## wpiea2019161

## Source details

**Canonical URL:** [wpiea2019161](https://www.imf.org/-/media/files/publications/wp/2019/wpiea2019161.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/wp/2019/wpiea2019161.pdf.md)
- [Structured JSON version](/-/media/files/publications/wp/2019/wpiea2019161.pdf.json)

---

### I. Introduction — research question and context
- Research question: Can the persistence of past practices and norms lead to unexpected economic policy consequences?
- Historical context: continued accumulation of gold reserves by non-US central banks after WWII under the Bretton Woods system — labelled the Bretton Woods “gold puzzle”.
- Puzzle highlighted:
  - Under Bretton Woods, banknotes were no longer legally redeemable in gold at the central bank (only the USA committed to convert dollars into gold on demand from other central banks).
  - Holding gold was costly relative to US government securities; nevertheless many monetary authorities accumulated gold.
- Literature background: behavioral economics and institutional memory relevant (Bertrand and Schoar (2003); Alesina and Fuchs-Schuendeln (2007); Malmendier and Nagel (2011, 2015); Koudijs and Voth (2016); Camerer and Malmendier (2007); Bouwman and Malmendier (2015)).

### II. Key empirical approach and dataset
- New dataset assembled using the IMF archives covering a large set of countries.
- Sample coverage statement: “Our sample represents, on average, 94 percent of gold reserves held outside of the United States over the Bretton Woods era.”
- Data construction and measurement:
  - Monetary statistics in contemporary domestic currency; reserves in current dollars; GDP from Penn World Table in today's local currency.
  - Nominal values converted into domestic currency of 1971 by tracking currency denomination changes and official exchange rates using IMF International Financial Statistics.
  - Sample covers 38 countries, for a total of 729 observations.
  - Age of central bank governors obtained for 85 percent of the sample (i.e. country-year observations).
  - Sample starts in 1950 and ends in 1970.

### III. Main empirical findings — money rather than trade
- Core estimating equation:
  - Gold_{i,t} = α + d_t + β T_{i,t} + γ M_{i,t} + θ Z_{i,t} + ε_{i,t}
  - Where Gold_{i,t} = level of gold reserves divided by GDP; T_{i,t} = trade openness; M_{i,t} = currency in circulation divided by GDP; Z_{i,t} = controls; d_t = year-fixed effects.
- Key empirical findings:
  - Currency in circulation and gold reserves are strongly positively correlated; without controls the estimate of γ is around 0.5 (Table 1, column 1).
  - The coefficient on trade (β) is lower and becomes insignificant once standard controls are added (Table 1, column 2).
  - Control variables have expected signs: countries accumulate more reserves when they have a larger current account surplus and an undervalued exchange rate, when exchange rates are less flexible, and when they are more financially open.
  - Country fixed-effects confirm significance and size of the currency coefficient (Table 1, column 3); deposits coefficient is small and insignificant.
  - Within variation explained by the currency variable is 0.22.
  - Estimated γ interpreted quantitatively: around 0.35 (between 0.27 and 0.43), implying that 35 percent of an additional unit of currency was backed by gold reserves.
  - Historical benchmark: in the gold standard the average cover ratio was around 40 percent (Bordo & Eichengreen 1998, Morys 2013).

### IV. Memory and gold — institutional memory, exposure, and Bretton Woods roles
- Hypothesis tested: countries more exposed to gold standard norms/practices before WWII were more likely to back currency by gold during Bretton Woods.
- Definition of GS: GS = 1 if country was in both classical gold standard (1880s–1913) and interwar gold standard (1920s–1936), zero otherwise.
- Results (Table 2):
  - Interaction term τ is positive and significant; backing gold reserves was more pervasive in countries fully exposed to the gold standard before WWII. Size of coefficient slightly below 0.5 (consistent with cover ratios).
  - Countries outside the gold standard also backed currency with gold but intensity much smaller (around 0.16).
- Robustness on colonies/dominions:
  - Setting GS = 0 for countries not independent before WWII; alternative coding (assign colonizer’s GS to dominions/colonies) yields slightly lower interaction (0.44 vs 0.48) but unchanged significance.
- Heterogeneity within gold-standard countries:
  - Using number of years spent in the gold standard as exposure measure, interaction coefficient is 0.01.
  - Quantitative examples supplied by authors:
    - Peru (17 years in gold standard) -> roughly 20 percent of an extra unit of currency covered by gold.
    - Portugal (37 years) -> 44 percent.
- Bretton Woods exposure:
  - Number of delegates at the 1944 Bretton Woods conference used as proxy for “Bretton Woods exposure”.
  - Interaction between currency and number of delegates is negative: countries more involved in Bretton Woods design were less likely to behave as in the gold standard.

### V. Personal experience vs institutional memory — distinguishing mechanisms
- Two transmission hypotheses:
  - Institutional memory: institutions have a memory; corporate or institutional culture transmits past practices.
  - Individual memory: personal formative experience of central bankers or governors matters.
- Evidence distinguishing the two:
  - Figure 3: correlation between average age of governors and number of years the institution spent in gold standard is flat.
  - Table 3, column 1: interaction of governor age with currency/GDP is positive and significant — older governors associated with higher gold-currency correlation.
  - When including both interactions (currency*gold-standard exposure and currency*age), both matter (Table 3, column 2).
  - Restricting to countries with at least one year in the gold standard (Table 3, column 3): the effect of number of years in the gold standard disappears, whereas interaction of currency and governor age remains significant and rises in magnitude (0.052 vs 0.015 in column 2).
  - Interpretation: age/personal experience has stronger predictive power than years-of-exposure within gold-standard countries; personal experience of central bankers independently influenced reserve management.
- Illustrative cases:
  - Maurice Frère (Belgium): born 1890, long professional exposure to gold standard era; Bank of Belgium maintained legal reference to gold reserves equal to 30 percent of notes issue.
  - Mats Lemne (Sweden): born 1919; Riksbank under Lemne had low gold cover and close links with postwar Ministry of Finance.

