## wpiea2023014-print-pdf

## Source details

**Canonical URL:** [wpiea2023014-print-pdf](https://www.imf.org/-/media/files/publications/wp/2023/english/wpiea2023014-print-pdf.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/wp/2023/english/wpiea2023014-print-pdf.pdf.md)
- [Structured JSON version](/-/media/files/publications/wp/2023/english/wpiea2023014-print-pdf.pdf.json)

---

### Overview of recent developments
- In Q3 2022, global central banks added US$20 billion of gold to their international reserve portfolios — the largest quarterly increase in official gold demand in fully 55 years according to the World Gold Council (2022).
- The increase reversed a multi-decade secular decline in gold’s share of global reserves that began earlier and started to reverse around the time of the 2008-9 Global Financial Crisis.
- The paper identifies which central banks and which economic, financial, and political factors account for the shift toward gold.

### Two principal hypotheses for increased official gold demand
- Safe-haven and portfolio-diversification hypothesis:
  - Gold is seen as a safe haven in periods of high economic, financial, and geopolitical uncertainty and when returns on reserve currencies are low.
  - Gold is popularly viewed as an inflation hedge and as a portfolio diversifier; it also benefits from custom and tradition.
  - Liquid markets for trading gold exist in London, New York, Shanghai, and a large over-the-counter market in Switzerland.
- Sanctions- and asset-freeze insulation hypothesis:
  - Gold is perceived as more insulated from financial sanctions and asset freezes than foreign-exchange reserves held abroad.
  - The G7 decision to freeze Bank of Russia foreign-exchange reserves and subsequent acceleration of Russian gold purchases (including repatriation and vaulting at home) focused attention on gold’s role as a sanctions-robust reserve asset.

### Costs and limitations of gold as a reserve asset
- Gold is expensive to transport, warehouse, and secure, and it does not bear interest.
- Gold can be lent out on term deposit and used in swaps, but counterparties typically expect gold to be vaulted at recognized depositories (Bank of England, Federal Reserve Bank of New York, or major exchange depositories), reintroducing sanctions risk.
- Repatriation of gold can limit its use in market transactions.

### Empirical findings on determinants of the gold share in reserves
- Historical inertia:
  - The gold share of reserves exhibits considerable inertia; lagged dependent variable coefficients are large and significant (e.g., Inertia 0.95***; ranges 0.80***–0.96*** across panels and subgroups).
- Relative returns and interest-rate effects:
  - Gold share responds positively to the return to stockpiling gold, proxied by the futures/spot price differential (gold basis; e.g., coefficient 0.70***).
  - Gold share responds negatively to the U.S. federal funds rate (shadow FFR), consistent with an opportunity cost relative to investing in short-term U.S. Treasuries (examples: -0.26* to -0.31***).
- Uncertainty, volatility, and geopolitical risk:
  - Gold share responds positively to Global Economic Policy Uncertainty (GEPU) and U.S. dollar volatility.
  - Some evidence of positive response to geopolitical risk (Caldara-Iacoviello acts subindex); advanced economies respond more to geopolitical acts while emerging markets respond more to economic policy uncertainty.
- Heterogeneity by country type:
  - Most regression results are driven by the emerging market subsample, consistent with emerging markets managing gold reserves more actively than advanced economies.
- Timing:
  - Price effects can raise gold share immediately; actual changes in gold volumes typically follow with a lag (roughly five months or more).

### Evidence on active diversifiers and magnitude of shifts
- Definition: “Active diversifiers” are countries that purchased gold and raised its share in reserves by at least 5 percentage points over the last two decades.
- Findings:
  - The paper identifies 14 active diversifiers; all 14 are emerging markets.
  - Eight active diversifiers purchased at least one million troy ounces: Kazakhstan, Belarus, Türkiye, Uzbekistan, Hungary, Iraq, Argentina, Qatar.
  - By contrast, an earlier IMF paper on foreign exchange reserves identified 46 active diversifiers (including advanced countries).
- Sanctions association:
  - Fully half of the largest year-over-year increases in central bank holdings of gold reserves since 2000 were associated with the risk of sanctions.
  - An indicator of financial sanctions imposed by the United States, United Kingdom, European Union, and Japan confirms sanctions have a positive impact on the share and volume of reserves held in gold.
  - Multilateral sanctions imposed jointly by these countries have a larger impact on gold shares than unilateral sanctions in share regressions; volume regressions show some specifications where unilateral sanctions have larger impacts.

### Aggregate econometric analysis: data, key variables, and summary statistics
- Monthly IFS data on gold’s share in global reserves (market valuation), 1980–2021; sample and covariates documented (inertia, gold returns, gold basis, U.S. policy rate (shadow FFR), U.S. CPI inflation, measures of volatility, GEPU, geopolitical acts subindex, liquidity proxies).
- Selected summary statistics (Table 5 excerpt):
  - Gold reserve share: N 504; Mean 0.221; S.D. 0.16; Min 0.0794; Max 0.652.
  - Gold reserve share (advanced): N 504; Mean 0.280; S.D. 0.159; Min 0.126; Max 0.727.
  - Gold reserve share (emerging and developing): N 504; Mean 0.120; S.D. 0.103; Min 0.0216; Max 0.373.
  - Gold basis: N 504; Mean 0.019; S.D. 0.018; Min -0.0166; Max 0.0903.
  - Gold return: N 504; Mean 0.065; S.D. 0.237; Min -0.375; Max 1.98.
  - Global Economic Policy Uncertainty: N 300; Mean 130; S.D. 67.9; Min 48.9; Max 430.
  - Geopolitical risk: N 504; Mean 72.5; S.D. 44.1; Min 20.4; Max 470.
  - U.S. Dollar appreciation: N 504; Mean 0.0336; S.D. 0.069; Min -0.105; Max 0.230.
  - Shadow FFR: N 504; Mean 0.0414; S.D. 0.044; Min -0.0299; Max 0.191.
  - Gold bid-ask spread: N 313; Mean 0.0114; S.D. 0.017; Min 0; Max 0.130.
  - US CPI inflation: N 504; Mean 0.0322; S.D. 0.025; Min -0.021; Max 0.148.

### Dynamic responses (local projections / impulse responses)
- GEPU shock:
  - A 1-point shock to GEPU produces a positive cumulative response in the gold share on impact, peaking after eight months.
  - Example calculation in text: GEPU rose by 140 points between February and March 2022; on impact this implies a 140*0.15 = 0.21 percent increase in the share of gold in central bank portfolios. The cumulative impact tops out in 7 months at about three times the immediate increase.
  - Volume response: holdings in fine troy ounces do not change on impact; buying begins about five months after the uncertainty event. A 140 point shock would result in a 0.14 percent increase (=140*0.1) in gold holdings (in volume) after 12 months.
- Geopolitical risk impulse:
  - Response peaks after 5–6 months and dissipates after a year.
- U.S. federal funds rate shock:
  - Response of gold share continues to grow between 6 and 12 months.

