## wp17115

## Source details

**Canonical URL:** [wp17115](https://www.imf.org/-/media/files/publications/wp/2017/wp17115.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/wp/2017/wp17115.pdf.md)
- [Structured JSON version](/-/media/files/publications/wp/2017/wp17115.pdf.json)

---

### Introduction — key results
- Documents evolution of cross-border holdings of financial assets and liabilities (international financial integration) since the global financial crisis and relates it to: (i) the global financial crisis and associated bank deleveraging; (ii) the euro area crisis of 2010–2012 and scaled down cross-border integration within the euro area; (iii) the post-crisis boom in emerging markets and subsequent fading; and (iv) the rapid increase in China’s size and role in the global economy.
- Main summary findings:
  - Growth in cross-border positions in relation to world GDP has come to a halt. Primary drivers:
    - Much weaker capital flows to and from advanced economies, especially financial centers, including diminished cross-border activity by banks in advanced economies and within the euro area.
    - Increase in the relative weight of emerging economies in global GDP, which have lower external assets and liabilities relative to advanced economies.
  - Cross-border FDI positions continued to expand, in contrast to positions in portfolio instruments and other investment; the FDI increase is primarily explained by FDI positions vis-à-vis financial centers and the role of special purpose vehicles (SPEs).
  - International assets and liabilities intermediated by financial centers complicate distinguishing “genuine” financial integration from positions reflecting multinational corporate structures or domiciliation of investment fund vehicles.
  - The share of domestic debt securities held by nonresidents is negatively related to the size of the domestic debt market and positively related to GDP per capita.
  - Global trend documented towards a higher share of nonresident ownership of debt securities issued by a country’s residents.

### The dataset — scope, construction, and classifications
- Coverage and period:
  - Estimates of external assets and liabilities (international investment position) for over 210 economies, for the period 1970–2015.
  - Close to 150 economies currently report IIP statistics to the IMF (up from 113 economies in 2007).
- Asset/liability categories follow BPM6: foreign direct investment; portfolio equity; portfolio debt; other investment; financial derivatives; plus foreign exchange reserves on the asset side (gold excluded).
- Estimation methods when IIP data are missing:
  - Portfolio debt liabilities: stock of debt securities held by nonresidents from external debt statistics; BIS data on international debt securities issued by residents; IMF CPIS holdings; World Bank data on foreign-held long-term bonds.
  - Portfolio debt assets: rely primarily on CPIS; partner-country datasets (U.S., Japan) are limited and may include reserves.
  - For small financial centers lacking full IIP reporting (e.g., Cayman Islands, Guernsey, Jersey), estimates use reported portfolio assets, mirror data, and specific external asset/liability series; for such centers only total external assets and liabilities are reported (not net positions).
- Country group aggregation for analysis:
  1. Financial centers
  2. Other advanced economies
  3. Emerging and developing economies
- Extended split of debt claims/liabilities into portfolio debt and other investment for the most recent twenty-year period.

### Dynamics of international financial integration — broad trends and group-by-group patterns
- Global aggregate:
  - After expansion up to a peak in 2007, cross-border positions have declined slightly in relation to world GDP; data coverage increases may understate the decline.
  - Financial centers were disproportionately large pre-crisis and remain so in external asset holdings.
- Reported shares and magnitudes:
  - 2007: financial centers ≈ 10 percent of world GDP but over 43 percent of global financial assets.
  - 2015: financial centers’ share in world GDP declined to 8 percent, but their global share of external assets remained around 43 percent.
  - Emerging and developing economies: 2007 — some 30 percent of world GDP and 10 percent of cross-border financial assets; 2015 — around 40 percent of world GDP and 13 percent of external assets.
- Instrumental shifts (2007–2015):
  - Contraction in debt instruments relative to world GDP.
  - Large increase in FDI and a smaller increase in financial derivatives.
- Stylized facts by group:
  - Financial centers:
    - Sharp contraction in “other investment” cross-border claims and liabilities (loans and deposits).
    - Large increase in FDI claims and liabilities; FDI now the largest component of their external portfolio (some 350 percent of these economies’ GDP in aggregate).
    - Aggregate position: net creditors with positive net FDI and sizable foreign exchange reserves; net creditor position in portfolio debt and net debtor position in portfolio equity.
  - Other advanced economies:
    - Shrinking cross-border positions in other investment.
    - Portfolio debt liabilities moderated by build-up of foreign exchange reserves by emerging economies.
    - Net position shows gradual deterioration driven by the United States; other advanced economies’ net positions improved.
  - Emerging and developing economies:
    - Lower stock of external assets and liabilities relative to GDP (around 60 percent of GDP).
    - Foreign exchange reserves large on the asset side; rising FDI on liabilities.
    - Net positions trended toward improvement; China is a net creditor and accounts for a growing share of group GDP (close to 40 percent in 2015, up from 22 percent in 2007).

### External balance sheet and global macro‑financial drivers
- Main drivers since the crisis:
  - Advanced economies and financial centers: large reduction in cross-border banking activity; euro area crisis and scaling down of intra-euro-area integration.
  - Emerging/developing economies: post-crisis EM boom and capital inflows (and later volatility), and China’s rapid growth and rising weight in global GDP.
- Banks:
  - Pre-crisis large international banks used internal channels to arbitrage funding conditions; substantial retreat from cross-border banking by major global banks (especially euro area, UK, Switzerland).
  - BIS-reporting banks’ external claims (locational basis): 2007 — accounted for some 56 percent of world GDP (and 28 percent of global external assets); 2015 — accounted for 36 percent of world GDP and less than 20 percent of global external claims.
  - Decline in claims between 2007 and 2015 concentrated in advanced economies and financial centers; claims on emerging and developing economies rose in dollar terms but declined as share of their GDP.
  - Regional heterogeneity: Japanese, Canadian, Australian, and Scandinavian banks increased cross-border claims as a share of their GDP; Chinese banks expanded overseas; foreign bank presence via locally-funded subsidiaries declined less than cross-border banking.
- Euro area crisis:
  - 2010–2012 crisis amplified scaling back of intra-euro-area financial integration; redenomination risk, national sovereign spreads, and capital controls mattered.
  - Intra-euro-area holdings continued to rise primarily via FDI and non-bank other investment; portfolio debt and bank loans shrank as a share of euro-area GDP.
- FDI measurement issues:
  - Financial centers’ role in global FDI:
    - End-2007: residents of financial centers held some 43 percent of world’s FDI claims abroad; FDI in financial centers accounted for around 40 percent of global FDI.
    - End-2014: FDI claims by financial centers rose to half of the world total; liabilities to 44 percent.
    - For these centers, claims and liabilities are each three times GDP in relation to size.
  - Special Purpose Entities (SPEs):
    - SPEs defined per OECD (2008): entities with “little or no employment; or operations, or physical presence in the jurisdiction in which they are created by their parent enterprises which are typically located in other jurisdictions (economies)”.
    - Netherlands and Luxembourg: vast majority of their FDI claims and liabilities are SPEs (over ¾ for Netherlands; over 90 percent for Luxembourg).
    - Total FDI claims by SPEs for Netherlands and Luxembourg grew by over $3.5 trillion between 2007 and 2014—over ¼ of the increase in global FDI claims during the same period.
  - Domiciliation/re-domiciling can mechanically raise global FDI stock (example: U.S. firm moving headquarters to Ireland increased measured Irish FDI claims by $600 billion between 2007 and 2014; by 2014 Irish FDI claims exceeded 5 times Irish GDP).
  - Measurement and structural factors complicate interpretation of FDI growth as reflecting real cross-border production versus financial/structural reclassification.

### Emerging and developing economies and the role of China
- Mechanisms:
  - Rising share of global GDP of emerging economies (and China) mechanically lowers aggregate cross-border assets relative to world GDP because these economies have lower external assets/liabilities ratios.
  - Structural changes within emerging markets: increased portfolio liabilities (sovereign and corporate issuance), official-sector accumulation (reserves; sovereign wealth funds), and outward investment by corporations and institutional investors.
- China:
  - Ratio of external assets/liabilities to GDP in China did not change dramatically since pre-crisis, but China’s share of aggregate emerging-economy GDP rose from 22 percent in 2007 to close to 40 percent in 2015 (market exchange rates).
  - China played a major role in the increase in foreign exchange reserves and in the rise in inward/outward FDI and other investment for the emerging/developing group.
  - For portfolio instruments, changes were larger in other emerging economies due to relatively closed Chinese portfolio markets.

