International Finance Discussion Papers: Accessible versions of figures for 1418

Geopolitical Risk and Global Banking

Accessible version of figures


Figure 1: U.S. Banks' Foreign Operations

Panel A: Foreign Exposures as a Share of Total Assets This line graph shows the ratio of foreign exposures to total assets by U.S. banks from 1990:Q1 to 2021:Q4. The y-axis ranges from 0 to 0.4, representing the share of foreign lending. The line showing the share hovers around 0.2 with a notable increase after the Great Financial Crisis around 2009 ending at a value of around 0.22 for the most recent quarter.

Panel B: Local Exposures as a Share of Foreign Exposures This graph shows the ratio of affiliate (local) foreign lending to total foreign exposures. The y-axis ranges from 0.0 to 0.8, while the x-axis spans 1990:Q1 to 2021:Q4. The ratio hovers roughly around 0.45 with a small increase around 2004 that reverts around 2012.

Panel C: Distribution of Foreign Exposure by Region, 2010:Q4 This graph shows a kernel density plot. The y-axis ranges from 0 to 6, while the x-axis ranges from 0 to 1. The graph shows four curves for Europe, Asia, Latin America, and Rest of World. The curve for Europe is fairly flat, starting at around 1.6 and going down continuously to zero from x=0.5 to x=1. The curves for Asia and Latin America have their mass between 0 and 0.1 where the y-axis reaches a value of 4. The curve for Asia hits 2 at x=0.1 and is close to 0 starting from x=0.4. The curve for Latin America behaves similarly, except that it exhibits a second small bump from x=0.4 to x=0.7. The curve for the Rest of World starts at around 2 and then falls slowly and continuously, reaching zero at around x=0.8.

Panel D: Distribution of Foreign Claims by Country for Selected Bank, 2019:Q4 This graph shows a kernel density plot. The y-axis ranges from 0 to 6, while the x-axis ranges from 0 to 1. The graph shows four curves for Europe, Asia, Latin America, and Rest of World. The curve for Europe is fairly flat, starting at around 1.8 and going down continuously to zero from x=0.4 to x=0.8. The curves for Asia and Rest of World have their mass between 0 and 0.1 where the y-axis reaches a value of 3. The curve for Latin America exceeds the value of 6 close to x=0 and then falls abruptly to around 1.8 at x=0.1. The curve then falls further, hovering at a small number ending at x=0.8.

Panel E Distribution of Foreign Claims by Country for Selected Banks, 2010:Q4

Date Bank name Country Country claims as a percent of the bank's total assets
12/31/2010 Bank of NY Mellon Australia 1.18
12/31/2010 Bank of NY Mellon France 2.53
12/31/2010 Bank of NY Mellon Germany 3.79
12/31/2010 Bank of NY Mellon Japan 1.33
12/31/2010 Bank of NY Mellon Netherlands 2.33
12/31/2010 Citigroup France 1.48
12/31/2010 Citigroup Germany 0.94
12/31/2010 Citigroup India 1.48
12/31/2010 Goldman Sachs China 1.62
12/31/2010 Goldman Sachs France 4.58
12/31/2010 Goldman Sachs Germany 2.54
12/31/2010 Goldman Sachs Japan 4.24
12/31/2010 Goldman Sachs United Kingdom 1.32
12/31/2010 State Street Australia 2.55
12/31/2010 State Street Netherlands 1.62
12/31/2010 State Street United Kingdom 2.76

Panel F Distribution of Foreign Claims by Country for Selected Banks, 2019:Q4

Date Bank name Country Country claims as a percent of the bank's total assets
12/31/2019 Bank of NY Mellon Belgium 1.69
12/31/2019 Bank of NY Mellon Canada 1.32
12/31/2019 Bank of NY Mellon Germany 5.40
12/31/2019 Bank of NY Mellon Japan 7.89
12/31/2019 Bank of NY Mellon United Kingdom 4.62
12/31/2019 Citigroup Germany 2.23
12/31/2019 Citigroup Japan 3.98
12/31/2019 Citigroup Mexico 3.69
12/31/2019 Citigroup South Korea 2.13
12/31/2019 Citigroup United Kingdom 3.54
12/31/2019 Goldman Sachs Canada 1.86
12/31/2019 Goldman Sachs France 1.93
12/31/2019 Goldman Sachs Germany 3.35
12/31/2019 Goldman Sachs Japan 8.42
12/31/2019 Goldman Sachs United Kingdom 6.30
12/31/2019 State Street Australia 1.37
12/31/2019 State Street Canada 1.32
12/31/2019 State Street Germany 7.21
12/31/2019 State Street Japan 4.60
12/31/2019 State Street United Kingdom 5.71