### VI. Extensions, robustness, and alternative explanations
- Alternative dependent variables (Table 4):
  - Non-gold reserves (foreign exchange): significant and positive but smaller coefficient on currency/GDP; not robust with country-fixed effects (Table 4, column 5).
  - Share of gold in total foreign reserves (goldshare): strong, significant, and robust coefficient on currency/GDP (Table 4, columns 3–4, 6).
  - Interaction of gold-standard exposure*currency is negative for non-gold reserves (column 2) and positive for gold share (column 4).
  - Conclusion: monetary authorities treated gold and foreign exchange differently; allocation driven more by persistence of past behavior than purely financial substitution; gold and foreign exchange were not perfect substitutes under Bretton Woods.
- Reverse causality and sterilization:
  - Potential reverse causality: incomplete sterilization could create positive correlation between reserves and currency if money supply increased to buy reserves.
  - Sterilization was common during Bretton Woods; reverse causality more likely at higher frequencies (monthly, quarterly) than annual.
  - Reverse-causality concern more relevant for foreign exchange reserves because gold was not used for private commercial/financial payment.
  - A deliberate policy to increase gold reserves to back rising domestic money supply is an intended policy choice, not an endogeneity bias.

### VII. Implications for the Bretton Woods system, literatures, and policy
- Implications for the international monetary system:
  - Central banks retained gold-standard norms and did not treat dollar and gold as perfect substitutes; old habits hindered the Bretton Woods system functioning as expected.
  - The main result concerns persistence of the practice to back currency by gold reserves, not merely path-dependency in the level of the gold stock.
  - After the dollar–gold link was abandoned in August 1971 and world inflation increased, gold-standard practices diminished.
- Contributions to literatures:
  - Reserves literature: first rigorous exploration of determinants of reserve holdings (gold or foreign exchange) between 1950 and 1970; challenges view that reserves accumulation during Bretton Woods was purely “trade-based”.
  - Behavioral and leadership literature: provides evidence on macroeconomic effects of corporate culture and individual experience of policymakers; supports behavioral models where decision makers overemphasize personal experience.
  - International monetary system design: contemporaries underestimated persistence of gold-standard practices; results inform understanding of safe reserve assets persistence.
- Policy-relevant insight:
  - Institutional history and personal experience of policymakers materially shaped postwar reserve management.
  - Alternative credible US policy or different institutional rules could have limited persistence of gold practices; persistence was not deterministic.

### VIII. Sample and selected descriptive statistics (selected)
- Sample: 38 countries; 729 observations.
- Selected means from Table A.1:
  - ARG: Years in the gold standard 25; Gold Share (mean, in percent) 41; Reserves/GDP (mean, in percent) 2; Gold Reserves/GDP (mean, in percent) 1; Currency/GDP (mean, in percent) 9.
  - BEL: Years in the gold standard 50; Years in the gold standard as independent countries 50; Gold Share (mean, in percent) 66; Reserves/GDP (mean, in percent) 13; Gold Reserves/GDP (mean, in percent) 8; Currency/GDP (mean, in percent) 20.
  - PER: Years in the gold standard 17; Gold Share (mean, in percent) 42; Reserves/GDP (mean, in percent) 4; Gold Reserves/GDP (mean, in percent) 2; Currency/GDP (mean, in percent) 6.
  - PRT: Years in the gold standard 37; Gold Share (mean, in percent) 61; Reserves/GDP (mean, in percent) 23; Gold Reserves/GDP (mean, in percent) 14; Currency/GDP (mean, in percent) 14.
  - SWE: Years in the gold standard 50; Gold Share (mean, in percent) 30; Reserves/GDP (mean, in percent) 4; Gold Reserves/GDP (mean, in percent) 1; Currency/GDP (mean, in percent) 8.
- Note on sample size: due to remaining data gaps, sample includes 729 observations rather than 798; nine additional countries had partial data but were excluded from some estimations.

*Source: wpiea2019161 - References and excerpts from the source PDF*

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

### References

### I. Introduction — research question and context
- Research question: Can the persistence of past practices and norms lead to unexpected economic policy consequences?  
- Literature background: behavioral economics and institutional memory relevant (Bertrand and Schoar (2003), Alesina and Fuchs-Schuendeln (2007), Malmendier and Nagel (2011, 2015), Koudijs and Voth (2016); Camerer and Malmendier (2007); Bouwman and Malmendier (2015)).  
- Historical context: continued accumulation of gold reserves by non-US central banks after WWII under the Bretton Woods system—a phenomenon labelled the Bretton Woods “gold puzzle”.  
- Puzzle highlighted: Under Bretton Woods, banknotes were no longer legally redeemable in gold at the central bank (only the USA committed to convert dollars into gold on demand from other central banks), and holding gold was costly relative to US government securities; nevertheless many monetary authorities accumulated gold.

### Key empirical approach and dataset
- New dataset assembled using the IMF archives covering a large set of countries.  
- Sample coverage statement: “Our sample represents, on average, 94 percent of gold reserves held outside of the United States over the Bretton Woods era.”

