The Safest Gold Is the Gold a Country Can Still Reach: Why Central Banks Are Rethinking Foreign Vaults

When Western governments immobilised hundreds of billions of dollars belonging to Russia’s central bank after the invasion of Ukraine in 2022, reserve managers around the world were given a reminder that the safety of a reserve asset depends on more than credit quality, market liquidity or the financial strength of the institution holding it. It also depends on jurisdiction. Russia could still own foreign securities and deposits in a legal sense while losing practical access to them almost immediately because those assets sat inside financial systems controlled by governments that had decided to impose sanctions. Its domestically held gold was different. The metal was still subject to the economic consequences of sanctions, including reduced market access and greater difficulty converting it internationally, but the physical bars themselves could not be immobilised by an instruction issued through a foreign banking system. That episode has pushed an old reserve management question into a different geopolitical environment. During the twentieth century, central banks often moved gold abroad because the principal danger was military invasion, and the safest reserve was one placed beyond the reach of an advancing army. In the twenty-first century, armies still matter, but reserve managers also have to plan for sanctions, asset freezes, changes in diplomatic recognition, financial restrictions, court orders and interruptions to cross border settlement. A vault can therefore be physically impregnable and still fail the sovereign that owns the gold if access to the asset ultimately depends on another government’s legal and political consent. The issue is no longer simply where gold is least likely to be stolen; it is where a country is most likely to retain control over it when relations with foreign powers deteriorate. JPMorgan Chase Chief Executive Jamie Dimon recently described telling officials from an unnamed country that withdrawing gold from the United States could make it less secure because several armies were located within a few hundred miles of their territory, while bullion held in America benefited from the protection of two oceans, the U.S. military and a highly developed legal system. His argument captures the original logic behind New York’s rise as a global gold custodian, and history provides ample support for it. The weakness is that modern reserve management cannot treat physical attack as the only scenario worth insuring against. A central bank that stores bullion domestically faces the risk that its own territory becomes insecure; one that stores bullion abroad gains physical security but adds foreign jurisdiction, political recognition and access risk to the equation. Russia Changed What Reserve Managers Mean by Safety Before 2022, the largest reserve managers generally regarded U.S. Treasury securities, deposits at major Western central banks and other highly liquid sovereign assets as close to the benchmark of financial safety because they combined low default risk with deep markets and immediate convertibility. Russia demonstrated that an asset can remain financially sound and still become unusable to its owner, and that distinction has become central to the debate over gold because physical bullion has one characteristic that foreign currency securities do not: when it is stored inside national territory, it can exist outside the direct operational reach of another country’s banking system, clearing infrastructure and sanctions authority. Gold held domestically does not make a country immune from economic pressure, nor does it guarantee that the metal can be sold or pledged freely when sanctions have closed major international markets. Its strategic value lies instead in preserving possession and decision-making authority. The country may still face a difficult choice about how to mobilise the metal, but another state cannot freeze the bars themselves by blocking an account at a custodian or prohibiting a transaction through its own financial institutions. For reserve managers watching Russia, the lesson is not that gold defeats sanctions, but that domestic custody removes one important channel through which sovereign assets can become inaccessible. The magnitude of central bank gold accumulation makes this question more consequential because gold is becoming more important inside reserve portfolios at the same moment that governments are reassessing the political risks attached to foreign financial assets. The World Gold Council has reported central bank purchases averaging around 1,000 tonnes annually in recent years, far above the pace of the previous decade, while reserve managers increasingly cite geopolitical risk, crisis performance and diversification as reasons for holding the metal. Venezuela Showed That Ownership and Control Can Separate Venezuela provides one of the clearest examples because the dispute concerned gold itself rather than securities or deposits. The Central Bank of Venezuela sought access to bullion held at the Bank of England, but the gold became entangled in a prolonged dispute over which Venezuelan authorities the United Kingdom recognised as legitimate. The bullion remained physically secure in London throughout the dispute, yet the institution claiming ownership could not simply instruct the custodian to return it. That episode illustrates why foreign custody creates a form of political optionality for the custodian jurisdiction. The Bank of England remains one of the world’s most respected reserve custodians, and London is still the dominant centre of physical gold trading because central banks can store bullion