### Foreign holdings of portfolio debt securities — trends and determinants
- Aggregate developments (2007–2015):
  - Aggregate stock of cross-border holdings of portfolio debt instruments increased by over $6 trillion for 2007–2015.
  - Portfolio debt claims (excluding holdings as foreign exchange reserves) increased much less (around $2 trillion), highlighting the role of reserve holdings in the aggregate increase.
  - Emerging and developing economies saw increased foreign holdings of portfolio debt issued by residents; aggregate ratio of portfolio debt liabilities to GDP for emerging/developing economies rose by 2 percentage points between 2007 and 2015.
- Cross-country heterogeneity and covariates examined for foreign share of government debt securities:
  - Market size: log of bond market capitalization (U.S. dollars) — expected inverse correlation with foreign share.
  - Level of development: GDP per capita (U.S. dollars) — expected positive correlation.
  - Common currency / euro-area membership: three euro-area groups distinguished (core euro; euro crisis countries; new euro members).
  - Net external position (NIIP): net borrowers expected to have higher foreign share.
  - Capital controls: overall restrictions and bond-specific restrictions (3-year average) — expected to lower foreign share.
  - Securities holdings by central banks: higher central bank holdings expected to lower foreign share.
- Empirical findings (cross-sectional, change, and panel regressions summarized):
  - Foreign share of domestic government debt securities is:
    - Negatively correlated with market size (log bond market capitalization).
    - Positively correlated with level of development (GDP per capita).
    - Negatively correlated with net international investment position (NIIP).
  - Euro-area dummies: nonresidents tend to hold higher shares in “core euro” countries, conditional on fundamentals; effect stable between 2007 and 2015.
  - Change regressions: new euro members show higher increases in foreign share (switch from negative in 2007 to positive in 2015 for some groups).
  - Capital controls measures often statistically insignificant and sometimes with counterintuitive signs; within-country variations in general capital controls are significantly and negatively correlated with foreign share in some samples.
  - Central bank holdings: changes in central bank share are not systematically correlated with changes in foreign share.

### Empirical specification and selected regression results (preserved coefficients and statistics)
- Cross-sectional and panel framework:
  - FORSHARE_it = α + β Z_it + ε_it, where FORSHARE is the share of foreign investors in total holdings of domestic government debt securities (ratio of portfolio debt liabilities to total debt securities outstanding).
  - Cross-sectional specifications for 2007 and 2015; change specification ΔFORSHARE = α + ΔZ β + ε for 2007–2015.
  - Regressors Z include: market size (log bond market cap), GDP per capita, euro-area subgroup dummies, NIIP, capital controls (total and bond-specific), central bank securities holdings.
- Selected exact coefficients and statistics from tables (preserved as in source):
  - Table 1 (Size of Global Debt Securities' Markets: 2007 vs 2014) — selected entries:
    - Total debt securities (columns: 2007 value, pct. of group GDP, 2015 value, pct. of group GDP):
      - Financial centers: 193.7 18.8 225.6 17.8
      - non-FC advanced economies: 165.0 103.3 180.3 96.4
      - Emerging and development economies: 36.2 10.0 50.3 19.4
        - China: 48.1 3.0 70.5 10.7
        - Others: 32.7 7.0 37.2 8.7
      - Total: 132.1 132.1 133.6 133.6
    - General government debt securities:
      - Financial centers: 35.1 3.4 60.4 4.8
      - non-FC advanced economies: 63.0 39.5 91.9 49.1
      - Emerging and development economies: 23.5 6.5 21.3 8.2
        - China: 20.3 1.3 14.2 2.2
        - Others: 24.5 5.2 25.8 6.1
      - Total: 49.4 49.4 62.1 62.1
  - Table 2 (Share of Debt Securities held by Nonresidents: 2007 and 2015) — selected entries:
    - 2007, General government securities: 0.48 (Mean AE), 0.50 (Median AE), 0.31 (Mean EM), 0.27 (Median EM)
    - 2007, Total debt securities: 0.44 (Mean AE), 0.50 (Median AE), 0.26 (Mean EM), 0.25 (Median EM)
    - 2015, General government securities: 0.51 (Mean AE), 0.54 (Median AE), 0.36 (Mean EM), 0.35 (Median EM)
    - 2015, Total debt securities: 0.45 (Mean AE), 0.44 (Median AE), 0.35 (Mean EM), 0.40 (Median EM)
    - Observations: 2007 — 32, 25; 2015 — 33, 32
  - Table 3 (Drivers of Foreign Share, cross-sectional) — selected coefficient extracts:
    - Log GDP per capita: 0.11*** (col 1, t-stat 5.66); 0.23*** (col 2, t-stat 4.53); 0.10*** (col 3, t-stat 3.50); 0.09*** (col 4, t-stat 2.70); 0.11 (col 5, t-stat 1.09); 0.10* (col 6, t-stat 1.84).
    - Log market size: 0.071*** (col 2); 0.055*** (col 3); -0.095*** (col 4, t-stat -5.75); -0.042*** (col 5, t-stat -5.05); -0.050*** (col 6, t-stat -3.43).
    - Euro_core dummy: 0.42*** (col 1); 0.40*** (col 2); 0.37*** (col 3); 0.38*** (col 4).
    - New euro members: -0.23*** (col 1); -0.19*** (col 2); 0.10 (col 3); 0.13 (col 4).
    - Euro crisis dummy: 0.28*** (col 1); 0.26*** (col 2); 0.015 (col 3); 0.033 (col 4).
    - Ratio NFA/GDP: -0.14*** (col 1); -0.15*** (col 2); -0.034 (col 3); -0.12*** (col 4); -0.12** (col 5); -0.20 (col 6).
    - Observations (as presented): 51, 28, 23, 59, 28, 31. Adjusted R2 reported: 0.75, 0.89, 0.50, 0.54, 0.6, 0.22.
  - Table 4 (Drivers of Change, 2007–2015) — selected coefficients:
    - Log diff. market size (avg growth rate 2007–15): -0.10*** (All, t-stat -4.37); -0.04 (AE); -0.19*** (EM, t-stat -4.18).
    - New euro members dummy: 0.39*** (EM, t-stat 5.21).
    - Euro crisis dummy: -0.17** (All, t-stat -2.18); -0.16** (AE, t-stat -2.21).
    - Change in overall restrictions: -0.43*** (All, t-stat -3.33); -0.49*** (AE, t-stat -3.51); -0.34* (EM, t-stat -1.90).
    - Observations: 52 (All), 25 (AE), 23 (EM). Adjusted R2: 0.49 (All), 0.54 (AE), 0.37 (EM).
  - Table 5 (Panel regressions, 2006–2015) — selected coefficients:
    - Log market size: -0.071*** (All, t-stat -5.51); -0.032** (AE, t-stat -2.31); -0.16*** (EM, t-stat -8.09).
    - Ratio NFA/GDP: -0.0036 (All); -0.013 (AE); -0.038 (EM).
    - Bond restrictions: 0.079* (All, t-stat 1.71); -0.017 (AE); 0.13** (EM, t-stat 2.19).
    - Overall restrictions: -0.094 (All); -0.29* (AE, t-stat -1.74); -0.17** (EM, t-stat -2.26).
    - Real GDP growth: 0.64*** (All, t-stat 4.28); 0.51** (AE, t-stat 2.02); 0.92*** (EM, t-stat 4.59).
    - New euro members: 0.21*** (All, t-stat 4.79); -0.019 (AE).
    - Euro crisis dummy: -0.23*** (All, t-stat -9.51); -0.19*** (AE, t-stat -9.21).
    - Central bank share: -0.40*** (All, t-stat -3.94).
    - Observations: 559.00 (All), 242.00 (AE), 279.00 (EM). R2 (within): 0.29 (All), 0.44 (AE), 0.39 (EM).
- Notes: Robust standard errors; t-statistics in parentheses; significance: * p<0.10; ** p<0.05; *** p<0.01.