### Main empirical findings
- Finding 1: The amount of gold that countries accumulated under Bretton Woods was still connected to their domestic money supply (currency in circulation), even after accounting for all other potential accumulation motives.  
- Finding 2: The intensity of the link between gold holdings and currency in circulation depended on past exposure to the pre-WWII gold standard:
  - Monetary authorities that had adhered to the gold standard were much more likely to back their currency in circulation under Bretton Woods than others.
  - Among countries exposed to the gold standard, the longer they had been on the gold standard before the war, the higher the correlation between their gold holdings and currency in circulation.
  - Magnitude: Countries that adhered to both classical and interwar gold standards continued to back between 40 percent and 50 percent of every new unit of currency issued with gold, thereby using a de facto cover ratio almost identical to those in effect during the gold standard.

### Hypotheses on transmission mechanisms (memory)
- Institutional memory hypothesis:
  - Institutions have a memory; corporate or institutional culture transmits past practices (references: Eichengreen & Temin 2000, Mouré 2002; James 1985, Capie 2010, Schenk 2010, Straumann 2010, Monnet 2018a).
- Individual memory (personal experience) hypothesis:
  - Personal formative experience of central bankers or governors matters (Romer & Romer 2004, Malmendier et al. 2017, Bordo & Istrefi 2018, Mishra & Reshef 2019).
  - Example: Governors born in the 1880s or 1890s (Germany, Austria, Mexico, Switzerland, Belgium) versus those born in the 1910s (Greece, Finland, Norway, Sweden) likely held different perceptions of the gold standard’s stability.

### Evidence distinguishing institutional vs. individual memory
- Data on identity and age of governors used to proxy for “individual memory.”  
- Empirical result: Central banks with an older governor were more prone to back currency in circulation with gold.  
- When restricting sample to gold standard countries, the age of the governor dominates the number of years spent in the gold standard, suggesting an independent and strong effect of personal experience in addition to institutional memory.

### Contributions to literatures
- Reserves literature:
  - First rigorous exploration of determinants of reserve holdings (gold or foreign exchange) between 1950 and 1970, bridging gaps between limited contemporary studies of the 1960s–1970s and later historical studies starting in the late 1970s.
  - Challenges the conventional wisdom that reserves accumulation during Bretton Woods was purely “trade-based” (Williamson 1973, Obstfeld et al. 2010).
- Behavioral and leadership literature:
  - Provides new evidence on macroeconomic effects of corporate culture and individual experience of policymakers; supports behavioral models where decision makers overemphasize personal experience (Gennaioli & Schleifer 2010; Malmendier and Nagel 2011, 2015; Koudijs and Voth 2016).
  - Aligns with empirical findings on leaders’ personal experience shaping organizational outcomes (Bertrand and Schoar 2003; Malmendier et al. 2011; Benmelech & Frydman 2015; Romer & Romer 2004; Malmendier et al. 2017; Bordo & Istrefi 2018).
- International monetary system design:
  - Implications for the role and persistence of safe reserve assets; contemporaries underestimated persistence of gold-standard practices (Machlup, 1966; Williamson, 1973).
  - Distinction: main result concerns persistence of the practice to back currency by gold reserves, not merely path-dependency in the level of the gold stock (contrast with Bordo and Eichengreen (1998) and Eichengreen et al. (2017)).

### Paper structure (as provided)
- Section II: Reviews the Bretton Woods system, differences with pre-WWII gold standards, and the resurgence of gold reserves after 1950.  
- Section III: Presents the data and econometric framework and preliminary results on the positive correlation between gold reserves and currency.  
- Section IV: Explores the role of institutional and personal memory.  
- Section V: Robustness checks.  
- Section VI: Further discussion of how memory prevented the Bretton Woods system from functioning as its founders expected.

*Source: wpiea2019161 - References (excerpt provided).*

### Section VI discusses the broader policy implications of our results. Section VII concludes.

### wpiea2019161 - Section VI discusses the broader policy implications of our results. Section VII concludes.

### II. Gold and the Bretton Woods System
- Bretton Woods System origins and timeline:
  - Created by the 1944 Articles of Agreement at Bretton Woods (USA) from July 1st to July 22nd, 1944.
  - Articles ratified in December 1945; system began implementation after the conference of Savannah in March 1946.
- Four departures from pre-WWI gold standard and interwar gold exchange standard:
  - Controls on capital and current accounts officially allowed to avoid destabilizing international movements of funds.
  - IMF created to provide short-term liquidity and conduct surveillance of national policies.
  - Pegged exchange rates defined in terms of dollars (not gold); only the US dollar convertible into gold at $35 per ounce.
  - Gold coinage and commitment to redeem banknotes in gold abandoned worldwide.
- Practical arrangements and limits:
  - Gold remained a final means of settlement between monetary authorities of major trading nations.
  - The United States held more than 90 percent of gold reserves held in the world by monetary authorities after the war.
  - Two official roles for gold: (i) US prepared to exchange gold for dollars on request by other central banks; (ii) IMF member countries could pay subscriptions in gold (not mandatory).
  - Strong limits on private gold holding in some countries (e.g., US and UK forbade private holding entirely).

### The Gold Puzzle (empirical observations 1950–1970)
- Key empirical facts from cross-country reserves data between 1950 and 1970:
  - Total gold held by foreign monetary authorities increased throughout the Bretton Woods period.
  - Demand for gold by non-US monetary authorities began to pressure the US gold stock starting in the mid-1950s; trend accelerated starting in 1958–1959 as most countries opened current accounts and made currencies fully convertible.
  - Countries that started with a lower gold ratio in 1950 often increased it over 1950–1970:
    - Figure 2 shows a negative relationship between cover ratio in 1950 and the change in cover ratio from 1950 to 1970.
    - Some countries with high postwar cover ratios reduced them (e.g., Switzerland, Great Britain, Canada), while several countries increased or kept them stable.
- Implication: Contrary to Bretton Woods' intentions, cover ratios did not universally decline toward very low values; heterogeneity across countries was large.