### Country-level determinants (annual panel, 144 economies, 1980–2021)
- Explanatory variables include: GDP growth, CPI inflation, fiscal balance, public-debt-to-GDP ratio, trade openness, local currency appreciation, domestic gold production, exchange rate regime indicator (floating = 1), financial sanctions indicators (USD, EUR, JPY, GBP), growth of reserve assets.
- Expected and observed relationships:
  - Gold shares tend to be higher when GDP growth is low and inflation is high (statistical significance varies by sample).
  - Trade openness associated with lower gold share (stronger for EMDEs).
  - Floating exchange rate associated with higher gold share.
  - Higher domestic gold production increases gold share.
  - Reserve growth associated with lower gold share, especially for advanced economies.
- Selected Tobit coefficient highlights (Table 10, All economies):
  - CPI inflation: 0.003*** (spec 1); 0.12** (spec 2); 0.27*** (spec 3); 0.21*** (spec 4).
  - Fiscal balance: 0.06** (spec 1); 0.16*** (spec 2); -0.46*** (spec 3); -0.47*** (spec 4).
  - Real GDP growth: -0.25***; -0.27***; -0.12; -0.01.
  - Trade openness: -0.02***; -0.02***; -0.03***; -0.03***.
  - Floating exchange rate: 0.07***.
  - Gold production: 0.06***; 0.04***.
  - Reserves growth (%): -0.01***; -0.01***.
  - N 3095; 2750; 1291; 1291; Pseudo R2 -0.74; -0.63; -0.28; -0.31.
- Advanced economies (Table 11) and EMDEs (Table 12) show differing coefficient signs/magnitudes; notable examples:
  - Advanced: Fiscal balance coefficients strongly negative (e.g., -1.67***), Reserves growth (%) negative (e.g., -0.10***).
  - EMDEs: CPI inflation positive and significant (e.g., 0.00***; 0.26***; 0.34***; 0.27***), Trade openness negative and significant (e.g., -0.02***), Floating exchange rate positive (0.07***).

### Effect of sanctions (country-level evidence, Tables 13–18; Figures 10–11)
- Qualitative patterns:
  - Many of the largest annual increases in gold share since 1999 coincided with sanctions or irregular political events; Table 13 shows 5 of top 10 annual increases involved sanctions in the same year or two preceding years.
  - Table 14 documents episodes of active gold diversification often coinciding with political, economic, or financial shocks (examples include Argentina, Belarus, Iraq, Hungary, Kazakhstan, Qatar, Russia, Türkiye, Uzbekistan, Bolivia, Egypt, Kyrgyz Republic, Mauritius, Mongolia, Serbia).
- Quantitative regression evidence (Tobit and volume specifications):
  - Table 15 (All economies; gold reserve shares): Big 4 Sanctions: 0.02**; EU Sanctions (individual): 0.04***; Japan Sanctions (individual): -0.04**; UK Sanctions (individual): 0.26***; Big 4 Sanctions (Multilateral): 0.04***; N 1293.
  - Table 16 (EMDEs; gold reserve shares): Big 4 Sanctions: 0.02***; EU Sanction (individual): 0.03***; UK Sanction (individual): 0.24***; Big 4 Sanctions (Multilateral): 0.05***; N 1086.
  - Table 17 (All economies; gold reserve volumes in mm troy ounces): Big 4 Sanctions: 4.17***; Big 4 Sanctions (Unilateral): 4.83***; Big 4 Sanctions (Multilateral): 3.25*; Constants: -10.89***; -10.55***; -10.97**; N 1291.
  - Table 18 (EMDEs; gold reserve volumes in mm troy ounces): Big 4 Sanctions: 3.72***; US Sanctions (individual): 2.30***; Japan Sanctions (individual): 13.35***; UK Sanctions (individual): -13.59***; Big 4 Sanctions (Unilateral): 1.55***; Big 4 Sanctions (Multilateral): 6.78**; N 1084.
- Magnitudes and interpretation:
  - For the full sample, Big Four financial sanctions are associated with an increase in gold volumes of 4.17 mm troy ounces (about 130 metric tons as described in text).
  - Sanctions variables account for around 2 percent of the R2 in the full sample R2 decomposition; in EMDEs sanctions variables contribute around 6 percent of the R2.
  - Domestic gold production, floating currency, and inflation account for large shares of explained variation (full-sample shares: 34, 27 and 13 percent of R2 respectively).

### Examples and case evidence (selected episodes)
- Top annual increases (2000–21) with concurrent events (Table 13 examples):
  - Türkiye 2020: Increase in gold share 21 ppt; Increase in gold volume 29%; Concurrent events: Sanctions by the US (2018) and EU (2019); COVID-19 pandemic (2020).
  - Belarus 2010: Increase in gold share 17 ppt; Increase in gold volume 57%; Concurrent events: Sanctions by Russia in 2010.
  - Lao PDR 2001: Increase in gold share 10 ppt; Increase in gold volume 328%; Concurrent events: A series of bomb blasts in 2000 before presidential elections in 2001.
  - Paraguay 2012: Increase in gold share 8 ppt; Increase in gold volume 1141%; Concurrent events: Sanctions by Mercosur and Unasur in 2012.
  - Sri Lanka 2009: Increase in gold share 8 ppt; Increase in gold volume 299%; Concurrent events: Purchase of gold from IMF following the global financial crisis.
- Table 14 documents numerous country episodes where gold accumulation coincided with sanctions, conflict, financial crises, or domestic policy initiatives (selected countries listed in paper).