### Euro‑crisis and euro‑area membership effects
- Countries more severely affected by the euro crisis:
  - 2007: coefficient positive and significant, albeit lower (indicating a share about 10 percentage points lower than for core countries).
  - 2015: coefficient drops drastically and loses statistical significance.
  - In the 2007–2015 change regressions, the euro-area crisis dummy is negative and economically and statistically very significant.
- New euro members:
  - Coefficient much higher for 2015 than for 2007 (switches from negative to positive), reflecting increased integration of their securities markets with the rest of the euro area.
  - New euro members dummy positive and very significant in change-in-foreign-share regressions.

### Capital controls, central bank holdings, and time effects
- Capital controls:
  - Both measures of capital controls are generally statistically insignificant and occasionally with the “wrong” sign; measures are imperfect proxies of intensity (example: bond restrictions variable equates China and India with Indonesia despite different actual access).
  - Within-country variations in the general capital controls measure are significantly and negatively correlated with the foreign share in some samples (Table 4).
- Central bank holdings:
  - Changes in the share of central bank holdings of government securities are not systematically correlated with changes in the foreign share; largest central bank share changes were in Japan and the United Kingdom, where foreign share was roughly unchanged between 2007 and 2015.
- Panel time-dummy estimates (Figure 15):
  - Trend increase common across countries: about 12 percentage points for advanced economies between 2006 and 2014; almost 20 percentage points for emerging and developing economies over the same period.

### Broader conclusions, policy implications, and outlook
- Level and evolution of foreign holdings of government debt securities are explained by a parsimonious set of variables:
  - Advanced economies: euro area membership, the euro area debt crisis, and market size are important.
  - Emerging and developing economies: level of development and net external position are important; within countries, foreign share tends to rise with growth rate and capital account liberalization.
- Since the financial crisis:
  - Very fast growth in cross-border positions relative to global GDP has come to a halt, reflecting retrenchment of cross-border banking activity and increased weight of less-financially-integrated emerging and developing economies in global GDP.
  - Disproportionate role of financial centers persists; growing but modest role of emerging and developing economies.
  - Retrenchment in cross-border banking and modest portfolio position increases have been offset by rapidly increasing FDI positions.
- Data and measurement issues:
  - Increasing difficulty assessing external exposures due to size of cross-border asset trade intermediated by financial centers, pass-through financial vehicles, and re-domiciliation.
  - Scope for progress on improving consolidated and portfolio/FDI data.
- Implications for future integration:
  - International banking factors: regulatory tightening, exit of marginal banks, expansion by emerging-economy banks, fixed costs and consolidation, growing role of non-bank intermediaries.
  - Euro area prospects: banking union and capital markets union should support sustainable flows, but high legacy debts and tighter regulation limit scope for net increases in debt flows; area-wide common deposit insurance remains necessary to realize full potential.
  - Convergence baseline: plausible that emerging economies’ share in the global financial system will match rising share in global output as restrictions are removed; domestic financial development needed to expand capacity for foreign liabilities/assets.
  - Risks for emerging economies: elastic international flows magnify domestic policy errors; vulnerability to self-fulfilling investor pessimism; less liquid systems tolerate shifts in global portfolio weights less well.
  - Mitigating signs: increased reliance on domestic-currency liabilities, more flexible exchange rate regimes, and stronger policy frameworks have arguably increased resilience to external shocks.

### Appendix 1 — Data and country classification (selected items)
- Financial centers (selected listing): Bahrain; Belgium; Cyprus; Hong Kong S.A.R. of China; Ireland; Luxembourg; Macao S.A.R. of China; Malta; Netherlands; Singapore; Switzerland; United Kingdom; Andorra; Bahamas; Barbados; Bermuda; British Virgin Islands; Cayman Islands; Curaçao; Gibraltar; Guernsey; Isle of Man; Jersey; Mauritius; Netherlands Antilles; Panama; San Marino; Turks and Caicos.
- Other advanced economies (selected listing): Australia; Austria; Canada; Czech Republic; Denmark; Estonia; Finland; France; Germany; Greece; Iceland; Israel; Italy; Japan; Korea; Latvia; Lithuania; New Zealand; Norway; Portugal; Slovak republic; Slovenia; Spain; Sweden; Taiwan province of China; United States.
- Emerging and developing economies (selected listing): Afghanistan; Albania; Algeria; Angola; Argentina; Bangladesh; Belarus; Belize; Benin; Bhutan; Bolivia; Bosnia and Herzegovina; Botswana; Brazil; Bulgaria; Burkina Faso; Burundi; Cambodia; Cameroon; Cape Verde; Central African Rep.; Chad; Chile; China; Colombia; Costa Rica; Croatia; Côte d'Ivoire; Djibouti; Dominica; Dominican Republic; Ecuador; Egypt; El Salvador; Equatorial Guinea; Eritrea; Ethiopia; Fiji; Gabon; Gambia; Georgia; Ghana; Grenada; Guatemala; Guinea; Guinea-Bissau; Guyana; Haiti; Honduras; Hungary; India; Indonesia; Iran, Islamic Republic of; Iraq; Jamaica; Jordan; Kazakhstan; Kenya; Kiribati; Kosovo; Kuwait; Kyrgyz Republic; Lao People's Dem.Rep; Lebanon; Lesotho; Liberia; Libya; Macedonia; Madagascar; Malawi; Malaysia; Maldives; Mali; Marshall Islands; Mauritania; Mauritius; Mexico; Micronesia; Moldova; Mongolia; Montenegro; Montserrat; Morocco; Mozambique; Myanmar; Namibia; Nauru; Nepal; New Caledonia; Nicaragua; Niger; Nigeria; Oman; Pakistan; Palau; Panama; Papua New Guinea; Paraguay; Peru; Philippines; Poland; Qatar; Romania; Russia; Rwanda; Samoa; Saudi Arabia; Senegal; Serbia; Seychelles; Sierra Leone; Sint Maarten; Solomon Islands; Somalia; South Africa; South Sudan; Sri Lanka; St. Kitts and Nevis; St. Lucia; St. Vincent & Grenadines; Suriname; Swaziland; Syrian Arab Republic; São Tomé & Príncipe; Tajikistan; Tanzania; Thailand; Timor-Leste; Togo; Tonga; Trinidad and Tobago; Tunisia; Turkey; Turkmenistan; Tuvalu; Uganda; Ukraine; United Arab Emirates; Uruguay; Uzbekistan; Vanuatu; Venezuela, Rep. Bol.; Vietnam; West Bank and Gaza; Yemen, Republic of; Zambia; Zimbabwe.
- Variable definitions (selected):
  - Market size: log of government bond market capitalization (U.S. dollars). Sources: BIS; Arslanalp and Tsuda (2014a, 2014b); national sources.
  - Level of development: Log GDP per capita (U.S. dollars). Source: IMF, World Economic Outlook database.
  - Euro core: dummy = 1 for Austria, Belgium, Finland, France, Germany, Luxembourg, Netherlands.
  - Euro crisis: dummy = 1 for Greece, Ireland, Italy, Portugal, Spain.
  - New euro members: dummy = 1 for Cyprus, Estonia, Latvia, Lithuania, Malta, Slovak Republic, Slovenia for all years they were part of the euro and the year before joining.
  - Ratio of NFA to GDP: net external position divided by domestic GDP. Source: Lane and Milesi-Ferretti, External Wealth of Nations database.
  - Bond restrictions / Overall restrictions: indices from Fernandez and others (2015); 3-year averages used; dataset available through 2013 (2013 used for 2014 and 2015 observations).
  - Central bank share: central bank holdings of domestic government debt securities as a share of total government debt outstanding. Sources: Arslanalp and Tsuda (2014a, 2014b), IFS, national sources.