securely and transfer ownership quickly inside a deep market. Those advantages are substantial, but they come with a trade-off because a country that stores gold in London gains market access and professional custody while accepting that the metal sits inside British law. Under normal diplomatic conditions that arrangement is highly efficient; under severe political stress, the legal environment surrounding the asset can become as important as the purity or value of the bars. Afghanistan Reinforced the Jurisdiction Lesson Afghanistan added another warning after the Taliban returned to power in 2021 and billions of dollars of central bank assets held abroad were frozen. The dispute was not over whether the Afghan state had once accumulated those reserves but over who could lawfully control and use them after political power changed hands. Part of the frozen assets was subsequently transferred into a Switzerland-based fund rather than returned directly to the authorities in Kabul. Russia, Venezuela and Afghanistan differ significantly in law, politics and circumstance, but for reserve managers they reveal the same operational vulnerability: an asset held abroad remains exposed to the legal interpretation, foreign policy decisions and recognition policies of the state where it is custodied. That risk may remain dormant for decades, but reserve policy exists partly to prepare for precisely the events that are unlikely during normal times and extremely costly when they occur. Central Banks Are Already Changing Where Gold Sits The change in thinking is increasingly visible in central bank custody policies. The shift is not a wholesale retreat from London or New York because both remain critical to official sector liquidity, trading and settlement, but more central banks are deliberately dividing holdings among domestic and foreign vaults rather than allowing historical arrangements to determine where national reserves remain for decades. That distinction is central to the economics of custody because London and New York retain advantages that domestic vaults cannot always replicate. Central banks may need to sell gold rapidly, swap it for foreign currency, use it as collateral or lend it into the market, and those operations are easier when the bullion is already sitting inside an established trading hub. The custody decision therefore involves three objectives that do not always point in the same direction: sovereign control, physical security and market liquidity. Poland Repatriated Gold Without Abandoning London Poland provides one of the clearest examples of a central bank deliberately increasing domestic control while retaining access to international markets. Narodowy Bank Polski repatriated 100 tonnes of gold from the Bank of England in 2019 after substantially increasing the country’s official gold reserves, presenting the move as part of a wider strategy to strengthen national financial security and confidence. The importance of the Polish case lies less in the symbolism of transporting bars back to Warsaw than in the resulting distribution of risk. Poland did not conclude that London had become economically irrelevant or unsafe; it decided that a larger share of national reserves should be physically available inside the country while still maintaining holdings abroad that could be traded quickly. That approach reflects an increasingly common reserve management principle: foreign markets remain useful, but strategic assets do not need to be concentrated entirely outside national control. Austria Built a Deliberate Three-Location System Austria adopted an even more explicit diversification strategy when its central bank decided that approximately half of the country’s gold should ultimately be held domestically, while the remainder would be divided between the United Kingdom and Switzerland. The policy was designed to reduce concentration risk without sacrificing the liquidity available in major international bullion markets. The Austrian case is important because it treats vault location in the same way a reserve manager would treat currencies, bond issuers or financial counterparties. Holding everything in one place, even an exceptionally secure place, creates a single point of failure. Dividing gold across domestic and foreign locations preserves more options if one market becomes inaccessible because of war, sanctions, political disputes, operational disruption or financial stress. Hungary Took the Sovereignty Argument Further Hungary adopted a more explicitly national approach when the Magyar Nemzeti Bank dramatically increased its gold holdings and brought physical bullion into the country. Its reserves rose from just over three tonnes to more than 30 tonnes in 2018 and subsequently increased several times further, with the central bank presenting domestic custody as part of a broader strategy around financial stability, national confidence and strategic independence. Hungary illustrates how some central banks increasingly view gold not only as a reserve asset but as part of financial sovereignty. That position has obvious advantages in terms of control, although it also concentrates physical and political risk inside the country. Domestic custody is therefore strongest when it is supported by secure infrastructure, independent central banking institutions, transparent inventories and political stability. Germany Chose a Balance Between Sovereignty and Liquidity Germany’s strategy remains one of the clearest examples of compromise. The Bundesbank decided that around half of Germany’s gold should be stored in Frankfurt, while a significant share would remain in New York and London. The reasoning was straightforward: domestic custody strengthened public confidence