Note: Panel (a) of the figure shows U.S. banks' average foreign exposures as a share of total assets from 1990:Q1 to 2021:Q4. Panel (b) shows U.S. banks' local exposures, or exposures through foreign offices, as a share of their total foreign exposures. Panels (c) and (d) illustrate the kernel density of the share of foreign operations in four regions|Europe, Asia, Latin America, and the rest of the world|in 2010:Q4 and 2019:Q4, respectively, across U.S. banks. Panel (e) and (f) illustrate the top countries by foreign claims size (expressed as a share of total assets) in 2010:Q4 and 2019:Q4, respectively, for four selected U.S. banks. Data source(s): FFIEC 009, FR Y9-C, and Call Reports for Panels (a){(d); public version of FFIEC 009/009a for Panels (e){(f).


ID id_rssd d_dt tstmp_ext tstmp_load cntry_cd xborder_claims local_claims_post_2013 local_claims_pre_2013 sub_name topholder_id topholder_name cntry_nm total_assets local_claims total_claims claims_as_pct_total_assets rank
1 35301 12/31/2010 20110219.01 20110219.01 12106 2555 0 19 STATE STREET B&TC 1111435 "State Street" Netherlands 158890.975 19 2574 1.619978731 3
2 35301 12/31/2010 20110219.01 20110219.01 13005 4386 0 0 STATE STREET B&TC 1111435 "State Street" United Kingdom 158890.975 0 4386 2.760383338 1
3 35301 12/31/2010 20110219.01 20110219.01 60089 3610 0 444 STATE STREET B&TC 1111435 "State Street" Australia 158890.975 444 4054 2.551435033 2
9 541101 12/31/2010 20110219.01 20110219.01 10804 6249 0 0 BANK OF NY MELLON 3587146 "Bank of NY Mellon" France 247159 0 6249 2.528331964 2
10 541101 12/31/2010 20110219.01 20110219.01 11002 7333 0 2046 BANK OF NY MELLON 3587146 "Bank of NY Mellon" Germany 247159 2046 9379 3.794723235 1
11 541101 12/31/2010 20110219.01 20110219.01 12106 5542 0 208 BANK OF NY MELLON 3587146 "Bank of NY Mellon" Netherlands 247159 208 5750 2.326437637 3
12 541101 12/31/2010 20110219.01 20110219.01 42609 2261 0 1025 BANK OF NY MELLON 3587146 "Bank of NY Mellon" Japan 247159 1025 3286 1.329508535 4
13 541101 12/31/2010 20110219.01 20110219.01 60089 2907 0 0 BANK OF NY MELLON 3587146 "Bank of NY Mellon" Australia 247159 0 2907 1.17616595 5
19 1951350 12/31/2010 20110601.01 20110601.01 10804 26401 0 2015 CITIGROUP 1951350 Citigroup France 1913902 2015 28416 1.484715518 1
20 1951350 12/31/2010 20110601.01 20110601.01 11002 17989 0 0 CITIGROUP 1951350 Citigroup Germany 1913902 0 17989 0.939912284 3
21 1951350 12/31/2010 20110601.01 20110601.01 42102 9850 0 18549 CITIGROUP 1951350 Citigroup India 1913902 18549 28399 1.483827281 2
27 2380443 12/31/2010 20111025.01 20111025.01 10804 41483 0 224 GOLDMAN SACHS GROUP THE 2380443 "Goldman Sachs" France 911330 224 41707 4.576498085 1
28 2380443 12/31/2010 20111025.01 20111025.01 11002 23121 0 0 GOLDMAN SACHS GROUP THE 2380443 "Goldman Sachs" Germany 911330 0 23121 2.537061218 3
29 2380443 12/31/2010 20111025.01 20111025.01 13005 12030 0 0 GOLDMAN SACHS GROUP THE 2380443 "Goldman Sachs" United Kingdom 911330 0 12030 1.32004872 5
30 2380443 12/31/2010 20111025.01 20111025.01 41408 14481 0 243 GOLDMAN SACHS GROUP THE 2380443 "Goldman Sachs" China 911330 243 14724 1.615660628 4