### Potential Motives for Gold Holding (conjectures and contemporary explanations)
- Suggested motives and critiques:
  - Monetary anchor hypothesis: central bankers used gold as an anchor for monetary policy in the uncertain postwar world, interpreting increases in money supply relative to gold (or FX) as inflationary signals.
  - Skeptical view: economists labeled such behavior “irrational” or “primitive” and few empirical studies investigated a money–gold link for this period.
  - Dollar credibility hypothesis: loss of US dollar credibility (notably in the 1960s due to Vietnam War spending and weak US stabilization policies) led foreign central banks to increase gold holdings to hedge against expected dollar devaluation.
    - This hypothesis does not fully explain uneven cross-country gold reconstructions.
  - Trade-openness hypothesis: gold holdings driven by same determinants as foreign exchange reserves, especially trade openness (exports + imports over GDP).
    - Evidence limited; Kenen (1963) noted central banks bought gold when reserves increased but did not sell gold when reserves decreased.
  - Political motives: 1960s gold purchases had political signaling (e.g., France converting dollars to gold in 1965–1966); US used coercion to limit gold purchases by Germany and Japan.
- Authors’ stance: challenge prevailing views that idiosyncratic political motives alone explain gold holdings; propose historical experience and monetary practices also important.

### III. Gold and Currency Under Bretton Woods — Empirical Framework
- Objective: formally explore motives behind gold holdings during Bretton Woods using a new cross-country database for 1950–1970.
- Core estimating equation (1):
  - Gold_{i,t} = α + d_t + β T_{i,t} + γ M_{i,t} + θ Z_{i,t} + ε_{i,t}
  - Where:
    - Gold_{i,t} = level of gold reserves divided by GDP of country i in year t.
    - T_{i,t} = trade openness (exports + imports over GDP).
    - M_{i,t} = currency in circulation (banknotes and coins) divided by GDP.
    - Z_{i,t} = set of control variables.
    - d_t = year-fixed effects; α = constant.
- Interpretation focus:
  - γ captures relationship between gold and currency in circulation; a positive γ of magnitude close to past cover ratios would support hypothesis that monetary authorities behaved as under the gold standard.
  - A positive β would indicate trade-based accumulation.
  - Idiosyncratic political motives would appear in residuals rather than as fundamental determinants.
- Control variables (Z_{i,t}) used:
  - Year-fixed effects to capture time-varying dollar credibility and opportunity cost of holding gold.
  - Capital account liberalization index of Quinn and Toyoda (2008): index ranges from 0 to 100 (higher = higher capital account openness).
  - Current account surplus (percent of GDP).
  - Exchange rate premium (difference between official exchange rate and black-market exchange rate in New York); positive premium implies overvaluation and expected devaluation.
  - Exchange rate regime de facto index (Reinhart and Rogoff 2004): lower value = harder peg.
  - Logarithm of population (proxy for country size/political power).
  - Other forms of money (deposits) following Obstfeld et al. (2010).
- Comparison with literature:
  - Post-Bretton Woods studies find correlation between total money supply (including deposits) and total foreign reserves, but not between the money base and reserves.
  - Authors hypothesize that during Bretton Woods, the money base (currency in circulation) correlated with foreign reserves (gold).

### Data Sources, Construction, and Sample
- New cross-country database covering Bretton Woods period (1950–1970) constructed by harmonizing reserves, monetary, and macroeconomic statistics.
  - Monetary statistics in contemporary domestic currency; reserves in current dollars; GDP from Penn World Table in today's local currency.
  - Converted nominal values into a consistent unit (domestic currency of 1971) by tracking currency denomination changes and official exchange rates using IMF International Financial Statistics (paper volumes since 1946).
  - Authors state they are the first to compile such a database for the Bretton Woods period.
- Sample characteristics and choices:
  - Sample starts in 1950 due to uneven monetary statistics and limited nominal GDP availability immediately postwar.
  - Data set ends in 1970 to exclude reactions to closure of US gold window in August 1971.
  - Comprehensive data obtained for 38 countries.
  - On average, these countries account for 94 percent of gold reserves held outside of the United States over the period.

*Source: wpiea2019161 - https://www.imf.org/-/media/files/publications/wp/2019/wpiea2019161.pdf*

### Appendix present the sample of countries and descriptive statistics.

### Appendix present the sample of countries and descriptive statistics.

### Sample of countries
- The appendix presents the sample of countries used in the analysis.

### Descriptive statistics
- Besides macroeconomic and financial statistics, the appendix compiles several indicators to measure past exposure of central bankers and central banks to the gold standard.

### Indicators of past exposure to the gold standard
- The compiled indicators are intended to be used in the analysis to capture historical exposure of central bankers and central banks to the gold standard.