and direct control, while overseas holdings in major trading centres preserved the ability to exchange gold for foreign currency rapidly when required. Germany subsequently transferred hundreds of tonnes from New York and Paris back to Frankfurt while retaining substantial stocks abroad. The policy recognises that gold performs two reserve functions simultaneously. It is an asset of sovereignty because it carries no issuer risk, but it is also an asset of liquidity because it can be exchanged, pledged or sold during financial stress. Domestic custody strengthens the first function; London and New York strengthen the second. France Brought Its New York Exposure Back to Europe France reached a similar outcome through a different route. Rather than physically transporting its remaining New York holdings across the Atlantic, the Banque de France sold older bars there and acquired replacement bullion in Europe, allowing the country to maintain the size of its gold reserves while relocating the relevant custody exposure closer to home. The French operation was presented primarily as a technical and reserve management decision rather than a geopolitical repudiation of the United States, but the outcome remains instructive because one of the world’s largest gold holders now exercises more direct geographical control over its reserves while continuing to operate inside international bullion markets. The Netherlands Shows Why Repatriation Is Not the Whole Story The Netherlands provides an especially useful counterweight to simplistic arguments that every bar should be returned home. De Nederlandsche Bank has rebalanced gold among several locations, including the Netherlands, London, New York and Ottawa, and has increased the importance of London because bullion stored there is highly liquid and can be mobilised quickly during a crisis. The Dutch approach shows that resilience can sometimes mean moving gold between foreign jurisdictions rather than bringing it home. A reserve asset that is physically secure but difficult to convert quickly can fail its liquidity function, just as one that is perfectly liquid but politically inaccessible can fail its sovereignty function. The objective is therefore not proximity for its own sake but optionality. India Shows Why Gold Must Remain Mobilisable India’s 1991 balance of payments crisis provides one of the strongest arguments against keeping every bar permanently isolated inside national vaults. With usable foreign exchange reserves approaching exhaustion, India pledged gold to foreign institutions to raise emergency financing and meet external obligations. The operation became one of the most famous episodes in the country’s modern economic history because it demonstrated that gold can become an emergency source of foreign currency when conventional financing channels close. The lesson is important because a gold reserve only becomes fully useful during a liquidity crisis if counterparties can accept, value and mobilise it quickly. Internationally recognised vaults and bullion markets make that process easier. India therefore reinforces the case for keeping a strategic core domestically while maintaining another portion within market infrastructure capable of converting metal into dollars, euros or other reserve currencies without delay. The Czech Republic Is Increasing Gold While Keeping It Tradable The Czech Republic represents the same logic in current reserve management. Its central bank has been increasing gold holdings rapidly but continues to store much of the metal in London because that location makes the bullion easier to trade, pledge or lend. The decision demonstrates that increasing gold and keeping gold at home are separate policy questions. A country can decide that it wants more gold because the asset provides diversification and geopolitical insurance while simultaneously deciding that some of those bars should remain abroad because reserve liquidity matters. The issue is therefore not whether London or New York are inherently unsafe, but whether a country has become excessively dependent on them. Norway Explains Why Gold Was Sent Abroad in the First Place The original case for foreign custody was compelling because central bank vaults could physically be captured during war. When Germany invaded Norway in April 1940, Norges Bank rapidly moved roughly 50 tonnes of gold out of Oslo using trucks, trains and ships before occupation forces could seize it, eventually transferring much of the bullion through Britain to North America. Norway’s successful evacuation demonstrated that reserves could survive the collapse of domestic territorial security if they were moved beyond the reach of an invading army. The experience helped reinforce the wider twentieth century logic behind placing gold in cities such as London, New York and Ottawa. Kuwait Shows What Happens When an Invading Army Reaches the Vault Kuwait experienced the opposite outcome during Iraq’s 1990 invasion, when Iraqi forces seized thousands of gold bars belonging to the Kuwaiti central bank. The bullion was eventually recovered following Iraq’s defeat and a United Nations supervised process, but the episode remains one of the clearest historical examples of the physical risk created by storing national monetary assets inside territory that can be occupied. The Kuwait case means Dimon’s military argument cannot simply be dismissed. For a small state located next to a militarily stronger potential adversary, storing at least some reserves abroad may be entirely rational because physical possession becomes a vulnerability if the state itself loses control of its territory. Libya Shows That Domestic Control Has Its Own Political Risk Libya complicates the argument