31 2380443 12/31/2010 20111025.01 20111025.01 42609 29932 0 8753 GOLDMAN SACHS GROUP THE 2380443 "Goldman Sachs" Japan 911330 8753 38685 4.244894824 2
4 35301 12/31/2019 20200219.15 20200219.18 11002 1332 16367 0 STATE STREET B&TC 1111435 "State Street" Germany 245610 16367 17699 7.206139815 1
5 35301 12/31/2019 20200219.15 20200219.18 13005 921 13110 0 STATE STREET B&TC 1111435 "State Street" United Kingdom 245610 13110 14031 5.71271528 2
6 35301 12/31/2019 20200219.15 20200219.18 29998 389 2841 0 STATE STREET B&TC 1111435 "State Street" Canada 245610 2841 3230 1.315093034 5
7 35301 12/31/2019 20200219.15 20200219.18 42609 2806 8488 0 STATE STREET B&TC 1111435 "State Street" Japan 245610 8488 11294 4.598346973 3
8 35301 12/31/2019 20200219.15 20200219.18 60089 622 2744 0 STATE STREET B&TC 1111435 "State Street" Australia 245610 2744 3366 1.370465372 4
14 541101 12/31/2019 20201209.17 20201209.18 10251 311 6140 0 BANK OF NY MELLON 3587146 "Bank of NY Mellon" Belgium 381508 6140 6451 1.690921291 4
15 541101 12/31/2019 20201209.17 20201209.18 11002 5300 15291 0 BANK OF NY MELLON 3587146 "Bank of NY Mellon" Germany 381508 15291 20591 5.397265588 2
16 541101 12/31/2019 20201209.17 20201209.18 13005 2306 15321 0 BANK OF NY MELLON 3587146 "Bank of NY Mellon" United Kingdom 381508 15321 17627 4.620348721 3
17 541101 12/31/2019 20201209.17 20201209.18 29998 3533 1490 0 BANK OF NY MELLON 3587146 "Bank of NY Mellon" Canada 381508 1490 5023 1.316617214 5
18 541101 12/31/2019 20201209.17 20201209.18 42609 812 29298 0 BANK OF NY MELLON 3587146 "Bank of NY Mellon" Japan 381508 29298 30110 7.892363987 1
22 1951350 12/31/2019 20200703.12 20200703.13 11002 27743 15748 0 CITIGROUP 1951350 Citigroup Germany 1951158 15748 43491 2.228984019 4
23 1951350 12/31/2019 20200703.12 20200703.13 13005 14181 54906 0 CITIGROUP 1951350 Citigroup United Kingdom 1951158 54906 69087 3.540820374 3
24 1951350 12/31/2019 20200703.12 20200703.13 31704 3723 68356 0 CITIGROUP 1951350 Citigroup Mexico 1951158 68356 72079 3.694165209 2
25 1951350 12/31/2019 20200703.12 20200703.13 42609 31814 45900 0 CITIGROUP 1951350 Citigroup Japan 1951158 45900 77714 3.982968063 1
26 1951350 12/31/2019 20200703.12 20200703.13 43001 8747 32786 0 CITIGROUP 1951350 Citigroup South Korea 1951158 32786 41533 2.128633355 5
32 2380443 12/31/2019 20200219.12 20200219.13 10804 18372 829 0 GOLDMAN SACHS GROUP THE 2380443 "Goldman Sachs" France 992996 829 19201 1.933643237 4
33 2380443 12/31/2019 20200219.12 20200219.13 11002 31625 1686 0 GOLDMAN SACHS GROUP THE 2380443 "Goldman Sachs" Germany 992996 1686 33311 3.354595587 3
34 2380443 12/31/2019 20200219.12 20200219.13 13005 10250 52339 0 GOLDMAN SACHS GROUP THE 2380443 "Goldman Sachs" United Kingdom 992996 52339 62589 6.303046538 2
35 2380443 12/31/2019 20200219.12 20200219.13 29998 17883 613 0 GOLDMAN SACHS GROUP THE 2380443 "Goldman Sachs" Canada 992996 613 18496 1.862645972 5
36 2380443 12/31/2019 20200219.12 20200219.13 42609 14223 69341 0 GOLDMAN SACHS GROUP THE 2380443 "Goldman Sachs" Japan 992996 69341 83564 8.415341049 1

Return to text.


