RESEARCH DOSSIER
Offshore Deposits
Abstract. This monograph investigates offshore deposits as monetary claims on banks whose booking geography, creditor residence, and currency denomination may differ. It integrates the historical development of eurocurrency markets, bank balance-sheet mechanics, deposit pricing, sectoral funding structure, foreign-currency liquidity, deposit protection, external accounts, and empirical research on international financial transparency. The analysis distinguishes conventional deposits from broader deposit-like funding statistics, gross liabilities from consolidated exposure, stocks from adjusted flows, immediate counterparties from beneficial owners, and statistical changes from causal policy effects. Hypothetical numerical examples explain accounting and liquidity mechanisms. Published findings on deposit responses to information exchange are interpreted within their study populations and periods. The original contribution is a structured economic synthesis connecting instrument-level analysis to global banking networks and policy evaluation. The research is a literature-based analysis, with no independently estimated global deposit stock or new named theory.
Keywords: offshore deposits; offshore banking; cross-border deposits; eurocurrency; Eurodollar; international banking; bank funding; liquidity risk; deposit insurance; locational banking statistics; financial transparency; beneficial ownership; tax compliance; external sector statistics
Introduction
Offshore deposits are deposit claims whose economic significance is shaped by differences between the jurisdiction of the deposit holder, the location of the bank office, and the monetary jurisdiction of the account currency. The relevant unit is an enforceable liability of a deposit-taking institution and the corresponding financial asset of its creditor. The study belongs to the economics of international finance because a claim denominated in one currency can be booked in a second economy, held by an entity resident in a third, and supported by a banking group managed in a fourth. These relations can coexist within a single account.
The term offshore has two historically important applications. A resident of one economy can maintain a deposit at a banking office in another; alternatively, a banking office can issue deposits denominated in a currency that is foreign to its own location. These applications overlap but do not exhaust each other. A domestic resident's foreign-currency deposit can participate in a eurocurrency market even though its immediate creditor and debtor are residents of the same economy. A foreign resident's local-currency deposit is a cross-border position even though its currency is domestic to the bank office.
The present research examines the intersection of these dimensions, with separate treatment of each measurable population. The subject includes deposit formation, contractual liquidity, bank funding, currency transformation, international statistical records, and the economic effects of financial transparency. It connects these topics through balance-sheet identities and a critical comparison of empirical studies. General offshore banking, credit allocation, securities issuance, legal doctrine, and estimation of all offshore wealth remain neighboring research objects. The evidence supports an analytical synthesis of existing scientific knowledge, not an independent estimate of the global stock of offshore deposits.
Three questions organize the inquiry. What does the economic object represent, and how can it be observed? What causes deposits to be booked across national and monetary boundaries, and what risks follow? What conclusions can be drawn from reported changes after monetary, prudential, or transparency shocks? Each question requires an appropriate population, because one series cannot simultaneously measure bank funding, private wealth, capital flight, and tax evasion.
The Deposit as a Financial Instrument
A bank deposit is a creditor's monetary claim against a bank. The contractual account balance, rather than the physical location of currency, establishes the claim. A demand deposit offers withdrawal or transfer on agreed terms; a time deposit establishes a maturity and may impose restrictions on early redemption. Interest may be fixed, floating, or absent. The denomination determines the units in which the bank promises performance. The bank's credit quality, legal entity, liquidity position, account terms, and applicable public guarantees determine the risk of performance.
The BIS statistical glossary (BIS, n.d.) groups nonnegotiable loans and deposits together in its locational banking statistics and includes some economically related funding arrangements. Negotiable certificates of deposit belong to its debt securities category. This makes a substantive difference: a funding statistic constructed by filtering the BIS loans-and-deposits instrument is wider than a population of conventional household bank accounts. In an account-level study the contract may be directly observed; in aggregate international statistics it must be inferred from the reported instrument and sector.
Deposits differ from custodial assets. A bank may provide services for securities owned by a customer, but securities held in custody ordinarily remain separate assets of that customer. Cash awaiting investment can be represented by a deposit or another contractual claim, depending on the arrangement. An investment fund share is a claim on a pool of assets rather than a deposit at its portfolio bank. The same private investor may therefore hold bank deposits, securities, and fund shares through one provider; economic measurement must separate the instruments.
The distinction between branches and subsidiaries also matters. A branch is an office of an existing legal entity; a subsidiary generally has its own corporate personality. An international brand does not determine the debtor in a particular contract. A banking group's consolidated balance sheet may identify the parent risk and internal exposures while the individual deposit contract determines the creditor's rights in an actual failure. Residence, ownership, and contractual liability have distinct functions.
The basic analytical distinction is between the financial claim, the institution that owes it, and the mechanism by which the institution funds and settles it. The deposit is held by its creditor; the bank can hold a separate correspondent deposit or reserve balance as an asset. Confusing these two claims counts the same transfer of money as a single asset at two economic levels or, conversely, suppresses a genuine chain of intermediation.
Jurisdiction, Currency, and Statistical Residence
The geography of offshore deposits requires a consistent reporting rule. In this research, the bank office's economy is the booking location, the creditor's economy is its immediate statistical residence, and the currency jurisdiction is the issuer's monetary area. The parent's headquarters and the beneficial owner's residence are additional attributes. Nationality, citizenship, location of a website, location of the account manager, and tax residence are separate concepts; a given data source may establish only some of them.
Consider four illustrative accounts. A resident of economy A holds A-currency at a bank in A: this is a local position in domestic currency. A resident of B holds A-currency at a bank in A: it is cross-border but in the bank's domestic currency. A resident of A holds B-currency at a bank in A: it is local in residence terms and foreign-currency in denomination. A resident of C holds B-currency at a bank in A: both characteristics apply. Each account has the same general economic form but a different exposure to international market conditions.