*Source: wp17115 - Appendix Tables (excerpted content).*

### References................................................................................................ 21

### wp17115 - References................................................................................................ 21

### Tables
- 1. The Size of Global Debt Securities' Markets: 2007 vs 2014 .............................................28
- 2. Share of Debt Securities held by Nonresidents: 2007 and 2015 ........................................29
- 3. Drivers of Foreign Share of Government Debt Securities, 2007 vs 2015
  Cross-sectional Regressions...................................................................................30
- 4. Drivers of Change in Foreign Share of Government Debt Securities, 2007-2015 ............31
- 5. Drivers of Foreign Share of Government Debt Securities, 2006-2015
  Panel Regressions....................................     ..............................................................32     

### Figures
- 1. Global External Assets and Liabilities.................................... ...............................33
- 2. The Composition of World GDP.................................... .......................................34
- 3. Changes in External Assets, Liabilities and World GDP, 2007-2015 ...............................35
- 4. Changes in External Assets and Liabilities, 2007-2015.................................... ....36
- 5. Gross and Net External Positions, Advanced Economies Excluding
  Financial Centers.................................... ...............................................................37
- 6. Net External Position: United States and Other Advanced Economies
  Excluding Financial Centers and United States.................................... .................38
- 7. Gross and Net External Positions, Financial Centers.................................... ........39
- 8. Gross and Net External Positions, Emerging and Developing Economies .......................40
- 9. Net External Position: China and Other Emerging and Developing Economies ...............41
- 10. External Assets of BIS-reporting Banks.................................... ............................42
- 11. Claims of BIS-reporting Banks by Destination.................................... .................43
- 12. Intra-euro Area Assets.................................... .......................................................44
- 13. Emerging and Developing Economies: Changes in Share of External Assets,
  Liabilities, and GDP to Global GDP......................................................................45
- 14. Emerging Markets: Change in Share of Portfolio Liabilities to GDP... ...........................46
- 15. Coefficients on Time Series Dummies.................................... ..............................47

### Appendices
- 1. Data and Country Classification.................................... ........................................48
- 2. FDI in Financial Centers: the cases of Ireland and the Netherlands ..................................50

*wp17115 - References................................................................................................ 21*

### Appendix Tables

### Appendix Tables

### Introduction — key results
- The paper documents evolution of cross-border holdings of financial assets and liabilities (international financial integration) since the global financial crisis and relates it to major macroeconomic and financial developments: (i) the global financial crisis and associated bank deleveraging; (ii) the euro area crisis of 2010–2012 and scaled down cross-border integration within the euro area; (iii) the post-crisis boom in emerging markets and subsequent fading; and (iv) the rapid increase in China’s size and role in the global economy.
- Main summary findings:
  - Growth in cross-border positions in relation to world GDP has come to a halt. Primary drivers:
    - Much weaker capital flows to and from advanced economies, especially financial centers, including diminished cross-border activity by banks in advanced economies and within the euro area.
    - Increase in the relative weight of emerging economies in global GDP, which have lower external assets and liabilities relative to advanced economies.
  - Cross-border FDI positions continued to expand, in contrast to positions in portfolio instruments and other investment. The FDI increase is primarily explained by FDI positions vis-à-vis financial centers and the role of special purpose vehicles (SPEs).
  - The disproportionate role of international assets and liabilities intermediated by financial centers complicates distinguishing “genuine” financial integration from positions reflecting multinational corporate structures or domiciliation of investment fund vehicles.
  - The share of domestic debt securities held by nonresidents is negatively related to the size of the domestic debt market and positively related to GDP per capita. The euro-area crisis altered the impact of a common currency on foreign holdings of domestic debt securities, but primarily for countries more severely affected by the crisis.
  - Documented a global trend towards a higher share of nonresident ownership of debt securities issued by a country’s residents.

### The dataset — scope, construction, and classifications
- Coverage and period:
  - Estimates of external assets and liabilities (international investment position) for over 210 economies, for the period 1970–2015.
  - Increased coverage relative to prior versions; close to 150 economies currently report IIP statistics to the IMF (up from 113 economies in 2007).
- Asset/liability categories follow BPM6: foreign direct investment; portfolio equity; portfolio debt; other investment; financial derivatives; plus foreign exchange reserves on the asset side (gold excluded).
- Estimation methods when IIP data are missing:
  - Portfolio debt liabilities estimated using: stock of debt securities held by nonresidents from external debt statistics; BIS data on international debt securities issued by residents; IMF CPIS holdings; and World Bank data on foreign-held long-term bonds.
  - Portfolio debt assets rely primarily on CPIS; partner-country datasets (U.S., Japan) are limited and may include reserves.
  - For small financial centers lacking full IIP reporting (e.g., Cayman Islands, Guernsey, Jersey), estimates use reported portfolio assets, mirror data, and specific external asset/liability series; for such centers only total external assets and liabilities are reported (not net positions) due to measurement error.
- Country group aggregation for analysis:
  1. Financial centers (Belgium; Hong Kong S.A.R.; Ireland; Luxembourg; Netherlands; Singapore; Switzerland; United Kingdom; Mauritius; Panama; Bermuda; Cayman Islands; etc.)
  2. Other advanced economies
  3. Emerging and developing economies
- Extended split of debt claims/liabilities into portfolio debt and other investment for the most recent twenty-year period.

### Dynamics of international financial integration — broad trends and group-by-group patterns
- Global aggregate:
  - After expansion up to a peak in 2007, cross-border positions have declined slightly in relation to world GDP; data coverage increases may understate the decline.
  - Financial centers played an outsized role pre-crisis and remain disproportionately large in external asset holdings.
- Shares and magnitudes (as reported in the source):
  - As of 2007: financial centers accounted for close to 10 percent of world GDP but over 43 percent of global financial assets.
  - By 2015: financial centers’ share in world GDP declined to 8 percent, but their global share of external assets remained around 43 percent.
  - Emerging and developing economies: accounted for some 30 percent of world GDP in 2007 but only 10 percent of cross-border financial assets; by 2015 their world GDP share increased to around 40 percent, but their share of external assets expanded only to 13 percent.
- Instrumental shifts between 2007 and 2015:
  - Contraction in debt instruments relative to world GDP.
  - Large increase in FDI and a smaller increase in financial derivatives.
- Stylized facts by country group:
  - Financial centers:
    - Sharp contraction in “other investment” cross-border claims and liabilities (loans and deposits), linked to downsizing of cross-border banking.
    - Large increase in FDI claims and liabilities; FDI now the largest component of their external portfolio (some 350 percent of these economies’ GDP in aggregate).
    - Aggregate position: net creditors with positive net FDI and sizable foreign exchange reserves; net creditor position in portfolio debt and net debtor position in portfolio equity (reflecting investment fund domiciliation effects).
  - Other advanced economies:
    - Shrinking cross-border positions in other investment.
    - Portfolio debt liabilities moderated by build-up of foreign exchange reserves by emerging economies.
    - Net position shows gradual deterioration driven by the United States; other advanced economies’ net positions improved.
  - Emerging and developing economies:
    - Lower stock of external assets and liabilities relative to GDP (around 60 percent of GDP).
    - Foreign exchange reserves large on the asset side; rising FDI on liabilities.
    - Net positions trended toward improvement; China is a net creditor and accounts for a growing share of group GDP (close to 40 percent in 2015, up from 22 percent in 2007).

### External balance sheet and global macro-financial trends — drivers
- Main drivers since the crisis:
  - Advanced economies and financial centers: large reduction in cross-border banking activity; euro area crisis and scaling down of intra-euro-area integration.
  - Emerging/developing economies: post-crisis EM boom and capital inflows (and later volatility), and China’s rapid growth and rising weight in global GDP.
- Banks:
  - Large international banks were central to pre-crisis capital flows, using internal channels to arbitrage funding conditions.
  - Result: substantial retreat from cross-border banking by major global banks (especially euro area, UK, Switzerland).
  - BIS-reporting banks’ external claims (locational basis):
    - 2007: accounted for some 56 percent of world GDP (and 28 percent of global external assets).
    - 2015: accounted for 36 percent of world GDP and less than 20 percent of global external claims.
  - Decline in claims between 2007 and 2015 concentrated in advanced economies and financial centers; claims on emerging and developing economies rose in dollar terms but declined as share of their GDP.
  - Regional heterogeneity: Japanese, Canadian, Australian, and Scandinavian banks increased cross-border claims as a share of their GDP; Chinese banks expanded overseas; foreign bank presence via locally-funded subsidiaries declined less than cross-border banking.
- Euro area crisis:
  - 2010–2012 crisis amplified scaling back of intra-euro-area financial integration; redenomination risk, national sovereign spreads, and capital controls (e.g., Cyprus, Greece) mattered.
  - Intra-euro-area holdings continued to rise primarily on account of FDI and non-bank other investment; portfolio debt and bank loans shrank as a share of euro-area GDP.
- FDI measurement issues:
  - Financial centers’ role in global FDI:
    - End-2007: residents of financial centers held some 43 percent of world’s FDI claims abroad; FDI in financial centers accounted for around 40 percent of global FDI.
    - End-2014: FDI claims by financial centers rose to half of the world total; liabilities to 44 percent.
    - For these centers, claims and liabilities are each three times GDP in relation to size.
  - Special Purpose Entities (SPEs):
    - SPEs: entities with “little or no employment; or operations, or physical presence in the jurisdiction in which they are created by their parent enterprises which are typically located in other jurisdictions (economies)” (OECD, 2008).
    - Netherlands and Luxembourg: the vast majority of their FDI claims and liabilities are SPEs (over ¾ for Netherlands; over 90 percent for Luxembourg).
    - Total FDI claims by SPEs for Netherlands and Luxembourg grew by over $3.5 trillion between 2007 and 2014—over ¼ of the increase in global FDI claims during the same period.
  - Domiciliation/re-domiciling of multinationals can mechanically raise global FDI stock (example: U.S. firm moving headquarters to Ireland increased measured Irish FDI claims by $600 billion between 2007 and 2014; by 2014 Irish FDI claims exceeded 5 times Irish GDP).
  - These measurement and structural factors complicate interpretation of FDI growth as reflecting real cross-border production versus financial/structural reclassification.