*Source: wpiea2019161 - Appendix present the sample of countries and descriptive statistics.*

### section III), namely (i) the number of years spent by a country in the gold standard (ii) the number

### wpiea2019161 - section III), namely (i) the number of years spent by a country in the gold standard (ii) the number

### Data and measurement
- Measures of past exposure to the gold standard taken from Officer (2008), Wandschneider (2008), and Mitchener & Weidenmier (2015).
- Number of delegates at the Bretton Woods conference comes from Schuler & Rosenberg (2012) and a more detailed list by Schuler & Bernkopf (2014).
- Age of central bank governors: obtained via a two-step process (identify governor name per year; obtain biographies/birth dates from central bank websites, regional/national Who's Who, monographs, Wikipedia, genealogical sites). Age found for 85 percent of the sample (i.e. country-year observations).
- Monetary and reserve data: foreign reserve data (gold and non-gold) from International Financial Statistics (IFS). Monetary series from IFS and Mitchell (2013). M1 used given data limitations for the Bretton Woods period.
- Nominal values converted to domestic currency of 1971 for comparability; foreign reserves in US dollars converted using IMF exchange rates (end of year).
- Sample covers 38 countries, for a total of 729 observations.

### Main empirical findings: Money rather than trade
- Currency in circulation and gold reserves are strongly positively correlated; without controls the estimate of 훾 is around 0.5 (Table 1, column 1).
- The coefficient on trade (훽) is lower and becomes insignificant once standard controls are added (Table 1, column 2).
- Control variables have expected signs: countries accumulate more reserves when they have a larger current account surplus and an undervalued exchange rate, when exchange rates are less flexible, and when they are more financially open.
- Country fixed-effects confirm significance and size of the currency coefficient (Table 1, column 3); deposits coefficient is small and insignificant.
- Within variation explained by the currency variable is 0.22.
- Estimated 훾 interpreted quantitatively: around 0.35 (between 0.27 and 0.43), implying that 35 percent of an additional unit of currency was backed by gold reserves.
- Historical benchmark: in the gold standard the average cover ratio was around 40 percent (Bordo & Eichengreen 1998, Morys 2013).

### Memory and gold: institutional memory, exposure, and Bretton Woods roles
- Hypothesis: countries more exposed to gold standard norms/practices before WWII were more likely to back currency by gold during Bretton Woods; tested via interaction of currency in circulation with exposure to gold standard in equation (2).
- Definition of GS: GS = 1 if country was in both classical gold standard (1880s–1913) and interwar gold standard (1920s–1936), zero otherwise.
- Key results (Table 2):
  - Interaction term 휏 is positive and significant; backing gold reserves was more pervasive in countries fully exposed to the gold standard before WWII. Size of coefficient slightly below 0.5 (consistent with cover ratios).
  - Countries outside the gold standard also backed currency with gold but intensity much smaller (around 0.16).
- Robustness on colonies/dominions: setting GS = 0 for countries not independent before WWII; alternative coding (assign colonizer’s GS to dominions/colonies) yields slightly lower interaction (0.44 vs 0.48) but unchanged significance.

### Heterogeneity across gold-standard countries: number of years and delegate exposure
- Within gold-standard countries, the intensity of past exposure matters:
  - Using number of years spent in the gold standard as exposure measure, interaction coefficient is 0.01.
  - Quantitative examples: Peru (17 years in gold standard) -> roughly 20 percent of an extra unit of currency covered by gold; Portugal (37 years) -> 44 percent.
- Falsification / Bretton Woods exposure:
  - Number of delegates at the 1944 Bretton Woods conference used as a proxy for “Bretton Woods exposure”.
  - Interaction between currency and number of delegates is negative: countries more involved in Bretton Woods design were less likely to behave as in the gold standard.

### Personal experience vs institutional memory
- Two hypotheses:
  - Institutional memory: persistent corporate culture or institutional habits inherited from prewar gold standard shape central bank behavior.
  - Individual memory: personal experience of central bankers (age/formative years) matters independently.
- Evidence:
  - Figure 3: correlation between average age of governors and number of years the institution spent in gold standard is flat (no strong mechanical relationship).
  - Table 3, column 1: interaction of governor age with currency/GDP is positive and significant — older governors associated with higher gold-currency correlation.
  - When including both interactions (currency*gold-standard exposure and currency*age), both matter (Table 3, column 2).
  - Restricting to countries with at least one year in the gold standard (Table 3, column 3): the effect of number of years in the gold standard disappears, whereas interaction of currency and governor age remains significant and rises in magnitude (0.052 vs 0.015 in column 2). This suggests age/personal experience has stronger predictive power than years-of-exposure within gold-standard countries.
- Case illustrations:
  - Maurice Frère (Belgium): born 1890, long professional exposure to gold standard era, strong adherence to gold practices; Bank of Belgium maintained legal reference to gold reserves equal to 30 percent of notes issue.
  - Mats Lemne (Sweden): born 1919, formative years after classical gold era; Sweden had adopted other monetary practices in 1930s; Riksbank under Lemne had low gold cover and close links with postwar Ministry of Finance.

### Extensions and robustness
- Alternative dependent variables (Table 4):
  - Non-gold reserves (foreign exchange): significant and positive but smaller coefficient on currency/GDP; not robust with country-fixed effects (Table 4, column 5).
  - Share of gold in total foreign reserves (goldshare): strong, significant, and robust coefficient on currency/GDP (Table 4, columns 3–4, 6).
  - Interaction of gold-standard exposure*currency is negative for non-gold reserves (column 2) and positive for gold share (column 4).
  - Conclusion: monetary authorities treated gold and foreign exchange differently; allocation driven more by persistence of past behavior than purely financial considerations; gold and foreign exchange were not perfect substitutes under Bretton Woods.
- Reverse causality and sterilization:
  - Potential reverse causality: incomplete sterilization could create positive correlation between reserves and currency if money supply increased to buy reserves.
  - Sterilization was common during Bretton Woods; reverse causality more likely at higher frequencies (monthly, quarterly) than annual.
  - Reverse causality concern more relevant for foreign exchange reserves because gold was not used for private commercial/financial payment; thus mechanism of banks exchanging foreign currency for domestic currency increasing central bank foreign assets (and currency if not sterilized) does not apply to gold reserves.
  - A deliberate policy to increase gold reserves to back rising domestic money supply is not an endogeneity bias but an intended policy choice.