A financial center that intermediates deposits of nonresidents may be economically specialized even if its banks operate under widely used prudential and reporting rules. The institutional label alone establishes neither illegality nor the purpose of a specific deposit. Absolute size reflects the scale of the economy and the banking system, while a ratio of external liabilities to domestic GDP can be large because the GDP denominator is small. Both measures require interpretation in conjunction with sector, currency, and balance-sheet assets.
Contemporary reallocation research, including the BIS (2026) offshore activity primer, illustrates the importance of tracing financial entities to corporate headquarters. For deposits, a parent-based reassignment can show the banking groups that owe the liabilities and may help interpret home-supervisory exposure. It cannot generally reveal the natural persons who ultimately control depositor entities. The identity of the creditor and the controlling parent of the debtor are different ends of the financial relation.
These observations establish the measurement discipline of this monograph: specify both parties and the instrument before applying a jurisdictional classification. Statistical residence is suitable for external accounts. Nationality of a banking group is suitable for group funding and risk. Beneficial ownership evidence is required for household wealth or compliance claims. A tax-residence record is necessary for assigning applicable reporting obligations.
Historical Formation of the Offshore Deposit Market
The development of eurocurrency deposits transformed the geographic reach of commercial banks. Schenk (1998), in The Origins of the Eurodollar Market in London, traces its early institutional development during 1955–1963. London provided an environment in which banks could intermediate balances denominated in dollars outside the United States. The emergence of this activity depended on commercial transactions, institutional arrangements, and banking incentives in a particular historical period. The prefix euro in Eurodollar identifies the historical market convention; it does not imply that the deposit is denominated in the euro.
McCauley, McGuire, and Wooldridge (2021) situate these developments in Seven Decades of International Banking. Differences in the costs and constraints imposed on intermediaries permitted offshore offices to compete for foreign-currency funding. Regulation, liberalization, innovation, and shifts in external financing needs subsequently transformed the market. The historical opportunity arose because currency denomination and the location of the bank liability could be separated.
A deposit in an offshore office was attractive to some institutions because the account could provide liquid balances in a widely accepted currency while their operating payments, trading partners, or liabilities were denominated in the same unit. Banks could also draw funds from other banks to expand foreign-currency lending. The resulting market was both a payment-related liquidity network and a mechanism for wholesale refinancing. It was susceptible to shifts in the relative regulatory and financing costs of competing banking offices.
The evolution of offshore banking should be distinguished from a narrative of funds permanently leaving a currency-issuing economy. A foreign bank can use a borrowed dollar to lend to an entity in the United States, to finance a transaction between two other economies, or to hold dollar assets for liquidity. He and McCauley (2012) emphasize in Eurodollar Banking and Currency Internationalisation that the development of the offshore dollar market has little necessary relation to the direction of net capital flows. The gross scale of currency intermediation can grow even when net external flows change little.
The historical record therefore supports an institutional rather than geographically deterministic account. Market size responds to differential funding costs, international business demand, the capacity of banks to settle obligations, and access to financial networks. Changes in domestic regulation can alter the original incentive without eliminating the commercial value of an established international network. Historical explanations of the creation of a market must be distinguished from explanations of its continued operation.
Balance Sheets, Money Creation, and Settlement
Bank balance sheets organize the mechanism independently of the location of the office. Assets equal liabilities plus equity. A deposit of 100 held at bank X is an asset of 100 for the creditor and a liability of 100 for X. If X receives funds from another bank and obtains a corresponding settlement asset, it records an increase in both assets and deposit liabilities. If X instead credits a borrower as part of a loan origination, the original increase can be a loan asset matched by a new deposit liability. These are different balance-sheet events.
Consider an illustrative closed chain. Household H initially holds a deposit of 100 at bank X. Economics A and B host X and Y, respectively. H transfers 40 to a deposit at Y. Where X and Y settle by moving an existing common settlement asset, X's liabilities to H and assets each decrease by 40; Y's settlement assets and liabilities to H each increase by 40. H's total deposits remain 100. Economy B records a new liability to a nonresident if H lives in A, while the external position of A depends on who holds and issues the settlement asset. Accounting for the settlement leg is indispensable.
Alternatively, Y grants a loan of 40 to a third party and credits its account. Y's loan assets and deposit liabilities both increase by 40; outstanding deposits in the specified banking system increase by that amount before subsequent transfers or repayments. This creation of a deposit does not free Y from settlement constraints. Withdrawal or payment to another bank requires transferable liquidity in the relevant currency, borrowing capacity, or access to a settlement correspondent.
Where the currency issuer's central bank does not provide direct access to an offshore bank, liquidity can be held as claims on eligible intermediaries and obtained through wholesale markets. The offshore deposit is an obligation of the offshore bank, not a direct central bank liability. A depositor may have a stable monetary unit of account and still bear the private bank's credit and settlement risk.
Consolidation changes the recorded chain. Suppose Y places 25 at Z, another bank, after receiving a customer deposit. The two banks' gross deposit-like liabilities are 125, of which 25 is interbank. At the combined banking-sector level, the interbank position offsets against the corresponding claim; external bank liabilities are 100 under the assumptions of this example. The 25 is a real financial obligation even though it cancels in consolidation. Gross network exposure and the net liability to outside sectors describe different analytical objects.
Foreign-currency lending additionally creates a distinction between the currency of assets and liabilities and the location of the borrower and depositor. A bank can match principal in the same currency and still incur a funding risk through unequal maturities. A matched notional currency position need not be a matched liquidity position. Settlement dates, pledged collateral, and counterparty withdrawal rights affect actual funding capacity.
Depositor Demand and International Pricing
Holding a deposit abroad can serve a range of economic functions: receiving export proceeds, paying suppliers, financing foreign operations, managing currency exposure, diversifying credit risk, and holding liquid savings. A corporation with expenses abroad may use a multicurrency treasury arrangement to reduce operational frictions. A household living across several economies may have legitimate payment needs in multiple monetary areas. Financial institutions can hold deposits for settlement, collateral management, or liquidity investment. These activities have distinct economic motivations even when their positions appear under a single cross-border label.