### Emerging and developing economies and the role of China
- Mechanisms:
  - Rising share of global GDP of emerging economies (and China) mechanically lowers aggregate cross-border assets relative to world GDP because these economies have lower external assets/liabilities ratios.
  - Structural changes within emerging markets: increased portfolio liabilities (sovereign and corporate issuance), official-sector accumulation (reserves; sovereign wealth funds), and outward investment by corporations and institutional investors.
- China:
  - Ratio of external assets/liabilities to GDP in China did not change dramatically since pre-crisis, but China’s share of aggregate emerging-economy GDP rose from 22 percent in 2007 to close to 40 percent in 2015 (market exchange rates).
  - China played a major role in the increase in foreign exchange reserves and in the rise in inward/outward FDI and other investment for the emerging/developing group.
  - For portfolio instruments, changes were larger in other emerging economies due to relatively closed Chinese portfolio markets.

### Foreign holdings of portfolio debt securities — trends and determinants
- Portfolio debt developments since 2007:
  - Aggregate stock of cross-border holdings of portfolio debt instruments increased by over $6 trillion for 2007–2015.
  - Portfolio debt claims (excluding holdings as foreign exchange reserves) increased much less (around $2 trillion), highlighting the role of reserve holdings in the aggregate increase.
  - Emerging and developing economies saw increased foreign holdings of portfolio debt issued by residents; aggregate ratio of portfolio debt liabilities to GDP for emerging/developing economies rose by 2 percentage points between 2007 and 2015.
- Cross-country heterogeneity in foreign penetration of debt markets:
  - Country-level drivers examined for foreign share of government debt securities in 2007 and 2015, and changes between 2007–2015.
  - Key covariates considered:
    - Market size (log of bond market capitalization in U.S. dollars): expected inverse correlation with foreign share.
    - Level of development (GDP per capita in U.S. dollars): expected positive correlation.
    - Common currency (especially euro-area membership): nuanced effects; three euro-area groups distinguished: (i) economies more severely affected by euro crisis (Greece, Ireland, Italy, Portugal, Spain); (ii) smaller economies that joined euro in 2007 or thereafter (Cyprus, Estonia, Latvia, Lithuania, Slovak Republic, Slovenia); (iii) remainder (Austria, Belgium, Finland, France, Germany, Luxembourg, Netherlands).
    - Net external position (NIIP): net borrowers expected to have higher foreign share.
    - Capital controls: overall restrictions on inflows and specific restrictions on bond inflows (average of preceding 3 years) expected to lower foreign share.
    - Securities holdings by central banks: higher central bank holdings expected to lower foreign share.
- Empirical findings (summary of regression results presented in the source):
  - The share of domestic government debt securities owned by nonresidents is:
    - Negatively correlated with market size (log bond market capitalization).
    - Positively correlated with level of development (GDP per capita).
    - Negatively correlated with net international investment position (NIIP).
  - Euro-area dummies are consistent with priors: nonresidents tend to hold higher shares in “core euro” countries, conditional on fundamentals; the coefficient is stable between 2007 and 2015.
  - Results for total debt securities are analogous (presented separately in the appendix of the source).

### Empirical specification (as presented in the source)
- Cross-sectional and panel regression framework:
  - FORSHARE_it = α + β Z_it + ε_it, where FORSHARE is the share of foreign investors in total holdings of domestic government debt securities (ratio of portfolio debt liabilities to total debt securities outstanding).
  - Cross-sectional specifications estimated for 2007 and 2015; change specification ΔFORSHARE = α + ΔZ β + ε to explain within-country changes between 2007 and 2015.
- Regressors Z include: market size (log bond market cap), GDP per capita, common currency dummies (euro-area subgrouping), NIIP, capital controls (total and bond-specific), and central bank securities holdings; definitions and sources provided in the Appendix of the source.

### Analysis implications and interpretation (from the source)
- Composition effects matter: rising weight of emerging markets and China in world GDP reduces aggregate cross-border holdings relative to world GDP independent of changes in advanced-economy behavior.
- Bank deleveraging materially reduced cross-border bank claims and global intermediated flows; this was partly offset by increased issuance in international bond markets.
- Financial centers remain central to global cross-border positions; much of measured FDI growth reflects SPEs and corporate domiciliation effects rather than conventional production-reallocation FDI.
- The rise in foreign holdings of government debt is driven by a combination of reserve accumulation (safe-asset demand), QE and central bank holdings, and increased issuance by emerging-market sovereigns and corporates; cross-country variation is shaped by market size, income, NIIP, capital controls, and currency arrangements.

*Source: wp17115 - Appendix Tables (excerpted content).*

### 2015. For the countries more severely affected by the euro crisis the coefficient for 2007 is

### wp17115 - 2015. For the countries more severely affected by the euro crisis the coefficient for 2007 is

### Euro-crisis effects on foreign share coefficients
- For countries more severely affected by the euro crisis the coefficient for 2007 is positive and significant, albeit lower (indicating a share about 10 percentage points lower than for core countries).
- The coefficient drops drastically for the 2015 regression, and loses statistical significance.
- A dummy for euro area countries most affected by the crisis (in the 2007–2015 change regressions) is negative and economically and statistically very significant.

### New euro members and market integration
- For smaller countries that joined the euro area later the coefficient is much higher for 2015 than for 2007 (it switches from negative in the first period to positive in the second period), reflecting increased integration of their securities markets with those in the rest of the euro area.
- The dummy for the new euro area members is positive and very significant in the change-in-foreign-share regressions.

### Capital controls and related measures
- Both measures of capital controls are generally statistically insignificant and occasionally with the “wrong” sign.
- Interpretation: these measures are poor proxies of the intensity of capital controls; example given for 2015:
  - the bond restrictions variable takes the same value for China and India (where foreign access is severely restricted and foreign share is tiny) and Indonesia (where nonresidents hold 65 percent of outstanding government debt securities).
- Within-country variations in the more general measure of capital controls are significantly and negatively correlated with the foreign share for all three samples (Table 4 result).

### Central bank holdings and foreign share changes
- Changes in the share of central bank holdings of government securities are not systematically correlated with changes in the foreign share.
- Possible reason: relatively limited number of countries with sizable variations in central bank share; largest changes were for Japan and the United Kingdom, both countries in which the foreign share was roughly unchanged between 2007 and 2015.

### Panel regressions (2006–2015) and time effects
- Table 5 presents panel regressions for the period 2006–2015 including country dummies and time dummies.
- Includes a dummy that equals one for Greece, Ireland, Italy, Portugal, and Spain for the years 2011–2015 to capture the impact of the euro area debt crisis.
- The time-series (panel) results confirm key cross-sectional findings:
  - impact of market size on foreign holdings;
  - impact of the euro area debt crisis on foreign holdings;
  - foreign share rises when a country’s growth rate rises.
- Capital account restrictions reduction is associated with a rise in foreign share, though this result is not as robust and holds only for overall capital account restrictions.
- For the sample of advanced economies, evidence that the foreign share declines in countries where the share of central bank holdings is rising.
- Figure 15 time-dummy estimates show a trend increase common across countries:
  - about 12 percentage points for advanced economies between 2006 and 2014;
  - almost 20 percentage points for emerging and developing economies over the same period.