further because its experience demonstrates that domestic custody does not eliminate sovereign risk; it changes its form. Before the 2011 uprising, Libya held roughly 143 tonnes of monetary gold, one of the largest official bullion reserves in Africa, but during the conflict its reported holdings fell by around 27 to 29 tonnes, generating years of speculation about whether part of the reserve had disappeared during the collapse of Muammar Gaddafi’s government. The available record indicates that the gold was not simply stolen from the central bank and left permanently unaccounted for. Libya’s newly appointed central bank leadership said after the conflict that the Gaddafi government had sold approximately 29 tonnes of bullion domestically to raise liquidity as war and international sanctions restricted access to cash. Later disputes over the decline in Libya’s reported reserves prompted the Central Bank of Libya to maintain that its remaining gold stock had not changed since August 2011 and that international audit work confirmed the balance. The Libyan experience is important precisely because it shows the opposite vulnerability to Venezuela. Venezuela demonstrated how gold stored abroad can become inaccessible when a foreign government disputes the authority of the institution requesting it; Libya showed how gold held domestically can become immediately available to the government controlling the central bank during political collapse, allowing the reserve to be sold rapidly to finance salaries, emergency spending or the survival of the regime. Domestic custody therefore does not automatically mean that gold is safer for the nation as distinct from the government of the day. If political institutions collapse, control over the vault can become control over national reserves, which creates an internal governance risk that foreign custody can sometimes constrain. Libya also shows why auditing, disclosure and institutional controls matter almost as much as physical location. A country can keep all of its bullion within national territory and still face uncertainty over what happened to part of it during a period of war if reserve transactions are not subject to transparent authorisation, independent audit and credible public reporting. Sovereign custody is strongest when the physical security of the bars is matched by strong institutions governing who can sell, pledge or transfer them. The Different Cases Point to Different Risks Taken together, Norway, Kuwait, Libya, Venezuela and Russia demonstrate why there is no single perfect location for monetary gold. Norway and Kuwait show the danger of military exposure; Libya shows the danger of internal political breakdown; Venezuela demonstrates foreign legal and recognition risk; Russia demonstrates sanctions and jurisdiction risk. India, meanwhile, shows why internationally mobilisable gold can become extremely valuable when a country urgently needs foreign currency. The modern reserve manager therefore has to protect against several different failure modes simultaneously. Keeping everything at home eliminates one set of risks while increasing others, just as storing everything abroad maximises international market access while surrendering part of the country’s control over the legal environment surrounding the asset. This is why diversification of custody is becoming as important as diversification of the reserve portfolio itself. Gold’s Rising Role Makes Custody More Important The debate over location would matter less if central banks were reducing exposure to gold, but the opposite is happening. Gold has become more prominent within official reserve strategies because it carries no sovereign credit risk, behaves differently from currencies and government bonds, and has historically performed well during periods of geopolitical and financial stress. A government bond is ultimately another state’s liability, while a bank deposit is another institution’s promise to repay. Physical gold is different because it is the asset itself, and that characteristic is one of the principal reasons countries hold it during periods of uncertainty. The economic logic becomes less compelling, however, if a country owns the bullion but cannot access it. The location of the bars therefore becomes part of the asset’s risk profile. Tanzania Now Has to Think About Custody as a Strategic Question This issue is becoming particularly important for African central banks because several countries are building reserves directly from domestic gold production. Tanzania has become one of the clearest examples after the Bank of Tanzania accumulated around 28 tonnes over approximately 18 months under a domestic purchase programme intended to diversify reserves, strengthen financial stability and deepen formalisation in the country’s gold market. The scale of that accumulation means custody policy can no longer be treated as a minor operational question. Tanzania now has to decide how much of its growing bullion stock should remain physically inside the country, how much should sit in recognised international trading centres and how quickly those assets need to be convertible into foreign exchange during periods of pressure. Domestic custody would give Tanzania an asset that remains physically under national control even if external financial relationships deteriorate, while keeping another portion in London or another recognised bullion centre would make it easier to sell, swap or pledge during a shortage of dollars. The appropriate strategy is therefore unlikely to be an absolute choice between Dar es Salaam and London. It is a portfolio decision about how much sovereignty Tanzania needs to preserve and how much international liquidity it wants to