The expected value of a deposit can be expressed through its contractual interest, expected exchange-rate change against the depositor's spending currency, fees, hedging costs, expected credit losses, and expected restrictions on access. For an unhedged deposit of principal P, denominated in foreign currency with interest i, whose exchange rate in home currency changes from e0 to e1, the nominal home-currency gross value before other costs is P(1+i)e1. Dividing by the initial home-currency value Pe0 gives a gross return of (1+i)e1/e0. This is an accounting identity rather than a prediction of currency movements.
For illustration, a deposit of 10,000 foreign currency units earns 4% over a year. If the foreign currency depreciates 8% against the home currency, the gross home-currency return factor is 1.04 × 0.92 = 0.9568, a loss of 4.32% before taxes and fees. With no depreciation the contractual nominal return is 4%. The example isolates the exchange-rate channel; actual decisions also reflect correlated income, transaction needs, and risk preferences.
The bank's pricing problem has a different perspective. The value of marginal deposits depends on the cost of competing liabilities, their expected persistence, the liquidity buffer needed to meet potential outflows, associated services, and the expected return on the assets financed. A deposit paying a low nominal rate may be expensive if it is readily transferable, highly concentrated, or costly to service. A higher-paying term deposit can reduce short-run rollover pressure while exposing the bank to refinancing risk at maturity.
Observable rate differences cannot by themselves identify tax or secrecy motives. A depositor selecting one office may respond to payment infrastructure, institutional reputation, currency availability, distance from operating markets, or risk of local banking disruption. A valid empirical comparison would control for these differences and compare the instrument's full costs and benefits. Where data on individual motives are absent, the research should establish mechanisms and population-level evidence rather than assign motives to account holders.
Sectoral Composition and Funding Networks
The structure of deposits matters at least as much as their aggregate size. Household balances supporting regular transactions can behave differently from corporate treasury concentrations. Some time deposits have predictable maturity dates; others carry penalties insufficient to prevent early redemption. A money market fund or bank can reallocate a substantial balance as market rates, collateral needs, or counterparty assessments change. The number of accounts alone is a weak measure of diversification because the distribution of balances and their correlation are more important during a shock.
Consider a bank with 100 deposit units. In one case it has one hundred accounts of one unit each; in another it has four independent corporate depositors of 25 units. The aggregate liabilities and average contractual maturity could be equal, but a single withdrawal in the second case removes 25% of funding. Independence cannot be assumed in either case: small accounts facing common news or corporate accounts operating in a common industry can react simultaneously. Diversification of creditor concentration and diversification of economic exposure must both be measured.
Hardy and Zhu (2023) examine international banks’ deposit funding using an intentionally broad instrument concept. Their analysis of early 2023 indicates a change in funding composition toward nonbank financial institutions after accounting for interbank and central-bank-balance-sheet effects. Their jurisdictional examples show why the boundary of the reporting population matters when balances move from internationally active to domestically focused banks. The study concerns bank funding in its chosen statistical scope; it is not a direct estimate of private offshore savings.
A household that withdraws money from a bank to purchase money market fund shares can initiate a chain in which the fund reinvests the proceeds in bank liabilities, government securities, or secured lending. A decline in household deposit statistics and an increase in nonbank financial funding can therefore occur simultaneously. The underlying bank funding terms change, as does the sector that can demand liquidity. The ultimate household's economic wealth need not change by the full amount of either gross statistical movement.
For deposit economics, this example yields two consequences. First, aggregate deposits should be accompanied by sector-specific measures and information about contracts or instruments. Second, a reduction in measured offshore deposits cannot be presumed to be a reduction in offshore exposure or total financial wealth: investors can switch holding forms, institutions, and contractual counterparties. These are observable alternatives to be evaluated, not instructions for avoiding financial reporting.
Bank Runs, Liquidity, and Currency Mismatch
A deposit provides a valuable degree of liquidity to a creditor while supporting the bank's transformation of liabilities into assets with different maturities. The classic research of Diamond and Dybvig (1983), Bank Runs, Deposit Insurance, and Liquidity, shows how demandable deposit contracts can give rise to a run equilibrium. Its stylized assumptions clarify the link between promised liquidity, withdrawal expectations, and costly asset liquidation. Cross-border banking adds currency and supervisory boundaries to this existing problem.
A bank can be solvent in the long run but unable to produce the required settlement asset on a particular day. Consider a simplified foreign-currency balance sheet: cash and eligible liquid assets 15, longer-term loans 85, deposits 90, and equity 10. The balance sheet is initially balanced at 100. If depositors request 30 and no new funding is available, the immediate liquidity shortfall is 15. If loans must be sold at 80 cents per unit of book value, generating 15 in cash requires disposing of 18.75 of loans; the loss is 3.75. Equity falls to 6.25 under these assumptions. Greater forced-sale losses can cause insolvency. This illustration distinguishes liquidity pressure from the asset loss through which the former can impair solvency.
The Basel Committee's Liquidity Coverage Ratio uses a thirty-calendar-day stress horizon, a defined portfolio of high-quality liquid assets, and prescribed net outflow assumptions. Its distinction between deposit categories shows why prudential analysis cannot treat all balances alike. Regulatory runoff coefficients are scenario parameters for the applicable rules. They are not predictions that a named offshore bank will experience exactly those withdrawal rates.
Foreign-currency funding adds the difficulty of acquiring a specific unit of settlement. A bank may possess liquid domestic-currency assets whose conversion incurs material cost or is constrained at the moment of need. Hedging against currency valuation changes using derivatives does not guarantee delivery of cash at the required time; margin calls, rollover requirements, and counterparties' own liquidity pressure may matter. A group with assets in several currencies must evaluate liquidity in each material currency as well as consolidated liquidity.
