trade = export + import, total_trade = total_export + total_import) ``` Let's visualize the trade volume over time. ```{r} #| label: fig-trade_volume #| fig-cap: "Korea's Trade Volume (in billion $)" trade_years <- trade %>% group_by(year) %>% summarise(total_trade_group_by = sum(trade, na.rm = T), total_export_group_by = sum(export, na.rm = T), total_import_group_by = sum(import, na.rm = T)) trade %<>% left_join(trade_years, by = "year") trade %>% ggplot(aes(year, total_trade_group_by/1e9)) + geom_line(color = "#2780e3") + #geom_line(aes(year, total_export_group_by/1e9), color = "#CF4446") + #geom_line(aes(year, total_import_group_by/1e9), color = "black") + scale_y_continuous(breaks = seq(100, 1500, 200), labels = scales::comma) + scale_x_continuous(breaks = seq(1948, 2023, 5)) + labs(#title = "Trade Volume of Korea (in billion $)", x = "Year", y = "Trade Volume") + ggthemes::theme_pander() + theme(axis.text.x = element_text(angle = 45, hjust = 1)) + geom_vline(xintercept = 1962, linetype = "dashed", color = "#CF4446") + annotate("text", x = 1962, y = 200, label = "Park Chung-hee's First\nFive-Year Economic Plan", color = "black") + geom_vline(xintercept = 1988, linetype = "dashed", color = "#CF4446") + annotate("text", x = 1988, y = 500, label = "Democratization", color = "black") ``` There is this myth that Korea's economic development was all thanks to Park Chung-hee's economic development plans which built on [export-led industrialization](https://www.much.go.kr/en/contents.do?fid=03&cid=03_7). Although we cannot deny his influence on Korea's industrialization and economic growth (see [@fig-gdp_growth]), but Korea's trade volume has grown exponentially since the 1990s following the country's liberalization and democratization. ::: {#fig-gdp_growth} ```{r} ##| label: fig-gdp_growth ##| fig-cap: "Korea's GDP Growth Rate" trade_gdp %>% ggplot(aes(year, gdp_growth)) + geom_line(color = "#2780e3") + geom_smooth() + scale_x_continuous(breaks = seq(1960, 2022, 5)) + # add percentages to y axis scale_y_continuous(labels = scales::percent_format(scale = 1)) + labs(#title = "Trade as % of GDP in Korea", x = "Year", y = "Korea's GDP Growth Rate") + ggthemes::theme_pander() + theme(axis.text.x = element_text(angle = 45, hjust = 1)) ``` Source: [WDI](https://github.com/vincentarelbundock/WDI) Korea's GDP Growth Rate ::: # Trade Dependence We live at an age of increasing great power rivalry in the Indo-Pacific region, even this relatively new geopolitical naming of the region causing controversies with China. For quite a long time, region's two great powers, China and Japan, outsourced [great power management](https://www.jstor.org/stable/48615950) [to the United States and ASEAN](https://academic.oup.com/book/5050). However, in recent years with [growing assertiveness of Chinese foreign policy](https://academic.oup.com/cjip/article/7/2/153/438673) and the [US's pivot to Asia](https://www.jstor.org/stable/i40179543), the region has seen the rise of [US-centric minilateralisms](https://asiasociety.org/policy-institute/are-minilaterals-new-multilateral-indo-pacific-0) that one way or another encircle China. For long, Korea, despite its iron-clad alliance with the US, has not taken an active part in most of these minilateralisms to avoid antagonizing China. Often two reasons are cited for this: 1) Korea's economic dependence on China, and 2) China's influence on North Korea, hence Korean Peninsula. Let's visualize Korea's trade dependence over time. ```{r} #| label: fig-trade_dependence #| fig-cap: "Korea's Trade Dependence with China, US, Japan, and Vietnam" trade %<>% mutate(trade_dep = trade / total_trade_group_by, label = ifelse(year == max(year), iso3c, NA)) trade %>% filter(iso3c %in% c("JPN", "CHN", "USA", "VNM")) %>% #, "SAU", "HKG", "TWN")) %>% ggplot(aes(year, trade_dep, color = iso3c)) + geom_line() + geom_label_repel(aes(label = label), nudge_x = 8, size = 8) + scale_x_continuous(breaks = c(seq(1948, 2023, 10), 2023)) + scale_y_continuous(breaks = seq(0, 1, 0.1), labels = scales::percent_format(scale = 100)) + labs(#title = "Trade Dependence of Korea", x = "Year", y = "Share in Korea's Trade Volume") + ggthemes::theme_pander() + theme(axis.text.x = element_text(angle = 45, hjust = 1)) + theme(legend.position = "none") ``` China has surpassed Japan and the United States as Korea's largest trading partner since the mid-2000s. Before COVID-19, China's share in Korea's trade reached 25%, declining since then to just above 20%, but still remaining as the largest trading partner. The US's share in Korea's trade is around 15%, while Japan and Vietnam's share has been 6% each. The following is the ranking of Korea's trading partners since its foundation. Below, I visualize the rankings of Korea's trading partners over time. ```{r} #| label: fig-trade_ranking #| fig-cap: "Ranking of Korea's Trading Partners: 1940s to 2020s" # give trade data countries rankings based on trade_dep. trade_decade <- trade %>% mutate(decade = floor(year / 10) * 10) trade_decade %<>% group_by(iso3c, decade) %>% summarise(trade_dep_decade = mean(trade_dep, na.rm = TRUE), .groups = 'drop') trade_decade %<>% group_by(decade) %>% arrange(decade, desc(trade_dep_decade)) %>% mutate(rank = row_number()) %>% ungroup() top10 <- trade_decade %>% filter(decade == 2020) %>% top_n(-10, rank) %>% pull(iso3c) first_appearance <- trade_decade %>% filter(iso3c %in% top10) %>% group_by(iso3c) %>% summarise(first_decade = min(decade[rank < 11])) %>% ungroup() # Join this back to the original data to identify where to place labels label_data <- trade_decade %>% inner_join(first_appearance, by = "iso3c") %>% filter(decade == first_decade) # using ggbump to visualize the rankings of trade partners over time. trade_decade %>% filter(iso3c %in% top10) %>% ggplot(aes(decade, rank, color = iso3c)) + ggthemes::theme_clean() + scale_x_continuous(limits = c(1935, 2025)) + scale_y_continuous(limits = c(1, 10), breaks = seq(1, 10, 1)) + geom_bump(size = 1.5) + geom_point(size = 6) + geom_text(data = label_data, aes(decade = decade - 0.1, label = iso3c), size = 5, hjust = 1.3) + geom_text(data = trade_decade %>% filter(iso3c %in% top10 & decade == max(decade)), aes(decade = decade + 0.1, label = iso3c), size = 5, hjust = -0.3) + scale_color_brewer(palette = "RdBu") + theme(legend.position = "none") + labs(title = "Ranking of Korea's Trading Partners", x = "Year", y = "Rank") ggsave("trade_ranking.png", width = 10, height = 6, dpi = 300) ``` # Concluding Remarks Korea's trade data is an important source for studies on Korean diplomacy, helping us understand trends and tendencies in Korean foreign policy. In this blog post, I explored the trade data in the [`kdiplo`](https://kdiplo.com) R package. I visualized the trade volume, Korea's trade dependence, and the rankings of Korea's trading partners over time. The data is available in the [`kdiplo`](https://kdiplo.com) package. In a future blog post, I will explore Korea's economic diplomacy bringing together the trade data with my other data on Korean presidents' bilateral diplomatic visits. Feel free to add comments next to this blog post, or make requests using the issues tab in the [`kdiplo` package's Github](https://kdiplo.com/issues). In line with my [post here](https://ayhan.phd/blog/2024-03-26%20kdiplo/), I invite you to explore and interact with this data. Let’s craft stories together using these datasets and connect through hashtags [`#kdiplo`](https://www.linkedin.com/feed/hashtag/?keywords=kdiplo), [#kdiploviz](https://www.linkedin.com/feed/hashtag/?keywords=kdiploviz), [`#kdata`](https://www.linkedin.com/feed/hashtag/?keywords=kdata), and [#kdataviz](https://www.linkedin.com/feed/hashtag/?keywords=kdiplo).