### Implications for the Bretton Woods system and conclusion
- Findings suggest central banks retained gold-standard norms and did not treat dollar and gold as perfect substitutes; old habits hindered the Bretton Woods system functioning as expected.
- This does not imply determinism: alternative credible US policy or different institutional rules could have limited persistence of gold practices.
- After the dollar–gold link was abandoned in August 1971 and world inflation increased, gold-standard practices diminished.
- Broader conclusion: institutional history and personal experience of policymakers materially shaped postwar reserve management; the demise of Bretton Woods led to loss of an anchor and has implications for understanding the institutional turn to fiat money and subsequent monetary policy evolution.

### Sample and key descriptive statistics (selected)
- Sample: 38 countries; 729 observations.
- Table A.1 (selected entries; means reported in source):
  - ARG: Years in the gold standard 25; Gold Share (mean, in percent) 41; Reserves/GDP (mean, in percent) 2; Gold Reserves/GDP (mean, in percent) 1; Currency/GDP (mean, in percent) 9.
  - BEL: Years in the gold standard 50; Years in the gold standard as independent countries 50; Gold Share (mean, in percent) 66; Reserves/GDP (mean, in percent) 13; Gold Reserves/GDP (mean, in percent) 8; Currency/GDP (mean, in percent) 20.
  - PER: Years in the gold standard 17; Gold Share (mean, in percent) 42; Reserves/GDP (mean, in percent) 4; Gold Reserves/GDP (mean, in percent) 2; Currency/GDP (mean, in percent) 6.
  - PRT: Years in the gold standard 37; Gold Share (mean, in percent) 61; Reserves/GDP (mean, in percent) 23; Gold Reserves/GDP (mean, in percent) 14; Currency/GDP (mean, in percent) 14.
  - SWE: Years in the gold standard 50; Gold Share (mean, in percent) 30; Reserves/GDP (mean, in percent) 4; Gold Reserves/GDP (mean, in percent) 1; Currency/GDP (mean, in percent) 8.
- Note: due to remaining data gaps, sample includes 729 observations rather than 798; nine additional countries had partial data but were excluded from some estimations.

*Source: https://www.imf.org/-/media/files/publications/wp/2019/wpiea2019161.pdf*

### References

### References

### Historical studies of monetary systems and gold
- Aizenman, J., & Inoue, K. (2013). Central banks and gold puzzles. Journal of the Japanese and International Economies, 28, 69–90.
- Bazot, G., Bordo, M. D., & Monnet, E. (2016). International shocks and the balance sheet of the Bank of France under the classical gold standard. Explorations in Economic History, 62, 87–107.
- Bordo M.D, (1993). The Bretton Woods International Monetary System: A Historical Overview, in A Retrospective on the Bretton Woods System, edited by Michael D. Bordo and Barry Eichengreen, pp. 3–98. Chicago: University of Chicago Press,  
- Bordo M. D. & Eichengreen B. ed. (1993). A Retrospective on the Bretton Woods System. Chicago: University of Chicago Press.  
- Bordo M. D. & Eichengreen B. (1998). The Rise and Fall of a Barbarous Relic: The Role of Gold in  the  International  Monetary  System. (No.  w6436).  National  Bureau  of  Economic Research.
- Bordo,  M.  D.,  &  McCauley,  R.  N.  (2018). Triffin:  dilemma  or  myth? (No.  w24195).  National Bureau of Economic Research.
- Bordo, M., Monnet, E., & Naef, A. (2017). The Gold Pool (1961-1968) and the Fall of the Bretton Woods System. Lessons for Central Bank Cooperation (No. w24016). National Bureau of Economic Research.
- Frère,  M.  (1960).  La  restauration  monétaire  de  la  Belgique. Revue  d'économie  politique,  70(6), 82–108.
- Gilbert, M. (1968). The Gold-Dollar system: conditions of equilibrium and the price of gold (Vol. 70). International Finance Section, Dept. of Economics, Princeton University.
- Helleiner, E. (2014). Forgotten foundations of Bretton Woods: International development and the making of the postwar order. Ithaca: Cornell University Press.
- James, H., (1985). The Reichsbank and Public Finance in Germany, 1924–1933: A Study of the Politics  of  Economics  during  the  Great  Depression (Schriftenreihe  des  Instituts  für Bankhistorische Forschung, number 5.) Frankfurt a/M: Fritz Knapp
- James,  H.,  (1996). International  Monetary  Cooperation  Since  Bretton  Woods. Washington DC: Oxford University Press & IMF.
- Machlup, F. (1964). Plans for the Reform of the International Monetary System. Princeton Essays in International Economics, n°3 Princeton University, International Finance Section.
- Machlup, F. (1966), The need for monetary reserves. PSL Quarterly Review 19–78.
- Monnet, E. (2014). Monetary policy without interest rates: Evidence from France's Golden Age (1948 to 1973) using a narrative approach. American Economic Journal: Macroeconomics 6 (4), 137–169.
- Monnet,  E. (2018a), Controlling  Credit.  Central  banking  and  the  planned  economy  in  postwar France, 1948-1973, Cambridge University Press.
- Monnet,  E.  (2018b).  Credit  controls  as  an  escape  from  the  trilemma.  The  Bretton  Woods experience. European Review of Economic History, 22(3), 349–360.
- Mouré, K. (2002). The Gold Standard Illusion: France, the Bank of France, and the International Gold Standard, 1914-1939. Oxford: Oxford University Press.
- Schenk, C. R. (2010). The decline of sterling: managing the retreat of an international currency, 1945–1992. Cambridge: Cambridge University Press.
- Triffin, R. (1960). Gold and the Dollar Crisis. New Haven, Conn.: Yale University Press.
- Van der Wee, H. (2012). “Belgian Monetary Policy under the Gold Standard during the Interwar Period”. In The Gold Standard Peripheries (pp. 130–144). Palgrave Macmillan, London.
- Wandschneider,  K.  (2008).  The stability  of  the  interwar  gold  exchange  standard:  Did  politics matter? The Journal of Economic History, 68(1), 151–181.
- Mouré, K. (2002). The Gold Standard Illusion: France, the Bank of France, and the International Gold Standard, 1914-1939. Oxford: Oxford University Press.
- Eichengreen, B., & Temin, P. (2000). The gold standard and the great depression. Contemporary European History, 9(2), 183–207.