### Summary statistics (Table 9 excerpt)
- Gold reserve share: N 5463; Mean 0.147; S.D. 0.208; Min 0; Max 0.958.
- Gold volume: N 5854; Mean 7.99; S.D. 34.4; Min 0; Max 402.8.
- Inflation: N 7351; Mean 0.438; S.D. 8.95; Min -0.727; Max 653.7.
- Overall fiscal surplus to GDP: N 5821; Mean -0.0286; S.D. 0.143; Min -5.58; Max 1.25.
- GDP growth: N 7369; Mean 0.0333; S.D. 0.0677; Min -0.667; Max 1.77.
- Reserves growth: N 6533; Mean 0.196; S.D. 1.25; Min -0.986; Max 58.5.
- Trade openness: N 5548; Mean 0.606; S.D. 0.442; Min 0.0126; Max 7.06.
- Currency appreciation: N 7093; Mean 1.87; S.D. 141.7; Min -0.993; Max 11800.
- Public debt to GDP: N 5140; Mean 0.560; S.D. 0.450; Min 0; Max 6.00.
- Gold production: N 2156; Mean 0.246; S.D. 0.580; Min 0; Max 4.54.
- Flexible exchange rate: N 8233; Mean 0.0949; S.D. 0.293; Min 0; Max 1.
- USD financial sanctions: N 8232; Mean 0.0862; S.D. 0.281; Min 0; Max 1.
- EUR financial sanctions: N 8232; Mean 0.0448; S.D. 0.207; Min 0; Max 1.
- JPY financial sanctions: N 8232; Mean 0.0102; S.D. 0.101; Min 0; Max 1.
- GBP financial sanctions: N 8232; Mean 0.0062; S.D. 0.0785; Min 0; Max 1.

### Conclusions and policy implications (Section VII summary)
- Aggregate evidence:
  - Reserve managers respond to relative costs and returns: they increase gold share when expected return on gold is high and return on financial assets (e.g., U.S. Treasury securities) is low.
  - Gold is viewed as a hedge against economic and geopolitical risks: gold shares in advanced countries and emerging markets increase with measures of economic uncertainty; advanced economies’ gold shares also increase with a measure of geopolitical risk.
- Sanctions-specific conclusion:
  - Reserve managers in emerging markets increase the share of reserves held in gold in response to sanctions risk.
  - Econometric results indicate both the volume and value of gold reserves increase with the imposition of sanctions from the U.S., UK, Euro Area, and Japan in the current or immediately preceding years.
- Uncertainty about future behavior:
  - Whether and to what extent central banks will increase gold shares going forward given recent events is uncertain; the reported results enable better-informed assessments.

*Source: wpiea2023014-print-pdf - References .....................................................................................................*

### References .............................................................................................................

### wpiea2023014-print-pdf - References .....................................................................................................

### Overview of recent developments
- In Q3 2022, global central banks added US$20 billion of gold to their international reserve portfolios — the largest quarterly increase in official gold demand in fully 55 years according to the World Gold Council (2022).
- The increase occurred against a multi-decade secular decline in gold’s share of global reserves that began decades earlier but began reversing around the time of the 2008-9 Global Financial Crisis.
- The paper identifies which central banks and which economic, financial, and political factors account for the shift toward gold.

### Two principal hypotheses for increased official gold demand
- Safe-haven and portfolio-diversification hypothesis:
  - Gold is seen as a safe haven in periods of high economic, financial, and geopolitical uncertainty and when returns on reserve currencies are low.
  - Gold is popularly viewed as an inflation hedge and as a portfolio diversifier; it also benefits from custom and tradition.
  - Liquid markets for trading gold exist in London, New York, Shanghai, and a large over-the-counter market in Switzerland.
- Sanctions- and asset-freeze insulation hypothesis:
  - Gold is perceived as more insulated from financial sanctions and asset freezes than foreign-exchange reserves held abroad.
  - The G7 decision to freeze Bank of Russia foreign-exchange reserves and subsequent acceleration of Russian gold purchases (including repatriation and vaulting at home) focused attention on gold’s role as a sanctions-robust reserve asset.

### Costs and limitations of gold as a reserve asset
- Gold is expensive to transport, warehouse, and secure, and it does not bear interest.
- Gold can be lent out on term deposit and used in swaps, but counterparties typically expect gold to be vaulted at recognized depositories (Bank of England, Federal Reserve Bank of New York, or major exchange depositories), reintroducing sanctions risk.
- Repatriation of gold can limit its use in market transactions.

### Empirical findings on determinants of the gold share in reserves
- Historical inertia:
  - The gold share of reserves exhibits considerable inertia.
- Relative returns and interest-rate effects:
  - The gold share is affected positively by the return to stockpiling gold, as proxied by the futures/spot price differential.
  - The gold share is affected negatively by the U.S. federal funds rate, indicating an opportunity cost relative to investing in short-term U.S. Treasuries and substitutes.
- Heterogeneity by country type:
  - Most regression results are driven by the emerging market subsample, consistent with emerging markets managing gold reserves more actively than advanced economies.
- Uncertainty, volatility, and geopolitical risk:
  - The gold share responds positively to global economic policy uncertainty and U.S. dollar volatility.
  - There is some evidence of a positive response to global geopolitical risk (index of interstate conflicts and terrorist attacks).
  - Advanced economies respond more to geopolitical risk; emerging markets respond more to economic policy uncertainty.

### Evidence on active diversifiers and magnitude of shifts
- Definition: “Active diversifiers” are countries that purchased gold and raised its share in reserves by at least 5 percentage points over the last two decades.
- Findings:
  - The paper identifies 14 active diversifiers; all 14 are emerging markets.
  - By contrast, an earlier IMF paper on foreign exchange reserves (Arslanalp, Eichengreen, and Simpson-Bell 2022) defined active diversifiers as shifting at least 5 percent of foreign exchange holdings into nontraditional reserve currencies and identified 46 active diversifiers (including advanced countries).
- Sanctions association:
  - Fully half of the largest year-over-year increases in central bank holdings of gold reserves since 2000 were associated with the risk of sanctions.
  - Using an indicator of financial sanctions imposed by the United States, United Kingdom, European Union, and Japan, the paper confirms that sanctions have a positive impact on the share of reserves held in gold.
  - There is evidence that multilateral sanctions imposed jointly by these countries have a larger impact on gold shares than unilateral sanctions.

### Regression and robustness notes (high-level)
- The gold share responds to:
  - Lagged inertia (persistence).
  - Futures/spot price differentials (positive effect).
  - U.S. federal funds rate (negative effect).
  - Global economic policy uncertainty and U.S. dollar volatility (positive effects).
  - Geopolitical risk (positive effect, more pronounced for advanced economies).
- Sanctions indicators (from the United States, United Kingdom, European Union, and Japan) are positively associated with higher country-level gold reserve shares and volumes, with larger effects for multilateral sanctions.