### Broader conclusions on cross-border holdings and integration
- Level and evolution of foreign holdings of government debt securities are well explained by a parsimonious set of variables:
  - For advanced economies: euro area membership, the euro area debt crisis, and market size are important.
  - For emerging and developing economies: level of development and the net external position are important cross-country determinants; within countries, the share tends to rise with growth rate and capital account liberalization.
- Since the financial crisis:
  - very fast growth in cross-border positions relative to global GDP has come to a halt, reflecting retrenchment of cross-border banking activity and increased weight of less-financially-integrated emerging and developing economies in global GDP.
  - disproportionate role of financial centers persists in total holdings; growing but modest role of emerging and developing economies.
  - retrenchment in cross-border banking and modest portfolio position increases have been offset by rapidly increasing FDI positions.
- Data and measurement issues:
  - increasing difficulty in properly assessing external exposures due to size of cross-border asset trade intermediated by financial centers and the role of pass-through financial vehicles and re-domiciliation.
  - scope for progress on improving consolidated and portfolio/FDI data.

### Implications for the future of international financial integration
- International banking factors: regulatory tightening, exit of marginal banks, expansion by emerging-economy banks, fixed costs and consolidation, growing role of non-bank intermediaries.
- Euro area prospects: banking union and capital markets union should support sustainable flows, but high legacy private and public debts and tighter regulation limit scope for net increases in debt flows; area-wide common deposit insurance remains necessary to realize full potential.
- Convergence expectations:
  - plausible baseline that emerging economies’ share in the global financial system will match their rising share in global output as financial-account restrictions are removed.
  - domestic financial development should expand capacity for foreign liabilities and assets and reduce asymmetries between advanced and developing countries.
  - much outward expansion may be regionally focused given gravity factors; regionalization could rise with independent monetary policies or regional exchange-rate regimes and RMB internationalization.
- Risks for emerging economies:
  - elastic international financial flows magnify domestic policy errors;
  - vulnerability to self-fulfilling investor pessimism;
  - less liquid systems tolerate shifts in global portfolio weights less well.
- Mitigating signs:
  - increased reliance on domestic-currency liabilities, more flexible exchange rate regimes, and stronger policy frameworks have arguably increased resilience to external shocks.

*Source: wp17115 - 2015. For the countries more severely affected by the euro crisis the coefficient for 2007 is*

### References

### References

### Major cited topics and contributions
- Do Global Banks Spread Global Imbalances?; Asset-Backed Commercial Paper During the Financial Crisis of 2007-2009 — Acharya and Schnabl, IMF Economic Review No. 58, No. 1, pp. 37–73.
- Cross-Border Banking in Europe: Implications for Financial Stability and Macroeconomic Policies — Allen, Beck, Carletti, Lane, Schoenmaker, Wagner (2011), Centre for Economic Policy Research.
- Tracking Global Demand for Advanced Economy Sovereign Debt — Arslanalp and Tsuda, IMF Economic Review No. 62, No. 3 (2014a).
- Tracking Global Demand for Emerging Market Sovereign Debt — Arslanalp and Tsuda, IMF Working Paper 14/39 (2014b).
- Breaking Free of the Triple Coincidence in International Finance — Avdjiev, McCauley, Shin, Economic Policy, Vol. 31 (June) pp. 409–45 (2016).
- Cross-Country Exposures to the Swiss Franc — Benetrix and Lane, Trinity Economic Paper No. 01-16 (2016).
- What Does Measured FDI Actually Measure? — Blanchard and Acalin, Policy Brief 16-17, Peterson Institute (October 2016).
- Capital Flows and the Current Account: Taking Finance (More) Seriously — Borio and Disyatat, BIS Working Paper No. 525 (October 2015).
- Waves of International Banking Integration: A Tale of Regional Differences — Bouvatier and Delatte, European Economic Review, Vol. 80 (November 2015), pp. 354–73.
- Rethinking the Effects of Financial Globalization — Broner and Ventura, Quarterly Journal of Economics, Vol. 131, No. 3 (March) pp. 1497–1542 (2016).
- Cross-Border Banking and Global Liquidity — Bruno and Shin, Review of Economic Studies, Vol. 82, No. 2 (April) pp. 535–64 (2015).
- US Investment in Global Bonds: As the Fed Pushes, Some EMEs Pull — Burger, Sengupta, Warnock, Warnock, Economic Policy, Vol 30, No. 84, pp. 729–766 (2015).
- Local Currency Bond Markets — Burger and Warnock, IMF Staff Papers, Vol. 53 (Special Issue) pp. 133–46 (2006).
- Rational contagion and the globalization of securities markets — Calvo and Mendoza, Journal of International Economics, Vol. 51, No. 1 (June 2000) pp. 79–113.
- Banking Globalization and Monetary Transmission — Cetorelli and Goldberg, Journal of Finance, Vol. 67, No. 5, pp. 1811–43 (2012).
- The Impact of the Global Financial Crisis on Banking Globalization — Claessens and Van Horen, IMF Economic Review, Vol. 63, No. 4 (2015).
- EME Banking Systems and Regional Financial Integration — Committee on the Global Financial System, CGFS Papers No 51 (March 2014).
- What Do We Know About Base Erosion and Profit Shifting? — Dharmapala, CES IFO Working Paper 4612 (2014).
- Profit Shifting of U.S. Multinationals — Dowd, Landefeld, Moore, Joint Committee on Taxation (January 2016), unpublished.
- OECD Benchmark Definition of Foreign Direct Investment: Fourth Edition — Economic Co-operation and Development (2008).
- Renminbi Internationalization: Achievements, Prospects, and Challenges — Eichengreen and Kawai, Brookings Institution Press (2015).
- Capital Controls: A New Dataset — Fernandez, Klein, Rebucci, Schindler, Uribe, NBER Working Paper 20970 (February 2015).
- Why Do Foreigners Invest in the United States? — Forbes, Journal of International Economics, Vol. 80, No. 1, pp. 3–21 (2010).
- The Holders and Issuers of International Portfolio Securities — Galstyan, Lane, Mehigan, Mercado, Journal of the Japanese and International Economies, Vol. 42 (December 2016), pp. 100–08.
- The Financial Crisis and The Geography of Wealth Transfers — Gourinchas, Rey, Truempler, Journal of International Economics, Vol. 88, No. 2 (November 2012), pp. 266–83.
- External Adjustment, Global Imbalances and Valuation Effects — Gourinchas and Rey, Handbook of International Economics, Vol. 4, pp. 585–645 (2014).
- The Internationalisation of the Renminbi as an Investing and a Funding Currency: Analytics and Prospects — He, Luk, Zhang, HKIMR Working Paper No. 01/2015 (January 2015).
- Explaining Foreign Holdings of Asia’s Debt Securities: The Feldstein-Horioka Paradox Revisited — Horioka, Terada-Hagiwara, Nomoto, NBER Working Paper No. 21734 (November 2015).
- World Economic Outlook — International Monetary Fund, October 2014; October 2016.
- Dollar Funding and the Lending Behavior of Global Banks — Ivashina, Scharfstein, Stein, Quarterly Journal of Economics, Vol. 130, No. 3, pp. 1241–81 (2015).
- Private Equity Funds, Capital Flows and the Foreign-Exchange Market — Jayswal, Kornvig, Skjærbæk, Monetary Review, Central Bank of Denmark (October 2006).
- Financial Globalisation and the Crisis; Capital Flows in Europe; Cross-Border Financial Linkages: Identifying and Measuring Vulnerabilities — Lane (multiple pieces 2013a, 2013b, 2014).
- The External Wealth of Nations Mark II; The Drivers of Financial Globalization; International Investment Patterns; Cross-Border Investment in Small International Financial Centers; The Cross-Country Incidence of the Global Crisis; External Adjustment and the Global Crisis — Lane and Milesi-Ferretti (2007, 2008a, 2008b, 2011a, 2011b, 2012).
- Global Imbalances and External Adjustment after the Crisis — Lane (2015) in Global Liquidity: Spillovers to Emerging Markets and Policy Responses.
- The Evolving Role of China and India in the Global Financial System — Lane and Schmukler, Open Economies Review, Vol. 18, No. 4 pp. 499–520 (2007).
- Globalization and Emerging Markets: With or Without Crash? — Martin and Rey, American Economic Review, Vol. 96, No. 5 (December) pp. 1631–651 (2006).
- The Great Retrenchment: International Capital Flows During the Global Financial Crisis — Milesi-Ferretti and Tille, Economic Policy, Vol. 26, No. 66, pp. 285–342 (2011).
- Does the Current Account Still Matter?; Financial Flows, Financial Crises, and Global Imbalances; Finance at Center Stage: Some Lessons of the Euro Crisis; Trilemmas and Tradeoffs: Living with Financial Globalization — Obstfeld (2012a, 2012b, 2013, 2015).
- Asian Capital Market Integration: Theory and Evidence — Park, ADB Economics Working Paper Series 351 (2013).
- Determinants of Financial Stress in Emerging Market Economies — Park and Mercado, Journal of Banking & Finance, Vol. 45(C) pp. 199-224 (2014).
- The Second Wave of Global Liquidity and Its Implications for Emerging Economies — Shin (2013), unpublished.
- The Future of International Liquidity and The Role of China — Taylor, NBER Working Paper 18771 (February 2013).