### International reserves, liquidity, and the international monetary system
- Aizenman J. & Lee L., (2008). Financial versus Monetary Mercantilism: Long-run View of Large International Reserves Hoarding. The World Economy, 31(5), pages 593–611, 05
- Cohen, B.J (1975). International reserves and liquidity. In International Trade And Finance, edited by P. B. Kenen, p.1–52. Cambridge: Cambridge University Press.
- Courchene, T.J., & G. M. Youssef, (1967). The demand for international reserves. The Journal of Political Economy, 75(4), 404–413.
- Despres, E., Kindleberger, C. P., & Salant, W. S. (1966). The dollar and world liquidity: a minority view. Washington DC: Brookings Institution.
- Grubel, H.G (1971). The  demand  for  international  reserves:  a  critical  review  of  the  literature. Journal of Economic Literature, 9(4): 1148–1166.
- Heller, H.R. (1976). International Reserves and World-Wide Inflation. Staff Papers-International Monetary Fund, 23(1), 61–87.
- Kenen, P. B. (1963). Reserve-asset preferences of central banks and stability of the gold-exchange standard (No.   10).   International   Finance   Section,   Dept.   of   Economics,   Princeton University.
- Kenen,  P.B.,  & Yudin E.B. (1965). The  demand  for  international  reserves. The  Review  of Economics and Statistics, 242–250.
- Makin, J. H.,  1971,  The  composition  of  international  reserve  holdings:  A  problem  of  choice involving risk. American Economic Review 61(5), 818–832.
- McKinnon,  R. I. (1969).  Private  and  Official  International  Money:  The  Case  for  the  Dollar. Princeton Essays in International n°74, Economics. Princeton University, International Finance Section.
- Obstfeld,  M., Shambaugh J.C, & Taylor, A.M (2010). Financial  Stability,  the  Trilemma,  and International Reserves. American Economic Journal: Macroeconomics 2(2) , 57–94.
- Penn Word Table: Feenstra, Robert C., Robert Inklaar and Marcel P. Timmer (2013), “The Next Generation of the Penn World Table” available for download at www.ggdc.net/pwt
- Rodrik, D.  (2006).  The  Social  Cost  of  Foreign  Exchange  Reserves. International  Economic Journal, 20(3): 253–66
- International Monetary Fund (IMF), 1971, International Reserves. Needs and Availability, Papers and Proceedings. Seminar at the International Monetary Fund, Washington DC.
- Gourinchas,  P.  O.,  Rey,  H.,  &  Sauzet,  M.  (2019).  The  International  Monetary  and  Financial System (No. w25782). National Bureau of Economic Research.
- Farhi, E., & Maggiori, M. (2017). A Model of the International Monetary System. The Quarterly Journal of Economics, 133(1), 295–355.

### Central banking, policy, and governance
- Aufricht, H. (Ed.). (1967). Central Banking Legislation: A Collection of Central Bank, Monetary, and Banking Laws, volume 2, International Monetary Fund.
- Capie, F. (2010). The Bank of England: 1950s to 1979. Cambridge: Cambridge University Press.
- Cassiers,  I.,  &  Ledent,  P.  (2006).  Learning  to  manage  external  constraints:  Belgian  monetary policy during Bretton-Woods era (1944–1971). IRES Discussion Paper, 2006/032.
- Horsefield,  K.,  ed. (1969). The  International  Monetary  Fund,  1945-1965:  Twenty  Years  of International Monetary Co-Operation. Vol. 3, Washington, D.C.: International Monetary Fund.
- Holtrop,  M.W. (1957). Method  of  monetary  analysis  used  by the Nederlandsche  Bank. Staff Papers-International Monetary Fund, 5(3), 303–316.
- Naef, A. (2017). Dirty float or clean intervention? The Bank of England on the foreign exchange market, 1952–72. University of Cambridge Mimeo.
- Williamson, J. (1973). Surveys in applied economics:  International liquidity. Economic Journal 83(3), 685–746.
- Schuler & Rosenberg (2012), “The Bretton Woods Transcripts”, Center  for  Financial  Stability, Paper in Financial History, New York.
- Schuler & Rosenberg (2014), “Who was at Bretton Woods”, Center for Financial Stability, Paper in Financial History, New York.
- Scott-Smith,  G.,  &  Rofe,  J.  S.  (Eds.).  (2017). Global  Perspectives  on  the  Bretton  Woods Conference and the Post-War World Order. Springer.
- Straumann, T. (2010). Fixed ideas of money: Small states and exchange rate regimes in twentieth-century Europe. Cambridge; Cambridge University Press.
- Straumann,  T.,  Kugler,  P.,  &  Weber,  F.  (2017).  How  the  German  crisis  of  1931  swept  across Europe: a comparative view from Stockholm. The Economic History Review, 70(1), 224–247.
- Wellens, R. (1976). Inventaire des papiers de Maurice Frère, gouverneur de la Banque nationale de Belgique. Archives générales du Royaume de Belgique.