retain. Ghana Shows Why More Gold Can Also Create Concentration Risk Ghana provides an equally important African case because its domestic gold purchase programme rapidly increased the role of bullion inside the central bank’s reserve portfolio. The initiative demonstrated how a gold producing country can transform part of its own mineral output directly into official reserves rather than exporting all of it and relying entirely on foreign currency accumulation. The Ghanaian experience also shows that reserve sovereignty cannot be reduced to accumulating the largest possible pile of gold. If bullion becomes too large a share of total reserves, the central bank can face concentration risk and may still need foreign currency to settle imports, repay external debt or intervene in exchange markets. The reserve portfolio therefore has to preserve a balance between gold, liquid currencies and internationally tradable securities. For African producers, this distinction is particularly relevant because buying domestic gold can reduce reliance on external borrowing for reserve accumulation and formalise mining supply chains, but the benefits are greatest when ownership, location and portfolio composition are treated as separate decisions. The Best Reserve Structure Is Built Around Redundancy The evidence from Russia, Venezuela, Afghanistan, Norway, Kuwait, Libya, India, Poland, Austria, Hungary, Germany, France, the Netherlands, the Czech Republic, Tanzania and Ghana points toward a more sophisticated conclusion than simply bringing every bar home. Countries need redundancy across assets, custodians and jurisdictions because each arrangement protects against a different failure. Domestic custody protects sovereign control and reduces exposure to foreign courts and sanctions. London provides exceptional market liquidity and access to the world’s dominant physical gold market. New York offers deep official sector infrastructure and extraordinary physical protection. Switzerland and other established bullion centres provide additional jurisdictional diversification. Holding gold across several locations reduces the probability that a single political dispute, financial disruption, invasion, domestic political crisis or operational failure renders the entire reserve unusable. The appropriate allocation will differ by country. A state facing a credible invasion risk may rationally keep more bullion abroad, while a geographically secure country with strong domestic vaulting infrastructure may prefer a larger local allocation. A country heavily dependent on emergency foreign currency liquidity may place more gold in London, while a government concerned about sanctions or geopolitical confrontation may assign greater value to maintaining a larger strategic core at home. The important question is no longer whether domestic or foreign custody is universally superior. It is whether the country has preserved enough options to function when one of those channels fails. A Reserve Is Only Fully Sovereign When Access Survives the Crisis The experience of the last several decades has changed what central banks should mean when they describe an asset as safe. Creditworthiness matters, but it is no longer sufficient. Liquidity remains essential, but it can become meaningless if legal access disappears. Physical security matters, but protection from invasion does not guarantee political availability, while keeping bullion inside national territory can itself create risks when governments collapse or institutions fail. Russia demonstrated how foreign reserves can become unusable without a bank default or market crash. Venezuela showed that gold can remain perfectly secure while its owner spends years unable to obtain it. Afghanistan showed how political recognition can transform the status of foreign reserves almost overnight. Norway and Kuwait demonstrated why physical security once pushed central banks toward overseas vaults, while Libya showed that domestic gold can be rapidly mobilised by whichever authority controls the vault during political breakdown. India showed why internationally accessible bullion can become invaluable during a foreign exchange crisis, and Germany, Austria, Poland and the Netherlands demonstrate why modern central banks increasingly prefer a mixture of domestic custody and international market access. For African gold producers such as Tanzania and Ghana, the debate is becoming more immediate because domestic mining is now feeding directly into official reserves. These countries have an opportunity to build reserve structures deliberately rather than inherit custody arrangements established in another geopolitical era, choosing consciously how much bullion should remain within national borders, how much should sit in international trading centres and what institutional protections should govern its use. Jamie Dimon is correct that Manhattan offers a level of physical protection few countries can match. A hostile army is extraordinarily unlikely to cross two oceans, penetrate the United States and carry another state’s bullion out of a vault beneath New York. But reserve managers now have to prepare for a different possibility as well: the bars remain untouched, the custodian remains solvent, the vault remains secure, and the country that owns the gold is nevertheless told that it cannot access it. They must also prepare for the opposite scenario, where the gold is physically available at home but control of the state itself has broken down. For a sovereign reserve, safety therefore cannot be defined by geography alone. The safest gold is the gold a country can still control, mobilise and account for when the political, financial or security environment turns against it.