### Key tables — exact figures and descriptive notes

- Table 1. The Size of Global Debt Securities' Markets: 2007 vs 2014
  - Columns/labels present: 2007, 2015, pct. of group GDP, pct. of world GDP.
  - Total debt securities:
    - Financial centers: 193.7 18.8 225.6 17.8
    - non-FC advanced economies: 165.0 103.3 180.3 96.4
    - Emerging and development economies: 36.2 10.0 50.3 19.4
      - China: 48.1 3.0 70.5 10.7
      - Others: 32.7 7.0 37.2 8.7
    - Total: 132.1 132.1 133.6 133.6
  - General government debt securities:
    - Financial centers: 35.1 3.4 60.4 4.8
    - non-FC advanced economies: 63.0 39.5 91.9 49.1
    - Emerging and development economies: 23.5 6.5 21.3 8.2
      - China: 20.3 1.3 14.2 2.2
      - Others: 24.5 5.2 25.8 6.1
    - Total: 49.4 49.4 62.1 62.1
  - Source note: Bank for International Settlements, national sources, and Arslanalp and others (2014).

- Table 2. Share of Debt Securities held by Nonresidents: 2007 and 2015
  - Advanced Economies / Emerging Markets; Mean / Median
  - 2007:
    - General government securities: 0.48 (Mean AE), 0.50 (Median AE), 0.31 (Mean EM), 0.27 (Median EM)
    - Total debt securities: 0.44 (Mean AE), 0.50 (Median AE), 0.26 (Mean EM), 0.25 (Median EM)
    - Observations: 32, 25
  - 2015:
    - General government securities: 0.51 (Mean AE), 0.54 (Median AE), 0.36 (Mean EM), 0.35 (Median EM)
    - Total debt securities: 0.45 (Mean AE), 0.44 (Median AE), 0.35 (Mean EM), 0.40 (Median EM)
    - Observations: 33, 32
  - Source note: Bank for International Settlements, national sources, and Arslanalp and other (2014).

- Table 3. Drivers of Foreign Share of Government Debt Securities, 2007 vs 2015 (Cross-sectional Regressions)
  - Dependent variable: ratio of foreign holdings of general government debt securities to GDP in 2007 (columns 1-3) and 2015 (columns 4-6).
  - Selected coefficient extracts (preserved signs, significance markers, and t-statistics placement as in source):
    - Log GDP per capita: 0.11*** (column 1, t-stat 5.66); 0.23*** (column 2, t-stat 4.53); 0.10*** (column 3, t-stat 3.50); 0.09*** (col 4, t-stat 2.70); 0.11 (col 5, t-stat 1.09); 0.10* (col 6, t-stat 1.84).
    - Log market size: 0.071*** (col 2); 0.055*** (col 3); -0.095*** (col 4, t-stat -5.75); -0.042*** (col 5, t-stat -5.05); -0.050*** (col 6, t-stat -3.43); -0.025 (col ? t-stat -3.28) [note: table formatting shows multiple log market size entries across columns].
    - Euro_core dummy: 0.42*** (col 1); 0.40*** (col 2); 0.37*** (col 3); 0.38*** (col 4).
    - New euro members: -0.23*** (col 1); -0.19*** (col 2); 0.10 (col 3); 0.13 (col 4).
    - Euro crisis dummy: 0.28*** (col 1); 0.26*** (col 2); 0.015 (col 3); 0.033 (col 4).
    - Ratio NFA/GDP: -0.14*** (col 1); -0.15*** (col 2); -0.034 (col 3); -0.12*** (col 4); -0.12** (col 5); -0.20 (col 6).
    - Bond restrictions: 0.17 (col 1); 0.44** (col 2); 0.12 (col 3)...
    - Overall restrictions: -0.14 (col1); 0.32 (col2)...
    - Constant and Observations: Observations 51, 28, 23, 59, 28, 31 (table shows Observations51 2823 592831 — formatting preserved as in source).
    - Adjusted R2: 0.75, 0.89, 0.50, 0.54, 0.6, 0.22.
  - Note: Robust standard errors, t-statistics in parentheses. * p<0.10; ** p<0.05; *** p<0.01.

- Table 4. Drivers of Change in Foreign Share of Government Debt Securities, 2007-2015
  - Dependent variable: change in the ratio of foreign holdings of general government debt securities to GDP between 2007 and 2015.
  - Selected coefficients:
    - Log diff. market size (average growth rate 2007-15): -0.10*** (All, t-stat -4.37); -0.04 (AE); -0.19*** (EM, t-stat -4.18).
    - New euro members dummy: 0.39*** (EM, t-stat 5.21).
    - Euro crisis dummy: -0.17** (All, t-stat -2.18); -0.16** (AE, t-stat -2.21).
    - Change in overall restrictions: -0.43*** (All, t-stat -3.33); -0.49*** (AE, t-stat -3.51); -0.34* (EM, t-stat -1.90).
    - Observations: 52 (All), 25 (AE), 23 (EM).
    - Adjusted R2: 0.49 (All), 0.54 (AE), 0.37 (EM).
  - Note: Robust standard errors, t-statistics in parentheses. * p<0.10; ** p<0.05; *** p<0.01.

- Table 5. Drivers of Foreign Share of Government Debt Securities, 2006-2015 Panel Regressions
  - Dependent variable: ratio of foreign holdings of general government debt securities to GDP. Panel regressions with fixed country and time effects.
  - Selected coefficients:
    - Log market size: -0.071*** (All, t-stat -5.51); -0.032** (AE, t-stat -2.31); -0.16*** (EM, t-stat -8.09).
    - Ratio NFA/GDP: -0.0036 (All); -0.013 (AE); -0.038 (EM).
    - Bond restrictions: 0.079* (All, t-stat 1.71); -0.017 (AE); 0.13** (EM, t-stat 2.19).
    - Overall restrictions: -0.094 (All); -0.29* (AE, t-stat -1.74); -0.17** (EM, t-stat -2.26).
    - Real GDP growth: 0.64*** (All, t-stat 4.28); 0.51** (AE, t-stat 2.02); 0.92*** (EM, t-stat 4.59).
    - New euro members: 0.21*** (All, t-stat 4.79); -0.019 (AE).
    - Euro crisis dummy: -0.23*** (All, t-stat -9.51); -0.19*** (AE, t-stat -9.21).
    - Central bank share: -0.40*** (All, t-stat -3.94).
    - Observations: 559.00 (All), 242.00 (AE), 279.00 (EM).
    - R2 (within): 0.29 (All), 0.44 (AE), 0.39 (EM).
  - Note: t-statistics in parentheses. * p<0.1; ** p<0.05; *** p<0.01. The coefficients and standard errors for time dummies are plotted in Figure 15.

