Accessible Version
Understanding Trade Fragmentation, Accessible Data
Figure 1: Effect of Geopolitical Distance on Trade Flows
Effect of Geopolitical Distance on Trade Flows over Time (Full Sample)
Coefficient on Total Geopolitical Distance
| End of 10-Year Window | Coefficient | Upper Bound | Lower Bound |
|---|---|---|---|
| 1999 | 0.0033673 | 0.0436022 | -0.0368677 |
| 2000 | 0.016123 | 0.0579097 | -0.0256637 |
| 2001 | 0.0302788 | 0.0727246 | -0.012167 |
| 2002 | 0.0493095 | 0.0910934 | 0.0075255 |
| 2003 | 0.057094 | 0.10169 | 0.012498 |
| 2004 | 0.0705325 | 0.116352 | 0.024713 |
| 2005 | 0.0964175 | 0.1537329 | 0.0391022 |
| 2006 | 0.0951084 | 0.1570315 | 0.0331853 |
| 2007 | 0.0736199 | 0.1298165 | 0.0174234 |
| 2008 | 0.0807207 | 0.1402636 | 0.0211778 |
| 2009 | 0.0734322 | 0.1314801 | 0.0153842 |
| 2010 | 0.0307859 | 0.0843768 | -0.022805 |
| 2011 | 0.0123421 | 0.0600303 | -0.035346 |
| 2012 | 0.0070725 | 0.0553483 | -0.0412034 |
| 2013 | -0.0326986 | 0.0246521 | -0.0900493 |
| 2014 | -0.0508124 | 0.003955 | -0.1055797 |
| 2015 | -0.0492514 | -0.009912 | -0.0885909 |
| 2016 | -0.0487024 | -0.0088686 | -0.0885363 |
| 2017 | -0.0283929 | 0.0128944 | -0.0696801 |
| 2018 | -0.0060949 | 0.0349962 | -0.047186 |
| 2019 | 0.0129663 | 0.0656637 | -0.039731 |
| 2020 | 0.0123392 | 0.0714432 | -0.0467647 |
| 2021 | 0.0064359 | 0.0687855 | -0.0559138 |
| 2022 | -0.020842 | 0.042224 | -0.083908 |
| 2023 | -0.0556456 | 0.0052134 | -0.1165046 |
Effect of Geopolitical Distance on Trade Flows over Time (ex. China)
Coefficient on Total Geopolitical Distance
| End of 10-Year Window | Coefficient | Upper Bound | Lower Bound |
|---|---|---|---|
| 1999 | -0.0038879 | 0.0334188 | -0.0411946 |
| 2000 | 0.0044601 | 0.0429399 | -0.0340198 |
| 2001 | 0.0219806 | 0.0613922 | -0.0174309 |
| 2002 | 0.0442968 | 0.0838838 | 0.0047098 |
| 2003 | 0.0481484 | 0.0894324 | 0.0068644 |
| 2004 | 0.0547742 | 0.0929242 | 0.0166241 |
| 2005 | 0.0760711 | 0.1193215 | 0.0328207 |
| 2006 | 0.0773693 | 0.1248114 | 0.0299271 |
| 2007 | 0.0421275 | 0.0839783 | 0.0002768 |
| 2008 | 0.0410373 | 0.0793379 | 0.0027368 |
| 2009 | 0.0436005 | 0.0759238 | 0.0112772 |
| 2010 | 0.0133971 | 0.0432042 | -0.0164101 |
| 2011 | 0.0260256 | 0.0554936 | -0.0034423 |
| 2012 | 0.0276238 | 0.0594769 | -0.0042292 |
| 2013 | 0.0029741 | 0.0390406 | -0.0330925 |
| 2014 | -0.0174743 | 0.0206573 | -0.0556059 |
| 2015 | -0.0552922 | -0.0181822 | -0.0924023 |
| 2016 | -0.0664024 | -0.0303743 | -0.1024305 |
| 2017 | -0.0651839 | -0.0284258 | -0.101942 |
| 2018 | -0.0579893 | -0.0233199 | -0.0926587 |
| 2019 | -0.074603 | -0.0398699 | -0.1093361 |
| 2020 | -0.0852909 | -0.0484155 | -0.1221663 |
| 2021 | -0.0986208 | -0.0622124 | -0.1350292 |
| 2022 | -0.1227615 | -0.0843382 | -0.1611848 |
| 2023 | -0.1398762 | -0.0937061 | -0.1860463 |
Note: Coefficient estimates from rolling-window PPML regressions (10-year window, 1990-2023), with 95% confidence intervals. Geopolitical distance measures are based on estimates from Bailey et al. (2017).
Source: Airaudo et al (2025), UN Comtrade, Authors’ calculations.
Figure 2: Import Penetration Across Advanced Economies and China, 2014-2024
Chinese Import Penetration in Advanced Economies
Percent of each country's GDP
| Country | 2014 | 2024 |
|---|---|---|
| U.S. | 2.7 | 2.1 |
| E.U. | 1.6 | 2.7 |
| U.K. | 2.1 | 1.4 |
| Canada | 2.9 | 2.8 |
Advanced Economies Import Penetration in China
Percent of Chinese GDP
| Country | 2014 | 2024 |
|---|---|---|
| U.S. | 1.2 | 0.2 |
| E.U. | 0.8 | 0.1 |
| U.K. | 2 | 0.2 |
| Canada | 1.2 | 0.2 |
Note: Key identifies in order from left to right.
Source: UN Comtrade, Authors’ calculations.