Figure 2: Global Geopolitical Risk and Other Risk Indices

Panel A: GGPR Indices Panel A has two plots. The top graph plots the standardized global GPR index GGPRN from Caldara and Iacoviello (2022) over time. The bottom graph shows the standardized global transcript-based GPR measure GGPRT provided by this paper over time. Both indexes spike with the Iraq war in 2003:Q1, the Russia-Ukraine War in 2022:Q1 and the Israel-Hamas War in 2023:Q4. GGPRN is more volatile than GGPRT showing larger fluctuations, but otherwise the indexes look fairly similar.

Note: Panel (a) shows two global geopolitical risk (GGPR) indices, which are aggregated from country-specific geopolitical risk (CGPR) indices, covering the period from 2002:Q1 to 2023:Q4. The top chart displays GGPR from Caldara and Iacoviello (2022) (GGPRN), and the bottom chart displays GGPR constructed by applying textual analysis to earnings-call transcripts using the NL Analytics platform (GGPRT ).

Panel B: Other Risk Indices Panel B has two plots. The top graph shows Country Risk CRI from Hassan et. al (2023). The bottom panel shows the World Uncertainty Index WUI from Ahir et al. (2022). CRI exhibits spikes around the Global Financial Crisis from 2008/2009, the European Sovereign Debt Crisis in 2012, and Covid in 2020. WUI does not spike with the Global Financial Crisis but with the European Sovereign Debt Crisis, Covid and the U.S. fiscal cliff in 2012. The correlation between the two series is not very strong based on visual inspection outside of the few common spikes.

Note: Panel (b) shows the aggregated country risk index (CRI) by Hassan et al. (2023) (top), and the World Uncertainty Index (WUI) by Ahir et al. (2022) (bottom). All the indices are standardized by their respective standard deviations within the sample.

Return to text.


Figure 3: Geopolitical Risk and Credit Risk: Russia-Ukraine Conflicts

This graph shows the estimated delta coefficients including 95 percent confidence intervals from Equation 3. The x-axis ranges from -4 to +4, indicating quarters before and after the event with the event occurring at x=0. The y-axis ranges from -1 to 4. All coefficient estimates for t=-4 to t=0 are below zero, taking values from around -0.8 and -0.2. The magnitude of the estimated coefficients jump at t=1 crossing the y=0 line. From t=1 to t=4, the estimated coefficient value ranges from 2.5 to 3. Except for the estimated coefficient value at t=4, the confidence intervals are small.

Note: The figure illustrates the effect of geopolitical risk shocks from the Crimea con ict in 2013:Q4 and the Russia-Ukraine war in 2022:Q1 on the log average probability of default of loans to Russian borrowers relative to loans to borrowers in other countries. It plots the coefficients $$\delta_{1k}$$ from Equation (3). Standard errors, shown in parentheses, are clustered at the country- time level. Data source: FR Y-14.

Return to text.


Figure A.1: Country-specific Geopolitical Risk and Other Risk Indices

This figure has three panels, one for Poland, the United Kingdom, and South Korea. Each panel consists of six plots. The three plots on the left show standardized CGPRN, CGPRT and CGPRTfin over time. The three plots on the right show standardized CRI, WUI, and CDS spreads over time.

Panel A: Poland For Poland the three different GPR measures are fairly similar spiking only with the Russia-Ukraine war in 2022:Q1 and are fairly flat otherwise. CRI and WUI exhibit some volatility but no notable spikes. CDS spreads spike at the height of the Global Financial Crisis and the European Sovereign Debt Crisis in 2011:Q4.

Panel B: United Kingdom For the United Kingdom, the three GPR measures exhibit some volatility over time. CGPRN spikes with the War on Terror and the invasion of Iraq in 2003:Q1 and the terrorist bombing in London in 2005:Q3 and the Russia-Ukraine war. CGPRT spikes with the terrorist bombing and additional terror threats in 2007:Q2. The CGPRTfin measure spikes with the bombing of British targets in Turkey in 2003:Q4 and the terror threats in 2007:Q2. CRI spikes with the Global Financial Crisis and the Brexit Vote. WUI spikes with the Global Financial Crisis, the Brexit Vote, and the Ratification of Brexit. CDS spreads spike with the Global Financial Crisis, the European Sovereign Debt Crisis and the release of the highest unemployment figures since 1994 in 2011:Q4.