The CGFS report on foreign currency funding (2026) identifies cross-border intragroup transfers, liquidity assistance, and prudential measures as issues in the resilience of international funding. A group's capacity to redistribute funds depends on internal arrangements and external restrictions. A locally solvent subsidiary and a globally solvent parent can face different liquidity constraints, particularly if transfers are limited during stress.
Profitability analysis needs to treat this liquidity support as costly insurance. A bank can improve reported interest margins by financing longer-duration assets with short liabilities. When funding concentration and currency mismatch increase, the expected private benefit must be evaluated beside the expected losses under stress and possible external cost to deposit protection or official liquidity facilities.
International Dollar Funding and Crisis Transmission
Offshore dollar deposits connect foreign banks with the global dollar system. McGuire and von Peter (2009), in their study of the dollar shortage, reconstructed national banking systems' consolidated positions and linked structural US-dollar funding needs to disruption of interbank borrowing and swaps during the global financial crisis. The mechanism is a reminder that assets denominated in a globally accepted currency do not necessarily secure refinancing in that currency.
An increase in the price of dollar liquidity can transmit through at least three channels. Banks that replace maturing wholesale deposits incur higher funding costs. Banks that hedge non-dollar funding into dollars can encounter more expensive or restricted swap markets. Banks selling assets to meet redemptions may affect prices used to value comparable portfolios elsewhere. These channels interact: the effort of one institution to increase its liquidity can increase the cost of liquidity for another.
Aldasoro, Ehlers, McGuire, and von Peter (2020) report in their analysis of dollar funding needs that the gross dollar liabilities of banks headquartered outside the United States were about 13 trillion US dollars at the end of 2019. That is a broad liability total for the indicated banking population and date, not an estimate of offshore deposit balances. The study compares short-term funding pressures with international central bank liquidity arrangements.
Swap lines allow a participating central bank to obtain a foreign currency from its issuing central bank on agreed terms and distribute liquidity through its own eligible channels. The instrument may reduce disorderly refinancing pressure but generally works through institutions and collateral rules, not a universal promise to every offshore depositor. The issuing currency's lender-of-last-resort capacity and the deposit contract's repayment jurisdiction thus occupy different institutional levels.
The international distribution of dollar deposits produces a tension between private network integration and public institutional boundaries. The banking system values the ability to allocate liquid balances across markets, whereas crisis authorities must consider local stability, counterparty eligibility, and public losses. In research, the relevant comparison is not merely the total scale of gross claims but their maturity, currency, collateral, concentration, and access to liquidity. This interpretation also applies to other internationally used currencies under their own institutional arrangements.
Deposit Protection and Resolution Economics
Deposit insurance can reduce the incentive of eligible depositors to withdraw funds merely because others might do so. It may also weaken the incentive to monitor a bank's risk if the guarantee is perceived as broad and certain. Policy design therefore balances confidence, loss allocation, administrative capacity, and the discipline of creditors. These are effects of institutions and contract coverage, rather than properties inherent in a cross-border account.
IADI's Core Principles for Effective Deposit Insurance Systems were updated in September 2025. They provide an international reference for evaluating deposit insurance arrangements. A particular account's eligibility still depends on the effective legislation, participating institution, type of claimant, instrument, currency, and relevant limits. A global brand and the nationality of its parent do not determine the protection of an account booked at one of its overseas offices.
Resolution questions cannot be settled by general statements about offshore centers. A subsidiary's insolvency estate and a branch's obligations may be treated differently under applicable arrangements. Cross-border recognition, allocation of assets, and coordination between authorities affect recovery and timing. This monograph therefore analyzes the economic consequences of uncertainty over legal rights without stating country-specific coverage or priority rules.
Consider two hypothetical deposits, both carrying a promised principal of 100 and interest of 3. One carries credible protection for the entire eligible principal under the assumptions of the example; the other is unprotected and has a possible recovery of only 60 if the bank fails. Equal contractual yields do not imply equal expected outcomes. If the failure probability were assumed to be 2%, and interest and timing were ignored for simplicity, the expected unprotected principal would be 99.2. This is an illustrative computation, not a probability estimate for any institution. A real comparison also includes delays, coverage limits, and currency conversion.
The public policy implications extend beyond an account's private payoff. Where public protection supports one jurisdiction's offices but funds are allocated to risks elsewhere, home-host coordination is important to understand the distribution of potential losses. Where protection is absent or limited, highly mobile funding can withdraw rapidly. Stability analysis therefore requires an explicit inventory of liabilities' contractual and institutional characteristics, with uncertainties carried through the stress scenario.
External Accounts and International Banking Statistics
The central public tool for comparative measurement is the BIS locational banking statistics. Its reporting population consists of participating jurisdictions' reporting banking offices, with positions recorded from the perspective of their booking locations. The BIS distinguishes local and cross-border counterparties, currencies, instruments, and sectors where reported. These dimensions allow analysts to isolate a defined subset of deposit-like external funding. They cannot be treated as interchangeable with a census of all bank accounts worldwide.
External accounts answer a related question from the economy's perspective. The ECB's (2025) balance of payments sources and methods identifies currency and deposits among other-investment instruments, subject to functional-category exceptions such as reserve assets or direct investment where applicable. A resident's deposit claim on a nonresident bank is generally an external financial asset; the nonresident bank's obligation corresponds to an external liability in its own economy. A domestic deposit denominated in foreign currency remains a domestic creditor-debtor position if both parties are residents of the same economy.
The IMF announced the seventh edition of its integrated external accounts manual in March 2025 through the release of new macroeconomic standards (IMF, 2025). Researchers must record which statistical methodology each reporting jurisdiction actually uses and when changes become effective in its published data. The availability of a new international manual does not imply immediate uniform compilation of historical and contemporary series.