### Key figures and dataset descriptors (titles and sources preserved)
- Figure 1. Global External Assets and Liabilities (ratio of world GDP). Source: authors’ calculations based on revised and extended EWN database.
- Figure 2. The Composition of World GDP (shares of world GDP). Source: authors’ calculations based on World Economic Outlook database.
- Figure 3. Changes in External Assets, Liabilities and World GDP, 2007-2015 (trillions US $). Source: authors’ calculations based on revised and extended EWN database.
- Figure 4. Changes in External Assets and Liabilities, 2007-2015 (percent of world GDP). Source: authors’ calculations based on revised and extended EWN database.
- Figure 5. Gross and Net External Positions, Advanced Economies Excluding Financial Centers (ratio of group GDP). Source: authors’ calculations based on revised and extended EWN database.
- Figure 6. Net External Position: United States and Other Advanced Economies Excluding Financial Centers and United States (ratio of group GDP). Source: authors’ calculations based on revised and extended EWN database.
- Figure 7. Gross and Net External Positions, Financial Centers (ratio of group GDP). Source: authors’ calculations based on revised and extended EWN database.
- Figure 8. Gross and Net External Positions, Emerging and Developing Economies (ratio of group GDP). Source: authors’ calculations based on revised and extended EWN database.
- Figure 9. Net External Position: China and Other Emerging and Developing Economies (ratio of group GDP). Source: authors’ calculations based on revised and extended EWN database.
- Figure 10. External Assets of BIS-reporting Banks (ratio of global GDP). Source: Bank for International Settlements.
- Figure 11. Claims of BIS-reporting Banks by Destination (billions US dollars). Source: Bank for International Settlements.
- Figure 12. Intra-euro Area Assets (percentage of euro area GDP). Source: authors’ calculations based on revised and extended EWN database. Note: Intra-euro area assets are calculated as the sum of external assets of individual euro area countries minus the external assets of the euro area as a whole.
- Figure 13. Emerging and Developing Economies: Changes in Share of External Assets, Liabilities, and GDP to Global GDP 2007-2015. Source: authors’ calculations based on revised and extended EWN database.
- Figure 14. Emerging Markets: Change in Share of Portfolio Liabilities to GDP 2007-2015. Source: authors’ calculations based on revised and extended EWN database. Note: countries are ordered by the size of their total portfolio liabilities in U.S. dollars in 2015.
- Figure 15. Coefficients on Time Series Dummies (panel regressions, Table 5). Note: Figures depict point estimates and 2-standard-error bands for time dummies in the panel regressions reported in Table 5.

*Source: wp17115 - References.*

### Appendix 1. Data and Country Classification

### Appendix 1. Data and Country Classification

### A. Classification of Economies
- Financial centers: Bahrain, Belgium, Cyprus, Hong Kong S.A.R. of China, Ireland, Luxembourg, Macao S.A.R. of China, Malta, Netherlands, Singapore, Switzerland, United Kingdom, Andorra, Bahamas, Barbados, Bermuda, British Virgin Islands, Cayman Islands, Curaçao, Gibraltar, Guernsey, Isle of Man, Jersey, Mauritius, Netherlands Antilles, Panama, San Marino, Turks and Caicos.
- Other advanced economies: Australia, Austria, Canada, Czech Republic, Denmark, Estonia, Finland, France, Germany, Greece, Iceland, Israel, Italy, Japan, Korea, Latvia, Lithuania, New Zealand, Norway, Portugal, Slovak republic, Slovenia, Spain, Sweden, Taiwan province of China, United States.
- Emerging and developing economies: Afghanistan, I.R. of; Albania; Algeria; Angola; Anguilla; Antigua and Barbuda; Argentina; Armenia; Aruba; Azerbaijan; Bangladesh; Belarus; Belize; Benin; Bhutan; Bolivia; Bosnia and Herzegovina; Botswana; Brazil; Brunei Darussalam; Bulgaria; Burkina Faso; Burundi; Cambodia; Cameroon; Cape Verde; Central African Rep.; Chad; Chile; China; Colombia; Comoros; Congo, Dem. Rep. of; Congo, Republic of; Costa Rica; Croatia; Côte d'Ivoire; Djibouti; Dominica; Dominican Republic; Ecuador; Egypt; El Salvador; Equatorial Guinea; Eritrea; Ethiopia; Fiji; French Polynesia; Gabon; Gambia; Georgia; Ghana; Grenada; Guatemala; Guinea; Guinea-Bissau; Guyana; Haiti; Honduras; Hungary; India; Indonesia; Iran, Islamic Republic of; Iraq; Jamaica; Jordan; Kazakhstan; Kenya; Kiribati; Kosovo; Kuwait; Kyrgyz Republic; Lao People's Dem.Rep; Lebanon; Lesotho; Liberia; Libya; Lithuania; Macedonia; Madagascar; Malawi; Malaysia; Maldives; Mali; Marshall Islands; Mauritania; Mauritius; Mexico; Micronesia; Moldova; Mongolia; Montenegro; Montserrat; Morocco; Mozambique; Myanmar; Namibia; Nauru; Nepal; New Caledonia; Nicaragua; Niger; Nigeria; Oman; Pakistan; Palau; Panama; Papua New Guinea; Paraguay; Peru; Philippines; Poland; Qatar; Romania; Russia; Rwanda; Samoa; Saudi Arabia; Senegal; Serbia; Seychelles; Sierra Leone; Sint Maarten; Solomon Islands; Somalia; South Africa; South Sudan; Sri Lanka; St. Kitts and Nevis; St. Lucia; St. Vincent & Grenadines.; Suriname; Swaziland; Syrian Arab Republic; São Tomé & Príncipe; Tajikistan; Tanzania; Thailand; Timor-Leste; Togo; Tonga; Trinidad and Tobago; Tunisia; Turkey; Turkmenistan; Tuvalu; Uganda; Ukraine; United Arab Emirates; Uruguay; Uzbekistan; Vanuatu; Venezuela, Rep. Bol.; Vietnam; West Bank and Gaza; Yemen; Republic of; Zambia; Zimbabwe.

### B. Variable Definitions (Regressions in Section V)
- Market size: log of government bond market capitalization measured in U.S. dollars. Source: Bank for International Settlements; Arslanalp and Tsuda (2014a, 2014b); and national sources.
- Level of development: Log GDP per capita in U.S. dollars. Source: IMF, World Economic Outlook database.
- Euro core: dummy taking the value of 1 for Austria, Belgium, Finland, France, Germany, Luxembourg, and the Netherlands.
- Euro crisis: dummy taking the value of 1 for Greece, Ireland, Italy, Portugal, and Spain.
- New euro members: dummy taking the value of 1 for Cyprus, Estonia, Latvia, Lithuania, Malta, Slovak Republic, and Slovenia for all years they were part of the euro as well as for the year before joining.
- Ratio of NFA to GDP: net external position divided by domestic GDP. Source: Lane and Milesi-Ferretti, External Wealth of Nations database.
- Bond restrictions: index of restrictions on foreign purchases of debt securities issued by the country. To measure the intensity of restrictions we use the average level of the variable for the preceding 3 years. Since the dataset goes only until 2013, we use data for that year for the 2014 and 2015 observations. Source: Fernandez and others (2015).
- Overall restrictions: index of restrictions on foreign purchases of domestic assets. To measure the intensity of restrictions we use the average level of the variable for the preceding 3 years. Since the dataset goes only until 2013, we use data for that year for the 2014 and 2015 observations. Source: Fernandez, and others (2015).
- Central bank share: Central bank holdings of domestic government debt securities as a share of total government debt securities outstanding. Sources: Arslanalp and Tsuda (2014a, 2014b), International Financial Statistics, and national sources.

### C. Regression Sample (Section V)
- Advanced economies: Australia, Austria, Belgium, Canada, Cyprus, Czech Republic, Denmark, Finland, France, Germany, Greece, Hong Kong S.A.R., Ireland, Israel, Italy, Japan, Korea, Latvia, Netherlands, New Zealand, Norway, Portugal, Slovenia, Spain, Sweden, Switzerland, United Kingdom, United States.
- Emerging economies: Argentina, Bangladesh, Brazil, Bulgaria, Chile, China, Colombia, Costa Rica, Dominican Republic, Egypt, El Salvador, Georgia, Guatemala, Hungary, India, Indonesia, Malaysia, Mexico, Nigeria, Pakistan, Paraguay, Peru, Philippines, Poland, Romania, Russia, South Africa, Thailand, Turkey, Ukraine, Uruguay.

*Source: Appendix 1. Data and Country Classification (wp17115).*

---


_Source: https://www.imf.org/-/media/files/publications/wp/2017/wp17115.pdf_