*Source: wpiea2023014-print-pdf - References .....................................................................................................*

### Section II of the paper reviews the related literature, while Section III provides a broad overview of the gold

### wpiea2023014-print-pdf - Section II/III Extracts: Gold as International Reserves

### Overview: recent trends in gold as official reserves
- Gold holdings of the official sector (measured in fine troy ounces) have remained relatively stable despite termination in 1971 of convertibility.
- The share of gold in reported foreign exchange reserves trended gently downward after the turn of the century but "still accounts for some 10 percent of total international reserves worldwide" and has shown "a slowly rising share in recent years."
- Since 1999:
  - Official gold holdings increased by 7 percent in volume terms.
  - This increase was "driven by a 130 percent increase in tons of gold held of EMDEs."
- Distinctive advanced vs EMDE patterns:
  - Advanced economies: gold holdings fell after the turn of the century; on average gold represents 17 percent of official reserves (down from 80 percent in 1950).
  - Emerging markets and developing economies (EMDEs): gold holdings trended strongly upward since the Global Financial Crisis; on average gold represents 7 percent of official reserves (down from 30 percent in 1950).

### Concentration of holdings (as of end-2021)
- International organizations (mainly the IMF and BIS) accounted for roughly 10 percent of official gold holdings.
- Two-thirds of non-international-organization holdings are held by advanced economies; one-third by EMDEs.
- Selected country-level holdings and shares (Table 1 excerpt):
  - United States: 261.5 million troy ounces; 23% of total; 66% percent of reserve assets; reserve assets as percent of GDP 3%.
  - Germany: 108.0 million troy ounces; 9% of total; 66% percent of reserve assets; reserve assets as percent of GDP 7%.
  - Italy: 78.8 million troy ounces; 7% of total; 63% percent of reserve assets; reserve assets as percent of GDP 11%.
  - France: 78.3 million troy ounces; 7% of total; 58% percent of reserve assets; reserve assets as percent of GDP 8%.
  - Russia: 74.0 million troy ounces; 6% of total; 21% percent of reserve assets; reserve assets as percent of GDP 36%.
  - China: 62.6 million troy ounces; 5% of total; 3% percent of reserve assets; reserve assets as percent of GDP 20%.
  - IMF: 90.5 million troy ounces (memo item).
- Gold accounted for more than 60 percent of reserves for Portugal, Kazakhstan, Germany, the United States, Italy, and Uzbekistan as of end-2021; more than 40 percent for France, Netherlands, Bolivia, Cyprus, Austria, Lebanon, and Greece.
- For the Euro Area (including the ECB), gold represents 53 percent of official reserves as of end-2021.

### Buyers and sellers, 1999–2021
- Top official-sector buyers (2000s–2021; Table 3 excerpt):
  - Russia: 60.7 million troy ounces; 28% of all buying/selling; change in gold share in official reserves -10%.
  - China: 49.9 million troy ounces; 23% of all buying/selling; change in gold share in official reserves 1%.
  - Türkiye: 17.4 million troy ounces; 8% of all buying/selling; change in gold share in official reserves 31%.
  - India: 12.7 million troy ounces; 6% of all buying/selling; change in gold share in official reserves -2%.
  - Kazakhstan: 11.1 million troy ounces; 5% of all buying/selling; change in gold share in official reserves 42%.
- Top official-sector sellers (2000s–2021; Table 3 excerpt):
  - Switzerland: -49.8 million troy ounces; 34% of all buying/selling; change in gold share in official reserves -34%.
  - France: -18.9 million troy ounces; 13% of all buying/selling; change in gold share in official reserves 17%.
  - IMF: -13.0 million troy ounces; 9% of all buying/selling.
  - Netherlands: -11.9 million troy ounces; 8% of all buying/selling; change in gold share in official reserves 7%.
  - United Kingdom: -10.6 million troy ounces; 7% of all buying/selling; change in gold share in official reserves -4%.
- Active diversifiers into gold, 1999–2021:
  - 14 EMDEs classified as "active diversifiers" (raised gold share by at least 5 percentage points).
  - Eight active diversifiers that purchased at least one million troy ounces: Kazakhstan, Belarus, Türkiye, Uzbekistan, Hungary, Iraq, Argentina, Qatar.
- State Street Advisors observation for 2008–17: "Russia and China together account for more than two-thirds of gross gold purchases and more than three quarters of net purchases." Over 2000–2021 they account for 51 percent of purchases.

### Institutional context: Central Bank Gold Agreements (CBGA)
- The first CBGA (Washington Agreement on Gold) announced in September 1999 and covered five years; it was renewed three times. The Agreement was eventually allowed to expire.
- After the GFC, advanced economies mostly stopped selling gold in substantial amounts (more than 1 metric ton a year), with exceptions: Germany (continues to sell regularly) and Canada (sold off gold in 2015-16). Ireland bought in 2021-22; Singapore bought in 2021.

### Aggregate econometric analysis: data, variables, and identification
- Monthly IFS data on gold’s share in global reserves (market valuation), 1980–2021.
- Key regression covariates and measures used:
  - Inertia: lagged dependent variable (instrumented via Griliches adjustment).
  - Return measures: Year-over-year gold return (LBMA spot), gold basis (6-month futures/spot ratio), U.S. dollar appreciation (12-month change in Nominal Broad Trade-Weighted Dollar Index; advanced and EMDE subsamples use respective trade-weighted indices), U.S. policy rate (Wu-Xia shadow federal funds rate from 1990; pre-1990 effective federal funds rate), U.S. CPI inflation.
  - Volatility: five-year rolling standard deviation of monthly gold returns; dollar volatility likewise.
  - Uncertainty: Caldara-Iacoviello (2022) geopolitical acts subindex; Global Economic Policy Uncertainty (GEPU) by Davis (2016); alternative: VIX.
  - Liquidity: gold bid-ask spreads (Bloomberg).
- Sample size and summary statistics (Table 5 excerpt):
  - Gold reserve share: N 504; Mean 0.221; S.D. 0.16; Min 0.0794; Max 0.652.
  - Gold reserve share (advanced): N 504; Mean 0.280; S.D. 0.159; Min 0.126; Max 0.727.
  - Gold reserve share (emerging and developing): N 504; Mean 0.120; S.D. 0.103; Min 0.0216; Max 0.373.
  - Gold basis: N 504; Mean 0.019; S.D. 0.018; Min -0.0166; Max 0.0903.
  - Gold return: N 504; Mean 0.065; S.D. 0.237; Min -0.375; Max 1.98.
  - Global Economic Policy Uncertainty: N 300; Mean 130; S.D. 67.9; Min 48.9; Max 430.
  - Geopolitical risk: N 504; Mean 72.5; S.D. 44.1; Min 20.4; Max 470.
  - U.S. Dollar appreciation: N 504; Mean 0.0336; S.D. 0.069; Min -0.105; Max 0.230.
  - Shadow FFR: N 504; Mean 0.0414; S.D. 0.044; Min -0.0299; Max 0.191.
  - Gold bid-ask spread: N 313; Mean 0.0114; S.D. 0.017; Min 0; Max 0.130.
  - US CPI inflation: N 504; Mean 0.0322; S.D. 0.025; Min -0.021; Max 0.148.