The BIS and external sector accounts may differ in coverage, classification, and the information on reporting institutions and counterparties. External accounts cover relevant resident institutional units beyond banks; BIS banking data emphasize banks and their reported balance sheets. Some instruments categorized together in one source may be separated in another. A discrepancy warrants investigation of the perimeter before attributing it to unrecorded or illicit wealth.
A useful empirical table would identify the reporting jurisdiction, counterparty residence, sector, currency, instrument, time, measure, reporting status, and source vintage. The researcher should retain suppressed observations as missing, record break markers, distinguish reported zero from unavailable data, and preserve denomination and conversion choices. Sample composition should remain explicit when comparing growth rates or cross-country rankings. A study measuring the organized international banking network can tolerate a different perimeter from a study estimating household wealth.
A resident's liability to a nonresident bank and that resident's claim on a nonresident bank may coexist. Netting them answers a net external financing question. Adding them measures gross exposure from the resident's perspective, subject to instrument and valuation consistency. Neither calculation establishes the presence or absence of a tax violation.
Stock–Flow Reconciliation and Aggregation
An outstanding balance is the value of a position at a specified observation date. A flow accounts for a change arising from transactions during a period. Changes in outstanding foreign-currency positions expressed in a common currency may reflect valuation, statistical breaks, or changes in coverage as well as transactions. A stock difference cannot be described as capital inflow or outflow without adjustment.
Suppose an offshore account contains 100 foreign currency units at two consecutive reporting dates. The exchange rate rises from 1.10 to 1.30 US dollars per unit. The reported dollar-equivalent balance rises from 110 to 130. There is a 20-dollar increase in the measured stock even though no deposit transaction occurred. This elementary example explains why international deposit series reported in US dollars must be interpreted with attention to exchange rates.
Now assume instead the account rises from 100 to 110 foreign currency units because of a net deposit transaction. At the second exchange rate its stock is 143 dollars, a rise of 33. Using a hypothetical average exchange rate of 1.20 gives a transaction approximation of 12 dollars, while the residual 21 relates to the conversion and timing assumptions. The full reconciliation would account for exact transaction dates, interest accrual, and any other valuation or classification changes. The BIS adjusted-change methodology explicitly addresses exchange rate movements and breaks in reported banking stocks.
Aggregation poses a separate difficulty. An individual deposits 100 with bank A. Bank A deposits 60 with bank B. Bank B places 40 with bank C. In this simplified chain, counting every deposit liability yields 200. Counting only the liability to the original household gives 100. Counting interbank funding yields 100. These are consistent answers to different questions. If the banks belong to one group, consolidated statistics can eliminate internal positions under the applicable perimeter; if they belong to separate groups, their mutual credit exposures remain relevant for contagion even if consolidated across the entire banking sector for another purpose.
A researcher should therefore choose the aggregation operator after defining the economic population. Total gross interbank links indicate network scale, nonbank liabilities indicate the sector's external funding as defined, and household-sector deposits indicate directly classified household positions. When a nonfinancial corporation acts as an intermediary for an individual, the immediate statistical sector may differ from the ultimate economic owner. An aggregate correction based solely on a corporation's jurisdiction would require unverified assumptions.
Reconciliation also requires distinguishing a changed series from a changed world. A reporting country joining the dataset, a bank being reclassified, or a currency breakdown becoming available can change the sample's total. Comparison on both an available sample and a consistent reporting cohort provides insight into whether results are driven by coverage. Researchers should disclose the extent to which such robustness tests are possible using the data actually obtained.
Financial Geography, Mobility, and Economic Development
Banks locate international deposit activity where a combination of monetary, organizational, informational, and institutional services is valuable. Network effects can arise when specialized offices attract firms with related legal, accounting, payment, and asset-management needs. These advantages can make a location attractive even if its domestic savings are limited. The location's domestic output and its offshore liability volume need not move proportionally.
Capital mobility can confer benefits on depositors through competition, payment efficiency, currency choice, and diversification. It can also create dependencies: banks in booking centers can rely on funds from economies exposed to common interest-rate shocks or changes in institutional rules. A high gross position can be associated with low local credit exposure if most corresponding bank assets are abroad. Conversely, local households and firms may be vulnerable if international deposits finance domestic credit and withdraw during a shock.
Economic impact should be assessed with a distinction between services produced locally and financial claims booked locally. Banking profits attributable to local services, employment, demand for professional services, infrastructure costs, tax revenues, and expected public liabilities are separate measures. Gross offshore balances are not the value added of offshore activity. An economy may be a booking location for large claims but retain only a limited share of the income generated by the network.
A comparative study of financial centers would therefore report multiple indicators: cross-border deposit-like liabilities relative to GDP, residence and sector composition, gross assets, net external positions, bank ownership, local financial-sector employment and value added, and evidence of local fiscal or contingent exposure. Country selection should follow the research question and available evidence; an inherently heterogeneous group of financial centers does not constitute a homogeneous treatment category.
Deposit mobility alters the power of both depositors and jurisdictions. A bank dependent on internationally mobile balances may compete intensely for them, while a government providing infrastructure or institutional credibility may gain an economic service sector. The effects depend on the stability of the funding base and how the resulting profits, wages, public revenues, and losses are distributed. This analysis is a set of ordinary economic deductions, not a new theory of offshore development.
Transparency, Tax Compliance, and Financial Privacy
International deposits can be lawful and fully declared. A bank account's location provides insufficient evidence about compliance. Tax obligations depend on applicable legal rules, the taxable person, the source and character of income, reporting obligations, and periods of assessment. The same financial instrument can support trade settlement, lawful savings, or conduct that violates tax or other laws. Research should distinguish a documented offense from a suspected or statistically inferred phenomenon.
The OECD's consolidated Common Reporting Standard (2025) describes the system of automatic annual exchange of financial-account information among implementing jurisdictions. The standard was adopted in 2014; the consolidated 2025 text incorporates amendments adopted after its review. Implementation timing, activated relationships, eligibility, and the relevant domestic provisions determine real-world coverage. The consolidated text's publication date is not a uniform effective date for all countries.