Figure 3: Global concentration of critical mineral supply chains, 2023
Extraction
| Critical Mineral | Country | Percent |
|---|---|---|
| Cobalt | Democratic Republic of Congo | 65.47122602 |
| Cobalt | Rest of world | 34.52877398 |
| Copper | Chile | 23.59728276 |
| Copper | Democratic Republic of Congo | 11.89882753 |
| Copper | Peru | 11.74599366 |
| Copper | Rest of world | 52.75789605 |
| Lithium | Australia | 43.39525284 |
| Lithium | Chile | 23.94220846 |
| Lithium | China | 17.54385965 |
| Lithium | Rest of world | 15.11867905 |
| Magnet rare earth elements | China | 61.1878453 |
| Rare earths | Rest of world | 29.97237569 |
| Rare earths | United States | 8.839779006 |
Processing
| Critical Mineral | Country | Percent |
|---|---|---|
| Cobalt | China | 76.72028597 |
| Cobalt | Finland | 8.400357462 |
| Cobalt | Rest of world | 14.87935657 |
| Copper | China | 43.84459413 |
| Copper | Rest of world | 56.15540587 |
| Lithium | Chile | 26.31877482 |
| Lithium | China | 64.60578559 |
| Lithium | Rest of world | 9.075439592 |
| Rare earths | China | 92.13630406 |
| Rare earths | Rest of world | 7.863695937 |
Note: Individual country shares shown when larger than 10 percent. RoW is rest of the world. DRC is Democratic Republic of the Congo, FIN is Finland, and PER is Peru.
Source: International Energy Agency.
Figure 4: Sectoral heterogeneity in the association between trade and geopolitical distance
5-Year Rolling Regression for FRB Forecasted Countries
Coefficient on Economic Geopolitical Distance
| End of 5-Year Window | Tech Class | Coefficient | Upper Bound | Lower Bound |
|---|---|---|---|---|
| 2006 | High Tech | -0.091491195 | -0.007070834 | -0.17591156 |
| 2007 | High Tech | -0.109163917 | -0.020491494 | -0.19783634 |
| 2008 | High Tech | -0.110698597 | -0.020162363 | -0.20123483 |
| 2009 | High Tech | -0.112916551 | -0.025578653 | -0.20025444 |
| 2010 | High Tech | -0.107145341 | -0.023217907 | -0.19107278 |
| 2011 | High Tech | -0.095139401 | -0.017129391 | -0.17314941 |
| 2012 | High Tech | -0.084794104 | -0.012051082 | -0.15753713 |
| 2013 | High Tech | -0.078934381 | -0.008054262 | -0.1498145 |
| 2014 | High Tech | -0.086689087 | -0.018891206 | -0.15448697 |
| 2015 | High Tech | -0.096329566 | -0.026300212 | -0.16635892 |
| 2016 | High Tech | -0.107071928 | -0.036077671 | -0.17806618 |
| 2017 | High Tech | -0.118906242 | -0.050734058 | -0.18707843 |
| 2018 | High Tech | -0.128864902 | -0.063071862 | -0.19465794 |
| 2019 | High Tech | -0.13180756 | -0.065051295 | -0.19856383 |
| 2020 | High Tech | -0.14479842 | -0.079127215 | -0.21046962 |
| 2021 | High Tech | -0.157496825 | -0.090319782 | -0.22467387 |
| 2022 | High Tech | -0.17812726 | -0.10312261 | -0.25313193 |
| 2023 | High Tech | -0.209607665 | -0.12331513 | -0.2959002 |
| 2024 | High Tech | -0.250903919 | -0.14646518 | -0.35534266 |
| 2006 | Low Tech | 0.064787094 | 0.18876655 | -0.059192367 |
| 2007 | Low Tech | 0.057032947 | 0.18779682 | -0.073730916 |
| 2008 | Low Tech | 0.06110541 | 0.1911331 | -0.068922274 |
| 2009 | Low Tech | 0.074687392 | 0.20596389 | -0.056589104 |
| 2010 | Low Tech | 0.092644697 | 0.22081837 | -0.035528969 |
| 2011 | Low Tech | 0.088212136 | 0.21105108 | -0.034626812 |
| 2012 | Low Tech | 0.067597297 | 0.18658791 | -0.051393319 |
| 2013 | Low Tech | 0.046036612 | 0.16124326 | -0.069170028 |
| 2014 | Low Tech | 0.016852029 | 0.12281093 | -0.089106873 |
| 2015 | Low Tech | -0.003376722 | 0.10218801 | -0.10894145 |
| 2016 | Low Tech | -0.015009463 | 0.088314041 | -0.11833297 |
| 2017 | Low Tech | -0.018662575 | 0.081566751 | -0.1188919 |
| 2018 | Low Tech | -0.018116996 | 0.080523349 | -0.11675734 |
| 2019 | Low Tech | -0.011616185 | 0.089689434 | -0.1129218 |
| 2020 | Low Tech | -0.014297373 | 0.083901607 | -0.11249635 |
| 2021 | Low Tech | -0.01622254 | 0.08194492 | -0.11439 |
| 2022 | Low Tech | -0.028208583 | 0.075070277 | -0.13148744 |
| 2023 | Low Tech | -0.042277596 | 0.071058281 | -0.15561347 |
| 2024 | Low Tech | -0.07029578 | 0.062666357 | -0.20325792 |
Note: Coefficient estimates from rolling-window regressions with 95% confidence intervals. Sector classification into high and low-tech comes from Airaudo et al (2025).
Source: Airaudo et al (2025), UN Comtrade, Authors’ calculations.
Last Update:
December 12, 2025