### Key econometric findings (aggregate monthly regressions)
- Strong inertia:
  - Lagged dependent variable coefficients large and significant across specifications (e.g., Inertia 0.95*** in Table 6, Columns (1)-(2); ranges 0.80***–0.96*** across panels and subgroups).
- Gold basis:
  - Positive and significant effect on gold reserve share (e.g., coefficient 0.70*** in Table 6, Column (1); remains significant across specifications and subsamples).
- Returns:
  - Gold return enters positively and significantly in extended specifications (e.g., 0.03*** and 0.04*** in Table 6 Columns (4)-(6)).
- U.S. policy rate (shadow FFR):
  - Negative effect on gold share (e.g., -0.26* to -0.31*** in Table 6; consistently negative in Tables 7 and 8 for advanced and EMDEs).
  - Interpretation: reserve managers shift toward gold when returns on reserve currencies, notably the dollar, are low.
- Dollar appreciation:
  - Generally negative or insignificant effects; not a robust positive driver of gold share.
- Volatility and liquidity:
  - U.S. dollar volatility is positively associated with gold share for advanced economies; gold volatility tends to reduce gold share for EMDEs (Table 8: gold volatility coefficient -0.57*).
  - Gold bid-ask spread not statistically significant in aggregate specifications.
- Uncertainty measures:
  - Global Economic Policy Uncertainty (GEPU) shows a positive and statistically significant effect on gold share (small coefficient magnitude but robust across specifications).
  - Geopolitical risk (aggregate) does not show a consistent significant effect; however, when disaggregated, the "acts" subindex is significantly positive in many specifications and driven by advanced countries.
- Subsample differences:
  - Advanced economies: more responsive to geopolitical acts and dollar volatility; gold returns matter positively.
  - EMDEs: respond more consistently to global economic policy uncertainty; gold volatility negatively affects EMDE gold shares.

### Dynamic responses (local projections / impulse responses)
- Response to a 1-point shock to GEPU (Figure 6):
  - Cumulative response of gold share is positive on impact and rises, peaking after eight months.
  - Example (textual calculation): GEPU rose by 140 points between February and March 2022; on impact (first month) this implies a 140*0.15 = 0.21 percent increase in the share of gold in central bank portfolios.
  - The cumulative impact tops out in 7 months at about three times the immediate increase.
- Volume (fine troy ounces) response to GEPU (Figure 7):
  - Gold holdings in volume do not change on impact but buying begins about five months after the uncertainty event.
  - A 140 point shock would result in a 0.14 percent increase (=140*0.1) in gold holdings (in volume) after 12 months.
- Geopolitical risk impulse (Figure 8):
  - Response peaks after 5–6 months and dissipates after a year.
- U.S. federal funds rate shock (Figure 9):
  - Response of gold share continues to grow between 6 and 12 months.

### Country-level determinants (panel analysis, annual country-level IFS data, 1980–2021)
- Panel covers 144 economies, unbalanced, 1980–2021.
- Explanatory variables used (annual):
  - GDP growth (year-on-year real GDP change, WEO).
  - Inflation (year-on-year consumer prices, WEO).
  - Fiscal balance (government net lending/borrowing divided by GDP, WEO).
  - Public-debt-to-GDP ratio (general government gross debt/GDP, WEO).
  - Trade openness (exports + imports of goods divided by GDP).
  - Local currency appreciation (current year change vs USD).
  - Domestic gold production (metric tons, World Gold Council).
  - Exchange rate regime indicator (1 for floating; Ilzetzki, Reinhart, Rogoff 2019).
  - Financial sanctions imposed by traditional reserve currency issuers (Global Sanctions Database).
  - Growth of reserve assets (annual percentage change in total reserve assets, gold at market value).
- Expected signs (literature-informed):
  - Gold shares expected to be higher when GDP growth is low, inflation is high, public debt is high, and when domestic gold production is larger.
  - Trade openness and pegged exchange-rate regimes expected to be associated with lower gold shares.
  - Rapid growth in total reserves expected to be associated with lower gold share.
- Results are reported separately for full sample, advanced economies, and EMDEs (detailed coefficient estimates not reproduced here but methodology and variables specified).

### Interpretations and policy-relevant inferences from analysis
- Central bank reserve portfolios exhibit strong inertia; month-to-month gold shares are highly persistent.
- Relative returns matter: lower U.S. policy rates and lower returns on reserve currencies are associated with higher gold shares.
- Market signals matter: a positive gold basis (futures/spot premium) and positive gold returns are associated with higher gold shares.
- Uncertainty and volatility:
  - Global Economic Policy Uncertainty (GEPU) is a robust positive driver of gold shares for both advanced economies and EMDEs.
  - Geopolitical acts appear to matter more for advanced economies; EMDEs respond more to economic policy uncertainty.
- Timing of responses:
  - Price effects can raise gold share immediately; actual changes in gold volumes (purchases) typically follow with a lag (roughly five months or more).
- Recent EMDE accumulation of gold is consistent with a combination of low returns on reserve currencies post-GFC, elevated economic policy uncertainty, and idiosyncratic geopolitical/economic concerns for specific countries.

*Italic: Source: IMF Working Paper extract (Sections II–V), wpiea2023014-print-pdf.*

### Section V, financial sanctions imposed by issuers of major reserve currencies should also encourage central

### Section V, financial sanctions imposed by issuers of major reserve currencies should also encourage central banks to hold a higher gold share.