Automatic exchange changes the information environment facing some depositors and authorities. In a basic economic interpretation, an increase in the expected probability of detecting undeclared taxable income can reduce the expected private benefit of noncompliance. Affected agents can respond through compliance, changes in asset allocation, consumption, or financial arrangements. The researcher must distinguish such responses using the observable outcome and the institutional context.
The cost of information reporting is not limited to deterrence. Banks and tax authorities incur implementation, data-quality, security, and administrative costs. Legitimate privacy is a valuable institutional feature, and improper use of confidential information can harm households. Effective public administration requires lawful access and data protection as well as accurate matching and enforcement. An empirical evaluation based solely on deposit balances cannot measure these welfare dimensions.
For cross-border deposits the distinction between immediate creditor, account holder, and controlling person has particular relevance. A statistical banking return can classify a corporate creditor by residence, while tax-reporting systems may have additional information about controlling persons in designated cases. These reporting systems serve different purposes and cannot be merged without legal authority and a consistent definition. Public aggregate data can establish network-level patterns; individual legal conclusions require appropriate evidence.
Evidence from International Transparency Reforms
The empirical literature offers several distinct research designs. Johannesen and Zucman (2014), The End of Bank Secrecy?, study the bilateral agreement wave associated with G20 pressure during the financial crisis. Their findings suggest relocation of balances among havens rather than broad repatriation. The paper is evidence of a response to the earlier treaty regime studied, not a quantitative evaluation of every subsequent automatic-reporting system.
O'Reilly, Parra Ramirez, and Stemmer (2019), in Exchange of Information and Bank Deposits, report that foreign-owned deposits in their IFC population fell by 24%, or USD 410 billion, from 2008 to 2019. Their analysis associates commencement of automatic exchange with an average decline of 22% in deposits from non-IFC counterparties, and earlier request-based exchange with a smaller association. The first statistic is a descriptive sample-level change; the second is an estimated policy association. Neither number identifies the amount of tax recovered.
Beer, Delgado Coelho, and Leduc (2019), Hidden Treasure, examine the period 1995–2018 using bilateral data for 39 reporting countries and more than 200 counterparties. Their IMF working paper reports an estimated average decline of 25% in the foreign-owned deposits of offshore jurisdictions linked to automatic exchange. They find no statistically significant effect of exchange upon request in their preferred reported analysis. The difference between results on request-based agreements in the two 2019 studies merits attention to specification, sample, timing, and definitions rather than arithmetic averaging.
Casi, Spengel, and Stage (2020), in Cross-Border Tax Evasion After the CRS, analyze the short-term impact of the CRS using deposit changes in haven and nonhaven jurisdictions around policy timing. The study's comparison is informative because it distinguishes treatment from a reference population. The choice of comparison group and the extent to which deposits can move between the groups are central to interpretation. This review does not assign a new numerical estimate to that article without a sufficiently retrieved final result table.
The descriptive and modeled findings from different studies are mutually informative, but the reported percentages should remain separate. Some use restricted banking statistics, country classifications differ, treatments include different instruments of information exchange, and the relevant windows span different phases of implementation. Similar estimates across those designs reinforce the plausibility of a deposit response; their common dependence on bilateral deposit statistics limits their ability to prove individual motives.
A direct comparison between a fall of 22% and one of 25% cannot establish that one regime was three percentage points more effective. The estimates use different populations and models. A defensible comparison would recover the exact dependent variable, weighting convention, treatment definition, timing rules, fixed effects, confidence intervals, and robustness checks in each study. The available records support the directional synthesis and scoped reported estimates, not an independently computed meta-analysis.
The literature also motivates a distinction between deterrence of new undeclared accumulation and adjustment of an existing offshore stock. A policy can reduce future contributions with little immediate change in balances if existing assets are retained. A large initial movement may reflect portfolio reallocation even if the long-run probability of noncompliance changes little. Observing stocks alone provides an incomplete picture of both margins.
Household Wealth, Distribution, and Asset Substitution
Bank deposits constitute a subset of household financial wealth. Private offshore assets can include securities, equity in entities, investment fund shares, and other instruments. Bank balances serve a transaction and liquidity function that differs from the long-term portfolio function of these assets. A dataset of bank deposits cannot be scaled into a global comprehensive wealth estimate using an assumed universal ratio without evidence that the composition assumption holds.
Alstadsæter, Johannesen, and Zucman (2018), in Who Owns the Wealth in Tax Havens?, analyze offshore household wealth and its consequences for inequality using macro evidence and distributional assumptions. This is a broader research object than deposits. The distribution of all offshore wealth cannot be assigned directly to a population of bank accounts; a high-balance bank account may belong to a corporate treasury, financial institution, or other immediate counterparty.
Consider a household with 100 financial wealth units initially recorded as 40 deposits and 60 securities. If it switches 20 units from deposits to securities held abroad, deposits fall by 50%, while its foreign financial wealth remains 100 under the assumptions of the transaction. If the securities are instead issued by a domestic entity, the external portfolio can change even if trading occurs through the same foreign custodian. Who owes the instrument, who holds it, and how it is classified determine the appropriate measure.
The distributional consequences of compliance reforms depend on which people are affected, the change in taxable income reported, and enforcement and penalties where lawful. A large movement of gross deposits may result from a limited number of large institutional depositors, while a small aggregate change in deposits can conceal significant individual disclosure. Inferring distributional effects requires a source that links relevant units to household characteristics.
The NBER working paper of Alstadsæter, Johannesen, Le Guern Herry, and Zucman (2026), Global Financial Transparency and Offshore Wealth Accumulation, uses Norwegian administrative data to study wealth flows around automatic exchange. This approach illustrates the value of research at a more granular level than bilateral bank aggregates. The present review records the design from available abstracts; it does not extrapolate a result from Norway to all countries or claim a completed assessment of the working paper's robustness.
