USITC Gravity Database Adapted From the International Trade and Production Database for Estimation (ITPD-E) and Dynamic Gravity Dataset (DGD)

This repository is unmaintained, please see the new repository https://github.com/pachadotdev/gravitydatasets.

About

The goal of usitcgravity is to provide data from The International Trade and Production Database for Estimation (ITPD-E) and the Dynamic Gravity Dataset (DGD) ready to be used in R (i.e. with the gravity package).

The package provides consistent data on international and domestic trade for 243 countries, 170 industries, and 17 years alongside variables for gravity estimation. The data sources are ITPD-E and DGD from the USITC, which use administrative data and intentionally does not include information estimated by statistical techniques, which makes the datasets suitable for gravity estimation.

usitcgravity can be installed by running

# install.packages("remotes")
install_github("pachadotdev/usitcgravity")

Sources

Borchert, Ingo & Larch, Mario & Shikher, Serge & Yotov, Yoto, 2020. The International Trade and Production Database for Estimation (ITPD-E). School of Economics Working Paper Series 2020-5, LeBow College of Business, Drexel University.

Gurevich, Tamara & Herman, Peter, 2018. The Dynamic Gravity Dataset: 1948-2016. USITC Working Paper 2018-02-A.

Differences with the original sources.

Since v0.1:

• Fixes duplicated ISO3 code + Dynamic code for Cambodia and West Samoa
• Fixes duplicated label "south_east_asia" vs "suth_east_asia"

Since v0.2:

• Fixes inconsistencies in the gravity table

The last point deserves an example. See the differences in the common_colonizer variable for Argentina-Chile-Peru and Spain. This variable should be the same, for example, for ARG-CHL or CHL-ARG, but it's not in the original dataset.

> tbl(con, "usitc_gravity") %>% 
+   filter(iso3_o == "ARG", iso3_d %in% c("CHL", "ESP", "PER"), year == 2015) %>%
+   select(iso3_o, iso3_d, colony_of_origin_ever, colony_of_destination_ever, colony_ever, common_colonizer)
# Source:   SQL [3 x 6]
# Database: postgres  [pacha@localhost:5432/tariff_man]
  iso3_o iso3_d colony_of_origin_ever colony_of_destination_ever colony_ever common_colonizer
  <chr>  <chr>                  <int>                      <int>       <int>            <int>
1 ARG    CHL                        0                          0           0                0
2 ARG    ESP                        0                          1           1                0
3 ARG    PER                        0                          0           0                0

> tbl(con, "usitc_gravity") %>% 
+   filter(iso3_d == "ARG", iso3_o %in% c("CHL", "ESP", "PER"), year == 2015) %>%
+   select(iso3_o, iso3_d, colony_of_origin_ever, colony_of_destination_ever, colony_ever, common_colonizer)
# Source:   SQL [3 x 6]
# Database: postgres  [pacha@localhost:5432/tariff_man]
  iso3_o iso3_d colony_of_origin_ever colony_of_destination_ever colony_ever common_colonizer
  <chr>  <chr>                  <int>                      <int>       <int>            <int>
1 CHL    ARG                        0                          0           0                1
2 ESP    ARG                        1                          0           1                0
3 PER    ARG                        0                          0           0                1

I corrected this by using the gravity table itself in two ways:

• By binding rows on a pairwise basis for all symmetrical variables (i.e., common_colonizer) and obtained the maximum for each pair.
• By obtaining a full join on a pairwise basis for all non-symmetrical variables (i.e., colony_of_origin_ever) and obtained the maximum for each pair.

To check this I ran some queries to verify, for example, that the populations for CHL-ESP are around c(20,40) and not c(40,40) (i.e., Spain doubles the population of Chile), and that for the same pair colony_of_origin_ever = 0 but for ESP-CHL colony_of_origin_ever = 1.