### Behavioral, managerial, and organizational influences on policy and risk
- Alesina, A., & Fuchs-Schündeln, N. (2007). Goodbye Lenin (or not?): The effect of communism on people's preferences. American Economic Review, 97(4), 1507–1528.
- Benmelech, E., & Frydman, C. (2015). Military CEOs. Journal of Financial Economics, 117(1), 43–59.
- Bertrand, M., & Schoar, A. (2003). Managing with style: The effect of managers on firm policies. The Quarterly journal of economics, 118(4), 1169–1208.
- Bouwman,  C.  H.,  &  Malmendier,  U.  (2015). Does  a  Bank's  History  Affect  Its  Risk-Taking? American Economic Review, 105(5), 321–25.
- Camerer, C.  & Malmendier,  U.  (2007). Behavioral  economics  of  organizations. In Behavioral Economics and Its Applications, edited by P.Diamond and H. Vartiainen, p. 235–290, Princeton: Princeton University Press.
- Gennaioli, N., & Shleifer, A. (2010). What comes to mind. The Quarterly journal of economics, 125(4), 1399–1433.
- Koudijs,  P.,  &  Voth,  H. J.  (2016).  Leverage  and  beliefs:  personal  experience  and  risk-taking  in margin lending. American Economic Review, 106(11), 3367–3400.
- Malmendier,  U.,  &  Nagel,  S.  (2011).  Depression  babies:  do  macroeconomic  experiences  affect risk taking? The Quarterly Journal of Economics, 126(1), 373–416.
- Malmendier, U., Tate, G., & Yan, J. (2011). Overconfidence and early life experiences: the effect of managerial traits on corporate financial policies. The Journal of finance, 66(5), 1687–1733.
- Malmendier, U., & Nagel, S. (2015). Learning from inflation experiences. The Quarterly Journal of Economics, 131(1), 53–87.
- Malmendier,  U.,  Nagel,  S.,  &  Yan,  Z.  (2017). The  making  of  hawks  and  doves:  Inflation experiences on the FOMC (No. w23228). National Bureau of Economic Research.
- Mendoza, P. (entry not present beyond listed items; included references limited to supplied items).
- Mishra,  P.,  &  Reshef,  A.  (2019).  How  Do  Central  Bank  Governors  Matter?  Regulation  and  the Financial Sector. Journal of Money, Credit and Banking, vol. 51, n° 2–3.
- Officer, L. H., & Willett T.D, (1969). Reserve-asset preferences and the confidence problem in the crisis zone. The Quarterly Journal of Economics, 83(4), 688–695.
- Officer, L. H. (2008). Gold standard, EH. net encyclopedia. http://eh.net/encyclopedia/article/officer. gold. standard.

### Empirical methods, country studies, and other contributions
- Jonung, L. (1979). Knut Wicksell's norm of price stabilization and Swedish monetary policy in the 1930's. Journal of Monetary Economics, 5(4), 459–496.
- Koudijs,  P.,  &  Voth,  H. J.  (2016).  Leverage  and  beliefs:  personal  experience  and  risk-taking  in margin lending. American Economic Review, 106(11), 3367–3400.
- Larsson, M., & Söderberg, G. (2017). Finance and the Welfare State: Banking Development and Regulatory Principles in Sweden, 1900–2015. Springer.
- Lipartito, K. (1995). Culture and the practice of business history. Business and Economic History, 24(2),1–41.
- Mitchener, K. J., & Weidenmier, M. D. (2015).  Was the classical gold standard credible on the periphery?  Evidence from  currency  risk. The  Journal  of  Economic  History,  75(2),  479–511.
- Naef, A. (2017). Dirty float or clean intervention? The Bank of England on the foreign exchange market, 1952–72. University of Cambridge Mimeo.
- Rathke,  A.,  Straumann,  T.,  &  Woitek,  U.  (2017).  Overvalued:  Swedish  monetary  policy  in  the 1930s. International Economic Review, 58(4), 1355–1369.
- Reinhart  C.M.  & Rogoff  K.S (2004).  The  Modern  History  of  Exchange  Rate  Arrangements:  A Reinterpretation. The Quarterly Journal of Economics.  119(1), pages 1–48, February.
- Romer,  C.  D.,  &  Romer,  D.  H.  (2002).  A  Rehabilitation  of  Monetary  Policy  in  the  1950's. American Economic Review, 92(2), 121–127.
- Romer, C. D., & Romer, D. H. (2004). Choosing the Federal Reserve chair: lessons from history. Journal of Economic Perspectives, 18(1), 129–162.
- Rowlinson, M., & Procter, S. (1999). Organizational culture and business history. Organization Studies, 20(3), 369–396.
- Scott-Smith,  G.,  &  Rofe,  J.  S.  (Eds.).  (2017). Global  Perspectives  on  the  Bretton  Woods Conference and the Post-War World Order. Springer.
- Van der Wee, H. (2012). “Belgian Monetary Policy under the Gold Standard during the Interwar Period”. In The Gold Standard Peripheries (pp. 130–144). Palgrave Macmillan, London.

*References as listed in the source PDF wpiea2019161 - References*

---


_Source: https://www.imf.org/-/media/files/publications/wp/2019/wpiea2019161.pdf_