A policy evaluation can combine aggregate banking data with authorized anonymized tax records, disclosure-program information, household wealth statistics, and portfolio data where legally and methodologically appropriate. Such triangulation might separate changes in recorded deposit holdings from changes in taxable income or declared wealth. The necessary linkage is a matter of research design, privacy, and access. No individual ownership inference is justified from a jurisdiction-level stock alone.
Empirical Research Design and Identification
A reproducible study of offshore deposit changes begins with a precise estimand. For example, the outcome might be exchange-rate-adjusted change in cross-border liabilities of reporting banking offices classified as loans and deposits vis-à-vis nonbank residents of an explicitly listed set of economies. A different study might estimate changes in conventional household deposits using an administrative source. The populations and interpretations must be documented separately.
A bilateral panel can index reporting location i, counterparty residence j, and quarter t, with currency and sector where available. A descriptive analysis can compare the levels and adjusted changes before and after an event while keeping reporting coverage fixed. To support causal interpretation, the researcher needs an explicit counterfactual: how would the same population have changed absent the policy or shock? Parallel trends in an appropriate outcome, lack of anticipatory changes under the intended design, and treatment timing are important assumptions rather than guaranteed properties.
A simple two-group two-period comparison illustrates the method. Suppose treated deposit balances decline from 100 to 80 and comparison balances decline from 100 to 95. The difference between their percentage changes is minus 15 percentage points in this deliberately equal-base example. A log specification would instead compare log(0.8) with log(0.95); the transformed result would have a different numerical interpretation. Neither design automatically identifies a policy effect if the treated group also faced a separate banking crisis.
The treatment in a transparency study could be the enactment of domestic legislation, first scheduled exchange, observed first exchange, or creation of a bilateral reporting relationship. Those events need not coincide. Anticipation and staggered adoption can bias a simplistic comparison if already-treated observations are used as controls for newly treated observations. Researchers should make treatment timing explicit, inspect pre-treatment trends, consider cohort-specific effects, and use a comparison strategy appropriate to the actual sequence. Alternative classifications of reporting centers should be tested rather than treated as immutable ground truth.
Spillovers create another identification challenge. If funds can move between the treated and comparison groups, the latter's outcomes can be affected by the same policy. The resulting difference can combine the decline in the treated group and the increase in the comparison group. The estimand may then capture a relative reallocation response rather than the global change that would occur under universal treatment. This is a substantive question that should be explored using flow evidence and institutionally justified groups.
The use of log outcomes raises an additional issue for zero balances. Logarithms of zero are undefined; replacing zero by an arbitrary small number can affect results, especially for small country pairs. Researchers should justify transformations, distinguish missing observations from true zeros, and report sensitivity to sample rules. If confidential data restrict verification, the paper should identify which assumptions can be externally assessed from published tables, and which remain dependent on the original researchers' access.
This monograph has not acquired a complete bilateral data extract, fitted a regression, or replicated the cited studies' coefficients. Its statistical analysis is methodological and illustrative. A future empirical test would preserve downloaded series versions, country and instrument filters, adjusted flow methods, the exact coding of treatment, analysis code, model diagnostics and a record of changes between revisions. That procedure would make independently produced estimates auditable without pretending that aggregate data answer an ownership question they were not designed to resolve.
Policy Evaluation and the Allocation of Risks
Deposit policy has several distinct objectives: efficient financial services, adequate liquidity, protection of eligible creditors, lawful financial transparency, fiscal integrity, and financial stability. One policy can improve one of these outcomes while worsening another. Restrictions on certain funding forms can increase resilience but raise the private cost of intermediation. Reporting can improve tax administration while imposing substantial implementation responsibilities. Guarantees can stabilize funding while shifting losses toward a collective arrangement.
Assessment consequently requires specifying the perspective. From the bank office's viewpoint, foreign deposits are liabilities and a source of funding. From the depositor's viewpoint they are assets whose value depends on liquidity and contractual performance. From a host jurisdiction's viewpoint they are potentially large external liabilities associated with domestic institutions. From a currency issuer's viewpoint, they form part of an international monetary network whose liquidity needs may become significant in stress. From the fiscal authority's viewpoint, they can be relevant to tax reporting without their face value establishing the taxable base.
There is no general reason for an increase in offshore deposits to be either beneficial or damaging across all perspectives. The relevant evidence includes how the funds are mobilized, the assets they finance, how the claims are distributed, maturity and currency risk, coverage by credible institutional safeguards, and compliance with relevant rules. Cross-border monetary coordination is justified by interdependence; the form and scale of such coordination depend on the evidence of stress transmission and legal mandates.
Policy benchmarks can be organized around matched indicators. Payment efficiency can be assessed through service costs and access metrics. Funding resilience can be assessed through liquidity and concentration measures. Compliance can be assessed through lawful reporting coverage, corrections of records, and documented tax outcomes. Distribution can be assessed through appropriate household and institutional data. Each requires a distinct corpus of evidence. Total deposits are informative but are rarely a sufficient outcome measure.
The reviewed empirical literature supports the conclusion that international information regimes can change the geographic distribution of reported deposit balances. It does not establish how much of every movement represents lawfully declared funds, undeclared income, substitution, or bank restructuring. A policy conclusion should therefore combine the direction of the reported treatment response with separate evidence about the mechanisms and welfare consequences rather than treat a balance-sheet change as a direct welfare measure.
Conclusion
Offshore deposits are a structural component of international banking because the jurisdiction of a financial claim, the location of its creditor, and the currency in which it is payable can be independently distributed. Their economic effects arise from the interaction of deposit contracts, banking balance sheets, international settlement, money markets, and national institutions. The relationship is visible in historical eurocurrency markets and persists in modern international funding networks.