### Summary statistics (Table 9)
- Gold reserve share: N 5463; Mean 0.147; S.D. 0.208; Min 0; Max 0.958
- Gold volume: N 5854; Mean 7.99; S.D. 34.4; Min 0; Max 402.8
- Inflation: N 7351; Mean 0.438; S.D. 8.95; Min -0.727; Max 653.7
- Overall fiscal surplus to GDP: N 5821; Mean -0.0286; S.D. 0.143; Min -5.58; Max 1.25
- GDP growth: N 7369; Mean 0.0333; S.D. 0.0677; Min -0.667; Max 1.77
- Reserves growth: N 6533; Mean 0.196; S.D. 1.25; Min -0.986; Max 58.5
- Trade openness: N 5548; Mean 0.606; S.D. 0.442; Min 0.0126; Max 7.06
- Currency appreciation: N 7093; Mean 1.87; S.D. 141.7; Min -0.993; Max 11800
- Public debt to GDP: N 5140; Mean 0.560; S.D. 0.450; Min 0; Max 6.00
- Gold production: N 2156; Mean 0.246; S.D. 0.580; Min 0; Max 4.54
- Flexible exchange rate: N 8233; Mean 0.0949; S.D. 0.293; Min 0; Max 1
- USD financial sanctions: N 8232; Mean 0.0862; S.D. 0.281; Min 0; Max 1
- EUR financial sanctions: N 8232; Mean 0.0448; S.D. 0.207; Min 0; Max 1
- JPY financial sanctions: N 8232; Mean 0.0102; S.D. 0.101; Min 0; Max 1
- GBP financial sanctions: N 8232; Mean 0.0062; S.D. 0.0785; Min 0; Max 1

### Country-level regression findings (Tables 10–12)
- General patterns:
  - Gold share is negatively related to GDP growth (statistical significance varies).
  - Gold share is negatively related to fiscal balance; effect larger for advanced countries than emerging markets.
  - Public debt is negatively related to gold share in emerging markets; no relationship in advanced economies.
  - Floating exchange rates associated with higher gold shares (apparent for both advanced and EMDEs).
  - Trade openness associated with lower gold share (larger and more consistently significant for emerging markets).
  - Higher domestic gold production increases gold share.
  - Reserve growth associated with lower gold share, especially for advanced economies.

- Selected coefficients from Table 10 (All economies, Tobit estimates; lower limit 0 and upper limit 1; year dummies included):
  - CPI inflation: 0.003*** (spec 1); 0.12** (spec 2); 0.27*** (spec 3); 0.21*** (spec 4)
  - Fiscal balance: 0.06** (spec 1); 0.16*** (spec 2); -0.46*** (spec 3); -0.47*** (spec 4)
  - Real GDP growth: -0.25***; -0.27***; -0.12; -0.01
  - Trade openness: -0.02***; -0.02***; -0.03***; -0.03***
  - Currency appreciation: 0.00001***; -0.01; 0.00; 0.00
  - Reserves growth (%): -0.01***; -0.01***; 0.00; 0.00
  - Public debt: 0.06***; 0.01; -0.01
  - Gold production: 0.06***; 0.04***
  - Floating exchange rate: 0.07***
  - Constant: 0.33***; 0.40***; 0.03*; 0.04**
  - Statistics: N 3095; 2750; 1291; 1291; Pseudo R2 -0.74; -0.63; -0.28; -0.31

- Selected coefficients from Table 11 (Advanced economies):
  - CPI inflation: -0.59***; -0.65***; 0.95; -0.64***
  - Fiscal balance: -1.67***; -1.76***; -2.52***; -1.74***
  - Real GDP growth: -0.35; -0.46*; -0.42; -0.47**
  - Reserves growth (%): -0.10***; -0.09***; 0.05; -0.08***
  - Gold production: 0.08*** (spec 3)
  - Floating exchange rate: 0.05** (spec 4)
  - Constant: 0.43***; 0.47***; 0.06; 0.47***
  - Statistics: N 714; 676; 207; 676; Pseudo R2 -7.67; -5.06; 1.79; -5.16

- Selected coefficients from Table 12 (Emerging Market and Developing Economies):
  - CPI inflation: 0.00***; 0.26***; 0.34***; 0.27***
  - Fiscal balance: 0.08***; 0.17***; -0.17*; -0.15
  - Real GDP growth: -0.19***; -0.15**; -0.01; 0.04
  - Trade openness: -0.02***; -0.01; -0.02**; -0.02***
  - Reserves growth (%): -0.01**; -0.01***; 0.00; 0.00
  - Public debt: 0.08***; -0.04**; -0.04**
  - Gold production: 0.02***; 0.01*
  - Floating exchange rate: 0.07***
  - Constant: 0.17***; -0.72***; 0.07***; 0.07***
  - Statistics: N 2381; 2074; 1084; 1084; Pseudo R2 -0.30; -0.24; -0.11; -0.13

### Effect of sanctions (Tables 13–18, Figures 10–11)
- Qualitative evidence:
  - Many of the largest annual increases in gold share since 1999 coincided with sanctions or irregular political events.
  - Table 13 (Top 10 annual increases, 2000–21) shows that in 5 of 10 cases the country was subject to sanctions in the same year or the two immediately preceding years.
  - Table 14 lists episodes of active gold diversification often coinciding with political, economic, or financial shocks (examples include Argentina, Belarus, Iraq, Hungary, Kazakhstan, Qatar, Russia, Türkiye, Uzbekistan, Bolivia, Egypt, Kyrgyz Republic, Mauritius, Mongolia, Serbia).

- Quantitative regression evidence on sanctions (Tobit estimates; covariates include inflation, fiscal balance, GDP growth, trade openness, currency appreciation, public debt, gold production, FX regime; year dummies included; standard errors robust):
  - Table 15 (All economies; gold reserve shares):
    - Big 4 Sanctions: 0.02**
    - EU Sanctions (individual): 0.04***
    - Japan Sanctions (individual): -0.04**
    - UK Sanctions (individual): 0.26***
    - Big 4 Sanctions (Multilateral): 0.04***
    - Constant: 0.04**; N 1293; Pseudo R2 -0.31; -0.27; -0.31

  - Table 16 (Emerging Market and Developing Economies; gold reserve shares):
    - Big 4 Sanctions: 0.02***
    - EU Sanction (individual): 0.03***
    - UK Sanction (individual): 0.24***
    - Big 4 Sanctions (Multilateral): 0.05***
    - Constant: 0.07***; N 1086; Pseudo R2 -0.14

  - Table 17 (All economies; gold reserve volumes in mm troy ounces):
    - Big 4 Sanctions: 4.17***
    - Big 4 Sanctions (Unilateral): 4.83***
    - Big 4 Sanctions (Multilateral): 3.25*
    - Constants: -10.89***; -10.55***; -10.97**
    - N 1291; Pseudo R2 0.04

  - Table 18 (Emerging Market and Developing Economies; gold reserve volumes in mm troy ounces):
    - Big 4 Sanctions: 3.72***
    - US Sanctions (individual): 2.30***
    - Japan Sanctions (individual): 13.35***
    - UK Sanctions (individual): -13.59***
    - Big 4 Sanctions (Unilateral): 1.55***
    - Big 4 Sanctions (Multilateral): 6.78**
    - Constant: 1.84; 2.28; 2.25
    - N 1084; Pseudo R2 0.04

- Interpretation and magnitudes:
  - Sanctions are associated with increases in gold share and gold volumes.
  - For the full sample, Big Four financial sanctions are associated with an increase in gold volumes of 4.17 mm troy ounces (about 130 metric tons as described in text).
  - Multilateral sanctions have a larger estimated effect than unilateral sanctions in share regressions; under some volume specifications unilateral sanctions have larger impacts.