Panel C: South Korea For South Korea, CGPRN and CGPRT exhibit a fair amount of volatility while CGPRTfin has several spikes but also periods where the value is flat. CGPRN spikes when North Korea withdraws from the nonproliferation treaty in 2003:Q1 and around 2017:Q4 when the country launches missile and nuclear tests. CGPRT spikes additionally with a wave of political protest in 2019:Q4. CGPRTfin spikes only around 2017:Q4 when the country begins missile and nuclear tests. CRI spikes around the collapse of Lehman Brothers in 2008:Q3 and the outbreak of Covid around 2020:Q1. WUI spikes at the beginning of the Global Financial Crisis in 2007. CDS spreads spike with the collapse of Lehman Brothers. CRI and WUI are fairly volatile, while the line for CDS spreads is fairly smooth with the exception of the spike around the Lehman Brothers collapse.

Note: Panels (a), (b), and (c) illustrate the country-specific geopolitical risk (CGPR) indices and other risk indices for Poland, the United Kingdom, and South Korea, respectively, covering the period from 2002:Q1 to 2023:Q4. In each panel, the left charts, from top to bottom, display CGPR from Caldara and Iacoviello (2022) (CGPRN), CGPR constructed by applying textual analysis to earnings-call transcripts using the NL Analytics platform (CGPRT ), and a sub-index of CGPRT constructed based solely on earnings-call transcripts of financial firms (CGPRT (fin). The right charts display the country risk index (CRI) by Hassan et al. (2023) (top), the World Uncertainty Index (WUI) by Ahir et al. (2022) (middle), and the five-year CDS spread (bottom) for the respective countries. All indices are standardized by their respective standard deviations within the sample.

Return to text.


Figure A.2: Bank-specific Geopolitical Risk Indices

The figure has two panels. Each panel plots a bank-specific GPR index over time evaluated at different percentiles, namely at the 25th, 50th and 75th percentiles.

Panel A: $$BGPR^N$$ This panel shows the bank-specific BGPR index that is constructed based on the country-specific GPR index from Caldara and Iacoviello (2022). The x-axis ranges from 1985:Q1 to 2023:Q4. The y-axis ranges from -2 to +2. The line for the 25th percentile is very flat at a value of around -1.8. The line for the 50th percentile lies naturally between the line for the 25th and 75th percentile. The line depicting values at the 75th percentile is much more volatile compared to the other lines and takes values from below -1 to nearly 2.

Panel A: $$BGPR^T$$ This panel shows the bank-specific BGPR index that is constructed based on the country-specific GPR index derived from earnings call transcripts. The x-axis ranges from 2002:Q1 to 2023:Q4. The y-axis ranges from -1 to +3. The line for the 25th percentile is very flat at a value of around -1.6. The line for the 50th percentile lies naturally between the line for the 25th and 75th percentile but is much closer to the 25th percentile line. The line depicting values at the 75th percentile is more volatile compared to the other lines and takes values from below -1.6 to nearly 3. All lines move up with the Russia-Ukraine war in 2022:Q2.

Note: Panels (a) and (b) show the bank-specific geopolitical risk (BGPR) indices constructed based on Equation (1) using CGPRN and CGPRT , respectively, over the periods of 1985:Q1 through 2023:Q4 and 2002:Q1 through 2023:Q4. See the notes under Appendix Figure A.1 for sources and definitions of the CGPR indices. Each panel illustrates the BGPR indices at the 25th, 50th, and 75th percentile. Data sources: FFIEC 009, FR Y-9C, and Call Reports.

Return to text.


Figure B.3: Banks' Cross-border and Local Exposures to Russia

Note: The figure illustrates cross-border claims (blue) and local claims (red) on Russia by the U.S. banking sector in Panel (a) and all BIS-reporting banking sectors in Panel (b). The vertical lines denote three geopolitical events: Russia's conflict with Georgia in 2008:Q3, Russia's annexation of Crimea in 2013:Q4, and Russia's invasion of Ukraine in 2022:Q1. Data sources: BIS Consolidated Banking Statistics and FFIEC 009.

Return to text.