The analysis establishes four substantive distinctions. A cross-border position and an offshore-currency position are overlapping categories, each with its own relevant statistical perimeter. A deposit and a security or custodial asset are distinct claims. A gross outstanding liability differs from a transaction flow and from the net position of a consolidated sector. The immediate statistical creditor differs from the beneficial owner when ownership is mediated through entities. These are established distinctions applied systematically to the offshore deposit research object.
The balance-sheet illustrations show that movement of a claim among banks can change its recorded geography without changing the depositor's total asset holdings. Interbank positions can produce multiple layers of real gross obligations without a corresponding increase in outside-sector deposits. Currency conversion can alter common-currency stocks in the absence of transactions. Maturity mismatch and foreign-currency settlement constraints can generate liquidity pressure even where a bank's accounting positions appear matched.
The evidence on international transparency reforms indicates that deposit holdings respond to changes in reporting institutions. Earlier bilateral arrangements and later automatic exchange must be evaluated in their respective policy and sample contexts. Published estimates are informative about the measured outcomes, and the differences among them warrant research into timing, country classification, and alternative mechanisms. They cannot establish the legal status or motivations of every account holder or serve as a direct measure of tax recovery.
The original contribution of this monograph is the integration of these results into an instrument-centered economic analysis. It connects deposit formation, funding networks, international statistics, and institutional reform through explicit accounting and study-specific interpretation. Its conclusions do not depend on a new named theory or an unverified estimate of the global stock. They establish that offshore deposits are best investigated as a class of internationally distributed bank liabilities and corresponding creditor assets, with measurement, risk, economic purpose, and accountability evaluated at the appropriate level of observation.
The future research priority is a transparent, reproducible cross-source examination of deposit positions and their evolution, with separate estimands for conventional deposits, wider bank funding, household assets, and reporting outcomes. Reliable interpretation requires making the institutional and statistical boundaries visible at every stage, from dataset construction to policy evaluation.
References
- Aldasoro, Iñaki, Torsten Ehlers, Patrick McGuire, and Goetz von Peter. 2020. Global Banks’ Dollar Funding Needs and Central Bank Swap Lines. BIS Bulletin 27, July 16.
- Alstadsæter, Annette, Niels Johannesen, and Gabriel Zucman. 2018. Who Owns the Wealth in Tax Havens? Macro Evidence and Implications for Global Inequality. Journal of Public Economics 162: 89–100.
- Alstadsæter, Annette, Niels Johannesen, Ségal Le Guern Herry, and Gabriel Zucman. 2026. Global Financial Transparency and Offshore Wealth Accumulation. NBER Working Paper 35505, July. DOI: 10.3386/w35505.
- Bank for International Settlements. 2026. International Finance Through the Lens of BIS Statistics: Offshore Activity. BIS Quarterly Review, March.
- Bank for International Settlements. n.d. Locational Banking Statistics: Overview and Glossary. BIS Data Portal. Accessed November 18, 2026.
- Basel Committee on Banking Supervision. n.d. Basel Framework: Liquidity Coverage Ratio (LCR). Accessed November 18, 2026.
- Beer, Sebastian, Maria Delgado Coelho, and Sebastien Leduc. 2019. Hidden Treasure: The Impact of Automatic Exchange of Information on Cross-Border Tax Evasion. IMF Working Paper 2019/286. DOI: 10.5089/9781513519975.001.
- Casi, Elisa, Christoph Spengel, and Barbara M. B. Stage. 2020. Cross-Border Tax Evasion After the Common Reporting Standard: Game Over? Journal of Public Economics 190: 104240.
- Committee on the Global Financial System. 2026. Foreign Currency Funding Risk and Cross-Border Liquidity. CGFS Papers 71, March 12.
- Diamond, Douglas W., and Philip H. Dybvig. 1983. Bank Runs, Deposit Insurance, and Liquidity. Journal of Political Economy 91 (3): 401–419. DOI: 10.1086/261155.
- European Central Bank. 2025. EU Balance of Payments and International Investment Position Statistical Sources and Methods. July 2025 edition.
- Hardy, Bryan, and Sonya Zhu. 2023. Unpacking International Banks’ Deposit Funding. BIS Quarterly Review, September.
- He, Dong, and Robert N. McCauley. 2012. Eurodollar Banking and Currency Internationalisation. BIS Quarterly Review, June.
- International Association of Deposit Insurers. 2025. Core Principles for Effective Deposit Insurance Systems. September 2025. Official core-principles publication page.
- International Monetary Fund. 2025. IMF and the Statistical Community Release New Global Standards for Macroeconomic Statistics. Press Release 25/072, March 20.
- Johannesen, Niels, and Gabriel Zucman. 2014. The End of Bank Secrecy? An Evaluation of the G20 Tax Haven Crackdown. American Economic Journal: Economic Policy 6 (1): 65–91. DOI: 10.1257/pol.6.1.65.
- McCauley, Robert N., Patrick McGuire, and Philip Wooldridge. 2021. Seven Decades of International Banking. BIS Quarterly Review, September.
- McGuire, Patrick, and Goetz von Peter. 2009. The US Dollar Shortage in Global Banking and the International Policy Response. BIS Working Papers 291.
- O’Reilly, Pierce, Kevin Parra Ramirez, and Michael A. Stemmer. 2019. Exchange of Information and Bank Deposits in International Financial Centres. OECD Taxation Working Papers 46. Paris: OECD Publishing. DOI: 10.1787/025bfebe-en.
- OECD. 2025. Consolidated Text of the Common Reporting Standard (2025): Standard for Automatic Exchange of Financial Account Information in Tax Matters. Paris: OECD Publishing. DOI: 10.1787/055664b1-en.
- Schenk, Catherine R. 1998. The Origins of the Eurodollar Market in London: 1955–1963. Explorations in Economic History 35 (2): 221–238. DOI: 10.1006/exeh.1998.0693.