- Relative importance of determinants (R2 decomposition; OLS used for decomposition):
  - Full sample (Figure 10): domestic gold production, floating currency, and inflation account for 34, 27 and 13 percent of the R2 respectively; sanctions variables account for around 2 percent as a group.
  - Emerging market and developing economies (Figure 11): inflation and currency regime remain important; domestic gold production contributes much less; sanctions variables contribute around 6 percent of the R2.

### Examples and case evidence (selected items from Tables 13–14)
- Table 13 (Top annual increases, 2000–21; increase in gold share in ppt of reserves; increase in gold volume in percent; concurrent events):
  - Türkiye 2020: Increase in gold share 21 ppt; Increase in gold volume 29%; Concurrent events: Sanctions by the US (2018) and EU (2019); COVID-19 pandemic (2020)
  - Belarus 2010: Increase in gold share 17 ppt; Increase in gold volume 57%; Concurrent events: Sanctions by Russia in 2010
  - Lao PDR 2001: Increase in gold share 10 ppt; Increase in gold volume 328%; Concurrent events: A series of bomb blasts in 2000 before presidential elections in 2001
  - Turkey 2017: Increase in gold share 9 ppt; Increase in gold volume 50%; Concurrent events: Coup attempt in 2016; Constitutional referendum in 2017
  - Sri Lanka 2009: Increase in gold share 8 ppt; Increase in gold volume 299%; Concurrent events: Purchase of gold from IMF following the global financial crisis
  - Paraguay 2012: Increase in gold share 8 ppt; Increase in gold volume 1141%; Concurrent events: Sanctions by Mercosur and Unasur in 2012
  - Hungary 2021: Increase in gold share 8 ppt; Increase in gold volume 200%; Concurrent events: COVID-19 pandemic in 2020
  - Belarus 2006: Increase in gold share 8 ppt; Increase in gold volume 25%; Concurrent events: Sanctions by the EU and US in 2006
  - Sri Lanka 2000: Increase in gold share 7 ppt; Increase in gold volume 437%; Concurrent events: 1997-98 Asian crisis; Assassination attempts before elections in 2000
  - Belarus 2008: Increase in gold share 7 ppt; Increase in gold volume 33%; Concurrent events: Sanctions by the EU and US in 2006
- Table 14 documents numerous country episodes where gold accumulation coincided with sanctions, conflict, financial crises, or domestic policy initiatives (examples include Argentina 2004; Belarus 2006–11, 2019, 2021; Iraq 2012–14, 2018; Hungary 2018 and 2021; Kazakhstan since 2012; Qatar 2007, 2015–20, since Feb 2022; Russia 2007–20; Türkiye since 2011; Uzbekistan at least since 2005; Bolivia 2010–11; Egypt since 2017; Kyrgyz Republic 2011–21; Mauritius 2009 and 2014–16; Mongolia 2018–19; Serbia since 2002).

### Conclusions and policy implications (Section VII summary)
- Aggregate evidence:
  - Reserve managers respond to relative costs and returns: they increase gold share when expected return on gold is high and return on financial assets (e.g., U.S. Treasury securities) is low.
  - Gold is viewed as a hedge against economic and geopolitical risks: gold shares in advanced countries and emerging markets increase with measures of economic uncertainty; advanced economies’ gold shares also increase with a measure of geopolitical risk.
- Sanctions-specific conclusion:
  - Reserve managers in emerging markets increase the share of reserves held in gold in response to sanctions risk.
  - Econometric results indicate both the volume and value of gold reserves increase with the imposition of sanctions from the U.S., UK, Euro Area, and Japan in the current or immediately preceding years.
- Uncertainty about future behavior:
  - Whether and to what extent central banks will increase gold shares going forward given recent events is uncertain; the reported results enable better-informed assessments.

*IMF Working Paper excerpt: Section V and associated tables and figures from "Gold as International Reserves: A Barbarous Relic No More?"*

### References

### References (wpiea2023014-print-pdf)

### Major themes covered by the references
- Gold as a hedge, safe haven, and component of central bank reserves (multiple empirical and theoretical studies).
- Currency composition of foreign exchange reserves and the evolving role of reserve currencies.
- Geopolitical risk, sanctions, and their effects on exchange rates, financial linkages, and reserve management.
- Measurement and drivers of uncertainty, global risk, and demand for gold by central banks.
- Historical and institutional analyses of gold in the international monetary system.

### Representative works and topics (selected from the reference list)
- Gold as hedge/safe haven and price dynamics:
  - Abid, Ilyes, Anderrazak Dhaoui, Stephanie Goutte and Khaled Guesmi (2020), “Hedging and Diversification Across Commodity Assets,” Applied Economics 52, pp.2472-2492.
  - Baur, Dirk and Brian Lucey (2009), “Is Gold a Hedge or a Safe Haven? An Analysis of Stocks, Bonds and Gold,” unpublished manuscript.
  - Baur, Dirk and Thomas McDermott (2010), “Is Gold a Safe Haven? International Evidence,” Journal of Banking and Finance 34, pp.1886-1898.
  - Beckman, Joscha, Theo Berger and Robert Czudaj (2019), “Gold Price Dynamics and the Role of Uncertainty,” Quantitative Finance 19, pp.663-681.
  - Ghosh, Dipak, Eric Levin, Peter MacMillan and Robert Wright (2004), “Gold as an Inflation Hedge?” Studies in Economics and Finance 22, pp.1-25.
  - Zulaica, Omar (2020), “What Share for Gold? On the Interaction of Gold and Foreign Exchange Reserve Returns,” BIS Working Paper No. 906 (November).

- Central bank demand for gold and reserve composition:
  - Ghosh, Amit (2016), “What Drives Gold Demand in Central Bank’s Foreign Reserve Portfolio?”, Finance Research Letters 17, pp. 146-150.
  - Gopalakrishnan, Balagopal and Sanket Mohapatra (2018a), “Turning over a Gold Leaf? Global Liquidity and Emerging Market Central Banks’ Demand for Gold after the Financial Crisis,” Journal of International Financial Markets Institutions and Money 57, pp.94-109.
  - Gopalakrishnan, Balagopal and Sanket Mohapatra (2018b), “Global Risk and Demand for Gold by Central Banks,” Applied Economics Letters 25, pp.835-839.
  - Oztunc, Hakan and Mehmet Orhan (2021), “Gold Demand by Central Banks: A Comparative Study of Emerging Market and Advanced Economies,” Emerging Markets Finance and Trade 57, pp.2687-2698.
  - World Gold Council (2021), “Central Bank Domestic ASGM Purchase Programmes,” London: World Gold Council (April).
  - World Gold Council (2022), “Gold Demand Trends Q3 2022: Central Banks,” London: World Gold Council (November).
  - Oktay, Beyza, Hakan Oztunc and Z. Vildan Serin (2016), “Determinants of Gold Reserves: An Empirical Analysis for G-7 Countries,” Procedia Economics and Finance 38, pp.8-16.