Usage

Estimating the gravity model of trade with exporter/importer time fixed effects for 4 sectors:

library(usitcgravity)
library(dplyr)
library(purrr)
library(fixest)

con <- usitcgravity_connect()
  
# run one model per sector
models <- map(
  tbl(con, "sector_names") %>% pull(broad_sector_id),
  function(s) {
    message(s)
    
    yrs <- seq(2005, 2015, by = 5)
    
    d <- tbl(con, "trade") %>% 
      filter(year %in% yrs, broad_sector_id == s) %>% 
      group_by(year, exporter_iso3, importer_iso3, broad_sector_id) %>% 
      summarise(trade = sum(trade, na.rm = T)) %>% 
      inner_join(
        tbl(con, "gravity") %>% 
          filter(year %in% yrs) %>% 
          select(iso3_o, iso3_d, contiguity, common_language, colony_ever, distance),
        by = c("exporter_iso3" = "iso3_o", "importer_iso3" = "iso3_d")
      ) %>% 
      collect()
    
    d <- d %>% 
      mutate(
        etfe = paste(exporter_iso3, year, sep = "_"),
        itfe = paste(importer_iso3, year, sep = "_")
      )
    
    feglm(trade ~ contiguity + common_language + colony_ever + 
            log(distance) | etfe + itfe,
          family = quasipoisson(),
          data = d)
  }
)

usitc_disconnect()

print(models)

[[1]]
GLM estimation, family = quasipoisson, Dep. Var.: trade
Observations: 246,359 
Fixed-effects: etfe: 666,  itfe: 685
Standard-errors: Clustered (etfe) 
                 Estimate Std. Error   t value   Pr(>|t|)    
contiguity      -1.107178   0.095210 -11.62882  < 2.2e-16 ***
common_language  0.904003   0.093483   9.67025  < 2.2e-16 ***
colony_ever     -0.574003   0.123660  -4.64179 4.1623e-06 ***
log(distance)   -2.296888   0.049558 -46.34779  < 2.2e-16 ***
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
                                           
  Squared Cor.: 0.995066                   

[[2]]
GLM estimation, family = quasipoisson, Dep. Var.: trade
Observations: 399,386 
Fixed-effects: etfe: 699,  itfe: 699
Standard-errors: Clustered (etfe) 
                 Estimate Std. Error   t value   Pr(>|t|)    
contiguity      -0.607615   0.066616  -9.12114  < 2.2e-16 ***
common_language  1.018309   0.122366   8.32184 4.5501e-16 ***
colony_ever     -0.349478   0.095461  -3.66094 2.7027e-04 ***
log(distance)   -1.231589   0.079559 -15.48011  < 2.2e-16 ***
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
                                           
  Squared Cor.: 0.993537                   

[[3]]
GLM estimation, family = quasipoisson, Dep. Var.: trade
Observations: 184,626 
Fixed-effects: etfe: 658,  itfe: 682
Standard-errors: Clustered (etfe) 
                 Estimate Std. Error   t value  Pr(>|t|)    
contiguity      -1.288373   0.104177 -12.36711 < 2.2e-16 ***
common_language  1.346375   0.132477  10.16305 < 2.2e-16 ***
colony_ever     -0.288694   0.222472  -1.29766   0.19486    
log(distance)   -2.298326   0.074100 -31.01657 < 2.2e-16 ***
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
                                           
  Squared Cor.: 0.992104                   

[[4]]
GLM estimation, family = quasipoisson, Dep. Var.: trade
Observations: 46,471 
Fixed-effects: etfe: 578,  itfe: 579
Standard-errors: Clustered (etfe) 
                 Estimate Std. Error    t value  Pr(>|t|)    
contiguity      -2.671951   0.183676 -14.547108 < 2.2e-16 ***
common_language  1.612853   0.127480  12.651800 < 2.2e-16 ***
colony_ever      0.139065   0.170706   0.814645   0.41561    
log(distance)   -2.267023   0.081150 -27.936349 < 2.2e-16 ***
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
                                           
  Squared Cor.: 0.998325