- Reserve currencies, dollar dominance, and shifts in international money:
  - Chinn, Menzie and Jeffrey Frankel (2007), “Will the Euro Eventually Surpass the Dollar as Leading International Reserve Currency?” in Richard Clarida ed., G7 Current Account Balances: Sustainability and Adjustment, Chicago: University of Chicago Press, pp.283-338.
  - Ito, Hiro and Robert McCauley (2020), “Currency Composition of Foreign Exchange Reserves,” Journal of International Money and Finance 102, pp.1-20.
  - Arslanalp, Serkan, Barry Eichengreen and Chima Simpson-Bell (2022), “The Stealth Erosion of Dollar Dominance: Active Diversifiers and the Rise of Nontraditional Reserve Currencies,” Journal of International Economics (forthcoming).
  - Iancu, A., G. Anderson, S. Ando, E. Bosell, A. Gamba, S. Hakobyan, L. Lusinyan N. Meads and Y. Wu (2020), “Reserve Currencies in an Evolving International Monetary System,” IMF Departmental Paper No. 2020/002 (November).
  - Bordo, Michael and Barry Eichengreen (2004), “The Rise and Fall of a Barbarous Relic: The Role of Gold in the International Monetary System” in Rudiger Dornbusch and Maurice Obstfeld eds, Money, Capital Mobility and Trade, Cambridge,Mass.: MIT Press.
  - Triffin, Robert (1960), Gold and the Dollar Crisis: The Future of Convertibility, New Haven: Yale University Press.
  - Chinn, Menzie, Hiro Ito and Robert McCauley (2022), “Do Central Banks Rebalance Their Currency Shares?” Journal of International Money and Finance 122, 102557.

- Sanctions, financial restrictions, and related datasets:
  - Felbermayr Gabriel, Aleksandra Kirilakha, Constantinos Syropoulos, Erdal Yalcin, and Yoto Yotov (2020), “The Global Sanctions Data Base,” European Economic Review 129, pp. 1–23.
  - Felbermayr, Babriel, T. Clifton Morgan, Constantinos Syropoulos and Yoto Yotov (2021), “Understanding Economic Sanctions: Interdisciplinary Perspectives on Theory and Evidence,” European Economic Review 135, pp.1-15.
  - Syropoulos, C., G. Felbermayr, A. Kirilakha, E. Yalcin, Erdal, and Y.V. Yatov (2022), “The Global Sanctions Data Base – Release 3: COVID-19, Russia, and Multilateral Sanctions,” School of Economics Working Paper Series 2022-11.
  - Eichengreen, Barry (2022), “Sanctions, SWIFT and China’s Cross-Border Interbank Payments System,” CSIS Policy Brief, Washington, D.C.: CSIS.
  - Heydarian, Samira, Mosayeb Pahlavani and Seyed Mirjalili (2021), “Financial Sanctions and Economic Growth: An Intervention Time-Series Approach,” International Economic Studies 51, pp.1-14.
  - Hufbauer, Gary, Jeffrey Schott and Kimberly Anne Elliott (2009), Economic Sanctions Reconsidered, Washington, D.C.: Peterson Institute of International Economics.

- Measurement, methods, and datasets used in related analyses:
  - Caldara, Dario and Matteo Iacoviello (2022), “Measuring Geopolitical Risk,” American Economic Review 112, pp. 1194-1225.
  - Davis, Steven, 2016, “An Index of Global Economic Policy Uncertainty,” NBER Working Paper No. 22740.
  - Jordà, Oscar (2005), “Estimation and Inference of Impulse Responses by Local Projections,” American Economic Review.95, pp.161-182.
  - Hüttner, Frank and Marco Sunder, 2011, “Decomposing R2 with the Owen value”, Working Paper No. 100, Universität Leipzig Faculty of Economics and Management Science.
  - Wong, Anna (2007), “Measurement and Inference in International Reserve Diversification,” Working Paper No. 07-6, Washington, D.C.: Peterson Institute for International Economics (July).
  - Cantu, Carlos and Yavuz Arslan (2019), “The Size of Foreign Exchange Reserves,” BIS Paper No.104, pp.1-23.

- Historical, institutional, and policy-oriented analyses:
  - Eichengreen, Barry (1990), “The Gold Exchange Standard and the Great Depression,” Chapter 10 of Elusive Stability: Essays in International Finance 1919-1939, Cambridge: Cambridge University Press, pp.239-270.
  - Eichengreen, Barry (1998), “The Euro as a Reserve Currency,” Journal of the Japanese and International Economies 12, pp.483-506.
  - Bordo, Michael and Robert McCauley (2019), “Triffin: Dilemma or Myth?” IMF Economic Review 67, pp.824-851.
  - Schenk, Catherine (2011), International Economic Relations Since 1945 – The Making of the Contemporary World, London: Routledge, second edition.
  - Monnet, Eric and Damien Puy (2020), “Do Old Habits Die Hard? Central Banks and the Bretton Woods Gold Puzzle,” Journal of International Economics 127, pp.1-20.
  - Lindert, Peter (1969), “Key Currencies and Gold, 1900-1913,” Princeton Studies in International Finance No. 24, International Finance Section, Department of Economics, Princeton University.

### Notable working papers, reports, and data sources cited
- IMF Departmental Paper No. 2020/002 (November): Iancu et al. (2020), “Reserve Currencies in an Evolving International Monetary System.”
- BIS Working Paper No. 906 (November): Zulaica (2020), “What Share for Gold? On the Interaction of Gold and Foreign Exchange Reserve Returns.”
- NBER Working Paper No. 22740 (2016): Davis, “An Index of Global Economic Policy Uncertainty.”
- Working Paper No. WP/23/14: Gold as International Reserves: A Barbarous Relic No More?

*Working Paper No. WP/23/14 — References*

---


_Source: https://www.imf.org/-/media/files/publications/wp/2023/english/wpiea2023014-print-pdf.pdf_